Assessment Issues in Child Neuropsychology

Assessment Issues in
Child Neuropsychology
Critical Issues in Neuropsychology
Series Editors:
Cecil R. Reynolds Antonio E. Puente
Texas A&M University University of North Carolina, Wilmington
Editorial Advisory Board:

Erin Bigler, University of Texas, Austin
Raymond S. Dean, Ball State University
Hans J. Eysenck, University of London
Charles J. Golden, Drexel University
John Gruzelier, University of London
Lawrence C. Hartlage, Fairfax University
Merril Hiscock, University of Saskatchewan
Lawrence Majovski, Huntington Memorial Hospital
Francis J. Pirozzolo, Baylor University School of Medicine
Karl Pribram, Stanford University
ASSESSMENT ISSUES IN CHILD NEUROPSYCHOLOGY

Edited by Michael G. Tramontana and Stephen R. Hooper
HANDBOOK OF CLINICAL CHILD NEUROPSYCHOLOGY

Edited by Cecil R. Reynolds and Elaine Fletcher-Janzen
MEDICAL NEUROPSYCHOLOGY: The Impact of Disease on Behavior

Edited by Ralph E. Tarter, David H. Van Thiel, and Kathleen 1. Edwards
NEUROPSYCHOLOGICAL FUNCTION AND BRAIN IMAGING

Edited by Erin D. Bigler, Ronald A. Yeo, and Eric Turkheimer
A Continuation Order Plan is available for this series. A continuation order will bring
delivery of each new volume immediately upon publication. Volumes are billed only
upon actual shipment. For further information please contact the publisher.
Assessment Issues in
Child Neuropsychology
Edited by
Michael G. Tramontana
Bradley Hospital
East Providence, Rhode Island
and Brown University
Providence, Rhode Island
and
Stephen R. Hooper
Clinical Center for the Study of Development and Learning
University of North Carolina School of Medicine
Chapel Hill, North Carolina
Springer Science+Business Media, LLC

Library of Congress Cataloging in Publication Data
Assessment issues in child neuropsychology / edited by Michael G. Tramontana and

Stephen R. Hooper.
p. cm. - (Critical issues in neuropsychology)
Includes bibliographies and index.
1. Developmental disabilitiesDiagnosis. 2. Neuropsychological tests for children.
I. Tramontana, Michael G. II. Hooper, Stephen R. III. Series.
[DNLM: 1. Child Development Disordersdiagnosis. 2. Neuropsychological Tests
in infancy & childhood. 3. Neuropsychology-in infancy and childhood. WS 340
A8453] RJ131.A789 1988
618.92 8588-dcl9
/
DNLM/DLC 88-22399
for Library of Congress CIP
ISBN 978-1-4757-9303-1 ISBN 978-1-4757-9301-7 (eBook)
DOI 10.1007/978-1-4757-9301-7
1988 Springer Science+Business Media New York

Originally published by Plenum Press, New York in 1988
Softcover reprint of the hardcover 1st edition 1988
All rights reserved
No part of this book may be reproduced, stored in a retrieval system, or transmitted

in any form or by any means, electronic, mechanical, photocopying, microfilming,
recording, or otherwise, without written permission from the Publisher
To Maryanne and our son, Michael Joseph, and to our
anticipated new arrival
M.G.T.
To my family
S.R.H.
Contributors
GLEN P. AYLWARD, Division of Developmental and Behavioral Pedi-

atrics, Departments of Pediatrics and Psychiatry, Southern Illinois
University School of Medicine, Springfield, Illinois
RUSSELL A. BARKLEY, Department of Psychiatry, University of Mas-
sachusetts Medical Center, Worcester, Massachusetts
ERIN D. BIGLER, Austin Neurological Clinic, and Department of Psychol-
ogy, University of Texas at Austin, Austin, Texas
THOMAS A. BOYD, Cleveland Metropolitan General Hospital, and Case
Western Reserve University School of Medicine, Cleveland, Ohio
ROBERT T. CONNOR, Kennedy Institute, Johns Hopkins School of Medi-
cine, Baltimore, Maryland
MICHAEL A. CRARY, Departments of Communicative Disorders and
Neurology, University of Florida Health Science Center, Gainesville,
Florida
RAYMOND S. DEAN, Neuropsychology Laboratory, Ball State University,
Muncie, Indiana, and Indiana University School of Medicine
DOROTHY EDGELL, Department of Psychology, Jack Ledger Child and
Adolescent Psychiatric Unit, Arbutus Society for Children, Victoria,
British Columbia, Canada
JANE M. FLYNN, Gundersen Medical Foundation, LaCrosse, Wisconsin
JEFFREY W. GRAY, Neuropsychology Laboratory, Ball State University,
Muncie, Indiana
NANCY J. HAAK, Departments of Communicative Disorders and Neu-
rology, University of Florida, Gainesville, Florida
STEPHEN R. HOOPER, Department of Psychiatry, University of North
Carolina School of Medicine, and Clinical Center for the Study of
vii
viii CONTRIBUTORS
Development and Learning, University of North Carolina, Chapel

Hill, North Carolina
GEORGE W. HYND, Departments of Educational Psychology and Psychol-
ogy, University of Georgia, Athens, Georgia, and Department of Neu-
rology, Medical College of Georgia, Augusta, Georgia
JOEL LEVY, Department of Psychology, Texas Institute for Rehabilitation
and Research, Houston, Texas
G. REID LYON, Departments of Neurology and Communication Science
and Disorders, University of Vermont, Burlington, Vermont, Gun-
dersen Medical Foundation, and Department of Special Education,
St. Michael's College, Winooski, Vermont
LOUISA MOATS, Associates in Counseling and Education, East Thetford,
Vermont
GRANT L. MORRIS, Department of Psychology, University of Northern
Colorado, Greeley, Colorado
NAOMI NIEVES, Kennedy Institute, Johns Hopkins School of Medicine,
Baltimore, Maryland
FRANCIS J. PIROZZOLO, Department of Neurology, Baylor College of
Medicine, and Neurology Service, Houston VA Medical Center,
Houston, Texas
ANTHONY H. RISSER, Department of Neurology, University of Wiscon-
sin Medical School, Mount Sinai Medical Center, Milwaukee, Wis-
consin
MICHAEL G. TRAMONTANA, Bradley Hospital and Department of Psy-
chiatry and Human Behavior, Brown University, East Providence,
Rhode Island
KYJTA K. S. VOELLER, Department of Neurology, University of Florida
Health Science Center, Gainesville, Florida
W. GRANT WILLIS, Department of Psychology, University of Rhode Is-
land, Kingston, Rhode Island
Foreword
Neuropsychology has its roots in clinical neurology. Reading case de-

scriptions by 19th century neurologists, such as Wernicke's painstakingly
detailed examinations of patients with the "aphasic symptom-complex,"
makes it obvious that neuropsychology is not a new discipline. Even the
marriage with psychology is not new; the neurologist Arnold Pick, for
example, was fully conversant with the developments in contemporary
psychological as well as linguistic research. However, the primary focus
of 19th and early 20th century psychology was on "general psychology,"
and only a small number of psychologists ventured into what then was
called "differential psychology" (the psychology of individual dif-
ferences) including a few who became attached to neurological research
and rehabilitation units after World War I. It remained until World War II
for psychologists to establish a more solid working relationship with
neurology.
What psychology had to offer to neurology was its experimental skill,
the development of a sophisticated methodology, and, for clinical work,
the development of psychometrics. On the whole, the marriage between
the two disciplines has been fruitful, leading to new insights, models, and
discoveries about brain-behavior relationships, documented in several
textbooks which appeared in rapid succession since the 1960s. In clinical
practice, neuropsychology has been inventive in some respects, in others
merely introducing psychometric rigor to already existing neurological
examinations.
As described in greater detail in this book, developmental neuropsy-
chology is of even more recent origin. Yet again, to a large extent this is
not a new field, but one which pediatric neurology and several allied
disciplines like neonatology, teratology, and embryology have developed
over the course of the century. In addition, developmental psychology
and developmental linguistics have developed a solid core of information
and theory, emphasizing a "psychobiological approach" for many years.
Integration of these many sources of knowledge into a neuropsychological
ix
x FOREWORD
framework has been attempted for individual topics during the last twen-
ty years and as integrated overviews only in recent years.
I think it is important to remember and acknowledge, as well as to
integrate existing knowledge into deve-Iopmental neuropsychology rather
than treat this field as terra nova. This book is a first criticall!Ppraisal of
the clinical application of child neuropsychology focusing on assess-
ment issues. It is to the credit of the editors and authors that they present
assessment issues not in isolation, but in the context of basic knowledge
about developmental neuropsychology, its history, and its neighbor disci-
plines. The book is not restricted to the critical evaluation of traditional
assessment methods, but includes a coverage of basic neurodevelopmen-
tal considerations, a comparison with pediatric, radiologic and elec-
trodiagnostic sources of information, different models of neuro-
psychological inference, and the implications of assessment for treatment.
It then moves on to the major topics of assessment of attention, memory,
functional laterality, language, and to more specific areas such as infancy,
early childhood, and learning disabilities. The successful integration of
these diverse areas is a major accomplishment of this book.
One recurrent problem is, of course, the practice of drawing on as-
sessment techniques developed for the adult patient, and extending these
methods downward. Within certain age limits and for some areas of func-
tion, this may be satisfactory. However, most techniques "bottom out" at
age 5 or 6, Le., at an age before the child becomes aquainted with the test-
taking attitude required in school. The preschool child has not yet ac-
quired these attitudes, and does not deliver a consistent performance.
More serious yet is the problem of developmental changes. To expect
the same tests to measure the changing component skills in the young
child which gradually develop into complex adult performance is haz-
ardous and disregards the accumulated knowledge of child development.
Hence, new and age-appropriate measuring instruments need to be devel-
oped and linked to neuropsychological theory. Most of the authors in this
book consider these problems carefully for their area of assessment and
suggest ways in which existing methods can be used in a neuro-
psychological framework; others find areas which are not covered suffi-
ciently and recommend a new design and a drastic re-development of
assessment methodology.
Finally, the limited behavioral repertoire of the infant, toddler, or
young child, allows fewer and fewer standardized psychological measure-
ments compared to what can be used with the adolescent and adult.
Observations of "baby holds rattle" or responses to light, sound,and touch
stimuli take the place of tests. Yet, observing the richness of the Brazelton
examination in newborns gives us an inkling of what is to come as the
child grows up. As Aylward describes in this book, the expansion of this
FOREWORD xi
basic repertoire should be examined carefully in the light of developmen-

tal neuropsychological theory.
A considerable portion of this book focuses on learning disabilities.
This rather complex area perhaps takes up so much of our interest be-
cause it started as a neurological theory, suggested by Pringle Morgan in
1896, as a "congenital" parallel to the aquired "word-blindness" as de-
scribed by James Hinshelwood and others. Although theories about the
forms and causes of learning disabilities have changed, and evidence
which substantiates the claim to neurological damage or dysfunction has
been found only recently, the neuropsychological focus has been present
throughout the century. In contrast, developmental neuropsychology has
been quite slow in exploring mental retardation, another area of important
neuropsychological inquiry which was discussed in a position paper by
Benton in 1974. Perhaps because of the very different social and educa-
tional demands of the mentally retarded, research in this field has devel-
oped in relative isolation, although a majority of this population suffers
from brain abnormalities. Only recently have some special syndromes,
e.g., Turner's, fragile X, and Down's syndrome attracted specific neuro-
psychological research.
Although the development of systematic neuropsychological assess-
ment methods for children remains the primary target of this book (and, I
am certain, of many future discussions arising from it), the future devel-
opment of such procedures should not exclude traditional assessments,
especially in the area of emotional, social and other adjustment areas.
Tramontana and Hooper's introduction and Bigler's and Hooper's chap-
ters make this point, but it is important to stress that emotional disorders
as well as cognitive disorders may arise from brain abnormalities so often
that the inclusion of an assessment of the adaptive status of the child is
almost mandatory unless neuropsychological assessment is viewed as
ancillary to other assessments.
It is also essential to see the child as it develops in its own special
environment, not just as a biological organism. For this reason, a careful
exploration of the influences and attitudes of family, neighborhood and
school, of resource availability, socio-economic status, and resour-
cesfulness of the parents often will explain some of the adjustment prob-
lems and even some of the cognitive and language problems of a given
child.
What emerges as the field of clinical child neuropsychology, then, is
a multifaceted composite. Neuropsychological assessment, almost by def-
inition, is not likely to account for more than a portion of the variance
presented by children. This portion is further reduced by the lack of direct
correspondence between lesion area and type of deficit as we have come
to expect from early studies of adults with clearly circumscribed damage,
xii FOREWORD
especially in gunshot and shrapnel victims. As outlined by Risser and

Edgell as well as by other authors in this book, brain lesions in children
are rarely clearly defined and circumscribed, nor do they show direct
cause-effect relationships even if age at onset is taken into account. A look
at Nichols and Chen's effort in 1981 to tease out the longitudinal effects of
pregnancy, delivery and early childhood factors in a population of 53,000
participants provides an appreciation of the small magnitude of variance
which can be explained by such antecedent variables as smoking and
convulsions of the mother during pregnancy, low fetal heart rate and
placental weight, head circumference at 12 months, and the Bayley Motor
Scales at 8 months among many others.
Progress in the neuropsychological assessment of the child often has
been preceded by the development of new research methods, such as
unique methods for determining lateralization in infancy. This process no
doubt will continue in the future. Yet, as we look for tools for clinical
neuropsychological assessment we also must be guided by what the edi-
tors describe as ecological validity in addition to construct and diagnostic
validity, i.e., by what the assessment contributes to the choice, delivery
and monitoring of treatment, and long-term prediction of outcome. This is
a difficult and lengthy task because it includes a consideration of different
treatment options and their effectiveness. Only a few authors, such as
Rourke and Bakker in the field of learning disabilities, have addressed
this task (also described by Lyon, Moats,and Flynn in this book). Ulti-
mately, however, the development of neuropsychology as a clinical disci-
pline cannot survive unless it provides answers to these questions.
Given the magnitude of the task of neuropsychological and related
assessments, one must marvel at the progress already shown in these
chapters. It is too early to say that child neuropsychological assessment
has come of age, but perhaps it has reached puberty.
Otfried Spreen
Victoria, B. C., Canada
Preface
Recent years have witnessed the emergence of child neuropsychology as a

growing subspecialty within clinical neuropsychology. The collective
efforts in this field over the last 15 years or so have been truly impressive.
They have succeeded in drawing attention not only to the special roles
that neuropsychological assessment can play in the clinical evaluation of
children, but also to the important differences in brain-behavior rela-
tionships that must be considered in evaluating children versus adults.
Much remains to be learned concerning the effects of brain dysfunction in
childhood, but the work to date has served to create a distinct base of
knowledge that is sure to grow.
The specialized nature of this field would not be apparent, however,
if one were only to inspect the current tools that are used in the neuro-
psychological assessment of children. Indeed, these largely represent
downward extensions of adult neuropsychological methods, most of
which lack a clear linkage with existing knowledge on the neuropsychol-
ogy of the developing brain. The range of well-validated assessment meth-
ods designed specifically for children is quite limited, especially for the
very young. New developments are needed, as the available methods
generally have not kept pace with the evolving aims and ever-widening
applications of child neuropsychological assessment.
The present volume is devoted specifically to an in-depth examina-
tion of assessment issues in child neuropsychology. Other key volumes
within the field have served to establish the importance of child neuro-
psychology and its potential contributions to the study and evaluation of
children with disordered brain function. It is time to pause and take
critical inventory of the current status of assessment in this field. This is
vitally important, as the future progress of child neuropsychology will
depend greatly on the availability of appropriate tools for assessing brain-
behavior relationships as they unfold in childhood. The intent of this
volume is not simply to lament over present shortcomings, but rather to
xiii
xiv PREFACE
help outline constructive ways in which progress can be achieved. It is

easy to criticize present assessment procedures, but it is important that
this be balanced by a genuine appreciation for the conceptual and prac-
tical complexities involved in the neuropsychological assessment of the
developing child. Emphasis is given not only to identifying critical areas
in need of further development, but also to providing informed opinions
as to which of the current methods offer particular promise in assessing
various aspects of child neuropsychological functioning.
This volume consists of 14 chapters divided into four parts. Parts I
and II provide an overview of assessment in child neuropsychology, both
in terms of the general approaches to assessment that are currently avail-
able as well as important conceptual issues that apply to the field as a
whole. These are followed by a special topics part (III) in which each
chapter focuses on a specific aspect of child neuropsychological assess-
ment. There were many more potential topics than possibly could have
been included in this section. We chose to emphasize certain aspects of
functioning that require particular consideration in childhood, including
attention, memory, functional laterality, and language development. The
assessment of learning disabilities is covered extensively, mainly because
it represents an area that has been the subject of intensive neuropsycho-
logical inquiry. This is balanced by a careful consideration of infant and
early childhood assessment-a topic that, by contrast, has received rela-
tively little attention in child neuropsychology. The final part (IV) of the
book offers an integration of the problems and prospects identified, high-
lighting crucial avenues in which a concerted effort in future work will be
especially needed.
The book is written at an advanced level and assumes that the reader
has some familiarity with basic brain-behavior relationships. It is
intended primarily for advanced students and professionals in the fields
of neuropsychology, clinical psychology, and school psychology. Its con-
tent also should be pertinent to professionals in other disciplines in-
volved in the study or care of children with brain dysfunction, including
neurologists, child psychiatrists, pediatricians, speech and language pa-
thologists, and special educators. Moreover, the emphasis on addressing
assessment issues in a fashion which integrates research in developmen-
tal neuropsychology should serve to provide useful guidelines for re-
searchers and practitioners alike.
We wish to extend our sincere appreciation to a number of indi-
viduals whose efforts helped us greatly in the completion of this book. We
are indebted especially to both Roseanne Rabideau and Elaine Nardolillo
for their superb assistance throughout the various stages of editing and
organizing chapter manuscripts. The dedication and countless hours that
they gave went well beyond the call of duty, and will be remembered
PREFACE xv
always. We also owe a special note of thanks to Maryanne Tramontana for

her patience and generosity in creating a very supportive atmosphere on
those many occasions when our work on this book extended late into the
night.
Stephen R. Hooper
Contents
Part I: INTRODUCTION
Chapter 1
Child Neuropsychological Assessment:
Overview of Current Status .................................... 3
Michael G. Tramontana and Stephen R. Hooper
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Historical Trends in Assessment ............................... 5
Current Approaches in Child Neuropsychological Assessment ... . 9
Fixed-Battery Approaches ................................... 9
Other General Approaches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Special-Purpose Measures ................................... 20
Applications ................................................. 21
Neurological Disorders ...................................... 21
Systemic Illness ............................................ 23
Psychiatric Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Learning Disabilities ........................................ 26
Rehabilitation and Assessing Recovery of Function. . . . . . . . . . . . 27
Conceptual and Practical Issues ..................... . . . . . . . . . . . 28
Summary.................................................... 32
References ................................................... 32
Part II: GENERAL ISSUES IN CHILD NEUROPSYCHOLOGICAL

ASSESSMENT
Chapter 2
Neuropsychology of the Developing Brain: Implications for
Neuropsychological Assessment ............................... 41
Anthony H. Risser and Dorothy Edgell
Introduction .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
xvii
xviii CONTENTS
Principles of Neural Development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

Proliferation and Migration .................................. 43
Differentiation .............................................. 44
Myelination ................................................ 47
Functional Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Summary .................................................. 51
Abnormalities in Neural Development. . . . . . . . . . . . . . . . . . . . . . . . . . 51
Developmental Neuropsychological Implications ................ 56
Conclusions .................................................. 59
References 60
Chapter 3
The Role of Neuropsychological Assessment in Relation to Other
Types of Assessment with Children ............................ 67
Erin D. Bigler
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Pediatric Neurological Exam (PNE) ............................. 68
Basic Components .......................................... 68
The Nature of Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Mental Status Characteristics of Neurologically Impaired
Children ............................................... 74
Relationship between Pediatric Neurological Exam and
Neuropsychological Assessment ......................... 75
Neuroradiological Tests ....................................... 75
Electrodiagnostic Tests ........................................ 80
Relationship between Electrodiagnostic Findings and
Neuropsychological Assessment ......................... 82
Psychological Assessment........................... .......... 84
Intellectual Assessment versus Neuropsychological Assessment
with Children: What's the Difference? .................... 86
Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
References ................................................... 88
Chapter 4
Neuropsychological Diagnosis with Children: Actuarial and
Clinical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
W. Grant Willis
Diagnostic Models ............................................ 94
Data-Diagnosis Contingency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Consideration of Data ....................................... 95
Relative Importance of Data ................................. 96
Actuarial Models ............................................. 97
Methods ................................................... 97
CONTENTS xix
Base-Rate Considerations ................................... 98

Validity of Diagnostic Criteria ............................... 100
Stability and GeneraIizability of Actuarial Rules. . . . . . . . . . . . . .. 101
Clinical Judgment ........................................... " 102
Interpretive Strategies ....................................... 103
Debiasing Techniques ....................................... 105
Assessment Design and Decision Rules ....................... 106
Summary .................................................... 107
References ................................................... 108
Chapter 5
From Assessment to Treatment: Linkage to Interventions with
Children ..................................................... 113
G. Reid Lyon, Louisa Moats, and Jane M. Flynn
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 113
Neuropsychological Assessment: Purposes and Measurement
Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 115
Relating Assessment to Treatment: Models and Studies .......... 119
Standardized Assessment Batteries: Implications for Treatment. .. 119
The Halstead-Reitan: Linkages to Treatment ............ '" . . .. 119
The Luria-Nebraska: Linkages to Treatment ................... 122
The K-ABC: Linkages to Treatment . . . . . . . . . . . . . . . . . . . . . . . . . .. 122
Selected Assessment Batteries: Linkages to Treatment ........... 123
Empirical Subtype Intervention Studies: The Lyon Research
Program. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 125
Clinical Subtype Intervention Studies: The Bakker Research
Program. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 131
Direct Assessment Intervention Studies: The Flynn Research
Program. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 132
Some Final Thoughts, Conclusions, and Directions .............. 135
Professional Preparation and Experience. . . . . . . . . . . . . . . . . . . . .. 135
Developmental Issues in Assessment ......................... 136
The Need for Dynamic Assessment. . . . . . . . . . . . . . . . . . . . . .. . . .. 137
The Need for Continued Classification Research. . . . . . . . . . . . . .. 138
References ................................................... 139
PART III: SPECIAL TOPICS IN ASSESSMENT
Chapter 6
Attention .................................................... 145
Russell A. Barkley
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 145
xx CONTENTS
Importance of Attention in Child Neuropsychology. . . . . . . . . . .. 146

Components of Attention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 148
Assessment of Attention. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 153
Behavior Rating Scales ...................................... 153
Psychometric Tests and Laboratory Measures ................. 159
Direct Observational Measures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 167
Discussion ................................................... 169
Summary .................................................... 170
References ................................................... 171
Chapter 7
Clinical Assessment of Memory in Children: A Developmental
Framework for Practice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 177
Thomas A. Boyd
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 177
Current Status of Memory Assessment with Children ............ 179
Models of Memory Development in Childhood ... . . . . . . . . . . . . . .. 185
Information-Processing Models .............................. 186
Interactionist Models of Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 188
Suggestions for a Children's Memory Battery: General Format .... 195
Suggestions for a Children's Memory Battery: Types of Tasks. . . .. 196
Clinical Relevance of the Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 199
Conclusions .................................................. 200
References ................................................... 201
Chapter 8
Assessing Functional Laterality ................................ 205
Jeffrey W. Gray and Raymond S. Dean
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 205
Methods of Laterality Assessment .............................. 206
Perceptual Asymmetry Techniques ........................... 207
Electrophysiological Measures ............................... 210
Lateral Preference Measures ................................. 211
Unimanual Performance Measures ........................... 214
Applications with Learning-Disabled/Language-Impaired
Children .............................................. 215
Clinical Implications .......................................... 218
References ................................................... 219
CONTENTS xxi
Chapter 9
Infant and Early Childhood Assessment ........................ 225
Glen P. Aylward
The Status of Neuropsychological Assessment in Infancy and
Early Childhood ........................................ 225
Conceptual Issues in Early Neuropsychological Assessment ...... 227
Differences in Assessing School-Age versus Younger
Children ............................................... 227
Consistency of Dysfunction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 229
Classification of Early Neuropsychological Findings ........... 231
Assessment Instruments ....... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 234
Newborn/Neonatal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 235
Infancy .................................................... 235
Early Childhood ............................................ 235
Useful Findings ............................................ 239
Directions .................................................... 239
Appendix .................................................... 245
References ................................................... 246
Chapter 10
Questions of Developmental Neurolinguistic Assessment. . . . . . . .. 249
Michael A. Crary, Kyjta K. S. Voeller, and Nancy J. Haak
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 249
Basics of Language Development ............................... 250
Prelinguistic Communication ................................ 251
Lexical Expansion Stage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 252
Grammatical Expansion Stage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 253
A Comment on Styles of Language Processing. . . . . . . . . . . . . . . .. 254
Relation of Neurological Maturation to Acquisition of Language
Skills ................................................. 254
Neuroanatomic Maturation .................................. 254
Hemispheric Functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 256
Language Assessment at Present-What's Covered,
What's Not? ........................................... 261
Prelinguistic Assessment (0 to 12+ Months) .................. 262
Lexical Expansion Assessment (12 to 24+ Months) ............ 263
Grammatical Expansion Assessment (2 to 4+ Years) ........... 264
Summary of Present Assessment Capabilities ................. 269
Guidelines and Questions Regarding Developmental
Neurolinguistic Assessment ........ . . . . . . . . . . . . . . . . . . . .. 269
xxii CONTENTS
A Concluding Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 273

References ................................................... 274
Chapter 11
Learning Disabilities Subtypes: Perspectives and Methodological
Issues in Clinical Assessment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 281
George W. Hynd, Robert T. Connor, and Naomi Nieves
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 281
Early Conceptualizations of Learning Disabilities ................ 282
Clinical Case Reports and Observations . . . . . . . . . . . . . . . . . . . . . .. 282
What's in a Name? .......................................... 283
Defining Learning Disabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 284
The Public Forum and PL 94-142 ............................ 284
Acknowledging the Neurological Etiology. . . . . . . . . . . . . . . . . . . .. 284
Multifactor Research .......................................... 285
Perspectives on the Single Factor Research ................... 285
Early Subtyping Studies ..................................... 286
Some Directions in Subtyping Research: Multivariate
Classification Approaches ............................... 287
Neurolinguistic Subtyping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 300
Neuroanatomical-Linguistic Perspectives. . . . . . . . . . . . . . . . . . . . . .. 301
The Wernicke-Geschwind Model. .. . . .. .. . . . . . . . . . . . . .. . . . .. 301
Clinical Implications for Process-Based Assessment ........... 304
Conceptual Framework for Clinical Evaluation .................. 305
Conclusions .................................................. 307
References ................................................... 308
Chapter 12
The Prediction of Learning Disabilities in the Preschool Child: A
Neuropsychological Perspective. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 313
Stephen R. Hooper
Introduction.. . . .. . . .. . . .. . . .. . . .. . . .. . . . . . . . . . . .. . . .. . . . . . . .. 313
The Importance of Early Prediction of Learning Disabilities ...... 314
Extent of the Problem ....................................... 315
Minimizing Educational Failure. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 316
Minimizing Social-Emotional Difficulties .................... 316
Effects of Environmental Influences .......................... 317
Summary .................................................. 318
The Prediction of Learning Disabilities: Current Status. . . . . . . . . .. 318
Recent Reviews of the Literature ............................. 318
Exemplary Studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 321
CONTENTS xxiii
Summary .................................................. 323

Issues and Directions ......................................... 324
Neurodevelopmental Theory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 325
Neuropsychological Constructs and Preschool Prediction ...... 328
Related Issues in the Prediction of Learning Disabilities. . . . . . .. 330
Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 331
References 332
Chapter 13
Electrophysiological Assessment in Learning Disabilities. . . . . . . .. 337
Grant 1. Morris, Joel Levy, and Francis J. Pirozzolo
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 337
Anatomical Correlates of Learning Disorders .................. 338
Neuroimaging Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 340
Electrophysiological Techniques and Findings .................. 343
Electroencephalography ..................................... 343
Evoked Potentials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 352
Conclusions .................................................. 362
References ................................................... 363
Part IV: COMMENT
Chapter 14
Problems and Prospects in Child Neuropsychological
Assessment .................................................. 369
Index ........................................................ 377

I
Introduction
1
Child Neuropsychological
Assessment
Overview of Current Status
MICHAEL G. TRAMONTANA and STEPHEN R. HOOPER
INTRODUCTION
There has been a tremendous growth of interest in child neuropsychology

as a distinct subspecialty within the broader field of clinical neuropsy-
chology. Whereas it was not too long ago that one could find only selected
chapters dealing with children in some of the major texts within the field
(e.g., Filskov & Boll, 1981; Reitan & Davison, 1974), there are now entire
volumes devoted exclusively to the topic (Gaddes, 1985; Hynd & Obrzut,
1981; Hynd & Willis, 1988; Ivan, 1984; Obrzut & Hynd, 1986a, 1986b;
Rourke, 1985; Rourke, Bakker, Fisk, & Strang, 1983; Rourke, Fisk, &
Strang, 1986; Rutter, 1983; Spreen, Tupper, Risser, Tuokko, & Edgell,
1984). Likewise, there has been a proliferation of symposia, workshops,
and journal articles covering various aspects of neuropsychological re-
search and practice with children. All of this certainly attests to the
emergence of child neuropsychology as a focus of vigorous interest and
inquiry.
Various factors have been cited as responsible for this increased in-
terest and activity in child neuropsychology. Hynd, Snow, and Becker
(1986), for example, underscored two factors that they saw as particularly
MICHAEL G. TRAMONTANA' Bradley Hospital and Department of Psychiatry and

Human Behavior. Brown University. East Providence. Rhode Island. STEPHEN R.
HOOPER. Department of Psychiatry. University of North Carolina School of Medicine and
Clinical Center for the Study of Development and Learning. University of North Carolina.
Chapel Hill. North Carolina
3
4 MICHAEL G. TRAMONTANA and STEPHEN R. HOOPER
important. The first had to do with the passage of the Education for All
Handicapped Children Act (Public Law 94-142, Federal Register, 1976)
and, among other things, its calling national attention to the fact that a
significant percentage of children suffer from learning disabilities and
other developmental handicaps presumed to have a neurodevelopmental
etiology. This, in turn, served to accentuate an important role for neuro-
psychological assessment both in the identification of specific neuro-
developmental disorders and in helping to devise appropriate educa-
tional plans. Another factor cited by Hynd et 01. concerns the increased
prevalence of children surviving neurological trauma. Advances in medi-
cal care have brought about a dramatic increase in the survival of children
whose conditions (or their treatment) have a potentially adverse impact
on the developing brain (e.g., very low birthweight associated with pre-
maturity, iatrogenic effects of childhood cancer treatment). Thus, whereas
mortality has decreased, there has been a relative increase in morbidity.
This has created a greater need for the careful assessment of the extent,
pattern, and developmental significance of possible neuropsychological
sequelae in these survivors of serious childhood illness.
Along with this increased demand for child neuropsychological as-
sessment has come the growing realization of important differences in
brain-behavior relationships between children and adults. Not only do
children differ in the types of brain insult commonly experienced, they
also differ with respect to the specificity of behavioral effects manifested,
the pattern and course of (re)acquisition of function after injury, the modi-
fying effects of ongoing developmental change, and the extent to which
deficits sometimes can be delayed or "silent" until later developmental
periods (Almli & Finger, 1984; Boll & Barth, 1981; Chadwick & Rutter,
1983; Rourke et 01., 1983). Developmentalists argue that a child should
never be viewed as simply a scaled-down version of an adult. Likewise,
an appropriate assessment of brain-behavior relationships in children
cannot be based simply on scaled-down versions of assessment methods
used with adults.
Unfortunately, many of the available assessment tools in child neuro-
psychology largely represent downward extensions of adult neuro-
psychological models and methods, many of which lack any clear linkage
with the neuropsychology of the developing brain. There is more to the
development of appropriate assessment procedures for children than sim-
ply gathering norms on adult tests for children of different ages (although
even this would be welcome in many instances). It also means more than
simply reducing the length or complexity of test procedures to make them
more practical for use with children. The measures themselves, as well as
their organization within an overall assessment strategy, must be tailored
to elucidate brain-behavior relationships as they unfold at different
points in a child's development. It is important that the field does not
OVERVIEW OF CURRENT STATUS 5
rush to meet the increased demand and new challenges it now faces
armed only with assessment tools borrowed from work with adults. New
challenges will require new approaches if the field of child neuropsychol-
ogy is to prosper and continue to grow.
This volume is devoted to underscoring important assessment issues
in child neuropsychology. In this chapter we will provide a general over-
view of historical trends within the field, the current approaches to as-
sessment, their clinical applications, and some of the key conceptual and
practical issues involved in the neuropsychological assessment of chil-
dren. We hope thus to set the stage for selected issues and topics to be
pursued at greater depth in subsequent chapters in this volume.
HISTORICAL TRENDS IN ASSESSMENT
The history of assessment in child neuropsychology can be dis-

tinguished into four distinct but overlapping stages of development.
These are similar to, but generally have lagged behind, the stages of devel-
opment that have characterized adult neuropsychology.
The first stage, the single-test approach, dominated the field from
about the mid-1940s to the mid-1960s. It was characterized by the use of
general, all-purpose measures for diagnosing brain damage or "organ-
icity." The approach was guided by the belief that brain damage, re-
gardless of its extent, location, or pathological process, manifested itself
in a unitary fashion-whether it was in terms of a loss of abstraction
abilities, perceptual-motor skills, or other factors. The goal was to differ-
entiate brain-damaged children from normals, and it was thought that a
single well-chosen measure could achieve that end. The issue of how
brain dysfunction was being manifested really did not matter because the
interest was more in the global differentiation of cases. Examples of tests
used in this fashion (although not necessarily with the encouragement of
their authors) included the Bender Visual Motor Gestalt Test (Bender,
1938; Koppitz, 1964), the Visual Retention Test (Benton, 1963), and the
Memory for Designs Test (Graham & Kendall, 1960).
The problems with the single-test approach have been well docu-
mented. For example, Herbert (1964) conducted a thorough review of the
tests in use up to the early 1960s and concluded that none achieved a
sufficient differentiation of brain-damaged and normal children to justify
its clinical use with individual cases. Even if there were such a justifica-
tion, there still would remain the issue of whether anything rEially is
gained by simply being able to classify a child as brain-damaged or not.
The single-test approach did have the historical significance of introduc-
ing psychologists into the role of making inferences regarding brain dys-
function. However, by today's standards, most neuropsychologists would
regard the continued reliance on this approach as constituting mal-

practice.
Next came the test battery/lesion-specification stage of neuro-
psychological assessment. Emhart and Graham are credited as being
among the first to apply a battery of psychological tests in assessing the
developmental outcomes of a heterogeneous group of children with docu-
mented brain damage (Emhart, Graham, Eichman, Marshall, & Thurston,
1963; Graham, Emhart, Craft, & Berman, 1963). They found that brain-
damaged children manifested deficits on verbal and conceptual measures
as well as on perceptual-motor tasks. Also, whereas no single measure
yielded a satisfactory discrimination of brain-damaged children, use of
the entire test battery did. This underscored the variability of brain
damage, and the need for a test battery covering a broad range of functions
to capture its effects. Shortly thereafter, Reitan and his colleagues (Reed,
Reitan, & Klove, 1965) reported on the successful discrimination of brain-
damaged children and normals using the Halstead-Reitan Neuropsycho-
logical Battery (HRNB). This battery (and its variants) quickly came to be
the predominant assessment approach of this period, at least in North
America.
There was another significant aspect to this stage of development in
neuropsychology. The work with adults had shown that test batteries
such as the HRNB not only provided a valid discrimination of brain-
damaged and normal subjects but also were reasonably accurate in dis-
tinguishing among brain-damaged patients who differed in terms of the
extent, location, and other characteristics of their lesions (see Reitan &
Davison, 1974). Indeed, neuropsychological test batteries achieved an
equal, if not superior, discrimination of lesion characteristics in com-
parison with many of the other neurodiagnostic methods available at the
time (Filskov & Goldstein, 1974). Inspired by this success with adults,
similar efforts were made with children, albeit with unimpressive results.
In the absence of significant hemimotor signs, even attempts to lateralize
early cerebral lesions on the basis of neuropsychological test performance
remains a controversial and poorly validated enterprise at best (Chadwick
& Rutter, 1983).
Both of the above stages constituted what Rourke (1982) has referred
to as the static phase of development in clinical neuropsychology. The
emphasis was on the detection and localization of brain lesions. The
approach (either with single tests or test batteries) was empirical, athe-
oretical, and geared heavily toward establishing cutoff scores and rules of
inference for the purpose of maximizing hit rates in categorical diagnosis.
These applications were quite appealing at the time because of the void
that existed in noninvasive neurodiagnostic technology until about the
mid-1970s.
This gave way to the functional profile stage, or what Rourke (1982)
has referred to as the cognitive phase of neuropsychological assessment.

Controversy over the validity of neuropsychological batteries to localize
brain lesions in children, coupled with the rapid development of other
noninvasive neurodiagnostic methods, resulted in a gradual de-emphasis
on using neuropsychological tests for making inferences regarding brain
lesions. The emphasis shifted more to the role of neuropsychological
assessment in specifying the behavioral effects of cerebral lesions. The
goal was to differentiate between spared and impaired abilities, func-
tional strengths and weaknesses, and so forth. The concern was not only
on the extent of impairment but also on the pattern of deficit and the
underlying components of impaired performance. It represented the "re-
psychologizing" of neuropsychology, in that the emphasis was on assess-
ing the psychological aspects of neurological insults and anomalies. Of all
the available neurodiagnostic methods, neuropsychological assessment
had a unique and complementary role to play in determining the mental
and behavioral effects of brain injury, in identifying functional assets and
deficits for the purpose of treatment planning, and in evaluating subse-
quent change.
A particularly important aspect of this period involved the neuropsy-
chological study of children with learning disabilities. Just as the field of
neuropsychology, in general, had moved away from unitary concepts and
single-test measures of brain dysfunction, so too the study of learning
disabilities progressed from single-factor research to multivariate research
and the identification of subtypes of disabled learners (see Chapter 11,
this volume). Neuropsychological test profiles, along with statistical clus-
tering techniques, were used in identifying more homogeneous groupings
of children having distinguishable patterns of disability. The work was
not exclusively descriptive and empirical but also entailed theory build-
ing and model testing. This was important not only because it involved
the extension of neuropsychological assessment into the realm of syn-
drome definition but also because it represented a major line of neuro-
psychological research focused on children. Furthermore, it promoted a
closer linkage between assessment and treatment, in that the differentia-
tion of learning disabilities into subtypes provided at least a theoretical
basis for the specification of differential approaches to treatment.
Although this stage of development represented a shift in the goals of
neuropsychological assessment, there were no dramatic changes or inno-
vations in the types of tests and measures being used. Basically, many of
the same measures that originally had been validated on their ability to
discriminate brain damage were being used now for the purpose of neuro-
psychological description and functional analysis. True, the existing
methods usually allowed for the assessment of a broad range of brain
function, but many of the measures never were designed to achieve a
detailed analysis of the underlying components of complex deficits. This
limited the specificity of neuropsychological description, and quite possi-

bly the degree of meaningful differentiation that could be achieved in
empirical studies of subtypes of neuropsychological disability.
The field now appears to have entered yet another stage of develop-
ment, with its chief characteristic being an emphasis on ecological valid-
ity. Issues identified in the previous stage continue to be addressed, but
now there is an added demand to relate assessment findings to an indi-
vidual's everyday functioning (Chelune & Edwards, 1981). Parents of
brain-impaired children and other consumers of neuropsychological ser-
vices want to be provided with more than just a delineation of the child's
deficits. They want to know precisely what these will mean in terms of
the child's everyday functioning and future potential. The emphasis is not
only on description but on making prescriptive statements regardiLg the
types of treatments and environments that will maximize adaptive
functioning.
Rourke (1982) has referred to this as the dynamic phase of develop-
ment in neuropsychological assessment, with the goal being to evaluate
the individual's current neuropsychological functioning in relation to the
specific environmental demands and developmental tasks that must be
faced. Rourke et 01. (1986) have incorporated this thinking in their model
of a treatment-oriented approach to the neuropsychological assessment of
children. According to the model, behavioral predictions and treatment
plans should be based on a careful consideration of "the interaction that
obtains between brain lesion(s) and the child's ability structure as this
impacts on adaptive behavior" (p. 251). This entails not only assessing
brain-behavior relationships within the child but also relating these to
the unfolding developmental demands of the child's immediate and long-
range environments. However, a potential limitation has to do with the
actual power of existing neuropsychological measures to reflect important
aspects of everyday functioning. As Rourke et 01. have pointed out, the
use of brief, homogeneous, narrow-band tests-although internally con-
sistent and stable-may be of little use in assessing the child's capacity to
meet complex environmental demands.
The foregoing overview has highlighted some of the major historical
trends in assessment that have characterized child neuropsychology. We
have seen the goals of assessment shift from a static emphasis on diagnos-
ing brain lesions to more of a focus on neuropsychological description and
the analysis of functional deficits, and, more recently, to an emphasis on
neuropsychological prescription that is relevant to everyday functioning.
Neuropsychologists will continue to make important contributions to diag-
nosing brain damage and delineating its effects, but in a fashion that
operates more from a biopsychosocial framework. The emphasis now is on
ecological validity, and the relationship between assessment results and
the individual's capacity to deal with important tasks of daily life. This is
one area in which the current status of neuropsychological assessment

with children actually appears to be somewhat ahead of work with adults.
That is, the major attention given to the neuropsychological assessment of
learning disabilities certainly has dealt with the child's adaptive capacity
(and future potential) in an important real-life context, namely, school.
However, as already noted, new developments are needed in assessment
measures, themselves, because these generally have not kept pace with the
evolving aims and ever-widening applications of neuropsychological
assessment.
CURRENT APPROACHES IN CHILD NEUROPSYCHOLOGICAL

ASSESSMENT
Fixed-Battery Approaches
A fixed-battery approach in neuropsychological assessment is one
that aims to provide a comprehensive assessment of brain function using
an invariant set of validated test procedures. The composition of the
battery is not tailored to the presenting characteristics of the individual
patient being assessed nor to the specific clinical hypotheses to be ad-
dressed. Rather, the emphasis is on administering as many of the desig-
nated procedures as the patient's condition will permit. Individual
variability is thought to be captured reasonably well so long as the battery
has been designed to tap a broad range of human capabilities. Moreover,
the use of a fixed battery across patients provides a standard data base on
which different clinical groups can be compared.
To date, fixed batteries such as the Halstead-Reitan Neuropsycho-
logical Battery (HRNB) and the Luria-Nebraska Neuropsychological Bat-
tery (LNNB) have represented the most commonly used approaches in
neuropsychological assessment (Hynd et aI., 1986). A detailed review of
these batteries is beyond the scope of this chapter, but some of the more
pertinent features of each are discussed below. The reader may wish to
refer to other available sources for a more thorough description of the
composition and validation of the HRNB (Boll, 1981; Reitan & Davison,
1974) and the children's revision of the LNNB (Golden, 1981, 1987). Also,
Hynd et al. (1986) have provided an excellent critical review of the valid-
ity and utility of both of these batteries in child neuropsychological
assessment.
Halstead-Reitan Neuropsychological Battery (HRNB)

Reitan and his colleagues developed two versions of the HRNB for
use with children: the Reitan-Indiana Neuropsychological Test Battery for
Children (ages 5 to 8) and the Halstead Neuropsychological Test Battery

for Children (ages 9 to 14). For convenience, both of these versions of the
battery, along with the supplemental tests that commonly are included,
will be referred to simply as the HRNB.
The HRNB for older children is a downward extension of the adult
version of the battery and includes the following procedures: Lateral
Dominance Examination, Aphasia Screening Test, Category Test (168
items), Tactual Performance Test (six blocks), Sensory-Perceptual Exam-
ination (including Tactile Finger Recognition, Tactile Form Recognition,
and Fingertip Number Writing), Grip Strength Test, Finger Oscillation
Test, Trail-Making Test, Speech-Sounds Perception Test (three-choice
format), and the Seashore Rhythm Test. The HRNB for younger children
consists of modified versions of the tests in the older children's battery,
but the Trail-Making Test, the Speech-Sounds Perception Test, and the
Seashore Rhythm Test are excluded. Six additional tests are included that
are suited specifically for use with younger children: Marching Test,
Color Form Test, Progressive Figures Test, Matching Pictures Test, Target
Test, and Individual Performance Test. Usually each version of the battery
is supplemented by the appropriate Wechsler scale for assessing intel-
ligence and by a standardized test of academic achievement. Performance
on the HRNB is evaluated in terms of Reitan's four methods of inference-
i.e., level of performance, pattern of performance, right-left differences,
and pathognomonic signs. For the older children's battery, these methods
of inference have been operationalized in an actuarial system of rules for
neuropsychological diagnosis (Selz & Reitan, 1979). Based on norms for 9-
to 14-year-olds, each rule converts raw scores on the various HRNB mea-
sures to scaled scores ranging from 0 to 3. However, the system uses only a
single set of norms for all children in this age range. Additional normative
data for 5- to 14-year-olds have been provided by Knights (1966) as well as
Spreen and Gaddes (1969).
The HRNB has been found to discriminate effectively between nor-
mal children and those with documented brain damage. This has been
demonstrated both with 5- to 8-year-olds (Klonoff, Robinson, & Thomp-
son, 1969; Reitan, 1974) and with 9- to 14-year-olds (Boll, 1974; Reed et
al., 1965). In each of these studies, brain-damaged children performed
more poorly on most of the test variables constituting the HRNB, with
their deficits spanning both verbal-conceptual and perceptual-perfor-
mance abilities. Interestingly, test variables drawn from the Wechsler In-
telligence Test for Children were consistently among the most discrimi-
nating measures.
In general, attempts to localize brain injury in children with the
HRNB have met with little success (Chadwick & Rutter, 1983). Perhaps
this is because children, unlike adults, are more likely to experience gen-
eralized as opposed to focal brain damage (Kolb & Whishaw, 1980). Also,
secondary pathological changes may serve to produce more generalized

neuropsychological features even among children with seemingly focal
lesions (Hynd et 01., 1986). Or it simply may be that the neuropsycholog-
ical deficits associated with localized brain damage are less specific in
children than in adults (Chadwick & Rutter, 1983). In any event, many of
the customary "rules" for localizing lesions in adults cannot be applied to
children. For example, differences between Wechsler Verbal IQ and Per-
formance IQ do not provide a valid means of lateralizing brain damage to
either the left or the right hemisphere in children with early lesions; if
anything, a pattern of lower Performance IQ than Verbal IQ seems to be
characteristic of children with generalized brain damage (Boll & Barth,
1981). Overall, in the absence of lateralized sensory or motor signs, there
is little evidence to support the validity of the HRNB in lateralizing child-
hood cerebral lesions.
In contrast to the groups of children considered so far, children with
learning disabilities have presumptive, as opposed to documented, evi-
dence of brain dysfunction. One would expect them to show less overall
impairment than a group of children with known brain damage, and to
pose more of a diagnostic challenge for a test battery to identify. Using the
HRNB, Selz and Reitan (1979) achieved an overall accuracy of about 73%
in correctly classifying 9- to 14-year-old children as normal, learning-
disabled, or brain-damaged. Misclassifications were almost entirely in the
direction of false negatives, because their system of rules for neuropsy-
chological diagnosis tended to underestimate dysfunction in both the
learning-disabled and brain-damaged groups. Not surprisingly, the learn-
ing-disabled children showed an intermediate level of impairment in
comparison with the other two groups; the hit rate rose to 87% when the
learning-disabled group was excluded and the remaining subjects were
classified as either brain-damaged or normal. Using a test battery based
largely on the HRNB, Rourke and his colleagues likewise have shown that
learning-disabled children can be distinguished reliably on the basis of
their neuropsychological results. Perhaps more importantly, different
subtypes of learning disability have been distinguished on the basis of
differential patterns of neuropsychological performance (e.g., Rourke &
Finlayson, 1978).
There has been very little investigation into either the construct va-
lidity of the individual measures of the HRNB or on the factorial composi-
tion of the battery as a whole. Crockett, Klonoff, and Bjerring (1969) con-
ducted a factor analysis of the younger children's version of the HRNB
administered to a sample of normal children. The factors that accounted
for the most variance included: (1) a perceptual-analytic factor (15.5%),
(2) motor speed (8.3%), and (3) verbal concept formation (6.9%). Although
Wechsler IQ scores were not included in the analysis, the Block Design
and Object Assembly subtests were the variables that contributed most
strongly to Factor I, whereas the Vocabulary subtest had the highest rank
of the variables loading on Factor III.
Several studies have examined the interdependence between intel-
ligence, as assessed on the Wechsler Intelligence Scale for Children-Re-
vised (WISC-R), and performance on the older children's version of the
HRNB (Klesges, 1983; Seidenberg, Giordani, Berent, & Boll, 1983; Tra-
montana, Klee, & Boyd, 1984). Taken together, the results of these studies
have indicated substantial overlap between general intelligence and over-
all performance on the HRNB (e.g., r = .59 in the Tramontana et al. study).
The greatest influence of IQ has been found on the tests tapping more
complex skills and problem-solving abilities, language functions, and au-
ditory-perceptual analysis. In contrast, measures of basic tactile percep-
tion, motor strength and speed, and right-left hand differences have not
been found to be affected by IQ and thus appear to represent the more
distinctive and nonredundant aspects of functioning assessed by the
HRNB. Moreover, there is some indication that the overlap between gen-
eral intelligence and performance on the HRNB is greater in brain-
damaged children than among normals (Boll & Reitan, 1972).
Critique. The HRNB clearly has been the dominant approach to child
neuropsychological assessment for over the last 20 years. It has been used
in research on childhood brain dysfunction not only by Reitan and his
colleagues but also in some fashion by many other investigators. Reitan
certainly must be credited for pioneering a rigorous clinical approach to
neuropsychological diagnosis through the use of a standard battery of
tests. The strict emphasis on standardization has had the benefit of pro-
moting replicability and comparability of findings across clinicians and
researchers alike. It also has provided a standard data base on which
different clinical groups can be compared.
The HRNB provides a reasonably accurate means of distinguishing
brain-damaged, learning-disabled, and normal children. To date, howev-
er. there is little evidence to support its validity in localizing early brain
lesions or in specifying other lesion characteristics in brain-damaged chil-
dren. Its discriminative power appears to be limited to fairly broad diag-
nostic categories that, to some degree, can be differentiated on the basis of
intelligence testing alone. It has been shown to overlap substantially with
IQ, with its nonredundant contributions to assessment mainly involving
measures of basic motor and sensory-perceptual abilities. Given this,
comparable accuracy probably could be achieved in discriminating brain-
damaged, learning-disabled, and normal children simply through the use
of standard measures of intelligence and academic achievement supple-
mented by selected tests of motor and sensory-perceptual abilities.
Consistent with the goals of neuropsychological assessment at the
time of its development, the composition of the HRNB was based heavily
on the selection of tests sensitive to brain damage. Although the battery

permits a broad differentiation of spared and impaired abilities, its com-
position does not lend itself readily to conducting a more detailed analy-
sis of a child's functional capabilities. Some tests require such a complex
interplay of abilities that it often is difficult to distinguish the precise
components of deficient performance. Language skills are assessed rather
crudely, and the response format for the Speech-Sounds Perception Test
is such that results often can be confounded by a child's reading or spell-
ing difficulties. Memory is not assessed adequately, and other important
areas of a child's functioning, such as attention, are not assessed at all. No
test battery can be expected to cover everything, but the range and speci-
ficity of coverage should be better for a test battery as time-consuming to
administer as the HRNB.
Luria-Nebraska Neuropsychological Battery-Children's Revision

(LNNB-CR)
Golden (1981) introduced a downward extension of the LNNB for
children ranging from 8 to 12 years of age. The selection of test items was
determined by administering the standard adult version of the battery
(Golden, Hammeke, & Purisch, 1980) to normal children of above-average
ability. Items that proved to be too difficult were eliminated, some new
items were added, and other modifications in administration and scoring
were made to adapt the battery for use with children. This resulted in a
battery consisting of 149 test items (versus 269 in the adult version) on
which normative data then were obtained from a sample of 125 normal
children (25 for each age level from 8 to 12).
The LNNB-CR is organized in terms of 11 summary scales similar to
those in the adult version: Motor, Rhythm, Tactile, Visual, Receptive
Speech, Expressive Speech, Writing, Reading, Arithmetic, Memory, and
Intellectual Processes. Also, like the adult version, these are supple-
mented by three second-order scales: the Pathognomonic (consisting of 13
items in the battery that provide the best discrimination of brain damage),
and the Left Sensorimotor and Right Sensorimotor scales (based on items
on the Motor and Tactile scales performed with the contralateral hand).
Each item in the battery is scored on a 3-point system (0 to 2), with higher
scores being indicative of impairment. Item scores are summed within
each scale, with each total then converted to a T score (mean = 50, stan-
dard deviation = 10). The usual rule for classifying a child as impaired is
based on the presence of elevations on at least two of the summary scales
that exceed a critical level cutoff adjusted for the child's age (Gustavson et
01.,1984). This is computed by the regression formula: 82.02 - (.14 x Age
in months). Modified criterion rules have been recommended for max-
imizing the discrimination of neuropsychological impairment in special
populations, including children with learning disabilities (Geary, Jen-

nings, Schultz, & Alper, 1984) or psychiatric disorders (Tramontana &
Boyd, 1986).
In addition to examining the level and pattern of scale elevations on
the LNNB-CR, interpretation is based on a careful analysis of performance
on the individual items making up the battery. In compairson with the
HRNB, the test items on the LNNB-CR are relatively simple and are geared
more toward assessing the component skills underlying broader dimen-
sions of function. It is this feature of the battery that is thought to incorpo-
rate Luria's (1966, 1973) emphasis on the qualitative analysis of compo-
nent skill deficits. However, the reliability of this level of analysis is
questionable in that the assessment of specific component skills generally
rests on very few test items. This is more of a problem on the LNNB-CR
than on the adult version of the battery because of the reduction of nearly
half of the total item pool.
The LNNB-CR has been found to discriminate effectively between
brain-damaged children and normal controls (Gustavson et 01., 1984;
Sawicki, Leark, Golden, & Karras, 1984; Wilkening, Golden, MacInnes,
Plaisted, & Hermann, 1981). In the Gustavson et 01. study, the presence of
two or more scales above critical level cutoff yielded a correct classifica-
tion rate of 79% for brain-damaged children, 89% for normals, and an
overall hit rate of 85%. To date, however, there is little evidence as to the
validity of the LNNB-CR in distinguishing different lesion characteristics
in brain-damaged children. It does discriminate effectively between learn-
ing-disabled children and normals, but, in controlling for IQ, its discrimi-
native power tends to be limited to the Expressive Speech, Writing, and
Reading scales (Geary & Gilger, 1984; Nolan, Hammeke, & Barkley, 1983).
It appears to be less sensitive to factors underlying math deficits than
reading or spelling deficits in learning-disabled children (Nolan et 01.,
1983), and its capacity to discriminate the severity of learning disability is
eliminated once IQ and overall academic achievement level are con-
trolled (Snow, Hynd, & Hartlage, 1984).
A number of studies have found the LNNB-CR to add significantly to
the discrimination obtained through the WISC-R alone. The overlap be-
tween the two measures appears to be greater among brain-damaged chil-
dren than in either children with psychiatric disorders or normal controls
(Sweet, Carr, Rossini, & Kaspar, 1986). Among children with psychiatric
disorders, impairment on the LNNB-CR has been found to be associated
with differences in brain density as assessed through computed tomogra-
phy (Tramontana & Sherrets, 1985), as well as in the severity of behavioral
disturbance (Tramontana et 01., in press), with neither set of findings being
attributable to differences in IQ alone. It provides a better prediction of
academic achievement for this population than WISC-R IQ, whether ex-
amined separately or in conjunction with behavioral and demographic
factors (Tramontana, Hooper, Curley, & Nardolillo, 1988). This is surpris-

ing, given that overall results on the WISC-R and LNNB-CR have been
found to correlate quite strongly in this population (- .69), with the cor-
relations between IQ and every LNNB-CR scale being significant and rang-
ing from - .35 to - .66 (Tramontana et al., 1984). Besides the fact that
specific neuropsychological deficits may contribute to learning impedi-
ments, the enhanced ability of the LNNB-CR to predict academic achieve-
ment on a standard measure such as the Woodcock-Johnson Tests of
Achievement can be attributed to its inclusion of a screening of basic
academic skills on the Writing, Reading, and Arithmetic scales (Hooper &
Tramontana, 1987).
Factor-analytic studies have been conducted examining both within-
scales factors (Gustavson, Wilkening, Hermann, & Plaisted, 1982; Karras
et al., 1987) and across-scales factors on the LNNB-CR (Karras et al., 1987;
Snow & Hynd, 1985). In the Karras et al. study, which was based on a
mixed sample of 719 children, a total of 11 factors were extracted from the
149 items of the LNNB-CR: (1) general academic skills, (2) spatial organi-
zation, (3) spatially based movement, (4) motor speed, (5) drawing quali-
ty, (6) drawing speed, (7) rhythm perception and reproduction, (8) soma-
tosensory function, (9) basic receptive language, (10) simple expressive
speech, and (11) abstract verbal thinking. This was different from the
factor structure that has been described for the HRNB (Crockett et al.,
1969). Specifically, the inclusion of a General Academic Skills factor,
together with the greater number of discrete factors that constitute the
LNNB-CR, points toward the presence of important differences in the
composition of these two batteries.
Nonetheless, the overall results of the HRNB and LNNB-CR have been
found to correspond very highly, with a 91 % rate of agreement in identify-
ing neuropsychological impairment in brain-damaged children (Berg et
a1., 1984) and as much as 86% agreement in children with psychiatric
disorders (Tramontana, Sherrets, & Wolf, 1983). In the Tramontana et a1.
(1983) study, however, differences in the results of the two batteries were
apparent at a closer level of inspection. For example, Speech-Sounds
Perception on the HRNB did not correlate significantly with the Receptive
Speech scale of the LNNB-CR; also, the two batteries yielded discrepant
appraisals of right-left hand differences.
Critique. A good deal of research on the LNNB-CR has been generated

in the relatively short time since its introduction as an experimental in-
strument in 1981. It appears to discriminate brain dysfunction in children
about as well as the HRNB, and, at least for general classification pur-
poses, the two batteries yield highly comparable results. This has impor-
tant practical implications, given that the LNNB-CR takes about half the
time to administer. There is some indication that it adds to the discrimi-
nant validity achieved by the WISC-R alone in children manifesting

milder forms of brain dysfunction. However, in the case of children with
learning disabilities, the discrimination probably is no better than what
would be achieved when standard measures of intelligence and academic
achievement are combined. As with the HRNB, the nonredundant contri-
butions to assessment by the LNNB-CR mainly appear to involve its mea-
sures of sensory-perceptual and motor abilities.
An attractive feature of the LNNB-CR is the relative simplicity and
range of coverage of its test items, which allow for more of an analysis of
component skill deficits. The item format lends itself rather nicely to
providing a detailed specification of functional strengths and deficits.
However, this very positive feature is weakened by a number of psycho-
metric problems exhibited by the present form of the battery. There sim-
ply are too few items tapping specific component skills for these to be
assessed reliably, and the inclusion of items on a given summary scale is
based largely on nominal factors rather than statistical confirmation of
internal consistency. The present norms are rather weak (based on only 25
subjects for each year level) and cover a limited age range. Also, what is
essentially a pathognomonic-sign approach to scoring items on the LNNB-
CR places a major constraint on the degree of variation possible in assess-
ments of change.
Probably the chief weakness of the LNNB-CR is its lack of a firm
grounding in developmental neuropsychology. For example, Hynd et 01.
(1986) have criticized the battery for its exclusion of items assessing front-
al lobe function based on Golden's (1981) assumption that the frontal
lobes do not begin to reach functional maturation until adolescence. This
is contrary to available evidence (e.g., PassIer, Isaac, & Hynd, 1986) that
suggests that frontal lobe functions develop in a stepwise fashion, with
some functions being developed by about 6 to 7 years of age and others
continuing to mature into adolescence. Even Luria (1973), whose theory
guided the construction of the LNNB-CR, has suggested that prefrontal
cortical zones begin to develop at about 4 years of age. Moreover, develop-
mental considerations hardly could have been taken into account in the
omission of test items specifically assessing attention and learning-these
being critical aspects of functioning to assess in the developing child who
has sustained brain injury. Contrary to Golden's (1981) original claims,
the construction of the LNNB-CR does not appear to have been based on a
careful consideration of developmental factors. Like the HRNB, it simply
represents a downward extension of a test battery primarily designed for
adults.
Other General Approaches

Three other general approaches to neuropsychological assessment
will be considered briefly in this section: (1) eclectic test batteries, (2)
qualitative approaches, and (3) process-oriented approaches. There are

major differences among these in the rationale and guiding principles of
assessment, and in the relative emphasis given to quantitative versus
qualitative methods of analysis. However, they all share in the conviction
that an effective neuropsychological assessment cannot be based on a
restricted or fixed set of test procedures as exemplified by the HRNB and
LNNB-CR.
Eclectic Test Batteries

This approach strives to preserve the quantitative nature of neuropsy-
chological assessment by selecting standardized tests that, when taken
together, cover a broad range of neuropsychological functions. There gen-
erally is at least an implicit outline of the relevant functions and abilities
that should be assessed routinely. However, any of a variety of available
tests may be selected to quantify the extent of deficit in each of the func-
tional areas of interest. The psychometric properties of individual tests
(e.g., adequacy of norms) as well as their complementarity when embed-
ded in a battery are usually important factors guiding specific test selec-
tion.
For example, Smith (1975) presented the rationale for the selection of
tests in the Michigan Neuropsychological Test Battery, which includes
the WISC-R, Visual Organization Test, Raven's Coloured Matrices, Benton
Visual Retention Test, Purdue Pegboard, Symbol Digit Modalities Test,
Peabody Picture Vocabulary Test, Color Naming/Recognition, and Memo-
ry for Unrelated Sentences. From a somewhat different perspective,
Gaddes (1985) provided an outline of the Victoria Battery, which includes
recommended measures for assessing a child's functioning in each of the
following areas: intelligence, visual-spatial-constructional abilities, au-
ditory perception, tactile perception, sensorimotor integration, body im-
age, motor function, cerebral dominance, memory, language, educational
achievement, and personality. Many other examples of eclectic batteries
for children could be cited, but the foregoing should give some idea of the
possibilities that exist.
Qualitative Approaches
Rather than an emphasis on quantifying the extent of deficit, qualita-
tive approaches are concerned more with determining how the individual
passes or fails a particular task. For example, although a child may recall a
memory item accurately, it is extremely important from a qualitative per-
spective to understand what strategies the child used to remember. Did
the child utilize cues based on the verbal, visual, or multimodal aspects of
the task? Were contextual ones (e.g., main themes or key words) em-
ployed? Perhaps rehearsal, chunking, verbal labeling, or other mnemonic
strategies were utilized to some degree. These kinds of data are extremely
important not only in distinguishing the precise nature of underlying
neuropsychological dysfunction but also in identifying effective treat-
ment strategies that can be used in helping the child to compensate for
deficits in particular skills.
Qualitative approaches are less concerned with test standardization
or in comparing an individual's performance against general norms than
with conducting a careful idiographic analysis of performance for each
case. Almost any task may be used or modified in such a way as to
facilitate an in-depth analysis of where and how a particular individual's
performance breaks down. Informal testing procedures typically are used,
but even standardized tests can be utilized from a qualitative perspective
so long as the emphasis is on observing how the individual performs
instead of focusing only on the test scores that are obtained.
Probably the most recognized approach to qualitative assessment is
exemplified in the work of Luria (1966,1973) as described by Christensen
(1975). There are three major parts to Luria's Neuropsychological Investi-
gation. The first part consists of an initial evaluation of the individual's
functioning utilizing a hypothesis-testing approach. Based on the pa-
tient's initial performance, the second part employs a more individual-
ized set of tasks to explore suspected deficits more fully. Finally, once the
qualitative information is obtained, the results are formulated according
to Luria's theory of functional systems. Christensen organized the various
examination tasks used by Luria into 11 major areas of function. Golden et
a1. (1980) later used these as the original pool of items from which the
summary scales of the LNNB were developed.
Process-Oriented Approaches
These approaches represent a hybrid of quantitative and qualitative
methods in neuropsychological assessment. For example, in the Boston
Process Approach (Milberg, Hebben, & Kaplan, 1986) the emphasis is on
understanding the qualitative nature of the behavior assessed by psycho-
metric tests. It draws selectively from a core set of standardized tests for
the purpose of gaining a quantified overview of the patient's general pat-
tern of spared and impaired functions. Depending on the initial results,
the examination thereafter is guided by clinical hypothesis-testing aimed
at pinpointing the precise nature of the individual's deficits. This is
achieved through the use of various "satellite tests," which may consist of
standardized tests, the addition of new components to published tests, or
a set of tasks designed specifically for each patient. The possibilities are
limited only by the examiner's knowledge of available tests and his/her
creativity in designing new tasks for assessing particular deficit areas.
Using a similar logic, Wilson (1986) has presented a branching hy-
pothesis-testing model of neuropsychological assessment designed spe-

cifically for preschool-aged children. An initial selection of measures is
used for a general assessment of complex cognitive functions. On the
basis of these results, additional measures are selected for a further assess-
ment of specific functional areas. These are organized according to a
scheme of neuropsychological constructs that includes language (audito-
ry integration, auditory cognition, short-term auditory memory, semantic
retrieval), visual functions (visual-spatial, visual cognition, short-term vi-
sual memory), and motor functions (fine-motor, graphomotor). Wilson
provides an extensive listing of tests currently available for use with pre-
school-aged children that can be drawn upon in assessing these various
aspects of neuropsychological functioning.
General Critique
There are distinct advantages associated with each of the approaches
to neuropsychological assessment considered in this section. Eclectic bat-
teries preserve the quantitative nature of neuropsychological assessment
but are not bound to any particular tests or measures in achieving that
end. Rather, the emphasis is on selecting tests that are best suited to assess
particular abilities, which, taken together, span a broad range of function.
Tests may be substituted as better ones become available, and specific
tests may be added or deleted in adapting the battery to different applica-
tions. Moreover, the approach lends itself nicely to neuropsychological
description if tests are selected deliberately so that they are aligned with
well-defined areas of ability. Qualitative approaches offer maximum flexi-
bility and a richness of clinical data that surpasses what possibly could be
obtained through the use of standardized tests alone. The emphasis on
identifying the cognitive strategies available to the patient provides an
obvious linkage between assessment and treatment planning. Of all the
approaches considered, it is the one that is most dependent on the skills
and conceptual framework of the individual examiner. Process-oriented
approaches attempt to integrate quantitative and qualitative methods of
assessment, taking the best from both worlds so to speak. Also, as in
qualitative approaches, the emphasis in process-oriented approaches on
careful case study serves to play an important role in theory building.
However, a major problem with each of these approaches is that there
has been insufficient validation research on which to judge their ade-
quacy. The absence of standardization in qualitative approaches is incom-
patible with conducting independent appraisals of reliability and valid-
ity. Many American neuropsychologists would argue that standardization
and quantification are defining features of neuropsychological assess-
ment, and that qualitative approaches are more characteristic of methods
in behavioral neurology. An eclectic battery may utilize well-validated
individual tests, but little is known about the validity and utility of the
battery as a whole. Indeed, this would be difficult to assess if the composi-
tion of the battery were subjected to constant change. There also is the
problem of making comparisons among the measures within a battery that
may differ in terms of their normative base, test construction, and other
factors. Even if they were more carefully researched, it is doubtful that the
present examples of eclectic batteries would not share some of the prob-
lems identified with the HRNB and LNNB-CR. The flexible nature of test
selection in process-oriented approaches is thought to result in greater
diagnostic efficiency, but this is yet to be demonstrated. As Rourke et a1.
(1983) have pointed out, a flexible or individualized approach would be
at least as time-consuming as a fixed battery if an assessment of compara-
ble breadth is to be achieved. Also, a fixed-battery approach serves to
assure a consistently comprehensive assessment that is not biased by
either referral complaints or the child's initial presentation.
The differences among the various approaches to neuropsychological
assessment seem to have gotten overstated sometimes for the sake of aca-
demic argument. In practice, there probably are relatively few neuropsy-
chologists who are purists with respect to one approach or another, and a
melding of approaches probably is quite common. This makes good sense
because the different perspectives can be combined in a complementary
fashion in an effort to maximize both breadth and depth of assessment. A
fixed battery of procedures that is brief, but nonetheless spans a broad
range of abilities, helps to assure a consistently comprehensive assess-
ment that can produce comparable results across times, patient groups,
and different research settings. It provides a horizontal analysis of general
functioning that, along with qualitative observation, can be used in tailor-
ing an in-depth or vertical analysis of specific functional areas based on a
more flexible selection of tests. Process-oriented approaches achieve this
to some extent, but the validity of initial screening decisions may be
compromised by the absence of a fixed set of procedures that could assure
a consistently broad-based assessment for all cases.
Special-Purpose Measures
In addition to the general, all-purpose approaches to neuropsycholog-
ical assessment that we have discussed, there are a host of individual
measures and specialized test batteries available to assess more specific
aspects of neuropsychological functioning in children. Examples include
Benton's Motor Impersistence Battery and other special-purpose tests
(Benton, Hamsher, Varney, & Spreen, 1983), the Goldman, Fristoe, and
Woodcock (1974) Auditory Skills Battery, the children's version of the
California Verbal Learning Test (Delis, Kramer, Kaplan, & Ober, 1986),
and the Wisconsin Card Sorting Test (Heaton, 1981), to name only a few.
Some measures have been validated on the basis of their prediction of

specific types of childhood outcomes, as, for example, the Florida Kinder-
garten Screening Battery, which appears to provide an effective means of
identifying children at risk for reading disabilities (Satz & Fletcher, 1982).
There also are instruments for assessing specific features in selected pop-
ulations. For example, the Glascow Coma Scale (Teasdale & Jennett, 1974)
can assess level of consciousness in cases of traumatic brain injury, and
the Rancho Los Amigos Levels of Cognitive Functioning Scale (Hagen,
1981) provides a useful means of gauging overall recovery of function in
rehabilitation settings.
The above examples are mentioned only to give some idea of the
special-purpose measures currently available in child neuropsychology.
The available tools for assessing critical aspects of functioning in chil-
dren, such as attention, learning and memory, cerebral dominance, and
language functions, will be dealt with more extensively in Part III of this
volume.
APPLICATIONS
As was apparent in the discussion of historical trends in assessment,

the goals of child neuropsychological assessment have evolved to encom-
pass a broad range of clinical applications. In general, these include: (1)
aiding in the detection of brain dysfunction for the purpose of differential
diagnosis, (2) providing a precise specification of the behavioral effects of
known brain injury, (3) helping to identify the specific underlying dimen-
sions of dysfunction in particular handicaps, (4) using assessment data to
help formulate effective treatment strategies, (5) helping to assess the
child's prognosis and risk for certain developmental outcomes, and (6)
conducting ongoing assessments of functional change over the course of
development and in response to particular interventions. The precise
form and relative importance of these different applications of neuropsy-
chological assessment depend, to some extent, on the particular clinical
population under consideration. Here we will consider important ap-
plications in four different but overlapping clinical populations,
including children with neurological disorders, systemic illness, psychi-
atric disorders, and learning disabilities. We also will touch upon applica-
tions in pediatric rehabilitation and in assessing (re)acquisition of func-
tion after brain injury.
Neurological Disorders
There are literally dozens of neurological disorders in childhood that
may require a careful neuropsychological assessment. These include ge-
netic disorders (e.g., Turner's syndrome), structural abnormalities of the

brain (e.g., agensis of the corpus callosum, hydrocephalis), traumatic inju-
ries, and a variety of neuropathological processes including anoxic epi-
sodes, viral and bacterial encephalitis, toxicity (e.g., carbon monoxide
poisoning, lead poisoning), metabolic disorders, demyelinating diseases
such as multiple sclerosis, neuromuscular disorders such as muscular
dystrophy, and, more rarely in children, tumors (e.g., medulloblastomas,
craniopharyngioma, cerebellar astrocytomas) and cerebral vascular acci-
dents. Although case studies have appeared in the literature depicting
neuropsychological findings for selected childhood neurological disor-
ders (e.g., Rourke et 01., 1983), the effects of most insults to the immature
nervous system are only beginning to be understood. Suffice it to say that
the precise effects depend on a complex interplay involving the type of
injury, individual variables, and environmental factors.
One area in which neuropsychological methods have been used ex-
tensively is in children with seizure disorders. This is an important group
of disorders to assess, not only because they are among the most prevalent
neurological disorders of childhood (Meighan, Queener, & Weitman,
1976) but because their frequent chronicity carries the potential for con-
tinued impact throughout a child's life. There is no such thing as a "typ-
ical" neuropsychological profile for seizure-disordered children (Boll &
Barth, 1981), although, as a group, their overall cognitive abilities tend to
fall toward the lower end of the normal range (Bolter, 1984), they tend to
be at greater risk for learning difficulties (Yule, 1980), and they tend to
show a higher incidence of psychopathology than the general population
(Dreisbach, Ballard, Russo, & Schain, 1982; Rutter, Graham, & Yule, 1970).
The variability among children with seizure disorders certainly under-
scores the importance of careful individual assessment.
Childhood head injury is another example of an important area for
neuropsychological assessment. There are many factors to consider in
evaluating the child with a head injury, including the specific nature of
the injury and its resultant process (e.g., shearing, contre coup), age at
onset, premorbid cognitive and emotional functioning, duration of im-
paired consciousness, and the rate of posttraumatic recovery. The effects
of a closed head injury tend to be less localized than in the case of brain
tumors or cerebral vascular accidents, with deficits spanning a wide vari-
ety of functions (Chadwick, Rutter, Brown, Shaffer, & Traub, 1981). As
with adults, the first 6 months postinjury seems to be the time when
neuropsychological deficits are most prominent in children (Levin &
Eisenberg, 1979), with the bulk of recovery occurring during the first year
(Brink, Garrett, Hale, Woo Sam, & Nickel, 1970). However, neuropsycho-
logical changes have been observed to occur over a much longer period,
which sometimes may extend up to 5 years (Klonoff, Low, & Clark, 1977).
Besides the child's general adjustment, educational functioning is one of
the most important areas to be monitored. The child's social-emotional

adjustment also appears to be a critical dimension to assess because chil-
dren with accidental head injuries appear to be at higher risk for develop-
ing psychiatric difficulties (Brown, Chadwick, Shaffer, Rutter, & Traub,
1981).
Some neurological disorders may require neurosurgical intervention,
as in the case of partial lobectomies or complete hemispherectomies for
the treatment of severe seizures. Probably the most common neurosurg-
ical intervention in childhood involves the insertion of a shunt for the
treatment of hydrocephalis. Hydrocephalic children successfully treated
by shunt insertion have been reported to show difficulties in attention
(Hurley, Laatsch, & Dorman, 1983), memory (Cull & Wyke, 1984), and
visual-motor functions (Soare & Raimondi, 1977). One study also found
academic lags and poor impulse control, but a relative sparing of verbal
fluency, motor coordination, and verbal naming abilities (Fennell et 01.,
1987). Difficulties may result from the actual insertion of the shunt, and
secondary complications sometimes may contribute to severe impairment
in cognitive and adaptive functioning. Further, the neuropathology of
hydrocephalis, which may include fiber stretching, delays in myelina-
tion, and interference with thalamocortical connections as well as limbic
structures (Spreen et 01., 1984), can result in a rather varied neuropsycho-
logical picture in these children. It is important that the effects of this
disorder as well as its treatment be assessed individually, with careful
follow-up assessment of functions associated with so-called "silent"
brain regions that normally would not emerge until later developmental
periods.
Systemic Illness
Interest in the neuropsychological functioning of children with sys-
temic illness has shown considerable growth recently (e.g., Hynd &
Willis, 1988). There are a variety of pediatric illnesses that can impact
negatively on the developing nervous system, including defects in specif-
ic organ systems (e.g., pulmonary disease, congenital heart disease, renal
dysfunction), metabolic disorders (e.g., elevated blood phenylalanine,
defects in amino acid metabolism, lipid storage diseases, and mucopoly-
saccharidosis), autoimmune disorders, and infections (e.g., meningitis,
cytomegalovirus, Reye's syndrome). Evidence regarding the neuropsycho-
logical sequelae of such childhood illnesses has begun to emerge.
For example, Holmes and Richman (1985) examined the cognitive
profiles of children with insulin-dependent diabetes and found that chil-
dren with early disease onset (7 years of age or younger) and a chronic
course (5 years or more) were more likely to show reading and memory
difficulties. They also tended to have a lower Performance IQ on the
WISC-R, but this appeared to be due largely to slow response time rather
than to a specific deficit in visual-spatial abilities. Similar findings have
been obtained by other researchers (Ryan, Vega, & Drash, 1985; Ryan,
Vega, Longstreet, & Drash, 1984). However, interacting factors apparently
must be considered as well; i.e., Rovet, Gore, and Ehrlich (1983) found
that whereas females with diabetes onset prior to 3 years of age tended to
experience deficits in visual-motor functioning, their male counterparts
did not.
A major area in which neuropsychological assessment can play a
crucial role is in pediatric oncology. For example, acute lymphocytic
leukemia (ALL) accounts for 35% of all malignancies of childhood, with
approximately 40% of these cases becoming disease-free survivors (George,
Aur, Mauer, & Simone, 1979). An important question concerns the possible
iatrogenic effects of the various treatments required to stabilize these
children medically. The evidence is inconclusive, but there is some indica-
tion that children with ALL who receive intracranial radiation and chemo-
therapy (Methotrexate) may suffer a variety of posttreatment neuropsycho-
logical deficits (Goff, 1982; Waber, Sollee, Wills, & Fischer, 1986) and are
more likely to be later diagnosed as having learning disabilities (Elbert,
Culbertson, Gerrity, Guthrie, & Bayles, 1985).
As was noted before, advances in medical care have brought about a
dramatic increase in the survival of children with serious illnesses. This
has created a greater need for the careful assessment of possible neuropsy-
chological sequelae associated with these diseases or their treatment. On-
going neuropsychological assessment may contribute effectively to
monitoring improvement, stability, or deterioration in the child's func-
tioning, and may indicate the need for other kinds of intervention (e.g.,
special education). Its utility in assessing children with various pediatric
illnesses is only beginning to be realized.
Psychiatric Disorders
There is strong evidence that brain dysfunction in childhood is associ-
ated with an increased risk for psychiatric disorder. The risk is much
greater than for children with other types of physical handicap (Seidel,
Chadwick, & Rutter, 1975), and the relationship appears to hold both for
children with documented brain damage (e.g., Brown et aI., 1981; Rutter et
aI., 1970) and for those with so-called soft neurological signs (e.g., Shaffer
et aI., 1985). Among children with accidental head injuries, the risk is
compounded by factors such as psychosocial adversity and any pre-exist-
ing tendencies toward behavioral or emotional disturbance (Brown et a1.,
1981). The relationship is not trivial, given that the effects appear to persist
and influence the child's long-range behavioral adjustment (Breslau &
Marshall, 1985; Shaffer et aI., 1985).
Conversely, there is a relatively high rate of neuropsychological dys-

function among children with psychiatric disorders, even when cases
with known brain damage are excluded (Tramontana, Sherrets, & Golden,
1980). Tramontana and Sherrets (1985) found that neuropsychological
abnormality in such cases actually corresponded to differences on various
indices of brain density as revealed by CT results, which was a remarkable
finding in view of the exclusion of cases having documented neurological
involvement. Moreover, the presence of neuropsychological deficits has
been found to be associated with more extensive behavior problems
among younger boys with psychiatric disorders, regardless of factors such
as IQ, socioeconomic status, and whether the deficits could be linked
specifically with a history of brain injury (Tramontana et 01., in press).
Thus, the presence of neuropsychological deficits in childhood appears to
constitute an important index of increased psychiatric risk.
A number of studies have begun to examine whether there are dis-
tinguishing neuropsychological features associated with specific forms of
child psychopathology, including autism (e.g., Fein, Waterhouse, Lucci, &
Snyder, 1985), attention deficit disorder (e.g., Chelune, Ferguson, Koon, &
Dickey, 1986; PassIer, Isaac, & Hynd, 1986), and depression (e.g., Wilson
& Staton, 1984). In a comprehensive review of this research, Tramontana
and Hooper (in press) concluded that there was little evidence of specific-
ity with respect to the type or pattern of brain dysfunction associated with
different forms of child psychopathology. Although there were some
promising leads, the greater weight of evidence suggested a largely non-
specific, indirect relationship between brain dysfunction and psycho-
pathology in childhood. There is absolutely no evidence to support earlier
thinking (e.g., Wender, 1971) regarding the existence of a behavioral ster-
eotype among children with brain dysfunction, consisting of symptoms
such as hyperactivity, inattention, and impulsivity. Symptoms such as
these do not distinguish children with either documented brain damage
(Brown et 01., 1981; Rutter et 01., 1970) or soft neurological signs (Shaffer
et 01., 1985) becauee they appear to be common features of psychiatric
disorders in general, regardless of whether neurological abnormality is
present (Rutter, 1977). If anything, it appears that internalizing rather than
externalizing symptoms are more distinctively tied to brain dysfunction
in childhood, with symptoms such as anxiety, withdrawal, and depres-
sion being among the more common outcomes associated with a history of
chronic handicap (Breslau & Marshall, 1985; Shaffer et 01., 1985). Child
neuropsychologists may play an important role in determining more pre-
cisely how this process unfolds so that it might be redirected more
positively, if not prevented.
There are a number of issues surrounding valid neuropsychological
diagnosis with this population that require special consideration. These
have been discussed extensively by Tramontana (1983). Briefly, it is
important to apply more conservative rules for diagnosing neuropsycho-

logical impairment in children with psychiatric disorders because there
appears to be a greater likelihood of false-positive errors in diagnosis with
this population. Following this line of thinking, Tramontana and Hooper
(1987) were able to distinguish adolescent psychiatric referrals with and
without documented brain damage at a 76% rate of overall accuracy,
which was a substantial improvement over the results obtained with the
use of standard cutoffs for defining impaired performance.
There also has been some work examining the use of brief screening
procedures in identifying child psychiatric cases who are likely to show
significant abnormalities on a comprehensive neuropsychological assess-
ment. Tramontana and Boyd (1986) derived a regression formula, based
on the WISC-R and Aphasia Screening Test (AST), for predicting im-
paired/nonimpaired overall performance on standard test batteries such
as the HRNB and LNNB-CR in child psychiatric referrals. This approach
capitalized on the common use of the WISC-R in clinical settings, on the
substantial overlap that has been found between the WISC-R and the
results of a comprehensive neuropsychological test battery (Tramontana
et a1., 1984), and on the brevity and efficiency of the AST in identifying
verbal and spatial-motor deficits (Wolf & Tramontana, 1982). Application
of the regression formula in a cross-validation study (Boyd, Tramontana,
& Hooper, 1986) yielded a hit rate of about 80% in correctly predicting
impaired/nonimpaired neuropsychological status. Although not a sub-
stitute for a full neuropsychological assessment, procedures such as these
may help clinicians in identifying child psychiatric cases for whom a
comprehensive neuropsychological assessment is indicated.
Learning Disabilities
As was noted before, the study of learning disabilities has been one of
the most intensive areas of investigation in child neuropsychology. The
explicit presumption of central nervous system dysfunction in current
definitions of learning disability (Hammill, Leigh, McNutt, & Larsen, 1981)
has served to underscore the important role of neuropsychological assess-
ment in this field. Neuropsychological assessment can contribute effective-
ly both in the identification of specific underlying dimensions of dysfunc-
tion in particular learning handicaps and in the formulation and evaluation
of specific educational plans. The work in learning disabilities-which has
dealt with the child's adaptive capacity (and future potential) in the real-
life context of school-has taken an exemplary lead in relating neuropsy-
chological assessment to questions of ecological validity. Much remains to
be learned, but major advances have been achieved in the area of syndrome
definition and subtype analysis, and in the identification of developmental
precursors of learning disabilities, possible neuroanatomical and neu-
rophysiological factors, as well as relevant aptitude x treatment interac-

tions. These topics are dealt with extensively in subsequent chapters in this
volume. They are mentioned here for the sake of highlighting this very
important area of application in child neuropsychology.
Rehabilitation and Assessing Recovery of Function

Child neuropsychologists are relative newcomers to the field of pedi-
atric rehabilitation. The professional disciplines involved in providing
rehabilitation services to brain-injured children typically have included
physical therapists, occupational therapists, speech and language pa-
thologists, special educators, physicians, social workers, and perhaps be-
havioral psychologists. Child neuropsychologists now are entering this
arena in increasing numbers, bringing with them a unique combination of
skills. The child neuropsychologist's background in neurocognitive func-
tions, mechanisms of recovery of function, and child development and
psychopathology, together with skills in behavioral treatment and psy-
chotherapy, certainly serves to enhance the range of expertise represented
on the pediatric rehabilitation team. Roles may include not only assess-
ment and treatment consultation but also the actual delivery of treatment
services, particularly with respect to cognitive, neuromuscular, behav-
ioral, social-emotional, and family intervention components (Incagnoli &
Newman, 1985). Furthermore, with the recent explosion in computerized
cognitive rehabilitation software, neuropsychologists have taken the lead
in calling for standards regarding their marketing, clinical utility, and
usage (Kay, Becker, Bleiberg, & Long, 1986). Current thinking suggests
that many of these cognitive retraining packages are useful, but that their
application should be restricted to use by trained clinicians as part of a
more comprehensive treatment program (Adamovich, Henderson, &
Auerbach, 1985).
A key role for neuropsychological assessment is in the monitoring of
the child's (re)acquisition of function after brain injury. There is a com-
plex interplay of factors that influence the actual recovery patterns and
developmental progress of children who suffer early brain lesions (Che-
lune & Edwards, 1981). Systematic knowledge regarding the prognostic
significance of different types of early brain lesions is only beginning to
emerge (e.g., Dennis, 1985a, 1985b). The factors affecting the child's prog-
nosis certainly are more complex than suggested simply by the concept of
plasticity and the enhanced potential for recovery of function that pre-
sumably characterizes the immature brain.
There are many clinical examples of children with seemingly cata-
strophic brain injuries who somehow manage to survive and do reasona-
bly well, whereas others with comparatively minor injuries may be im-
peded by significant handicaps. Craft, Shaw,and Cartlidge (1972) showed
the importance of monitoring developmental progress by documenting

that even mildly brain-injured infants, who were described as fully re-
covered, continued to manifest cognitive, behavioral, and sensorimotor
deficits several years following their injury. Whereas deficits involving
motor function tend to be relatively stable over time, cognitive function-
ing may show considerable variability from infancy through early child-
hood (Aylward, Gustafson, Verhulst, & Colliver, 1987). Moreover, as
Rourke et a1. (1983) have pointed out, deficits involving so-called "silent"
brain regions may not become apparent until later developmental stages.
The child may develop normally at first but later may appear to "grow
into a deficit" when faced with new developmental demands. All of this
underscores the importance of ongoing neuropsychological assessment in
monitoring the child's developmental progress. Early prognostic formula-
tions must be viewed as tentative, at best, and subject to revision on the
basis of periodic reassessment of the child's actual functioning.
Neuropsychological assessment also can playa key role in helping to
formulate treatment plans, monitor treatment progress, and adjust the
child's rehabilitation program as needed. Rourke et a1. (1986) have pro-
vided a general framework for integrating neuropsychological assessment
in the treatment planning process. To date, however, little is known re-
garding the prescriptive significance of different patterns of neuropsycho-
logical data in the actual selection of treatments. Treatment options tend
to be selected on the basis of theory (e.g., Luria, 1966, 1973), personal
experience, or practical issues such as the availability of particular
therapeutic resources. There has been very little research that could serve
to link neuropsychological assessment and treatment more directly. This
is a critical new frontier for child neuropsychology that is greatly in need
of systematic inquiry. Issues involved in bridging the gap between assess-
ment and treatment are discussed more fully in Chapter 5, this volume.
CONCEPTUAL AND PRACTICAL ISSUES
We have seen that the effects of brain damage tend to be less specific
in children than in adults (Chadwick & Rutter, 1983). Because brain
damage primarily seems to impair new learning and achievement, decre-
ments in general intelligence often are among the chief manifestations of
early childhood brain injury (Boll & Barth, 1981; Rourke et a1., 1983).
Hynd et a1. (1986) have suggested that the effects of brain damage begin to
take on a more specific character, resembling those seen in adults, when
injuries are sustained after the functional organization of the cerebral
hemispheres has become fairly well established, at about 5 to 7 years of
age. Nonetheless, there is considerable overlap or redundancy between
standard measures of intelligence and current test batteries used for as-
sessing neuropsychological functioning in children (Seidenberg et 01.,

1983; Sweet et 01., 1986; Tramontana et 01., 1984). What, then, defines
neuropsychological assessment and distinguishes it from simply a general
assessment of a child's mental abilities? Is it something intrinsic to the
procedures, themselves, that constitute neuropsychological tests? Or is it
the conceptual perspective that is brought to bear when interpreting per-
formance data on measures of whatever type?
Certainly, many of the standard instruments available for assessing
children's abilities, such as the WISC-R or McCarthy Scales of Children's
Abilities (McCarthy, 1972), or an instrument like the Kaufman Assess-
ment Battery for Children (Kaufman & Kaufman, 1983), which permits a
differentiation of sequential and simultaneous processing abilities, can be
used quite effectively in evaluating a child's neuropsychological func-
tioning (Wilson, 1986). This is especially true when the use of such in-
struments is accompanied by careful qualitative observations of the
child's performance. However, most of these tests were neither developed
from a neuropsychological perspective nor designed to facilitate neuro-
psychological inferences regarding a child's functioning. Procedural mod-
ifications usually must be included to facilitate hypothesis-testing, such
as introducing certain cues or altering the response mode on a task once
the standard administration of the test is completed. There is no question
that this can be an effective means of neuropsychological assessment,
provided that the process is guided by a clear conceptual framework of
developmental brain-behavior relationships. However, the process
would be more efficient if the basic composition of the tests in question
were aligned more closely with relevant aspects of neuropsychological
functioning.
Thus, in contrast to standard cognitive testing, the tests or measures
making up a neuropsychological assessment are selected, designed, or
adapted to facilitate the conceptualization of a child's performance in
terms of known or hypothesized brain-behavior relationships. Although
virtually any measure of a child's abilities may be used in making neuro-
psychological inferences, some measures are better than others in reveal-
ing specific aspects of neuropsychological functioning. It is this feature,
not simply their sensitivity to childhood brain damage, that is a defining
characteristic of neuropsychological tests. Moreover, to be considered as
components of a comprehensive neuropsychological assessment, the tests
must be embedded within an overall assessment strategy that is organized
according to current knowledge of brain-behavior relationships and the
process by which these unfold over the course of development. The latter
feature is particularly critical in child neuropsychological assessment in
in that a given test or measure (e.g., name writing) may tap very different
functions at different points in a child's development.
Various conceptual frameworks have been proposed for organizing
the clinical assessment of brain-behavior relationships in childhood (e.g.,

Rourke et a1., 1986; Wilson, 1986). These are important in that they serve
to anchor the assessment in terms of well-defined neuropsychological
constructs. To date, however, these achieve only a partial incorporation of
a developmental perspective into the assessment process. Applying a de-
velopmental perspective in neuropsychological assessment means more
than simply assuring that the tests and measures are appropriate for the
age range of children being assessed. It also entails more than utilizing
measures that are sensitive to the child's increasing competence over
time. It involves understanding how brain functions develop normally, as
well as under various pathological conditions. A child not only grows
more competent with age but also develops new and more efficient strat-
egies for solving problems. This is a fundamental tenet of developmental
psychology, and it is also reflected in Luria's (1966, 1973) ontogenetic
theory of functional systems. When applied to child neuropsychology,
this perspective requires that a conceptual framework for assessment in-
dicate not only what areas of function should be assessed but how these
should be assessed at different points in development. It requires, quite
frankly, a more complete and dynamic understanding of normal and ab-
normal brain development than the field currently can provide. Much
remains to be learned regarding the neuropsychology of the developing
brain and its implications for clinical assessment (see Chapter 2, this
volume).
Besides the issue of developmental change, there are a host of other
factors that can obscure and complicate the assessment of brain-behavior
relationships in children. These include: (1) problems in specifying the
precise time of onset for various forms of brain pathology, (2) the absence
of a referential baseline of premorbid functioning in cases of early brain
damage, (3) the sometimes blurred distinction between neurodevelop-
mental anomalies and normal variations in the rate and pattern of acquisi-
tion of function, (4) the extent to which deficits sometimes can be delayed
or "silent" until later developmental periods, (5) the absence of a stan-
dard nosology for classifying childhood neurodevelopmental disorders,
(6) the interacting effects of various nonneurological attributes within the
child that may serve to compound or mitigate the effects of brain dysfunc-
tion, and (7) the impact of environmental factors, including the family, in
shaping the child's outcomes. In addition, there are a number of practical
complications that often may occur in assessing the brain-impaired child,
including problems in maintaining behavioral compliance and sustained
attention for prolonged periods of assessment. Also, the younger child
may be unable to provide a reliable subjective report of deficits, and the
circumstances in which these are more or less likely to be problematic.
Reports of parents and teachers often are crucial in providing accounts of
the child's abilities outside of the testing situation. Although some of the
above issues may apply in the evaluation of adults, they are far more
likely to arise as complicating factors in the neuropsychological assess-
ment of children.
Nowhere are the age-specific constraints on assessment more appar-
ent than in the evaluation of children during the first 3 or 4 years of life.
Perhaps this is why, with few exceptions, the bulk of research and prac-
tice in child neuropsychology has dealt almost exclusively with children
5 years of age and older. The relatively limited response repertoire of
infants or toddlers that can be used in assessing their abilities, together
with the degree to which performance is state-dependent, places major
constraints on formal neuropsychological assessment (see Chapter 9, this
volume). This is largely an uncharted territory for neuropsychologists
that, up to this point, mainly has involved the work of developmental
psychologists and pediatric neurologists. At present, there is little re-
semblance between what is done in assessing brain function in infancy or
early childhood (which tends to rely heavily on the evaluation of reflexes,
sensorimotor development, and other developmental milestones) and
what ordinarily would constitute a neuropsychological assessment in
later years. The extent to which there is continuity in the neuropsycholog-
ical meaning of the assessment data obtained over these developmental
periods is an open question.
Finally, consideration must be given to the unique challenges that
arise in the assessment of children with special handicapping conditions
such as sensory loss, physical deformities, motor disabilities, and lan-
guage impairment. The nature of the handicap as well as its impact on the
child's general functioning must be assessed, but care must be taken to
devise an appropriate assessment strategy that will yield a balanced pic-
ture of the child's overall abilities. It would be inappropriate to utilize
tasks that unduly tax the child's specific handicap, and then go on and
use these results to draw conclusions about the child's more general func-
tioning. For example, the assessment of nonverbal abilities in a language-
impaired child would be biased if this were based on tasks that depended
heavily on the comprehension of spoken instructions. Conducting a com-
prehensive assessment that effectively works around the child's handicap
and maintains his or her motivation to perform requires a great deal of
skill, ingenuity, and patience on the part of the examiner. It also requires a
knowledge of any unique neurodevelopmental features associated with
specific handicaps. For example, a blind child is likely to show delays in
sensorimotor development, and may use different cognitive strategies in
performing certain tasks than a sighted child. Although inconclusive,
there also is evidence to suggest that children with a specific sensory loss
in early life may develop an atypical pattern of cerebral organization of
function (e.g., Gibson & Bryden, 1984). Differential neuropsychological
features may be associated with other types of handicap as well. This is an
important area of application in child neuropsychology, one that ulti-

mately may prove to require a highly specialized background of knowl-
edge and skill.
SUMMARY
This chapter has provided a general overview of child neuropsycho-

logical assessment. We began by tracing the major historical trends in
assessment that have characterized the field, culminating with the present
emphasis on using neuropsychological methods prescriptively in identi-
fying factors that could maximize the brain-impaired child's capacity to
deal effectively with important tasks of daily life. Next, we considered the
current approaches to neuropsychological assessment, outlining their re-
spective strengths and weaknesses, their similarities and differences, as
well as possible ways in which the various approaches might be com-
bined in an efficient and complementary fashion. We then discussed
important applications of neuropsychological assessment in treatment
planning and in monitoring (re)acquisition of function, and highlighted
its potential contributions in the assessment of selected clinical popula-
tions, including children with neurological disorders, systemic illness,
psychiatric disorders, and learning disabilities. Last, we addressed some
key definitional issues that pertain to neuropsychological assessment, the
need for conceptual guidance from a developmental framework of brain-
behavior relationships, and some of the common complications that can
arise in the neuropsychological assessment of children. Although our
coverage in this chapter was selective, and undoubtedly a number of
important topics were omitted (e.g., mental retardation), we believe that it
nonetheless provides a reasonably thorough overview of the current sta-
tus of the field. With this, the stage now should be set for selected issues
and topics in child neuropsychological assessment to be pursued at great-
er depth in the subsequent chapters of this volume.
REFERENCES
Adamovich, B. B., Henderson, J. A., & Auerbach, S. (1985). Cognitive rehabilitation of closed
head injured patients. San Diego, CA: College-Hill Press.
Almli, C. R., & Finger, S. (1984). Early brain damage. Orlando, FL: Academic Press.
Aylward, G. P., Gustafson, N., Verhulst, S. J., & Colliver, J. A. (1987). Consistency in the
diagnosis of cognitive, motor, and neurologic function over the first three years. Journal
of Pediatric Psychology, 12, 77-98.
Bender, 1. (1938). A visual motor gestalt test and its clinical use (Research Monograph No.
3). New York: American Orthopsychiatric Association.
Benton, A. 1. (1963). The Revised Visual Retention Test: Clinical and experimental applica-
tions. New York: Psychological Corporation.
Benton, A. 1., Hamsher, K., Varney, N. R., & Spreen, O. (1983). Contributions to neuropsy-
chological assessment: A clinical manual. New York: Oxford University Press.
Berg, R A., Bolter, J. F., Ch'ien, 1. T., Williams, S. J., Lancaster, W., & Cummins, J. (1984).
Comparative diagnostic accuracy of the Halstead-Reitan and the Luria-Nebraska Neuro-
psychological Adult and Children's Batteries. Clinical Neuropsychology. 6(3), 200-204.
Boll, T. J. (1974). Behavioral correlates of cerebral damage in children aged 9-14. In R M.
Reitan & 1. A. Davison (Eds.), Clinical neuropsychology: Current status and applica-
tions (pp. 91-120). New York: Wiley.
Boll, T. J. (1981). The Halstead-Reitan Neuropsychology Battery. In S. B. Filskov & T. J. Boll
(Eds.), Handbook of clinical neuropsychology (pp. 577-607). New York: Wiley.
Boll, T. J., & Barth, J. T. (1981). Neuropsychology of brain damage in children. In S. B.
Filskov & T. J. Boll (Eds.)' Handbook of clinical neuropsychology (pp. 418-452). New
York: Wiley.
Boll, T. J., & Reitan, R M. (1972). Comparative ability interrelationships in normal and brain-
damaged children. Journal of Clinical Psychology, 28, 152-156.
Bolter, J. F. (1984). Neuropsychological impairment and behavioral dysfunction in children
with chronic epilepsy. Unpublished doctoral dissertation, Memphis State University.
Boyd, T. A., Tramontana, M. G., & Hooper, S. R (1986). Cross-validation of a psychometric
system for screening neuropsychological abnormality in older children. Archives of
Clinical Neuropsychology, 1, 387-391.
Breslau, N., & Marshall, 1. A. (1985). Psychological disturbance in children with physical
disabilities: Continuity and change in a 5-year follow-up. Journal of Abnormal Child
Psychology, 13, 199-216.
Brink, J., Garrett, A., Hale, W., Woo Sam, J., & Nickel, V. (1970). Recovery of motor and
intellectual function in children sustaining severe head injuries. Developmental Medi-
cine and Child Neurology, 12, 565-571.
Brown, G., Chadwick, 0., Shaffer, D., Rutter, M., & Traub, M. (1981). A prospective study of
children with head injuries: III. Psychiatric sequelae. Psychological Medicine, 11,63-
78.
Chadwick, 0., & Rutter, M. (1983). Neuropsychological assessment. In M. Rutter (Ed.), Devel-
opmental neuropsychiatry (pp. 181-212). New York: Guilford Press.
Chadwick, 0., Rutter, M., Brown, G., Shaffer, D., & Traub, M. (1981). A prospective study of
children with head injuries: II. Cognitive sequelae. Psychological Medicine, 11, 49-
61.
Chelune, G. J., & Edwards, P. (1981). Early brain lesions: Ontogenetic-environmental consid-
erations. Journal of Consulting and Clinical Psychology, 49, 777-790.
Chelune, G., Ferguson, W., Koon, R, & Dickey, T. (1986). Frontal lobe dysinhibition in
attention deficit disorder. Child Psychiatry and Human Development, 16, 221-234.
Christensen, A. 1. (1975). Luria's neuropsychological investigation. New York: Spectrum.
Craft, A., Shaw, D., & Cartlidge, N. (1972). Head injuries in children. British Medical Journal,
4, 200-203.
Crockett, D., Klonoff, H., & Bjerring, J. (1969). Factor analysis of neuropsychological tests.
Perceptual and Motor Skills, 29, 791-802.
Cull, c., & Wyke, M. A. (1984). Memory function of children with spina bifida and shunted
hydrocephalus. Developmental Medicine and Child Neurology, 26, 177-183.
Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1986). The California Verbal Learning
Test: Children's Version. New York: Psychological Corporation.
Dennis, M. (1985a). Intelligence after early brain injury: I. Predicting IQ scores from medical
variables. Journal of Clinical and Experimental Neuropsychology, 7, 526-554.
Dennis, M. (1985b). Intelligence after early brain injury: II. IQ scores of subjects classified on
the basis of medical history variables. Journal of Clinical and Experimental Neuropsy-
chology, 7, 555-576.
Dreisbach, M., Ballard, M., Russo, D. c., & Schain, R J. (1982). Educational intervention for
34 MICHAEL G. TRAMONTANA and STEPHEN R HOOPER
children with epilepsy: A challenge for collaborative service delivery. Journal of Special
Education. 16,111-121.
Elbert, J. c., Culbertson, J. 1., Gerrity, K. M., Guthrie, 1. J., & Bayles, R (1985, February).
Neuropsychological and electrophysiologic follow-up of children surviving acute lym-
phocytic leukemia. Paper presented at the annual meeting of the International Neuro-
psychological Society, San Diego.
Emhart, C. B., Graham, F. K., Eichman, P. L., Marshall, J. M., & Thurston, D. (1963). Brain
injury in the preschool child: Some developmental considerations. II. Comparison of
brain-injured and normal children. Psychological Monographs, 77(Whole No. 574), 17-
33.
Fein, D., Waterhouse, 1., Lucci, D., & Snyder, D. (1985). Cognitive subtypes in developmen-
tally disabled children: A pilot study. Journal of Autism and Developmental Disorders,
15, 77-95.
Fennell, E. B., Eisenstadt, T., Bodiford, C., Rediess, S., de Bijl, M., & Mickle, J. (1987,
February). The assessment of neuropsychological dysfunction in children shunted for
hydrocephalus. Paper presented at the Fifteenth Annual Meeting of the International
Neuropsychological Society, Washington, D.C.
Filskov, S. B., & Boll, T. J. (Eds.). (1981). Handbook of clinical neuropsychology. New York:
Wiley.
Filskov, S. B., & Goldstein, S. G. (1974). Diagnostic validity of the Halstead-Reitan Neuropsy-
chological Battery. Journal of Consulting and Clinical Psychology, 42, 383-388.
Gaddes, W. H. (1985). Learning disabilities and brain function: A neuropsychological ap-
proach (2nd ed.). New York: Springer-Verlag.
Geary, D. c., & Gilger, J. W. (1984). The Luria-Nebraska Neuropsychological Battery-Chil-
dren's Revision: Comparison of learning disabled and normal children matched on full
scale IQ. Perceptual and Motor Skills, 58, 115-118.
Geary, D. C., Jennings, S. M., Schultz, D. D., & Alper, T. G. (1984). The diagnostic accuracy of
the Luria-Nebraska Neuropsychological Battery-Children's Revision for 9 to 12 year old
learning disabled children. School Psychology Review, 13, 375-380.
George, S. 1., Aur, R J. A., Mauer, A. M., & Simone, J. V. (1979). A reappraisal of the results
of stopping therapy in childhood leukemia. New England Journal of Medicine, 300,
269-273.
Gibson, C. J., & Bryden, M. P. (1984). Cerebral laterality in deaf and hearing children. Brain
and Language, 23, 1-12.
Goff, J. R (1982). Memory deficits and distractibility in survivors of childhood leukemia.
Paper presented at the annual meeting of the American Psychological Association.
Golden, C. J. (1981). The Luria-Nebraska Children's Battery: Theory and formulation. In G.
W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child:
Issues and perspectives (pp. 277-302). New York: Grune and Stratton.
Golden, C. J. (1987). Manual for the Luria-Nebraska Neuropsychological Battery-Children's
Revision. Los Angeles, CA: Western Psychological Services.
Golden, C. J., Hammeke, T. A., & Purisch, A. D. (1980). Manual for the Luria-Nebraska
Neuropsychological Battery. Los Angeles: Western Psychological Services.
Goldman, R, Fristoe, M., & Woodcock, R W. (1974). Manual for the Goldman-Fristoe-
Woodcock Auditory Skills Test Battery. Circle Pines, MN: American Guidance Service.
Graham, F. K., Emhart, C. B., Craft, M., & Berman, P. W. (1963). Brain injury in the preschool
child: Some developmental considerations: I: Performance of normal children. Psycho-
logical Monographs, 77(Whole No. 573), 1-16.
Graham, F. K., & Kendall, B. S. (1960). Memory-for-Designs-Test: Revised general manual.
Perceptual and Motor Skills Monograph (Supplement, No.2-VII), 11, 147-188.
Gustavson, J. 1., Golden, C. J., Wilkening, G. N., Hermann, B. P., Plaisted, J. R, MacInnes, W.
D., & Leark, R A. (1984). The Luria-Nebraska Neuropsychological Battery-Children's
Revision: Validation with brain-damaged and normal children. Journal of Psychoeduca-
tional Assessment, 2, 199-208.
Gustavson, J. 1., Wilkening, G. N., Hermann, B. P., & Plaisted, J. R. (1982). Factor analysis of
the children's revision of the Luria-Nebraska Neuropsychological Battery. Unpublished
manuscript.
Hagen, C. (1981). Language disorders secondary to closed head injury: Diagnosis and treat-
ment. Topics in Language Disorders, 1, 73-87.
Hammill, D. D., Leigh, J. E., McNutt, G., & Larsen, S. C. (1981). A new definition of learning
disabilities. Learning Disability Quarterly, 4, 336-342.
Heaton, R K. (1981). Manual for the Wisconsin Card Sorting Test. Odessa, FL: Psychological
Assessment Resources.
Herbert, M. (1964). The concept and testing of brain damage in children: A review. Journal of
Child Psychology and Psychiatry,S, 197-216.
Holmes, C. S., & Richman, 1. C. (1985). Cognitive profiles of children with insulin-depen-
dent diabetes. Journal of Developmental and Behavioral Pediatrics, 6, 323-326.
Hooper, S. R, & Tramontana, M. G. (1987). The relationship between neuropsychological
variables and academic achievement in child psychiatric patients. Paper presented at
the 95th Annual Convention of the American Psychological Association, New York.
Hurley, A. D., Laatsch, 1. K., & Dorman, C. (1983). Comparison of spina bifida, hydro-
cephalic patients, and matched controls on neuropsychological tests. Zeitschrift fUr
Kinderchir, 17, 65-70.
Hynd, G. W., & Obrzut, J. E. (Eds.). (1981). Neuropsychological assessment and the school-
age child: Issues and procedures. New York: Grune and Stratton.
Hynd, G. W., Snow, J., & Becker, M. G. (1986). Neuropsychological assessment in clinical
child psychology. In B. Lahey & A. Kazdin (Eds.). Advances in clinical child psychology
(Vol. 9). New York: Plenum.
Hynd, G. W., & Willis, W. G. (1988). Pediatric neuropsychology. New York: Grune and
Stratton.
Incagnoli, T., & Newman, B. (1985). Cognitive and behavioral interventions. International
Journal of Clinical Neuropsychology, 7, 173-182.
Ivan, L. P. (Ed.). (1984). Pediatric neuropsychology. St. Louis: Warren H. Green.
Karras, D., Newlin, D. B., Franzen, M. D., Golden, C. J., Wilkening, G. N., Rothermel, RD., &
Tramontana, M. G. (1987). Development of factor scales for the Luria-Nebraska Neuro-
psychological Battery-Children's Revision. Journal of Clinical Child Psychology, 16,
19-28.
Kaufman, A. S., & Kaufman, N. 1. (1983). Manual for the Kaufman Assessment Battery for
Children. Circle Pines, MN: American Guidance Service.
Kay, G. G., Becker, B., Bleiberg, J., & Long, C. J. (1986, August). Cognitive rehabilitation
software: Therapy or adjunct? A call for standards. Paper presented at the Annual
Meeting of the American Psychological Association, Washington, D.C.
Klesges, R C. (1983). The relationship between neuropsychological, cognitive, and behav-
ioral assessments of brain functioning in children. Clinical Neuropsychology, 1, 28-32.
Klonoff, H., Low, M. D., & Clark, C. (1977). Head injuries in children with a prospective 5-
year follow-up. Journal of Neurology, Neurosurgery and Psychiatry, 40, 1211-1219.
Klonoff, H., Robinson, G. C., & Thompson, G. (1969). Acute and chronic brain syndromes in
children. Developmental Medicine in Child Neurology, 11, 198-213.
Knights, R M. (1966). Normative data on tests for evaluating brain damage in children from 5
to 14 years of age (Research Bulletin No. 20). London, Ontario: University of Western
Ontario.
Kolb, B., & Wishaw, I. Q. (1980). Fundamentals of human neuropsychology. San Francisco:
W. H. Freeman.
Koppitz, E. M. (1964). The Bender Gestalt Test for Young Children. New York: Grune and
Stratton.
Levin, H. S., & Eisenberg, H. M. (1979). Neuropsychological impairment after closed head
injury in children and adolescents. Journal of Pediatric Psychology, 4, 389-402.
Luria, A. R. (1966). Higher cortical functions in man. New York: Basic Books.
Luria, A. R. (1973). The working brain. New York: Basic Books.

McCarthy, D. (1972). Manual for the McCarthy Scales of Children's Abilities. New York:
Psychological Corporation.
Meighan, S. S., Queener, L., & Weitman, M. (1976). Prevalence of epilepsy in children of
Multnomah County, Oregon. Epilepsia, 17, 245-256.
Milberg, W. P., Hebben, N., & Kaplan, E. (1986). The Boston process approach to neuropsy-
chological assessment. In 1. Grant & K. M. Adams (Eds.). Neuropsychological assess-
ment of neuropsychiatric disorders (pp. 65-86). New York: Oxford University Press.
Nolan, D. R., Hammeke, T. A., & Barkley, R. A. (1983). A comparison of the patterns of the
neuropsychological performance in two groups of learning disabled children. Journal of
Clinical Child Psychology, 12, 13-21.
Obrzut, J. E., & Hynd, G. W. (1986a). Child Neuropsychology: Theory and Research. (Vol. 1).
New York: Academic Press.
Obrzut, J. E., & Hynd, G. W. (1986b). Child Neuropsychology: Clinical Practice. (Vol. 2). New
York: Academic Press.
Passier, M., Isaac, W., & Hynd, G. W. (1986). Neuropsychological development of behavior
attributed to frontal lobe functioning in children. Developmental Neuropsychology, 1,
349-370.
Reed, H. B. C., Reitan, R. M., & Klove, H. (1965). Influence of cerebral lesions on psychologi-
cal test performance of older children. Journal of Consulting Psychology, 29, 247-
251.
Reitan, R. M. (1974). Psychological effects of cerebral lesions in children of early school age.
In R. M. Reitan & L. A. Davison (Eds.). Clinical neuropsychology: Current status and
applications (pp. 53-90). New York: Wiley.
Reitan, R. M., & Davison, L. A. (Eds.). (1974). Clinical neuropsychology: Current status and
applications. New York: Wiley.
Rourke, B. P. (1982). Central processing deficiencies in children: Toward a developmental
neuropsychological model. Journal of Clinical Neuropsychology, 4, 1-18.
Rourke, B. P. (1985). Neuropsychology of learning disabilities: Essentials of subtype analy-
sis. New York: Guilford Press.
Rourke, B. P., Bakker, D. J., Fisk, J. L., & Strang, J. D. (1983). Child neuropsychology: Intro-
duction to theory, research and practice. New York: Guilford Press.
Rourke, B. P., & Finlayson, M. A. J. (1978). Neuropsychological significance of variations in
patterns of academic performance: Verbal and visual-spatial abilities. Journal of Abnor-
mal Child Psychology, 6, 121-133.
Rourke, B. P., Fisk, J. L., & Strang, J. D. (1986). Neuropsychological assessment of children: A
treatment-oriented approach. New York: Guilford Press.
Rovet, J., Gore, M., & Ehrlich, R. (1983). Intellectual and behavioral deficits associated with
early onset diabetes mellitus. Diabetes, 32 (Supplement 1), 17.
Rutter, M. (1977). Brain damage syndromes in childhood: Concepts and findings. Journal of
Child Psychology and Psychiatry, 18, 1-21.
Rutter, M. (Ed.). (1983). Developmental neuropsychiatry. New York: Guilford Press.
Rutter M., Graham, P., & Yule, W. (1970). A neuropsychiatric study in childhood (Clinics in
Developmental Medicine Nos. 35-36). London: Spastics International Medical Publica-
tions/Heinemann Medical Books.
Ryan, C., Vega, A., & Drash, A. (1985). Cognitive deficits in adolescents who developed
diabetes early in life. Pediatrics, 75, 921-927.
Ryan, c., Vega, A., Longstreet, C., & Drash, A. (1984). Neuropsychological changes in adoles-
cents with insulin-dependent diabetes. Journal of Consulting and Clinical Psychology,
52, 335-342.
Satz, P., & Fletcher, J. M. (1982). Florida Kindergarten Screening Battery. Odessa, FL: Psy-
chological Assessment Resources.
Sawicki, R. F., Leark, R., Golden, C. J., & Karras, D. (1984). The development of the pathog-
nomonic, left sensorimotor, and right sensorimotor scales for the Luria-Nebraska Neuro-
psychological Battery-Children's Revision. Journal of Clinical Child Psychology, 13(2),
165-169.
Seidel. U. P., Chadwick, 0., & Rutter, M. (1975). Psychological disorders in crippled chil-
dren: A comparative study of children with and without brain damage. Developmental
Medicine and Child Neurology, 17, 563-573.
Seidenberg, M., Giordani, B., Berent, S., & Boll, T. J. (1983). IQ level and performance on the
Halstead-Reitan Neuropsychological Test Battery for Older Children. Journal of Con-
sulting and Clinical Psychology, 51, 406-413.
Selz, M., & Reitan, R. M. (1979). Rules for neuropsychological diagnosis: Classification of
brain function in older children. Journal of Consulting and Clinical Psychology, 47(2).
258-264.
Shaffer, D., Schonfeld, 1., O'Connor, P. A., Stokman, C., Trautman, P., Shafer, S., & Ng, S.
(1985). Neurological soft signs. Archives of General Psychiatry, 42, 342-351.
Smith, A. (1975). Neuropsychological testing in neurological disorders. In W. J. Friedlander
(Ed.). Advances in neurology (Vol. 7). New York: Raven Press.
Snow, J. H., & Hynd, G. W. (1985). Factor structure of the Luria-Nebraska Neuropsychologi-
cal Battery-Children's Revision. Journal of School Psychology, 23, 271-275.
Snow, J. H., Hynd, G. W., & Hartlage, 1. C. (1984). Difference between mildly and more
severely learning-disabled children on the Luria-Nebraska Neuropsychological Battery-
Children's Revision. Journal of Psychoeducational Assessment, 2, 23-28.
Soare, P. 1., & Raimondi, A. J. (1977). Intellectual and perceptual-motor characteristics of
treated myelomeningocele children. American Journal of Diseases of Children, 131,
199-204.
Spreen, O. J., & Gaddes, W. H. (1969). Development norms for 15 neuropsychological tests
age 6-15. Cortex, 5, 171-191.
Spreen, O. J., Tupper, D., Risser, A., Tuokko, H., & Edgell, D. (1984). Human developmental
neuropsychology. New York: Oxford University Press.
Sweet, J. J., Carr, M. A., Rossini. E., & Kaspar, C. (1986). Relationship between the Luria-
Nebraska Neuropsychological Battery-Children's Revision and the WISC-R: Further ex-
amination using Kaufman's factors. International Journal of Clinical Neuropsychology,
8, 177-180.
Teasdale, G., & Jennett, B. (1974). Assessment of coma and impaired consciousness. Lancet,
11,81-84.
Tramontana, M. G. (1983). Neuropsychological evaluation of children and adolescents with
psychopathological disorders. In C. J. Golden & P. J. Vicente (Eds.). Foundations of
clinical neuropsychology (pp. 309-340). New York: Plenum.
Tramontana, M. G., & Boyd, T. A. (1986). Psychometric screening of neuropsychological
abnormality in older children. International Journal of Clinical Neuropsychology, 8,
53-59.
Tramontana, M. G., Hooper, S. R., & Nardolillo, E. M. (in press). Behavioral manifestations of
neuropsychological impairment in children with psychiatric disorders. Archives of
Clinical Neuropsychology
Tramontana, M. G., & Hooper, S. R. (1987). Discriminating the presence and pattern of
neuropsychological impairment in child psychiatric disorders. International Journal of
Tramontana, M. G., & Hooper, S. R. (in press). Neuropsychology of child psychopathology.
In C. R. Reynolds (Ed.). Handbook of child clinical neuropsychology. New York:
Plenum.
Tramontana, M. G., Hooper, S. R., Curley, A. D., & Nardolillo, E. M. (1988, January). Determi-
nants of academic achievement in children with psychiatric disorders. Paper presented
at the Annual National Conference of the International Neuropsychological Society,
New Orleans.
Tramontana, M. G., Klee, S. H., & Boyd, T. A. (1984). WISC-R interrelationships with the
Halstead-Reitan and Children's Luria Neuropsychological Batteries. International Jour-
nal of Clinical Neuropsychology, 6, 1-8.
Tramontana, M. G., & Sherrets, S. D. (1985). Brain impairment in child psychiatric disorders:
Correspondencies between neuropsychological and CT scan results. Journal of the
American Academy of Child Psychiatry, 24, 590-596.
Tramontana, M. G., Sherrets, S. D., & Golden, C. J. (1980). Brain dysfunction in youngsters
with psychiatric disorders: Application of Selz-Reitan rules for neuropsychological di-
agnosis. Clinical Neuropsychology, 2, 118-123.
Tramontana, M. G., Sherrets, S. D., & Wolf, B. A. (1983). Comparability of the Luria-Nebraska
and Halstead-Reitan Neuropsychological Batteries for older children. Clinical Neuro-
psychology,S, 186-190.
Waber, D., Sollee, N., Wills, K., & Fischer, R (1986, February). Neuropsychological effects of
two levels of cranial radiation in children with acute lymphoblastic leukemia (ALL).
Paper presented at the Fourteenth Annual Meeting of the International Neuropsycholog-
ical Society, Denver, Colorado.
Wender, P. (1971). Minimal brain dysfunction in children. New York: Wiley.
Wilkening, G. N., Golden, C. J., Macinnes, W. D., Plaisted, J. R, & Hermann, B. P. (1981,
August). The Luria-Nebraska Neuropsychological Battery-Children's Revision: A pre-
liminary report. Paper presented at the meeting of the American Psychological Associa-
tion, Los Angeles.
Wilson, B. C. (1986). An approach to the neuropsychological assessment of the preschool
child with developmental deficits. In S. B. Filskov & T. J. Boll (Eds.), Handbook of
clinical neuropsychology (Vol. 2). New York: Wiley.
Wilson, H., & Staton, R D. (1984). Neuropsychological changes in children associated with
tricyclic antidepressant therapy. International Journal of Neuroscience, 24, 307-312.
Wolf, B. A., & Tramontana, M. G. (1982). Aphasia Screening Test interrelationships with
complete Halstead-Reitan test results for older children. Clinical Neuropsychology, 4,
179-186.
Yule, W. (1980). Educational achievement. In B. M. Kulig, M. Meinhardi, & G. Stores (Eds.),
Epilepsy and behavior. Lisse, The Netherlands: Swets & Zeitlinger.
II
General Issues in Child

Neuropsychological Assessment
2
Neuropsychology of the Developing

Brain
Implications for Neuropsychological
Assessment
ANTHONY H. RISSER and DOROTHY EDGELL
INTRODUCTION
Childhood neuropsychological assessment contributes interpretative data

on the relation between overt behavior and the functional systems of the
child's brain. These data are obtained in the expectation that they will be
useful in interpretating the significance of the child's manifest problems.
The assessment provides a comprehensive analysis of test-delimited
neuropsychological systems, such as language comprehension or visual
short-term memory, and of the integrated functioning of these systems.
Neuropsychological interpretation occurs in a diagnostic milieu that also
might include clinical neurological, neuroimaging, electrophysiological,
psychological, and educational data.
As with any psychological assessment, neuropsychological assess-
ment is performed with an awareness of other important factors, such as
the child's history, family environment, and expected normative psycho-
social and biological development. Any mediating factor may influence
behaviors manifested on assessment. There is a defining neuropsycholog-
ical dimension of this interpretative context, however; namely, obtained
data are viewed in light of knowledge about the central nervous system
ANTHONY H. RISSER Department of Neurology, Mount Sinai Medical Center, University

of Wisconsin Medical School, Milwaukee, Wisconsin. DOROTHY EDGELL Depart-
ment of Psychology, Jack Ledger Child and Adolescent Psychiatric Unit, Arbutus Society for
Children, Victoria, British Columbia, Canada.
41
42 ANTHONY H. RISSER and DOROTHY EDGELL
(eNS) substrate, which is presumed to account to some extent for the

obtained performances. The developmental neuropsychologist views
data-both behavioral and neurological-that are transient, albeit charac-
teristic features of evolving systems at the time of assessment. A three-part
focus is created by this context for the analysis of neuropsychological
findings: (1) as possible consequences of past occurrences, (2) as func-
tions representing current status, and (3) as influences upon further
development.
The major purposes for obtaining neuropsychological data include
(1) exploring the cognitive and behavioral consequences of a known neu-
rological insult to determine the extent of its impact, (2) exploring possi-
ble neurological processes that may underlie abnormal cognitive or
behavioral manifestations to address possible cause, and (3) creating treat-
ment or management plans. In all instances, the assessment requires
knowledge of the developing brain as it relates to a performance charac-
teristic or to a syndrome (i.e., a set of correlated characteristics). Although
appropriate childhood neuropsychological assessments can be accom-
plished without special reference to underlying neural function (Taylor,
Fletcher, & Satz, 1984), an elementary premise of neuropsychological
practice is that relations between brain and behavior do exist, although
they are not likely to be isomorphic in nature.
A formal neuropsychology of the developing brain-that is, one that
can integrate features of neurological and psychological development-is
a contemporary goal rather than an established feat (Benton, 1982). How-
ever, one need only compare compilations of contemporary empirical
findings, such as this volume, with those of only a decade ago, such as
Witelson's (1977) germinal evaluation of issues surrounding the develop-
ment of hemispheric specialization, to realize how the scope and depth of
our basic knowledge has changed. This advance has been particularly true
for brain-behavior phenomena that encompass the first 2 years of life. To
provide an outline that may be of use as a guide when examining the
specific neurobehavioral topics of later chapters, this chapter provides a
basic overview of the normal and pathological developing brain, with an
emphasis on development from the prenatal period through the first years
of life.
PRINCIPLES OF NEURAL DEVELOPMENT
Two dimensions define brain development. First, as individual cells

are created in accordance with genetic instruction, they develop along
predetermined parameters of growth. Second, aspects of this maturation
show degrees of flexibility or plasticity in adaptive response to the en-
NEUROPSYCHOLOGY OF THE DEVELOPING BRAIN 43
vironment (broadly defined) as cell populations evolve into committed

functional systems. Growth in the developing brain proceeds along dis-
tinctive temporal-spatial dimensions-cell populations differing in their
ultimate destinies mature through developmental stages at different times
and in different regions of the brain. These stages are proliferation, migra-
tion, differentiation, and myelination.
Each stage will be introduced below; in the interest of brevity, pro-
liferation and migration will be discussed together. Gross normative sta-
tus at birth will be discussed in the section on differentiation. The func-
tional organization of brain-behavior relations also will be introduced.
The distinctions between stages are somewhat arbitrary, and others have
employed different criteria to distinguish them. Kandel and Schwartz
(1981), in their edited text on neural science, for example, separate as-
pects of neuronal development from aspects of trans neuronal synaptic
development. The reader is referred to works of general embryonic devel-
opment (e.g., Thomas, 1968), developmental neurobiology (e.g., Jacobson,
1978; Lund, 1978; Williams & Caviness, 1984), and child neuropsychol-
ogy (e.g., Rourke, Bakker, Fisk, & Strang, 1983; Spreen, Tupper, Risser,
Tuokko, & Edgell, 1984) for discussions of neural development from sev-
eral different perspectives.
Proliferation and Migration

The creation of embryonic cells by cell division results in the ap-
pearance of the neuroectoderm (a thickened, elongated central region of the
embryonic disk) in the third gestational week. Neuroectodermal demarca-
tion of brain-cell precursors from other embryonic cells evolves into the
distinctive neural tube during the fourth week. Accelerated brain-cell
creation occurs shortly thereafter, during the period of embryo-wide organ
formation in the fourth and fifth weeks. The internal wall of the neural
tube, which surrounds a fluid-filled cavity, is the birth site of neuroblasts
(destined to become neurons) and glial-cell precursors. Where along this
elongated wall, and when during this time period, a cell is created plays a
critical role in the cell's ultimate destiny in the mature CNS. Different cell
populations reside in specific regions of the expanding neural tube. The
tube's anterior end bulges and bends at particular sites of extensive cell
creation, such as the forebrain, midbrain, and hindbrain vesicles-precur-
sor areas for cell populations destined to become the brain. Generally,
neurons are created before glial cells, neurons found in the deeper layers of
the cortex are created prior to those destined for superficial levels, and
cerebellar neurons are created late in proliferation. Golgi class I neurons,
which have large axons and form major pathways, generally proliferate
before Golgi class II neurons (microneurons), which form connections
within brain areas. Brain-cell proliferation slows by the end of the first
trimester of pregnancy; most neurons have been created by the beginning of
the final trimester.
Once a population of neuroblasts has stopped multiplying, its con-
stituents migrate as true neurons outward from their birth sites. Young
neurons migrate in sheets (laminae) of similar cells created at similar
times toward specific zones of the outer layer of the neural tube, which
eventually will become the multilayered cortical "gray matter" and the
various subcortical nuclei. Axons that will enter or leave this outer layer
on route to their target synaptic zones form an intermediate layer destined
to become "white matter." Rakic (1981) has provided a detailed descrip-
tion of migratory mechanisms and their roles in configuring the cortex.
The goal of migration is to emplace neurons at their mature functional
sites, creating the gridwork for subsequent cellular differentiation and
synapse formation.
Characteristic physical changes in the fetal brain during this time
involve its regional distinctiveness: the rapid growth of the cerebral hemi-
spheres, the initial appearance of sulcal and gyral landmarks, and cere-
bellar development. As early as the sixth gestational week, features such
as the cortex and the internal capsule are evident. By the 23rd week, major
sulcal and gyrallandmarks on the cortical surface are evident (Chi, Dool-
ing, & Gilles, 1977). Figure 1 presents sulcal development in relation to
brain weight over the course of prenatal development.
Differentiation
Differentiation is a multifactorial growth period during which maturation
of the structure of existing cells is accomplished. Differentiation permits
cells to function as committed constituents of specific neural systems.
The major components of differentiation include (1) formation of the den-
dritic field and distal axonal growth, (2) maturation of cell-body struc-
tures that are required for proper metabolic functioning, (3) creation and
maturation of synaptic architecture and the establishment of functional
transmission links, (4) maturation of enzyme and neurotransmitter syn-
thesis and storage, and (5) selective retrogressive processes (e.g., neuronal
death) to streamline the neural substrate. Differentiation of neuronal pop-
ulations begins in earnest once they have migrated to their mature place-
ment sites. Differentiation is a characteristic feature of fetal brain develop-
ment, but it also continues beyond birth. Some elements of differentiation
become life-span features of neurological maintenance. For example, syn-
aptic differentiation is extensive during fetal development when synaptic
turnover (i.e., creation and elimination) is great, but it also may continue
as a modulating mechanism in response to change throughout the life-
span (Cotman & Nieto-Sampedro, 1982, 1984).
400
360
320
280
(i)
E
~ 240
!!J
~
J:
S? 200
w
~
Z
160
a:
CD
120
80
40
0
110 150 190 270
BIRTH
GESTATIONAL AGE (days)
FIGURE 1. The relation between brain weight and the development of cerebral sulci. (Adapt-
ed from Lemire, Loeser, Leech, & Alvord, 1975,)
The destiny of a neuron population to become, for example, the sub-

strate for brainstem nuclei, primary sensory or motor cortex, or the re-
gions of association cortex, unfolds genetically during fetal proliferation.
Subsequent neural growth is the result of differentiation, representing
vast increases in the size and complexity of existing cells. It is during
differentiation that neurons become active as committed members of spe-
cialized functional systems, as relevant synaptic connections between
neurons become established and they begin to communicate. Activity
may remain immature or poorly integrated for some time. This is particu-
larly true for neurons that have axons yet to myelinate and for systems
that continue to evolve in infancy and childhood prior to their ultimate
commitment to subserve particular functions. Aspects of differentiation,
such as the growth and pruning of synaptic spines on dendrites (receptive
postsynaptic regions), appear to have a strong environmental component,
occurring in response to the experience of the organism in its postnatal

environment (e.g., Lund, 1978).
Until neuronal populations are committed to subserve certain func-
tions, they may display some degree of plasticity to subserve other func-
tions if required to do so. The mechanisms of brain-behavior plasticity
and commitment remain to be established and are of key research con-
cern. Basic neuroscientific models of eNS plasticity exist; the most suc-
cessful has been the primate model of dorsolateral prefrontal cortical
development (e.g., Goldman, 1979; Goldman & Galkin, 1978; Goldman-
Rakic, Isserhoff, Schwartz, & Bugbee, 1983). The dorsolateral prefrontal
cortex subserves specific higher cognitive and integrative abilities in pri-
mates. These abilities require time for adult performance levels to be
attained and are subserved by subcortical regions prior to cortical com-
mitment. Lesioning the dorsolateral prefrontal cortex in the adult monkey
results in a characteristic behavioral deficit on delayed-response tasks
that require the animal to remember the location of a salient stimulus,
such as food, over brief time intervals (Jacobsen, 1936). Lesioning this
region during either fetal development or postnatally until the first 2
months of life, however, does not result in a performance deficit when
lesioned animals are subsequently tested as adults. This model's utility in
understanding plasticity in the human brain with regard to behavioral
implications derives from its analogous fit to higher cortical development
in its relative timing, early-life experiential component, and requisite
adult cortical commitment.
Once neurons have ceased proliferating, they become irreplaceable.
As neurons continue to differentiate, their structure and their functional
features also become unique. Although other neuronal populations may
supplement or even assume the responsibilities of other neurons if re-
quired to do so (simple descriptions of proposed mechanisms have been
presented by Gazzaniga, Stein, & Volpe, 1979), destroyed neurons cannot
be replaced directly by a newly created population. The number of neu-
rons created during proliferation is in excess of what is needed in the
normal brain. This provides the evolving system with a degree of initial
redundancy. During differentiation, this redundancy dissipates. A com-
petition among neurons for limited aspects of their synaptic target zones
results in a "weeding out" of those neurons that lose the competition
either by not establishing synapses at the proper time or by somehow
synapsing in improper fashion (Cowan, Fawcett, O'Leary, & Stanfield,
1984; Purves & Lichtman, 1980).
Although differentiation is the most ubiquitous of all growth stages,
peak periods of differentiation occur at different times in different brain
regions. Subcortical structures that maintain the vegetative state of the
organism and cortical areas composed predominantly of Golgi class I
neurons, such as sensorimotor cortex, differentiate early in the second
trimester. Other areas, such as the cerebellum and the vast regions of
microneuron-composed association cortex, continue to differentiate well
beyond birth, concurrent with the substantial functional acquisitions in
language and motor control that the infant achieves during the first 2
years of life. The neural differentiation that occurs in the context of medi-
ating normal psychological learning remains an enigma, addressed spec-
ulatively in classic works such as that by Hebb (1949) and more recently
by Bakker (1984).
Birth occurs while the majority of neuronal populations are still dif-
ferentiating. As the sine qua non for further brain-behavior development,
birth provides the evolving brain with the environmental context to guide
subsequent maturation. The full-term newborn's brain is a functioning
organ, although historically it was considered to be quiescent. A cardinal
change in all developmental disciplines has been the relatively recent
understanding of the functioning of the neonatal brain. This has entailed
its recognition as more than the modest precursor to the adult brain. The
neonatal brain has matured by birth to encounter the real-time ontogenet-
ic adaptations it must make to ensure continued normative development,
albeit predominantly through relatively more mature reflexive subcortical
systems. These adaptations have been described by Oppenheim (1981)
and Turkewitz and Kenny (1982). Examples of these adaptations are neo-
natal reflexes and modal sensory abilities. Many observed neonatal reflex-
es, once considered to be the primitive features of mature behavior, are
now considered to be prenatal neurobehavioral adaptations to increase
the probability that the fetus will be in the lower-risk, normal vertex birth
position. Likewise, sensory abilities at any given time during develop-
ment are attuned to adaptive ontogenetic requirements for different com-
binations of intramodal and intermodal performances, and are not simply
the immature versions of adult sensory capabilities.
Myelination
Prior to myelination, neuronal communication is characterized by
slow transmission rates and long refractory periods (Bronson, 1982). Al-
though many neuronal populations are structurally mature prior to their
myelination, true functional maturity occurs only after this process of
exogenous axonal sheathing has occurred. The relative rates of myelina-
tion in different brain regions do serve as a rough index of the sequence in
which components of the eNS reach functional adult levels (Yakovlev &
Lecours, 1967), although a myelin sheath is not necessary for conduction
of action potentials and some neuronal populations do not require mye-
lination at all (Bronson, 1982; Purpura & Reaser, 1974). The population of
glial cells that produce myelin have late proliferation periods and mature
postnatally. The bulk of axonal myelination, therefore, occurs during in-
fancy and early childhood. New glia and their myelin product are the
prime factors (along with continued structural differentiation of neuron
populations) that account for the dramatic postnatal growth in the size of
the brain (Bronson, 1982). Brain weight at birth is approximately 350 g,
about one-quarter of approximate adult weight. However, by age 2, brain
weight is approximately 1300 g-close to the adult range of 1400-1500 g.
Myelination, like other dimensions of brain development, proceeds
in a general temporal-spatial pattern. This was noted as early as 1901 by
Flechsig. Yakovlev and LeCours (1967) and LeCours (1975) have provided
a detailed description of human myelogenetic development (Figure 2).
Beginning with the sensory and motor roots of the spinal cord and pro-
ceeding to axons of neurons in "primordial zones," such as the primary
sensorimotor cortical areas, myelination starts prenatally and continues
through early infancy. In the pyramidal tract, for example, the major
efferents from the motor cortex begin to undergo myelination 1 month
prior to birth and are essentially completed by the end of the first year of
life. The slower, postnatal myelination of cortical-subcortical pathways,
callosal fibers, and association cortex Golgi class II interneurons is related
to the gradual emergence of language, cognitive, and motor skills from late
infancy through middle childhood.
Better clinical appreciation of normal (and aberrant) patterns of mye-
IGestationlFirst Year! 1-10 Years 10-30 Years

,..--
Sensory Roots I
Medial Lemniscus
>en I ~ Somesthetic Radiation

a::;;
Ow -=:::::::::J Statoacoustic, Tectrum, Tegmentum
en I-
zen Colliculus Inferior
w>
en en Acoustic Radiation
f::::J Optic Nerve and Tract, Colliculus Superior
f<::J OptiC Radiation I
I-- I
a: en Motor Roots I
i
0:;; Cerebellar Peduncles
I- w
oli; Pyramidal Tracts
:;;>
en
T Corpus Striatum
t--
I ~
Reticular Formation
w
:=:en 1 Fronto-Pontine Tract
1-:;;
w
a: I-
T Non-specific Thalamic Radiation
(!len
w>
I-en
I Cerebral Commissures
:!: I
'--- I Iintra-cortic Assoc. Area
FIGURE 2. Myelogenetic cycles of regional maturation in the human brain. (Adapted from
Yakovlev & LeCours, 1967.)
lination may be substantially enhanced by the introduction of magnetic

resonance imaging (MRI) of the brain (Risser, 1984). MRI is the best avail-
able clinical imaging modality for visualizing white matter and gray-
white matter delineation (Lee et 01., 1986). MRI data samples have been
providing atlaslike delineations of the course of normal (McArdle et 01.,
1987) and abnormal (Johnson et 01., 1987) myelination from preterm birth
through early childhood. This imaging modality should prove invaluable
for correlational research comparing structural images and indices of
brain development, such as myelination patterns, with neurobehavioral
features in individual children.
Functional Organization
Postnatally, during periods of continued differentiation and myelina-
tion, brain development is paralleled by elaboration of the organism's
adaptive behavioral repertoire. Exactly how structure and function are
related is unclear, but it appears that the two are mutually dependent.
Structural changes appear in many ways to produce modifications in
behavior, whereas various dimensions of brain structure are shaped, to
some degree, by behavioral experiences that may contribute to individual
differences in brain-behavior patterns (e.g., Greenough & Juraska, 1979).
The traditional topograpical representation of different structural areas on
the lateral and medial surfaces of the brain and their related functions are
illustrated by Brodmann's map of the cytoarchitecture of the cortex (Fig-
ure 3), which defines areas by cell structure and cellular layers. To the
neuropsychologist, units of interest are composed by the functional inte-
gration of these regions. Components of functionally integrated units at-
tain maturity at different rates and at different times during early develop-
ment. As a result, assessment of children depends upon measures that
take into account the pattern of structural and functional maturation at
any given time in a changing, maturing brain.
The "behavioral geography of the brain" (Lezak, 1983, p. 41) reflects
the discipline of adult behavioral neurology and is well presented in
many neuropsychological resources (e.g., Walsh, 1978). One of several
ways to approach this behavioral geography, which incorporates develop-
ment, was presented by Luria (1973). As part of Luria's neuropsychologi-
cal model of three functional units delineating the CNS (Le., arousal,
informational interaction, and programming units), three functional cor-
tical zones may be distinguished and described ontogenetically. These are
the primary, secondary, and tertiary zones.
Primary zones are modality specific and receive input from the
senses or control motor activity. These zones are somatotopically ar-
ranged and possess an abundance of neurons in afferent cell layer IV.
Their characteristic topographical organization and modality specificity
Function
VISION
primary 17
secondary 18.19
AUDITORY
pnmary 41
secondary 22.24
BODY SENSES
prImary 1.2.3
secondary 5.7
SENSORY. TERTIARY 7.21.22.37.39.40
MOTOR
primary 4
secondary 6
eye movement 8
speech 44
MOTOR, TERTIARY 9.10. 11 .45.46.47
FIGURE 3. Brodmann's cytoarchitectonic areas of the brain. (From Elliott, 1969.)
are helpful diagnostically because lesions in these zones lead to fairly

stereotypic deficits. According to Luria's developmental approach, these
zones become fully functional by the end of the first year of life and form
the basis for successful secondary zone expression. Secondary zones lie
adjacent to primary ones and integrate modality-specific information into
perceptive information. Afferent cell layers II and III predominate and are
composed of associative microneurons. Lesions in these areas give rise to
perceptual deficits that typically are restricted to specific modalities, but
the deficits are less stereotypic. These areas become fully functional with-
in the first 5 years of life and form the basis for successful tertiary zone
expression. Tertiary zones are associative, supramodal areas encompass-
ing the borders of parietal, temporal, and occipital zones as well as the
vast prefrontal region with its abundant cortical and subcortical (e.g.,
thalamus) connections. These zones generally serve to integrate informa-
tion across modalities and to control executive, purposive, and conative
aspects of functioning. Supramodal regions become functional between
the ages of 5 and 8, whereas the prefrontal region becomes functional
later. PassIer, Isaac, and Hynd (1985) suggested that this prefrontal devel-
opment is dynamic, occurring in stages, with functional maturity by
age 12.
Over the course of development, not only is a hierarchical rela-
tionship of zones expressed, but a lateralization of the functioning of
these zones also is obtained between the two hemispheres. Lateralization
of hemispheric functional specialization, of course, has held a predomi-

nant position in adult and developmental neuropsychological research,
given compelling evidence for both structural and cognitive asymmetries
(Bryden, 1982; Corballis & Morgan, 1978; Geschwind & Galaburda, 1985a,
1985b, 1985c). Conscious activity occurs through the integrated working
of hierarchical and lateralized specialized zones within the three-unit
functional system that is the CNS.
Summary
If made more elaborate, this selective overview of the most rapid and
critical periods of growth observed in the developing brain would fall far
afield of the overall focus of this volume. Yet it is during this period of
normal brain development that the bulk of the neural framework for sub-
sequent neuropsychological functioning is created. It is also here, per-
haps, that a subclinical anoxic-ischemic episode or some microstructural
or neurochemical anomaly may set the stage for a neuropsychological
disorder, such as a specific learning disability or an attentional deficit,
that may not be evident until later in childhood.
Proposals of the sort described above have a notable history (Kno-
bloch & Pasamanick, 1966) and have been presented on the basis of selec-
tive neuropathological findings (Fuller, Guthrie, & Alvord, 1983; Towbin,
1971), but they remain speculative. Speculative relations of this type,
tested on a case-study postmortem basis, have gained an invaluable docu-
mentational aid in Galaburda's cytoarchitectonic studies of the
postmortem brains of dyslexic individuals (Galaburda & Eidelberg, 1982;
Galaburda & Kemper, 1979; Galaburda, Sherman, Rosen, Aboitiz, &
Geschwind, 1985). In group analyses, however, such speculative relations
are usually found to account for only a very small portion of outcome
variability. Naeye and Peters (1987), for example, found that the presence
of fetal and perinatal hypoxia accounted for only 2% of IQ variation on the
Wechsler Intelligence Scale for Children (WISC) at age 7. In sum, although
at a tangent to the practical concerns of the clinician, explorations of the
neural side of normal brain-behavior development do, and shall continue
to, contribute to a fuller understanding of normal cognitive and behav-
ioral development and provide a firmer basis for appreciating the signifi-
cance of deviations from normal development.
ABNORMALITIES IN NEURAL DEVELOPMENT
An insult to brain tissue at any time may result in impairments in

behavior, and these impairments may be permanent. Damage to the young
CNS, however, may cause quantitatively and qualitatively different im-
pairments (with differences in permanence) than would damage to the
adult eNS. Although the bulk of the literature addressing this issue has
been delineated in terms of age, it is now recognized that reference to the
age of the eNS may be more confusing than helpful in seeking to under-
stand those eNS mechanisms that account for differences in the develop-
ing versus the mature brain's response to injury. More precisely, damage
to the functionally uncommitted brain may not be the same as damage to
the committed brain (Goldman-Rakic et 01.,1983). Relevant experimental
evidence about the traditional role of age dates back to Stoltmann in 1876
(cited by Goldman-Rakic et 01., 1983) and has generated two heuristic
guidelines about the timing of a brain insult. It also has created an arena
for critical debate about recovery of function following a brain insult
(Isaacson, 1975; Schneider, 1979; St. James-Roberts, 1979).
The two guidelines that have emerged from this literature are the
Kennard principle (Kennard, 1940, 1942) and the Dobbing hypothesis
(Dobbing & Smart, 1974). The ablation studies in monkeys by Kennard
(1942) led to the general conclusion that brain lesions early in life pro-
duced fewer deficits than similar lesions in later life. The Kennard princi-
ple states, therefore, that the degree of impaired function subsequent to
brain damage is proportional to the age at which the injury occurred. The
Dobbing hypothesis, tested extensively upon rat models of adult brain-
behavior consequences of early-life malnutrition or irradiation (Dobbing
& Sands, 1971), states that brain damage during development has its great-
est impact upon those cell populations and/or individual growth param-
eters that show the greatest rate of development at the time of the insult.
Insults occurring during cell proliferation limit the number of cells in the
mature brain; insults during differentiation do not necessarily alter the
number of cells in the brain but do influence cell size-a gross index of
altered cellular architecture (Winick & Rosso, 1969).
Euphemistically, Kennard's principle suggests that earlier rather than
later brain damage would be preferable in sparing function, whereas Dob-
bing's hypothesis suggests the opposite-later, rather than earlier, brain
damage would be preferable to avoid permanent impairment. The fact
that either may have relevance at different levels of analysis (e.g., cellular,
systemic, behavioral) and in different contexts (e.g., language develop-
ment, electrophysiology) points toward an underlying complexity that
evades a simple explanation. The behavioral consequences of brain
damage are made more complex by findings that suggest that plasticity at
a structural level is not necessarily associated with improved function
(Isaacson, 1975).
Perhaps the simplest statement that can be made from an integration
of both rules is that brain damage during infancy and early childhood may
prove less devastating in outcome than damage either during fetal devel-
opment or in the mature brain. The timing of an insult to the developing
brain will influence the nature and extent of any resultant deficit and the
probability for the acquisition of the normal repertoire of abilities. Inter-

ference with neural development at a given time will have a characteristic
influence on the normal status of different cell populations depending
upon what stage(s) of growth they are experiencing. This will have a
compounding impact upon the organism's ability to recover losses, devel-
op further, and reach levels of ultimate maturity. Given the previously
outlined stages of brain development, it is important to recognize that an
early lesion is a fetal lesion, not necessarily one that occurs during infan-
cy or childhood. This recognition has not always proved of benefit in
neuropsychology for two reasons.
First, traditional examinations of age issues in neuropsychology have
compared adult ("late") with child or infant ("early") lesions. Age is a
comparison with practical value. For example, adult soldiers of different
ages who suffered war-related brain injury showed different levels of
resiliency to long-term behavioral losses (Teuber, 1975). In addition, chil-
dren, like adults, usually have some degree of vocal and cooperative in-
teraction with the trained assessor. Although the assessor may not use
tests created from a developmental neuropsychological perspective, avail-
able adult neuropsychological tests have been revised, renormed, or rein-
terpreted to permit purposeful administration to young children and can
be supplemented by the use of established child psychological instru-
ments, such as the McCarthy ability scales (McCarthy, 1972). Examples of
downwardly extended adult tests include the Halstead-Reitan Battery
(Reitan & Davison, 1974), the Neurosensory Center Comprehensive Exam-
ination of Aphasia (Gad des & Crockett, 1973), and the Token Test for
Children (DiSimoni, 1978). Despite the practicality of using such "bor-
rowed" tests, their interpretation remains that of how children perform on
adult-derived tests rather than how children perform on tests that have
been constructed to reflect purer developmental neuropsychological mea-
sures of age-relevant behaviors. How one can collect infant and preschool
neuropsychological data, which are valuable both on their own and in
making a comparison with adolescent and adult findings, remains a chal-
lenging investigational question (see Chapter 9, this volume).
Second, many of the truly early insults (e.g., lesions that occur during
the prenatal periods of proliferation and migration) may result in spon-
taneous abortion or create individual cases that are not usually accessible
to neuropsychologists. Prenatal lesions can have a devastating impact
upon the integrity of the CNS. Survivors of such early insults are more
frequently seen in neurological, rather than neuropsychological, practice
and usually are left with severe mental and sensorimotor deficits. On the
other hand, determining whether subtle prenatal insults may evade iden-
tification at onset but leave some discernible "thumbprint" on subse-
quent neuropsychological development awaits a suitable methodology for
study.
Timing will never be the sole dimension of an insult to be considered

in trying to appreciate its consequences. The nature, severity, and loca-
tion of the insult, control over secondary aspects of the insult (e.g., in-
creased intracranial pressure or seizures), and treatment all need to be
considered, as is true for any neurological insult (Adams & Victor, 1985).
However, timing of insult is the obvious defining developmental dimen-
sion. The significance of timing as an isolated dimension in altering de-
velopment may be seen more clearly in a neuropathology laboratory than
in a neuropsychology clinic. The general timing of an insult during neural
migration provides an example in that it may alter the number of neocor-
tical cell layers (e.g., Lund, 1978). These consequences range from schiz-
encephaly (agenesis of entire regions, which can result from an insult
during the second gestational month) to the mildest migratory defect (re-
sulting from insults prior to the seventh month) in which all six neocor-
tical layers are present, but deviant collections of cells (heterotopias) are
found buried in white matter, not having migrated to their proper loca-
tions.
Timing, in a neuropsychological context, also is correlated with age-
related, ontogenetic changes in the epidemiology of neurological insults
across the life-span, the extent of the repertoire of neuropsychological
abilities available to be assessed, the probability of functional restitution,
and the availability of reliable and valid assessment devices (Boll & Barth,
1981; Chelune & Edwards, 1981; Pape & Fitzhardinge, 1981; Rourke et a1.,
TABLE 1. Selective List of Possible Etiologic Agents, Neural Abnormalities,

and Neuropsychological Consequences a
Neuropsychological
Etiologic agent Neural abnormality consequence
Genetic Microcephaly Mental retardation

Nutritional Regional agenesis Learning disability
Traumatic Migratory defect Attentional disorder
Infectious Axon/ dendrite Sensory loss
Metabolic damage Motor deficit
Intoxicant Synaptic loss Language deficit
Systemic disease Synaptic aberrance Speech deficit
Maternal disease Tissue infarction Memory disorder
Anoxic-hypoxic Tissue ischemia Emotional alteration
Neoplastic Hemorrhage Isolated defects
Environmental deprivation Demyelination Developmental delay
Idiopathic Aberrant cell growth Normal-range functioning
Gliosis
Diffuse
Unidentifiable
"Categories are not meant to be exhaustive in delineation or representative of any classification system.
TABLE 2. Selective List of Mediating

Variables Influencing the Impact of an
Etiologic Agent upon Neural Substrate and
Neuropsychological Functioning
Timing, nature, and severity of insult

Age at insult or gestational age at birth
Location, extent, velocity of lesion formation
Plastic, homeostatic, and defense features
Neural commitment
Functional plasticity
Genetic compliment
Diagnosis, treatment, and management
Control over secondary consequences of insult
Prevention
1983). In practice, however, only gross distinctions in timing (i.e., pre/

peri/postnatal) can usually be made for insults that place the child at risk
for altered behavioral development (Kopp, 1983; Kopp & Parmelee, 1979).
Table 1 presents a selective listing of etiologies, neural abnormalities,
and neuropsychological consequences; Table 2 presents a selective listing
of mediating variables that may relate to them. Neuropsychological syn-
dromes are a consequence of neural abnormalities that, themselves, are
due to some underlying etiology. Variations in the outcome of both neuro-
psychological and neural consequences are due to multidimensional me-
diating variables that encompass systemic, behavioral, and educational
influences on development. Agents that impair the normal progression of
brain development are varied. Identification of agent etiology, when pos-
sible, is part of the medical diagnostic process, even if the diagnosis
occurs years after the insult and can be stated only in probabilistic terms.
Identifying the consequences of an insult on the neural substrate is a
feature of the neurological diagnostic process. Comprehensive descrip-
tions of these identifiable consequences are provided in child neurology
texts (e.g., Menkes, 1985). Neuropsychological consequences span the
observable, assessable aspects of neuropsychological syndromes. General
descriptions of most of these syndromes are readily available-e.g.,
Robinson and Robinson's text (1976) on mental retardation, Gaddes
(1985) on learning disabilities, Barkley (1981) or Weiss and Hechtman
(1979) on hyperactivity.
Genetic diseases (i.e., hereditary and chromosomal aberrations) are
the most frequent causes of neural abnormalities, but, because they are so
often incompatible with continued embryonic or fetal development, only
a minority of such fetuses may actually reach birth. Defined sets of phys-
ical and intellectual features (e.g., Down's and Turner's syndromes) char-
acterize many of the survivable genetic diseases (Robinson & Robinson,
1976). Although the predominant worldwide cause of acquired defective
brain development and neurobehavioral morbidity is protein-calorie mal-
nutrition and its attendant milieu of deprivation and infection (Pollitt &
Thomson, 1977), the most frequent identifiable cause in industrialized
societies is traumatic brain damage. Traumatic brain damage in late infan-
cy and childhood is typically the result of head injuries caused by falls
and motor vehicle or recreational accidents (Annegers, Grabow, Kurland,
& Laws, 1980).
In an earlier era, perinatal brain trauma was a fairly frequent phe-
nomenon. Although medical advances have resulted in a significant de-
crease in birth-related mechanical injury, other frequent perinatal causes
of neural abnormalities usually result from the spectrum of risks associ-
ated with preterm and postterm births. This is particularly true for the
very-Iow-birth-weight (VLBW; defined as a birth weight at or under 1500
g) premature newborn (Levene & Dubowitz, 1982). The VLBW neonate is
at particular risk for periventricular and intraventricular hemorrhages
from the germinal matrix and the choroid plexus (Goddard-Finegold,
1984). Other risks for the premature newborn include anoxic-ischemia,
seizures, infections, and systemic disease. Generally, however, single fac-
tors are less important than combinations of complications in determin-
ing outcome. Although survival rates have improved, neurobehavioral
morbidity is a frequent outcome. Performances of VLBW infants usually
are poorer than normal-birthweight infants on neurobehavioral measures
at follow-up (Kitchen et a1., 1986).
DEVELOPMENTAL NEUROPSYCHOLOGICAL IMPLICATIONS
Many of the challenges in contemporary assessment occur because

(1) an insult's nature and timing may be silent at its onset and on medical
evaluation, making it an intangible influence on functioning, (2) neuro-
psychological problems may appear only once the child has "grown up"
into a challenging academic environment, where the deficit is forced to
reveal itself, (3) the neuropsychological consequences of a prior insult
may dissipate via natural compensatory mechanisms and not manifest
themselves as might be expected, (4) similar pathological events may lead
to different disorders depending on mediators, such as the time of onset,
and (5) it is not always possible to distinguish pathological from imma-
ture (Le., still evolving) performances on evaluation.
Except in the context of addressing a referral of a child with a known
neurological etiology, it is common for many neuropsychological prob-
lems to present themselves without an overt cause and without gross
brain damage (children with learning problems being the archetypical

example). Many neuropsychological problems may not have an etiology
that can be identified from the child's history and performance data.
These problems are neuropsychological manifestations of what seems to
be a dysfunctional brain, in the absence of other, nonneuropsychological
explanations that could provide a more valid, parsimonious understand-
ing of the observed anomalies. Contemporary controversies exist over
several basic neuropsychological explanations. Two controversies of rele-
vance to developmental neuropsychological practice are (1) the validity of
the concept of "soft" neurological signs, and (2) the value of comparative
statements about the "dysfunctional" brain, such as deficit, delay, lag, or
retardation in development, used to interpret poor neuropsychological
performances.
Clinical concern exists over the ability to identify and attribute soft
signs to disordered brain functioning (Shafer, Shaffer, O'Connor, & Stok-
man, 1983). Soft signs include dysdiadokinesia, mirror movements, dys-
graphesthia, astereognosis, and choreiform movements. Soft signs may be
a consequence of early acquired lesions due to obstetric difficulties
(Prechtl & Stemmer, 1962). Conversely, they may represent normal genet-
ic variants (Nichols & Chen, 1981), they may be generated by stress or be
learned behaviors (Prechtl & Stemmer, 1962), or, finally, they may be
transient phenomena observed during neurological maturation (Peters,
Romaine, & Dykman, 1975). These hypotheses are all generally compati-
ble with a developmental course in which soft signs become more diffi-
cult to elicit with increasing age.
The presence of soft signs is sometimes stated to be diagnostically
helpful and to possess some predictive usefulness. Some studies have
found that, in children of normal intelligence without focal neurological
signs, the presence of soft signs is associated with cognitive dysfunction,
learning difficulties, and behavioral disturbances (Shafer et al., 1986;
Shaffer, O'Connor, Shafer, & Puris, 1983). The usefulness of soft signs as
predictors of later neurological and psychiatric status has been the subject
of longitudinal research (Sarazin & Spreen, 1986; Shafer et a1., 1986;
Shaffer, 1978; Stokman et al., 1986). These findings usually have reported
persistent soft signs in male adolescents with accompanying behavioral
disturbances. Other research findings, however, have been equivocal and
contradictory. Some of the contradictions can be explained by mis-
classification, the reliability of measurement, the confounding effects of
focal neurological findings or low levels of cognitive functioning, and
biases in sampling created by relying upon clinic-referred children
(Shaffer et al., 1983).
Despite the literature that documents their existence and correlates,
soft signs are considered to be of minimal value. In general, the base rate
of neurological soft signs in children who do not show problems is so high
as to suggest that these phenomena are of only limited clinical value for
screening and prediction. Although these are associated with learning
and behavior problems, even on a long-term basis in certain children, they
also are observed frequently in normal children without being correlated
with cognitive deficits or maladaptive outcomes.
Assessing whether the term dysfunctional actually can be employed to
implicate covert aberrant brain function, which can only be detected upon
proper and sensitive (typically neuropsychological) evaluation, has a legit-
imate basis, despite its various controversies (Benton, 1975). Among the
many available works, Taylor et al. (1984) provide a concise review of the
concept of dysfunction in the evolution of developmental neuropsychol-
ogy; a thoughtful paper by Denckla (1978), which addresses key conceptual
concerns, is also of value. One realm of developmental neuropsychological
research has been directed at determining the existence of brain-behavior
relations in the absence of both overt brain damage and a recognized
neurological etiology. Examples of this research include the search for
biochemical explanations of minimal brain damage (Wender, 1971), struc-
tural hemispheric deviations from the range of normative findings in
computed tomography (CT) (Haslam, Dalby, Johns, & Rademaker, 1981;
Hier, LeMay, Rosenberger, & Perla, 1978; LeMay, 1981) and in MRI imaging
data of the brains of dyslexic children (Rumsey et a1., 1986), and the
detailed cytoarchitectonic study of the brains of dyslexics at autopsy (e.g.,
Galaburda & Kemper, 1979).
Another aspect of concern is whether observed dysfunctions are truly
evidence of neurological defects or instead are indicators of a develop-
mental delay. The concept of delayed development infers that there is a
slower rate of maturation of otherwise normal brain regions and behav-
ioral manifestations of brain functioning (Satz, Taylor, Friel, & Fletcher,
1978). Mental retardation in the absence of a genetic or acquired struc-
tural etiology might be considered an example of delayed maturation in
its most severe form. A maturational lag is based on the assumption that
structure and function normally mature according to an identifiable pat-
tern, are hierarchically organized, and develop sequentially by a system of
associated and interrelated components. According to Rodier (1984),
there is little evidence of a genuine delay (Le., a retardation) in brain
morphology. A brain that has sustained injury during development does
not appear immature or "unfinished." Rather, the regions forming during
the injury appear abnormal. The implication of a delay (Le., that at some
point in the future, development will eventually be finished) appears
inappropriate to describe morphologically based dysfunctions.
The concept of delayed maturation has been questioned on the
grounds that if a delay in function is due to a maturational lag, then
psychological development should otherwise be normal, except for the
fact that its timing is off (Knights & Bakker, 1976). However, a 12-year-old
with a reading problem does not read as if he or she were 7 years old.
Research with reading-disabled children suggests that learning strategies
used by these children are not merely delayed but are truly aberrant, thus
suggesting a deficit in skills rather than skill slowness. An early work by
Witelson (1976), for example, posited an association between reading
problems and inadequate hemispheric specialization for both linguistic
and spatial functioning. The same finding appears true for children with
language delays. In addition to a delay in language function, these chil-
dren often are found to show deviant aspects of language function (Men-
yuk, 1978).
A key aspect in the debate about developmental delays is found in the
learning disability follow-up literature. Studies have demonstrated that
although learning-disabled readers may improve their skills somewhat
over time, they continue to be disabled readers, never "catching up" to
their peers in reading skills (Rutter, Tizard, Yule, Graham, & Whitmore,
1976; Satz & Sparrow, 1970; Satz, et al., 1978; Spreen, 1978). In fact, it is
likely that the presence of delay itself may act as an influence upon the
child, thus altering the form and quality of continued growth (Zigler,
1969). Conceivably, a developmental delay could result in faulty organi-
zation of cognitive skills, causing defective and aberrant levels of perfor-
mance. On the other hand, the presence of a frank deficit does not imply
that some gains cannot be anticipated over time or that aggressive inter-
vention may not prove a significant aid in improving the child's abilities
to approximate better adaptive functioning. Validation of an acceptable
construct of delay will await the development of direct measures of in
vivo brain maturation, and a clearer understanding of the relation be-
tween function and brain maturation.
CONCLUSIONS
Assessment of children to evaluate suspected neuropsychological

deficits has proceeded for many years in the absence of a formal neuro-
psychology of the developing brain (Benton, 1982). A work by E. M.
Taylor (1959), as one example, provided detailed instructions concerning
the evaluation of psychological consequences of pediatric neurological
insults, prior to the appreciation of the value of developmental neuropsy-
chology that occurred during the 1970s. Further, many successful child
neuropsychology clinics have been functioning for well over 15 years.
Practical clinical demands cannot await the results of systematic basic
research, but they can generate interest in executing such work. One
prime goal of the emergent neuropsychology of the developing brain is to
provide procedures and objective brain-behavior validations to increase
both the efficiency and the effective scope of the neuropsychological as-
sessment of the child (Spreen et a1., 1984). In the context of advanced

imaging technologies (MRI and ultrasound), evolving neuropsychological
and educational interventions, and a multidisciplinary professional com-
munity, all bodes well for the future of a developmental neuropsychology.
REFERENCES
Adams, R. D. & Victor, M. (1985). Principles of neurology (3rd ed.). New York: McGraw-Hill.
Annegers, J. F., Grabow, J. D., Kurland, 1. T., & Laws, E. R. (1980). The incidence, causes, and
secular trends of head trauma in Olmsted County, Minnesota, 1935-1974. Neurology,
30, 912-919.
Bakker, D. J. (1984). The brain as a dependent variable. Journal of Clinical Neuropsychology,
6, 1-16.
Barkley, R. A. (1981). Hyperactive children: A handbook for diagnosis and treatment. New
York: Guilford Press.
Benton, A. (1975). Developmental dyslexia: Neurological aspects. Advances in Neurology, 7,
1-27.
Benton, A. (1982). Child neuropsychology: Retrospect and prospect. In J. de Wit & A. Benton
(Eds.), Perspectives in child study: Integration of theory and practice (pp. 41-61). Lisse:
Swets and Zeitlinger.
Boll, T. J., & Barth, J. T. (1981). Neuropsychology of brain damage in children. In S. B.
Filskov & T. J. Boll (Eds.)' Handbook of clinical neuropsychology (pp. 418-452). New
York: Wiley Press.
Bronson, G. (1982). Structure, status and characteristics of the nervous system at birth. In P.
Stratton (Ed.), Psychobiology of the human newborn (pp. 99-114). New York: Wiley
Press.
Bryden, M. P. (1982). Laterality: Functional asymmetry in the intact brain. New York:
Academic Press.
Chelune, G. J., & Edwards, P. (1981). Early brain lesions: Ontogenetic environmental consid-
Chi, J. G., Dooling, E. C., & Gilles, F. H. (1977). Gyral development of the human brain.
Annals of Neurology, 1, 86-93.
Corballis, M. C., & Morgan, M. J. (1978). On the biological basis of human laterality: 1.
Evidence for a maturational left-right gradient. Behavioral and Brain Sciences, 2, 261-
336.
Cotman, C. W., & Nieto-Sampedro, M. (1982). Brain function, synapse renewal, and plas-
ticity. Annual Review of Psychology, 33, 371-401.
Cotman, C. W., & Nietro-Sampedro, M. (1984). Cell biology of synaptic plasticity. Science,
225, 1287-1294.
Cowan, W. M., Fawcett, J. W., O'Leary, D. D. M., & Stanfield, B. B. (1984). Regressive events
in neurogenesis. Science, 225, 1258-1265.
Denckla, M. B. (1978). Minimal brain dysfunction. In J. S. ChaJl & A. F. Mirsky (Eds.),
Education and the brain, 77th yearbook of the National Society for the Study of Educa-
tion (pp. 223-268). Chicago: University of Chicago Press.
DiSimoni, F. (1978). The Token Test for Children, manual. Hingham, MA: Teaching Re-
sources Corp.
Dobbing, J., & Sands, J. (1971). Vulnerability of developing brain: IX. The effect of nutritional
growth retardation on the timing of the brain growth spurt. Biology of the Neonate, 19,
363-378.
Dobbing, J., & Smart, J. L. (1974). Vulnerability of developing brain and behavior. British
Medical Journal, 30, 164-168.
Elliott, H. (1969). Textbook of anatomy. Philadelphia: Lippincott.
Flechsig, P. (1901). Developmental (myelogenetic) localization in the cerebral cortex of the
human subject. Lancet, 2, 1027-1029.
Fuller, P. W., Guthrie, R. D., & Alvord, E. C. (1983). A proposed neuropathological basis for
learning disabilities in children born prematurely. Developmental Medicine and Child
Neurology, 25, 214-231.
proach (2nd ed.). New York: Springer-Verlag.
Gaddes, W. H., & Crockett, D. J. (1973). The Spreen-Benton aphasia tests, normative data as
a measure of normal language development (Research Monograph No. 25). Victoria, BC:
University of Victoria.
Galaburda, A. M., & Eidelberg, D. (1982). Symmetry and asymmetry in the human posterior
thalamus. II. Thalamic lesions in a case of developmental dyslexia. Archives of Neu-
rology, 39, 333-336.
Galaburda, A. M., & Kemper, T. L. (1979). Cytoarchitectonic abnormalities in developmental
dyslexia: A case study. Annals of Neurology, 6, 94-100.
Galaburda, A. M., Sherman, G. F., Rosen, G. D., Aboitiz, F., & Geschwind, N. (1985). Devel-
opmental dyslexia: Four consecutive patients with cortical anomalies. Annals of Neu-
rology, 18, 222-233.
Gazzaniga, M. S., Stein, D., & Volpe, E. T. (1979). Functional neuroscience. New York:
Harper & Row.
Geschwind, N., & Galaburda, A. M. (1985a). Cerebrallateralization: Biological mechanisms,
associations, and pathology. A hypothesis and a program for research. Archives of
Neurology, 42, 428-459.
Geschwind, N., & Galaburda, A. M. (1985b). Cerebrallateralization: Biological mechanisms,
Neurology, 42, 521-552.
Geschwind, N., & Galaburda, A. M. (1985c). Cerebrallateralization: Biological mechanisms,
Neurology, 42, 634-654.
Goddard-Finegold, J. (1984). Periventricular, intraventricular hemorrhages in the premature
newborn: Update on pathologic features, pathogenesis, and possible means of preven-
tion. Archives of Neurology, 41, 766-771.
Goldman, P. S. (1979). Development and plasticity of frontal association cortex in the in-
frahuman primate. In C. L. Ludlow & M. E. Doran-Quine (Eds.). The neurological bases
of language disorders in children: Methods and directions for research (NIH Publication
No. 79-440) (pp. 1-16). Washington, DC: U.S. Government Printing Office.
Goldman, P. S., & Galkin, T. W. (1978). Prenatal removal of frontal association cortex in the
fetal rhesus monkey: Anatomical and functional consequences in postnatal life. Brain
Research, 152, 451-485.
Goldman-Rakic, P., Isserhoff, A., Schwartz, M. L., & Bugbee, N. M. (1983). The neurobiology
of cognitive development. In M. M. Haith & J. J. Campos (Eds.). Infancy and develop-
mental psychobiology; Vol. 2 of P. H. Mussen (Ed.). Handbook of child psychology (4th
ed., pp. 281-344). New York: Wiley Press.
Greenough, W. T., & Juraska, J. M. (1979). Experience-induced changes in brain fine struc-
ture: Their behavioral implications. In M. E. Hahn, C. Jensen, & B. C. Dudek (Eds.).
Development and evolution of brain size (pp. 295-320). New York: Academic Press.
Haslam, R. H. A., Dalby, J. T., Johns, R. D., & Rademaker, A. W. (1981). Cerebral asymmetry
in developmental dyslexia. Archives of Neurology, 38, 679-682.
Hebb, D. O. (1949). Organization of behavior. New York: Wiley Press.
Hier, D. B., LeMay, M., Rosenberger, P. B., & Perlo, V. P. (1978). Developmental dyslexia:
Evidence for a subgroup with a reversal of cerebral asymmetry. Archives of Neurology,
35, 90-92.
Isaacson, R. L. (1975). The myth of recovery from early brain damage. In N. R. Ellis (Ed.),
Aberrant development in infancy (pp. 1-25). Potomac: Erlbaum.
Jacobsen, c. F. (1936). Studies of cerebral function in primates. Comparative Psychological
Monographs, 13, 1-68.
Jacobson, M. (1978). Developmental neurobiology (2nd ed.). New York: Plenum Press.
Johnson, M. A., Pennock, J. M., Bydder, G. M., Dubowitz, L. M. S., Thomas, D. J., & Young,l.
R. (1987). Serial MR imaging in neonatal cerebral injury. American Journal of Neu-
roradiology, 8, 83-92.
Kandel, E. R. & Schwartz, J. H. (1981). Principles of neural science. New York: Else-
vier/North-Holland.
Kennard, M. A. (1940). Relation of age to motor impairment in man and subhuman primates.
Archives of Neurology and Psychiatry, 44, 377-397.
Kennard, M. A. (1942). Cortical reorganization of motor function: Studies on series of
monkeys of different ages from infancy to maturity. Archives of Neurology and Psychia-
try, 47, 227-240.
Kitchen, W. H., Rickards, W. H., Ryan, M. M., Ford, G. W., Lissenden, J. V., & Boyle, L. W.
(1986). Improved outcome to two years of very low birthweight infants: Fact or artifact.
Developmental Medicine and Child Neurology, 28, 579-588.
Knights, R. M., & Bakker, D. J. (Eds.). (1976). The neuropsychology of learning disorders:
Theoretical approaches. Baltimore: University Park Press.
Knobloch, H., & Pasamanick, B. (1966). Prospective studies on the epidemiology of re-
productive casualty: Methods, findings, and some implications. Merrill-Palmer Quar-
terly of Behaviour and Development, 12, 27-43.
Kopp, C. B. (1983). Risk factors in development. In M. M. Haith & J. J. Campos (Eds.), Infancy
and developmental psychobiology; Vol. 2 of P. H. Mussen (Ed.), Handbook of child
psychology (4th ed., pp. 1081-1188). New York: Wiley Press.
Kopp, C. B., & Parmelee, A. H. (1979). Prenatal and perinatal influences on infant behavior.
In J. Osofsky (Ed.), Handbook on infant behavior (pp. 29-74). New York: Wiley-
Interscience.
LeCours, A. R. (1975). Myelogenetic correlates of the development of speech and language.
In E. H. Lenneberg & E. Lenneberg (Eds.), Foundations of language development: A
multidisciplinary approach (Vol. 1, pp. 121-135). New York: Academic Press.
Lee, B. C. P., Lipper, E., Nass, R., Ehrlich, M. E., deCiccio-Bloom, E., & Auld, P. A. M. (1986).
MRI of the central nervous system in neonates and young children. American Journal of
Neuroradiology, 7, 605-616.
LeMay, M. (1981). Are there radiological changes in the brains of individuals with dyslexia?
Bulletin of the Orton Society, 31, 135-141.
Lemire, R. J., Loeser, J. D., Leech, R. W., & Alvord, E. C. (1975). Normal and abnormal
development of the human nervous system. New York: Harper & Row.
Levene, M. I., & Dubowitz, L. M. S. (1982). Low-birth-weight babies long-term follow-up.
British Journal of Hospital Medicine, 24, 487-491.
Lezak, M. (1983). Neuropsychological assessment (2nd ed.). New York: Oxford Press.
Lund, R. D. (1978). Development and plasticity of the brain: An introduction. New York:
Oxford Press.
Luria, A. R. (1973). The working brain: An introduction to neuropsychology. Middlesex:
Penguin Press.
McArdle, C. B., Richardson, C. J., Nicholas, D. A., Mirfakhree, M., Hayden, C. K., & Amparo,
E. G. (1987). Developmental features of the neonatal brain: MR imaging: Part I. Gray-
white matter differentiation and myelination. Radiology, 162, 223-229.
McCarthy, D. (1972). McCarthy scales of children's abilities. New York: Psychological

Corporation.
Menkes, J. H. (1985). Textbook of child neurology (3rd ed.). Philadelphia: Lea & Febiger.
Menyuk, P. (1978). Linguistic problems in children with developmental dysphasia. In M. A.
Wyke (Ed.), Developmental dysphasia (pp. 135-157). London: Academic Press.
Naeye, R L., & Peters, E. C. (1987). Antenatal hypoxia and low IQ values. American Journal
of Diseases of Childhood, 141,50-54.
Nichols, P., & Chen, T. (1981). Minimal brain dysfunction: A prospective study. Hillsdale,
NJ: Lawrence Erlbaum.
Oppenheim, R W. (1981). Ontogenetic adaptations and retrogressive processes in the devel-
opment of the nervous system and behavior: A neuroembryological perspective. In K. J.
Connolly & H. F. R Prechtl (Eds.), Maturation and development: Biological and psycho-
logical perspectives (pp. 73-101). Philadelphia: Lippincott.
Pape, K. E., & Fitzhardinge, P. M. (1981). Perinatal damage to the developing brain. In A.
Milusky, E. A. Friedman, & L. Gluck (Eds.), Advances in perinatal medicine (Vol. 1, pp.
45-86). New York: Plenum Press.
Passier, M. A., Isaac, W., & Hynd, G. W. (1985). Neuropsychological development of behav-
ior attributed to frontal lobe functioning in children. Developmental Neuropsychology,
1, 349-370.
Peters, J. E., Romaine, J. S., & Dykman, R A. (1975). A special neurological examination of
children with learning disabilities. Developmental Medicine and Child Neurology, 17,
63-78.
Pollitt, E., & Thomson, C. (1977). Protein-caloric malnutrition and behavior: A view from
psychology. In R. J. Wurtman & J. J. Wurtman (Eds.), Nutrition and the brain (Vol. 2, pp.
261-304). New York: Raven Press.
Prechtl, H. F. R, & Stemmer, C. H. (1962). The choreiform syndrome in children. Develop-
mental Medicine and Child Neurology, 4,119-127.
Purpura, D. P., & Reaser, E. P. (1974). Methodological approaches the study of brain matura-
tion and its abnormalities. Baltimore: University Park Press.
Purves, D., & Lichtman, J. W. (1980). Elimination of synapses in the developing nervous
system. Science, 210, 153-157.
Rakic, P. (1981). Developmental events leading to laminar and areal organization of the
neocortex. In F. Schmitt (Ed.), The organization of the cerebral cortex (pp. 7-28).
Cambridge: M.I.T. Press.
Reitan, R M., & Davison, L. A. (Eds.). (1974). Clinical neuropsychology: Current status and
applications. New York: Wiley Press.
Risser, A. H. (1984, February). Nuclear magnetic resonance imaging: A primer for the neu-
ropsychologist. Paper presented at the XIIth annual meeting of the International Neuro-
psychological Society, Houston, TX.
Robinson, H., & Robinson, N. (1976). The mentally retarded child (2nd ed.). New York:
McGraw-Hill.
Rodier, P. M. (1984). Exogenous sources of malformations in development. In E. S. Gollin
(Ed.), Malformations of development: Biological and psychological sources and conse-
quences (pp. 287-313). New York: Academic Press.
Rourke, B. P., Bakker, D. J., Fisk, F. L., & Strang, J. D. (1983). Child neuropsychology: An
introduction to theory, research, and clinical practice. New York: Guilford Press.
Rumsey, J. M., Dorwart, R, Vermess, M., Denckla, M. B., Kruesi, M. J. P., & Rapoport, J. L.
(1986). Magnetic resonance imaging of brain anatomy in severe developmental dyslexia.
Archives of Neurology, 43, 1045-1046.
Rutter, M., Tizard, J., Yule, W., Graham, P., & Whitmore, K. (1976). Research report: Isle of
Wight studies 1964-1974. Psychological Medicine, 6, 313-332.
Sarazin, F. F-A. & Spreen, O. (1986). Fifteen-year stability of some neuropsychological tests
in learning disabled subjects with and without neurological impairment. Journal of

Clinical and Experimental Neuropsychology, 8, 190-200.
Satz, P., & Sparrow, S. (1970). Specific developmental dyslexia: A theoretical formulation. In
D. J. Bakker & P. Satz (Eds.), Specific reading disability: Advances in theory and method
(pp. 17-39). Rotterdam: Rotterdam University Press.
Satz, P., Taylor, H. G., Friel, J., & Fletcher, J. M. (1978). Some developmental and predictive
precursors of reading disabilities: A six year follow-up. In A. Benton & D. Pearl (Eds.),
Dyslexia: An appraisal of current knowledge (pp. 313-348). New York: Oxford Press.
Schneider, G. E. (1979). Is it really better to have your brain disease early? A revision of the
"Kennard principle." Neuropsychologia, 17, 557-584.
Shafer, S. Q., Shaffer, D., O'Connor, P. A., & Stokman, C. J. (1983). Hard thoughts on neu-
rological "soft signs." In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 133-
143). New York: Guilford Press.
Shafer, S. Q. Stokman, C. J., Shaffer. D., Ng, S. K-C., O'Connor, P. A., & Schonfield, I. S.
(1986). Ten year consistency in neurological test performance of children without focal
neurological deficit. Developmental Medicine and Child Neurology, 28, 417-427.
Shaffer, D. (1978). "Soft" neurological signs and later psychiatric disorder-A review. Jour-
nal of Psychology and Psychiatry and Allied Disciplines, 19, 63-65.
Shaffer, D., O'Connor, P. A., Shafer, S. Q., & Puris, S. (1983). Neurological soft signs: Their
origins and significance for behaviour. In M. Rutter (Ed.), Developmental neuropsychia-
try (pp. 144-164). New York: Guilford Press.
Spreen, O. (1978). Learning disabled children growing up (Final report to Health and Wel-
fare Canada, Health Programs Branch). Ottawa: Health and Welfare Canada.
Spreen, 0., Tupper, D., Risser, A., Tuokko, H., & Edgell, D. (1984). Human developmental
neuropsychology. New York: Oxford Press.
St. James-Roberts, I. (1979). Neurological plasticity, recovery from brain insult, and child
development. In H. W. Reese & P. Lipsitt (Eds.). Advances in child development and
behavior (Vol. 14, pp. 253-319). New York: Academic Press.
Stokman, C. J., Shafer, S. Q., Shaffer, D., Ng, S. K-C., O'Connor, P. A., & Wolff, R W. (1986).
Assessment of neurological soft signs in adolescents: Reliability studies. Developmental
Medicine and Child Neurology, 28, 428-439.
Taylor, E. M. (1959). Psychological appraisal of children with cerebral defects. Cambridge:
The Commonwealth Fund.
Taylor, H. G., Fletcher, J. M., & Satz, P. (1984). Neuropsychological assessment of children.
In G. Goldstein & M. Hersen (Eds.), Handbook of psychological assessment (pp. 211-
234). New York: Pergamon Press.
Teuber, H-L. (1975). Recovery of function after brain injury in man. Outcome of severe
damage to the central nervous system, Ciba Foundation Symposium 34 (new series)
(pp. 159-186). Amsterdam: Elsevier.
Thomas, J. (1968). Introduction to human embryology. Philadelphia: Lea & Febiger.
Towbin, A. (1971). Organic causes of minimal brain dysfunction: Perinatal origin of minimal
cerebral lesions. Journal of the American Medical Association, 217, 1207-1214.
Turkewitz, G., & Kenny, P. (1982). Limitations on input as a basis for neural organization and
perceptual development: A preliminary theoretical statement. Developmental Psycho-
biology, 15, 357-368.
Walsh, K. W. (1978). Neuropsychology: A clinical approach. Edinburgh: Churchill Living-
stone.
Weiss, G., & Hechtman, C. (1979). The hyperactive child syndrome. Science, 205, 1348-
1354.
Wender, P. H. (1971). Minimal brain dysfunction in children. New York: Wiley-Interscience.
Williams, R, & Caviness, V. B. (1984). Normal and abnormal brain development. In R E.
Tarter & G. Goldstein (Eds.), Advances in clinical neuropsychology (Vol. 2, pp. 1-62).
New York: Plenum Press.
Winick, M., & Rosso, P. (1969). The effect of severe early malnutrition on cellular growth of
human brain. Pediatric Research, 3, 181-187.
Witelson, S. F. (1976). Abnormal right hemisphere specialization in developmental dyslexia.
In R. M. Knights & D. J. Bakker (Eds.), The neuropsychology of learning disorders:
Theoretical approaches (pp. 233-256). Baltimore: University Park Press.
Witelson, S. F. (1977). Early hemisphere specialization and interhemisphere plasticity: An
empirical and theoretical review. In S. J. Segalowitz & F. A. Gruber (Eds.), Language
development and neurological theory (pp. 213-287). New York: Academic Press.
Yakovlev, P. 1., & LeCours, A. R. (1967). The myelogenetic cycles of regional maturation of
the brain. In A. Minkowski (Ed.), Regional development of the brain in early life (pp. 3-
69). Oxford: Blackwell Scientific.
Zigler, E. (1969). Development versus difference theories of mental retardation and the
problem of motivation. American Journal of Mental Deficiency, 73, 536-556.
3
The Role of Neuropsychological

Assessment in Relation to Other
Types of Assessment with Children
ERIN D. BIGLER
INTRODUCTION
Neuropsychological assessment represents an important aspect in the

overall evaluation of the child with neurological disorder. The efficacy of
neuropsychological findings, however, can be greatly enhanced by a bet-
ter understanding of the interrelationship between neuropsychological
testing and other aspects of the neurological and psychological evalua-
tion. With recent advances in neurodiagnostics, particularly in the area of
brain imaging and related assessment techniques, there is a greater abun-
dance of clinical information that must be integrated with neuropsycho-
logical findings. This has resulted in further overlap between specialty
areas in the clinical neurosciences and has furthered the need for one
specialty to understand the significance of findings in the others. To that
end, this chapter will attempt to relate and contrast neuropsychological
test findings with other assessment practices.
First, the pediatric neurological exam will be reviewed. This has been
a standard part of any neurological evaluation with children, and fre-
quently it is the first examination undertaken with the child presenting
neurobehavioral symptoms. Accordingly, it is important to understand
the relationship between pediatric neurological exam findings and neuro-
psychological assessment. There has been a marked increase in the uti-
lization of brain imaging and electrophysiological techniques in the as-
ERIN D. BIGLER Austin Neurological Clinic, Austin, Texas, and Department of Psychol-
ogy, University of Texas at Austin, Austin, Texas.
67
68 ERIN D. BIGLER
sessment of neurological disorders of children. The next two sections of

the chapter will address these areas in relationship to neuropsychological
assessment. Last, the unique role of neuropsychological assessment in the
context of traditional psychological testing will be reviewed and dis-
cussed.
PEDIATRIC NEUROLOGICAL EXAM
Basic Components
The pediatric neurological exam (PNE) naturally varies with the age
of the child. For the purpose of this chapter, the neurological exam out-
lined will be based on the child 4 years of age and older since it is at this
age that the exam begins to take on the characteristics of the more formal,
systematic exam of the adolescent or adult patient (Hartlage & Krawiecki,
1984). In general, the components of the exam include an evaluation of
gait and station, motor and sensory function, reflex integrity, and physical
development, along with a brief overview of mental status. These topic
areas will be briefly reviewed below.
Station and Gait

Normal gait consists of smooth, effortless movement that is fluid,
symmetric, and rhythmic (Brown, 1982). Leg position should be narrow-
based and arm swing should be even. Such features also should be present
when gait is stressed such as in tiptoe and heel walking, balancing, or
running. Abnormal motor findings constitute a deviation from this. For
example, spastic gait typically is characterized by the child who has a
shuffling gait, with flexion at the knees and hips, along with increased
adductor tone. These abnormal movements are greatly exaggerated when
the child is asked to run or when gait is otherwise stressed. These can
result from damage to the primary motor system at various levels or re-
gions. Gait disturbance secondary to cerebellar involvement is quite dif-
ferent, with the prominent features being a wide-based, veering, and un-
steady ataxic gait in which the child has great difficulty in controlling or
adjusting to postural change (i.e., turning around). With peripheral motor
involvement, gait disturbance typically takes the form of a waddling or
steppage gait. The examination for disturbance in gait and station is made
through direct inspection of the child performing various movements.
Motor Function
Muscle tone and strength are tested by the child's strength of grip and
limb strength resistence in various positions of the extremities. Muscle
NEUROPSYCHOLOGICAL AND OTHER TYPES OF ASSESSMENT 69
tone is visually and tactually inspected, as well as the child's response to

reflex assessment in upper and lower extremities. Hyper- or hyporeflexia,
or abnormal muscle tone activity, may be a sign of peripheral or central
nervous system involvement. The differential diagnosis is based on the
pathognomonic signs present in relation to other findings (e.g., hyper-
reflexia associated with spasticity in hemiplegia of cortical origin) and is
thoroughly discussed in the work of Brown (1982). Fine motor control
typically is examined by having the child attempt to make rapid, repetitive
finger-oscillatory movements between the index finger and thumb, fol-
lowed by successive finger-tapping movements where each finger is touch-
ed in succession by the thumb. By the age of 6, these movements should be
mastered and executed with little difficulty (Herskowitz & Rosman, 1982).
By itself, impaired performance on these fine motor tasks has no particular
localizing significance other than indicating impaired motor system func-
tioning. Finger-to-nose and heel-to-shin movements provide an index for
assessing cerebellar integrity. Cerebellar tremor may be seen in conjunc-
tion with cerebellar movement disorder, with the tremor present on inten-
tion or a loss of rhythmicity and smoothness of movement during action.
Cranial Nerve Function

The cranial nerve exam historically has been the foundation of the
neurological exam and has changed very little in the past century (Van
Allen, 1969). For children, cranial nerves 1, 5, and 7 through 12 are tested
essentially as in the adult. Optic nerve (second cranial nerve) function is a
critical part of any neurological exam but represents some obstacles for
the younger child. Simple visual acuity is usually measured first, fol-
lowed by visual field testing, which may require some innovative meth-
ods for younger children (see Touwen, 1979). Cranial nerves 3, 4, and 6
control ocular motility and are tested by observing eye movements.
Younger children may have difficulty in following commands, and in
such cases the examiner will have to depend on observation of spon-
taneous performance.
Sensory-Perceptual Function
Simple touch usually is tested by the child's ability to report light
touch. Two-point tactile discrimination is another test of simple tactile
perception. By the time a child is 6 years of age, formal testing of stereog-
nosis (tactile perception of a coin, paper clip, or key placed in the palm of
the hand) also can be undertaken. By 8 years of age, formal graphesthesic
perception can be tested using letters or numbers "written" on the palm
or fingertips (Herskowitz & Rosman, 1982). These tests, along with the
perception of double simultaneous tactile stimulation, constitute the as-
70 ERIN D. BIGLER
sessment of higher parietal function. Abnormalities found on any of these

assessment techniques typically imply impaired parietal lobe functioning
contralateral to the side on which the tactile errors were made. Auditory
screening is typically based on the child's ability to hear whisper speech
by the examiner and the perception of double simultaneous auditory
stimulation. Abnormalities within this aspect of the examination typ-
ically do not have clear pathognomonic significance but indicate the need
for further investigation of the auditory system. Finally, double simul-
taneous visual stimulation is utilized to examine visual inattention/neg-
lect in one hemifield or the other, and it is sometimes included in the
cranial nerve evaluation when visual functioning is examined. Visual
inattention/neglect/extinction in one visual field suggests contralateral
occipital lobe involvement (and/or adjacent association cortical areas).
The Nature of Findings

The PNE is, at best, a screening exam and review of systems. It repre-
sents an overview of central nervous system (CNS) functioning from
which some inferences can be made concerning the integrity or dysfunc-
tion of neurological systems. Pathological indicators on the standard PNE
typically represent focal versus generalized dysfunction. For example, a
child with a history of severe encephalitis may show generalized motor,
sensory, and reflex abnormalities that point toward nonspecific cerebro-
cortical dysfunction, as opposed to the child who sustained a focal brain
injury with resultant hemiplegia (a focal sign indicating contralateral
frontal involvement) as the only neurological abnormality. Similarly, ab-
normalities found on the PNE typically imply some direct index or patho-
gnomonic sign of underlying neurological deficit, the so-called hard neu-
rological signs.
Anomalous Physical Development as Markers for eNS Pathology

Congenital physical anomalies may be indicators of abnormal brain
development due to altered embryogenesis (genetic) or an insult (e.g.,
trauma, toxicity) early in pregnancy. The occurrence of CNS anomalies in
relation to age of embryo or fetus is represented in Table 1. The presence
of such physical anomalies is a direct indication of aberrant CNS organi-
zation and development (Williams & Caviness, 1984).
Table 2 presents the Yale Neuropsychoeducational Assessment
Scales-Stigmata schedule, which overviews the potential physical stig-
mata that may be present. Although the presence of one of these anomalies
is not necessarily clinically significant, numerous studies (see review by
Shaywitz, 1982) have suggested that the presence of three or more of these
minor anatomical defects may be associated with a variety of neu-
~
~
en
~
t""
TABLE 1. Occurrence of eNS Anomalies in Relation to Age of Embryo or Fetus a ~

Days 18-19-20-21-22-23-24-25-26-27-28-29-30-31-32-3 3-34-3 5-36-3 7-38-39-40-41-42-43-44-45-46-4 7-48- Months 3-4-5- 6- 7-8- 9-10
?:
49-50-51-52-53-54-55-56
~
o
Embryogenesis Fetogenesis
~
~
Dysraphias , 27 days
(Anencephaly, spina bifida)
Holoprosencephalon , 29 days ~en
A genesis of interhemispheric connections , 6th month o'"'l
Abnormal cellular migration in the cerebral cortex ,7th month
(Pachygyria, lissencephaly, heteropia of gray matter) >-en
Porencephaly - - 9th month g;
Hydrencephaly - 9th month en
en
oThe length of each arrow corresponds to the age of the embryo or fetus during which the causative (teratogenic) injury to the nervous system is believed to
occur. For example, it is suggested that the insult to the brain that results in anencephaly occurs between day 18. which is the date of the appearance of the ~
neural plate, and day 27. when the tubulation of the neural plate is terminated. The time indicators shown here for the production of anomalies are tentative;
they are suggested on the basis of animal experiments and a few rare observations of embryos and fetuses presenting with a particular anomaly. (From
Duckett. 1981.)
'-l
....
TABLE 2. Anomaly Grading Scale of the Yale
Neuropsychoeducational Assessment Scales
Hair: Frontal bossing:
Whorls: (1) lor a line '3 1" long)
Electric hair: (Normal) (Fine: soon awry) (Very
fine; will not comb)
Coarse, brittle:
Eyes: Strabismus:
Ptosis:
Crusting of lid:
Epicanthus: (absent) (partial) (deep)
Ears: Malformed
Asymmetric:
Pits or tags:
Soft and pliable:
Low set (above) (';;0.5 cm below) (>0.5 cm below)
line through nosebridge and eye
Lobe extension: (below attachment) (straight back)
(toward crown)
Nose: Antiverted nostrils:
Hypoplastic tip:
Hypoplastic philtrum:
Mouth: Palate: (normal arch) (flat steeple ri,) (high
steeplerl)
Cleft palate-soft or hard:
Tongue furrowed:
Tongue with smooth-rough spots:
Teeth-enamel hypoplasia; carious:
Teeth small:
Upper vermillion thin:
Jaw: Prominent maxilla:
Micrognathia:
Prognathia:
Midfacial hypoplasia:
Torso: Webbed neck:
Pectus excavatum:
Pectus carinatum:
Nipples widely spaced:
Heart murmur:
Vertebral defects:
Hernia-umbilical, diaphragm, groin;
diastasis recti
Hypospadias:
Skin: Thin skin-visible vascular pattern:
Dry, coarse, rough:
Cafe-au-lail:
Hemangiomata:
Nails-hypoplastic or dysplastic:
Hands: Palm-coarse skin:
Palm-single transverse crease:
Small, broad hands:
Thin, tapering hands:
Joint limitation:
Clinodactyly:
Broad first toe or thumb:
Feet: Gap between first and second toes:
Partial syndactylia of 2nd and 3rd toes:
Length of 2nd and third toes:
(2nd> 3rd) (2nd = 3rd) (2nd < 3rd)
Note. From Shaywitz, 1982.

robehavioral syndromes, including mental retardation and associated in-

tellectual deficits, learning disability, developmental delay, and attention
deficit disorder. The presence of a minor congenital anomaly (MCA) ,
although signifying the potential presence of some neurological irreg-
ularity, lacks specificity with respect to locus of neurological dysfunction.
Whereas MCA findings appear to represent a "marker," with the presence
of the:;;e findings being indicative of possible incomplete or irregular neu-
ronal development, only limited clinical inferences can be made beyond
this point. Such early factors may have a deleterious effect on subsequent
development, as well as on the development of related systems (e.g., early
dysfunction of perceptual systems may affect motor control, although no
direct motor involvement is present). Thus, in children with MCAs, there is
a greater likelihood of underlying aberrant neuronal development, which,
in turn, may be associated with a greater likelihood of associated neu-
rological abnormalities. However, the presence of MCAs should not be
construed as pathognomonic indicators per se.
"Hard" versus "Soft" Neurological Findings

A standard PNE usually includes an examination of minor neu-
rological dysfunction (Touwen, 1979). Whereas abnormalities of gait and
station, reflex, motor strength, or primary sensation may indicate focal
"hard" neurological signs, a variety of minor neurological abnormalities
(the so-called soft neurological signs) may be elicited in some children
whose neurological exam is otherwise normal. Minor neurological abnor-
malities also may coexist with more specific neurological abnormalities.
The most common minor neurological abnormalities include dys-
diadochokinesia, associated movements (variously known as synkinetic
movements, comovements, and mirror movements), and the so-called
choreiform twitch. Associated movements are normally present in the
infant and young child and decrease with age. In fact, the disappearance
of the elicitation of associated movements is taken as a sign of functional
maturation of the CNS (Herskowitz & Rosman, 1982). The presence of the
"minor" neurological abnormalities beyond the age of 8 or 9, however, is
felt to be suggestive of neuropathology, although the meaning of such
signs in these older children remains in doubt. Numerous studies tend to
indicate a higher incidence of cognitive dysfunction, learning difficulties,
attention deficit disorder, and other psychiatric disturbance in children
with neurological "soft signs" and MCAs (Kaufman, 1981; Rie & Rie,
1980; Shaffer et 01., 1985; Shafer, Shaffer, O'Connor, & Stokman, 1983).
But as Shaffer, O'Connor, Shafer, and Propis (1983) state, "the rela-
tionship of 'soft signs' to ... specific syndromes is less certain."
Research to date suggests that the origin of minor neurological abnor-
malities is likely multifactorial. In some children, the presence of these
74 ERIN D. BIGLER
characteristics may be a consequence of cerebral injury in the prenatal or

perinatal period (see Table 1). For others, the presence of these irreg-
ularities is under genetic control and expression, representing develop-
mental differences. In still others, there may be an interactive effect (e.g.,
motor maladroitness may interfere with critical time periods of motor
development). Yet no research has demonstrated any direct relationship
between the presence of such findings and specific neuropsychological
impairments. Rather, these neurological "soft signs" appear to be "mark-
ers" suggestive of underlying neurological irregularity, and clinical cor-
relation with presenting problem and other neurological findings is
needed on an individual case basis.
Mental Status Characteristics of Neurologically Impaired

Children
Behavioral observation and the evaluation of the child's interaction
with the examiner are integral aspects of the PNE. First, the child's ap-
pearance should be noted with respect to appropriateness of dress, groom-
ing, and hygiene. As children grow older, they tend to be more concerned
about their appearance, but some children with neurological impairment
may remain somewhat oblivious and uncaring about how they appear. An
assessment of the child's general level of psychomotor activity is essen-
tial, particularly in terms of general activity level, degree of impulsivity,
and attention. Some children with underlying neurological dysfunction
tend to be more active, impulsive, and inattentive, characteristics also
seen in attention deficit disorder.
There is some tendency in children with neurological impairment to
be concrete in their interpretation of proverbs and to have difficulty with
any mental task requiring verbal abstraction. This also serves as an infor-
mal assessment of the child's verbal intelligence. Likewise, speech con-
tent may provide further insight into the child's thought processes. Cer-
tain child psychiatric disorders, as with adults, will be manifested via
distorted, illogical thinking. The mental status exam always should exam-
ine for delusions, hallucinations, tangentiality, or looseness of thinking as
well as confusion (checking for orientation to time, place, and person).
For the normal child, speech should be flowing, logical, and in direct
relationship to questions asked. Further, it should contain spontaneous
content rich in information concerning the child's interests, hobbies, and
general sense of well-being. Insight and judgment are also elemental as-
pects of a mental status exam, and neurologically impaired children fre-
quently will display some deficits in these realms. Shaffer et 01. (1985), in
a prospective study of children with minor neurological abnormalities
first identified at age 7 and followed up 10 years later, found this group to
be more anxious, withdrawn, and depressed.
Relationship between Pediatric Neurological Exam and

The PNE constitutes a review of neurological systems, but by no
means is it an exhaustive evaluation of them. Rather, the neurological
exam is more of a screening procedure, the results of which direct the
nature and type of further investigations. Much of the PNE is involved in
assessing brain function that would be described as rudimentary and
reflexive. Accordingly, a given patient may have a "normal" PNE but
could have underlying cerebral damage/dysfunction. This is particularly
true when cerebral involvement is in "silent" association cortical areas
where no specific motor, sensory, or language deficit is present.
Neuropsychological assessment, on the other hand, provides a com-
prehensive evaluation of functions that are related to, or dependent on,
higher cerebral processes. Neuropsychological assessment likewise pro-
vides an index of functional integrity of more basic cerebral processes.
Accordingly, the two techniques overlap to a certain extent but for the
most part are examining brain function at different levels and to different
degrees.
The PNE, particularly when minor neurological abnormalities are
found, may give only a hint that a neurological deficit may be present. In
such cases, the neuropsychological evaluation may be the essential as-
sessment tool in documenting the nature and type of impairment present.
Also, a child that is experiencing learning problems, developmental de-
lays, or change in conduct frequently will be taken first to the pediatrician
to determine if there is some "medical" explanation for these behaviors
(Dworkin, 1985; Schimschock, Milford-Cooley, & Cooley, 1984). In such
cases, the PNE is the initial assessment tool, which in turn may lead to the
more comprehensive neuropsychological assessment or other neurodiag-
nostic tests.
Several studies (see Touwen, 1979) have attempted to utilize PNE
findings to predict neuropsychological outcome. These studies have been
disappointing in that there appears to be little direct relationship between
PNE findings and neuropsychological impairment. This should not be
surprising, though, when one considers the complexity of human behav-
ior, particularly cognitive processes, and the rather elementary nature of
the PNE. These studies do support the role of the PNE as an initial assess-
ment procedure that represents an overview of basic CNS integrity in
which abnormal findings direct further studies.
NEURORADIOLOGICAL TESTS
Brain imaging via the methods of computerized axial tomography

(CT) and magnetic resonance imaging (MRI) have revolutionized diag-
76 ERIN D. BIGLER
FIGURE 1. MRI and CT depictions of neuroanatomy. The MRI scan on top is taken in the
sagittal plane and depicts normal brain anatomy. Major anatomic structures are labeled
numerically (l-corpus callosum. 2-fornix. 3-thalamus, 4-hypothalamus, 5-mamillary bodies,
6-tegmentum, 7-tectum, 8-aqueduct of Sylvius, 9-fourth ventricle, 10-pons, ll-cerebellum,
12-frontal lobe, 13-parietal lobe, 14-occipitallobe, 15-temporallobe). Note the clarity and
resolution that can be obtained with MRI imaging. The CT scan on the bottom is in the
horizontal plane. This patient was an 18-year-old male who had suffered a severe head
injury 1 year prior to the CT. This scan was selected because it demonstrates frontal atrophy
and ventricular enlargement, which permits better visualization of some of the major ana-
tomi~ structures that are labeled numerically in the adjacent drawing (l-frontal lobe, 2-
interhemispheric fissure, 3-anterior horn, lateral ventricle, 4-caudate nucleus, 5-internal
capsule, 6-thalamus, 7-putamen/globus pallidus complex, 8-third ventricle, 9-posterior
horn, lateral ventricle, 10-Sylvian fissure, ll-temporallobe, 12-occipitallobe.
nostic neurology (Oldendorf, 1980). Current generation CT and MRI scan-

ners can provide exceptional detail with respect to gross anatomy of the
brain (see Figure 1). Both of the procedures are excellent in depicting
major structural anomalies of the brain such as dysplasic errors in devel-
opment (see Figure 2), hydrocephalus (see Figure 3), tumors (see Figure
4), degenerative effects due to infectious disorder (see Figure 5), con-
FIGURE 2. Cerebral dysplasia. Note the asymme-

try between the two cerebral hemispheres and the
density difference in the left occipital region. The
decreased density in the left occipital region rep-
resents either a failure of neuronal development
or early traumatic injury. This patient was
dyslexic.
genital porencephaly (see Figure 6), and a host of other childhood disor-
ders (see Bigler, 1984).
With the advent of these techniques, there was a period of great
optimism concerning the advancement of "lesion-localization" work ex-
amining specific functions of specific brain regions (Kertesz, 1983). Prior
FIGURE 3. MRI scan depicting hydrocephalus in a patient with a pituitary tumor.

78 ERIN D. BIGLER
FIGURE 4. CT scan depicting tumor (astro-

cytoma) in a l6-year-old male patient who pre-
sented with a lO-year history of learning and be-
havioral problems.
FIGURE 5. Degenerative changes depicted in a CT scan in a 13-year-old male patient who

suffered from diffuse encephalitis. The CT on the left demonstrates compression of the
ventricular system due to diffuse cerebral edema. As a consequence of the edema as well as
the brain infection, diffuse neuronal degeneration evolved, as depicted in the CT on the right
taken 3 months later. Premorbidly the patient's intellectual level was assumed to be in the
normal range. Postinfection his WISC-R Full Scale IQ score was 64.
FIGURE 6. CT scan depicting porencephaly. De-

spite the absence of the left frontal and most of
the left temporal and parietal lobes, this child
had normal intellectual and language develop-
ment (see Bigler & Naugle, 1984). Intellectual
testing at age 4 demonstrated a Verbal IQ (VIQ)
score of 102, a Performance IQ (PIQ) score of
100, and a Full Scale IQ (FSIQ) score of 101.
Four years later, the results were almost identi-
cal (VIS = 98, PIQ = 98, FSIQ = 98). The child
had no speech or language deficit, and achieve-
ment scores were at age- and grade-appropriate
levels. This case clearly demonstrates cerebral
plasticity in the immature nervous system.
to these imaging techniques, only invasive procedures were available for

examining brain structure, or one had to wait for postmortem examina-
tion. Initially it was thought that CT analysis would provide the essential
background for the lesion-localization work (Bigler, 1980). However, al-
though there are some general relationships that have been demonstrated
between neuropsychological outcome and CT and MRI findings, the rela-
tionships are certainly far from linear. For example, Bigler and colleagues
(Bigler, Hubler, Cullum, & Turkheimer, 1985; Cullum & Bigler, 1986) have
demonstrated only a modest relationship between overall neuropsycho-
logical performance and the degree of cerebral atrophy as a result of trau-
ma or degenerative disease. A major confounding factor with child stud-
ies examining the relationship between anatomic irregularities and
function is the cerebral plasticity issue. For example, Figure 7 depicts the
CT results of a child who had major structural defects in posterior aspects
of the cerebrum. Despit Despite having essentially no visual cortex or
posterior parietal-temporal association cortical areas, the child was able
to read at both age- and grade-appropriate levels, showed normal recep-
tive language abilities, and had a Verbal IQ in the near superior range!
None of these neuropsychological parameters would have been predicted
from her grossly abnormal CT results.
In light of the various interpretive problems alluded to above, the best
clinical approach to integrating CT and MRI data with neuropsychologi-
cal outcome appears to be in the use of CT IMRI in the depiction of struc-
tural abnormalities and neuropsychological results for the depiction of
abnormalities of function. It is tempting to imply functional abnormalities
on the basis of CT IMRI information. Likewise, it is alluring to infer struc-
80 ERIN D. BIGLER
FIGURE 7. CT scan of a 12-year-old female pa-

tient with a history of a congenital occipital en-
cephalocele, which was surgically removed at 4
years of age. Despite the marked structural ab-
normalities in the posterior aspect of both cere-
bral hemispheres and near complete absence of
both occipital lobes, this child's Verbal IQ score
was 119 on the Wechsler Intelligence Scale for
Children-Revised. Despite being cortically
blind for objects, she could read at age- and
grade-appropriate levels [Wide Range Achieve-
ment Test-Revised Reading score was BB (63%);
Boder Test of Reading-Spelling Patterns Read-
ing Quotient was 125, and on this measure she
was reading at the 10th-grade level]. This case
clearly demonstrates unexpectedly normal
neuropsychological findings in a child with
marked structural abnormalities.
tural abnormalities on the basis of neuropsychological test results. How-

ever, the clinician must remember that brain imaging with CT/MRI is not
a measure of cognitive functioning, and that neuropsychological perfor-
mance is not a direct measure of anatomic localization. The CT IMRI and
neuropsychological results should be used to integrate information about
a child's neurological status, both contributing to different aspects of neu-
rological integrity.
In the future some of the problems associated with integrating struc-
h.:ral data with functional information may be aided by such measures as
regional cerebral blood flow (RCBF; Risberg, 1986) and positron emission
tomography (PET) scanning (Heiss, Herholz, Pawlik, Wagner, & Wien-
hard, 1986). With these techniques it may be possible to determine the
dynamic relationship between anatomical structure and function. Unfor-
tunately, to date there have been few studies done with children, with the
majority of the work being done with adults having some type of acquired
neurological disorder.
ELECTRODIAGNOSTIC TESTS
The electroencephalogram (EEG) may be considered a measurement

of the ambient electrical activity of the brain. Of clinical interest is the
change or discontinuities that may occur within the four general EEG
frequency bands-delta (0-4 Hz), theta (4-8 Hz), alpha (8-12 Hz), and
beta (greater than 12 Hz). Pathology is frequently manifested when slow
activity (delta, theta) is augmented or fast activity (alpha, beta) is dimin-
ished. For example, a focal abnormality may alter the EEG in such a
fashion as to produce predominantly delta waves in that region, with
other recording sites displaying relatively normal activity. This would be
contrasted with diffuse slowing of the EEG, which would be seen in
generalized disorders (see Figure 8; Kiloh, McComas, & Osselton, 1979).
The clinical parameters of pediatric EEG have been well established
for over two decades (Kiloh et al. 1979). However, despite numerous
studies examining the relationship between EEG variables and neuropsy-
chological features, the results have generally lacked specificity (Brandeis
& Lehmann, 1986; Capute, Niedermeyer, & Richardson, 1968; Hughes,
1971; Hughes & Denckla, 1978). Although there is a greater incidence of
EEG abnormalities in children with learning disability, hyperactivity, psy-
chiatric disorder, and developmental deficits (Golden, 1982), there are no
distinctive EEG findings that are diagnostic of such conditions. The case
illustrated in Figure 8 is representative of this problem.
Evoked potential (EP) testing may reveal the intactness of visual,
auditory, and somatosensory pathways, and such findings may be useful
in cerebral localization studies (Luders, Dinner, Lesser, & Morris, 1986).
FIGURE 8. EEG in a 12-year-old right-handed male patient with a history of learning disor-
der. Asynchronous slow-wave (underlined) activity in the fight parietal region is noted. This
type of irregularity is seen with greater frequency in children with minor neurological
abnormalities. It does not indicate any specific abnormality and is not epileptogenic. Such
abnormalities frequently do not relate to any specific neurological or neuropsychological
pattern. In this child his neuropsychological studies suggested language-based deficits that
would implicate posterior left hemisphere dysfunction, but, as indicated, the EEG abnormal-
ity was in the right parietal region. Thus, such EEG abnormalities are a variant from normal
but do not necessarily correlate with the type of neuropsychological impairment present,
only with the presence of such impairment.
82 ERIN D. BIGLER
EP studies also have enjoyed great research popularity in a variety of

psychiatric disorders, including attention deficit disorder (Klorman et aI.,
1983) and autism (Courschesne, Kilman, Galambos, & Lincoln, 1984).
However, the clinical application of EP findings in relationship to cog-
nitive processes and neuropsychological outcome remains specious at
this time and under intense experimentation (Brandeis & Lehmann, 1986;
Visser, Njiokiktjien, & De Rijke, 1982).
Traditional interpretation of EEG or EP studies is and has been ac-
complished by visual inspection. However, the advent of computer-as-
sisted data analysis has permitted the development of a variety of inno-
vative methods for analysis (John et al., 1977). To date, the procedure that
appears to have the greatest promise is the Brain Electrical Activity Map-
ping (BEAM) technique pioneered by Duffy and colleagues (Duffy, Denck-
la, Bartels, & Sandini, 1980). The BEAM technique is illustrated in Figure
9. In a study examining BEAM findings in dyslexic boys and age-matched
normal readers, Duffy and McAnulty (1985) demonstrated demarcated
regional differences. The dyslexic children showed abnormalities in the
medial frontal (supplementary motor area), the left anterolateral lobe
(Broca's area), and the left posterior quandrant speech-associated areas
(Wernicke's area, angular gyrus, posterior temporal lobe, and parietal
lobe), with minor differences noted in the right parietal lobe. On the basis
of these findings, Duffy and McAnulty have speculated that dyslexic indi-
viduals manifest some degree of dysfunction throughout cerebral regions
involved in language-related functions.
Relationship between Electrodiagnostic Findings and

In the era prior to brain imaging and computerized EEG (Le., roughly
pre-1970s), routine EEG studies were frequently used to establish the
PATIENT
FIGURE 9. Representative BEAM study demonstrating an

abnormality in the left occipital-parietal region. The pa-
tient had a history of cerebral trauma and subsequent cog-
nitive deficit but had a normal routine EEG. Note how the
BEAM technique aids in the localization/lateralization of
22-1-250 t'ISEC EEG abnormalities.
locus of dysfunction according to lateralized electro physiological abnor-

malities. Neuropsychological studies were then undertaken to explore
lesion-localization and lateralization relationships. We now realize that
this procedure was a rather primitive one at best, and although the focal
EEG abnormality may be indiative of specific dysfunction in that area or
region, it did not address the likelihood of secondary abnormalities and
effects. This is exactly what the BEAM research has demonstrated-that
the electrophysiological abnormalities in learning-disabled children and
other children with minor neurological abnormalities may be more wide-
spread than once considered. These findings are not at variance with
anatomical studies (Galaburda, 1984; Kemper, 1984) that have demon-
strated more delimited microanatomical lesions in critical language and
association cortical regions in dyslexic subjects. It is likely that these
"lesions" are strategically situated so as to interfere with neuronal inter-
connections, thereby producing effects distant from the locus of anatomic
aberration. Alternately, there also may be multiple sites of dysfunction.
Neuropsychological studies with learning-disabled children have fre-
quently demonstrated that in addition to the specific learning deficit ex-
hibited, they may perform less well than normal controls on a variety of
neuropsychological measures unrelated to the primary deficit (Nussbaum
& Bigler, 1986; Nussbaum, Bigler, & Koch, 1986; Rourke, Bakker, Fisk, &
Strang, 1983). These neuropsychological findings support the BEAM re-
search implicating multifocality and/or multiple system involvement in
the neurological dysfunction associated with learning disorder.
Thus, neuropsychological findings and electrophysiological results
may be expected to differ just as electrophysiological results do not di-
rectly correspond to underlying anatomic abnormalities (Kiloh et al.,
1979). Accordingly, one should examine for consistencies in neuropsy-
chological findings and electrophysiological results so as to identify uni-
fying concepts of underlying neurological dysfunction. But where one
deviates from the other, this does not necessarily imply a lack of con-
sistency or superiority of one approach over the other. Neuropsychologi-
cal findings will always be more descriptive of functional levels of perfor-
mance (Gaddes, 1985), whereas electrophysiological techniques will
remain a direct, objective measure of the electrical activity of the brain,
with no particular reference as to function. For example, the current elec-
trophysiological methods cannot be utilized to separate groups of chil-
dren with average IQ from those with above-average IQ, but psychometric
testing can. Similarly, whereas neuropsychological studies may be able,
psychometrically and statistically, to differentiate between children with
and without seizure disorder, such studies are poor at clinically differ-
entiating seizure type, which is a rather straightforward task utilizing EEG
parameters (e.g., 3 per second spike-and-wave activity is diagnostic of
petit mal epilepsy; Kiloh et a1., 1979; O'Donohoe, 1981).
84 ERIN D. BIGLER
PSYCHOLOGICAL ASSESSMENT
Traditional psychological evaluation of the child typically has in-

cluded an assessment of intelligence, personality, academic performance,
and emotional state (Krahn, 1985). The presence of "neurological" or
"organic" signs sometimes has been an indirect objective of such evalua-
tions. This is exemplified by the Draw-A-Person (DAP) test. Frequently,
DAP results are rich in clinical information (Dileo, 1983), but such results
also have been utilized as an index of intelligence and development (Har-
ris, 1963). To a certain extent, the drawings also provide an index of
maturation (Scott, 1981), and, because of this, irregularities in drawing
may also indicate some form of perceptual-motor disturbance that is neu-
rologically based (Carr, 1985). Hence, some of the paper-and-pencil tests
such as the DAP have been viewed as demonstrating not only potential
personality indicators but also neurological dysfunction.
Similarly, the development of the Bender Visual-Motor Gestalt test
(BVMG) had its origin more in the realm of a maturational and personality
measure rather than as a neurological measure (Bender, 1938). However,
clinicians were quick to note that deficits in graphomotor ability occurred
significantly more often in children with "organic" dysfunction, and
hence the BVMG became the screening measure for organicity (Lacks,
1984). The era from 1930 to 1970 was a time period where the concept of
"organicity" (Small, 1980) remained in vogue, but as neurodiagnostic
sophistication increased, it became apparent that such screening mea-
sures as the BVMG (Bigler & Erfurth, 1981) and the concept of organicity,
per se, were obsolete (Bigler, 1984).
The reason that some psychological tests have sensitivity to neu-
rological processes is based on the direct relationship between develop-
mental stages of psychological growth and physical neuromaturational
growth (Rourke et al., 1983). This has been well demonstrated in an ele-
gant study by PassIer, Isaac, and Hynd (1985). They examined normal
male and female children at four age levels between 6 and 12 years of age.
They were particularly interested in examining the relationship between
maturation and the mastery of tasks that had been shown to be sensitive to
frontal lobe functioning. These tasks were selected because frontal lobe
maturation is thought to be incomplete until adolescence (Nauta & Feir-
tag, 1986). The authors found that the greatest development occurred at
the 6- and 8-year-old levels in terms of the child's performance on verbal
and nonverbal tasks requiring proactive and retroactive inhibition, and
tasks that elicited perseveration. By the age of 10, they found that the
ability to inhibit attention to irrelevant stimuli and perseveratory re-
sponses was almost complete, with mastery present by age 12. If such a
developmental model is used, failure to develop certain abilities by spec-
ified developmental stages may be a sign of neuromaturational deficit or
specific neurological impairment (Whelihan & Lesher, 1985). Conversely,

the expression of normal developmental staging probably represents nor-
mal neurological development. Thus, psychological measures that are
sensitive to developmental stages may indirectly be associated with neu-
romaturational factors as well.
Based on extensive research and standardization over the past two
decades, neuropsychological assessment has developed as an integrated
and comprehensive approach to evaluating behavioral aspects of brain
function (Bigler, 1984; Lezak, 1983). As a result, no single psychometric
test is appropriate for assessing brain function. Rather, neuropsycho-
logical assessment encompasses a wide selection of standardized tests
that carefully assess motor (including strength, dexterity, integrative
control, and visuopraxic/graphomotor abilities), sensory (basic and
complex aspects of visual, auditory, and tactile perception), language
(both expressive and receptive abilities, as well as the psychoeduca-
tional aspects of reading, spelling, writing, and math abilities), memory
(both verbal and nonverbal, as well as short- and long-term memory
functions), and general cognition (including measures of reasoning,
problem solving, and judgment, as well as standardized intellectual
tests). Many tests that are now incorporated into comprehensive neuro-
psychological test batteries were originally designed for other purposes.
This overlap between traditional psychological assessment and neuro-
psychological application can be exemplified by two tests, the Raven
(1965) Coloured Progressive Matrices (CPM) and the Beery (1967) Devel-
opmental Test of Visual-Motor Integration (VMI). The CPM test was orig-
inally designed as a nonverbal, motor-free measure of intelligence. How-
ever, the CPM, from the neuropsychological standpoint, has found
acceptance as a measure of visual-spatial reasoning that is more affected
by posterior cerebral lesions than by anterior (Lezak, 1983). Similarly,
the VMI test was originally designed as a "sensorimotor" test of visual-
motor development, and an actual developmental index was established.
Neuropsychologists have continued to use the developmental aspects of
this measure, but they utilize it as a direct measure of constructional
praxis as well.
In its infancy, clinical neuropsychology emerged from clinical psy-
chology and neurology, and accordingly had to utilize the assessment
methods employed at that time by those two disciplines. This is why
there was so much early reliance on traditional psychometric tests for
examining "brain function." Currently, clinical neuropsychology is in the
process of carefully delineating those tests that are specific behavioral
measures of brain function and creating novel measure that are based on
current neuroscience theory. The field is beginning to move away from
dependence on past psychometric measures that were developed for other
purposes (Delis, Kramer, Kaplan, & Ober, 1986).
86 ERIN D. BIGLER
Intellectual Assessment versus Neuropsychological Assessment

with Children: What's the Difference?
Test development in the measurement of intelligence predated the
emergence of the field of human clinical neuropsychology. Nonetheless,
because various measures of childhood intelligence are, in part, sensitive
measures of development, and because cognitive functioning is related to
brain function, these measures are also measures of cerebral integrity.
The prototype and standard of childhood intellectual testing is the
Wechsler Intelligence Scale for Children-Revised (Wechsler, 1974). The
development of the WISC-R-and its predecessor, the WISC (Wechsler,
1949)-took place in the embryonic stages of clinical neuropsychology.
This was long before the specialization of hemispheric functioning was
understood, and it is only fortuitous that Wechsler designed his test with
verbal and performance test items. Thus, the WISC was developed as a
measure of intelligence, not as a measure of neurological functioning.
However, studies soon began to demonstrate its efficacy in distinguishing
differential aspects of impaired ability in children with neurological dis-
orders (Kaufman, Long, & O'Neal, 1986).
Kaufman et a1. (1986) review the utilization of the WISC-R from a
neurological standpoint. Factor-analytic studies have demonstrated three
factors that have potential neuropsychological significance-a verbal
comprehension factor, a perceptual organization factor, and a freedom
from distractability factor. The first two loosely correspond to aspects of
hemispheric specialization-namely, language versus manipulospatial
abilities (Le., left and right hemispheric functions). Despite this rela-
tionship, VIQ-PIQ differences on the WISC may not hold the same sig-
nificance as that seen with adults (Boll & Barth, 1981). In part this is due
to developmental and cerebral plasticity issues of the immature brain.
Thus, whereas left hemisphere damage in the adult may affect VIQ more
than PIQ, with the opposite holding true for right hemisphere damage,
similar relationships may not be present with children (see Figure 6). For
example, as reviewed by Boll and Barth (1981), children with congenital
hemiplegic syndromes do not necessarily have VIQ-PIQ differences that
correspond to the affected side (Le., VIQ < PIQ in left-brain-damaged,
right hemiplegic children). To examine this problem further, Riva and
Cazzaniga (1986) studied children's performance on the WISC who had
sustained unilateral cerebral lesions before and after 1 year of age. Their
results indicated that left hemisphere lesions, regardless of whether they
occurred before or after the first year of life, affected both verbal and
performance scores, but with a greater negative impact when the lesion
occurred early in life. Right hemisphere damage, irrespective of whether
the lesion occurred prior to or after 1 year of age, affected only PIQ. Riva
and Cazzaniga suggest that recovery from unilateral deficits in children is
related to a hierarchy in cognitive development and that verbal skills are
usually spared at the expense of nonverbal skills (see also Bigler & Naugle,
1984).
Accordingly, some neuropsychological inferences can and should be
made from WISC-R results. However, the WISC-R is incomplete from the
neuropsychological perspective because it does not provide a direct as-
sessment of specific motor or sensory abilities, only incompletely assesses
language, and provides no in-depth assessment of memory. Thus, al-
though the WISC-R (like other measures of intelligence) is extremely valu-
able and worthwhile in the assessment of the child, it can be viewed only
as an adjunct to a comprehensive neuropsychological test battery. Such a
battery should assess cortical function in a thorough fashion, fully exam-
ining motor, sensory-perceptual, spatial-perceptual, memory (both verbal
and nonverbal), graphomotor, language, and cognitive functions.
Recently, the Kaufman Assessment Battery for Children (K-ABC) was
developed to overcome some of these problems (Kaufman & Kaufman,
1983). Instead of the Verbal-Performance schema, the K-ABC used the
dichotomy of simultaneous versus sequential/successive processing. This
differentiation is based, in part, on Luria's (1973) research that suggested
sequential processing to be primarily a frontotemporal activity and simul-
taneous processing to be more related to parietoccipital functioning. How-
ever, in preliminary studies (Morris & Bigler, 1987) the results are equiv-
ocal at this point in terms of the advantages of this measure over the
WISC-R in children with neurological disorders. Morris and Bigler found
that the K-ABC Simultaneous Processing score demonstrated greater sen-
sitivity for a variety of nonverbal, spatial-perceptual-motor deficits than
did the PIQ measure from the WISC-R. However, this did not hold up for
the Sequential score, with the WISC-R VIQ measures demonstrating a
greater sensitivity for verbal-language deficits in neurologically impaired
children.
Thus, traditional intellectual assessment methods yield only partial
neuropsychological information. Accordingly, these methods should be
considered as only a subset and not a substitute for the more complete
neuropsychological assessment. This is certainly well demonstrated by
the case illustrated in Figure 6. In this case, the child had shown average
intellectual ability over a span of 4 years, with no significant VIQ-PIQ
difference, despite the absence of two-thirds of her left cerebral hemi-
sphere. However, she did have a moderate right-sided spastic hemiplegia
and tactile sensory loss.
CONCLUSIONS
There was a time when neuropsychological testing was done in rela-

tive isolation with respect to other neurodiagnostic procedures. In fact,
neuropsychological testing measures utilized prior to 1970 were as sen-
88 ERIN D. BIGLER
sitive (or more sensitive) and less invasive than other techniques for as-
sessing "organic" dysfunction (Small, 1980). However, with the diag-
nostic improvements in neuroradiology and electrophysiology, as well as
the better understanding of minor neurological abnormalities in relation
to a variety of neurobehavioral disorders, neuropsychology must continue
to integrate its clinical findings in relation to these other areas. To that
end, this chapter has attempted to integrate neuropsychological test find-
ings in children with other assessment techniques. Despite certain limita-
tions, particularly with respect to children, neuropsychological assess-
ment techniques remain the best functional assessment tools for the
evaluation of higher cortical functioning. They provide a crucial perspec-
tive that should be integrated with other neurodiagnostic and psychologi-
cal procedures in gaining a comprehensive understanding of the neu-
rologically impaired child.
REFERENCES
Beery, K. E. (1967). Developmental Test of Visual-Motor Integration. Chicago: Follett Educa-
tional Corporation.
Bender, 1. A. (1938). A visual motor gestalt test and its clinical use (Research Monographs,
3). New York: American Orthopsychiatric Association.
Bigler, E. D. (1980). Neuropsychological assessment and brain scan results: A case study
approach. Clinical Neuropsychology, 2, 13-24.
Bigler, E. D. (1984). Diagnostic clinical neuropsychology. Austin: University of Texas Press.
Bigler, E. D., & Ehrfurth, J. W. (1981). The continued inappropriate singular use of the Bender
Visual Motor Gestalt Test. Professional Psychology, 12, 562-569.
Bigler, E. D., Hubler, D. W., Cullum, C. M., & Turkheimer, E. (1985). Intellectual and memory
impairment in dementia: CT volume correlations. Journal of Nervous and Mental Dis-
ease, 173. 347-352.
Bigler, E. D., & Naugle, R. I. (1984). Case studies in cerebral plasticity. International Journal
of Clinical Neuropsychology, 7, 12-23.
Boll, T., & Barth, J. T. (1981). Neuropsychology of brain damage in children. In S. B. Filskov
& T. J. Boll (Eds.)' Handbook of clinical neuropsychology (pp. 418-452). New York:
Wiley.
Brandeis, D., & Lehmann, D. (1986). Event-related potentials of the brain and cognitive
processes: Approaches and applications. Neuropsychologia, 24, 151-168.
Brown, S. B. (1982). Neurologic examination of the older child. In K. F. Swaiman & F. S.
Wright (Eds.), The practice of pediatric neurology (pp. 110-129). St. Louis: C. V. Mosby.
Capute, A. J., Neidermeyer, E. F. 1., & Richardson, F. (1968). The EEG in children with
minimal cerebral dysfunction. Pediatrics, 41,1104-1114.
Carr, A. C. (1985). Psychological testing of personality. In H. I. Kaplan & B. J. Sadock (Eds.),
Comprehensive textbook of psychiatry (4th ed., pp. 514-535). New York: Williams &
Wilkins.
Courchesne, E., Kilman, B. A., Galambos, R., & Lincoln, A. J. (1984). Autism: Processing of
novel auditory information assessed by event-related potentials. Electroencephalogra-
phy and Clinical Neurophysiology: Evoked Potentials, 59, 238-248.
Cullum, C. M., & Bigler, E. D. (1986). Ventricle size, cortical atrophy and the relationship
with neuropsychological status in closed head injury: A quantitative analysis. Journal of
Clinical and Experimental Neuropsychology, 8, 437-452.
Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1986). The California Verbal Learning
Test. Orlando, FL: Harcourt Brace Jovanovich.
Dileo, J. H. (1983). Interpreting children's drawings. New York: Brunner/Maze!'
Duckett, S. (1981). Neuropathological aspects: I. Congenital malformations. In P. Black (Ed.),
Brain dysfunction in children: Etiology. diagnosis and management (pp. 17-46). New
York: Raven Press.
Duffy, F. H., Denckla, M. B., Bartels, D. H., & Sandini, G. (1980). Dyslexia: Regional dif-
ferences in brain electrical activity by topographic mapping. Annals of Neurology, 7,
412-420.
Duffy, F. H., & McAnulty, G. B. (1985). Brain electrical activity mapping (BEAM): The search
for a physiological signature of dyslexia. In F. H. Duffy & N. Geschwind (Eds.), Dyslexia:
A neuroscientific approach to clinical evaluation (pp. 105-122). Boston: Little, Brown.
Dworkin, P. H. (1985). Learning and behavioral problems of schoolchildren. Philadelphia:
W. B. Saunders.
Gaddes, W. H. (1985). Learning disabilities and brain function (2nd ed.). New York: Spring-
er-Verlag.
Galaburda, A. M. (1984). Anatomical asymmetries. In N. Geschwind & A. M. Galaburda
(Eds.), Cerebral dominance (pp. 11-25). Cambridge, MA: Harvard University Press.
Golden, G. S. (1982). Neurobiological correlates of learning disabilities. Annals of Neu-
rology, 12,409-418.
Harris, D. B. (1963). Children's drawings as a measure of intellect. New York: Harcourt Brace
& World.
Hartlage, P. L., & Krawiecki, N. S. (1984). The neurological assessment of children. In R. E.
Tarter & G. Goldstein (Eds.), Advances in clinical neuropsychology (Vol. 2, pp. 63-77).
Heiss, W. D., Herholz, K., Pawlik, G., Wagner, R., & Wienhard, K. (1986). Positron emission
tomography in neuropsychology. Neuropsychologia, 24, 141-149.
Herskowitz, J., & Rosman, N. P. (1982). Pediatrics, neurology and psychiatry-Common
ground. New York: Macmillan.
Hughes, J. R. (1971). EEG and learning disabilities. In H. R. Myckleburst (Ed.), Progress in
learning disabilities (Vo!. 2, pp. 196-223). New York: Grune and Stratton.
Hughes, J. R., & Denckla, M. B. (1978). Outline of a pilot study of electroencephalographic
correlates of dyslexia. In A. L. Benton & D. Pearl (Eds.), Dyslexia: An appraisal of
current knowledge (pp. 217-245). New York: Oxford University Press.
John, E. R., Karmel, B. Z., Corning, W. D., Easton, P., Brown, D., Ahn, H., John, M., Harmony,
T., Prichep, L., Toro, A., Garson, I., Bartlett, F., Thatcher, R.,Kaye, H., Valdes, P. &
Schwartz, E. (1977). Neurometrics. Science, 196. 1393-1410.
Kaufman, A. S., & Kaufman, N. L. (1983). Kaufman Assessment Battery for Children. Circle
Pines. MN: American Guidance Service.
Kaufman, A. S., Long, S. W., & O'Neal, M. R. (1986). Topical review of the WISC-R for
pediatric neuroclinicians. Journal of Child Neurology. 1,89-98.
Kaufman, D. M. (1981). Clinical neurology for psychiatrists. New York: Grune and Stratton.
Kemper, T. L. (1984). Asymmetrical lesions in dyslexia. In N. Geschwind & A. M. Galaburda
(Eds.), Cerebral dominance (pp. 75-89). Cambridge, MA: Harvard University Press.
Kertesz, A. (1983). Localization in neuropsychology. New York: Academic Press.
Kiloh, L. G., McComas, A. J., & Osselton, J. W. (1979). Clinical electroencephalography.
London: Butterworths.
Klorman, R., Salzman, L. F., Bauer, L. 0., Coons, H. W., Borgstedt, A. D., & Halpern, W. I.
(1983). Effects of two doses of methylphenidate on cross-situational and borderline
hyperactive children's evoked potentials. Electroencephalography and Clinical Neu-
rophysiology, 56, 169-185.
Krahn, G. L. (1985). The use of projective assessment techniques in pediatric settings. Jour-
nal of Pediatric Psychology, 10. 179-193.
90 ERIN D. BIGLER
Lacks, P. (1984). Bender Gestalt screening for brain dysfunction. New York: Wiley-Inter-
science.
Lezak, M. D. (1983). Neuropsychological assessment. New York: Oxford University Press.
Luders, H., Dinner, D. S., Lesser, R P., & Morris, H. H. (1986). Evoked potentials in cortical
localization. Journal of Clinical Neurophysiology, 3, 75-84.
Luria, A. R (1973). The working brain. London: Penguin Press.
Morris, J. M., & Bigler, E. D. (1987). Hemispheric functioning and the K-ABC: Results in
neurologically impaired children. Developmental Neuropsychology, 3, 67-79.
Nauta, W. J. H., & Feirtag, M. (1986). Fundamental neuroanatomy. New York: W. H.
Freeman.
Nussbaum, N. L., & Bigler, E. D. (1986). Neuropsychological and behavioral profiles of
empirically derived subgroups of learning disabled children. International Journal of
Nussbaum, N. L., Bigler, E. D., & Koch, W. (1986). Neuropsychologically derived subgroups
of learning disabled children: Personality/behavioral dimensions. Journal of Research
and Development in Education, 19, 57-67.
O'Donohoe, N. V. (1981). Epilepsies of childhood. London: Butterworths.
Oldendorf, W. H. (1980). The quest for an image of brain. New York: Raven Press.
PassIer, M. A., Isaac, W., & Hynd, G. W. (1985). Neuropsychological development of behav-
ior attributed to frontal lobe functioning in children. Developmental Neuropsychology,
1, 349-370.
Raven, J. C. (1965). The Coloured Progressive Matrices Test. London: H. K. Lewis.
Rie, H. E., & Rie, E. D. (1980). Handbook of minimal brain dysfunction. New York: Wiley.
Risberg, J. (1986). Regional cerebral blood flow in neuropsychology. Neuropsychologia, 24,
135-140.
Riva, D., & Cazzaniga, L. (1986). Late effects of unilateral brain lesions sustained before and
after age one. Neuropsychologia, 24, 423-428.
Rourke, B. P., Bakker, D. J., Fisk, J. L., & Strang, J. D. (1983). Child neuropsychology. New
Schimschock, J. R, Milford-Cooley, M., & Cooley, N. (1984). Practical management of chil-
dren with apparent learning disabilities. Neurologic clinics. Philadelphia: W. B.
Saunders.
Scott, L. H. (1981). Measuring intelligence with the Goodenough-Harris Test. Psychological
Bulletin, 89, 483-505.
Shafer, S. Q., Shaffer, D., O'Connor, P. A., & Stokman, C. J. (1983). Hard thoughts on neu-
rological "soft signs." In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 133-
Shaffer, D., O'Connor, P. A., Shafer, S. Q., & Propis, S. (1983). Neurological "soft signs":
Their origins and significance for behavior. In M. Rutter (Ed.), Developmental neurop-
sychiatry (pp. 144-163). New York: Guilford Press.
Shaffer, D., Schonfeld, 1., O'Connor, P. A., Stokman, C., Trautman, P., Shafer, S., & Ng, S.
(1985). Neurological soft signs, their relationship to psychiatric disorder and intel-
ligence in childhood and adolescence. Archives of General Psychiatry, 42, 342-352.
Shaywitz, S. E. (1982). The Yale Neuropsychoeducational Assessment Scales. Schizo-
phrenia Bulletin, 8, 360-424.
Small, L. (1980). Neuropsychodiagnosis in psychotherapy (2nd ed.). New York: Brun-
ner/Mazel.
Touwen, B. C. L. (1979). Examination of the child with minor neurological dysfunction (2nd
ed.). Philadelphia: J. B. Lippincott.
Van Allen, M. W. (1969). Pictorial manual of neurologic tests. Chicago: Year Book Medical
Publishers.
Visser, S. L., Njiokiktjien, Ch. J., & De Rijke, W. (1982). Neurological condition at birth in
relation to the electroencephalogram (EEG) and visual evoked potential (VEP) at the age
of 5. Electroencephalography and Clinical Neurophysiology, 54, 458-468.
Wechsler, D. (1949). Wechsler Intelligence Scale for Children. New York: Psychological
Corporation.
Wechsler, D. (1974). Wechsler Intelligence Scale for Children-Revised. New York: Psycho-
logical Corporation.
Whelihan, W. M., & Lesher, E. L. (1985). Neuropsychological changes in frontal functions
with aging. Developmental Neuropsychology, 1, 371-380.
Williams, R., & Caviness, V. S. (1984). Normal and abnormal development of the brain. In R.
E. Tartar & G. Goldstein (Eds.), Advances in clinical neuropsychology (Vol. 2, pp. 1-62).
4
Neuropsychological Diagnosis with

Children
Actuarial and Clinical Models
W. GRANT WILLIS
Neuropsychological diagnosis with children is a highly specialized en-

deavor. It draws heavily from disciplines that encompass neuropsychol-
ogy, functional psychodiagnosis, and pediatrics. It also departs from these
established disciplines in some important ways. For example, unique
issues in child neuropsychological diagnosis such as referral bias, assess-
ment technique, and poor taxonomy distinguish it from functional
psychodiagnosis, especially with adults. These issues, which serve to
militate against high degrees of diagnostic accuracy, have been well ar-
ticulated by Achenbach (1985), Fletcher and Taylor (1984), and Rapoport
and Ismond (1984).
In terms of diagnostic accuracy, it is well established that actuarial
(Le., statistical) models are superior to clinical models (e.g., Goldberg &
Werts, 1966; Meehl, 1973; Phelan, 1964; Wallach & Schoof, 1965). As
nearly all practicing clinicians can attest, however, the solution is not so
simple as to abandon clinical models in favor of actuarial ones. Rather,
both models of diagnosis are not only necessary but also desirable. Thus,
the focus in this chapter is to discuss the assets and liabilities of both of
these models with special reference to the unique issues encountered in
child neuropsychological diagnosis.
W. GRANT WILLIS Department of Psychology, University of Rhode Island, Kingston,

Rhode Island.
93
94 W. GRANT WILLIS
DIAGNOSTIC MODELS
The essential distinction between actuarial and clinical models of

diagnosis is in terms of the ways that assessment data are combined in
order to aid clinicians in arriving at some kind of decision (usually a
diagnostic prediction). For actuarial models, multivariate statistical tech-
niques are used to combine data; for clinical models, professional judg-
ments are used for this task. A common misconception is that actuarial
models require objective psychometric scores as data whereas clinical
models allow for nonparametric clinical judgments. The kinds of assess-
ment data that are collected and the methods that are used for data collec-
tion, however, are irrelevant to the distinction between actuarial and
clinical models. Clinical interview data and nominal child characteristics
are legitimate input for actuarial diagnostic rules, and standardized psy-
chometric scores are easily (albeit sometimes inaccurately) incorporated
into clinical judgments.
The relative superiority of one model over the other in terms of asso-
ciated diagnostic accuracy has been studied and debated for many years
(see Meehl, 1954). As already noted, the evidence at this point is clear:
Actuarial models yield consistently higher probabilities of accurate diag-
nostic predictions than do clinical models, even when those predictions
are made by highly experienced clinicians. The reason for this, however,
cannot be due to the kinds of assessment data that are used in the two
models; both models are operative with the same pool of information.
Rather, the essential difference between these models centers on methods
of integrating assessment data.
There are at least three important distinctions between actuarial and
clinical models in terms of data integration. A thorough understanding of
these distinctions can help to highlight some sources of inaccuracy com-
monly associated with clinical decision making.
Data-Diagnosis Contingency
In order to know how useful a particular source of assessment data is
in terms of arriving at some diagnosis, clinicians must first understand the
relationship (Le., contingency) between those data and the diagnosis. For
actuarial models, the contingency between assessment data and the diag-
nostic decision is estimated through statistical techniques. These tech-
niques are typically multivariate in nature and may include, for example,
discriminant analysis, multiple regression, and cluster analysis. Because
data-diagnosis contingencies are established empirically, the probability
of making an accurate diagnostic prediction can be determined and subse-
quently evaluated by the clinician. In contrast, for clinical models, the
contingency between assessment data and the diagnostic decision is esti-
NEUROPSYCHOLOGICAL DIAGNOSIS WITH CHILDREN 95
mated by the clinician on the basis of factors such as experience and

training. Because data-diagnosis contingencies are established subjec-
tively, probabilities of diagnostic accuracy are assumed rather than deter-
mined and are not, therefore, subject to external evaluation.
In order to establish contingencies between assessment data and diag-
noses, clinicians are forced to rely on their own judgment concerning the
degree to which a particular combination of neuropsychological signs
found on examination of the patient matches some neuropsychological
syndrome or disorder. This is an arduous task because relying on profes-
sional experience and training (regardless of extent) is likely to bias judg-
ment owing to selective retrievability of previous cases. Selective re-
trievability is associated with bias because it is affected by factors such as
salience and familiarity, as opposed to more appropriate factors (in the
sense of leading to more accurate diagnoses) such as the prevalence of a
particular neuropsychological syndrome or disorder in some population
of interest (d. Tversky & Kahneman, 1973).
Consideration of Data
A second distinction between the models concerns the ways in which
different sources of assessment data are considered. For actuarial models,
data are considered simultaneously whereas in clinical models, data are
typically considered independently. Because different sources of assess-
ment data are usually interrelated (i.e., they are not truly independent),
each source does not contribute entirely unique information to the diag-
nostic prediction. Thus, independent consideration of data frequently
leads to diagnostic errors.
For example, if the degree of contingency between a particular neuro-
psychological sign (e.g., a test result) and a diagnosis is 0.7, then, in the
absence of any other information about the patient, the clinician knows
0.7 2 x 100% or 49% of the total amount of information that is necessary in
order to make a completely accurate diagnosis. It does not follow, howev-
er, that given another neuropsychological sign (e.g., age-inappropriate
gait), for which the contingency with the diagnosis is say 0.4, that the
clinician now knows 49% + 0.4 2 X 100%, or 65%, of the information
required for certainty with respect to the diagnosis. Rather, the actual
proportion of diagnostic information that the clinician has obtained is
reduced by an amount that corresponds to the degree that the neuropsy-
chological signs are themselves interrelated. These intercorrelations are
accounted for when data integration is accomplished statistically. When
data integration is accomplished clinically, however, the intercorrelations
among neuropsychodiagnostic signs must be estimated. The difficulty of
these estimations, of course, increases dramatically with the number of
signs considered and is rarely undertaken in clinical practice.
96 W. GRANT WILLIS
An undesirable outcome of independent consideration of signs is that

two highly correlated sources of assessment data may be weighted doubly
in terms of their contributions to a diagnostic prediction. Moreover, when
assessment data are highly congruent (i.e., correlated), clinicians often
tend to express higher degrees of confidence in diagnoses than when such
data are less dependent. It is much more sobering to view these situations
in terms of redundancy as opposed to congruence. When they are viewed
in the context of redundancy, it is clear why diagnoses that are based on
simultaneously considered sources of assessment data are more likely to
be accurate than sources that are overly weighted owing to independent
consideration.
Relative Importance of Data

A final distinction between actuarial and clinical models concerns
the relative importance (i.e., weights) attributed to particular assessment
data across cases. For actuarial models, assessment data are weighted
consistently across similar cases (where those weights are partially deter-
mined by the amount of diagnostic variance for which a datum accounts).
In contrast, for clinical models, clinicians have the freedom to use assess-
ment data differentially depending on factors that are unique to the case
in point. For example, a result of an IQ test might be weighted relatively
less in terms of its contribution to the diagnosis of a minority child than of
an Anglo child; an assessed constructional apraxia might be weighted
more in terms of its contribution to the diagnosis of a child with prior
formal educational experience than to the diagnosis of a child without
such experience.
This kind of differential weighting of signs is often couched in expla-
nations regarding equality and nondiscriminatory assessment practices.
The implicit assumption made when differential weights are assigned to
individual sources of data contingent on the presence of putative moder-
ator variables (e.g., race, educational experience), however, is that those
sources of data are differentially predictive of the diagnosis depending on
the moderator variable (Thorndike, 1982). In the absence of empirical
verification of such differential predictability, assigning different weights
to assessment data is more likely to introduce bias into the assessment
rather than to eliminate it (Jensen, 1973; Mercer & Lewis, 1978; Mitchell,
1967; Scarr-Salapatek, 1971). This may be due, in part, to the phe-
nomenon of illusory correlation that is common in situations where clini-
cians are required to perceive the covariation between assessment data
and diagnoses without the benefit of empirical techniques (Chapman,
1967; Chapman & Chapman, 1967, 1969; Golding & Rorer, 1972; Kurtz &
Garfield, 1978; Lueger & Petzel, 1979; Starr & Katkin, 1969).
ACTUARIAL MODELS
For actuarial models of diagnosis, rules (i.e., equations) are devel-

oped that specify how assessment data are to be combined. Applications
of these rules result in empirically determined representations of diag-
nostic predictions or some other outcome for individual patients. Rules
are typically multivariate (however, d. Goldstein & Shelly, 1973). This
means that several sources of assessment data can be considered concomi-
tantly. Perhaps because of the relatively large amounts of assessment data
that are typically collected for neuropsychological evaluations, many ac-
tuarial diagnostic models have been proposed, but mainly for adult popu-
lations. As the discipline of child neuropsychology continues to mature,
it is reasonable to expect that diagnostic applications of the model to this
population will also increase. Currently, the model is popular in the child
psychopathology literature; however, most of the reports employing the
model with child neuropsychological cases are exploratory rather than
diagnostic at the present time.
Methods
There are numerous statistical methods that can be used to derive
actuarial diagnostic rules, but three of the most popular methods are
discriminant analysis, multiple regression, and cluster analysis (see
Huberty, 1975, 1984; McDermott, 1982; Pedhazur, 1982; Tryon & Bailey,
1970). In all of these statistical methods, the variability among multiple
sources of assessment data is analyzed.
In order to use discriminant analysis and multiple regression meth-
ods to derive an actuarial diagnostic rule, particular diagnostic outcomes
must be specified prior to the analysis. For discriminant analysis meth-
ods, these outcomes are scaled in nominal units (e.g., diagnostic classifi-
cations), and the result of the analysis is an equation that indicates how
the various sources of assessment data should be weighted and combined
in order to discriminate maximally among the outcomes. Subsequently,
clinicians can evaluate an empirically derived probability that suggests
how closely a particular pattern of assessment data collected from a pa-
tient corresponds to similar data collected from some previously diag-
nosed group. For multiple regression methods, outcomes may be scaled in
metric units; consequently, degrees of some criterion (e.g., level of impair-
ment or prognostic information) can be predicted from the combined
assessment data. The result of a multiple regression analysis is an equa-
tion that indicates how the various sources of assessment data should be
weighted and combined in order to best predict the degree of some out-
come of interest.
98 W. GRANT WILLIS
For cluster analysis methods, outcome variables such as diagnostic

classification and prognostic information are not specified prior to the
analysis. Instead, multiple sources of assessment data from each of a large
number of patients are analyzed in order to determine if there are sub-
groups of patients that demonstrate similar patterns of assessment results.
These subgroups are determined by successive comparisons in which
patterns of patient assessment data are clustered such that, with reference
to these data, the variability within subgroups is minimized whereas the
variability among subgroups is maximized. Subsequently, similar to dis-
criminant analysis, a clinician can evaluate how closely a particular pat-
tern of assessment data collected from an individual patient corresponds
to the pattern demonstrated by some natural subgroup. This method is
essentially that which Achenbach (see Achenbach & Edelbrock, 1978)
employed in order to establish empirical diagnostic categories in the field
of child psychopathology. The method also has been readily embraced by
researchers who are exploring potential relationships between specific
learning disabilities in children and neuropsychological status (see Hoop-
er & Willis, in press).
Given the relative assets of actuarial models of diagnostic decision
making, it logically follows that diagnostic accuracy in child neuropsy-
chology would improve as a function of the availability and application of
actuarial rules. Responsible use of actuarial rules, however, demands crit-
ical evaluation. Clinicians should evaluate the technical aspects of actu-
arial rule derivation just as carefully as they would evaluate, for example,
the psychometric properties of a test prior to its adoption in a case. In this
respect, three important issues are (a) base rates in particular populations,
(b) validity of criteria, and (c) stability of rules.
Base-Rate Considerations
When an actuarial diagnostic rule is published, the hit rate is usually
presented as an index of that rule's accuracy. For example, Selz and
Reitan (1979; see also Reitan, 1984) developed an actuarial rule to classify
children aged 9 through 14 as normal, learning-disabled, or brain-dam-
aged; their data are presented in Table 1. There were 25 children in each
group who had been diagnosed independently according to the criteria for
these categories. Using only the actuarial rule, 24 of the 25 normal chil-
dren were correctly classified, 14 of the 25 learning-diasbled children
were correctly classified, and 17 of the 25 brain-damaged children were
correctly classified. Thus, from a total of 75 children, 24 + 14 + 17, or
55/75, were correctly classified; the hit rate in this case was 73.3%.
In order to evaluate a particular hit rate appropriately, it must be
compared to a base rate. In the example, the base rates for each of the three
diagnostic classifications were 25/75, or approximately 33.3%. This
TABLE 1. Classification Accuracy Table a
Actual diagnosis
Prediction Normal Learning-disabled Brain-damaged
Normal 24 8 4
Learning-disabled 1 14 4
Brain-damaged 0 3 17
aFrom Selz and Reitan (1979). Copyright 1979 by the American Psychological
Association. Adapted by permission.
means that if a clinician were to diagnose every patient in this population

as, for example, brain-damaged, that diagnosis would be accurate for ap-
proximately 33.3% of the cases and inaccurate for approximately 66.7% of
the cases. Because base rates establish some chance criterion of diagnostic
accuracy, actuarial rules must, at the minimum, improve on those proba-
bilities. Thus, the hit rate of 73.3% noted in the example would be com-
pared with the base rate of 33.3% in order to determine improvement over
chance.
In order to assess the significance of a difference between a hit rate
and a base rate, clinicians can apply one of two criteria proposed by
Huberty (1984). For one assessment, an overall hit rate is compared with
the base rate for the largest diagnostic group. In the example taken from
Selz and Reitan (1979), because the diagnostic groups are of equal size
(i.e., 25 each), their associated base rates, of course, are also equal (i.e.,
33.3% each); however, in many other instances base rates are not equal. In
such instances, comparisons are made against the highest base rate. A
standardized normal test statistic (i.e., z) for hit-rate-base-rate com-
parisons is
(o-e)VN
z= (1)
VerN - e)
where 0 is the observed number of hits (ef. hit rate), e is the expected
number of hits (i.e., the number of cases in the largest diagnostic group; d.
base rate), and N is the total number of cases in the sample. Substituting
values in the example taken from Selz and Reitan for variables in Equa-
tion 1, z = 7.35 (p < .001). Thus, the hit rate obtained from using the
actuarial rule significantly improved on the chance assignment of patients
to diagnostic classifications.
Huberty (1984) also proposed a different kind of comparison that is
useful when clinicians are interested in classification accuracy for a par-
ticular diagnostic group. For example, one might be interested in how
100 W. GRANT WILLIS
accurately a brain-damaged child might be identified by using the actu-

arial rule proposed by Selz and Reitan (1979). A standardized normal
statistic for this comparison is
z = (ngg - eg) Vn~

(2)
Veg(n g - eg)
where ngg is the number of correctly identified patients within the diag-
nostic group of interest (d. 0, hit rate), e" is the number of patients within
that group expected to be diagnosed correctly by chance (calculated as
ng2!N; d. e, base rate), and ng is the number of patients within the diag-
nostic group of interest (d. N). Substituting values from Selz and Reitan
for variables in Equation 2, z = 3.69 (p < .001). Thus, the accuracy associ-
ated with identifying brain-damaged children was significantly better
than chance.
Finally, Huberty (1984) further proposed an index that is useful in
determining the degree to which an actuarial rule can improve diagnostic
accuracy over chance levels. This index is
(3)
where Ho is the observed hit rate, and He is the hit rate expected by chance
(i.e., base rate). This improvement-over-chance index (I) can be used in
the context of either Equation 1 or 2. For example, Is = 0.60 and 0.52, for
the overall and the brain-damaged classification comparisons, respec-
tively. Thus, the use of the actuarial rule proposed by Selz and Reitan
(1979) improved diagnostic accuracy overall by 60% and within the brain-
damaged group by 52%.
Although comparisons between base rates and hit rates are relatively
straightforward and were, in fact, proposed by Meehl and Rosen (1955)
over three decades ago, they are frequently absent from current reports of
neuropsychological actuarial rules. For example, when evaluated in this
context, published neuropsychological actuarial rules often do not signifi-
cantly improve on chance levels of diagnostic accuracy; worse yet, many
yield a level of diagnostic accuracy that is significantly poorer than
chance (see Willis, 1984). Thus, it is important for clinicians to evaluate
actuarial rules in terms of hit-rate to base-rate comparisons in order to
make informed decisions about the utility of those rules.
Validity of Diagnostic Criteria

Another methodological issue that must be considered concerns the
validity of the criteria (e.g., for diagnostic or prognostic outcomes) that
were used to establish the actuarial rule. The validity of any actuarial rule
is necessarily limited by the validities of the criteria used to establish that
rule. Moreover, the validities of those criteria are necessarily limited by
their associated reliabilities. The problem is compounded in the case of
children because of the often invasive nature of definitive diagnostic cri-
teria for neuropsychological disorders, in addition to the limited data base
describing the course and sequelae of particular disorders. Further, given
the special circumstance that child (versus adult) psychological symp-
toms are usually pathological only in the context of developmental de-
lays, diagnostic criteria used to establish actuarial rules for children
should be referenced against particular age-level expectations.
The level of inference adopted by researchers who develop actuarial
rules is another consideration of some importance with reference to valid-
ity. For example, particular subtypes of specific developmental disorders
(SDD, d. specific learning disabilities, SLD) are often assumed to be
organically (neuropsychologically) based. This assumption may be at
least partially predicated on the (perhaps erroneous) idea that if a particu-
lar form of observable neurological impairment can cause severe forms of
a disorder such as epilepsy or cerebral palsy, then milder forms of such
impairment may result in milder forms of disorders such as hyperactivity
or SDDISLD (Fletcher & Taylor, 1984; however, see Rosen, Sherman, &
Galaburda, 1986). It is often the case that groups of children of varying
ages are diagnosed as SDDISLD according to DSM-III-R (American Psychi-
atric Association, 1987) or federal, state, and local educational agency
criteria (primarily on the basis of psychoeducational assessment), and
that such children are subsequently subtyped or differentiated from oth-
ers on the basis of patterns of neuropsychological assessment data (Fisk &
Rourke, 1979, 1983; Fletcher, Smidt, & Satz, 1979; Lyon, Stewart, & Freed-
man, 1982; Selz & Reitan, 1979). Two limitations inherent in these ap-
proaches are that (a) the validity of the external SDDISLD diagnosis may
be, in itself, questionable (see Shepard, Smith, & Vojir, 1983) and (b)
because the diagnosis of these disorders is contingent on developmental
comparisons, merging diagnosed children of greatly different chronologi-
cal ages for the purpose of statistical analysis is probably inappropriate.
Stability and Generalizability of Actuarial Rules

Even when the criteria upon which actuarial rules are based are valid
(and therefore also reliable), the actuarial rule can still lack validity. This
is because the reliabilities of rules, per se, are relatively independent of
criterion-related considerations. Thus, to the extent that actuarial rules
lack reliability, they also lack validity. Actuarial rules are derived by
analyzing the variability among multiple sources of assessment data via
powerful multivariate methods. These methods, however, capitalize on
102 W. GRANT WILLIS
all sources of variability present, even unspecified sources due to random

factors (Le., error). It is entirely possible, therefore, that, even when care-
ful attention has been paid to issues concerning criterion validity, a par-
ticular actuarial rule could classify patients primarily on the basis of
random variation among assessment data.
The problem is intensified in the clinical specialty of child neuropsy-
chology because rules are developed typically with relatively large
amounts of assessment data collected from each patient and relatively
small numbers of patients in distinct diagnostic criterion groups. These
circumstances potientiate the possibility that resulting actuarial rules can
be largely determined on the basis of chance. For example, when the ratio
of the number of patients to the number of assessment measures is low
(Le., approaches 1), application of the actuarial rule to new patients is
associated with a higher probability of a diagnostic error than when that
ratio is relatively higher (Fletcher, Rice, & Ray, 1978).
In order to address this problem properly, actuarial rules should be
cross-validated. In the context of actuarial models of diagnosis, cross-
validation is a procedure in which an actuarial rule, derived with one
group of patients, is subsequently applied to a second (different) group of
patients. The actuarial diagnostic or prognostic predictions from this sec-
ond group of patients are then compared with independently determined
diagnoses or prognoses. An index of relationship between the predicted
and independent outcomes is then determined. This index is lower than
that between the aggregate of assessment data and the predicted outcome
in the initial analysis, and is a more stable, representative estimate of the
true data-outcome contingency.
Many neuropsychological actuarial rules are not cross-validated; as
such, the established hit rates are spuriously high owing to multivariate
capitalization on random variation among assessment data. This meth-
odological issue was well illustrated by Willson and Reynolds (1982),
who conducted a secondary analysis of nine reports (a majority with child
subjects) of actuarial rules for diagnosis. After estimates of cross-valida-
tion were made, only about half of the original 12 indexes of relationship
remained statistically significant. Clearly, clinical applications of actu-
arial rules that have not been cross-validated are premature. In the
absence of actual cross-validation, which often may be difficult to accom-
plish because of low numbers of patients in diagnostic groups, estima-
tions can be made (see Cattin, 1980; Herzberg, 1969; Lachenbruch, 1967;
Mosier, 1951).
CLINICAL JUDGMENT
There are at least three assumptions inherent in the responsible use of

actuarial rules for child neuropsychological diagnosis. The first is that
these rules are free from the methodological problems described. The
second is that actuarial rules are readily available for clinical evaluation.
The third is that when diagnostic errors are made (as they inevitably will
be), the costs associated with different kinds of error are constant both
across cases and for an individual patient (or at least match those esti-
mated by the developer of the rule). In reality, certain kinds of error are
much more costly (Le., serious) than others in terms of a variety of criteria.
Moreover, the current state of the field is such that, with children, the first
two assumptions would often be violated as well.
In situations where these assumptions are violated, child neuro-
psychologists cannot depend on actuarial rules to guide diagnostic deci-
sion making. Instead, assessment data must be aggregated clinically, and,
as such, without thoughtful consideration of the cognitive processes in-
volved, decisions are subject to a high degree of error. Moreover, even
when methodologically sound actuarial rules are available, clinicians ide-
ally will use these models of data integration as guides to help clarify the
diagnostic or prognostic decision-making process, as opposed to an ex-
clusive dependence on the application of rules per se. In the final analy-
sis, of course, clinicians diagnose patients; rules and assessment data do
not. The ultimate responsibility for accurate decisions lies with clinicians
rather than with the tools they choose to help them think. As already
noted, however, thoughtful clinical judgment demands systematically or-
ganized thinking. An examination of how interpretive strategies, debias-
ing techniques, assessment design, and decision rules can be useful in
this respect follows.
Interpretive Strategies
Clinical decisions are often biased (and therefore erroneous) owing to
the inconsistent use of particular strategies for interpreting assessment
data (see also McDermott, 1981). Two major interpretive strategies are
normative and ipsative. For normative strategies, each source of assess-
ment data collected from a patient is compared with a reference group,
and marked discrepancies from this normative distribution are consid-
ered as abnormal. In contrast, for ipsative strategies, the patient's own
baseline (or average) level of functioning is taken as the reference point
against which specific areas of functioning are compared. The latter in-
terpretive strategy has become increasingly more widespread in recent
years, especially in the pediatric literature (e.g., Kaufman & Kaufman,
1983; Reynolds, 1982; Reynolds & Clark, 1983; Reynolds & Gutkin, 1981;
Reynolds & Kaufman, 1985).
These two interpretive strategies are well illustrated by Reitan's (e.g.,
Reitan & Davison, 1974) approach to neuropsychological test interpreta-
tion, which comprises four methods of inference (see also Selz, 1981).
104 W. GRANT WILLIS
Two of these methods are distinctly normative whereas the other two are
ipsative. The two normative strategies concern level of performance and
pathognomonic signs. Level-of-performance measures (e.g., numbers of
errors on the Category Test) are compared against norms established for
children of similar age levels in order to determine deficiencies in perfor-
mance abilities. Although useful in that context, these measures are simi-
larly limited, as are other kinds of normative data in child neuropsycho-
logical evaluations, because of difficulties in distinguishing developmen-
tal delays from deficits. Such distinctions are important in terms of diag-
nosis and prognosis as well as treatment planning. Additionally, premor-
bid levels of functioning are neglected sources of information when as-
sessment data are considered primarily in a normative sense. Thus, an
average level of performance may be interpreted mistakenly as asymptom-
atic in those cases where a premorbid level was actually superior. Pathog-
nomonic signs (e.g., errors on Imperception Tests) are also normative
sources of data in the sense that their presence is compared with that of
some reference group. They are less susceptible to interpretive errors,
however, because such signs can be (but are not always) symptomatic
independent of developmental considerations. The two ipsative strategies
concern pattern-of-performance measures (e.g., a particular configuration
of test scores) and right-left differences (e.g., differences in finger-tapping
rates between the hands). Both of these strategies involve intraindividual
as opposed to normative group comparisons.
Most ipsative strategies are predicated on the assumption that each
source of assessment data possesses an adequate degree of specificity for
individual interpretation. In this context, specificity refers to that propor-
tion of the variance of a particular source of assessment data that is reli-
able and unique (i.e., is independent of other sources). If a particular
source of assessment data lacks specificity, then it should not ordinarily
be interpreted in the typical ipsative sense as representing a unique func-
tion. Ipsative interpretations of most subtests from the Kaufman Assess-
ment Battery for Children (Kaufman & Kaufman, 1983) and the Wechsler
Intelligence Scale for Children-Revised (Wechsler, 1974), for example, are
well established strategies because adequate degrees of subtest specificity,
for the most part, have been documented (Kaufman & Kaufman, 1983;
Reynolds & Kaufman, 1985).
When clinicians are consistent in applying interpretive strategies
(whether normative, ipsative, or both) to assessment data, they are less
likely to bias diagnostic and prognostic decisions. For example, an initial
prejudicial choice of a particular approach (either for an individual pa-
tient or for a particular source of assessment data) can help to determine
the outcome of a case. In these instances, diagnostic decisions may be
inappropriately influenced by methodological rather than more important
substantive issues.
Debiasing Techniques
Research investigating judgmental heuristics has delineated several
cognitive biases associated with diagnostic inaccuracy as well as several
possible debiasing techniques (Achenbach, 1985; Arkes, 1981). As noted
previously, one common source of bias is an inappropriate conceptualiza-
tion or assessment of symptom-diagnosis covariation. Table 2 is a pro-
totypical contingency table of the relationship between a particular symp-
tom (25-point discrepancy between the Verbal and Performance IQs on
the WISe) and neurologic insult (presence vs. absence of brain damage).
Entries in the table are estimated calculations based on data presented by
Holroyd and Wright (1965), assuming a total of 100 cases and a liberal
incidence estimate for childhood brain damage of 10%. These hypo-
thetical data show that 30% of a sample of non brain-damaged children
demonstrated a discrepancy of 25 or more points between Verbal and
Performance IQ scores and 80% of a sample of brain-damaged children
demonstrated this same symptom. Similar data (e.g., Black, 1976) are
often misinterpreted as evidence that this size of discrepancy may be
symptomatic of brain damage in children (e.g., Sattler, 1982, p. 460; how-
ever, d. Sattler, 1988, pp. 705-706).
These kinds of interpretations of symptom-diagnosis contingencies
are often made when some of the available data (e.g., the proportion of
cases in which the symptom is absent but the diagnosis is present) are
considered to be less important than others (e.g., the proportion of cases
in which both the symptom and the diagnosis are present; see Arkes &
Harkness, 1983; Nisbett & Ross, 1980). In fact, all data are equally impor-
tant in determining the degree of relationship between the symptom and
the diagnosis. An accurate assessment of this relationship for the data
presented in Table 2 indicates that the interpretation noted is perhaps not
as clear as it may seem.
Several pieces of information are required in order to determine the
probability that the diagnosis of brain damage is correct or even if this
symptom is likely to be from a brain-damaged child. This information
includes (a) prior odds, (b) likelihood ratio, and (c) posterior odds (see
Arkes, 1981). The ratio representing prior odds is simply the likelihood
TABLE 2. Symptom-Diagnosis Contingency Table
Diagnosis
Symptom Brain-damaged Non brain-damaged
2:25-point VIQ-PIQ discrepancy 8 27

<25-point VIQ-PIQ discrepancy 2 63
106 W. GRANT WILLIS
that the diagnosis is correct divided by the likelihood that the diagnosis is
incorrect and is thus comparable to a base-rate index. Prior odds are
subsequently modified when information concerning a symptom is ob-
tained. Such information can be expressed as a likelihood ratio, which is
the probability of obtaining that symptom if the diagnosis were correct
divided by the probability of obtaining that symptom if the diagnosis were
incorrect. Finally, the ratio representing posterior odds is the probability
that the diagnosis is correct given the symptom and it is determined by
the product of the prior odds and the likelihood ratio.
Substituting values from Table 2, the prior odds are (8 + 2)/(27 + 63)
or 10/90. The likelihood ratio is (8/[8 + 2])/(27/[27 + 63]) or 0.8/0.3.
Consequently, the posterior odds are (10/90)(0.8/0.3) or 8/27 (Le.,0.3).
Thus, the probability that a diagnosis of brain damage is correct when
based on the 25-point VIQ-PIQ discrepancy is only 30%; the probability
that a 25-point VIQ-PIQ discrepancy is from a brain-damaged child is
only 8/(8 + 27) or 23%. Results fare even worse given more conservative
incidence estimates for childhood brain damage. When all data in con-
tingency tables are given proper consideration, symptom-diagnosis rela-
tionships can be perceived much more accurately.
In addition to this kind of Bayesian analysis of symptom-diagnosis
covariation, clinicians should entertain alternative diagnostic or prog-
nostic possibilities until all have been actively considered. This is be-
cause another source of error in clinical decision making is that, once a
decision is made, the processing of subsequent assessment data may be
biased. In this respect, clinicians should also endeavor to decrease the
degree to which they rely on memory when assessment data are being
aggregated. For example, biased decisions may result because unpre-
sented symptoms that are consistent with a particular diagnosis tend to be
remembered as having been presented, whereas symptoms that are actu-
ally presented, yet are inconsistent with the diagnosis, tend to be forgot-
ten (Arkes & Harkness, 1980). Thus, it is not surprising that the most
accurate clinicians are those who arrive at diagnostic decisions relatively
later than less accurate clinicians (Elstein, Shulman, & Sprafka, 1978).
Assessment Design and Decision Rules

Careful consideration of how the neuropsychological assessment pro-
cess is designed also can help to improve the accuracy of diagnostic and
prognostic predictions. Data should be collected from a variety of settings,
observers, and techniques in order to establish a comprehensive descrip-
tion of the neuropsychological status of the child. Additionally, clinicians
should resist the temptation to evaluate only a particular domain of func-
tioning because this may bias the ultimate decision from the outset. In this
respect, the multiaxial system outlined in DSM-III-R (American Psychi-
atric Association, 1987) is useful for designing a comprehensive evalua-
tion. It is the multiaxial process that is perhaps more useful (especially for
childhood disorders) than the DSM-III-R diagnostic codes per se.
Consider the assessment situation that, from the point of view of
accuracy, would be ideal: use of multiple techniques such as tests, observa-
tions, and interviews to collect information in each of the five DSM-III-R
domains (i.e., clinical syndromes, developmental disorders, physical con-
ditions, psychosocial stressors, global adaptive functioning). Further, a
variety of assesssors would be used in the multiple settings in which the
child interacts (e.g., home, school, clinic). This kind of design, in which
technique, domain, assessor, and setting are crossed, would serve to mini-
mize biased diagnostic and prognostic decisions by eliminating confound-
ing influences inherent in more typical assessment situations. In a more
typical situation, for example, a child suspected of being learning-disabled
might be examined by a neuropsychologist in a clinic where the focus of
that examination is primarily related to DSM-III-R Axis II domains. Infor-
mation also might be solicited from the teacher at school via rating scales
and work samples regarding academic performance, and from the mother
via an interview regarding home-related and parental concerns. Here,
potential differences among these three major sources of assessment infor-
mation could be a function of technique, domain, assessor (or informant),
setting, or some combination of factors. The confounding influences can
only be reduced to the extent that multiple techniques, domains, assessors,
and settings are used. Given a variety of constraints, idealized assessment
designs are impractical, of course, but it is incumbent that clinicians be
sensitive to the ways in which their diagnostic and prognostic decisions are
influenced by deviations from such designs.
Finally, when child neuropsychologists are able to approach evalua-
tions from a multidimensional perspective, they can expect that the large
amount of assessment data collected often will be difficult to aggregate
without the guidance of actuarial rules. Applying decision rules can be a
useful strategy in such circumstances. Decision rules help to reduce bias
by specifying a prescribed sequence of decisions that should be made, as
well as the criteria (usually empirically derived) upon which those deci-
sions should be based. McDermott and Watkins (1985) developed a com-
prehensive set of these rules for the diagnosis of childhood psychological
disorders (see also Aaron, 1981; Spitzer, Skodol, Gibbon, & Williams,
1981). Similar to multivariate actuarial rules, however, decision rules
should be carefully evaluated prior to their use because associated relia-
bility and validity issues will limit the utility of the clinician's diagnostic
prediction.
SUMMARY
There are a number of professional issues unique to the discipline of

child neuropsychological diagnosis. Many of these issues can be ad-
108 W. GRANT WILLIS
dressed via a thorough consideration of the diagnostic models used by

clinicians in order to aid decision making. For actuarial models, neuro-
psychological assessment data are integrated statistically, whereas for
clinical models, data are integrated more subjectively.
Actuarial models have been traditionally associated with higher de-
grees of diagnostic accuracy than clinical models. In addition to meth-
odological shortcomings of many actuarial rules for neuropsychological
diagnosis, however, the general availability of those rules and relative
costs of different kinds of associated errors are considerations that some-
times limit utility. There are assets and liabilities associated with both
diagnostic models. Indeed, the two models are complementary in many
respects. An integrated diagnostic approach is necessary in order to
achieve the highest possible degree of accuracy for the neuropsychologi-
cal diagnosis of children.
REFERENCES
Aaron, P. G. (1981). Diagnosis and remediation of learning disabilities in children-A neuro-
psychological key approach. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological
assessment and the school-age child (pp. 303-333). New York: Grune and Stratton.
Achenbach, T. M. (1985). Assessment and taxonomy of child and adolescent psycho-
pathology. Beverly Hills, CA: Sage.
Achenbach, T. M., & Edelbrock, C. S. (1978). The classification of child psychopathology: A
review and analysis of empirical efforts. Psychological Bulletin, 85, 1275-1301.
American Psychiatric Association. (1987). Diagnostic and statistical manual of mental dis-
orders (3rd ed., rev.). Washington, DC: Author.
Arkes, H. R (1981). Impediments to accurate clinical judgment and possible ways to mini-
mize their impact. Journal of Consulting and Clinical Psychology, 49, 323-330.
Arkes, H. R, & Harkness, A. R (1980). The effect of making a diagnosis on subsequent
recognition of symptoms. Journal of Experimental Psychology: Human Learning and
Memory, 6, 568-575.
Arkes, H. R, & Harkness, A. R (1983). Estimates of contingency between two dichotomous
variables. Journal of Experimental Psychology: General, 112, 117-135.
Black, F. W. (1976). Cognitive, academic, and behavioral findings in children with suspected
and documented neurological dysfunction. Journal of Learning Disabilities, 9, 182-187.
Cattin, P. (1980). Note on the estimation of the squared cross-validated multiple correlation
of a regression model. Psychological Bulletin, 87, 63-65.
Chapman, 1. (1967). Illusory correlation in observational report. Journal of Verbal Learning
and Verbal Behavior, 6, 151-155.
Chapman, L., & Chapman, J. (1967). Genis of popular but erroneous psychodiagnostic obser-
vations. Journal of Abnormal Psychology, 72, 193-204.
Chapman, L., & Chapman, J. (1969). Illusory correlation as an obstacle to the use of valid
psychodiagnostic signs. Journal of Abnormal Psychology, 74, 271-280.
Elstein, A. S., Shulman, A. S., & Sprafka, S. A. (1978). Medical problem solving: An analysis
of clinical reasoning. Cambridge, MA: Harvard University Press.
Fisk, J. 1., & Rourke, B. P. (1979). Identification of subtypes of learning-disabled children at
three age levels: A neuropsychological, multivariate approach. Journal of Clinical Neu-
ropsychology, 1, 289-310.
Fisk, J. L., & Rourke, B. P. (1983). Neuropsychological subtyping of learning-disabled chil-

dren: History, methods, implications. Journal of Learning Disabilities, 16, 529-531.
Fletcher, J. M., Rice, W. J., & Ray, R M. (1978). Linear discriminant analysis in neuropsycho-
logical research: Some uses and abuses. Cortex, 14, 564-577.
Fletcher, J. M., Smidt, R K., & Satz, P. (1979). Discriminant function strategies for the
kindergarten prediction of reading achievement. Journal of Clinical Neuropsychology,
1, 151-166.
Fletcher, J. M., & Taylor, H. G. (1984). Neuropsychological approaches to children: Towards
a developmental neuropsychology. Journal of Clinical Neuropsychology, 6, 39-56.
Goldberg, L. R, & Werts, C. E. (1966). The reliability of clinical judgments: A multitrait-
multimethod approach. Journal of Consulting Psychology, 30, 199-206.
Golding, S. G., & Rorer, L. (1972). Illusory correlation and subjective judgment. Journal of
Abnormal Psychology, 80, 249-260.
Goldstein, G., & Shelly, C. H. (1973). Univariate vs. multivariate analysis in neuropsycholog-
ical test assessment of lateralized brain damage. Cortex, 9, 204-216.
Herzberg, P. A. (1969). The parameters of cross-validation. Psychometrika, Monograph Sup-
plement, No. 16.
Holroyd, J., & Wright, F. (1965). Neurological implications of WISC verbal-performance
discrepancies in a psychiatric setting. Journal of Consulting Psychology, 29, 206-212.
Hooper, S. R, & Willis, W. G. (in press). Learning disability subtyping: Neuropsychologi-
cal foundations, conceptual models, and issues in clinical differentiation. New York:
Plenum.
Huberty, C. J. (1975). Discriminant analysis. Review of Educational Research, 45, 543-598.
Huberty, C. J. (1984). Issues in the use and interpretation of discriminant analysis. Psycho-
logical Bulletin, 95, 156-171.
Jensen, A. (1973). Education and group differences. New York: Harper & Row.
Kaufman, A. S., & Kaufman, N. L. (1983). Kaufman Assessment Battery for Children. Circle
Pines, MN: American Guidance Service.
Kurtz, R M., & Garfield, S. L. (1978). Illusory correlation: A further exploration of Chapman's
paradigm. Journal of Consulting and Clinical Psychology, 46,1009-1015.
Lachenbruch, P. A. (1967). An almost unbiased method of obtaining confidence intervals for
the probability of misclassification in discriminant analysis. Biometrics, 23, 639-645.
Lueger, R J., & Petzel, T. P. (1979). Illusory correlation in clinical judgment: Effects of
amount of information to be processed. Journal of Consulting and Clinical Psychology,
47, 1120-1121.
Lyon, R, Stewart, N., & Freedman, D. (1982). Neuropsychological characteristics of em-
pirically derived subgroups of learning disabled readers. Journal of Clinical Neuropsy-
chology, 4, 343-365.
McDermott, P. A. (1981). Sources of error in the psychoeducational diagnosis of children.
Journal of School Psychology, 19, 31-44.
McDermott, P. A. (1982). Actuarial assessment for the grouping and classification of school
children. In C. R Reynolds & T. B. Gutkin (Eds.), The handbook for school psychology
(pp. 243-272). New York: Wiley.
McDermott, P. A., & Watkins, M. W. (1985). McDermott Multidimensional Assessment of
Children. Cleveland: OH: Psychological Corporation.
Meehl, P. E. (1954). Clinical versus statistical prediction: A theoretical analysis and a
review of the evidence. Minneapolis: University of Minnesota Press.
Meehl, P. E. (1973). Psychodiagnosis: Selected papers. Minneapolis: University of Min-
nesota Press.
Meehl, P. E., & Rosen, A.(1955). Antecedent probability and the efficiency of psychometric
signs, patterns, or cutting scores. Psychological Bulletin, 52, 194-216.
Mercer, J. R, & Lewis, J. F. (1978). System of Multicultural Pluralistic Assessment. New
York: Psychological Corporation.
110 W. GRANT WILLIS
Mitchell, B. (1967). Predictive validity of the Metropolitan Readiness Test and the Murphy-
Durrell Reading Readiness Analysis for white and negro pupils. Educational and Psy-
chological Measurement, 27, 1047-1054.
Mosier, C. I. (1951). Problems and designs of cross-validation. Educational and Psychologi-
cal Measurement, 11, 5-11.
Nisbett, R. E., & Ross, L. (1980). Human inference: Strategies and shortcomings of social
judgment. Englewood Cliffs, NJ: Prentice-Hall.
Obrzut, J. E., Hynd, G. W., & Obrzut, A. (1983). Neuropsychological assessment of learning
disabilities: A discriminant analysis. Journal of Experimental Child Psychology, 35, 46-
55.
Pedhazur, E. J. (1982). Multiple regression in behavioral research (2nd ed.). New York: Holt,
Rinehart & Winston.
Phelan, J. G. (1964). Rationale employed by clinical psychologists in diagnostic judgment.
Journal of Clinical Psychology, 20, 454-458.
Rapoport, J. L., & Ismond, D. R. (1984). DSM-III training guide for diagnosis of childhood
disorders. New York: Brunner/Mazel.
Reitan, R. M. (1984). An impairment index of brain functions in children. Perceptual and
Motor Skills, 58, 875-881.
Reitan, R. M., & Davison, L. A. (Eds.). (1974). Clinical neuropsychology: Current status and
applications. Washington, DC: V. H. Winston.
Reynolds, C. R. (1982). Determining statistically reliable strengths and weaknesses in the
performance of single individuals on the Luria-Nebraska Neuropsychological Battery.
Journal of Consulting and Clinical Psychology, 50, 525-529.
Reynolds, C. R., & Clark, J. H. (1983). Assessment of cognitive abilities. In K. D. Paget & B.
Bracken (Eds.), The psychoeducational assessment of preschool children (pp. 163-
189). New York: Grune and Stratton.
Reynolds, C. R., & Gutkin, T. B. (1981). Statistics for the interpretation of Bannatyne re-
organization of WPPSI subtests. Journal of Learning Disabilities, 14,446-467.
Reynolds, C. R., & Kaufman, A. S. (1985). Clinical assessment of children's intelligence with
the Wechsler scales. In B. B. Wolman (Ed.), Handbook of intelligence: Theories, mea-
surements, and applications (pp. 601-661). New York: Wiley.
Rosen, G. D., Sherman, G. F., & Galaburda, A. M. (1986). Biological interactions in dyslexia
(pp. 155-173). In J. E. Obrzut & G. W. Hynd (Eds.), Child neuropsychology: Vol. 1.
Theory and research. Orlando, FL: Academic Press.
Sattler, J. M. (1982). Assessment of children's intelligence and special abilities (2nd ed.).
Boston: Allyn & Bacon.
Sattler, J. M. (1988). Assessment of children (3rd ed.). San Diego, CA: Jerome M. Sattler.
Scarr-Salapatek, S. (1971). Race, social class, and IQ. Science, 174, 1285-1295.
Selz, M. (1981). Halstead-Reitan neuropsychological test batteries for children. In G. W.
Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child
(pp. 195-235). New York: Grune and Stratton.
Selz, M., & Reitan, R. M. (1979). Rules for neuropsychological diagnosis: Classification of
brain function in older children. Journal of Consulting and Clinical Psychology, 47,
258-264.
Shepard, L. A., Smith, M. L., & Vojir, C. P. (1983). Characteristics of pupils identified as
learning disabled. American Educational Research Journal, 20, 309-331.
Spitzer, R. L., Skodol, A. E., Gibbon, M., & Williams, J. B. W. (1981). DSM-III case book.
Washington. DC: American Psychiatric Association.
Starr, J. G., & Katkin, E. (1969). The clinician as an aberrant actuary: Illusory correlation and
the Incomplete Sentences Blank. Journal of Abnormal Psychology, 74, 670-675.
Thorndike, R. L. (1982). Applied psychometrics. Boston: Houghton Mifflin.
Tryon, R. c., & Bailey, D. E. (1970). Cluster analysis. New York: McGraw-Hill.
Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and
probability. Cognitive Psychology,S, 207-232.
Wallach, M. S., & Schoof, K. (1965). Reliability of degree of disturbance rating. Journal of
Clinical Psychology, 21, 273-275.
Wechsler, D. (1974). Wechsler Intelligence Scale for Children-Revised. New York: Psycho-
Willis, W. G. (1984). Reanalysis of an actuarial approach to neuropsychological diagnosis in
consideration of base rates. Journal of Consulting and Clinical Psychology, 52, 567 -569.
Willson, V. 1., & Reynolds, C. R. (1982). Methodological and statistical problems in deter-
mining membership in clinical populations. Clinical Neuropsychology, 4, 134-138.
5
From Assessment to Treatment

Linkage to Interventions with Children
G. REID LYON, LOUISA MOATS, and JANE M. FLYNN
INTRODUCTION
Within the past decade, child clinical neuropsychologists have been

called upon increasingly to make relevant and informed recommenda-
tions for the treatment of both documented (Le., traumatic head injury)
and putative (Le., learning disabilities) neurologically based developmen-
tal disorders. This increase in requests for specific therapeutic recommen-
dations reflects a change in how the role of the child clinical neurop-
sychologist is perceived and, in particular, how the data obtained from
neuropsychological assessments are used.
Historically, neuropsychological assessment practices were applied
primarily to identify the presence, nature, and possible site(s) of brain
damage or dysfunction (Beaumont, 1983). However, the emphasis in
clinical neuropsychology has shifted from assisting in the diagnosis of
lesion type and location to the assessment of the functional capacities of
the child in order to select and implement efficacious management, re-
habilitation, and/or remediation programs.
There appear to be several reasons for this transition in the practice of
clinical neuropsychology. For example, recent technical advances in spe-
cialized neurodiagnostic procedures for brain imaging gradually are re-
placing neuropsychological approaches to lesion classification (Stoddart
G. REID LYON Departments of Neurology and Communication Science and Disorders.

University of Vermont. Burlington. Vermont. Gundersen Medical Foundation. and Depart-
ment of Special Education. St. Michael's College. Winooski. Vermont. LOUISA
MOATS Associates in Counseling and Education. East Thetford. Vermont. JANE M.
FLYNN Gundersen Medical Foundation. LaCrosse. Wisconsin.
113
114 G. REID LYON et a1.
& Knights, 1986). Less documentable, but equally influential, are the so-
cial, political, and educational trends that have created and fostered a
major role for neuropsychological treatment approaches in the fields of
learning disabilities and cognitive rehabilitation. Models for treatment
that claim to derive validity from biological science are appealing in the
field of learning disabilities (LD), in particular, where previously em-
ployed aptitude-treatment interaction models have failed to generate the
foundations of a clinical science (Lyon, 1987; Lyon & Moats, in press).
The heterogeneity of the LD population, although now recognized, con-
tinues to beg classification and validation (Lyon & Risucci, 1988). Thus,
treatment models that invoke biological explanations for intellectual and
behavioral differences are welcome in the void. A focus on intrinsic bio-
logic variables permits us to minimize the importance of educational,
cultural-familial, and societal causes for individual differences (Chall &
Mirsky, 1978) and, thus, to empower ourselves by defining treatment
problems in terms of simple, often dichotomous neurobehavioral con-
structs (e.g., right vs. left brain learners, sequential vs. simultaneous
processing).
Taken at face value and for whatever reasons, the shift in emphasis
from neuropsychological assessment for classification purposes (i.e.,
presence or absence of brain damage) to assessment for prescriptive pur-
poses (i.e., treatment) reflects a possibly productive transition toward
enhanced clinical relevance. Indeed, Alfano and Finlayson (1987) and
others (Lyon & Moats, in press; Newcomb, 1985) have pointed out re-
cently that the power of the clinical contributions that evolve from neuro-
psychological practice ultimately depends upon the field's capacity to (1)
delineate neuropsychological strengths and weaknesses in a reliable and
valid fashion, (2) predict the extent to which these information-process-
ing characteristics influence recovery of function and/or learning, and (3)
generate testable hypotheses concerned with remedial methodologies for
individuals who either lost or did not acquire information because of
known or suspected neural insult.
Perusal of the current literature in both child and adult clinical neu-
ropsychology reveals significant support for the concept of establishing
valid linkages between the information generated from neuropsychologi-
cal assessment practices and the development of management, rehabilita-
tion, and remediation programs (Rourke, Fisk, & Strang, 1986). However,
such support is generally given in the form of rhetorical and testimonial
presentation, case study information, or anecdotal reports. Although
these particular forums underscore the value of attempting to develop
valid neuropsychologically based treatment programs, we feel that popu-
lar views regarding the clinical benefits of such practices go beyond the
data. Given this observation, it should be made clear that our purpose in
this chapter is not to advocate dismissal of the concept of neuropsycho-
FROM ASSESSMENT TO TREATMENT 115
logical aptitude-treatment interaction as a useful clinical possibility. On

the contrary, our goal is to clarify what actually is known about the rela-
tionships between neuropsychological assessment outcomes and treat-
ment decisions and to make relevant recommendations regarding con-
tinued research and clinical practice.
Within this context, we first examine neuropsychological assessment
principles frequently used with children with an eye toward delineating
the purposes for which they were developed as well as their measurement
and content characteristics. In doing so, we also attempt to point out how
such characteristics relate or do not relate to the content of the clinical
treatment process. Following this overview, we describe specific neuro-
psychological assessment models that have been reported to be useful in
generating treatment programs for children with a range of neuropsycho-
logical deficiencies. Our purpose is to address the question of which
models are most efficacious for the treatment of particular clinical popu-
lations (i.e., learning-disabled) and which ones hold the most promise for
developing even more robust linkages between assessment outcome and
instructional decision making. Finally, we present what we believe are
the major limitations in clinical neuropsychological assessment practices
that impede our ability to formulate relevant and powerful intervention
programs for children with neurodevelopmental disorders. This discus-
sion serves as the basis for what we hope are productive recommenda-
tions for enhancing the descriptive, predictive, and ecological validities
of assessment practices as they relate to treatment methodologies.
NEUROPSYCHOLOGICAL ASSESSMENT: PURPOSES AND

MEASUREMENT CHARACTERISTICS
In order to provide a context for conceptualizing how linkages can be

forged between assessment and treatment, it is necessary to examine first
the purposes (goals), psychometric properties, and task domains common
to neuropsychological batteries and approaches. Given this information,
one then can attempt to determine if the inferential process inherent in
neuropsychological assessment can be related efficiently to specific treat-
ment methodologies. The reader should note that concepts relevant to
neuropsychological assessment have been covered in depth throughout
this volume. Thus, what follows is a brief discussion of factors that guide
the measurement of brain-behavior relationships.
One's purpose for conducting neuropsychological assessment ob-
viously influences how brain-behavior relationships are measured and
assessment results interpreted. As Boll (1981) pointed out, the primary
purpose or goal of neuropsychology is to describe brain-behavior rela-
tionships in a reliable and valid manner. Boll also indicated that the
116 G. REID LYON et 01.
ultimate, but yet to be realized goal is "the development of remediation

and rehabilitation procedures based upon the empirically validated un-
derstanding of the behavioral consequences specific to the condition in
question in each patient" (p. 582). Although it is noteworthy that Boll
(1981) and others (Beaumont, 1983; Obrzut & Hynd, 1986; Rourke et a1.,
1986) have emphasized the need for neuropsychology to establish valid
relationships between assessment and treatment, it seems clear that the
majority of batteries and allied procedures designed to infer brain func-
tion from behavior have been constructed by incorporating psychometric
principles and task content primarily useful for descriptive purposes, not
necessarily prescriptive purposes.
To clarify this point, consider the following. Prominent neuropsycho-
logical assessment batteries and diagnostic procedures have been devel-
oped and refined over the past 50 years by selecting or developing tasks
and applying inferential interpretive methods to (1) understand the im-
pact of brain damage or dysfunction on a range of human abilities, (2)
differentiate reliably those individuals who present with brain damage
and dysfunction from those who do not, and (3) discern the specific
behavioral effects of different types of neuropathology (e.g., tumor vs.
stroke vs. head injury).
It is reported frequently in the neuropsychology literature that these
clinical outcomes are realized most effectively when (1) the assessment
procedures consist of objective, standardized, and quantitative measures
of an individual's neuropsychological ability structure (Alfano & Finlay-
son, 1987; Reitan, 1966; Reitan & Wolfson, 1985; Rourke, 1981; Rourke et
a1., 1986); (2) the assessment procedures include measures that are scaled
psychometrically to measure abilities on a continuous scale rather than
on an interval scale (Golden, Hammeke, & Purisch, 1978); (3) the assess-
ment tasks and measures are valid and reliable reflections of cerebral
dysfunction and are not confounded by the effects of age and education
(Finlayson, Johnson, & Reitan, 1977); and (4) the assessment tasks sample
a broad range of abilities to include measures of general intellectual abil-
ity, the ability to retain verbal and nonverbal information, motor and
psychomotor abilities, sensory-perceptual functions, receptive and ex-
pressive language skills, attentional skills, analytical reasoning and con-
cept formation, and personality, behavioral, and emotional status (Alfano
& Finlayson, 1987; Reitan & Wolfson, 1985).
A number of studies have shown that neuropsychological assessment
batteries and allied procedures that have been developed according to
these principles are valid for the purposes of identifying the presence of
brain damage or dysfunction in both adults (Boll, 1981; Golden et a1.,
1978; Reitan & Davison, 1974) and children (Hynd & Obrzut, 1986; Rou-
rke, 1981; Rourke et a1., 1986; Teeter, 1986). Further, there are data indi-
cating that widely used neuropsychological batteries (e.g., Halstead-
Reitan, Luria-Nebraska) are capable of describing the nature of the neural

insult (e.g., type of lesion, site of lesion), particularly when applied to
adult clinical populations and interpreted by skilled clinicians. There is
also some evidence, albeit limited in scope, that the information derived
from adult neuropsychological assessment batteries can be useful in con-
structing some remediation and rehabilitation programs (Diller & Gor-
don, 1981a, 1981b; Diller & Weinberg, 1977; Finlayson, Gowland, &
Basmajian, 1986; Luria, 1966; Luria & Tzetkova, 1968; Rao & Bieliauskas,
1983).
Despite the success achieved by traditional neuropsychological as-
sessment practices in the diagnosis and description of neuropathology in
both adult and pediatric populations, the clinical utility and validity of
such practices in designing treatment programs for children remains
questionable. This appears to be the case for several reasons. First, a large
portion of the assessment tasks that constitute the most widely used stan-
dardized neuropsychological batteries for children are downward exten-
sions of batteries initially developed and validated on adult clinical popu-
lations. This is particularly true of the Halstead Neuropsychological Test
Battery for Children (Reitan & Davison, 1974), the Reitan-Indiana Neuro-
psychological Test Battery (Reitan, 1969), and the Luria-Nebraska Battery-
Children's Revision (Plaisted, Gustavson, Wilkening, & Golden, 1983).
Likewise, the type of stimuli (content) used in tasks to assess specific
brain-behavior relationships are downward extensions of stimulus items
presented to adults.
These test development practices could compromise a battery's power
in predicting which treatment methods are most efficacious for particular
children because (1) the tasks employed and their content are based pri-
marily on models of adult brain function and dysfunction that occur
following a period of normal development; (2) many tasks are designed to
assess the effects of focal neuropathology typically seen in adults (e.g.,
tumors, cerebral vascular accidents, penetrating head wounds) rather than
the generalized neural disorders usually observed in children (e.g., closed
head injury, anoxia, epilepsy, perinatal trauma); and (3) the neuropsycho-
logical task content may bear minimal relationship to the ecological de-
mands that the child is facing in home and school environments. For
example, even though many widely used children's batteries contain tasks
assessing reading, mathematics, and writing skills, such tasks rarely pos-
sess adequate content validity. Consider that the Wide Range Achievement
Test, a staple of many child neuropsychological batteries and procedures,
assesses only the oral reading of single words, mathematics calculation,
and spelling, leaving abilities in reading comprehension, math reasoning,
and written language open to question.
Second, and related to the previous points, some neuropsychological
assessment procedures employed with children use tasks that yield static
measures of competence in neuropsychological ability structures. The

data obtained from such measures reflect only a child's past and current
declarative knowledge of perceptual, linguistic, cognitive, psychomotor,
and academic skills, not how they use or do not use such abilities in their
daily lives (Brown & Campione, 1986; Lyon & Moats, in press). A notable
exception is the Category Test, known for its sensitivity to abstract con-
cept formation, mental efficiency, and the ability to assess new learning
(Boll, 1981).
Third, there is increasing concern that tasks making up neuropsycho-
logical assessment batteries for children primarily assess general cog-
nitive ability, not distinct neuropsychological processes (Hynd & Obrzut,
1986; Seidenberg, Giordani, Berent, & Boll, 1983; Tramontana, Klee, &
Boyd, 1984). Given this possibility, administering time-consuming bat-
teries beyond administration of a Wechsler Intelligence Scale for Chil-
dren-Revised (WISC-R) may net redundant information. Further, the con-
sistent finding that the WISC-R is not particularly useful for the
development of instructional or remediation programs (Ysseldyke & AI-
gozzine, 1982; Ysseldyke & Mirkin, 1981) does not bode favorably for the
use of redundant neuropsychological batteries for the same purpose.
Fourth, issues related to development and brain maturation may ob-
scure some possible benefits that accrue from administering standard
neuropsychological batteries to children for the purposes of designing
treatment programs. For instance, Hynd and Obrzut (1986) reported that a
number of neuropsychological tasks simply are not age-appropriate, and
no neuropsychological test battery has yet to establish adequate cross-
sectioned norms. Further, these authors concluded that, without such
norms, "it becomes nearly impossible to provide any accurate appraisal of
the possible impairment of developing abilities" (p. 10).
Finally, as Lyon and Toomey (1985) have stressed, describing brain-
behavior relationships through the application of assessment procedures
does not ensure successful remediation of brain~based deficiencies.
Whereas neuropsychological assessment may help to clarify the physio-
logical correlates of dyslexia for example, altering the underlying neuro-
pathology or identifying alternate intact processing routes via remedia-
tion may not be possible.
There is little doubt that attempts to use neuropsychological assess-
ment data for the purposes of planning treatment programs reflect produc-
tive movement toward an exceedingly important clinical goal. However,
forming clinically valid linkages between assessment and treatment meth-
ods may be hindered by the application of diagnostic procedures that
were developed initially for descriptive rather than prescriptive purposes.
Moreover, difficulties in identifying such linkages are exacerbated by in-
adequacies in the content, predictive, and ecological validities of stan-
dard neuropsychological tasks used with children, as well as limitations
in understanding normative development. These points are elaborated in

the following sections.
RELATING ASSESSMENT TO TREATMENT: MODELS AND

STUDIES
In general, neuropsychological models of developmental disorders

conceptualize a child's learning strengths and weaknesses as manifesta-
tions of efficient or inefficient brain regions and/or systems (Gaddes,
1980; Hartlage & Telzrow, 1983; Obrzut & Hynd, 1986; Rourke, Bakker,
Fiske, & Strang, 1983). A variety of standard neuropsychological batteries
as well as selected neuropsychological assessment procedures have been
employed to elucidate such patterns of strengths and weaknesses.
Despite the meager data base, it may be useful to examine specific
findings that have been obtained from both clinical and empirical study of
the assessment-treatment interface for no other reason than to explicate
current clinical trends and future research possibilities. For clarity, we
have organized our review of the extant literature in this area first to
address treatment approaches and models that are linked to data obtained
from standard neuropsychological assessment batteries. This review is
followed by a review of intervention studies associated with the use of
selected batteries. Treatment approaches will be examined with respect to
their orientation (e.g., remediating strengths, weaknesses, or both) as well
as to their demonstrated efficacy.
STANDARDIZED ASSESSMENT BATTERIES: IMPLICATIONS

FOR TREATMENT
A number of recent papers have discussed the application of stan-

dardized neuropsychological assessment methods in formulating treat-
ment programs for children (Alfano & Finlayson, 1987; Gunnison, 1984;
Teeter, 1986). These efforts typically involve the use of assessment data
obtained from the Halstead-Reitan Neuropsychological Test Batteries and
Allied Procedures (Reitan, 1980), the Luria-Nebraska Battery-Children's
Revision (Plaisted et al., 1983), and the Kaufman Assessment Battery for
Children (K-ABC; Kaufman & Kaufman, 1983). A discussion of these par-
ticular assessment methods and their relationship to treatment follows.
The Halstead-Reitan: Linkages to Treatment

The Halstead-Reitan assessment procedures (Boll, 1981; Hartlage &
Hartlage, 1977; Teeter, 1986) are a mainstay of clinical neuropsychologi-
cal practice with children. The Halstead-Reitan procedures have been

reported to be sensitive to brain dysfunction in a number of developmen-
tal disorders, including asthma (Dunleavy & Beade, 1980), autism
(Dawson, 1983), Gilles de la Tourette's syndrome (Bornstein, King, & Car-
roll, 1983), juvenile delinquency (Yendall, Fromm-Auch, & Davies, 1982),
and epilepsy (Herman, 1982). An abundance of data show that the
Halstead-Reitan batteries are valid for the differential diagnosis of brain
damage in children (Boll & Reitan, 1972; Reed, Reitan, & Klove, 1965;
Reitan, 1979). Further, the batteries have been found useful for the neuro-
psychological classification of minimal brain dysfunction in young chil-
dren (ages 5-8) (Reitan & Boll, 1973) and learning disabilities in older
children (ages 9-14) (Reitan, 1980: Selz & Reitan, 1979).
In 1980, Reitan initiated formal attempts to relate neuropsychological
assessment data explicitly to treatment through the development of a
program titled Reitan Evaluation of Hemispheric Abilities and Brain Im-
provement Training (REHABIT). According to Reitan (1979, 1980), the
efficacy of REHABIT for the purposes of remediation is dependent upon
(1) a comprehensive neuropsychological evaluation (using the Halstead-
Reitan procedures) that clearly identifies areas of brain-related strengths
and weaknesses and, (2) a determination from the assessment data as to
whether the particular neuropsychological deficits reflect neural-cog-
nitive deficiencies or generalized cognitive problems affecting several
functional systems.
According to Reitan (1980), direct linkages between assessment and
treatment are forged by the training concepts inherent in the REHABIT
model. For example, REHABIT proposes treating the general area of
neuropsychological deficit directly by using alternate forms of neuropsy-
chological tests as training items. According to Alfano and Finlayson
(1987), such an approach seems reasonable because challenging the areas
measured by neuropsychological tasks could provide direct stimulation
of the wide range of neural functions that they assess. (For an alternate
point of view, refer to Mann, 1979; Mann & Sabatino, 1985.) Following
this general form of deficit training, remediation in five specific areas
(tracts) is carried out using previously developed educational materials
and tasks. The tracts include (1) Tract A, materials for the development of
expressive and receptive language and verbal skills; (2) Tract B, materials
to develop abstract language functions to include verbal reasoning, verbal
concept formation, and verbal organization; (3) Tract C, materials de-
signed to enhance general reasoning capabilities; (4) Tract D, materials for
developing abstract visual-spatial and temporal-sequential concepts; and
(5) Tract E, materials designed to promote understanding of basic visual-
spatial and manipulation skills. Thus, Tracts A and B are generally linked
to left hemisphere functions, Tracts D and E to right hemisphere func-
tions, and Tract C to general logical analysis and reasoning functions
subserved by all functional systems.
Data to support the REHABIT rehabilitation and remediation con-

cepts are difficult to find. Reitan (1979, 1980) does report a few case
studies, but the information provided in them cannot be construed as
empirical validation for the REHABIT model In fact, reviews of similar
neuropsychological process remediation models (Lyon & Moats, in press;
Mann, 1979) have indicated that such practices suffer from a lack of both
construct and ecological validity, particularly with respect to their ap-
plication with children who display academic achievement deficits with-
out demonstrable brain injury. In the absence of empirical validation for
the REHABIT model, the clinician is ultimately responsible for judging
whether the time spent in assessment and training activities is in the best
interests of the child.
Rourke and his colleagues (Rourke et a1., 1983, 1986) have argued
convincingly that the aims, content, and style of neuropsychological as-
sessments are improved significantly when a comprehensive battery of
neuropsychological tasks are administered to children and the data in-
terpreted according to several frames of reference (level of performance,
pathognomonic signs, differential [pattern] score approach, comparisons
of performance on two sides of the body, pre- and postlesion com-
parisons). Rourke's (1975, 1981) orientation to assessment practices and
how assessment data relate to treatment is influenced significantly by his
use of Reitan's neuropsychological measurement concepts and modes of
clinical interpretation (Rourke, 1981). As such, Rourke's concepts of how
brain-related deficiencies in children should be assessed and related to
remediation are presented in this section.
Rourke et a1. (1986) propose that linkages are best formed between
assessment and remediation when a developmental neuropsychological
model is employed. Within the context of such a model, specific informa-
tion related to the child's neuropsychological ability structure is collected
and interpreted in relation to (1) the immediate demands in the environ-
ment (e.g., school and social demands); (2) hypothesized long-range de-
mands (e.g, occupational and social functioning); (3) specific short- and
long-term behavioral outcomes that best characterize the child with re-
spect to developmental status, information-processing strengths and
weaknesses, and neuropsychological status; (4) an ideal remediation pro-
gram for the child given the above information; and (5) the development
of a realistic remediation program given the child's characteristics and the
actual availability of remedial sources for family, school, and child.
Rourke's developmental neuropsychological remediation/habilita-
tion model appears to have potential for linking assessment data to treat-
ment because it stresses a comprehensive analysis of how child variables
interact systematically with environmental factors and the pragmatics of
clinical service delivery. However, as with other models incorporating the
use of Halstead-Reitan tasks and modes of clinical interpretation (Le.,
REHABIT), there simply are not the empirical data to support the validity
and clinical efficacy of Rourke's (Rourke et aI., 1983, 1986) model. Clearly,
however, the model does offer well-reasoned and comprehensive guide-
lines for intervention that can serve as clinical frames of reference.
The Luria-Nebraska: Linkages to Treatment

A. R. Luria's (1973, 1980) seminal conceptualization of the human
brain as being composed of functional units has led to the development of
a standardized neuropsychological assessment battery for use with both
adults and children. The Luria-Nebraska Battery-Children's Revision
(Plaisted et a1., 1983) has been found to be sensitive in detecting demon-
strable neuroencephalopathy in children (Teeter, 1986). Some recent
studies also have shown that the battery can discriminate between learn-
ing-disabled children and normally achieving students (Geary & Gilger,
1984; Nolan, Hammeke, & Barkley, 1983).
To date, no formal attempts have been made to relate assessment data
obtained from the Luria-Nebraska to structured remediation programs for
children. However, Luria's (1973) concepts of brain-behavior relation-
ships can be clinically useful if applied to intervention practices in an
informed manner. This conclusion may be a reasonable one for at least
two reasons. First, Luria's model incorporates concepts related to both
brain systems and their development. As such, a dynamic theoretical
basis exists from which to make predictions about outcome and potential
for remediation. Second, Luria's model argues that disturbances in com-
plex cognitive functions can be related to a wide variety of brain-related
deficiencies. For example, failure to learn to write could be attributable to
deficits in any of several brain systems. Thus, children who display writ-
ten language deficits may not respond equally well to the same remedia-
tion procedure.
One additional point is in order. Luria (1963, 1973, 1980) advocated
the use of dynamic, nonstandardized assessment methods that could vary
across patients depending upon the nature of the clinical question. He
supported the use of these procedures with substantial clinical case-study
data. Further, Luria (1963,1980) presented a rationale for applying assess-
ment procedures directly to the treatment and rehabilitation process and
reported case-history data to substantiate his point of view. It is possible
that attempts to standardize Luria's dynamic assessment methods could
reduce their power in relating assessment findings to treatment program
planning. The reader should keep in mind that this possibility remains an
open question.
The K-ABC: Linkages to Treatment

Another standardized assessment tool that relies on neuropsycholog-
ical constructs is the Kaufman Assessment Battery for Children (K-ABC;
Kaufman & Kaufman, 1983). Emphasizing a dual-processing model of

cognition, the K-ABC Test purports to measure simultaneous and suc-
cessive information-processing strengths and weaknesses in children up
to 12 years of age. In addition to scores on the Simultaneous and Suc-
cessive scales, a third Achievement test cluster is used to measure ac-
quired knowledge and verbal learning ability. The test user then is en-
couraged to formulate hypotheses regarding remediation of academic
deficiencies that emphasize the preferred processing mode of the subject.
Unique to the K-ABC remediation framework (Gunnison, 1984) is the
specificity of these recommendations to academic domains-reading,
arithmetic reasoning, and written language-and the well-elaborated
models of intervention that attempt to code both learner behavior and task
demands along the simultaneous-sequential dichotomy.
Unfortunately, the usefulness of the K-ABC even for descriptive and
classification purposes has not been accepted uniformly. For example,
Sternberg (1984) argued that the test lacks construct validity, a problem
that may be related to the authors' misrepresentation or misreading of the
evidence supporting a simultaneous-successive processing dichotomy.
Further, in equating processing style with scores on selected tasks, the test
fails to assess constructs that pertain to dynamic problem solving. Select-
ing and conducting remediation on the basis of K-ABC results thereby
would constitute a tenuous practice.
Empirical support for remediation based on the K-ABC, as with other
neuropsychological approaches reviewed in this chapter, is sparse. Al-
though Gunnison and her colleagues (research in press), cited in Gun-
nison, 1984) have shown meaningful gains in reading for children taught
with methods described as simultaneous or sequential in emphasis, both
the assumptions underlying the aptitude-treatment linkages and the data
base supporting them are weak (Ayres & Cooley, 1986; Salvia & Hritchko,
1984). Where the logical-intuitive classification of child responses and
teaching strategies appears to have most value is in the provision of a
conceptual framework for diagnostic teaching. The concept of dual-pro-
cessing modes may simply encourage the clinician to behave in a flexible
manner when alternative representations of concepts are needed by the
learner.
SELECTED ASSESSMENT BATTERIES: LINKAGES TO

TREATMENT
The majority of studies that have attempted to identify empirically

the linkages between assessment data and treatment options have mea-
sured children's neuropsychological characteristics with tasks that are
not standardized in battery form. Generally, the tasks are selected on the
basis of their relevance to a particular theoretical test of a research ques-
tion. Because the tasks included in selected assessment batteries lack a

common standardization sample from which to derive scores, control
groups matched on relevant variables generally are assessed along with
the clinical group of interest.
In the main, studies employing this type of assessment approach are
conducted for the purpose of establishing a classification scheme for chil-
dren who are included in heterogeneous clinical populations (Le., learning
disabilities). Within this classification context, children are assigned to
different subtypes on the basis of their performance on the selected neuro-
psychological tasks. Once a classification solution is obtained, it must be
validated internally and externally. Internal validation is achieved by
ensuring that the subtypes identified are reliable, replicable, and robust
enough to include most members of the clinical population of interest.
External validity is examined by determining whether the classification
solution is useful for description, prediction, and clinical practice. One
specific way to address external validity is to determine whether subtypes
differ from one another in response to treatment (Lyon & Risucci, 1988). It
is this area of classification research that is relevant to this chapter.
To date, a number of research programs have reported preliminary
data that suggest that subtypes respond differently to various forms of
remediation. Although all of the published investigations have been car-
ried out with learning-disabled readers (dyslexics), the studies differ with
respect to theoretical orientation, assessment tasks used to form subtypes,
and classification methodology. For example, Lyon and his colleagues
have identified several subtypes by applying empirical multivariate quan-
titative clustering methods to information-processing task scores obtained
by large samples of LD readers. External validity studies have then in-
volved attempts to teach the disabled learners and to determine subtype-
teaching method interactions. In contrast, Bakker (1983) has classified
dyslexics into two major subtypes according to clinical criteria, the most
important of which is left/right ear asymmetrics in dichotic listening
tasks. External validation consisted of hemisphere-specific stimulation
via presentation of words to right and left visual fields and identifying
whether subtypes responded differently to both the site of presentation
and the type of stimulus used. Flynn and her associates (Flynn, 1987)
have concentrated on clinically identifying dyslexic subtypes on the basis
of their reading and spelling error patterns. She has presented compelling
pilot data showing that children with particular patterns respond well to
specific methods of reading instruction. Each of these research programs
is reviewed in more detail in this section. Emphasis is placed on describ-
ing the theoretical orientation that drives the selection of assessment tasks
used in the various research programs, the types of remediation pro-
cedures employed, and the clinical relationship between tasks and inter-
ventions.
Empirical Subtype Intervention Studies: The Lyon Research

Program
Lyon and his associates (Lyon, 1983, 1985a, 1985b; Lyon, Stewart, &
Freedman, 1982; Lyon & Watson, 1981) have questioned the appropri-
ateness of a single-deficit classification model for reading disability and
hypothesized that LD readers (dyslexics) constitute a population that is
composed of a number of subtypes, each of which is defined by its own
particular array of linguistic, perceptual, and reading characteristics. The
theoretical background underlying Lyon's research can be viewed as a
logical extension of Luria's (1966, 1973) clinical neuropsychological theo-
ry and Benson and Geschwind's (1975) multiple syndrome model of alex-
ia. For example, Lyon (1983) proposed that reading development is a
complex process that requires the concerted participation of cognitive,
linguistic, and perceptual subskills. As such, deficiencies in anyone sub-
skill can limit the acquisition of fluent decoding and/or comprehension
abilities.
Within this theoretical context, an initial series of studies was con-
ducted (Lyon, Rietta, Watson, Porch, & Rhodes, 1981; Lyon & Watson,
1981) in which a battery of tasks designed to assess linguistic and percep-
tual skills related to reading development was administered to 100 LD
readers and 50 normal readers matched for age (11-12 years) and IQ (M =
104). The data were submitted to a series of cluster analyses to test the
hypothesis that subtypes could be identified. Six distinct subtypes were
delineated and characterized by significantly different patterns of lin-
guistic and perceptual deficits. The six-subtype solution remained stable
across internal validation studies employing different variable subsets
and clustering algorithms. Further, 94% of subjects were recovered into
similar subtypes in a cross-validation study using a new subject sample
(Lyon, 1983). A brief description of each of the subtypes' information-
processing characteristics is provided here, followed by an overview of
the intervention program. Readers are referred to cited references for spe-
cific details.
Children who were assigned empirically to subtype 1 (n = 10) exhib-
ited significant deficits in language comprehension, the ability to blend
phonemes, visual-motor integration, visual-spatial skills, and visual
memory skills, with strengths in naming and auditory discrimination
skills. Analysis of the reading and spelling errors made by members of
subtype 1 indicated significant deficits in the development of both a sight
word vocabulary and word attack skills.
Children in subtype 2 (n = 12) also exhibited a pattern of mixed
deficits but in a milder form than observed in subtype 1. Specifically,
significant problems in language comprehension, auditory memory, and
visual-motor integration were observed and may have been related to the
reading problems of these subjects. No deficits were seen in these young-

sters' performance on naming, auditory discrimination, sound blending,
visual-spatial, and visual memory tasks. Subtype 2 members produced
mixed visual and phonetic errors when reading but to a much milder
degree than did subtype 1 children.
Members of subtype 3 (n = 12) manifested selective deficits in lan-
guage comprehension and sound blending, with corresponding strengths
in all other linguistic and visual-perceptual skills measured. The oral
reading errors made by subtype 3 youngsters were primarily phonetic in
nature, as would be expected from their diagnostic profile.
Children in subtype 4 (n = 32) displayed significant deficiencies on
visual-motor integration tasks and average performance on all other mea-
sures. These youngsters displayed an assorted sample of oral reading
errors, though most errors were made when attempting to read phonet-
ically irregular words.
Subtype 5 (n = 12) members displayed significant deficits in lan-
guage comprehension, auditory memory, and sound blending, with corre-
sponding strengths in all measured visual-perceptual and visual-motor
skills. These characteristics appeared related to the severity of their oral
reading and written spelling errors. The major academic characteristic
that distinguished subtype 5 youngsters from the other children was their
consistently poor application of word attack (phonetic) skills to the read-
ing and spelling process.
The pattern of scores obtained by members of subtype 6 (n = 16)
indicated a normal diagnostic profile. These results were unexpected. It is
quite possible that these children were reading poorly for reasons that
were not detected by the assessment battery.
Following this subtype identification study, an external validation
investigation (Lyon, 1983) was carried out to determine whether subtypes
would respond differently to reading instruction. However, because of the
relatively small sample size, a standard aptitude-by-treatment study
could not be designed appropriately. Therefore, it was decided to explore
the possibility that the six subtypes might respond differently to one
teaching condition. Since the children for this exploratory study had to be
matched for pre intervention achievement levels and other relevant vari-
ables (age, IQ, sex, socioeconomic status), the initial subject pool available
from the subtype identification study was reduced to 30. Thus, random
assignment of children from each of the six subtypes to several teaching
conditions was not feasible.
In light of these logistical difficulties, five subjects were selected from
each of the six subtypes. They were matched on their ability to read single
words, age, IQ, race, and sex. All 30 subjects were white males ranging in
age from 12.3 years to 12.7 years and in Full Scale IQ from 103.5 to 105.
Preintervention grade equivalents on the Reading Recognition subtest of
the Peabody Individual Achievement Test (PlAT) ranged from 3.0 to 3.3,
with percentile ranks ranging from 4 to 8. It was not possible to control for
the amount and type of previous reading instruction experienced by the
children, their present curriculum, and the amount of time spent in class-
rooms for learning-disabled youngsters. Thus, the results obtained from
this study must be evaluated in light of these confounding features.
The teaching method selected for the study was a synthetic phonics
program (Traub & Bloom, 1975). This program was chosen because of its
sequenced format, its coverage of major phonics concepts, and its famil-
iarity to the teachers in training who were providing the instruction. All
subjects were provided 1 hour of reading instruction per week (in addi-
tion to their special and regular classroom instruction) for 26 weeks.
Following the 26 hours of phonics instruction, the 30 children were
posttested with the PlAT Reading Recognition subtest, and gain scores
employing percentile ranks and grade equivalents were computed. A one-
way analysis of variance indicated significant differences among the six
subtypes for both types of gain scores achieved from preintervention to
posttesting. An analysis of subtype gain scores and subsequent pairwise
comparisons indicated that members of subtype 6 made the most progress
(mean percentile rank gain = 18.0)' followed by members of subtype 4
(mean percentile rank gain = 8.2). On the other hand, subtypes 1, 2, 3, and
5 made minimal gains in percentile ranks and were not significantly dif-
ferent from one another in terms of gains achieved. Subtypes 6 and 4 were
both significantly different from one another and from all other subtypes
with respect to their improvement in the oral reading of single words.
The data obtained in this subtype remediation study indicate that, for
some subtypes, a synthetic phonics teaching intervention appeared to
enhance significantly the ability to read single words accurately. Clearly,
members of subtypes 6 and 4 demonstrated robust improvements in their
decoding capabilities. Whether or not the absence of auditory-verbal defi-
cits in these two subtypes was associated with their good response to
instruction cannot be answered clearly at this time, but one could hypoth-
esize that this might be the case. This hypothesis is made more tenable by
the observation that those subtypes with the most severe auditory recep-
tive and auditory expressive language deficits made either minimal gains
(e.g., subtypes 2 and 3) or no gains (e.g., subtypes 1 and 5) in the ability to
pronounce single words accurately and efficiently.
In a related program of research carried out with younger disabled
readers (Lyon, 1985a; Lyon et a1., 1982), five LD subtypes were identifed
and validated internally and externally by using different variable sub-
tests, clustering algorithms, and subtype-teaching method interaction
studies. Again, a brief description of each of the subtypes' information-
processing characteristics is provided, followed by an overview of the
external validation intervention program.
Children assigned to subtype 1 (n = 18) manifested significant defi-

cits in visual perception, visual-spatial analysis and reasoning, and visu-
al-motor integration. Visual memory was also below average but not sig-
nificantly so. All measured auditory receptive and auditory expressive
skills were within the average range. The reading errors made by members
of subtype 1 appeared to be related to their diagnostic deficit profile.
Frequent mispronunciations due to confusion of visually similar words
were noted, as were reading errors involving medial vowels and vowel
combinations.
Children in subtype 2 (n = 10) displayed selective deficits in mor-
phosyntactic skills, sound blending, language comprehension, auditory
memory, auditory discrimination, and naming ability, with correspond-
ing strengths in all measured visual-perceptual skills. These deficits
across auditory receptive and auditory expressive language domains ap-
peared to seriously impede their ability to decode single words and to
apply decoding principles to the pronunciation of nonsense words.
Members of subtype 3 (n = 12) scored in the normal range on all
diagnostic measures and, thus, can be compared with youngsters in the
subtype identified by Lyon and Watson (1981) that scored significantly
below normal on reading tasks without concomitant low performance on
diagnostic test batteries. It is possible that members of subtype 3 read
inefficiently for social or affective reasons rather than because of inherent
oral language or perceptual deficiencies. It is also quite possible that the
diagnostic battery employed did not assess effectively all skills relevant to
the developmental reading process. As was the case with Lyon and Wat-
son's (1981) subtype 6 (normal diagnostic profile), members of subtype 3
scored higher than all other subgroups on the reading measures. These
youngsters did have relatively more difficulties in comprehending read-
ing passages than in the other measured reading skills. No systematic
patterns of errors could be identified from analysis of their performance
on word recognition and word attack measures.
Children in subtype 4 (n = 15) displayed significant deficiencies in
sound blending, language comprehension, auditory memory, naming
ability, and some aspects of visual perception. The difficulties manifested
by subtype 4 members in remembering, analyzing, synthesizing, and cor-
rectly sequencing verbal and visual information appeared to have a signif-
icant effect on their ability to decode phonetically regular real and non-
sense words. For example, in measures of oral reading and word attack
skills, a large proportion of subtype 4 youngsters could not approximate
the correct pronunciation of many words.
Members of subtype 5 (n = 9) manifested significant mixed deficits in
morpho syntactic skill, sound blending, visual perception, visual-motor
integration, visual-spatial analysis, and visual memory. These youngsters
committed primarily "visual" errors when reading single words (both real
and nonsense), apparently reflecting their deficiencies in visual analysis

and memory.
Following the subtype identification phase with younger disabled
readers, Lyon (1985a) carried out a pilot remediation study. Similar to the
Lyon (1983) subtype remediation study, a relatively low sample size and
other logistical difficulties (funding, sample migration) prohibited any
attempts to assign members randomly from each of the five identified
subtypes to a variety of teaching approaches. However, rather than teach-
ing all subtype members with the same general methods and materials, as
was done in the first intervention study, one subtype (subtype 2) was
split, with half of the members receiving reading instruction via a synthet-
ic phonics approach and the other half receiving instruction through a
combined whole-word and analytic phonics methodology.
Although this approach represents a significant departure from the
experimental design necessary for an aptitude (subtype)-treatment (teach-
ing method) interaction study, Lyon attempted to gain preliminary infor-
mation about how children who are very similar to one another diag-
nostically would respond to different teaching methods. Subtype 2 (n =
10) was chosen as the target subtype for this pilot study because all of its
members displayed both significant receptive language deficits (auditory
discrimination), auditory comprehension, auditory memory) and audito-
ry expressive deficits (retrieval, syntax, sequencing) within the context of
robust visual perceptual-motor-memory strengths. Because all of the sub-
type 2 members also manifested significant difficulties reading single
words and connected language, the opportunity existed to determine how
two different reading approaches affected these skills in the presence of a
number of linguistic subskill impairments.
For this pilot study, five children were randomly assigned to a syn-
thetic phonics approach (Traub & Bloom, 1975), whereas the remaining
five were placed randomly in a combined sight word, contextual analysis,
structural analysis, and analytic phonics group. Preintervention assess-
ment using the Woodcock Reading Mastery Word Identification subtest
indicated that the five children in each remediation group were reading
between the 8th and 10th percentile ranks for age. The mean percentile
ranks for the two groups were not significantly different (Mann-Whitney
Z > .05) prior to the initiation of the remediation programs.
Both remediation groups received approximately 30 hours of indi-
vidualized instruction (3 hours a week for 10 weeks). Unfortunately, it
was not possible to control for the type of previous exposure to reading
instruction or for the type of ongoing regular and special class instruction
the children were receiving in their typical school day. Thus, as in the
Lyon (1983) study, any conclusions drawn from the results of this study
must be interpreted in light of these confounding factors.
The synthetic phonics remediation group was taught via the scope
and sequence presented in the Traub and Bloom (1975) reading program.
A brief description of the instructional format for this approach was pre-
sented earlier. The combined remediation group learned to label whole
words (three nouns, three verbs) rapidly by first pairing the word with
pictures, then recognizing the names of the words (by a pointing re-
sponse), and then finally reading the words in isolation. Following the
development of rapid reading ability for these six words, function words
(the, is, was, are) were introduced and taught. Following stable reading of
these words, short sentences using combinations of the sight and function
words were constructed and read in order to introduce the concept of
contextual analysis and to develop metalinguistic awareness of reading as
a meaningful language skill. Following contextual reading drills, the com-
bined group received instruction in structural analysis and the reading
and comprehension of the morpho syntactical markers -ed, -s, and -ing.
These morphemes were written on anagrams and introduced into context
so that the children could readily grasp their effect on syntax and mean-
ing. Finally, analytic phonics drills were initiated to develop letter-
sound correspondences with the context of whole words. Specifically,
phonetically regular words that could be read rapidly by sight were pre-
sented, and children were first asked to recognize a particular letter-
sound correspondence ("Point to the letter that makes the /a/ sound") and
then to give a recall response ("What sound(s) does this letter make?"). As
children became more adept at recalling grapheme-phoneme relations,
drills in auditory analysis and blending were initiated.
Following the 30 hours of remediation, children in both groups were
posttested with an alternate form of the Woodcock Reading Mastery Word
Identification subtest. Significant differences were found between the two
remediation groups with respect to postintervention reading percentile
rank scores (Mann-Whitney Z < .0003). Children within the combined
remediation group gained, on the average, 11 percentile rank points,
whereas members of the synthetic phonics group gained approximately 1
percentile rank point.
There is little doubt that subtype 2 members responded significantly
differently to two forms of reading instruction. Apparently, the auditory
receptive and auditory expressive language deficits that characterized
each member of the subtype 2 impeded their response to a reading in-
structional method that required learning letter-sound correspondences
in isolation followed by blending and contextual reading components. A
tentative hypothesis might be that subtype 2 children did not have the
auditory language subskills necessary for success with this approach but
could deploy their relatively robust visual-perceptual and memory skills
more effectively with whole words, as seen within the combined remedia-
tion. A more tenable hypothesis is that whole-word reading placed far less
linguistic demands on these readers than alphabetic approaches that re-
quire a phonological awareness of sound structure and acoustic bound-
aries and the relationship of these units to letter sequences. Thus, whereas
subtype 2 members learned to read whole words in structured, isolated
context, their ability to generalize phonological and orthographic con-
cepts to read new words is most likely to remain limited.
In general, the data derived from this series of subtype identification
and remediation studies support a model of dyslexia that presumes that a
number of diverse information-processing deficits can have specific read-
ing disability as a common correlate. Although the results from these
basic research endeavors are interesting, the findings have limited clinical
utility for a number of reasons. First, the kinds of subtypes identified and
their descriptions are limited by the range and quality of the tests that
provide the data for cluster analysis. For example, the tasks selected for
use in Lyon's assessment batteries did not provide adequate coverage of
some linguistic factors (particularly phonology) implicated in the devel-
opmental reading process. Second, the specific nature of the relationship
between subtype assessment characteristics and response to reading in-
struction is difficult to determine because the assessment tasks are indi-
rect measures of associated symptomatology. Third, it is not well under-
stood if the correlated information-processing deficits constitute
necessary and/or sufficient conditions for reading disability. Fourth,
methodological limitations in sample size and the number and type of
dependent reading measures preclude adequate interpretations and confi-
dent generalization of the subtype-teaching method interaction studies.
Finally, even though particular teaching (treatment) approaches had dif-
ferential effects for some subtypes, it is difficult to determine if the effects
should be attributed to subtype characteristics, the instructional program,
the interaction between the two, the teacher, time spent in remediation, or
previous or concomitant educational experiences.
Clinical Subtype Intervention Studies: The Bakker Research

Program
Bakker and his colleagues (Bakker, 1983; Bakker, Moerland, & Goe-
koop-Hoefkens, 1981; Bakker & Vinke, 1985) have developed a classifica-
tion model of reading disability that stresses the importance of balance in
the functional development of the two hemispheres of the brain. He hy-
pothesized that each of two dyslexic subgroups represents functional
overdevelopment of one hemisphere for reading behavior. Thus, L-type
dyslexics, who are hypothesized to be overreliant on a left hemisphere
strategy for reading, are fast and inaccurate decoders of text, whereas P-
types are believed to be overreliant on a right hemisphere strategy re-
flected in slow, relatively accurate reading. The groups are classified in-
dependently through dichotic listening tasks, with L-types showing a
right ear advantage and P-types showing a left ear advantage.
Although a subtyping system that uses dichotic listening tasks as a
132 G. REID LYON et oJ.
key variable for classification might be constrained by unreliability (Lyon

& Risucci, 1988), Bakker's theoretical constructs have been validated ex-
ternally by intervention studies (Bakker & Vinke, 1985). In these studies,
structured presentation of reading stimuli produced changes in both
event-related potentials and reading behaviors in the direction predicted
by the balance model. The intervention paradigm assessed the effects of
both direct hemisphere stimulation through right or left visual-field pre-
sentation of words, and indirect stimulation through the presentation of
modified text formats. Thus, L-types were forced to attend to the graphic
features of text with mixed and altered typefaces, whereas P-types were
required to attend to the alphabetic code by having pictures and titles
removed from the text, as well as having to rhyme and perform temporal
order tasks. Significant changes in both event-related potentials and read-
ing improvement in experimental groups followed stimulation of the un-
derused hemisphere of the given subtype.
Although Bakker's work is promising in its demonstration of the-
oretically predicted changes in reading behavior as a consequence of
clearly conceived treatment methodologies, the work needs to be in-
terpreted with caution. Subject description and identification continue to
be problematic; for example, in the Bakker and Vinke (1985) study, sub-
jects were selected with IQs of 70 and above, and the mean IQ was a low
average of 85. There were no group differences reported with respect to
discrepancies between ability and achievement. This becomes important
only because the results may not generalize to dyslexic samples chosen by
commonly used sample marker variables. Further, the sample was teach-
er-selected, remediation sessions occurred only once per week for 22
sessions, and no control for concurrent reading instruction of the subjects
in their classrooms was reported. The L-type and P-type dyslexics were
not well differentiated according to reading behaviors. Without further
internal validation, it will be difficult to evaluate the similarities between
Bakker's classification system and others, as well as to apply Bakker's
model more generally in clinical treatment.
Direct Assessment Intervention Studies: The Flynn Research

Program
The work of one of the authors (Flynn, 1987) suggests that direct
assessment of reading and spelling behaviors within a neuropsychologi-
cal framework provides useful linkages to treatment. Assessment tasks in
Flynn's studies were chosen on the basis of a conceptualization of reading
as an interactive process composed of lower-level processes (phonology,
memory, perception, attention) and higher-level processes (syntax, se-
mantics, experiential knowledge, executive monitoring) that operate si-
multaneously and synergistically to produce fluent reading. Ecological
assessment of intact, deficient, and compensatory reading strategies in-
cluded multiple oral reading samples using the reading model to classify
errors and the Boder Test of Reading-Spelling patterns (BTRSP; Boder &
Jarrico, 1982). Thus, test content was directly related to the tasks that
children face on a daily basis.
The BTRSP was chosen as a direct measurement of word-recognition
skills involving phonology, memory, and perception. Dyslexic children
were classified as dysphonetic (deficient in sequential auditory pro-
cesses), dyseidetic (deficient in visual-spatial skills), or mixed (deficient
in both processes), according to an adaptation of the Boder procedures
(Boder & Jarriso, 1982). The initial thrust of this research was to investi-
gate the construct validity of the BTRSP through external validation stud-
ies involving quantitative neurophysiology and subtype-treatment re-
mediation programs. As the reader will note, results of the studies
indicate that a reconceptualization of the dyslexic subtypes originally
proposed by Boder may offer greater explanatory power and lead more
directly to testable reading remediation hypotheses.
With this direct assessment model, subtypes of children who demon-
strated distinct profiles of reading behaviors were identified. One sub-
type, described as deficient in phonetic development, demonstrated diffi-
culty with syntactical and alphabetic skills. Generally, semantics and
sight word recognition skills were areas of strength. Oral reading was
characterized by global substitutions, difficulty with sound-symbol rela-
tionships, and semantic substitutions. Often, comprehension was rela-
tively intact despite numerous reading inaccuracies. Conversely, children
with presumed deficits in whole word recognition skills displayed excel-
lent phonetic analysis skills. Characteristically slow but accurate readers,
they often failed to understand the meaning of a passage.
The question of whether these clinical description of reading profiles
represent reliable, replicable subtypes that have ecological and predictive
validity (i.e., lead to testable remediation hypotheses and prognoses) was
investigated through external validation studies, including quantitative
neurophysiology studies and a subtype-treatment remediation project. In
the first neurophysiology study (Flynn & Deering, 1987),44 children (ages
7-10) were assessed using direct measures of reading and spelling behav-
iors, a neurological evaluation, and spectral analysis of theta and alpha
brain waves during cognitive tasks. The results suggested that distinct
subtypes of reading-disabled children could be formed using direct mea-
sures of word recognition skills. The data also suggested that reading
disability in the subtype of children identified variously as dyseidetic
(Boder, 1971, 1973), L-type dyslexics (Bakker & Vinke, 1985), visual or
visual-spatial (Johnson & Myklebust, 1967; Mattis, French, & Rapin, 1975;
Pirozzolo, 1981) may be attributable to overutilization of early-developing
linguistic skills (phonological decoding) rather than deficient visual per-
ceptual processes.
Although this study clearly differentiated the dyseidetic dyslexics
134 G. REID LYON et al.
from other groups on three of the six cognitive tasks, the direction and
location of neurophysiologic difference was initially quite surprising
given previous reports of this subtype's characteristics. For example, dys-
eidetic children's reading disabilities have been attributed to deficient
visual-spatial abilities referable to atypical, right hemisphere develop-
ment with concomitant strength in phonetic skills (Boder, 1971, 1973;
Boder & Jarrico, 1982). In the Flynn and Deering (1987) investigation,
however, dyseidetic children demonstrated significant increases from
resting baseline in left temporal-parietal theta compared with the other
groups. That this difference occurred in the area of the angular gyrus,
presumed to be important in phonetic decoding (Hynd & Hynd, 1984),
suggested that the reportedly normal phonetic skills of dyseidetic chil-
dren may, in fact, indicate overutilization of a processing strategy that
results in, or is caused by, relative inefficiency of right hemisphere visual
gestalt abilities.
A second study of 64 children (ages 8-9), subtyped according to
modified Boder procedures (Boder & Jarrico, 1982) and oral reading pat-
terns, provided replicated evidence of increased left theta activity in dys-
eidetic children during task engagement (Flynn & Deering, 1987). Further
external validation studies investigated the efficacy of the classification
system for prescribing remediation (Flynn, 1987). Using the neuropsycho-
logical principle of teaching to intact or compensatory processes, a sub-
type-treatment research program was initiated. The study involved 22
first-, second-, and third-graders classified as dysphonetic or dyseidetic.
Children were randomly assigned to a treatment group and received a full
year of remediation, three sessions per week. The reading approaches
included (1) a language experience, analytic phonetic approach using the
Initial Teaching Alphabet; (2) Distar, a synthetic phonics approach; and
(3) a multisensory, analytic phonetic approach using regular orthography.
Data derived from the reading remediation project provided addi-
tional validation of ecologically based assessment procedures for subtyp-
ing dyslexic children (Flynn, 1987). However, the data also demonstrated
the inadequacy of a design that implies a simple match between reading
subtype and remediation system. Specifically, within each treatment con-
dition, there were some children who benefited whereas others with the
same reading profile made less than average gains.
Data from the neurophysiology and remediation studies suggest that
not all children fail to develop fluent reading behaviors for the same
reason, that distinct subtypes of dyslexic children can be identified on the
basis of direct measurements of reading and spelling behaviors, and that,
to some degree, responses of subtyped children to specific reading ap-
proaches can be predicted. The data also suggest that more dynamic,
interactive models of the teaching-learning situation must be developed
in order to describe and predict adequately how the child's processing
deficits covary with instructional variables (e.g., the reading system and
the learning context).
SOME FINAL THOUGHTS, CONCLUSIONS, AND DIRECTIONS
There is little doubt that establishing clinically valid linkages be-

tween neuropsychological assessment and the development of treatment
options for children in need of rehabilitation and remediation is a lauda-
ble goal. Ecologically valid assessment data could provide a framework
for the systematic development of hypotheses to determine which specif-
ic treatment methods and materials have the highest probability of suc-
cess for a particular child with a particular array of neuropsychological
characteristics. However, our review of efforts to achieve this goal indi-
cates that a substantial distance remains to be traveled before one can
conclude that the time spent in carrying out neuropsychological assess-
ment contributes significantly to the treatment process. The first step in
closing this distance is actually twofold. First, shortcomings inherent in
neuropsychological assessment for treatment purposes must be recog-
nized, and, second, a productive course and set of refinements to help
correct our clinical weaknesses must be suggested.
Throughout this chapter, we have noted that linkages between assess-
ment and treatment are weakened by factors associated with the limited
construct, content, and ecological validities of the test batteries employed.
We also have pointed out that a number of persistent difficulties in relat-
ing assessment to treatment stem from our use of assessment tasks that
were not designed initially to predict and guide treatment options. In
addition, specific issues associated with clinical training, developmental
appropriateness of assessment tasks used with children, and the problem
of static versus dynamic assessment also need to be addressed if clinical
progress is to be made regarding the assessment-treatment interface. Our
intent in the remainder of this chapter is to provide some meaningful
direction with respect to these issues.
Professional Preparation and Experience

A majority of clinical neuropsychologists feel unprepared to under-
take the task of recommending treatment or intervention options on the
basis of assessment data. This may be expected, as Craig (1979) and others
(Hynd & Obrzut, 1986) have pointed out, because clinical neuro-
psychologists spend significantly greater clinical time in assessment ac-
tivities than practicing in an intervention context. Although this lack of
experience with treatment issues is problematic, a more fundamental rea-
son for clinical naivete in treatment situations is limited preparation. As
Hynd and Obrzut (1986) have reported, clinical neuropsychologists re-

ceive little or no formal training in intervention practices, particularly
educational practices. Training programs typically emphasize assess-
ment, diagnosis, and consultation with only general exposure to treat-
ment methodologies. Moreover, when neuropsychologists are exposed to
intervention and treatment approaches, such experiences usually take
place in settings somewhat removed from the populations and problems
that ultimately will demand expert clinical treatment services. More spe-
cifically, the types of intervention and treatment strategies applied with
adults in hospital settings often are not generalizable to the school-related
difficulties seen in pediatric populations.
If, in fact, clinical neuropsychology is to undertake a responsible role
in applied settings where treatment is not only a logical but necessary
outcome of assessment, then substantial improvements in formal clinical
training must occur. Supervised practica and internships under the tu-
telage of master teachers and clinicians, preceded by relevant coursework
in applied academic content areas, would seem helpful in this regard.
Developmental Issues in Assessment

In addition to improvements in professional preparation, advances in
establishing clinically meaningful linkages between assessment and treat-
ment could be fostered by emphasizing an understanding of developmen-
tal factors in child clinical practice. Test design, interpretation of data,
and the application of instructional methodologies must be informed by a
developmental perspective. A prime example of a neuropsychologically
based assessment tool that could benefit from a reexamination of its de-
velopmental appropriateness is the Boder Test of Reading-Spelling Pat-
terns (BTRSP; Boder & Jarrico, 1982). This instrument, which is used in
some studies to assign dyslexics to dysphonetic, dyseidetic, and mixed-
deficit subtypes, is a clinical-inferential tool yielding a classification
based on direct assessment of reading and spelling behavior. It also is
employed in clinical practice as a basis for remedial prescriptions. Al-
though the test is designed ostensibly for all school-aged children, the
methods for error analysis and classification of children to subtypes do
not change with age. There is insufficient consideration given to the man-
ner in which spelling strategies shift with normal development. Thus, a
diagnostic term such as dysphonetic may capture the essence of the prob-
lem at one beginning stage of reading and spelling but mask the essence of
the difficulty at more advanced developmental stages. Indeed, spelling
strategies do shift in dyslexics over time (Cook, 1981), both in response to
instruction and as a function of intelligence (Moats, 1983). This is but one
example of how assessment data can be confounded by the lack of an
informed developmental perspective.
No doubt, the majority of neuropsychological batteries that are down-

ward extensions of adult batteries and which do not have adequate cross-
sectional norms suffer from the same developmental inadequacies. Cer-
tainly, it would be erroneous to prescribe remedial treatment techniques
on the basis of these types of assessment data until external validation
studies are conducted to evaluate the relationships among assessment
strategies, clinical subtypes, developmental level, and response to
instruction.
The Need for Dynamic Assessment

As mentioned throughout this chapter, most neuropsychological tests
are static measures of competence designed to allow standardized com-
parisons of a child with other children in specified populations. Such
comparisons yield only indirect and nonspecific information regarding
the nature of the child's problem as it is manifested in his learning con-
text. The test data do not address motivational factors and their impact on
learning, nor do they inform the teacher or clinician about the child's
spontaneous use of strategies or procedural knowledge in learning.
One notable exception is the Halstead Category Test, one of the most
sensitive measures of brain dysfunction, which does assess the child's
ability to learn from corrective feedback on one kind of nonverbal prob-
lem-solving task. Its potential as a dynamic assessment tool is great, but it
usually is used merely as a normative measure to add to the evidence for
or against the presence of brain dysfunction. In fact, the power of the
Category Test lies in its ability to sample the domains of impairment most
commonly associated with brain dysfunction, such as adaptation, gener-
alization, memory for recent experience, conceptual organization, con-
sistency of response capability, and ability to reason abstractly. As a test,
it successfully departs from truncated, structured, single ability measures
that do not sample adequately the cognitive characteristics most relevant
to academic performance and adaptation in general. More tools of this
kind are needed in our assessment batteries if neuropsychology is going to
provide evaluations relevant for the individualization of therapy or
teaching.
At the present time, one of the more promising examples of dynamic
neuropsychological assessment is the work of Ylvisaker and his col-
leagues (Szekeres, Ylvisaker, & Cohen, 1987) in closed head injury re-
habilitation. Their delineation of variables to consider in the selection
and training of compensatory strategies (Haasbauer-Krupa, Henry,
Szekeres, & Ylvisaker, 1985) acknowledges the complex interaction of
cognitive, motivational, and personality factors in recovery from head
injury. Their approach to rehabilitation involves dynamic observations of
patient response to real-life situations and meaningful activities in natural
settings. Ylvisaker and associates also systematically evaluate patients'

performances according to (1) the efficiency or rate of performance, (2) the
level of conceptualization or mastery that the patient can comprehend, (3)
the scope or variety of contexts in which performance can be maintained,
and (4) the manner of performance (impulsive-reflective, flexible-rigid,
active-passive, dependent-independent). Functional integrative perfor-
mance is the appropriate target of therapy, rather than the direct re-
habilitation of basic cognitive processes such as attention and memory,
which may result in improvements on neuropsychological tests but
which may not generalize to everyday life.
Given the arguments against static measurement (failure of tests to
predict real-life adaptation, to identify satisfactorily the impact of specific
cognitive impairments on complex adaptive behavior, and to predict re-
sponses to specific interventions), one might ask if child neuropsycholog-
ical test batteries, as currently designed, have any purpose or relevance in
the treatment of documented or putative developmental disorders.
Should we delete their use in rehabilitation settings? Should we bypass
neuropsychological testing of learning-disabled students, for example, in
favor of curriculum-based assessment or analysis of the learner's charac-
teristics during instruction in a given context? We do not believe so. A
number of advances have been made in the development of neuropsycho-
logical assessment batteries that consist of tasks that are ecologically valid
and direct measurements of processes crucial to the development of read-
ing and spelling behavior (Flynn, this chapter; Hynd, 1986). There is also
an increased awareness that establishing valid linkages between assess-
ment data and treatment options will require tasks and modes of clinical
interpretation that are as dynamic and flexible as the learning process
itself (Lyon & Moats, in press).
The Need for Continued Classification Research

In order to relate assessment findings ultimately to treatment options
and predicted outcomes, the phenotypic or behavioral expression of dif-
ferent developmental disorders must be clearly defined, characterized,
and categorized. To this end, subtype research must capitalize on what we
know about theoretically driven dynamic assessment procedures, the de-
velopmental nature of both brain function and human learning, and the
methodological requirements for taxonomic research (Lyon & Risucci,
1988). Some subtyping efforts (e.g., Bakker, 1983; Flynn, 1987; Flynn &
Deering, 1987; Hynd, 1986; Lyon, 1985b; Lyon & Risucci, 1988) are mak-
ing some advances in these directions with an eye toward treatment rele-
vance for children. Significant work remains to be done, however, before
any classification scheme has precise descriptive, communicative, and
predictive power. An informed understanding of some of the issues raised
in this chapter may help clinicians fill the therapeutic breech until sys-
tematic and reliable linkages between assessment and treatment are

forged.
REFERENCES
Alfano, D. P., & Finlayson, M. A. J. (1987). Clinical neuropsychology in rehabilitation.
Clinical Neuropsychologist, 1, 105-123.
Ayres, R R, & Cooley, E. J. (1986). Sequential versus simultaneous processing on the K-ABC:
Validity in predicting learning success. Journal of Psychoeducational Assessment, 4,
211-220.
Bakker, D. (1983). Hemispheric specialization and specific reading retardation. In M. Rutter
(Ed.), Developmental neuropsychiatry (pp. 203-232). New York: Guilford Press.
Bakker, D., & Vinke, J. (1985). Effects of hemisphere specific stimulation on brain activity
and reading in dyslexics. Journal of Clinical and Experimental Neuropsychology, 7,
505-525.
Bakker, D. J., Moerland, R, GoeKoop-Hoefkens, M. (1981). nlltcts of hemisphere-specific
stimulation on the reading performance of dyslexic boys: A pilot study. Journal of
Clinical Neuropsychology 3, 155-159.
Beaumont, J. G. (1983). Introduction to neuropsychology. New York: Guilford Press.
Benson, D. F., & Geschwind, N. (1975). The alexias. In D. J. Vinkis & G. W. Bruyn (Eds.),
Handbook of clinical neurology (Vol. 4). New York: American Elsevier.
Boder, E. (1971). Developmental dyslexia: Prevailing diagnostic concepts and a new diag-
nostic approach. In H. Mykelbust (Ed.), Progress in learning disabilities (Vol. 2). New
York: Grune and Stratton.
Boder, E. (1973). Developmental dyslexia: A diagnostic approach based on three atypical
reading-spelling patterns. Developmental Medicine and Child Neurology, 15,663-687.
Boder, E. D., & Jarrieo, S. (1982). The Boder Test of Reading-Spelling Patterns. A diagnostic
screening test for subtypes of reading disability. Orlando, FL: Grune and Stratton.
Boll, T. J. (1981). The Halstead-Reitan Neuropsychological Test Battery. In S. B. Filskov & T.
J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 577-607). New York: Wiley.
Boll, T. J., & Reitan, R M. (1972). Comparative ability interrelationships in normal and brain-
damaged children. Journal of Clinical Psychology, 28, 152-156.
Bornstein, R A., King, G., & Carroll, A. (1983). Neuropsychological abnormalities in Gilles
de la Tourette's syndrome. Journal of Nervous and Mental Disease, 171,497-502.
Brown, A. 1., & Campione, J. C. (1986). Psychological theory and the study of learning
disabilities. American Psychologist, 14, 1059-1068.
Chall, J. S., & Mirsky, A. F. (1978). The implications for education. In J. S. Chall & A. F.
Mirsky (Eds.), Education and the brain: The seventy-seventh yearbook of the National
Society for the Study of Education. Chicago: University of Chicago Press.
Cook, 1. (1981). Misspelling analysis in dyslexia. Observation of developmental strategy
shifts. Bulletin of the Orton Society, 31, 123-134.
Craig, D. 1. (1979). Neuropsychological assessment in public psychiatric hospHals: The
current state of practice. Clinical Neuropsychology, 1, 1-7.
Dawson, G. (1983). Lateralized brain dysfunction in autism: Evidence from the Halstead-
Reitan Neuropsychological Battery. Journal of Autism and Developmental Disorders,
13, 269-286.
Diller, 1., & Gordon, W. A. (1981a). Intervention for cognitive deficits in brain-injured adults.
Journal of Consulting and Clinical Psychology, 49, 822-839.
Diller, L., & Gordon, W. A. (1981b). Rehabilitation and clinical neuropsychology. In S. B.
Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 702-7330). New
York: Wiley.
Diller, 1., & Weinberg, J. (1977). Hemi-inattention in rehabilitation: The evolution of a
140 G. REID LYON et al.
rational remediation program. In E. A. Weinstein & R. P. Friedland (Eds.), Advances in

neurology (Vol 18, pp. 62-82). New York: Raven Press.
Dunleavy, R. A., & Baade, 1. A. (1980). Neuropsychological correlates of severe asthma in
children 9-14 years old. Journal of Consulting and Clinical Psychology, 48, 564-577.
Finlayson, M. A. J., Gowland, C., & Basmajian, J. V. (1986). Neuropsychological predictors of
treatment response following stroke. Journal of Clinical and Experimental Neuropsy-
chology, 7, 647. (Abstract)
Finlayson, M. A. J., Johnson, K. A., & Reitan, R. M. (1977). Relationship of level of education
to neuropsychological measures in brain-damaged and non brain-damaged adults. Jour-
nal of Consulting and Clinical Psychology, 45, 536-542.
Flynn, J. (1987). Neurophysiologic characteristics of dyslexic subtypes and response to
remediation. Grant awarded by the Initial Teaching Alphabet Foundation, Roslyn, New
York.
Flynn, J. & Deering, W. (1987). Subtypes of dyslexia: Investigation of Boder's system for
identification and classification of dyslexic children using quantitative neurophys-
iology. Manuscript submitted for publication.
proach. New York: Springer-Verlag.
Geary, D. C., & Gilger, J. W. (1984). The Luria-Nebraska Neuropsychological Battery-Chil-
dren's Revision: Comparison of learning disabled and normal children matched on Full
Scale IQ. Perceptual and Motor Skills, 58, 115-118.
Golden, C. J., Hammeke, T., & Purisch, H. (1978). Diagnostic validity of a standardized
neuropsychological battery derived from Luria's neuropsychological tests. Journal of
Consulting and Clinical Psychology, 46, 1258-1265.
Gunnison, J. A. (1984). Developing educational intervention from assessments involving the
K-ABC. Journal of Special Education, 18, 325-344.
Haasbauer-Krupa, J., Henry, K., Szekeres, S. & Ylvisaker, M. (1985). Cognitive rehabilitation
therapy, late stages of recovery. In M. Ylvisaker (Ed.), Head injury rehabilitation: Chil-
dren and adolescents. San Diego: College Hill Press.
Hartlage, 1. C., & Hartlage, P. 1. (1977). Application of neuropsychological principles in the
diagnosis of learning disabilities. In 1. Tamapol & M. Tamapol (Eds.), Brain function
and reading disabilities. Baltimore: University Park Press.
Hartlage, 1. C., & Telzrow, K. F. (1983). The neuropsychological basis of educational inter-
vention. Journal of Learning Disabilities, 16, 521-526.
Herman, B. P. (1982). Neuropsychological function and psychopathology in children with
epilepsy. Epilepsia, 23, 545-554.
Hynd, C. R. (1986). Educational intervention in children with developmental learning disor-
ders. In J. E. Obrzut & G. W. Hynd (Eds.)' Child neuropsychology (Vol. 2, pp. 265-297).
Orlando, FL: Academic Press.
Hynd, G., & Hynd, C. (1984). Dyslexia: Neuroanatomicalineurolinguistic perspectives. Read-
ing Research Quarterly, 29, 482-498.
Hynd, G. W., & Obrzut, J. E. (1986). Clinical child neuropsychology: Issues and perspectives.
In J. E. Obrzut & G. W. Hynd (Eds.). Child neuropsychology (Vol. 2, pp. 3-14). Orlando,
FL: Academic Press.
Johnson, D. J., & Mykelbust, H. R. (1967). Learning disabilities: Educational principles and
practices. New York: Grune and Stratton.
Kaufman, A. S., & Kaufman, N. 1. (1983). Kaufman Assessment Battery for Children: In-
terpretive manual. Circle Pines, MN: American Guidance Service.
Luria, A. R. (1963). Restoration of function after brain injury. New York: Macmillan.
Luria, A. R. (1966). Higher cortical functions in man. New York: Basic Books.
Luria, A. R. (1973). The working brain: An introduction to neuropsychology. New York:
Basic Books.
Luria, A. R. (1980). Higher cortical functions in man, (2nd ed). New York: Basic Books.
Luria, A. R, & Tzetkova, L. S. (1968). The re-education of brain-damaged patients and its
psychopedagogical application. In J. Hellmuth (Ed.), Learning disorders (pp. 139-154).
Seattle: Special Child Publications.
Lyon, G. R (1983). Subgroups of learning disabled readers: Clinical and empirical identifica-
tion. In H. R Mykelbust (Ed.), Progress in learning disabilities (Vol. 5, pp. 103-134).
New York: Grune and Stratton.
Lyon, G. R (1985a). Educational validation of learning disability subtypes: In B. P. Rourke
(Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis (pp. 228-
Lyon, G. R (1985b). Identification and remediation of learning disability subtypes: Prelimi-
nary findings. Learning Disabilities Focus, 1, 21-35.
Lyon, G. R (1987). Learning disabilities research: False starts and broken promises. In S.
Vaughn & C. Bos (Eds.), Research in learning disabilities: Issues and future directions
(pp. 69-85). San Diego: College Hill Press.
Lyon, G. R, & Moats, L. C. (in press). Critical issues in the instruction of the learning
disabled. Journal of Consulting and Clinical Psychology.
Lyon, G. R, Rietta, S., Watson, B., Porch, B., & Rhodes, J. (1981). Selected linguistic and
perceptual abilities of empirically derived subgroups of learning disabled readers. Jour-
nal of School Psychology, 19, 152-166.
Lyon, G. R, & Risucci, D. (1988). Classification issues in learning disabilities. In K. A. Kavale
(Ed.), Learning disabilities: State of the art and practice. San Diego: College Hill Press.
Lyon, G. R, Stewart, N., & Freedman, D. (1982). Neuropsychological characteristics of em-
pirically derived subgroups of learning disabled readers. Journal of Clinical Neuropsy-
chology, 4, 343-365.
Lyon, G. R, & Toomey, F. (1985). Neurological, neuropsychological, and cognitive-develop-
mental approaches to learning disabilities. Topics in Learning Disabilities, 2, 1-15.
Lyon, G. R, & Watson, B. (1981). Empirically derived subgroups of learning disabled readers:
Diagnostic characteristics. Journal of Learning Disabilities, 14, 256-261.
Mann, 1. (1979). On the trail of process. New York: Grune and Stratton.
Mann, 1., & Sabatino, D. A. (1985). Foundations of cognitive process in remedial and special
education. Rockville, MD: Aspen.
Mattis, S., French, J., & Rapin, I. (1975). Dyslexia in children and young adults. Three
independent neuropsychological syndromes. Developmental Medicine and Child Neu-
rology, 17, 150-163.
Moats, 1. (1983). A comparison of the spelling errors of older dyslexic and normal second
grade children. Annals of Dyslexia, 33, 121-140.
Newcomb, F. (1985). Neuropsychological qua interface. Journal of Clinical and Experimen-
tal Neuropsychology, 7. 663-681.
Nolan. D. R, Hammeke. T. A., & Barkley, R A. (1983). A comparison of the neuropsychologi-
cal performance in two groups of learning disabled children. Journal of Clinical Child
Obrzut. J. E., & Hynd. G. W. (1986). Child neuropsychology (Vol. 2). Orlando. FL: Academic
Press.
Pirozzolo. F. (1981). Language and brain: Neuropsychological aspects of developmental
reading disability. School Psychology Review. 10. 350-355.
Plaisted, J. R, Gustavson, J. 1., Wilkening, G. N., & Golden, C. J. (1983). The Luria-Nebraska
Neuropsychological Battery-Children's Revision: Theory and current research findings.
Journal of Clinical Child Psychology, 12, 13-21.
Rao. S. M., & Bieliauskas, 1. A. (1983). Cognitive rehabilitation two and one-half years post
right temporal lobectomy. Journal of Clinical Neuropsychology. 5. 313-320.
Reed. H. B. C., Reitan, R M., & Klove, H. (1965). Influence of cerebral lesions in psychologi-
cal test performance of older children. Journal of Consulting Psychology, 29. 247-251.
Reitan, R. M. (1966). A research program on the psychological effects of brain lesions in
human beings. In R. M. Ellis (Ed.), International review of research in mental retarda-

tion (pp. 153-218). New York: Academic Press.
Reitan, R. M. (1969). Manual for administration of neuropsychological test batteries for
adults and children. Indianapolis: Author.
Reitan, R. M. (1979). Neuropsychology and rehabilitation. Tucson: Author.
Reitan, R. M. (1980). REHABIT-Reitan evaluation of hemispheric abilities and brain im-
provement training. Tucson: Reitan Neuropsychological Laboratory and University of
Arizona.
Reitan, R. M., & Boll, T. J. (1973). Neuropsychological correlates of minimal brain dysfunc-
tion. Annals of the New York Academy of Sciences, 205, 65-88.
Reitan, R. M., & Davison, L. A. (1974). Clinical neuropsychology: Current status and applica-
tions. Washington, DC: V. H. Winston.
Reitan, R. M., & Wolfson, D. (1985). The Halstead-Reitan Neuropsychological Test Battery:
Theory and clinical interpretation. Tucson: Neuropsychology Press.
Rourke, B. P. (1975). Brain-behavior relationships in children with learning disabilities: A
research program. American Psychologist, 30, 911-920.
Rourke, B. P. (1981). Neuropsychological assessment of children with learning disabilities.
In S. B. Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 453-478).
New York: Wiley.
Rourke, B. P., Bakker, D., Fiske, J. L., & Strang, J. D. (1983). Child neuropsychology: An
introduction to theory, research, and clinical practice. New York: Guilford Press.
Rourke, B. P., Fiske, J. L., & Strang, J. D. (1986). The neuropsychological assessment of
children: A treatment-oriented approach. New York: Guilford Press.
Salvia, J., & Hritcko, T. (1984. The K-ABC and ability training. The Journal of Special
Education, 18, 345-356.
Selz, M. J., & Reitan, R. M. (1979). Rules for neuropsychological diagnosis: Classification of
brain function in older children. Journal of Consulting and Clinical Psychology, 47,
258-264.
Sternberg, R. J. (1984). An information processing analysis and critique. Journal of Special
Education, 18, 269-279.
Stoddart, C., & Knights, R. M. (1986). Neuropsychological assessment of children: Alter-
native approaches. In J. E. Obrzut & G. W. Hynd (Eds.), Child neuropsychology (Vol. 2,
pp. 229-244). Orlando, FL: Academic Press.
Szekeres, S., Ylvisaker, M., & Cohen, S. (1987). A framework for cognitive rehabilitation
therapy. In M. Ylvisaker & E. M. R. Gobble (Eds.), Community re-entry for head-injured
adults. Boston: Little, Brown.
Teeter, P. A. (1986). Standard neuropsychological test batteries for children. In J. E. Obrzut &
G. W. Hynd (Eds.), Child neuropsychology (Vol. 2, pp. 187-227). Orlando, FL: Academ-
ic Press.
Tramontana, M. G., Klee, S. N., & Boyd, T. A. (1984). WISC-R interrelationships with the
Halstead-Reitan and Children's Luria Neuropsychological Batteries. Clinical Neuropsy-
chology, 6, 1-8.
Traub, N., & Bloom, C. (1975). Recipe for reading. Cambridge, MA: Educators Publishing
Service.
Yendall, L. T., Fromm-Auch, D., & Davies, P. (1982). Neuropsychological improvement of
persistent delinquency. Journal of Nervous and Mental Disease, 170,257-265.
Ysseldyke, J. E., & Algozzine, B. (1982). Critical issues in special and remediation education.
Boston: Houghton Mifflin.
Ysseldyke, J. E., & Mirkin, P. K. (1981). The use of assessment information to plan instruc-
tional interventions: A review of the research. In C. Reynolds & T. Gutkin (Eds.), A
handbook for school psychology (pp. 113-132). New York: Wiley.
III
Special Topics in Assessment

6
Attention
RUSSELL A. BARKLEY
INTRODUCTION
Attention is a multidimensional construct referring to a variety of rela-

tionships between environmental stimuli or tasks and behavioral re-
sponses. Although it has enjoyed widespread investigation in both
humans and animals, this was not always so. Much of our understanding
of this complicated psychological construct comes from investigations
and reviews of the literature published since 1960, although it received
some consideration by earlier prominent psychologists such as James
(1890) and Titchener (1924).
Attempts at defining a global construct of attention have proven diffi-
cult because of its involvement with so many different activities or behav-
iors across a diversity of tasks and contexts. Instead, more recent investi-
gators (see Egeth & Bevan, 1973; Hale & Lewis, 1979) have subdivided it
into more specific components, such as selective or sustained attention,
distractibility, orientation, search, and alertness. Even these components,
however, are not easily operationalized and one typically must examine
the nature of the task used to assess the construct or component before
fully appreciating what actually is being measured. Moreover, these com-
ponents overlap with other mental or neuropsychological constructs,
such as memory and its aspects, motor planning and execution, arousal,
impulsivity, and the regulation of behavior by rules, to name but a few.
This often has led to confusion in efforts to infer the nature of attention
and its deficits in individuals with neuropsychological impairment in
whom other related brain functions also may be deficient.
Given this theoretical and experimental guagmire, it is easy to see
RUSSELL A. BARKLEY Department of Psychiatry, University of Massachusetts Medical

Center, Worcester, Massachusetts.
145
146 RUSSELL A. BARKLEY
why relatively simple measures of attention typically have not been in-
cluded in the test batteries frequently given by either clinical child psy-
chologists (Rosenberg & Beck, 1986) or child neuropsychologists (Plai-
sted, Gustavson, Wilkening, & Golden, 1983; Rourke, 1981). What types of
attention should be assessed and how to do so conveniently in routine
clinical practice remain unresolved issues in clinical child neuropsychol-
ogy. Yet much of the research literature on attention has been conducted
with children, and a considerable amount of this with children having a
quasi-neuropsychological disorder known as attention deficit disorder
(ADD), with or without hyperactivity (American Psychiatric Association,
1980). In fact, ADD is probably the most well studied of childhood psy-
chological disorders (Barkley, 1988; Ross & Ross, 1982; Weiss &
Hechtman, 1979).
It is the purpose of this chapter to draw upon this wealth of research
to provide (1) an appreciation for the importance of attention in child
neuropsychological assessment, (2) an understanding of its components,
(3) a brief survey of currently available methods of clinical assessment,
and (4) awareness of some of the critical issues that remain to be ad-
dressed in each of these domains. Psychophysiological measures or corre-
lates of attention will not be reviewed in this chapter owing to their rare
inclusion in neuropsychological assessments of children and their indi-
rectness as indices of attention as a cognitive or behavioral construct.
Instead, the focus of this chapter is upon behavioral measures of attention
and its components.
Importance of Attention in Child Neuropsychology

To understand the importance of attentional deficits and their assess-
ment, one need only examine the prevalence of these deficits within those
populations likely to be seen by child psychologists and neuropsycholo-
gists.
Prevalence of Attention Deficits

Epidemiological surveys suggest that as many as 49% of boys and
27% of girls are described as inattentive by their teachers (Lapouse &
Monk, 1958; Werry & Quay, 1971). Clearly, being inattentive is a relatively
common occurrence in normal children. Serious deficits in attention (e.g.,
greater than 1.5 to 2 standard deviations from the mean) appear to occur
in at least 3 to 10% of the school-age population of children, making them
among the most prevalent of all childhood neuropsychological disorders
(Barkley, 1981; Ross & Ross, 1982). Most of these children are diagnosed
as having an attention deficit disorder-a disorder characterized by de-
velopmentally inappropriate levels of inattention, impulsivity, and, in
many cases, overactivity.
ATTENTION 147
Attentional deficits are also commonly associated with other child

psychiatric and developmental disorders, such as autism, pervasive de-
velopmental disorders, such as autism, pervasive developmental disor-
der, depression, conduct disorders, and learning disabilities (Mash & Ter-
dal, 1988). Further, childhood neurological disorders often have attention
deficits among their sequelae. Children experiencing closed head injury
(Boll, 1983), leukemia and its treatments (Brouwers, Riccardi, Poplack, &
Fedio, 1984), epilepsy and its treatments (Holdsworth & Whitmore, 1974;
Wolf & Forsythe, 1978), anoxia/hypoxia (0 'Dougherty, Nuechterlein, &
Drew, 1984), tic disorders and Gilles de la Tourette's syndrome (Barkley,
1988b), lead and other toxin poisonings (Needleman et 01., 1979), or ex-
posure to biological hazards during the perinatal period (Nichols, 1980)
all have an increased incidence of attentional deficits-some as high as
70% (e.g., Tourette's syndrome). Child neuropsychologists are therefore
likely to encounter a large case load of children with some type and
degree of deficiency in attention and must be knowledgeable of its
assessment.
Interaction of Attention with Other Neuropsychological

Functions
The necessity for an understanding and assessment of attention in
child neuropsychology is further underscored by the almost ubiquitous
influence of attention on other cognitive or neuropsychological abilities
and their assessment. It virtually goes without saying that few standard
psychological or neuropsychological tests can be administered without a
modicum of selective and sustained attention by the child. Attention
plays a key role not only in the evaluation of other neuropsychological
functions but also in the proficiency of these functions. Deficits in some
types of short-term memory problem-solving strategies, effectiveness of
both perceptual and memory search strategies, and motor planning, coor-
dination, and execution are commonly seen in children with attention
deficits (Douglas, 1983). These attentional deficits may have negative ef-
fects on the child's social interactions with parents, peers, and teachers
(Barkley, 1985; Whalen, Henker, & Dotemoto, 1980). Given these findings,
it is clear that deficits in attention can have wide-ranging effects on the
child's cognitive, academic, and social functioning. Surely such an influ-
ential variable as attention deserves greater emphasis in our evaluation of
children.
Use of Attentional Measures in Clinical Practice

Regrettably, the use of standardized or systematic measures of atten-
tion in clinical practice has not been commonplace. Recent surveys of
clinical child psychologists have shown that the vast majority rely on
parent interviews, informal observations of behavior in the office, intel-

ligence measures, or drawing-and-copying tests to assess attentional defi-
cits (Rosenberg & Beck, 1986). As will be shown later, each of these
sources of information is highly flawed as a method of evaluating atten-
tional deficits. Standard practice in child neuropsychology is not much
better in that descriptions of neuropsychological assessment batteries
(Boll, 1981; Plaisted et aI., 1983) do not typically include systematic or
standardized assessments of attention and its components. In general,
current clinical practices for evaluating attention either are highly impres-
sionistic or rely on poorly validated measures.
Components of Attention
Numerous attempts have been made to define attention, many of
which are lacking in precision and operational terms (Gibson & Rader,
1979). Early philosophers viewed attention as a mental faculty dispensed
by organisms from a limited pool or capacity. In such conceptualizations,
attention is "paid" or deployed by the organism as if it were withdrawn
from a limited cognitive bank account or staging area that could be de-
pleted (Mackworth, 1976). Other theorists speculate that there is not just a
limited reservoir of attention but also a limited range of stimuli or ac-
tivities to which an organism can optimally attend simultaneously. What
determines these limits is not clear in these formulations.
Current conceptualizations as well as experimental research with
children suggest that attention may be best understood as a multidimen-
sional construct having numerous components (Piontrowski & Calfee,
1979). There exists at least some consensus for the following dimensions:
Alertness or arousal appears to refer to the degree of general sensitivity of
the child to the environment. Alertness seems to mean a generalized
wakefulness or state of responsivity, but to nothing in particular. Selec-
tive or focused attention most commonly refers to the child's ability to
focus on critical stimuli, or those essential to the task, while ignoring
unessential elements. Deficits in this component may be thought of as
inattention. It is closely linked to distractibility, which refers to the extent
to which a child responds to the unessential or irrelevant aspects of a task.
Although it is not typically considered a component of attention, exten-
sive research of ADD children (Douglas & Peters, 1979; Milich & Kramer,
1984) suggests that impulsivity closely overlaps with these aspects of
attention. It often means the speed and accuracy with which a child reacts
to an event or task. Rapid, inaccurate responding is defined as impulsivity
(Milich & Kramer, 1984). Sustained attention means the duration of a
child's responding to a task or stimulus and is sometimes referred to as
vigilance. It seems more accurate, however, to view vigilance as an aspect
of sustained attention in which the child persists in directing his sensory
ATTENTION 149
perception to a task so as to detect and respond to the sporadic occurence

of a target stimulus. Span of apprehension has been used to refer to the
number of stimuli or amount of information to which a child can effec-
tively attend simultaneously. The term search denotes the strategies or
rules that a child may employ in interacting with or attending to a stim-
ulus.
Hemi-inattention, or neglect, is a neuropsychological term referring
to a rare disorder of selective attention involving failure to respond to
target stimuli occurring in a particular sensory half-field or quadrant
while responding to the same target stimulus in the opposing half-field or
remaining quadrants (see Heilman & Valenstein, 1985). In such cases, it
must be clear that the individual retains sensory perception in the partic-
ular sensory field in question (e.g., is not blind or does not have a visual
field cut) but fails to respond to certain stimuli placed within that field. It
is more often evident in adults than in children after lesions to the right
posterior cerebral hemisphere, has most often been reported in vision as
compared with other sensory modalities, and is more commonly found in
left versus right sensory fields. When asked to copy a cross on a page,
patients with hemi-inattention ignore the left half of the cross, drawing
only its right side. Such male patients may shave only the right side of the
face or dress only the right side of the body. When seen in children,
neglect is usually an acute symptom noted immediately following a le-
sion. It resolves rapidly, usually within a matter of hours. Although, in its
most dramatic form, neglect is a rare condition, little research has been
conducted on whether lesser degrees of hemi-inattention exist in adults
or children with a history of posterior lesions of the right hemisphere or
impairment in functions subserved by this cortical region. Such a deficit
might appear as a slight decrease in the probability of detecting stimuli in
one hemifield versus another over numerous trials on a task.
These components of attention do not appear to be independent.
Deficits in one component of attention, such as vigilance, are often associ-
ated with deficits in others, such as impulsivity, selective attention, and
distractibility (Bremer & Stern, 1976; Klee & Garfinkel, 1983). The rela-
tionships, although statistically significant, are of a low to moderate de-
gree. Whereas overlap exists across the components of attention, they are
not identical and deserve recognition as separate, semi-independent
components.
Most standard approaches to psychological or neuropsychological
assessment rarely assess all of the components of attention. This is not to
say that every aspect of attention should be evaluated in every child
referred for assessment. Nevertheless, both clinician and researcher need
to be sensitive to the existence of these components, assessing selective
components more objectively where history, current concerns, or clinical
impressions suggest potential deficits.
Developmental Considerations
Because children's mental and physical abilities seem almost in a
constant state of change and maturation, attention and its components
should be found to vary considerably with development. Ample research
documents an increase in the specificity of perception or focus of atten-
tion with increasing age (Gibson & Rader, 1979). This progressive differ-
entiation of selective attention with experience and maturation results in
greater economy of effort as children learn to distinguish "signal" from
"noise" more efficiently, or relevance from irrelevance in stimulus dis-
plays. Also with maturation comes a lengthening of children's attention
span, such that more time is spent looking at, exploring, manipulating,
and generally interacting with an object than occurred in prior stages of
development (Milich, 1984). Children demonstrate increased vigilance or
duration of sustained perception to a signal detection task as well as a
commensurate decrease in impulsivity with increased maturation (Gor-
don, McClure, & Post, 1983; Salkind & Nelson, 1980). After age 12 years,
response time remains relatively unchanged whereas accuracy continues
to improve, implying an ongoing development in efficiency of attentional
search strategies at these later ages.
Seemingly at odds with the progressive specificity of attention with
age, children also display greater breadth of attention or span of ap-
prehension with development. In other words, they are able to attend to a
far larger array of stimuli simultaneously as they mature (Gibson & Rader,
1979). This should not be surprising, however, given that greater specific-
ity or sharpness of attention may permit older children to monitor and
respond simultaneously to larger and more complex stimulus arrays with
heightened efficiency. Along with these changes appears to come a de-
cline in their proneness to distraction by task-irrelevant stimuli (Wood-
cock, 1976). Condry, McMahon, and Levy (1979) have shown that older
children also seem better able to effectively utilize warning stimuli, or
information preceding a task or activity, than do younger children. This
suggests a greater preparedness to attend with increasing age in children
(Gibson & Rader, 1979). Perhaps related to this, children acquire and
utilize progressively more efficient, complex, and effective search strat-
egies over time so as to expend attention more economically during prob-
lem-solving activities (Salkind & Nelson. 1980).
With these changes in attention, particularly those related to greater
sustained attention, the duration of children's compliance with parental
commands and requests increases (Barkley, Karlsson, & Pollard, 1985).
This is accompanied by a corresponding decline in the amount of direc-
tion, commands, and control exerted by caregivers over the children's
play and task performance activities. Such findings underscore the impor-
tance of attention in social functioning. Deficits in the components of
ATTENTION 151
attention emanate into social exchanges with others, resulting in re-

ciprocal changes in the manner in which such children are treated by
others (Whalen & Henker, 1980).
A Behavioral Conceptualization of Attention

In a functional analysis, attention is not viewed as a behavior, cog-
nitive ability, or mental construct. It refers to a relationship between
stimuli (e.g., tasks or events) and behavioral responses. As such, it de-
scribes a correlation, or a lack of one, between the environment and chil-
dren's activities. Labeling a lack of or weak correlation between two vari-
ables as a "deficit" or "disorder" in the child, as in attention deficit
disorder, is ludicrous, for we have simply blamed a low-order correlation
as the basis for the children's problems. It would seem more useful to
study the variables that may account for these relationships because these
variables will shed greater light on what may be neuropsychologically
impaired in inattentive children (Barkley, in press).
Attention refers to the probability that a particular behavior will oc-
cur in the presence of a given stimulus. This is referred to as stimulus
control, but stimuli do not actually control the behavior in any physical
sense. They merely set the occasion for a change in response probabilities
(Skinner, 1953, 1969). Different components of attention refer to different
aspects of stimulus control. When children fail to respond to a task or to
sustain or inhibit their responding (inattention and impulsivity, respec-
tively), this simply indicates that the instructions given to the children
and or the task itself exert little stimulus control over the children's be-
havior. It does not tell us why this relationship has failed to occur or to be
maintained over time by some children when it has occurred or been
maintained successfully by others.
Whether or not stimulus control exists is dependent upon a number
of factors (e.g., the integrity and maturation of the nervous system, the
physical properties of the stimulus). However, one of the greatest determi-
nants is the learning history of the children. Behaviors that have been
reinforced or punished in the presence of a stimulus will subsequently
occur with greater or lesser probability, respectively, when that stimulus
is presented again. Such children are said to be developing selective at-
tention, in the former case, or impulse control, in the latter. The sustained
performance or inhibition of those behaviors in the continued presence of
that stimulus is chiefly a function of the types and schedules of conse-
quences occurring in that situation. Where strong or potent reinforcers or
punishers are provided in a frequent schedule for that behavior, its perfor-
mance will be sustained or inhibited for longer durations than when weak
or infrequent consequences occur. This suggests that some types of inat-
tention to certain tasks may arise from an inadequate reinforcement histo-

ry surrounding that task (i.e., the child simply has not had an adequate
opportunity to learn the task).
However, when children provided with an adequate learning history
fail to develop adequate attention to certain tasks, it is possible that dis-
turbances in certain neurological substrates are involved. Neurological
insults that disrupt previously established stimulus-response relations or
preclude their adequate development will appear as inattention, im-
pulsivity, or poor sustained attention. A behavioral analysis suggests that
some types of inattention may result from disruption of neural substrates
for certain sensory analyzers, whereas others arise from impairments in
the substrates mediating reinforcement and punishment and their con-
nections with the somatosensory or motor cortex or their association
areas. The type of deficit in attention may eventually prove to have some
localizing significance in neuropsychological assessment. Similarly, the
manner in which certain drugs affect different components of attention
may shed light on which functional parameters (e.g., detection, reinforce-
ment, schedules) and neurological substrates mediate these drug effects.
Unfortunately, previous neuropsychological research on attention
has seriously neglected the role of response consequences and their
schedules in the development and maintenance of attention. As a result,
the neurological substrates that mediate different aspects or components
of attention are not well understood. Present research on neurological
centers related to memory and reinforcement and drug effects on rein-
forcement centers, however, may offer promising developments in our
understanding of attention in child neuropsychology (Gray, 1972, 1982).
Implications for Assessment

This functional analysis of attention suggests that current methods of
assessment may be focusing exclusively upon the controlling effects of
various stimuli or task parameters (e.g., color, shape, complexity) on be-
havior without assessing the role of response consequences in attention to
tasks. Recent research (Douglas, 1983; Draeger, Prior, & Samson, 1986)
clearly shows that performance on such well-respected measures of atten-
tion as the continuous performance task are greatly affected by the moti-
vational factors in the setting. Attention deficits may be seen to appear or
disappear as response consequences are manipulated. Moreover, recent
research on stimulant drugs suggests that they affect attention by enhanc-
ing the reinforcement value of stimuli (Stein, 1985). Measures must there-
fore be developed that systematically evaluate the role of such conse-
quences and their schedules in the functional relations we call "attention"
if a more complete assessment of this construct is to occur.
Methods must also be developed that attempt to discriminate more
ATTENTION 153
accurately among the components of attention if the localizing signifi-

cance of the components is to be realized. For instance, poor sustained
attention may have more to do with problems with rapid extinction to the
reinforcing properties of tasks, whereas impulsi vi ty may reflect underreac-
tivity in the substrates mediating punishment. Behavioral reward and
punishment systems appear to be mediated by different neurological sub-
strates. If measures of sustained attention can be divorced from the effects
of impulsivity, they may yield greater neuropsychological utility than has
heretofore been the case with such commonly used measures as vigilance
tasks. Similarly, poor selective attention may actually be a function of the
parameters of the task stimuli themselves, such as sequential versus spa-
tial arrangement, symbolic versus concrete content, sensory modality,
color, etc. This implies that there may be many subtypes or patterns of
poor selective attention having differing neuropsychological significance.
ASSESSMENT OF ATTENTION
Many methods exist for assessing different aspects of attention in

children. Most are only in their early stages of development, lacking ade-
quate research on their psychometric properties or norms. This section
will discuss some of the more commonly used measures of attention since
a complete description of the numerous methods used in research is be-
yond the scope and intent of this chapter. Like measures of other con-
structs, measures of attention can be valuable for establishing the degree
of statistical deviance of children's attention from some norm and per-
haps can serve as a criterion for a particular diagnosis, as in attention
deficit disorders. They also can be used to select subjects for research on
attentional deficits, formulate treatment planning, assess changes as a
function of treatment or recovery after injury, and further study qualita-
tive and quantitative changes in attention throughout development. These
measures can be categorized into behavior rating scales, psychometric
tests and laboratory measures, and direct observational systems, all hav-
ing their own benefits and limitations.
Behavior Rating Scales

Behavior rating scales and checklists have become quite popular meth-
ods for assessing certain dimensions of child behavior in research and
clinical practice. Over 70% of clinical child psychologists and 60% of
school psychologists now employ rating scales as routine measures in their
assessment of children with attention deficit disorders (Rosenberg & Beck,
1986). Child neuropsychologists would be wise to do the same. Such
questionnaires have numerous advantages (see Barkley, 1987). They are
able to (1) evaluate a variety of different types of behavior in a very short

time (usually minutes), (2) collapse observations of children's attention
over lengthy time intervals (weeks to months) that would not be possible
with other tests or measures, (3) collect information about the child's
attentional characteristics in the natural environment without the need for
cumbersome and expensive direct observations, (4) assess the views of
important caregivers in the child's daily life concerning the nature of the
child's attention to various activities, (5) establish the statistical diviance of
the child's attention from peer-referenced normative data available for
most rating scales, (6) be repeated frequently over time to assess treatment
effects and recovery of function, and (7) provide quantification of
qualitative properties of children's attention through the use of Likert
scaling of such qualitative items.
Nevertheless, like other types of measures, rating scales have their
own inherent limitations. These include such things as bias or confound-
ing of ratings by characteristics of the rater, the loss of information on the
functional parameters (i.e., contexts, consequences) surrounding atten-
tion by rating only the behaviors involved, the ambiguity of the meaning
of items and their scaling reference points, and other disadvantages (see
Barkley, 1987). Such drawbacks often can be overcome by employing
several different types of measures (tests and observations) along with
rating scales. Those scales assessing some aspect of attention are shown in
Table 1 and are briefly described below.
Conners Rating Scales

The most widely used rating scales of attention in children have been
the Conners scales for parents and teachers (Conners, 1969, 1970; Goyette,
Conners, & Ulrich, 1978). The original Conners Parent Rating Scale (Con-
ners, 1970) consists of items of various behavior problems in children,
such as aggression, restlessness, fears, enuresis, learning problems, and
psychosomatic complaints. The Revised Parent Rating Scale (Goyette et
a1., 1978) was reduced in length. Each item on either scale is scored on a
4-point dimension of severity (not at all, just a little, pretty much, and
very much). Several of the items on each scale pertain to inattention,
distractibility, and impulsivity. These can be examined for their degree of
rated occurrence as well as combined to constitute a separate subscale of
attention. However, factor analyses of the parent scales have not revealed
a separate factor of inattention apart from hyperactivity and impulsivity
(Conners, 1970; Goyette et al., 1978). This suggests that parents view these
behaviors as falling along a homogeneous dimension of disruptive behav-
ior labeled impulsive-hyperactive rather than distinguishing inattention
as a separate problem. Normative data are available for the original (C. K.
ATTENTION 155
TABLE 1. Rating Scales for Assessing Attention in Children
Rating N of Ages
scale a Rater items Norms (years) Relevant scales
CPRS Parent 73 Yes 3-17 Hyperactive

CPRS-R Parent 48 Yes 3-17 Hyperactive
CTRS Teacher 39 Yes 5-17 Inattentive
CTRS-R Teacher 28 Yes 5-17 Inattentive
CBCL Parent 118 Yes 4-16 Hyperactive
CBCL-TRF Teacher 118 Yes 6-16 Inattentive
CAP Teacher 12 Yes 6-16 Inattentive
RBPC Parent and teacher 89 Yes 5-16 Attention
PBQ Teacher 30 Yes 3-5 Hyperactive-distractible
ACTeRS Teacher 24 Yes 5-11 Attention
"CPRS = Conners Parent Rating Scale. CPRS-R = Conners Parent Rating Scale-Revised. CTRS = Conners
Teacher Rating Scale. CTRS-R = Conners Teacher Rating Scale-Revised. CBCL = Child Behavior Check-
list. CBCL-TRF = Child Behavior Checklist-Techer Report Form. CAP = Child Assessment Profile. RBPC
= Revised Behavior Problem Profile. PBQ = Preschool Behavior Questionnaire. ACTeRS = Add-H
Comprehensive Teacher Rating Scale.
Conners, Department of Psychiatry, Children's Hospital National Medical

Center, 111 Michigan Avenue NW, Washington, D.C.) and revised ver-
sions (Goyette et a1., 1978).
The Conners Teacher Rating Scale (Conners, 1969) contains items
constituting a variety of behavioral and learning problems. Its more recent
shortened revision (Goyette et al., 1978), like the original scale, is scored
along a 4-point dimension of frequency/severity identical to the parent
scales (see Barkley, 1981). Three of the items appear to assess attentional
deficits. Unlike the parent scales, these items, along with five others, do
load on a single dimension that has been labeled Inattentive-Passive. Nor-
mative data are available for the original Teachers Rating Scale (Trites,
Blouin, & LaPrade, 1982) and the revised scale (Goyette et a1., 1978). Users
should keep in mind that the Inattentive-Passive factor also comprises
items that do not necessarily reflect attention deficits, and so this scale
would not constitute a pure or homogeneous rating of attentional prob-
lems.
Both the parent and teacher scales have satisfactory reliability and
validity, correlate well with other rating scales of similar behaviors, and
have been shown to discriminate children with attention deficit disorders
(ADD) from non-ADD clinic-referred children and normal children (see
Barkley, 1987). They also have been demonstrated repeatedly to be sen-
sitive to stimulant drug effects on ADD children (Cantwell & Carlson,
1978).
Child Behavior Checklist (CBCL)

Achenbach and Edelbrock (1983) have developed a global rating scale
of child psychopathology having both parent and teacher report forms.
The parent rating scale items are scored on a 3-point scale (0, 1, 2). They
yield information concerning the child's social competence as well as
psychopathology. One of these scales has been labeled Hyperactive and
comprises items assessing inattention, impulsivity, and overactivity. Like
the Conners parent scales, research on this scale shows that parent ratings
result in a single dimension comprising these three symptoms rather than
separate factors for each. Reliability and validity are well established
(Achenbach & Edelbrock, 1983; Barkley, 1987), and normative data are
available. The scale is useful for assessing ADD in children but is limited
for evaluating more pure deficits in the components of attention dis-
cussed above.
The teacher report form of this scale also yields information on differ-
ent dimensions of psychopathology. One of these scales has been labeled
Inattention and another Nervous-Overactive. The Inattention scale ap-
pears to be quite useful for assessing attentional deficits in children and,
along with the Nervous-Overactive scale, is extremely helpful in estab-
lishing diagnoses of ADD (with or without hyperactivity) in children.
Normative data are available (Edelbrock & Achenbach, 1984). Reliability
and validity of the scales appear satisfactory (see Barkley, 1987). As with
the Conners teacher scales, however, the Inattention factor on this scale
is partially confounded with items assessing behaviors other than in-
attention. Again, the scale seems more useful in assessing ADD in children
as compared with the more selective deficits in the components of atten-
tion.
Edelbrock Child Assessment Profile (CAP)

Edelbrock (personal communication) has extracted seven items from
the Inattention scale and five from the Overactive scale of the CBCL teach-
er form that have the highest loadings on these factors, respectively. These
have been cast into a separate rating scale shown in Figure 1. Normative
data for 1100 children ages 6 through 16 years by sex are provided in
Table 2. Scores of 11 or higher on the Inattention sub scale would place the
child above the 98th percentile relative to normal children. This rating
scale seems useful for assessing ADD and its subtypes (with and without
hyperactivity). But, unlike other rating scales, the inattention sub scale
also is helpful in evaluating a more pure construct of attention uncon-
founded by items concerning other behavioral problems, such as imma-
turity or affective symptoms.
ATTENTION 157
Child's Name: FOR OFFICE USE ONLY

Today's Date:
Filled Out By:
Below is a list of items that describes pupils. For each item that describes the pupil now or
within the past week, check whether the item is Not True, Somewhat or Sometimes True, or
Very or Often True. Please check all items as well as you can, even if some do not seem to
apply to this pupil.
Somewhat Very
or or
Not Sometimes Often
True True True
1. Fails to finish things he/she starts ............ . [I [I [I
2. Can't concentrate, can't pay attention for long .. [I [I [I
3. Can't sit still, restless, or hyperactive ......... . [I [I [I
4. Fidgets ..................................... . [I [I [I
5. Daydreams or gets lost in his/her thoughts .... . [I [I [I
6. Impulsive or acts without thinking ........... . [I [I [I
7. Difficulty following directions ............... . [I [I [I
8. Talks out of turn ............................ . [I [I [I
9. Messy work ................................ . [I [I [I
10. Inattentive, easily distracted ................. . [I [I [I
11. Talks too much ............................. . [I [I [I
12. Fails to carry out assigned tasks .............. . [I [I [I
Please feel free to write any comments about the pupil's work or behavior in the last week.
FIGURE 1. CAP rating scale. (cl Copyright 1986 Craig Edelbrock, Psychiatry,
UMMS, Worcester, MA 01605.
Revised Behavior Problem Checklist (RBPC)

This is a recent revision of one of the more widely used rating scales
in research on childhood behavioral disorders (the Behavior Problem
Checklist). The RBPC uses a 3-point scale (0, 1, and 2) for each item, and
scoring is based on item placement within empirically derived factors.
The RPBC now has a subscale for assessing attention problems/immatur-
ity separate from a motor excess scale. Normative data are available for
grades K through 12, using teachers as raters, and for 248 children, ages 5
to 16 years, using mothers' ratings (Quay & Peterson, 1983). The scale
seems to have satisfactory reliability and validity (Barkley, 1987). Like
those scales reviewed above, this one seems most helpful for evaluating
the major symptoms of ADD in children as opposed to assessing more
specific deficits in various components of attention.
TABLE 2. Normative Cutoff Points for the

Inattention, Overactivity, and Total Scores
for the Edelbrock CAP Scale
Total Boys Girls

Cutoff points (1100)0 (550) (550)
Inattention b
Median 1 2 0
69th percentile 3 4 2
93rd percentile 8 9 7
Overactivity
Median 0 1 0
Total score
Median 2 4 1
oNumbers in parentheses are sample sizes.
blnattention score is the sum of items 1, 2, 5, 7, 9, 10, and 12.
The Overactivity score is the sum of items 3, 4, 6, 8, and 11.
Table entries are raw scores that fall at or below the desig-
nated percentile range. The 93rd percentile is the recom-
mended upper limit of the normal range. Scores exceeding
this cutoff are in the clinical range. All scores are based on
teacher reports.
Preschool Behavior Questionnaire (PBQ)

The PBQ was designed for teachers' use with preschool-age children
as a screening measure for those who might be at risk for later adjustment
problems and psychopathology in school. It consists of 30 items and is
completed by teachers or their aides. A 3-point scale (0, 1, 2) is used to
rate each item, many of which deal with aggressiveness, anxiety, conduct
problems, hyperactivity, and distractibility, As with many other scales,
factor analysis of the items has yielded a hyperactivity-distractibility fac-
tor that can be scored separately. Normative data are based on a sample of
496 children (Behar & Stringfield, 1974). Reliability and validity appear to
be satisfactory (Barkley, 1987). Like the Conners scales, the PBQ has little
utility in delineating subtypes of attentional deficits since the hyperac-
ATTENTION 159
tivity scale is confounded with items pertaining to restlessness and imma-

ture conduct.
ADD-H Comprehensive Teacher Rating Scale (ACTeRS)

This scale was developed to provide more refined ratings of the major
symptoms of ADD in children: attention, hyperactivity, social problems,
and oppositional behavior (Ullman, Sleator, & Sprague, 1984). Items were
chosen that loaded on only one factor to avoid contamination of the scales
with items also correlating with the other behavior scales. This scale
relies on teachers reports and uses a 5-point scale for rating each item.
Normative data are provided with the manual for 1347 children. Only
recently developed, the scale has some preliminary findings on its various
reliabilities that appear to be satisfactory. Further studies of its validity
and utility are awaited. Like the Edelbrock CAP Scale, this scale is useful
not only for assessing subtypes of ADD in elementary-age children but
also for evaluating a more pure construct of attention using the Attention
Subscale.
Psychometric Tests and Laboratory Measures

A plethora of laboratory measures and tests have been devised to
evaluate various aspects of attention, seemingly limited only by the
creativity of the investigator and the technology available. In fact, because
practically every test or laboratory task requires the subject's attention
and cooperation before conclusions can be drawn about performance,
virtually every test serves in some fashion as a measure of attention.
However, as already noted, to the extent that the measure requires other
neuropsychological functions to perform it satisfactorily, the more cau-
tious must be the conclusions about attentional deficits from poor perfor-
mance on it. Not surprisingly, then, tests of attention usually involve
quite simple instructions, stimuli, and motor responses so as to minimize
confounding by other cognitive deficits. A few of the more widely used or
promising laboratory tasks and tests are reviewed below. The reader in-
terested in a review of other measures should consult the excellent re-
views by Douglas (1983; Douglas & Peters, 1979) dealing with attentional
deficits in learning-disabled and ADD children.
Compared with rating scales, psychometric tests and lab measures
have the ability to conduct a finer analysis of the components of attention
and the effects of manipulating various task parameters upon them.
Where adequate normative data are available, these measures can assist in
providing an objective index of the children's deviance from normal at-
ten tiona I processes and are less affected by potential rater bias that can
easily influence behavior rating scales. Many of the measures reviewed

below are brief, can be easily administered, and have a number of research
articles dealing with their reliability and validity. Most are hampered,
however, by some confounding with other cognitive abilities necessary to
conduct the task. This can be partially overcome by using a battery of tests
that sample a variety of the major neuropsychological functions with
some redundancy across tests in the functions that they evaluate. Perfor-
mance on these other tests may then provide clues as to the basis for the
poor performance on the attentional measure(s).
Another limitation of many of these tests is their low to moderate
relationship with other measures of attention taken in the children's natu-
ral environment, such as direct observations in the classroom or parent
ratings. There is a tendency among newly trained neuropsychologists to
attribute greater credibility to the findings from a test than from those
derived from parent or teacher reports or direct observations, perhaps
because the former seem somehow to be more objective. It is worth recall-
ing that tests often utilize very brief samples of behavior taken in settings,
such as offices, that are not especially representative of the children's
more natural habitats. In such cases, where test results conflict with other
sources of information about attention, further investigation of the dis-
crepancy is clearly required. In some of these instances, it may be the
"objective" test results that should be disregarded. As noted earlier, the
use of multiple measures of attention derived from multiple sources can
overcome many of these limitations.
Reaction Time (RT) Tasks

Probably one of the oldest measures of assessing vigilance or atten-
tion span is the reaction time task. These tasks can be configured into a
variety of formats, some of which simply involve pressing or releasing a
button as soon as a signal appears. Others utilize multiple signals and
multiple response keys corresponding with the signals. Some may in-
clude a warning stimulus to the subject with a fixed or variable time
interval prior to the onset of the signal. In all of these, the primary depen-
dent measure is typically the speed with which the subject presses/re-
leases the button after the signal appears. Often measured in hundredths
of seconds, the subject's RT may be elicited over multiple trials, with the
average reaction time across trials serving as the score. Other investigators
also have calculated the standard deviation of scores around this mean to
evaluate variability of performance. Monitoring the number of times the
subject presses the button before the signal or at the incorrect signal also
can serve as a measure of poor inhibition or impulse control (often re-
ferred to as errors of commission). These tasks require the maintenance of
ATTENTION 161
vigilance to simple stimuli or the inhibition of responses to competing

stimuli over prolonged periods of time.
Studies ,often find children with ADD to have greater mean reaction
times, greater variability of reaction times, and more errors of commission
than matched normal children (Douglas, 1983; Douglas & Peters, 1979).
Other studies have found impaired reaction times in long-term survivors
of childhood leukemia (Brouwers et a1., 1984) and as sequelae of closed
head injury in children (Levin, Benton, & Grossman, 1982). In studies
using a warning stimulus prior to signal onset, ADD, brain-damaged, and
mentally retarded children have all been noted to profit less from this
warning than normal children, resulting in slower reaction times within
these groups (Czudner & Rourke, 1972; Gosling & Jenness, 1974). Many
studies have shown both stimulant drugs and various reinforcement con-
tingencies to improve, if not normalize, these reaction time deficits in
children (Barkley, 1977a, 1977b; Douglas, 1983).
To the casual observer, reaction time tasks may seem as simple and
pure a measure of vigilance as one is likely to find. Although this may be
true, the interpretation of reaction time scores as reflecting attentional
deficits must be done with some caution. It is possible that deficits in
sensory perception, motor programming and execution, language com-
prehension, and rule-governed behavior (e.g., following experimental in-
structions) may impair reaction time performance. However, where such
deficits do not exist, then problems with vigilance may be concluded from
poor performances on these tasks.
Continuous Performance Tasks (CPT)

The CPT is similar to an RT task except that many different stimuli
may be used and the subject is required to respond to only one type while
inhibiting responses to others (Rosvold, Mirsky, Sarason, Bransome, &
Beck, 1956). For instance, flashing letters of the alphabet on a screen and
instructing the child to respond only to the letter X after it follows an A
may serve as a simple CPT. Most of the tasks available last from 8 to 15
minutes.
Measures often derived from these tasks are total correct responses,
errors of commission (false hits), and errors of omission (missed hits)
(Douglas, 1983). Commission errors are believed to reflect impulsivity
whereas omission errors reflect inattention. Reaction time mayor may not
be taken in these tasks. Others (Buchsbaum & Sostek, 1980) have applied
signal detection theory to analysis of the results of the CPT, calculating
two additional dependent measures. Sensitivity, or d', refers to the atten-
tiveness component of vigilance and is calculated as Z (hit rate) - Z (false
alarm rate). Hit rate is total hits/total signals, whereas false alarm rate is
false hits/(total of target signals - total of all signals). Response bias, or

Beta, refers to a subject's tendency to respond when a signal has occurred
instead of responding when one has not occurred. It is believed to reflect
impulsivity and motivational factors, and it is calculated as Y (hit rate)/Y
(false alarm rate).
Some investigators (Buchsbaum & Sostek, 1980) have developed a
CPT in which the interval between stimulus presentations varies as a
function of the prior success or failure in detecting the last signal present-
ed. If the subject gets a correct hit, the speed of presentation of the stimuli
increases, whereas an incorrect hit results in a decrease in the rate of
presentation. This permits calculation of an additional dependent mea-
sure known as the interstimulus interval (lSI). lSI is the average interval
between stimuli over the trials of the task.
Besides RT tasks, CPTs have been demonstrated to be some of the
most discriminating measures between ADD and normal children and to
be quite sensitive to medication effects (Barkley, 1977a; Douglas, 1983;
Swanson & Kinsbourne, 1979). They also correlate to modest but signifi-
cant degrees with other tests of attention and impulsivity, teacher ratings
of hyperactivity in the classroom, and clinic playroom observations of on-
task behavior (Klee & Garfinkel, 1983; Prinz, Tarnowski, & May, 1984).
With developments in microcomputer technology, it has become
quite feasible to create software capable of running CPTs on personal
computers. V. Berman (personal communication, July 18, 1985; Univer-
sity of Nebraska-Lincoln, Department of Special Education, 301 Barkley
Memorial Center, Lincoln, NE 68583-0783) has developed a CPT for the
Apple lIe that utilizes geometric shapes rather than letters as the stimuli.
The target stimulus is an airplane. As in the Buchsbaum and Sostek (1980)
CPT, the interval between stimulus presentations varies as a function of
prior success at detecting signals. This task also incorporates distracting
stimuli that flash to the left or right of the target stimulus display area,
permitting a measure of vigilance during distracting and nondistracting
conditions. Within the distraction trials, scores can be tabulated for dis-
tractions to the left and right visual fields separately. This task would
seem to have some promise in assessing hemiinattention in addition to
vigilance. Normative data for the lSI measure are available based only on a
sample of 230 children in grades K through 9. Others (Klee & Garfinkel,
1983; Prinz et 01., 1984) have also developed computerized CPT pro-
grams, but these do not yet have normative data available.
Gordon et 01. (1983) has provided the first commercially marketed
CPT for use in the assessment and diagnosis of ADD children. The device
is a portable CPT housed in a metal child-proof box containing a micro-
processor. Numbers are flashed on a display screen, and the subject is to
respond by pushing a button only when the number 1 (ages 3 to 5 years) or
the numbers 1 then 9 (ages 6 to 16 years) appear. The device computes the
ATTENTION 163
measures of number correct, number of commissions, and number of

omissions. Normative data are available for more than 1400 children be-
tween 3 and 16 years of age. A recent modification of the task now permits
it to flash distracting numbers to the left and right of the target display
area for assessment of vigilance with and without distractors. The advan-
tages of Gordon's CPT task, besides its construction, are the availability of
normative data on a large sample of children over a wide age range, its
satisfactory discrimination of ADD from non-ADD children (Gordon et aI.,
1983), its sensitivity to stimulant drug effects (Barkley, Fischer, Newby, &
Breen, 1988). and its significant correlation with teacher ratings of inat-
tentive and hyperactive behavior (Gordon, 1983).
In general, CPT tasks appear to be a convenient way of assessing
vigilance and impulsivity in children. The availability of software pro-
grams for personal computers and normative data on some versions per-
mits the use of these tasks for clinical and research purposes.
Cancellation Tasks
Several paper-and-pencil versions of CPT tasks have been used as
methods of assessing attention, primarily vigilance and impulsivity.
These tasks typically involve having the subject scan a series of symbols
(i.e., letters, numbers, shapes) in rows on paper. The subject is typically
required to draw a line through or under the target stimulus using a
pencil. One such task is the Children's Checking Task (CCT), developed
by Margolis (1971). The task consists of a series of 15 numerals per row
printed in 16 rows on a page. There are seven pages to the task. A tape
recorder reads off the numbers in each row, and the child is required to
draw a line through each number as it is read. Discrepancies between the
tape and the printed page are to be circled. There are seven discrepancies
between the tape and the printed pages. The task takes about 30 minutes.
Scores are derived for errors of omission (missed discrepancies) and er-
rors of commission (numbers circled when no discrepancy exists). The
task discriminates educationally handicapped from normally achieving
students (Keough & Margolis, 1976) and ADD from reading-disabled chil-
dren (Brown & Wynne, 1982). It also appears to correlate to a modest but
significant degree with other measures of attention (Keough & Margolis,
1976).
Another cancellation task is the Underlining Test developed by
Doehring and used extensively in neuropsychological studies by Rourke
and his colleagues (Rourke & Gates, 1980; Rourke & Orr, 1977). The task
consists of 14 subtests. Each subtest varies as to whether it involves num-
bers, geometric shapes, letters or letter combinations, or words. In each
subscale, the child is required to scan rows of these symbols on a page and
to underline a particular one that matches a sample provided at the top of
the page. The child's speed and accuracy in performing the task are re-
corded. The subtests range in degree of difficulty, with subtests 1 and 13
being identical to assess practice effects, and another subtest (14) involving
stimuli (rectangles) that are all identical, and all of which the child must
underline. This is designed to control for motor speed. Normative data are
available (Rourke & Gates, 1980), and impairments on the tests have some
relationship to reading and spelling deficits in children (Rourke & Orr,
1977). The test would seem to require adequate visual perception, spatial
discrimination, number and letter facility, and adequate motor speed and
coordination. Where these are intact but performance on the task is im-
paired, deficits in selective attention may be hypothesized.
Children's Embedded Figures Test (CEFT)

A commonly used test of selective attention and visual discrimina-
tion has been the CEFT (Witkin, Oltman, Raskin, & Karp, 1971). Some
have interpreted the test as a measure of "coming to attention" (Keough &
Margolis, 1976), field dependence-independence and global versus ana-
lytic perception (Schain, Ward, & Guthrie, 1977). The test involves 27
pictures of figures within which a simple figure (tent or house) is embed-
ded. The scores are the mean time to respond and the number of correct
responses. Normative data are available for ages 5 to 12 years, and reliabil-
ity of the test is quite satisfactory (Schain et 01., 1977). The test has been
used to assess selective attention in studies of hyperactive, hypoxic, edu-
cationally handicapped, and epileptic children (Brown & Wynne, 1982;
Keough & Margolis, 1976; Schain et 01., 1977). It appears to correlate
significantly and moderately with the Children's Checking Test, de-
scribed above, and Kagan's Matching Familiar Figures Test (see below).
Hence, it would appear that the task assesses impulsivity and vigilance as
well as selective attention.
Mazes
Maze performance has frequently been used in child neuropsycho-
logical research as a measure of motor planning, speed, and execution, as
well as impulsivity and sustained attention (Reitan & Davison, 1974; Rou-
rke, Bakker, Fisk, & Strang, 1983). Poor maze performances have been
noted in ADD children (Milich & Kramer, 1984), children surviving
closed head injury (Klonoff, Low, & Clark, 1977), and epileptic children
undergoing treatments with anticonvulsants (Schain et 01., 1977). Such
tests also have proven sensitive to stimulant drug effects in ADD children
(Barkley, 1977a, 1977b).
ATTENTION 165
Probably the most commonly used maze test is the Porteus Mazes
(1965). Although originally intended as a measure of intelligence, the test
has been most typically used to assess impulsivity, sustained attention,
and the motor parameters noted above. In this paper-and-pencil task, the
printed mazes progress across a range of increasingly difficult configura-
tions. A year level, test quotient, and Q score are obtained, the latter
believed to assess impulsivity. Another paper-and-pencil maze test is an
optional subtest of the Wechsler Intelligence Scale for Children-Revised
(1974), for which normative data are available. A maze test often used in
neuropsychological evaluations is part of the Wisconsin version of the
Halstead-Reitan Neuropsychological Test Battery (Matthews & Klove,
1965). It involves the child's using a metal stylus to traverse a raised metal
maze. The child's speed of performance, number of contacts with the
sides of the maze, and time in contact with the sides is recorded.
Matching Familiar Figures Test (MFFT)

The most widely employed measure of impulsivity is undoubtedly
Kagan's MFFT (Kagan, Rosman, Day, Albert, & Phillips, 1964). The test
involves presenting the child with a sample figure (i.e., a house) and
simultaneously displaying six very similar figures, one of which is identi-
cal to the sample figure. The child is to point to the figure that matches the
sample. The time to the first response and the number of errors are re-
corded. Scores are then derived for mean time to response and total errors
across the 12 trials. Normative data are available for ages 5 through 12
years for a sample of over 2800 children (Salkind & Nelson, 1980). Impair-
ments in test performance have been reported for ADD children (Douglas
& Peters, 1979; Milich & Kramer, 1984), impulsive educationally handi-
capped children (Keough & Margolis, 1976), behavior-disordered children
(Brown & Quay, 1977), and epileptic children receiving anticonvulsants
(Schain et al., 1977). The test scores correlate to a low, but significant,
degree with continuous performance tests, Arithmetic and Coding sub-
tests from the WISC-R, the Children's Checking Test, and the Children's
Embedded Figures Test (Keough & Margolis, 1976; Klee & Garfinkel,
1983). Numerous studies have shown the test to be sensitive to stimulant
drug effects in ADD children (Barkley, 1977b; Cantwell & Carlson, 1978).
Direct Reinforcement of Latency (DRL) Tasks

A relatively recent attempt has been made by Gordon (1979; Gordon
et a1., 1983) to utilize a DRL paradigm to assess impulse control in chil-
dren. The task requires a child to sit before a metal box containing a
microprocessor that administers and scores the test. The child is required
to press a button, wait a while, and press it again. If the child has waited
long enough, a point is earned. Cumulative points are recorded on a
display screen. The child is not informed of the delay interval (6 seconds)
but is reinforced with points on each trial that exceeds the standard inter-
val. The test is brief (8 minutes), and normative data are available for more
than 1400 children. The test discriminates ADD from non-ADD clinic-
referred children, and it correlates moderately and significantly with the
MFFT, the Conners parent and teacher rating scales, and the Child Behav-
ior Checklist Teacher Report Form (Gordon e1 01., 1983). Although prom-
ising, the task has not proven sensitive to stimulant drug effects with ADD
children (Barkley e1 01., 1988).
Span of Apprehension Test

The concept of span of apprehension has been used to refer to the
amount of information a child is able to attend to simultaneously. The
child is shown matrices on a display screen. Some of the 16 cells contain a
T, an F, or noise letters. The number of noise or nonsignalletters is varied
across trials (2,4, or 8 noise letters), as is the location of one of the signal
or target letters (T, F). Matrices are presented via a tachistoscope for 100-
msec exposures while the child focuses upon a fixation point. The
number of letters scanned by the child is the span of apprehension (A)
and is calculated as A = D (2 Pc - 1), where D is the number of letters in
the display, and Pc the percent correct of recognition of target stimuli.
Studies have shown the task to discriminate ADD from normal children
(Denton & McIntyre, 1978) and to correlate with direct observations of
attention during a clinic playroom task (Prihz e1 01., 1984). However,
normative data for the task are lacking, and the necessity of using a ta-
chistoscope for presentation makes the task unwieldy for young children
and inconvenient for clinical practice.
Goldman-Fristoe-Woodcock (GFW) Selective Attention Test

One of the few tests of auditory, as opposed to visual, selective atten-
tion is the GFW (Woodcock, 1976). This test has been used occasionally in
studies of learning-disabled, ADD, and normal children (Foch & Plomin,
1980; Plomin & Foch, 1981). The test requires the child to wear earphones
and listen to words presented on a tape recording. The child is to point to
printed pictures of the words provided on the tape. Part of the test in-
volves no background noise on the tape during word presentation, where-
as other parts have increasing background noise (white noise and cafeteria
sounds). So constructed, the test appears to be more a measure of distrac-
tibility to extraneous sounds while reading than one of selective attention.
ATTENTION 167
Freedom From Distractibility (FFD) Factor of the WISC-R

Factor analysis of the WISC-R by Kaufman (1975) demonstrated a
three-factor solution now widely accepted as the predominant cognitive
constructs assessed by this intellectual test. One of these three factors
comprised the Arithmetic, Digit Span, and Coding subtests and was
named Freedom From Distractibility (FFD). It was so named in part out of
tradition because a comparable factor emerged in analyses of other
Wechsler batteries and "because of research with hyperactive children
showing that drug therapy leads to decreased distractibility and improved
memory and arithmetic skills in these youngsters" (Kaufman, 1980, p.
179).
Since its introduction over a decade ago, clinicians and researchers
alike have interpreted deficits on this factor as reflecting distractibility or
poor attention. The subtests constituting the factor do show low, but
significant, correlations with other measures of attention, such as vig-
ilance and impulsivity (Klee & Garfinkel, 1983). Evidence is conflicting,
however, as to whether deficits on these subtests discriminate ADD from
non-ADD or reading-disabled children (Brown & Wynne, 1982; Milich &
Loney, 1974). Deficits in these subtests have been found in long-term
survivors of acute lymphoblastic leukemia (Goff, Anderson, & Cooper,
1980). However, where deficits in FFD subtests are found, it is not clear
that they necessarily reflect deficiencies in attention (e.g., distractibility).
These subtests also appear to require adequate short-term memory, nu-
merical facility (Arithmetic and Digit Span), arithmetic calculation, and
visuospatial constructional skills as well as visual-motor speed (Coding
subtest). As a result, a number of investigators (Kaufman, 1980; Ownby &
Matthews, 1985; Stewart & Moely, 1983) have urged caution in interpret-
ing these subtests as measures of distractibility, believing its label to be an
oversimplification and misleading (Ownby & Matthews, 1985). Complex
cognitive processes appear to be involved in these tests, precluding any
straightforward interpretation of performance deficits in them as reflect-
ing inattention.
Direct Observational Measures

Although often not included in standard neuropsychological test bat-
teries with children, systematic direct observational recordings of behav-
ior during various situations readily reveal deficits in attention and
impulse control. They have most often been used in studying ADD chil-
dren, especially during task performances (Abikoff, Gittelman-Klein, &
Klein, 1977; Ross & Ross, 1982). Direct observations of behavior systemat-
ically recorded alleviate many of the limitations of other assessment
methods, such as unrepresentativeness or rater bias. Such observational

systems are exquisitely sensitive to stimulant drug treatments and behavior
modification interventions (Barkley, 1977b; Barkley et a1., 1988; Mash &
Dalby, 1979). They are more ecologically representative of the children's
actual attentional deficits when taken in natural settings than are laborato-
ry measures or psychometric tests. Moreover, they permit close analysis of
the environmental variables of which the inattentiveness may be a func-
tion. For instance, deficits in sustained attention in ADD children may not
always be apparent in free-play settings, but they are commonly noted
when the children are assigned tasks to perform by others (Routh &
Schroeder, 1976). These attentional deficits may be readily improved by
the addition of frequent reinforcement or punishment (Pfiffner, Rosen, &
O'Leary, 1985), or by altering the salience of the stimuli to which the
children must attend (Ross & Ross, 1982).
Probably the most commonly used behavioral observation measure of
attention is the direct recording of "on-task" or "off-task" behavior, either
in the child's regular classroom (Abikoff et a1., 1977; Pfiffner et a1., 1985)
or in clinic analogue situations using one-way observation mirrors (Bark-
ley, 1977a; Barkley et a1., 1988). Coding may involve simply using a
stopwatch to time the duration of the child's visual focusing on the task
(Barkley, 1977a; Prinz et a1., 1984). or recording whether or not on-task or
off-task behavior was noted during a series of brief time intervals, say 15
seconds (Barkley et a1., 1988). In the latter case, the dependent measure is
expressed as the percentage of the total observation intervals in which the
child demonstrated these behaviors.
Such observational systems correlate well with observer ratings when
taken in the same classroom (Kazdin, Esvelt-Dawson, & Loar, 1983), and
with laboratory measures of vigilance, impulsivity, and span of apprehen-
sion when taken in clinic playroom settings (Prinz et a1., 1984). Clinic
playroom observations during task performance also have shown surpris-
ingly high stability over a 2-year interval (Milich, 1984) and appear to be
as sensitive to dose effects of stimulant medication as those taken in
classroom settings (Barkley et a1., 1988; Rapport, Stoner, DuPaul, Bir-
mingham, & Tucker, 1985). In these analogue situations, measures also are
taken of the children's work productivity as a behavioral by-product of
the children's on-task activities (Prinz et a1., 1984).
Other investigators have recorded the number of toy changes or the
mean duration of play with each toy during free-play situations in clinic
playrooms as reflections of sustained attention (Barkley, 1977a; Routh &
Schroeder, 1976). These tend to have poor test-retest reliability (Plomin &
Foch, 1981). In some cases, the number of times a child looked at a
particular stimulus, such as a television monitor, served as the measure of
attention (Prinz et a1., 1984).
Direct observational measures seem to be convenient to employ, es-
ATTENTION 169
pecially if clinic analogue settings are used. They also are particularly
sensitive to treatment manipulations. However, their main drawback at
present remains the lack of normative data for use in establishing de-
viance of attention during initial clinic evaluations. This is less of an issue
in child neuropsychological research, but it is a major problem for clinical
practice. Nevertheless, such recordings may still prove useful in measur-
ing within-subject changes in behavior due to stimulant medication or
other interventions in clinical practice.
DISCUSSION
This brief review of assessment techniques suggests that many mea-

sures are flawed in their evaluation of attention, often because of confound-
ing of attentional components with each other or with other neuropsycho-
logical constructs. Most rating scales do not evaluate pure components of
attention and are hampered by possible sources of rater bias. Although
convenient, they lack the precision desired in a finer-grain analysis of
attentional components within neuropsychological research. Still, as
crude initial screening measures to detect attentional deficits in large
populations of children, they would seem quite valuable. They also offer
the advantage of identifying those natural settings in which the attentional
deficits are creating the greatest difficulties in adjustment and perfor-
mance. This suggests that the inclusion of such rating scales in research
and practice would be prudent but that more objective. systematic, and
specific laboratory measures of attentional components also must be
included.
Among the psychometric and laboratory measures, reaction time and
vigilance tasks appear to have the greatest sensitivity in detecting atten-
tion deficits, particularly in sustained attention and impulsivity. Mea-
sures of selective attention and distractibility are not as well developed or
investigated, and those of span of apprehension and attentional search
have received minimal research in the neuropsychological literature. Just
as measures of memory and its components must be more refined and
standardized (see Chapter 7, this volume), those of attention must be more
narrow or specialized in the type of attention assessed. Differences in the
components of attention within each sensory modality, between spatial
and sequentially presented material, and between symbolic and concrete
contents probably exist yet are unappreciated by the crude methods now
employed. The application of signal detection theory to measures of vig-
ilance has proven helpful in this regard, separating inattention from im-
pulsivity in the analysis of performance scores. Similar approaches to
other components of attention might prove fruitful.
Of even greater promise would appear to be a functional analysis of
attention as a term describing numerous contingent relations between

stimuli and responses that are a function of not only stimulus parameters
and task instructions but also the consequences for responding and their
schedules. This implies that a full appreciation of attention and its com-
ponents will come only when the manner in which variations in response
consequences affect behavior are systematically assessed. Although this is
evaluated to a limited degree on tasks such as the Wisconsin Card Sort
Test and the Halstead-Reitan Categories task, the way in which response
consequences affect subsequent responding is not controlled in a suffi-
ciently systematic or standardized way across subjects to assess this as-
pect of attention. More promising would seem to be the use of very simple
behavioral paradigms, such as vigilance tasks, in which response conse-
quences are systematically manipulated to evaluate different rates of be-
havioral extinction (poor sustained attention) or disinhibition (im-
pulsivity) under differing schedules of consequences.
The acceptance of attention as multidimensional implies that the
separate components of attention could be used to construct profiles of
performance for each child. A taxonomy of more homogeneous subtypes
of children having disorders of attention could then be created from these
typologies. Present research suggests that children with deficits in selec-
tive attention may constitute a different subtype from those with poor
sustained attention, who in turn may differ from those with deficits in
impulse control. Each may stem from different etiologies, be mediated by
different neurological subtrates, have differing developmental courses,
and respond differently to various treatments. As research on reading
disorders has unfolded to reveal more homogeneous subtypes (see Chap-
ter 11, this volume), each having different developmental courses and
treatment implications, so too will this likely be seen in the field of atten-
tion if as much effort and research interest can be brought to bear on the
problem.
SUMMARY
Attention is a multidimensional construct that refers to a variety of

functional relationships between environmental stimuli or tasks and chil-
dren's behaviors. Deficits in attention are quite prevalent within normal,
psychiatric, and neurological populations of children, yet they are often
ignored in the systematic assessment of such children. Even when they
are evaluated, the methods for doing so are often highly impressionistic or
of questionable validity. More appropriate measures are available in the
recent research literature but have yet to be adopted to any widespread
degree in clinical practice. Such adoption is strongly encouraged.
The major components of attention appear to be arousal and alert-
ATTENTION 171
ness, selective or focused attention, distractibility, sustained attention,

impulsivity, span of apprehension, and search. Developmental research
indicates increasing economy, efficiency, and effectiveness in these com-
ponents with maturation well into the adolescent years.
A functional analysis of these components suggests that they are un-
der different aspects of stimulus control. Stimulus control occurs as a
function of the type of stimulus or task, its inherent reinforcement proper-
ties, the nature of rules given with the task, the type and schedule of
consequences for responding to the task, and the integrity of the develop-
ing nervous system. Disturbances in any or all of these may disrupt "at-
tention." As such, deficits in the components of attention may result from
improper task construction, inadequate learning histories, or neurological
impairments in the sensory association cortex, in the reinforcement or
punishment centers, in the feedback loops between these centers and the
sensory-motor association areas, or in the prefrontal zones (executive
cortex).
Numerous methods of assessing attention were reviewed. These are
categorized into behavior rating scales, laboratory measures, psycho-
metric tests, and direct observational methods. Although each type offers
unique strengths and limitations, the former can be capitalized upon and
the latter overcome by the inclusion of at least one measure of each type in
the clinical assessment of children. It may eventually prove important to
assess each component of attention within specific sensory modalities,
once adequate measures of such are developed and studied for their sig-
nificance. What is now required is a concerted effort of increased re-
search, such as that seen with the learning disabilities. Such an effort
should incorporate the views of behavioral analysis to advance our under-
standing of this perplexing construct and to better serve the large segment
of the childhood population afflicted with deficits in the components of
attention.
REFERENCES
Abikoff, H., Gittelman-Klein, R., & Klein, D. (1977). Validation of a classroom observation
code for hyperactive children. Journal of Consulting and Clinical Psychology, 45, 772-
783.
Achenbach, T. M., & Edelbrock, C. (1983). Manual for the Child Behavior Checklist and
Revised Child Behavior Profile. Burlington, VT: Thomas Achenbach.
American Psychiatric Association. (1980). Diagnostic and statistical manual of mental dis-
orders (3rd ed.). Washington, D.C.
Barkley, R. (1977a). A review of stimulant drug research with hyperactive children. Journal
of Child Psychology and Psychiatry, 18, 137-165.
Barkley, R. (1977b). The effects of methylphenidate on various measures of activity level and
attention in hyperkinetic children. Journal of Abnormal Child Psychology, 5, 351-369.
Barkley, R. A. (1981). Hyperactive children: A handbook for diagnosis and treatment. New
Barkley, R. A. (1985). Family interaction patterns in hyperactive children: Precursors to
aggressive behavior? In M. Wolraich & D. Routh (Eds.), Advances in behavioral pedi-
atrics (pp. 117-150). Greenwich, CT: JAI Press.
Barkley, R. A. (1987). A review of child behavior checklists and rating scales for research in
child psychopathology. In M. Rutter, A. H. Tuma, & 1. S. Lann (Eds.), Assessment and
diagnosis in child psychopathology. New York: Guilford Press.
Barkley, R. A. (1988a). Attention deficit disorders. In E. Mash & 1. Terdal (Eds.), Behavioral
assessment of childhood disorders (2nd ed.). New York: Guilford Press.
Barkley, R. A. (1988b). Tic disorders and Tourette's syndrome. In E. Mash & 1. Terdal (Eds.),
Behavioral assessment of childhood disorders (2nd ed.). New York: Guilford Press.
Barkley, R. A. (in press). The problem of stimulus control and rule-governed behavior in
attention deficit disorder with hyperactivity. In 1. Bloomingdale & J. Swanson (Eds.),
Attention deficit disorders (vol. 3). New York: Pergamon.
Barkley, R., Fischer, M., Newby, R. F., & Breen, M. J. (1988). Development of a multi-method
clinical protocol for assessing stimulant drug responding in ADD children. Journal of
Clinical Child Psychology, 17,14-24.
Barkley, R., Karlsson, J., & Pollard, S. (1985). Effects of age on the mother-child interactions
of hyperactive boys. Journal of Abnormal Child Psychology, 13,631-637.
Behar, L., & Stringfield, S. (1974). A behavior rating scale for the preschool child. Develop-
mental Psychology, 10, 601-610.
Boll, T. J. (1981). The Halstead-Reitan Neuropsychological Test Battery. In S. Filskov & T.
Boll (Eds.), Handbook of clinical neuropsychology (Vol. 1, pp. 577-607). New York:
Wiley.
Boll, T. J. (1983). Minor head injury in children-Out of sight but not out of mind. Journal of
Bremer, D. A., & Stern, J. A. (1976). Attention and distractibility during reading in hyperac-
tive boys. Journal of Abnormal Child Psychology, 4, 381-387.
Brouwers, P., Riccardi, R., Poplack, D., & Fedio, P. (1984). Attentional deficits in long term
survivors of childhood acute lymphoblastic leukemia (ALL). Journal of Clinical Neuro-
psychology, 6, 325-336.
Brown, R. T., & Quay, 1. C. (1977). Reflection-impulsivity in normal and behavior-disor-
dered children. Journal of Abnormal Child Psychology,S, 457-462.
Brown, R. T., & Wynne, M. E. (1982). Correlates of teacher ratings, sustained attention, and
impulsivity in hyperactive and normal boys. Journal of Clinical Child Psychology, 11,
262-267.
Buchsbaum, M. S. & Sostek, A. J. (1980). An adaptive continuous performance test: Vigilance
characteristics and reliability for 400 male students. Perceptual and Motor Skills, 51,
707-713.
Cantwell, D. P., & Carlson, G. (1978). Stimulants. In J. Werry (Ed.), Pediatric psychophar-
macology (pp. 171-207). New York: Brunner/Mazel.
Condry, S., McMahon, M., & Levy, A. (1979). A developmental investigation of selective
attention to graphic, phonetic, and semantic information in words. Perception and
Psychophysics, 25, 88-94.
Conners, C. K. (1969). A teacher rating scale for use in drug studies with children. American
Journal of Psychiatry, 126, 884.
Conners, C. K. (1970). Symptom patterns in hyperkinetic, neurotic, and normal children.
Child Development, 41, 667-682.
Czudner, G., & Rourke, B. P. (1972). Age differences in visual reaction time of "brain-
damaged" and normal children under regular and irregular preparatory interval condi-
tions. Journal of Experimental Child Psychology, 13, 516-526.
ATTENTION 173
Denton, C. 1., & Mcintyre, C. W. (1978). Span of apprehension in hyperactive boys. Journal
of Abnormal Child Psychology, 6, 19-24.
Douglas, V. I. (1983). Attention and cognitive problems. In M. Rutter (Ed.), Developmental
neuropsychiatry (pp. 280-329). New York: Guilford Press.
Douglas, V. 1., & Peters, K. G. (1979). Toward a clearer definition of the attentional deficit of
hyperactive children. In G. A. Hale & M. Lewis (Eds.), Attention and cognitive develop-
ment (pp. 173-248). New York: Plenum.
Draeger, S., Prior, M., & Sanson, A. (1986). Visual and auditory attention performance in
hyperactive children: Competence or compliance Journal of Abnormal Child Psychol-
ogy, 14,411-424.
Edelbrock, C., & Achenbach, T. A. (1984). The Teacher Version of the Child Behavior Profile:
I. Boys aged 6-11. Journal of Consulting and Clinical Psychology, 52, 207-217.
Egeth, H., & Bevan, W. (1973). Attention. In B. B. Wolman (Ed.), Handbook of general
psychology. Englewood Cliffs, NJ: Prentice-Hall.
Foch, T. T., & Plomin. R (1980). Specific cognitive abilities in 5- to 11-year-old twins.
Behavior Genetics, 10, 507-520.
Gibson, E., & Rader, N. (1979). Attention: Perceiver as performer. In G. Hale & M. Lewis
(Eds.), Attention and cognitive development (pp. 1-22). New York: Plenum.
Goff, J. R, Anderson, H. R, & Cooper, P. F. (1980). Distractibility and memory deficits in
long-term survivors of acute lymphoblastic leukemia. Developmental and Behavioral
Pediatrics, 1, 158-163.
Gordon, M. (1979). The assessment of impulsivity and mediating behaviors in hyperactive
and nonhyperactive children. Journal of Abnormal Child Psychology, 7, 317-326.
Gordon, M., McClure, F. D., & Post, E. M. (1983). Interpretive guide to the Gordon Diagnostic
System. DeWitt, NY: Gordon Systems.
Gosling, H., & Jenness, D. (1974). Temporal variables in simple reaction times of mentally
retarded boys. American Journal of MentaI' Deficiency, 79, 214-224.
Goyette, C. H., Conners, C. K., & Ulrich, R F. (1978). Normative data for Revised Conners
Parent and Teacher Rating Scales. Journal of Abnormal Child Psychology, 6, 221-236.
Gray, J. A. (1972). Learning theory, the conceptual nervous system, and personality. In V. D.
Neblitsin & J. A. Gray (Eds.), Biological bases of individual behavior (pp. 370-398).
Gray, J. A. (1982). The neuropsychology of anxiety: An enquiry into the functions of the
septo-hippocampal system. New York: Oxford University Press.
Hale, G., & Lewis, M. (1979). Attention and cognitive development. New York: Plenum.
Heilman, K., & Valenstein, F. (1985) Clinical neuropsychology (2nd ed.). New York: Oxford
University Press.
Holdsworth, 1., & Whitmore, K. (1974). A study of children with epilepsy attending ordinary
schools. I: Their seizure patterns, progress, and behaviour in school. Developmental and
Child Neurology, 16, 746-758.
James, W. (1890). The principles of psychology. New York: Holt.
Kagan, J., Rosman, B., Day, D., Albert, J., & Phillips, W. (1964). Information processing in the
child: Significance of analytic and reflective attitudes. Psychological Monographs,
78,(1, Whole No. 578).
Kaufman, A. S. (1975). Factor analysis of the WISC-R at eleven age levels between 6V, and
16V, years. Journal of Consulting and Clinical Psychology, 43, 135-147.
Kaufman, A. S. (1980). Issues in psychological assessment: Interpreting the WISC-R intel-
ligently. In B. Lahey & A. Kazdin (Eds.), Advances in clinical child psychology (Vol. 3,
pp. 177-214). New York: Plenum.
Kazdin, A. E., Esveldt-Dawson, K. & Loar, 1. 1. (1983). Correspondence of teacher ratings and
direct observations of classroom behavior of psychiatric inpatient children. Journal of
Abnormal Child Psychology, 11, 549-564.
Keough, B. K., & Margolis, J. S. (1976). A component analysis of attentional problems of

educationally handicapped boys. Journal of Abnormal Child Psychology, 4, 349-359.
Klee, S. H., & Garfinkel, B. D. (1983). The computerized continuous performance task: A new
measure of inattention. Journal of Abnormal Child Psychology, 11, 487-496.
Klonoff, H., Low, M., & Clark, C. (1977). Head injuries in children: A prospective five year
follow-up. Journal of Neurology, Neurosurgery, and Psychiatry, 40,1211-1219.
Lapouse, R, & Monk, M. (1958). An epidemiological study of behavior characteristics in
children. American Journal of Public Health, 48, 1134-1144.
Levin, H. S., Benton, A. L., & Grossman, R G. (1982). Neurobehavioral consequences of
closed head injury. New York: Oxford University Press.
Mackworth, J. (1976). Development of attention. In V. Hamilton & M. D. Vernon (Eds.). The
development of cognitive processes. New York: Academic Press.
Margolis, J. S. (1971). Academic correlates of sustained attention. Unpublished doctoral
dissertation, University of California, Los Angeles.
Mash, E. J., & Dalby, J. T. (1979). Behavioral interventions for hyperactivity. In R Trites (Ed.),
Hyperactivity in children: etiology, measurement, and treatment implications (pp.
161-216). Baltimore University Park Press.
Mash, E., & Terdal, L. (1987). Behavioral assessment of childhood disorders (2nd ed.). New
Matthews, C. G., & Klove, H. (1965). Manual for the Wisconsin adaptation of the Halstead-
Reitan Neuropsychological Test Battery. Madison: University of Wisconsin Hospitals.
Milich, R (1984). Cross-sectional and logitudinal observations of activity level and sustained
attention in a normative sample. Journal of Abnormal Child Psychology, 12, 261-276.
Milich, R, & Kramer, J. (1984). Reflections on impUlsivity: An empirical investigation of
impulsivity as a construct. In K. Gadow & I. Bialer (Eds.), Advances in learning and
behavioral disabilities (Vol. 3, pp. 57-94). Greenwich, CT: JAI Press.
Milich, R, & Loney, J. (1974). The factor composition of the WISC for hyperkinetic/MBD
males. Journal of Learning Disabilities, 12, 67-71.
Needleman, H. L., Gunnoe, c., Leviton, A., Reed, R, Peresie, H., Maher, C., & Barrett, P.
(1979). Deficits in psychologic and classroom performance of children with elevated
dentine lead levels. New England Journal of Medicine, 300, 689-695.
Nichols, P. (1980). Earlv antecedents of hyperactivity. Neurology, 30, 439.
O'Dougherty, M., Nuechterlein, K. H., & Drew, B. (1984). Hyperactive and hypoxic children:
Signal detection, sustained attention, and behavior. Journal of Abnormal Psychology,
93, 178-191.
Ownby, R L., & Matthews, C. G. (1985). On the meaning of the WISC-R third factor: Relations
to selected neuropsychological measures. Journal of Consulting and Clinical Psychol-
ogy, 53, 531-534.
Pfiffner, L. J., Rosen. L. A., & O'Leary, S. G. (1985). The efficacy of an all-positive approach to
classroom management. Journal of Applied Behavior Analysis, 18,257-262.
Piontrowski, D., & Calfee, R (1979). Attention in the classroom. In G. Hale & M. Lewis (Eds.),
Attention and cognitive development (pp. 297-330). New York: Plenum.
Plaisted, J. R, Gustavson, J. L., Wilkening, G. N., & Golden, C. J. (1983). The Luria-Nebraska
Neuropsychological Battery-Children's Revision: Theory and current research findings.
Journal of Clinical Child Psychology, 12, 13-21.
Plomin, R, & Foch, T. T. (1981). Hyperactivity and pediatrician diagnoses, parental ratings,
specific cognitive abilities, and laboratory measures. Journal of Abnormal Child Psy-
chology, 9, 55-64.
Porteus, S. (1965). Porte us Maze Tests: Fifty years' application. Palo Alto, CA: Pacific Books.
Prinz, R J., Tarnowski, K. J., & Nay, S. M. (1984). Assessment of sustained attention and
distraction in children using a classroom analogue task. Journal of Clinical Child Psy-
chology, 13, 250-256.
ATTENTION 175
Quay, H. C., & Peterson, D. R. (1983). Interim manual for the Revised Behavior Problem
Checklist. Unpublished manuscript, University of Miami.
Rapport, M. D., Stoner, G., DuPaul, G. J., Birmingham, B. K, & Tucker, S. (1985). Meth-
ylyphenidate in hyperactive children: Differential effects of dose on academic learning,
and social behavior. Journal of Abnormal Child Psychology, 13, 227-244.
Reitan, R. & Davison, L. (1974). Clinical Neuropsychology: current status and applications.
New York: Wiley.
Rosenbert, R. P., & Beck, S. (1986). Preferred assessment methods and treatment modalities
for hyperactive children among clinical child and school psychologists. Journal of
Ross, D. M., & Ross, S. A. (1982). Hyperactivity: Current issues, research, and theory (2nd
ed.). New York: Wiley.
Rosvold, H. E., Mirsky, A. F., Sarason, I., Bransome, E. D., & Beck, 1. H. (1956). A continuous
performance test of brain damage. Journal of Consulting Psychology, 20, 343-350.
Rourke, B. P. (1981). Neuropsychological assessment of children with learning disabilities.
In S. Filskov & T. Boll (Eds.), Handbook of clinical neuropsychology (Vol. 1, pp. 453-
480). New York: Wiley.
Rourke, B. P., Bakker D. J., Fisk, J. L., & Strang, J. D. (1983). Child neuropsychology. New York:
Guilford Press.
Rourke, B. P., & Gates, R. D. (1980). Underlining test: Preliminary norms. Windsor, Ontario:
Authors.
Rourke, B. P., & Orr, R. R. (1977). Prediction of the reading and spelling performances of
normal and retarded readers: Four-year follow-up. Journal of Abnormal Child Psychol-
ogy, 5, 9-20.
Routh, D. K., & Schroeder, C. S. (1976). Standardized playroom measures as indices of
hyperactivity. Journal of Abnormal Child Psychology, 4, 199-207.
Salkind, N. J., & Nelson, C. F. (1980). A note on the developmental nature of reflection-
impulsivity. Developmental Psychology, 16, 237-238.
Schain, R. J., Ward, J. W., & Guthrie, D. (1977). Carbamazine as an anticonvulsant in chil-
dren. Neurology, 27, 476-480.
Skinner, B. F. (1953). Science and human behavior. New York: Macmillan.
Skinner, B. F. (1969). Contingencies of reinforcement: A theoretical analysis. New York:
Appleton-Century-Crofts.
Stein, L. (1985, October). Methylphenidate and the effects of reinforcement. Paper presented
at the Hogg Foundation Conference on Attention Deficit Disorders, San Antonio, TX.
Stewart, K J., & Moely, B. E. (1983). The WISC-R third factor: What does it mean? Journal of
Consulting and Clinical Psychology, 51, 940-941.
Swanson, J., & Kinsbourne, M. (1979). The cognitive effects of stimulant drugs on hyperac-
tive children. In G. Hale & M. Lewis (Eds.), Attention and cognitive development (pp.
249-274). New York: Plenum.
Titchener, E. B. (1924). A textbook of psychology. New York: Macmillan.
Trites, R. L., Blouin, A. G. A., & LaPrade, K. (1982). Factor analysis of the Conners Teacher
Rating Scale based on a large normative sample. Journal of Consulting and Clinical
Ullmann, R. K, Sleator, E. K., & Sprague, R. L. (1984). A new rating scale for diagnosis and
monitoring of ADD children. Psychopharmacology Bulletin, 20, 160-164.
Wechsler, D. (1974). The Wechsler Intelligence Scale for Children-Revised. New York: Psy-
chological Corp.
Weiss, G., & Hechtman, L. (1979). The hyperactive child syndrome. Science, 205, 1348-
1354.
Werry, J., & Quay, H. (1971). The prevalence of behavior symptoms in younger elementary
school children. American Journal of Orthopsychiatry, 41, 136-143.
Whalen, C., & Henker, B. (1980). Hyperactive children: The social ecology of identification
and treatment. New York: Academic Press.
Whalen, C., Henker, B., & Dotemoto, S. (1980). Methylphenidate and hyperactivity: Effects
on teacher behaviors. Science, 208, 1280-1282.
Witkin, H. A., Oltman, P., Raskin, E., & Karp, S. (1971). Manual for the Children's Embedded
Figures Test. Palo Alto, CA: Consulting Psychologists Press.
Wolf, S. M., & Forsythe, A. (1978). Behavior disturbance, phenobarbital, and febrile seizures.
Pediatrics, 61, 728-731.
Woodcock, R. W. (1976). Goldman-Fristoe-Woodcock Auditory Skills Test Battery Tech-
nical Manual. Circle Pines, MN: American Guidance Service.
7
Clinical Assessment of Memory in

Children
A Developmental Framework for Practice
THOMAS A. BOYD
INTRODUCTION
An assessment of memory functions has long been considered an essen-

tial component of the neuropsychiatric mental status exam, as well as the
more comprehensive, standardized neuropsychological evaluation. The
vast proliferation of literature on the experimental investigation of human
learning and memory (Craik, 1979; Horton & Mills, 1984) amply attests to
the interest and importance attached to understanding and measuring this
realm of cognitive activity. Russell (1981) has pointed out that this rich
heritage of research provides a sound practical and theoretical foundation
for the construction of clinical memory tests but has not led to parallel
advances in their actual construction. Not even such clinically derived
theories as Luria's (1973, 1976, 1980) on the neuropsychological mecha-
nisms underlying memory and its disorders have led to the construction
of a comprehensive and valid psychometric assessment of memory.
The lack of a sound psychometric tool for assessing memory func-
tions is a well-recognized deficiency in the field of adult neuropsychology
(Erickson & Scott, 1977; Russell, 1981). Lezak's (1983) authoritative re-
view of memory functions and associated adult assessment procedures
amply illustrates the lack of theoretically derived, well-validated, and
extensively normed instruments available to clinical practitioners. The
woeful state of current practice is, perhaps, best illustrated by clinicians'
THOMAS A. BOYD Cleveland Metropolitan General Hospital, and Case Western Reserve
University School of Medicine, Cleveland, Ohio.
177
178 THOMAS A. BOYD
continued reliance on the Wechsler Memory Scale (WMS; Stone &

Wechsler, 1945; Wechsler, 1945) despite its numerous clinical and psy-
chometric limitations (Prigitano, 1977, 1978). It is difficult to understand
the efforts to salvage this instrument since it seems highly unlikely that
repeated efforts to revise (Russell, 1975; Stones, 1979), restandardize
(Hulicka, 1966; Ivison, 1977; Klonoff & Kennedy, 1966), or revamp the
WMS will do much to advance the efforts of clinical practice to keep pace
with theoretical and technical developments.
When one considers the singular importance of learning and memory
to the developmental tasks of the school-age child, the dearth of appropri-
ate instruments for assessing the memory performance of children repre-
sents an even larger gap between current knowledge and clinical practice.
The topic of memory in childhood has become one of the most active
areas of research in cognitive development since the late 1960s (Kail &
Hagen, 1982), and an increasing number of reviews have appeared in the
literature that provide an overview and synthesis of advances in the field
(d. Brown, 1975, 1979; Brown, Bransford, Ferrara, & Campione, 1983;
Hagen, Jongeward, & Kail, 1975; Kail & Hagen, 1982; Pressley & Brainerd,
1985). Concurrently, though unrelated, child neuropsychology has
emerged as a subspecialty within clinical neuropsychology with its own
growing body of literature. Yet not a single text devoted to the discipline
of child neuropsychology has specifically addressed the topic of either
memory development or memory assessment (e.g., Gaddes, 1980; Hynd &
Obrzut, 1981; Obrzut & Hynd, 1986; Rourke, Bakker, Fisk, & Strang, 1983;
Spreen, Tupper, Risser, Tuokko, & Edgell, 1984; Tarter & Goldstein, 1984).
Ironically, then, it seems that memory assessment has been a forgot-
ten area in clinical neuropsychology, particularly in the subspecialty de-
voted to children. Present trends in clinical neuropsychology are actively
promoting movement away from merely descriptive diagnostic endeavors
toward prescriptive, treatment-oriented assessments that bridge the gap
between functional neuropsychological diagnosis and rehabilitation (e.g.,
Rourke, Fisk, & Strang, 1986; Wedding, Horton, & Webster, 1986). This is
a propitious time, then, for taking the steps necessary for developing
effective memory assessment techniques that incorporate mainline trends
in clinical neuropsychology.
This chapter is intended as a preliminary step in the process neces-
sary for the construction of a developmentally based, prescriptive assess-
ment of memory functions in children. Beginning with a critical review of
currently available clinical assessment instruments, this chapter will pro-
ceed with a selected review of research models on the development of
memory in childhood. This will provide a heuristic framework for under-
standing the cognitive-developmental changes and situational variables
that influence mnemonic performance. Implicit to these models is a set of
guidelines or principles that can be used profitably in the selection and
CHILDREN'S MEMORY 179
construction of memory tasks sensitive to developmental changes in

memory performance. The chapter will conclude with a description of
clinically relevant tasks and procedures recommended for inclusion in a
children's memory battery.
CURRENT STATUS OF MEMORY ASSESSMENT WITH

CHILDREN
Table 1 presents a selected listing of published tests and procedures

available for assessing memory abilities in children. Inclusion in the list
was based primarily on popularity and high frequency of use in general
clinical child and psycho educational practice and, secondarily, on a test's
representativeness of a genre of procedures commonly employed in
practice.
Because memory abilities exhibit a clear and reliable developmental
progression over a wide age span (Hagen et al., 1975), tests of memory
have been included in standardized intelligence batteries since their in-
troduction (Anastasi, 1982). This fact is recognized in each of the four
major intelligence tests currently in use, particularly the McCarthy Scales
and the newly revised Stanford-Binet, each of which permits the com-
putation of a short-term memory score that can be compared with other
broad areas of a child's functioning. Cooper (1982) has developed a
qualitative appraisal of memory based on asking the child to recall vari-
ous portions of the Wechsler Intelligence Scale for Children-Revised
(WISC-R) following its administration. It is surprising to note that, except
for the WISC-R, each of the listed intelligence tests devotes proportionally
more attention to the assessment of memory abilities than either of the
two major standardized neuropsychological batteries listed in Table 1.
Memory assessment appears to be a rather incidental component to the
Halstead-Reitan procedures, whereas the memory scale of the Luria-Ne-
braska and its children's revision possess limited item pools and poten-
tially serious psychometric weaknesses that severly limit their interpre-
tive value (Lezak, 1983).
Diagnostic psychoeducational batteries frequently contain subtests or
procedures that tap children's memory abilities, such as the Detroit Tests
of Learning Aptitude (DTLA; Baker & Leland, 1967). The DLTA has been
revised recently (DTLA-2; Hammill, 1985) and deserves mention here
because of the battery's inclusion of complementary verbal and nonverbal
measures of short-term memory. Although the revision represents an im-
provement, the DTLA-2 shares a number of problems with its predeces-
sor. These include poor design and limited reliability for some subtests,
particularly some of the memory sub tests , and unsupported claims for
validity (Radencich, 1986).
TABLE 1. Selected Listing of Published and Procedures for Assessing Children's Memory
Assessment instrument Stimulus Response Age range

subtest Task description input output (years-months) Comment
Wechsler Intelligence
Scale for Children-Re-
vised (Wechsler, 1974)
Information Long-term verbal recall of general Verbal Verbal 6-0 to 16-11 Based on Guilford's (1967)
information Structure of Intellect
Digit Span Short-term memory span for digits Verbal Verbal! Model.
pointing Vocabulary also may be a
Arithmetic Numerical reasoning, long-term Verbal Verbal measure of long-term ver-
memory bal memory.
Digit Span can be adapted to
pointing (Smith, 1975).
Cooper (1982) has developed
a Post-Wechsler Memory
Scale.
Stanford-Binet Intelligence
Scale (Thorndike,
Hagen, & Sattler, 1986)
Bead Memory Replication of bead sequences Visual Motor 2-0 to 18 Short-term memory com-
from memory prises one of three broad-
Memory for Sentences Sentence repetition Verbal Verbal band factors in the tests'
Memory for Digits Short-term memory span for digits Verbal Verbal hierarchical model of cog-
Memory for Objects Sequential pointing to objects pre- Visual Motor nitive abilities. A standard
viously displayed Age Score is derived for
short-term memory. Over-
comes limitations of the
1960 Stanford-Binet (Her-
man & Merrill, 1960/1973)
owing to discontinuous
presentation of memory
tasks across age levels.
McCarthy Scales of Chil-
dren's Abilities (McCar-
thy, 1972)
Pictorial Memory Short-term memory for objects Visual! Verbal 2-6 to 8-6 Difficult to make com-
verbal parisons between subtest
Tapping Sequence Short-term nonverbal sequential Visual! Motor scores.
Verbal Memory I and memory auditory
II Word and sentence repetition; Verbal Verbal
Numerical Memory I short-term story recall
and II Short-term memory span for digits Verbal Verbal
Kaufman Assessment Bat-
tery for Children (Kauf-
man & Kaufman, 1983a,
1983b)
Face Recognition Short-term memory for faces Visual Motor 2-6 to 4-11 All of the achievement sub-
Hand Movements Visual Motor 2-6 to 12-5 tests are considered to in-
Short-term recall of sequential volve long-term memory
Number Recall hand movements Verbal Verbal 2-6 to 12-5 (Kaufman & Kaufman,
Word Order Short-term recall of sequential Verbal Verbal! 4-0 to 12-5 1983b).
hand movements motor
Spatial Memory Short-term memory span for digits Visual Motor 5-0 to 12-5
Sequential touching of previously
named objects
Recall locations of pictures ar-
ranged randomly on a page
Detroit Tests of Learning
Aptitude (2nd ed.)
DTLA-2 (Hammill,
1985)
Sentence Imitation Sentence repetition Verbal Verbal 6-0 to 17-11 Revision of the DTLA (Baker
Oral Directions Perform orally directed series of Verbal Motor & Leland, 1967).
actions Poor subtest reliability (Ra-
(continued)
TABLE 1. (Continued)
Assessment instrument Stimulus Response Age range

subtest Task description input output (years-months) Comment
Word Sequences Repeat string of unrelated words Verbal Verbal dencich, 1986).
Design Reproduction Draw geometric form from Visual Motor
Object Sentences memory Visual Motor/
Written recall of series of pictured verbal
Letter Sequences objects Visual Motor/
Written recall of sequence of let- verbal
ters
Halstead-Reitan Neuropsy-
chological Batteries (Re-
itan & Davison, 1974)
Target Test Reproduction of visual-spatial Visual Visual- 5 to 8 No specific measures of ver-
configurations motor bal memory appear in the
Tactual Performance Draw shapes and location of stim- Tactile Visual- 5 to 14 battery other than those
Test ulus blocks from memory motor associated with WISC-R.
Category Test Recall of previously learned deci- Visual Motor/ 9 to 14
sion principles nonverbal
Luria-Nebraska Neuropsy-
chological Battery
(Golden, Hammeke, &
Purisch, 1980)
Memory Scale Eleven memory tasks (13 test Verbal Verbal 13 to adult Test is normed on an adult
items) assessing short-term ver- sample but frequently ap-
bal and nonverbal memory with Visual Verbal! plied to adolescents.
and without interference motor
Luria-Nebraska Neuropsy-
chological Battery Chil-
dren's Revision (Golden,
1981)
Memory Scale Eight memory tasks (8 items) as- Verbal Verbal 8 to 12 Test items are based on
sessing short-term verbal and adult version of the
nonverbal memory with and Visual Verbal! LNNB.
without interference motor
Denman Neuropsychologi- 11 subtests measuring immediate Verbal Verbal 10 to adult Format of battery highly
cal Memory Scale (Den- recall, short-term memory, and similar to Wechsler Intelli-
man, 1984) memory for remotely stored in- Visual- Visual- gence Scales.
formation across verbal (6 sub- motor motor Major weaknesses apparent
tests) and nonverbal (5 subtests) in test construction. Poor
domains Auditory Verbal norms and validity.
Benton Revised Visual Re-
tention Test (Benton,
1974)
Administration A 10-second exposure to geometric Visual Motor 8 to adult Norms for age 8 to 14 for
shapes with immediate recall Administration A and C
reproduction only.
Administration B 5second exposure with immedi- 15 to adult Norms take into account age
ate recall and intelligence level. No
Administration C Copy shapes; no recall norms for Administration
Administration D 10-second exposure with recall af- D.
ter 15-second delay 3 alternate forms.
Memory for Designs Test 15 geometric designs, exposed one Visual Motor 8-6 to 60 Corrections for age and in-
(Graham & Kendall, at a time for 5 seconds followed telligence are recom-
1960) by immediate recall! mended for children.
reproduction Scoring system of question-
able reliability.
Auditory-Verbal Learning Immediate recall of 15-word list Verbal Verbal 5 to 15 Limited norms available.
Test (Rey, 1964) wi th repeated trials Adults
184 THOMAS A. BOYD
The Benton Visual Retention Test (Benton, 1974), the Memory for
Designs Test (Graham & Kendall, 1960), the Auditory Verbal Learning Test
(Rey, 1964), and other similar single-task procedures represent the main-
stay of specific memory assessment in clinical practice. Beyond re-
strictions in their task composition and complexity, most of these pro-
cedures possess distinct psychometric weaknesses, such as poor
reliability, as well as inadequacies in their standardization and limitations
in age-based norms. Further, systematic comparison across these various
measures is invalidated by the differences in their standardization and
psychometric properties.
The Denman Neuropsychology Memory Scale (DNMS; Denman,
1984), to date, does not appear to be a frequently employed test in clinical
practice. Nevertheless, it has been included in Table 1 because it repre-
sents a contemporary effort to address some of the deficiencies cited in
the preceding paragraph regarding single-task procedures. The DNMS
comprises six Verbal subtests and five Nonverbal subtests packaged in a
format similar to the Wechsler intelligence scales. Scores are derived for
each subtest (M = 10, SD = 3) and for Verbal, Nonverbal, and Full Scale
Memory Quotients (M = 100, SD = 15). The subtests of the DNMS have
been based on existing assessment methods drawn from clinical tradition,
and their combined standardization within a single battery ostensibly
permits valid systematic comparisons across the various subtests and be-
tween broad domains of memory (i.e., Verbal and Nonverbal). Unfortu-
nately, the DNMS is a seriously flawed psychometric instrument that,
even upon casual review, is severely limited in its clinical utility (Hooper
& Boyd, 1987). The subtests of the DNMS provide solely a quantitative
assessment of memory functions to the exclusion of important qualitative
factors (e.g., there are no cross-modal pairings of stimulus materials). The
normative base for the battery is extraordinarily weak at present and not
representative of the general population. In general, it does not appear
that appropriate standards for test construction (i.e., the establishment of
reliability and validity) were followed. Aslo, it represents a rather am-
bitious attempt to assess memory functions across nearly a 60-year age
span (age 10 and older) without recourse to a theoretical model of memory
development.
Although the DNMS is not by any means unique in the kind and
seriousness of its deficiencies, it amply illustrates the fallacy inherent to
any modern effort to develop a battery of memory tests simply by collect-
ing new norms for a set of unimodal traditional methods. Memory pro-
cesses are not simple, and measurement tools consisting of a variety of
separate unimodal measures, no matter how venerated, will only con-
tinue the tradition of piecemeal assessment and will not adequately eval-
uate the complexities of memory performance and its dysfunctions.
The foregoing caveats regarding memory tests and their construction
are pertinent to both child- and adult-oriented instruments. A further

caution is necessary, though, when considering the composition of a chil-
dren's memory battery. Here, perhaps even more than in other areas of
child assessment, one must be strongly cautioned against the simple
adoption of scaled-down adult procedures. Whereas clinical models of
memory functions may be appropriate to the assessment of acquired dis-
orders of memory in adults (e.g., Luria, 1976), these models ignore the
important developmental influences operative in the growing child. The
central nervous system of the child is undergoing rapid change, and both
quantitative and qualitative differences in the organization and pathology
of higher cognitive functions should be expected from those found in
adults (Hooper & Boyd, 1986). Accordingly, appropriate assessment in-
struments must employ a cognitive-developmental framework that takes
into consideration the profound quantitative and qualitative changes in
mnemonic performance across the age span. Assessment approaches that
ignore these important issues may lack any meaningful diagnostic and
prescriptive relevance.
MODELS OF MEMORY DEVELOPMENT IN CHILDHOOD
In this section, a selected review of models of memory that contain

developmental implications will be presented. The intent is not to be
exhaustive but, rather, to highlight trends within the literature that have
both developmental and clinical relevance. Following a brief introduc-
tion, this section will discuss the information-processing models of mem-
ory proposed by Atkinson and Shiffrin (1968) and Craik and Lockhart
(1972), the tetrahedral model of memory proposed by Jenkins (1979), and,
finally, a developmentally sensitive model of memory performance pro-
posed by Brown (1975).
One dominant theme is evident throughout the literature: The devel-
opmental changes that mark the progression toward mnemonic compe-
tence are attributable to the child's growing proficiency in the use of
strategies to aid the encoding and retrieval of information. As summarized
by Kail and Hagen (1982), the attainment of mnemonic competence fol-
lows an invariant developmental pattern: (1) infrequent use of strategies
in children aged 6 and younger, (2) a transitional phase from 7 to 10 years
of age when strategies begin to emerge, are implemented with increasing
consistency, and become gradually refined, and (3) the beginning of ma-
ture strategy use with further gradual refinement in the effectiveness and
flexibility with which they are implemented.
In fact, Kail and Hagen (1982) argued that the developmental progres-
sion in strategic competence is what accounts for the nature of individual
differences in children's memory, rather than the concept of a "general
186 THOMAS A. BOYD
memory ability." For example, Stevenson, Parker, and Wilkinson (1975)

administered 11 memory tasks to 255 5-year-olds and found a median
correlation of only .14, with less than half of the correlations reaching
statistical significance. Similarly, Kail (1975) found a median correlation
of only .18 among eight memory tasks administered to a sample of 8- and
9-year-olds. Of course, cross-validation of these results and replication of
the findings with other measures of memory and with different samples
will be necessary before firm conclusions can be drawn. However, the
findings do suggest that the concept of a "general memory ability" does
not adequately account for individual differences in children's memory
performance.
From a different perspective, Kail (1979) tested the hypothesis that
children of the same age may vary in the accuracy of their memory perfor-
mance depending on whether or not they are efficient and consistent in
the execution of mnemonic strategies. He tested 8- and ii-year old chil-
dren on a battery of memory tasks that included measures for which
strategies were appropriate, as well as measures presumed to be relatively
free of strategic influence. The data were then factor-analyzed. Consistent
with the hypothesis, the three strategy measures loaded on a single factor
for the ll-year-old sample, but on three separate factors for the B-year-
olds. Thus, strategic competence appeared to emerge as a source of indi-
vidual differences with increasing age.
Information-Processing Models
Although no universally accepted model of memory development
has appeared to date, information-processing models of memory are ex-
tant. These embrace the concept of memory as a dynamic phenomenon
that undergoes predictable changes with development. To be discussed
are the multistore model of memory proposed by Atkinson and Shiffrin
(1968) and the levels-of-processing model proposed by Craik and Lock-
hart (1972).
The Multistore Model of Memory

The multi store model of memory accommodates developmental
changes in two ways. A distinction is made between structural features of
memory (Le., the physical system and built-in processes of memory that
impose the biological limitations of memory abilities) and control pro-
cesses, such as mnemonic strategies. which are under the direction and
control of the individual. The structural features are composed of three
components: (1) a sensory register that provides a literal copy of informa-
tion but is subject to rapid decay, (2) a short-term store that functions as
the individual's working memory and, although greatly limited in capaci-
ty, can maintain information by rehearsal, and (3) a long-term store with
unlimited capacity. Changes in memory ability that occur during infancy
and early childhood are usually attributed to changes in the child's struc-
tural features that possibly reflect developments in neural structure (Kail
& Hagen, 1982). Age-related improvements in school-age children are
attributed to increasing sophistication in the child's available control pro-
cesses. These self-directed processes control the flow of information
among the three structural components and are described as "transient
phenomena under the control of the subject; their appearance depends on
such factors as instructional set, the experimental task, and the past histo-
ry of the subject" (Atkinson & Shiffrin, 1968, p. 106). This latter point has
important clinical significance because it anticipates the model proposed
by Jenkins (1979), which will be discussed later in this section.
The Levels-of-Processing Model of Memory

Craik and Lockhart (1972) have proposed a second information-pro-
cessing model of memory that is also quite compatible with the strategic-
developmental hypothesis. This model abandons the multistore struc-
tural components of the Atkinson and Shiffrin model in favor of a depth
or levels-of-processing approach, which places greater emphasis on the
role of subject-controlled processing in memory. Accordingly, a memory
trace is the product of perceptual analyses of incoming stimuli that pro-
ceed through a number of levels. The persistence of a memory trace is
dependent on the level of processing. Highly transient traces result from
initial levels when only physical or sensory features of a stimulus are
processed. More durable and resistant traces result from deeper levels of
processing which are concerned with pattern recognition and the ex-
traction of meaning, and which are more cognitive and more semantic in
nature. The individual memorizer plays an active role in directing the
amount and type of processing employed in response to the particular
material and the nature of the task demands. Developmentally, the model
is consistent with the hypothesis that with increasing maturity an indi-
vidual attains greater control and flexibility over the activation of deeper
levels of processing (Le., strategy use).
Both of the information-processing models discussed were proposed
to explain adult memory behavior, and they predate much of the impor-
tant research on memory development in children. However, the relative
emphasis placed on conscious and strategic mnemonic activity, particu-
larly in the levels-of-processing model, render both theories highly com-
patible with observed developmental memory phenomena. Nevertheless,
few direct links have been drawn in the literature between general memo-
ry theory and developmental research (Kail & Hagen, 1982). Although
various authors [e.g., Brown, 1975; Brown et 01., 1983; Hagen et 01., 1975;
188 THOMAS A. BOYD
Kail, 1979; Ornstein, 1978; Paris & Lindauer, 1982) have used these and
other general models of memory as frameworks for organizing the devel-
opmentalliterature on memory, the models possess few specific develop-
mental implications and do not furnish any particular insights to the task
of constructing a clinically relevant battery of memory tests.
Interactionist Models of Memory

In what follows, two additional models will be described that can
provide a more specific heuristic framework for classifying developmen-
tal memory phenomena. Implicit to these models are principles that can
be used to guide the construction of a clinical memory battery for chil-
dren. These include the Tetrahedral Model (Jenkins, 1979) and Brown's
(1975) Developmentally Sensitive Memory Model.
The Tetrahedral Model of Memory

The first of these is a general model proposed by Jenkins (1979) that
has been adapted specifically by Brown and her colleagues (1983) to
organize the developmental literature on memory. The tetrahedral model
is illustrated in Figure 1 as a pyramid in which each of the four vertices
represents a major factor that should be considered in the construction of
a learning or memory task. The four factors are (1) the nature of the
materials to be learned, (2) the criterial task, (3) the learner's activities,
and (4) the characteristics of the learner. The first two of these factors
define the assessment paradigm, whereas the latter two define the mne-
monic activities and competencies of the child being assessed. The
tetrahedral model accounts for the complexity of memory operations, and
consequently their assessment, by recognizing that each of the four ver-
tices interacts with all others in determining the outcome of a memory
task. This interdependence of factors was anticipated by the information-
processing models of Atkinson and Shiffrin (1968) and Craik and Lock-
hart (1972). Each of the four factors will be discussed briefly below, with a
special focus on highlighting interactions that elicit developmental dif-
ferences in memory performance.
Stimulus Materials. The first point of the pyramid concerns the

nature of the materials that must be remembered. Materials may vary
according to a variety of dimensions such as the modality of presentation,
the physical and psychological structure, degree of conceptual difficulty,
sequence of presentation, and even the number of stimuli presented. Vari-
ations along any of these dimensions interact with the other factors in
important ways. For example, the degree of prior familiarity with the type
CHARACTERISTICS OF
THE LEARNER
Skills
Knowledge
Intent
MotivatIOn
Strategic competence
Metamemory
Cognitive level
Memory capacity
etc.
CRITERIAL TASKS
LEARNING ACTIVITIES
Recognition
Orienting task Recall
Task instructions Problem solving
Strategies Transfer of skills
etc, etc.
STIMULUS MATERIALS
Sensory modality
Physical structure
Psychological structure
Conceptual difficulty
Sequence of materials
Number of stimuli
Cues
etc.
FIGURE 1. An organizational framework for considering variables important to memory as-

sessment. (Adaptedfrom Jenkins, 1979, and Brown, Bransford, Ferrara, & Campione, 1983.)
of stimulus material (learner variable) may affect the child's selection and
the relative success of particular strategic approaches (activity variable).
Also, although recognition memory is generally assumed to be less
difficult than verbatim recall (task variables), the difference may depend
considerably on whether the material is familiar (e.g., words) or not (e.g.,
nonsense syllables) (Jenkins, 1979). Likewise, changing the structure of
the material while holding the type of material constant can alter a child's
success on a memory task (Moley, 1977; Ornstein & Corsale, 1980; Perl-
mutter & Myers, 1979). Accordingly, the speed of identifying the second
word of a pair is greater if the words are drawn from the same categories
(e.g., dog-lion) as opposed to different ones (e.g., dog-piano) (Sperber,
McCauley, Ragin, & Weil, 1979). Further, compared with younger and/or
impaired learners, older and/or efficient learners show a preference for
taxonomic (e.g., horse-cow) groupings over thematic (e.g., horse-saddle)
190 THOMAS A. BOYD
groupings of materials that enhance both their learning rate (Smiley &
Brown, 1979) and memory performance (Overcast, Murphy, Smiley, &
Brown, 1975).
Criteria1 Tasks. The second point of the pyramid, the criterial task, is
the dependent measure in memory investigations. As noted above,
whether the criterial task requires recognition or recall may impact signif-
icantly on a child's relative success. The interaction between the criterial
task and the learner's activities is at the heart of the levels-of-processing
paradigm in memory research (Jenkins, 1979). If the learner's activities are
focused toward deeper, semantic levels of processing, then superior recall
will be expected. However, it is possible to construct memory tasks in a
manner that precludes particular strategies, such as using materials that
are not semantically related or introducing activities that intervene be-
tween learning and recall as a means of circumventing rehearsal strat-
egies. The degree of mastery over the material to be learned also can be
assessed by using criterial tasks that require inferential problem solving,
or transferring acquired knowledge to new contexts.
Learning Activities. The third point of the tetrahedral model, the

activities of the learner, represents a major source of variance in any
developmental model of memory. As stated previously, developmental
changes in memory performance are attributable in large part to the
child's growing competency in strategic mnemonic activity. To the extent
that the examiner's selection of either materials or criterial tasks mini-
mizes opportunities for employing strategies, developmental differences
in memory performance will not be observed; that is, older children will
produce levels and patterns of performance similar to those of younger
children and slower learners (Brown et a1., 1983).
Even an understanding that there is a need to remember appears to
follow a developmental progression. Appel et a1. (1972) found that in
response to different task instructions such as "to look at" or "to re-
member," younger children did not necessarily modify their cognitive
activity in relation to the stimuli. Their level of performance was nearly
identical on the two tasks. In contrast, older children recognized the dis-
tinct cognitive implications of the "to remember" condition and engaged
in a variety of mnemonic strategies that enhanced their performance.
However, Istomina (1975) dramatically demonstrated that the pur-
pose of a memory task can effectively alter the younger child's success on
it. Children were able to recall nearly twice as many items from a five-
word list under a "naturalistic" play situation where the words repre-
sented items to be bought at a grocery store than was the case when the
items were presented solely for the purpose of learning and remembering
them. That is, the younger child was more likely to remember when
learning was an incidental by-product of an intrinsically interesting ac-

tivity, rather than when learning was presented as an end in itself.
Characteristics of the Learner. The fourth point in the pyramid, char-

acteristics of the learner, purposely has been left for last. As a heuristic
model, one could classify a memory experiment by noting the interactions
among the four factors of the tetrahedron. This process could begin by
entering the pyramid from any point because there is no hierarchical
ranking of factorial interactions, nor is there any fixed direction to the
path of interactions. However, when we move from the realm of classify-
ing research experiments to the task of constructing a battery of memory
tasks that are differentially sensitive to developmental phenomena, we
necessarily impose a certain hierarchy and direction to the factorial in-
teractions of interest in the model. When we engage in clinical assess-
ment, we are in pursuit of information about only one factor in the pyra-
mid, namely, the characteristics of the learner. This information is
unknown at the start, and by systematically varying and controlling in-
teractions involving the stimulus materials, the criterial task, and the
learner's activities, our intent is to determine the mnemonic charac-
teristics intrinsic to the learner.
That this is possible to accomplish is evident in the nomothetic en-
deavors of developmental memory research. It follows logically that this
should be no less possible in the idiographic task of assessing individuals
clinically. Through the use of a battery of memory tests that systemat-
ically vary levels across the first three factors described above, it is prac-
ticable to determine clinically the relative mnemonic competence of a
given individual.
With the possible exception of motivation, which may be more sub-
ject to situational variation, each of the learner characteristics listed in
Figure 1 changes over the course of development. The mature learner
should differ from the immature learner by possessing more skills and by
having a greater fund of knowledge and experience, a greater clarity of
intent in learning situations, and a more advanced level of cognitive de-
velopment. The issue of whether memory capacity, also referred to as
working memory or attentional capacity, undergoes a developmental in-
crease remains controversial in the literature. Although it is well recog-
nized that manifest performance on memory-span tasks such as digit span
increases with age (Wechsler, 1974), what has been disputed is whether
this enhanced performance reflects an increase in the short-term store
capacity of the child or whether it is related to other age-dependent
changes in cognitive operations.
Pascual-Leone (1970) hypothesized that increased memory span cor-
responded with the growth in capacity of a central computing space that
he termed M-space, and some investigators have provided supportive
192 THOMAS A. BOYD
evidence for his theory (Burtis, 1982). Case, Kurland and Goldberg (1982),
however, have proposed that the total processing space required for short-
term memory operations remains constant throughout development. The
increments in functional memory capacity result from reductions in the
space required for executing memory operations. This is made possible by
the growing speed, efficiency, and automaticity of these basic processes.
Chi (1976) suggested that restrictions in the young child's knowledge
base, which may limit the richness of associations of various memory
chunks in the system, may underlie these basic processing inefficiencies.
Brown et 01. (1983) consider the capacity-development issue as moot
and conclude that there is insufficient evidence to support the contention
that total processing capacity increases with age. Rather, they propose
that the interaction of developmental changes in the child's knowledge
base, efficiency of basic processes, and the use of strategies accounts for
the observed increments in memory span. These, in turn, are dependent
on the factorial interactions of the tetrahedral model.
The importance of strategic competence to developmental models of
memory has been stressed throughout this section and will not be further
elaborated here. However, one final characteristic of the learner listed in
Figure 1 remains to be discussed. The concept of metamemory (Flavell,
1970; Flavell & Wellman, 1977; Kreutzer, Leonard, & Flavell, 1975)-that
is, knowing about knowing-is an important developmental variable in-
fluencing interactions involving all factors of the tetrahedral model. The
model serves to remind clinical and experimental investigators that
knowledge of interactions within the tetrahedron are essential to an effec-
tive understanding of memory abilities. Similarly, the maturing child also
needs to know something about his or her own characteristics as a learner,
the significance of various learning activities, the demand characteristics
of various tasks, and the inherent nature of various materials in order to
become an effective learner, or to "learn how to learn" (Brown et 01.,
1983).
The multifactorial nature of the tetrahedral model cannot be over-
emphasized in its importance, although the sheer number of possible
interactions places a cumbersome burden on the practical limitations of
time and needs for efficiency in clinical practice. In addition to six 2-
factor interactions, the model allows for three 3-factor interactions and
one 4-factor interaction; each of these would be subject to systematic
variation across the several characteristics of each factor in order to ex-
haust all the possibilities permitted by the model. A set of guidelines is
required that would assist the clinician in selecting which permutations
of the tetrahedral interactions are likely to be most sensitive to develop-
mental changes in memory, or in discriminating competent from incom-
petent memory abilities. Fortunately, a second model, to be described
below, provides just such a set of guidelines.
A Developmentally Sensitive Model of Memory
Ann Brown (1975) has proposed a model of developmental changes

in memory that takes into account important themes in the developmental
literature on memory. As shown in Figure 2, these themes are expressed
as dichotomies, although they more realistically should be thought of as
continua. Each theme influences the degree to which a particular task is
developmentally sensitive.
The first theme is the strategy versus no strategy distinction. As pre-
viously noted, to the extent that a given memory task requires the efficient
use of strategic operations for success, it will discriminate immature and
inefficient learners from mature and efficient learners. A second dichoto-
my pertains to the difference between mediation and production defi-
ciencies (Flavell, 1970). A mediation deficiency is implicated when a
child is unable to employ a potential mediator or strategy spontaneously
for a given task, and cannot be induced to do so by the examiner even
when given specific instructions or demonstrations. A production defi-
ciency refers to the situation where a child can be induced through train-
ing or instruction to use a strategy despite previous failure to employ that
strategy spontaneously.
The third dichotomy involves the distinction between episodic and
semantic memory systems. Episodic memory refers to memory for di-
rectly experienced instances of discrete perceptual cues that are relatively
meaningless and cannot be deduced inferentially or generalized, and the
criterial task is a reproduction of the actual stimulus (in other words, the
prototypical laboratory memory task). In contrast, semantic memory is
concerned with meaningful holistic units experienced in context that can
be retrieved through imaginative reconstruction of the ideas fundamental
to the stimulus information. One can even retrieve knowledge not directly
experienced but inferred or generalized from the semantic information
presented. Recall for categorical lists or paragraphs are examples of the
semantic memory system.
In particular, tasks constructed according to the factorial combina-
tions of strategic-nonstrategic and episodic-semantic distinctions
should possess differential sensitivity to developmental effects. Analo-
gous to an analysis of variance model, two major types of developmental
effects can be seen. Level differences, reflecting improved performance
(e.g., more items recalled, more accurate recall) with age, would be akin to
a main effect for a developmental stage. Pattern differences refer to in-
teractions between a developmental stage and task variables, such as is
illustrated by the growing preference for taxonomic over thematic catego-
rization by older children (Overcast et al., 1975).
Examination of the four types of tasks permitted by the model yields a
set of general predictions about their relative degree of developmental
YES
Episodic Differences in
> ~ levels, not patterns
Possible dif-
Semantic ferences, pri-
L - - - - - - - - - t.... marily in
(Mediation deficiency)
difference in leve Is patterns
and patterns
Task
Episodic Least sensitive to

developmental differences
-_ ... _.... - Possible differences,

~ primarily in patterns
FIGURE 2. A model for selecting developmentally sensitive memory tasks. (From Brown,1975.]
sensitivity (Brown, 1975). The nonstrategic-episodic combination will in-

clude tasks that minimize demands for active retrieval or acquisition
strategies, and are not dependent on the semantic memory system for
success. Tasks of this type should be the least sensitive to developmental
trends and, therefore, should not discriminate between younger and older
children or unsophisticated and sophisticated learners. The strategic-epi-
sodic pairing should elicit differences in the sheer level of performance
(i.e., more items recalled, more accurate recall), but not in the pattern of
performance (e.g., the order of stimulus presentation is varied across trials
so as to disrupt primacy vs. recency effects for older children). Non-
strategic-semantic tasks should be sensitive primarily to pattern dif-
ferences that reflect the congruence or incongruence between the material
to be learned and the child's level of cognitive development. If the child
can assimilate the material, then effective learning should occur. The
strategic-semantic variation should disclose differences in both perfor-
mance levels and patterns.
SUGGESTIONS FOR A CHILDREN'S MEMORY BATTERY:

GENERAL FORMAT
As noted earlier, implicit to the memory models devised by Jenkins

(1979) and Brown (1975) are a set of principles that can be used to guide
the construction of a developmentally based clinical battery of memory
tests. Before embarking on a description of the types of tasks that might be
nominated for inclusion in such a battery, we should consider several
points about the general format of such a battery.
First, because the hallmark of strategy use is improved performance
over trials, it is essential that tasks incorporate multiple trials in order to
elicit evidence of this important source of developmental differences
(Murphy, Puff, & Campione, 1977). An interesting variation of the tradi-
tional multiple trial learning task is the selective reminding procedure
developed by Buschke (1973, 1974) for verbal word recall, and recently
modified by Gilliam, Fletcher, and Levin (1986) for use with nonverbal
material. The technique permits simultaneous analysis of a child's stor-
age, retention, and retrieval abilities, and it could be adapted easily to
incorporate the strategic-nonstrategic and episodic-semantic dimen-
sions proposed by Brown (1975). Such a procedure may be highly sen-
sitive to differences between normal and disabled learners (Howe, Brain-
erd, & Kingman, 1985).
Another format issue relates to prescriptive planning and the need to
build in test mechanisms that provide clues for remediation. For example,
pairwise combinations of visually and auditorally presented materials
can be arranged in ways that either minimize or facilitate strategy use,
196 THOMAS A. BOYD
thus permitting a continuum of developmental sensitivity. Relative dif-

ferences in performance on paired recall and unimodal recall could also
provide a potentially important clue for prescriptive planning.
However, perhaps the most important source of information relevant
to planning prescriptive interventions may come from an assessment of
the child's metamemory status. The child's awareness of various
mnemonic phenomena and their interrelations may impact significantly
on his or her level and pattern of performance. When the level of a child's
metamemory status is below that expected for his or her chronological
age, immature levels and patterns of performance also can be expected.
Increasing a child's awareness of metamemory variables may help im-
prove memory abilities (Brown, 1978; Meichenbaum & Asarnow, 1978).
Accordingly, a metamemory interview, such as that developed by Kreut-
zer et a1. (1975), should constitute the initial portion of the memory as-
sessment. For example, does the child understand that a longer interval
between learning and recall increases task difficulty, or that verbatim
recall is more difficult than paraphrased recall? This should be supple-
mented by requests for predictive, concurrent, and retrospective ver-
balizations occurring, respectively, before, during, and after task perfor-
mance (Brown et a1., 1983).
A third format issue relates to the importance of attentional abilities
for successful performance on all psychometric tests (see Chapter 6, this
volume), and perhaps memory tests in particular. Disturbances in atten-
tion can interfere significantly with a child's demonstration of memory
skills (Cecci & Tishman, 1984), and thus an assessment of attention
should precede the administration of memory tasks.
SUGGESTIONS FOR A CHILDREN'S MEMORY BATTERY:

TYPES OF TASKS
Turning now to the four ideal tasks outlined by Brown (1975), atten-
tion must be paid to the sequence of their presentation so as to minimize
the potential contamination effects between tasks. In general, for reasons
elaborated below, the sequence of presentation should proceed through a
continuum from nonstrategic to strategic tasks, first with semantic mate-
rials followed by an examination of episodic memory abilities.
Nonstrategic-semantic tasks should reflect learning that arises as the
automatic results of meaningful activity. Essentially, these tasks assess
incidental memory, and because the orienting instructions would not
direct the child toward intentional learning, these tasks must precede
those that would induce specific learning sets. If the material of these
tasks matches the child's level of cognitive development, then memory for
the substance of the material will be involuntary (Brown, 1975, 1979;

Brown et aJ., 1983).
Relatively little of the research literature has examined this type of
learning, although several task paradigms seem appropriate. Basically, the
child is asked to interact with a set of stimulus materials that permit
semantic encoding for a purpose that is not ostensibly for memorizing,
and recall of the material then is assessed. For example, the child could be
asked to categorize a series of pictured objects, or to evaluate a list of
words semantically (e.g., pleasant-unpleasant). Coding tasks also could
be employed, where the child's recall for the association between coded
pairs would be assessed following completion of a clerical coding task
(e.g., numbers and shapes). A task similar to the WISC-R Picture Arrange-
ment subtest also could be adapted. Following a child's successful se-
quencing of pictorial stories, a verbal recall of the stories would be ob-
tained. Finally, a visual-motor task similar to the Bender Visual-Motor
Gestalt Test could be used to assess incidental recall for designs. How-
ever, to comply with the semantic requirements of this category, the
shapes to be copied should be compatible with simple verbal labels. The
designs also should be very simple in order to minimize the graphomotor
skills required.
The selection of tasks for the strategic-semantic category is compli-
cated by the fact that semantic memory appears to occur as the automatic
result of comprehension (Brown, 1975). Wickens's (1972; Esrov, Hall, &
LaFaver, 1974) "release from interference" paradigm is one useful tech-
nique for determining the salience of various levels of semantic related-
ness to the encoding abilities of the child. A series of "release" tasks that
utilize a range of semantic categories could be employed for this purpose.
"False recognition procedures" (Bach & Underwood, 1970) that contain a
variety of distractors (e.g., acoustic, thematic, taxonomic) also can be used
to assess the relative dominance of semantic encoding strategies of the
child. Of course, the use of sentences and paragraphs provides a rich
domain of semantic material that can be adapted in a variety of ways to
assess the child's strategic competence (ef., Brown, 1975; Brown et al.,
1983; Hagen et al., 1975; Kail & Hagen, 1982).
The nonstrategic-episodic category of tasks should not demand any
obvious strategies for efficient performance and should not include stim-
uli that can be semantically encoded. These tasks would be relatively
insensitive to developmental differences. Nevertheless, these types of
tasks should be included because they would provide the best measure of
basic memory capacity against which performance on developmentally
sensitive tasks can be compared. Three types of tasks are particularly
suited to the nonstrategic-episodic criteria.
In a recognition memory task, the child is shown a lengthy series of
198 THOMAS A. BOYD
unrelated pictures sequentially, several of which will appear a second

time in the series. The child is asked to identify whether a picture has
been seen previously or not. The procedure could also be modified for use
with an orally presented list of unrelated words. Although such a pro-
cedure suffers from a possible ceiling effect, the additional use of a judg-
ment-of-recency task overcomes this difficulty, since even adults would
not show a ceiling effect. Here, the child again is shown a lengthy series of
pictures (or list of words). Later, pairs of pictures (or words) from the list
are presented to the child, who is asked to judge which member of the pair
was presented more recently in the list. To minimize developmental ef-
fects predicated on mnemonic strategies such as rehearsal, the series
should contain 20 or more items (Kail & Hagen, 1982). The third type of
task appropriate to the nonstrategic-episodic category involves memory
search rates (Naus & Ornstein, 1977). The child is presented a target stim-
ulus consisting of a set of digits, letters, pictures, or shapes, and then is
asked to search from memory for the target stimulus among an array of
similar configurations.
The strategic-episodic combination of tasks are constructed in a man-
ner such that the use of strategies appropriate to the task should enhance a
child's performance. However, consideration should be given to minimiz-
ing the potential for semantic encoding. Five task types are considered
here. The first two represent modifications of the judgment-of-recency
task described above. With shorter lists, perhaps 7 to 12 pictures, rehears-
al strategies become a useful mnemonic and enhance performance in
older, more efficient learners. Similarly, the provision of cues, such as
color-coding early, middle, and later members of the stimulus series
(Brown, Campione, & Gilliard, 1974), also can aid the judgment of strate-
gic learners.
The third task nominated for inclusion in this category is a modifica-
tion of the memory search paradigm discussed earlier. When the target
stimuli consist of mixed sets of letters and numbers, older children are
able to conduct their search approximately 25% more rapidly than when
the set contains only letters or numbers (Naus & Ornstein, 1977). The
fourth strategic-episodic task consists of recall for serial position (Hagen &
Kingsley, 1968), where the child is shown briefly a series of eight cards,
one at a time, which are placed face down in front of the child in the same
sequence as presented. Then a cue card is shown by the examiner, and the
child is asked to point to the face-down card that matches the cue card.
Multiple trials are employed, but the order of the cards is varied across
trials, and the child is tested for each of the eight serial positions. Results
on this task can be contrasted with a similar task that requires verbal
labeling of the pictures as they are presented. Developmental interactions
between labeling and no-labeling conditions and recall of serial position
(primacy vs. regency) can provide useful information on the influence of
strategy use on performance, A fifth type of task consists of traditional

memory-span tasks, such as digit span, or variations of traditional multi-
ple trial learning tasks using series of related words,
The foregoing list of representative tasks is not intended to be ex-
haustive. Creative application of the principles illustrated in the develop-
mental models of memory potentially could generate a wide array of valid
memory tasks exhibiting differential sensitivity to developmental factors.
CLINICAL RELEVANCE OF THE MODELS
The memory models reviewed in this chapter and the corresponding

tasks selected to illustrate their application were derived on the basis of
developmental considerations. The models predict that use of the four
types of ideal tasks will permit an assessment of memory functions that
discriminates older from younger, or sophisticated from unsophisticated,
learners. However, the implications of these models for assessing patho-
logical conditions of memory are less direct, although several hypotheses
can be made. First, children with impaired mnestic functions should
more closely resemble younger children than same-age peers in both the
level and, increasingly with age, the pattern of their performance. Second,
impairment on developmentally insensitive memory tasks, such as those
comprising the combination of nonstrategic and episodic features, might
herald deficits in basic-level encoding processes symptomatic of prob-
lems in attention/concentration or of damage to subcortical memory
areas. A corollary to this would be that developmentally sensitive tasks
(i.e., strategic-episodic, nonstrategic-semantic, strategic-semantic), in the
absence of these more basic deficits in encoding, likely would be more
sensitive to disturbances in areas of higher cortical functioning.
Prediction of the usual effects on memory performance relative to
right versus left hemisphere damage would form the basis of a third hy-
pothesis. That is, one would expect that recall of stimulus materials that
can be semantically organized would be generally poorer in children with
left hemisphere injuries, whereas performance with nonsemantic mate-
rials would be relatively unaffected. Conversely, right hemisphere inju-
ries would be expected to impair differentially the recall of nonverbal
(i.e., cannot be coded linguistically) materials.
A fourth hypothesis pertains to the strategic-nonstrategic dichotomy.
Tasks that incorporate strategic dimensions and, therefore, involve the
planning, implementation, and evaluation of mnemonic activities, may be
particularly sensitive to anterior brain injuries or perhaps frontal lobe
immaturity. In contrast, nonstrategic tasks may emphasize the more per-
ceptually based mnemonic activities of posterior cerebral regions. Explo-
ration of production and mediation deficiencies also may prove helpful in
200 THOMAS A. BOYD
making diagnostic and prescriptive discriminations. For example, a pro-

duction deficiency may not be indicative of brain pathology, per se, and
has obvious implications for remedial interventions (e.g., the child could
be tutored in the use of mnemonic strategies). Mediation deficiencies, on
the other hand, would require further analyses to determine the nature
and cause of the disturbance and, by implication, would warrant a variety
of perhaps quite different remedial interventions depending on the partic-
ular findings.
The formation of more specific hypotheses must await the actual
construction of a battery of memory tasks for children. Whereas the guide-
lines presented here can be profitably applied to such a task, many deci-
sions about the particular content and format of such a battery will have to
be made empirically.
CONCLUSIONS
This chapter has presented a critique of contemporary approaches to

clinical memory assessment, particularly in the domain of clinical child
neuropsychology. Current practice is viewed as seriously flawed and in
need of a complete overhaul. As a preliminary step in this endeavor, an
approach that takes advantage of the rich heritage of developmental re-
search on memory has been proposed. Recommendations have been made
for the types of tasks which will best sample a clinically relevant range of
developmental memory phenomena, and which should be considered for
inclusion in a children's memory battery. Several hypotheses also are
offered regarding the potential clinical relevance of such a battery in
assessing memory disturbances in children.
REFERENCES
Anastasi, A. (1982). Psychological testing. New York: Macmillan.
Appel, 1., Cooper, R., McCarrell, N., Sims-Knight, J., Yussen, S., & Flavell, J. (1972). The
development of the distinction between perceiving and memorizing. Child Develop-
ment, 43, 1365-1381.
Atkinson, R., & Shiffrin, R. (1968). Human memory: A proposed system and its control
processes. In K. W. Spence & J. T. Spence (Eds.), The psychology of learning and
motivation (Vol. 2). New York: Academic Press.
Bach, M., & Underwood, B. (1970). Developmental changes in memory attributes. Journal of
Educational Psychology, 61, 292-296.
Baker, H., & Leland, B. (1967). Detroit Tests of Learning Aptitude. Indianapolis: Bobbs-
Merrill.
Benton, A. L. (1974). Revised Visual Retention Test: Clinical and experimental applications
(4th ed.). New York: Psychological Corporation.
Bransford, J. D. (1979). Human cognition: Learning, understanding, and remembering. Bel-

mont, CA: Wadsworth.
Brown, A. (1975). The development of memory: Knowing, knowing about knowing, and
knowing how to know. In H. W. Reese (Ed.), Advances in child development and
behavior (Vol. 10, pp. 103-152). New York: Academic Press.
Brown, A. (1978). Knowing when, where and how to remember: A problem in metacogni-
tion. In R. Glaser (Ed.), Advances in instructional psychology. Hillsdale, NJ: Erlbaum.
Brown, A. (1979). Theories of memory and the problems of development: Activity, growth,
and knowledge. In L. S. Cermak & F. I. M. Craik (Eds.), Levels of processing in human
memory (pp. 225-258). Hillsdale, NJ: Erlbaum.
Brown, A., Bransford, J., Ferrara, R., & Campione, J. (1983). Learning, remembering, and
understanding. In P. Mussen, J. Flavell, & E. Markman (Eds.)' Handbook of child psy-
chology (Vol. 3, pp. 77-166). New York: Wiley.
Brown, A., Campione, J., & Gilliard, D. (1974). Recency judgments in children: A prodUction
deficiency in the use of redundant background cues. Developmental Psychology, 10,
303.
Burtis, P. (1982). Capacity increase and chunking in the development of short-term memory.
Journal of Experimental Child Psychology, 34, 387-413.
Buschke, H. (1973). Selective reminding for analysis of memory and learning. Journal of
Verbal Learning and Verbal Behavior, 12, 543-550.
Buschke, H. (1974). Components of verbal learning in children: Analysis by selective re-
minding. Journal of Experimental Child Psychology, 18, 488-496.
Case, R., Kurland, D., & Goldberg, J. (1982). Operational efficiency and the growth of short-
term memory span. Journal of Experimental Child Psychology 33, 386-404.
Cecci, S., & Tishman, J. (1984). Hyperactivity and incidential memory: Evidence for atten-
tional diffusion. Child Development, 55, 2192-2203.
Chi, M. (1976). Short-term memory limitations in children: Capacity on processing deficits?
Memory and Cognition, 4, 559-572.
Cooper, S. (1982). The Post-Wechsler Memory Scale. Journal of Clinical Psychology, 38,
380-387.
Craik, F. I. M. (1979). Human memory. Annual Review of Psychology, 3D, 63-102.
Craik, F., & Lockhart, R. (1972). Levels of processing: A framework for memory research.
Journal of Verbal Learning and Verbal Behavior, 11,671-684.
Denman, S. B. (1984). Denman Neuropsychology Memory Scale. Charleston, SC: Sidney B.
Denman.
Erickson, R. C., & Scott, M. L. (1977). Clinical memory testing: A review. Psychological
Bulletin, 34, 1130-1149.
Esrov, L., Hall, J., & LaFaver, D. (1974). Preschoolers conceptual and acoustic encoding as
evidenced by release from PI. Bulletin of the Psychonomic Society, 4, 89-90.
Flavell, J. (1970). Developmental studies of mediated memory. In H. W. Reese & L. P. Lipsitt
(Eds.), Advances in child development and behavior (Vol. 5, pp. 181-211). New York:
Academic Press.
Flavell, J., & Wellman, H. (1977). Metamemory. In R. V. Kail & J. W. Hagen (Eds.), Perspec-
tives on the development of memory and cognition. Hillsdale, NJ: Erlbaum.
proach. New York: Springer-Verlag.
Gilliam, D., Fletcher, J., & Levin, H. (1986, February). Development of verbal, nonverbal and
recognition memory in children. Paper presented at the Annual Meeting of the Interna-
tional Neuropsychological Society, Denver, CO.
Golden, C. (1981). The Luria-Nebraska Children's Battery: Theory and formulation. In G.
Hynd & J. Obrzut (Eds.), Neuropsychological assessment and the school-age child (pp.
277-302). New York: Grune and Stratton.
202 THOMAS A. BOYD
Golden, c., Hammeke, T., & Purisch, A. (1980). The Luria-Nebraska Neuropsychological
Battery manual. Los Angeles: Western Psychological Services.
Graham, F., & Kendall, B. (1960). Memory for Designs Test: Revised general manual. Percep-
tual and Motor Skills, Monograph Supplement No. 2-VIU, 11, 147-188.
Guilford, J. P. (1967). The nature of human intelligence. New York: McGraw-Hill.
Hagen, J., Jongeward, R, & Kail, R (1975). Cognitive perspectives on the development of
memory. In H. W. Reese (Ed.). Advances in child development and behavior (Vol. 10,
pp. 57-101). New York: Academic Press.
Hagen, J., & Kingsley, P. (1968). Labeling efforts in short-term memory. Child Development,
39, 113-121.
Hammill, D. (1985). Detroit Tests of Learning Aptitude-DTLA-2. Austin, TX: PRO-ED.
Hooper, S. R, & Boyd, T. A. (1986). Neurodevelopmentallearning disorders. In J. E. Obrzut &
G. W. Hynd (Eds.). Child neuropsychology, Volume 2: Clinical Practice. New York:
Academic Press.
Hooper, S., & Boyd, T. (1987). Test review: The Denman Neuropsychology Memory Scale.
International Journal of Clinical Neuropsychology, 9, 141-144.
Horton, D. L., & Mills, C. B. (1984). Human learning and memory. Annual Review of Psychol-
ogy, 35, 361-394.
Howe, M., Brainerd, C., & Kingman, J. (1985). Storage-retrieval processes of normal and
learning-disabled children: A stages-of-learning analysis of picture-word effects. Child
Development, 56, 1120-1133.
Hulicka, 1. M. (1966). Age differences in Wechsler Memory Scale scores. Journal of Generic
Hynd, G. W., & Obrzut, J. E. (Eds.). (1981). Neuropsychological assessment and the school-
age child: Issues and procedures. New York: Grune and Stratton.
Istomina, Z. (1975). The development of voluntary memory in preschool-age children. Sovi-
et Psychology. 13, 5-64.
Ivison, D. J. (1977). The Wechsler Memory Scale: Preliminary findings toward an Australian
standardization. Australian Psychologist, 12, 303-312.
Jenkins, J. J. (1979). Four points to remember: A tetrahedral model of memory experiments.
In 1. S. Cermak & F. 1. M. Craik (Eds.). Levels of processing in human memory (pp. 429-
446). Hillsdale, NJ: Erlbaum.
Kagan. J., Klein. R. Finley, G.. Rogoff. B., & Nolan, E. (1979). A cross-cultural study of
cognitive development. Monographs of the Society for Research in Child Development,
44(5. Serial No. 180).
Kail, R (1975). Interrelations in children's use of mnemonic strategies. Unpublished doc-
toral dissertation, University of Michigan.
Kail, R (1979). Use of strategies and individual differences in children's memory. Develop-
mental Psychology, 15, 251-255.
Kail, R. & Hagen. J. W. (1982). Memory in childhood. In B. Wolman, G. Stricker. S. Ellman,
P. Keith-Spiegel, & D. Palermo (Eds.). Handbook of developmental psychology (pp.
350-366). Englewood Cliffs, NJ: Prentice-Hall.
Kaufman, A. & Kaufman. N. (1983a). Administration and scoring manual for the Kaufman
Assessment Battery for Children. Circle Pines, MN: American Guidance Service.
Kaufman, A., & Kaufman, N. (1983b). Interpretive manual for the Kaufman Assessment
Battery for Children. Circle Pines, MN: American Guidance Service.
Klonoff. H., & Kennedy, M. A. (1966). A comparative study of cognitive functioning in old
age. Journal of Gerontology, 21, 239-243.
Kreutzer, M., Leonard, C., & Flavell, J. (1975). An interview study of children's knowledge
about memory. Monographs of the Society for Research in Child Development, 40.{1,
Whole No. 159).
Lezak. M. D. (1983). Neuropsychological assessment (2nd ed.). New York: Oxford University
Press.
Luria, A. R. (1973). The working brain. New York: Basic Books.

Luria, A. R. (1976). The neuropsychology of memory. New York: Wiley.
Luria. A. R. (1980). Higher cortical functions in man (2nd ed.). New York: Basic Books.
Psychological Corporation.
Meichenbaum, D., & Asarnow, J. (1978). Cognitive behavior modification and metacognitive
development: Implications for the classroom. In P. Kendall & S. Hollon (Eds.), Cog-
nitive-behavioral interventions: Theory, research, and procedures. New York: Academ-
ic Press.
Moley, B. E. (1977). Organizational factors in the development of memory. In R. V. Kail & J.
W. Hagen (Eds.), Perspectives on the development of memory and cognition. Hillsdale,
NJ: Erlbaum.
Murphy, M. D., Puff, C. R., & Campione, J. C. (1977). Clustering measures and organization.
Paper presented at the meeting of the Society for Research on Child Development, New
Orleans.
Naus, M. J., & Ornstein, P. (1977). Developmental differences in the memory search of
categorized lists. Developmental Psychology, 13, 60-68.
Obrzut, J. E., & Hynd, G. W. (Eds.). (1986). Child Neuropsychology. Volume I: Theory and
research. Volume II: Clinical practice. New York: Academic Press.
Ornstein, P. A. (Ed.). (1978). Memory development in children. Hillsdale, NJ: Erlbaum.
Ornstein, P., & Corsale, K. (1980). Organizational factors in children's memory. In C. R. Puff
(Ed.), Memory organization and structure. New York: Academic Press.
Overcast, T., Murphy, M., Smiley, S., & Brown, A. (1975). The effects of instruction on recall
and recognition of categorized lists in the elderly. Bulletin of the Psychonomic Society,
5, 339-341.
Paris, S., & Lindauer, B. (1982). The development of cognitive skills during childhood. In B.
Wolman, G. Stricker, S. Ellman, P. Keith-Spiegal, & D. Palermo (Eds.). Handbook of
developmental psychology. Englewood Cliffs, NJ: Prentice-Hall.
Pascual-Leone, J. (1970). A mathematical model for the transition rule in Piaget's develop-
mental stages. Acta Psychologica, 32, 301-345.
Perlmutter, M., & Myers, N. (1979). Development of recall in 2- to 4-year-old children.
Developmental Psychology, 15, 73-83.
Pressley, M., & Brainerd, C. (Eds.). (1985). Cognitive learning and memory in children. New
York: Springer-Verlag.
Prigatano, G. P. (1977). The Wechsler Memory Scale is a poor screening test for brain
dysfunction. Journal of Clinical Psychology, 33, 772-777.
Prigatano, G. P. (1978). Wechsler Memory Scale: A selective review of the literature. Journal
of Clinical Psychology, 34, 816-832.
Radencich, M. C. (1986). Test review: Detroit Tests of Learning Aptitude (DTLA-2). Journal
of Psychoeducational Assessment, 4,173-181.
Reitan, R. M., & Davison, 1. A. (1974). Clinical neuropsychology: Current status and applica-
tions. New York: Wiley.
Rey, A. (1964). L'examen clinique en psychologie. Paris: Presses Universitaires de France.
Rourke, B. P., Bakker, D. J., Fisk, J. 1., & Strang, J. D. (1983). Child neuropsychology: An
introduction to theory, research and clinical practice. New York: Guilford Press.
Rourke, B., Fisk, J., & Strang, J. (1986). Neuropsychological assessment of children: A treat-
ment-oriented approach. New York: Guilford Press.
Russell, E. W. (1975). A multiple scoring method for the assessment of complex memory
functions. Journal of Consulting and Clinical Psychology, 43, 800-809.
Russell, E. W. (1981). The pathology and clinical examination of memory. In S. B. Filskov &
T. J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 287-319). New York:
Wiley.
Rutter, M. (Ed.). (1983). Developmental neuropsychiatry. New York: Guilford Press.
204 THOMAS A. BOYD
Smiley, S. S., & Brown, A. L. (1979). Conceptual preference for thematic and taxonomic
relations: A nonmonotonic age trend from preschool to old age. Journal of Experimental
Child Psychology, 28, 249-257
Smith, A. (1975). Neuropsychological testing in neurological disorders. In W. J. Friedlander
(Ed.), Advances in neurology (Vol. 7). New York: Raven Press.
Sperber, R. D., McCauley, C., Ragin, R., & Weil, C. (1979). Semantic priming effects on
picture and word processing. Memory and Cognition, 7, 339-345.
Spreen, 0., Tupper, D., Risser, A., Tuokko, H., & Edgell, D. (1984). Human developmental
neuropsychology. New York: Oxford University Press.
Stevenson, H., Parker, T., & Wilkinson, A. (1975). Ratings and measures of memory pro-
cesses in young children. Unpublished manuscript, University of Michigan.
Stone, C. P., & Wechsler, D. A. (1945). Wechsler Memory Scale Form II. New York: Psycho-
Stones, M. J. (1979). Rekitting the Wechsler paired-associate task: The Waterford Index.
Journal of Clinical Psychology, 35, 626-630.
Tarter. R. E., & Goldstein, G. (1984). Advances in clinical neuropsychology (Vol. 2). New
York: Plenum Press.
Terman, L., & Merrill, M. (1973). Stanford-Binet Intelligence Scale. Boston: Houghton
Mifflin. (Original work published 1960)
Thorndike, R., Hagen, E., & Sattler, J. (1986). Guide for administering and scoring the Stan-
ford-Binet Intelligence Scale: Fourth Edition. Chicago: Riverside.
Wechsler, D. A. (1945). A standardized memory scale for clinical use. Journal of Psychology,
19, 87-95.
Wechsler, D. A. (1974). Manual for the Wechsler Intelligence Scale for Children-Revised.
New York: Psychological Corporation.
Wedding, D., Horton, A. M., & Webster, J. (Eds.). (1986). The neuropsychology handbook:
Behavioral and clinical perspectives. New York: Springer.
Wickens, D. (1972). Characteristics of word encoding. In A. W. Melton & E. Martin (Eds.),
Coding processes in human memory. Washington, D.C.: D. H. Winston.
8
Assessing Functional Laterality

JEFFREY W. GRAY and RAYMOND S. DEAN
INTRODUCTION
A considerable amount of research evidence suggests that the cerebral

hemispheres of the human brain serve independent cognitive functions.
Neuropsychological assessments rely upon this concept in making in-
ferences concerning the integrity of the brain. Because individual dif-
ferences exist in the degree to which functions are lateralized to one
hemisphere or the other, most neuropsychological approaches include
some measure of laterality. Although this is a rather complex matter with
adults, developmental factors complicate the procedure further with
children.
A number of investigators have suggested that microanatonomical
and psychophysiological differences in the hemispheres of the brain are
present as early as the 30th week of gestation (Molfese, Freeman, & Paler-
mo, 1975; Wada, Clarke, & Hamm, 1975). However, whereas elementary
lateralization appears to be present in newborns, more complex patterns
of hemispheric specialization continue to develop throughout childhood
(e.g., Satz, Bakker, Teunissen, Goebel, & van der Vlught, 1975). Investiga-
tors also have suggested that the failure to establish consistent functional
lateralization may well predict a number of behavioral anomalies.
As early as the 19th century, researchers examined the complex link
between psychological functions and their localization within the brain.
On the basis of clinical observations of brain-injured patients, Jackson
(1874/1932) proposed that two different, yet coexisting, modes of cog-
nitive processing exist and follow hemispheric lines. Congruent with
Broca's (1864/1960) earlier observations, Jackson (1874/1932) argued that
JEFFREY W. GRAY Neuropsychology Laboratory, Ball State University, Muncie, Indi-

ana. RAYMOND S. DEAN Neuropsychology Laboratory, Ball State University, Mun-
cie, Indiana, and University of Indiana School of Medicine.
205
206 JEFFREY W. GRAY and RAYMOND S. DEAN
language/verbal processing was served by the left cerebral hemisphere.

Further support for this notion came with investigations of patients who
had undergone surgery that partially severed the major nerve connections
between the hemispheres (corpus callosum; Gazzaniga, 1970; Gazzaniga &
Sperry, 1962; Sperry, Gazzaniga, & Bogen, 1969). Specifically, results of
such investigations suggested that in most right-handed and many left-
handed individuals, linguistic functions are served primarily by the left
hemisphere of the brain (Gazzaniga, 1970), whereas visual-spatial abili-
ties are mediated primarily by the right hemisphere (Sperry, et a1., 1969).
Investigations with normal subjects confirmed that although there is
considerable interhemispheric communication, the two cerebral hemi-
spheres of the brain serve specialized functions (e.g., Milner, 1971;
Witelson, 1976). From this point of view, the left hemisphere in most
individuals is seen to be better prepared to process information in a more
analytical, temporal, logical, or sequential fashion than is necessary for
the production and decoding of language (Gruber & Segalowitz, 1977).
Conversely, the right hemisphere would seem more able to process infor-
mation in a holistic, simultaneous, or concrete fashion, and thus it is
better able to deal with information of a spatial, manipulative nature (e.g.,
Bogen & Gazzaniga, 1965).
In this chapter, important aspects of laterality assessment with chil-
dren will be discussed. A number of methods of laterality assessment will
be reviewed with a focus on their clinical utility. In addition, clinical
guidelines for assessing cerebral laterality in children will be examined,
with an emphasis on both administration and interpretation. Finally, a
review and integration of the literature as it applies to the clinical signifi-
cance of inconsistent hemispheric lateralization for language disorders
will be presented.
METHODS OF LATERALITY ASSESSMENT
Clearly, the most predominant difficulty in assessing cerebral later-

ality is the inaccessibility of the brain. For this reason, the majority of our
information regarding hemispheric specialization is based on inferential
methods. The measures utilized have ranged from intracarotid sodium
amytal injections (Milner, Branch, & Rasmussen, 1964) to an assessment
of children's right or left turning hair whorl (Tjossem, Hansen, & Ripley,
1961). A number of behavioral methods are currently used to assess cere-
brallaterality. The most commonly used and clinically useful methods of
laterality assessment include perceptual asymmetry techniques, elec-
trophysiological procedures, self-report methods, and measure of uni-
manual performance on various types of motor tasks.
LATERALITY ASSESSMENT 207
Perceptual Asymmetry Techniques

As is the case with most measures of cerebral laterality, perceptual
asymmetry techniques infer a cerebral difference in organization based on
a lateral difference in performance. The two most commonly used percep-
tual asymmetry procedures include dichotic listening and split visual
half-field techniques.
Dichotic Listening
Considered to be one of the most widely used methods of investigat-
ing cerebral laterality, the dichotic listening technique was originally in-
troduced by Broadbent in 1954 and subsequently refined by Kimura
(1961a, 1961b, 1963). This technique involves presenting two different
auditory stimuli to each ear simultaneously. Because contralateral audito-
ry pathways have a greater number of ear-to-hemisphere nerve fibers than
do the ipsilateral connections (Majkowski, Bochenck, Bochenck, Knapik-
Fijalkowska, & Kopec, 1971), and the contralateral input "blocks" or oc-
cludes information passing along the ipsilateral path (Gruber & Segalo-
witz, 1977), the dominant hemisphere for a particular stimulus should
produce more correct recall for input to the contralateral ear. Thus, infor-
mation from the right ear would reach the left hemisphere more quickly
than it would the right hemisphere, whereas the opposite would be true
for a left ear input. This being the case, if a subject is consistently faster
and more accurate in identifying information presented to a particular
ear, it is concluded that the cerebral hemisphere opposite the more profi-
cient ear has a specialized function with regard to that type of stimulus.
The dichotic listening task has proven to be a reliable clinical pro-
cedure in assessing cerebral laterality in children (e.g., Hynd & Obrzut,
1977). Easily administered, this method simply involves presenting the
child with pairs of auditory stimuli (e.g., digits, nonsense words, sen-
tences, consonant-vowel-consonant syllables) to each ear simultaneously,
and asking the subject to recall or recognize the stimulus perceived. Stim-
uli are presented by means of a dual-channel tape recorder equipped with
stereophonic headphones. Thus, the dichotic listening task may be ad-
ministered in most clinical settings. However, portability may be prob-
lematic.
As may be expected, early research in the area of dichotic listening
was conducted with brain-damaged subjects (Kimura, 1961a). Along these
lines, Kimura (1961a) found that patients who had undergone temporal
lobe excision showed a depressed performance on the ear contralateral to
the damaged area. More important, she also found that patients with
documented left hemispheric speech lateralization (from sodium amytal
tests) showed a right ear advantage (REA). However, those subjects with
right hemisphere specialization for language were found to have a left ear
superiority, regardless of their hand preference. Similarily, Kimura
(1961b) examined the dichotic listening performances of normal right-
handed subjects and again found a significant right ear superiority. On the
basis of these findings, she concluded that the ear contralateral to the
hemisphere specialized for language was more efficient at processing ver-
bal stimuli.
In support of Kimura's (1961a, 1961b) earlier findings, more recent
evidence from dichotic listening studies has indicated that normal adults
(e.g., Dean & Hua, 1982) and children (e.g., Hynd & Obrzut, 1977) show an
REA for verbal material. Among the most robust and replicated findings
in contemporary experimental psychology, these data are often used to
support the inference of left hemisphere specialization for language. In-
deed, numerous authors have argued that the dichotic listening paradigm
is the most valid noninvasive method of inferring hemisperic lateraliza-
tion for language (Witelson, 1977). At the same time, however, this pro-
cedure is not without methodological problems (Satz, 1976). Factors such
as attentional bias, perceptual set, and memory (Bryden, 1982), acoustic
variables (i.e., signal intensity, signal-to-noise ratio; Berlin & Cullen,
1977), and response distortions (e.g., Birkett, 1977) all have been impli-
cated as contributing to major methodological deficiencies.
In response to such methodological problems, a number of pro-
cedural improvements have been suggested (e.g., Studdert-Kennedy &
Shankweiler, 1970). One such alteration involves the presentation of sin-
gle pairs of consonant-vowel-consonant (CVC) nonsense syllables rather
than whole stimulus lists. The use of single pairs is seen as "reducing the
memory load on the subject ... and eliminating many of the response
organization factors that complicate the interpretation of list data"
(Bryden, 1982, pp. 46-47). In a different approach, Geffen, Traub, and
Stierman (1978) introduced large word-pair lists in which a target word
was randomly presented to one ear or the other. The subject was in-
structed to respond via a button-press upon perception of the predeter-
mined target word. Under these conditions, the collection of both ac-
curacy and latency data was possible. Importantly, normal subjects have
been shown to have a right ear advantage with regard to both accuracy and
speed of responding.
Perhaps the most important methodological refinement, the atten-
tional monitoring procedure differs from those previously discussed in
that subjects are required to attend and respond to dichotic stimuli only in
one prespecified ear. This directed dichotic task eliminates the possibility
of order effects because stimulus items are presented in counterbalanced
fashion. It has been argued that when used in conjunction with single-pair
stimulus material, this procedure may overcome the majority of the most
frequently cited methodological problems (Bryden, 1982). Thus, this may
be the most clinically useful dichotic listening procedure. More recently,
high-technological tapes also have been developed to control for a num-

ber of these acoustic variables (e.g., Auditech).
Visual Half-Field Procedures

Similar to the dichotic listening paradigm in terms of its neurological
assumptions, the visual half-field (VHF) procedure involves either bilat-
erally or unilaterally presenting stimuli to the visual half-fields. Because
of the physiological structure of the human visual system, stimuli pre-
sented to one VHF are received first by the contralateral cerebral hemi-
sphere. Thus, stimuli presented to the right VHF are transmitted directly
to the left visual cortex, whereas material presented to the left VHF is
initially transmitted to the right visual cortex.
Although analogous to the dichotic listening procedure, the VHF
technique has not proven to be as clinically useful. Indeed, a number of
methodological difficulties make the interpretation of data obtained from
such procedures somewhat ambiguous. Importantly, the possibility of eye
movement may jeopardize the validity of this paradigm. In this regard, a
number of modifications have been suggested. One early approach in-
volved the fixation of the stimulus on the retina, thus assuming a valid
VHF projection. To this end, a number of investigators have used contact
lenses (Pritchard, Heron, & Hebb, 1960) or prolonged afterimage para-
digms (MacKinnon, Forde, & Piggens, 1969).
In a more common approach, stimuli are presented tachistoscopically
to both VHFs simultaneously. It has been fairly well established that very
brief exposure durations (i.e., less than 200 msec) reduce the possibility of
eye movement during presentations (e.g., Marcel & Rajan, 1975). Howev-
er, it is important to note that the validity of this procedure is dependent
upon the subject's cooperation in initially fixating on a central point (see
Kimura & Durnford, 1974). Methodological deficiencies notwithstanding,
a left VHF (right hemisphere) advantage has typically been found for
nonverbal visual-spatial tasks whereas a right VHF (left hemisphere) su-
periority has been shown for more verbally oriented stimuli (e.g., Witel-
son, 1977).
Following the lead of Heron (1957), much of the early research in-
volving this paradigm used a unilateral presentation of material (see
Kershner, 1977). However, more recent investigations have emphasized
the methodological problems of this approach and stressed a bilateral
method of stimulus presentation (see Dean, 1981). Although this method
partially controls for fIxation effects, the possibility of directional scan-
ning of visually encoded material cannot be ruled out. Consequently, it is
possible that verbal stimuli such as words or letters may evoke postex-
posure attentional scanning (Kershner, 1977). For this reason, clinical
inferences regarding hemisphere specialization made on the basis of a
subject's VHF performance may be tentative at best.
Electrophysiological Measures
As a result of the inherent methodological difficulties in perceptual
asymmetry paradigms, a number of researchers and clinicians have opted
to use more direct measures of hemispheric lateralization. For various
reasons, most have relied upon measures of the electrical activity in the
brain. Two of the most common indices include the ongoing electroen-
cephalogram (EEG) and the average evoked potential (AEP). Both pro-
cedures involve the examination of electrical activity in a general area of
the cerebral cortex as measured from EEG leads. An alpha wave pattern
(8-13 Hz) has most often been associated with a relaxed, resting state and
is seen as reflecting a decrease in electrical activity in a particular area of
the brain. Beta waves (greater than 13 Hz), on the otcer hand, are found
with an alert or aroused state (e.g., attending to a stimulus). Given these
assumptions, one would expect differential alpha activity in the two cere-
bral hemispheres depending upon the task at hand.
Whereas early research involving these procedures focused on lateral
differences in electrical activity when the subject was at rest, recent em-
phasis has been upon the examination of brain wave activity of subjects
performing various tasks (Bryden, 1982). These investigations have
shown less activity (Le., a predominance of alpha) in postcentral areas of
the right hemisphere when normal right-handed subjects are involved in
verbal analytic tasks (Doyle, Ornstein, & Galin, 1974; Galin & Ornstein,
1972; Morgan, McDonald, & McDonald, 1971). In contrast, a predomi-
nance of alpha waves has been found in the left hemisphere of normal
individuals when they are required to perform more spatially oriented
tasks (Doyle et a1., 1974; Galin & Ornstein, 1972; Morgan et a1., 1971). The
combined results of these investigations support the existence of a verbal-
analytic versus spatial-holistic processing difference between the two
hemispheres. However, in an investigation that examined subjects' block
design performance, Galin, Johnstone, and Herron (1978) found increased
beta activity in the left hemisphere concomitant with increased task com-
plexity. Interestingly, it appeared that as this predominantly right hemi-
spheric task became more difficult, the analytical, sequential mode of
processing most often attributed to the left hemisphere was activated. It is
important to note that this also may apply to dichotic listening and visual
half-field paradigms. Thus, the clinician must pay close attention to the
particular stimuli and task demands used with such techniques, and
should interpret the data with caution.
Consistent with EEG findings, the majority of investigations using the
evoked potential paradigm have shown a significantly larger AEP in the
left hemisphere for verbal stimuli and an increased right hemisphere AEP
for nonverbal musical stimuli (e.g., Shucard, Shucard, & Thomas, 1977).
Of particular interest, Molfese and his colleagues (Molfese & Hess, 1978;
Mo1ese & Molfese, 1979a, 1979b) have shown a similar hemispheric

asymmetry in preschool children and newborns.
Although these seemingly more direct physiological measures of lat-
erality have appeal, they are not without methodological difficulties. In-
deed, such extraneous variables as skull thickness and electrode place-
ment may affect electrophysiological asymmetry (Galin & Ornstein, 1972)
and, consequently, reliability. Thus, increased alpha activity may be asso-
ciated as much with procedural components as it is with hemispheric
lateralization of functions. In this regard, a number of investigations have
suggested that the use of EEG procedures in conjunction with behavioral
techniques of laterality assessment may offset many of these meth-
odological problems (e.g., Davidson & Schwartz, 1977). At the present
time, however, little data exist that support the valid use of EEG measures
of laterality on a large-scale clinical basis. Moreover, problems with avail-
ability and portability make the routine clinical use of these highly tech-
nical procedures impractical.
Lateral Preference Measures

Because the performance on unimanual motor activities has been
shown to be controlled primarily by the contralateral cerebral hemi-
sphere, it has been suggested that the consistency in lateral preference
may be a behavioral expression of hemispheric dominance. Probably re-
lated in part to its deceptive ease in assessment, hand preference may be
the most clinically utilized index of cerebral laterality (Dean, 1981, 1984).
Indeed, most clinical examinations and neuropsychological batteries in-
corporate some method of assessing handedness. Related to this clinical
preoccupation with handedness is the long-held notion that mixed-hand
preference may be a predictor of mixed or confused hemispheric later-
alization of language (e.g., Orton, 1937).
Estimates of handedness based on large samples suggest that about
90% of normally functioning individuals are right-hand-dominant for
unimanual activities. The remaining 10% consist of individuals who are
considered as either consistently left-handed or mixed-dominant (Old-
field, 1971). However, a number of investigators have argued that more
recent liberal societal views on left-handedness may well be associated
with an increase in the prevalence of left-handedness (Bryden, 1982).
Hand preference is most often assessed by means of a structured se1-
report format, and a number of inventories for assessing handedness are
available (Annett, 1970; Crovitz & Zener, 1962; Oldfield, 1971; Reitan,
1969). Generally speaking, these query the subject's hand preference for a
number of everyday activities. Moreover, the subject often is asked to
perform each task physically.
Perhaps the most commonly used measure, the Edinburgh Handed-

ness Inventory (EHI) was developed by Oldfield (1971). In the initial
phase of construction, 20 items that asked about the subject's hand use for
simple unimanual activities were selected from an existing measure of
handedness (Humphrey, 1951). From the original pool of 20 items, 10
were selected on the basis of their ability to discriminate between left- and
right-handed subjects. In its present form, subjects are asked to respond
by indicating their left-right preference for 10 commonly performed
tasks.
A "laterality quotient" (LQ) is then calculated by summing across
each hand, subtracting the total of the left hand from that of the right
hand, dividing this difference by the total of the two hands, and multiply-
ing by 100 (see Oldfield, 1971). Because the EHI is so simple to administer
and score, it is not surprising to find that it is the most commonly used
group measure of hand preference.
Although handedness inventories like the EHI appear to have some
clinical utility, they are not without their methodological difficulties.
Generally speaking, these inventories have failed to demonstrate clinical-
ly acceptable estimates of item stability and concurrent validity. Using
items from the three most commonly used inventories (Le., Annett, 1970;
Crovitz & Zener, 1962; Oldfield, 1971), Raczkowski, Kalat, and Nebes
(1974) found that only 74% of their subjects responded reliably between
two administrations. Moreover, only 78% of the subjects' actual perfor-
mances were consistent with their initial self-report of hand preference
for the various tasks. Somewhat surprisingly, only 1 of the 23 items pro-
duced perfect correlations between subjects' initial response and both the
actual performance and second response to the item. It may be that a
number of the items on these inventories represent activities that are not
performed routinely, making it difficult for the subjects to anticipate their
responses (Raczkowski et aI., 1974).
More important to the focus of this chapter, it has become obvious
that simple handedness does not relate directly to the lateralization of
higher mental functions (Dean, 1982; Kinsbourne & Hiscock, 1977). In-
deed, a review of the recent literature leads to the conclusion that the
relationship between handedness and hemispheric specialization is
weak. In fact, the clinician is more likely to be correct if he or she simply
assumes that language is served by the left hemisphere in all subjects
regardless of their hand preference (Dean, 1984). In support of this, Lake
and Bryden (1976) showed that a large proportion of both right- and left-
handed individuals are left-hemisphere-dominant for language. More spe-
cifically, Milner (1974) reported that in as many as 95% of all right-
handed, and 70% of left-handed individuals, language is served by the left
cerebral hemisphere. In contrast to early notions of a right hemispheric
language dominance for all left-handed individuals, Milner (1974)
showed that only 30% of all left-handed individuals have right hemi-
sphere or inconsistent hemispheric specialization for language.
Consistent with Milner's (1974) findings, Satz (1972, 1976) has sug-
gested that differences between left-handed persons in language later-
alization may be attributed to the differences in the etiology of the left-
hand preference. He has argued in favor of both a genetically or environ-
mentally predisposed left-handedness and a form of left-handedness that
arises out of early brain damage and/or a developmental anomaly in the
left hemisphere. In light of more recent findings (see Satz, 1976), the early
hypotheses (e.g., Orton, 1937) that offered handedness as a definitive
indicator of language lateralization now seem rather simplistic.
With this in mind, it is not surprising to find that investigations
examining the relationship between left/mixed handedness and chil-
dren's learning problems have produced inconsistent results. Indeed,
such findings have led a number of authors to conclude that the rela-
tionship between hand preference and language or scholastic achieve-
ment is minimal at best (Beaumont & Rugg, 1978; Benton, 1975; Hardyck,
Petrinovich, & Goldman, 1977). Supporting this argument, Dunlop, Dun-
lop, and Fenelon (1973) found that simple hand preference accounted for
little of the total variance associated with language lateralization as mea-
sured by dichotic listening. In contrast, these authors showed a clear
relationship between eye/hand use and chronic learning problems.
Although measures of lateral preference (e.g., handedness) have been
used most often to classify subjects in a nominal fashion (left, right,
mixed), it has been argued that lateral preference data may be more
clinically useful if they are treated in a continuous manner (see Dean,
1978,1981,1982,1984; Shankweiler & Studdert-Kennedy, 1975). Similar-
ly, it seems reasonable that whereas simple handedness may not reliably
reflect cerebral organization, coherently lateralized cerebral systems (e.g.,
visually guided fine motor performance) may offer more clinical utility in
the examination of cerebrallateralization (Dean, 1982). Indeed, a number
of investigations have shown preference for visually guided motor tasks
(e.g., picking up a penny) to be most sensitive to other measures of neuro-
psychological impairment (Dean, Schwartz, & Smith, 1981; Gray, Dean, &
Seretny, in press). This point of view tends to support Luria's (1966)
position that lateral dominance of the hemispheres varies as a function of
the cerebral system under investigation. Thus, lateral preference patterns
differ not only between individuals but within individuals as a function
of the task variable under investigation.
Dean (1978) developed a multifactorial measure of lateral preference
that treats lateral preference in a continuous fashion (Laterality Preference
Schedule; LPS). This 49-item self-report measure requires respondents to
indicate their lateral preference for unimanual tasks on a 5-point Likert
scale. All items produced both stability and predictive validity estimates
of at least 0.80. An individual's score for each item is computed using

weights for each response category (Le., right always = 1, right mostly =
2, right and left equally often = 3, left mostly = 4, left always = 5). Scores
for each of the six-factor analytically derived subscales are obtained by
multiplying the total number of responses for a given category by the
appropriate weight and then summing across the categories (Dean, 1982).
The six factors were identified as general laterality, visually guided motor
activities, eye preference, ear use, strength, and foot preference. Prefer-
ence scores for the general index (across all six subs cales) range from 49
(right always on each item) to 245 (left always on each item).
In contrast to measures of simple hand preference, the summary
scores on the LPS allow the placement of a subject on a left-right con-
tinuum. From a practical point of view, the LPS can be administered on
either a group or an individual basis. Moreover, there are reasonably good
normative data on children available to compare an individual's scores.
Although the LPS may be limited by its inventory format, it seems to
offer a cost-effective and psychometrically sound method of assessing
laterality. The LPS subscales have been shown to have a differential rela-
tionship with a number of cognitive (Dean et a1., 1981) and personality
(Dean, Schwartz, & Hua, in press) variables, and to correlate significantly
with dichotic listening results (Dean & Hua, 1982). Nonetheless, clini-
cians should interpret the results for individual patients with caution.
The validity and reliability of the LPS are jeopardized when it is used
with severely reading-disabled or very young children. However, it is
important to note that if the instructions are read to the child, these
potential problems may be at least partially remedied. Even so, the LPS is
perhaps not appropriate for clinical use with children under 9 years of
age.
Unimanual Performance Measures

A number of authors have argued that unimanual performance on
psychomotor/motor tasks may provide the most accurate assessment of
laterality (Reynolds, Hartlage, & Haak, 1981). Such procedures simply
involve the observation and comparison of the actual performance of the
two sides of the body on various tasks. Performance measures have ranged
from ratchet-twirling speed to dart-throwing accuracy (Provins & Cunliffe,
1972). However, the most frequently utilized performance measures in-
clude tasks of relative hand strength (e.g., Woo & Pearson, 1927) and
motor speed (Annett, 1970, 1976; Peters & Durding, 1978). For the clinical
setting, it is often recommended that multiple performance measures be
used (e.g., motor speed and manual dexterity).
Many test batteries of neuropsychological functioning include tasks
that lend themselves to comparison of right side versus left side perfor-
mance. For example, the Halstead-Reitan Neuropsychological Battery

(Reitan, 1969) contains several tasks that allow comparisons between per-
formance on the left and right sides of the body. The Finger Oscillation
Test is a measure of overall motor speed and manual dexterity. As such, it
lends itself to a comparison of the tapping speed of the dominant and
nondominant hands, thus allowing the assessment of lateral differences.
The dynamometer, a measure of grip strength, also allows comparison of
the preferred and nonpreferred hands.
Although performance measures appear to be clinically useful, they
present a number of methodological problems. As was the case with mea-
sures of simple handedness, many performance measures may lack the
reliability that is necessary for clinical use. In an attempt to delineate this
issue, Sappington (1980) examined the test-retest stability of three popu-
lar performance measures (i.e., dowel balancing, peg placement, and grip
strength). He found stability estimates ranging from only 0.29 to 0.62.
Although two of the three coefficients were statistically significant, they
were too low to be of practical value. In a related investigation, Provins
and Cunliffe (1972) found stability estimates ranging from 0.15 to 0.94.
However, only handwriting speed and manual dexterity produced signifi-
cantly consistent performances. An examination of the literature in this
area leads to the conclusion that measures of manual dexterity (e.g., finger
tapping) may hold the most clinical utility (Peters & Durding, 1978;
Provins & Cunliffe, 1972). Moreover, a number of investigators have rec-
ommended the combined use of performance measures and other mea-
sures of lateral preference (Bryden, 1982; Dean, 1984). The rationale for
this is based on data showing low concordance (i.e., < 0.50) between
laterality measures when different methodologies are utilized. Such re-
sults question the view of laterality as a unitary factor and suggest that
lateralization may be viewed better as system-specific (Dean, 1982). Thus,
the use of multiple measures of laterality permits more specific state-
ments to be made with respect to the individual's particular pattern of
lateral dominance.
APPLICATIONS WITH LEARNING-DISABLED/LANGUAGE-

IMP AIRED CHILDREN
The assessment of cerebral laterality is an integral component of most

neuropsychological evaluations. Although measures of laterality may
serve a number of purposes, they are utilized most often to infer the
lateralization of language. Because unimanual activities are served pri-
marily by the contralateral cerebral hemisphere, inconsistent perfor-
mance in such activities historically has been seen to reflect confused
(incomplete) lateralization of cortical functions (Annett, 1976; Orton,
1937; Zangwill, 1962). Moreover, a number of investigators have offered

data suggesting a strong relationship between language disorders and bi-
lateral representation of language functions in the brain (e.g., Zangwill,
1960, 1962). Thus, the clinical preoccupation with lateralization of lan-
guage has a rich historical base.
Indeed, as early as the 19th century researchers were interested in the
relationship between children's learning problems and neurological ab-
normalities Oackson, 1874/1932). Most notably. Orton (1937) attributed a
number of language difficulties in children to a failure to establish cere-
bral dominance. On the basis of his clinical observation of a higher inci-
dence of dyslexia in children with inconsistent handedness, Orton (1937)
hypothesized a similar confusion at the cerebral level. He argued that the
failure of one of the cerebral hemispheres to achieve functional superi-
ority may well be related to these children's inability to process informa-
tion adequately. Although Orton's early notions may have been somewhat
simplistic (Critchley, 1970), the relationship between inconsistent later-
alization and learning disorders in children remains one of the most re-
searched and controversial issues in education and neuropsychology.
Moreover, because evidence has shown that the failure to attain a clear
hemispheric dominance may be related to numerous maladjusted behav-
iors (Dean et a1., 1981; Walker & Birch. 1970; Zangwill, 1962), most child
neuropsychologists include some form of laterality measure in their as-
sessment battery.
In light of Orton's (1937) original hypothesis of inconsistent cerebral
lateralization in children with learning problems, it would be expected
that children with reading and language difficulties would fail to show
the consistent REA for speech stimuli found in normals. Contrary to this
notion, the majority of studies in this area actually have found REA for
dyslexic (Sparrow & Satz, 1970; Witelson, 1977) and learning-disabled
children (Hynd, Obrzut, Weed, & Hynd, 1979). Taken together with pre-
vious findings with normals, such results suggest that the left cerebral
hemisphere is specialized for linguistic functions in both learning-disor-
dered and normally functioning children.
Although a number of investigations have failed to show REA for
speech in children with learning disorders (Le., incomplete left hemi-
sphere dominance), it is important to note that the vast majority of these
results may be attributed to factors other than inconsistent cerebral later-
ality (Summers & Taylor, 1972; Witelson & Rabinovitch, 1972; Zurif &
Carson, 1970). For instance, Zurif and Carson (1970) interpreted their
findings as suggesting inconsistent cerebral laterality for poor readers.
Clearly, this conclusion was questionable, given that normal subjects in
this study also failed to show an REA. In the same manner, Witelson and
Rabinovitch (1972) speculated that their results were indicative of in-
complete cerebral dominance in a group of reading-disabled children.
However, like Zurif and Carson (1970), they failed to find an REA for good
readers.
As a result of the popular conceptualization that linguistically dis-
abled children may suffer delayed cerebrallateralization (e.g., Critchley,
1970), numerous investigations have attempted to examine the develop-
mental implications of ear asymmetry for a number of clinical popula-
tions when compared with normal controls (Bakker, 1973; Hynd et al.,
1979; Satz, 1976; Satz, Rardin, & Ross, 1971; Sparrow & Satz, 1970). Over-
all, few studies have reported the expected ear-by-age interaction for
groups. Thus, it appears that there is little evidence to support the devel-
opmentallag hypothesis of cerebrallateralization in children with learn-
ing disorders.
In contrast, the majority of investigations have shown a less pro-
nounced right VHF dominance for linguistic material in language-dis-
abled children when compared with normal controls (Kershner, 1977;
Marcel, Katz, & Smith, 1974; Marcel & Rajan, 1975; McKeever & Van
Deventer, 1975). Using both a bilateral and a unilateral presentation,
McKeever and Van Deventer (1975) compared the VHF performance of
dyslexic and normal controls. As a control for eye movements, subjects in
this investigation were required to focus on and subsequently identify a
"fixation number" prior to their reporting any experimental stimuli. Al-
though both groups showed an overall right visual half-field advantage
under the unilateral presentation, dyslexic subjects performed much
more poorly on the word recognition task than did the normal controls. Of
particular interest, dyslexic subjects differed more from their normal co-
horts on RVF than on LVF presentations. Under the bilateral presentation,
however, dyslexic children showed a normal right VHF advantage. Al-
though McKeever and Van Deventer acknowledged the inconsistency of
these findings, they interpreted the results as indicating an impaired "left
hemisphere functioning" for dyslexic subjects in this study. Clearly,
whereas this interpretation was consistent with the unilateral presenta-
tion, it was not supported by the results under the bilateral condition.
These findings suggest that the use of bilateral versus unilateral presenta-
tion can make a difference in measurement.
In a second experiment, these investigators presented subjects with
verbal stimuli via a visual half-field procedure as well as a dichotic listen-
ing paradigm. Interestingly, their dyslexic subjects showed a normal left
hemispheric superiority for the auditory presentation (REA) but a reduc-
tion in asymmetry for stimuli presented visually. Moreover, there was
little correspondence between the ear advantage and VHF performance of
the dyslexic subjects. In a related investigation, Tomlinson-Keasey and
Kelly (1979) found a significant relationship between right hemispheric
specialization for words and linguistic disabilities in third-grade chil-
dren. Conversely, as a left VHF advantage (right hemisphere) for pictures
increased, a concomitant increase in mathematical skill occurred. Thus,

in general, VHF research indicates less of a left hemispheric specializa-
tion for children with learning problems when linguistic material is pre-
sented visually.
Using a multifactor measure of lateral preference, Dean et a1. (1981)
found a clear association between inconsistent lateralization of visually
guided motor performance and learning disabilities. In contrast, however,
they found little difference between learning-disabled children and nor-
mal controls when lateral preference patterns were summed across fac-
tors. Thus, as previously mentioned, it is important to consider not only
individual differences but also the individual tasks utilized to infer cere-
brallaterality. Dean (1979) also showed that children with stronger verbal
than nonverbal abilities presented a more anomalous pattern in their cere-
bral lateralization than children with either similar verbal-nonverbal
abilities or verbal skills less than nonverbal functioning. Taken together,
these findings provided partial support for Hines and Satz's (1974) argu-
ment suggesting a functional disassociation in visual-verbal systems that
covaries with the degree of lateral symmetry.
CLINICAL IMPLICATIONS
Although inconsistent cerebral lateralization of language may be re-

lated to some learning disorders, the relationship is not as simple as was
once thought (Dean, 1981). Indeed, it appears that although evidence of
normal language lateralization exists when measured in auditory fashion
(Le., dichotic listening), many language-disabled children show less clear
language lateralization when assessed visually (Le., VHF). Beaumont and
Rugg (1978) have attempted to reconcile these seemingly contradictory
findings. Expanding upon Pizzamiglio's (1976) and Geschwind's (1974)
hypotheses, Beaumont and Rugg have argued that learning problems asso-
ciated with confused lateralization have a basis in a functional disassocia-
tion in visual and verbal systems that covaries with the degree of lateral
symmetry. Thus, for learning-disabled children, the ability to move com-
fortably between verbal and nonverbal strategies may be impaired. In this
regard, Dean (1979), Levy (1969), and Hines and Satz (1977) have shown a
difficulty in the integration of visual and verbal strategies that varies with
the individual lateral preference patterns.
The assessment of cerebral laterality will continue to be an important
component of neuropsychological assessment. However, clinicians need
to be aware of the many practical assessment and complex theoretical
issues when using these data in clinical settings. As discussed in this
chapter, a number of methods of laterality assessment exist. One of the
most well known, the dichotic listening paradigm, appears to have con-
siderable clinical appeal. However, this procedure may not offer the por-
tability necessary for some clinical settings. Similarly, the visual half-
field procedure is both bulky and costly. Although seen as a more direct
measure of laterality, EEG techniques lack the data necessary for large-
scale clinical use. Measures of handedness may be the most widely used
index of laterality; however, in light of the interaction between hand
preference and both social learning and environmental effects, the rela-
tionship between simple handedness and cerebrallateralization is equiv-
ocal. Problems with inventory techniques notwithstanding, multifactorial
inventories of laterality may offer the single most economical and psycho-
metrically sound method of evaluating lateral preference patterns in most
clinical settings. However, as noted before, these may not be appropriate
for use with reading-disabled or younger children. Regardless of the pro-
cedure used, it is important to interpret the data obtained from such
measures in light of other clinical and psychometric findings. Perhaps the
most reliable method of laterality assessment involves the use of multiple
measures. With the agreement of multiple measures comes the clinical
assurance that generalizations concerning the patient's functional later-
alization are warranted. Conversely, important implications may be
drawn on the basis of the pattern of findings and how they vary according
to the procedure, mode of input, and type of task used.
REFERENCES
Annett, M. (1970). A classification of hand preference by association analysis. British Jour-

nal of Psychology, 61, 303-321.
Annett, M. (1976). A coordination of hand preference and skill replicated. British Journal of
Bakker, D. J. (1973). Hemispheric specialization and stages in the learning-to-read process.
Bulletin of the Orton Society, 23, 15-27.
Beaumont, J. G., & Rugg, M. D. (1978). Neuropsychological laterality of function and dyslex-
ia: A new hypothesis. Dyslexia Review, 1, 18-21.
Benton, A. 1. (1975). DeY'llopmental dyslexia: Neurological aspects. In W. J. Friedlander
(Ed.), Advances in neurology (Vol. 7, pp. 195-230). New York: Raven Press.
Berlin, C. I., & Cullen, J. K. (1977). Acoustic problems in dichotic listening tasks. In S. J.
Segalowitz & F. A. Gruber (Eds.), Language development and neurological theory (pp.
75-88). New York: Academic Press.
Birkett, P. (1977). Measures of laterality and theories of hemispheric process. Neuropsycho-
logia, 15, 693-696.
Bogen, J. E., & Gazzaniga, M. S. (1965). Cerebral commissurotoma in man: Minor hemisphere
dominance for certain visuo-spatial functions. Journal of Neurosurgery, 23, 394-399.
Broadbent, P. E. (1954). The role of auditory localization in attention and memory span.
Journal of Experimental Psychology, 47, 191-196.
Broca, P. (1960). Remarks on the seat of the faculty of articulate language, followed by an
observation of aphemia. In G. Von Bonin (Ed). Some papers on the cerebral cortex.
Springfield, IL: Thomas. (Original work published 1865)
Bryden, M. P. (1982). Laterality: Functional asymmetry in the intact brain. New York:
Academic Press.
Critchley, M. (1970). The dyslexic child (2nd ed.). London: Heinemann.
Crovitz, H. F., & Zener, K. (1962). A group-test for assessing hand and eye dominance.
American Journal of Psychology, 75, 271-276.
Davidson, R. J., & Schwartz, G. (1977). The influence of musical training on patterns of EEG
asymmetry during musical and non-musical self-generalization tasks. Psycho-
physiology, 14, 58-63.
Dean, R. S. (1978). Cerebral laterality and reading comprehension. Neuropsychologia, 16,
633-636.
Dean, R. S. (1979). Cerebral laterality and verbal-performance discrepancies in intelligence.
Journal of School Psychology, 17, 145-150.
Dean, R. S. (1981). Cerebral dominance and childhood learning disorders: Theoretical per-
spectives. School Psychology Review, 10, 373-378.
Dean, R. S. (1982). Assessing patterns of lateral preference. Journal of Clinical Neuropsy-
chology, 4, 124-128.
Dean, R. S. (1984). Functionallateralization of the brain. Journal of Special Education, 18,
240-256.
Dean, R. S., & Hua, M. S. (1982). Laterality effects in cued auditory asymmetries. Neuropsy-
chologia, 20. 685-690.
Dean, R. S., Schwartz, N. H., & Hua, M. (in press). Lateral preference patterns in schizo-
phrenia and affective disorders, Journal of Abnormal Psychology.
Dean, R. S., Schwartz, N. H., & Smith, 1. S. (1981). Lateral reference patterns as a discrimi-
nator of learning disabilities. Journal of Consulting and Clinical Psychology, 49, 227-
235.
Doyle, J. C., Ornstein, R., & Galin, D. (1974). Lateral specialization of cognitive mode: II. EEG
frequency analysis. Psychophysiology, 11, 567-578.
Dunlop, D. B., Dunlop, P., & Fenelon, B. (1973). Vision laterality analysis in children with
reading disability: The results of new techniques of examination. Cortex, 9, 227-236.
Galin, D., Johnstone, J., & Herron, J. (1978). Effects of task difficulty on EEG measures of
cerebral engagement. Neuropsychologia, 16, 461-472.
Galin, D., & Ornstein, R. (1972). Lateral specialization of cognitive mode: An EEG study.
Psychophysiology, 9, 412-418.
Gazzaniga, M. S. (1970). The bisected brain. New York: Appleton-Century-Crofts.
Geffen, G., Traub, E., & Stierman, I. (1978). Language laterality assessed by unilateral ECT
and dichotic monitoring. Journal of Neurology, Neurosurgery, and Psychiatry, 41, 354-
360.
Geschwind, N. (1974). Selected papers on language and the brain. The Netherlands: D.
Reidal.
Gray, J. W., Dean, R. S., & Seretny M. 1. (1986). Lateral preference as a predictor of cognitive
rigidity, Journal of Clinical Psychology, 42, 956-960.
Gruber, F. A., & Segalowitz, S. J. (1977). Some issues and methods in the neuropsychology of
language. In S. J. Segalowitz & F. A. Gruber (Eds.)' Language development and neu-
rological theory (pp. 3-19). New York: Academic Press.
Hardyck, C., Petrinovich, 1. F., & Goldman, R. D. (1977). Left-handedness and cognitive
deficit. Cortex, 17, 266-279.
Heron, W. (1957). Perception as a function of retinal locus and attention. American Journal
of Psychology, 70, 38-48.
Hines, D., & Satz, P. (1977). Cross-model asymmetries in perception related to asymmetry in
cerebral function. Neuropsychologia, 12, 239-247.
Humphrey, M. (1951). Handedness and cerebral dominance. Bachelor of Science thesis,
Oxford University.
Hynd, G. W., & Obrzut, J. E. (1977). Effects of grade level and sex on the magnitude of the
dichotic ear advantage. Neuropsychologia, 15, 689-692.
Hynd, G. W., Obrzut,]. E., Weed, W., & Hynd, C. R. (1979). Development of cerebral domi-
nance: Dichotic lstening asymmetry in normal and learning-disabled children. Journal
of Experimental Child Psychology, 2B, 445-454.
]ackson,]. H. (lB74). On the duality of the brain. Medical Press, 1, 19. Reprinted in]. Taylor
(Ed.), Selected writings of John Hughlings Jackson (Vol. 2). London: Hodder and
Stoughton, 1932.
Kershner, ]. B. (1977). Cerebral dominance in disabled readers, good readers, and gifted
children: Search for a valid model. Child Development, 4B, 61-67.
Kimura, D. (1961a). Cerebral dominance and the perception of verbal stimuli. Canadian
Journal of Psychology, 15, 166-171.
Kimura, D. (1961b). Some effects of temporal lobe damage on auditory perception. Canadian
Journal of Psychology, 15, 156-165.
Kimura, D. (1963). Right temporal lobe damage. Archives of Neurology, B, 264-271.
Kimura, D., & Durnford, M. (1974). Normal studies on the function of the right hemisphere in
vision. In S. ]. Dimond & ]. G. Beaumont (Eds.), Hemisphere functions in the human
brain (pp. 159-175). London: Elek Scientific Books.
Kinsbourne, M., & Hiscock, M. (1977). Does cerebral dominance develop? In S.]. Segalowitz
& F. A. Gruber (Eds.), Language development and neurological theory (pp. 171-191).
Lake, D., & Bryden, M. (1976). Handedness and sex differences in hemispheric asymmetry.
Brain and Language. 3, 266-282.
Levy,]. (1969). Possible basis for the evaluation of lateral specialization of the human brain.
Nature, 224, 614-615.
Luria, A. R. (1966). Human brain and psychological processes. New York: Harper & Row.
MacKinnon, G. E., Forde, ]., & Piggens, D. ]. (1969). Stabilized images, steadily fixated
figures, and prolonged after images. Canadian Journal of Psychology, 23, 184-195.
Majkowski, ]., Bochenck, Z., Bochenck, W., Knapik-Fijalkowska, D., & Kopec, ]. (1971).
Latency of average evoked potentials to contralateral and i psalateral auditory stimula-
tion in normal subjects. Brain Research, 25, 416-419.
Marcel, T., Katz, 1., & Smith, M. (1974). Laterality and reading proficiency. Neuropsycho-
logia, 12, 131-139.
Marcel, T., & Rajan, P. (1975). Lateral specialization for recognition of words and faces in
good and poor readers. Neuropsychologia, 13,489-497.
McKeever, W F., & Van Deventer, A. D. (1975). Dyslexic adolescents: Evidence of impaired
visual and auditory language processing associated with normallateralization and visu-
al responsivity. Cortex, 11, 361-378.
Milner, B. (1971). Interhemispheric differences and psychological processes. British Medical
Bulletin, 27, 272-277.
Milner, B. (1974). Hemispheric specialization: Scope and limits. In F. O. Schmitt & F. G.
Warden (Eds.), The neurosciences: Third study programme (pp. 75-89). Cambridge,
MA: M.LT. Press.
Milner, B., Branch, c., & Rasmussen, T. (1964). Observations on cerebral dominance. In A. V.
S. DeRenck & M. O'Connor (Eds.), Disorders of language (CIBA Foundation Sym-
posium) (pp. 200-241). London: Churchill.
Molfese, D. 1., Freeman, R. B., & Palermo, D. (1975). The ontogeny of brain lateralization for
speech and nonspeech stimuli. Brain and Language, 2, 356-368.
Molfese, D. 1., & Hess, T. M. (1978). Hemispheric specialization for VOT perception in the
preschool child. Journal of Experimental Child Psychology, 26, 71-84.
Molfese, D. 1., & Molfese, V.]. (1979a). Hemisphere and stimulus differences as reflected in
the cortical responses of newborn infants to speech stimuli. Developmental Psychology,
15,505-511.
Molfese, D. 1., & Molfese, V.]. (1979b). VOT distinctions in infants: Learned or innate. In H.
A. Whitaker & H. Whitaker (Eds.), Studies in neurolinguistics (Vol. 4, pp. 21-35). New
Morgan, A., McDonald, P. J., & McDonald, H. (1971). Differences in bilateral alpha activity as
a function of experimental tasks, with a note on lateral eye movements and hypno-
tizability. Neuropsychologia, 9, 459-469.
Oldfield, R. C. (1971). The assessment and analysis of handedness: The Edinburgh invento-
ry. Neuropsychologia, 9, 94-113.
Orton, S. T. (1937). Specific reading disability-strephosymbolia. Journal of the American
Medical Association, 90, 1095-1099.
Peters, M., & Durding, B. (1978). Handedness as continuous variable. Canadian Journal of
Pizzamiglio, L. (1976). Cognitive approach to hemispheric dominance. In R. M. Knights & D.
Bakker (Eds.), The neuropsychology of learning disorders (pp. 265-272). Baltimore:
University Park Press.
Pritchard, R. M., Heron, W., & Hebb, D. O. (1960). Visual perception approached by the
method of stabilized images. Canadian Journal of Psychology, 14, 66-77.
Provins, K. A., & Cunliffe, P. (1972). Motor performance tests of handedness and motivation.
Raczkowski, D., Kalat, J. W., & Nebes, R. (1974). Reliability and validity of some handedness
questionnaire items. Neuropsychologia, 12, 43-47.
Reitan, R. M. (1969). Manual for administration of neuropsychological test batteries for
adults and children. Indianapolis: Author.
Reynolds, C. R., Hartlage, L. C., & Haak, R. (1981). The relationship between lateral prefer-
ence, as determined by neuropsychological test performance, and aptitude-achieve-
ment discrepancies. Clinical Neuropsychology, 3, 19-21.
Sappington, J. T. (1980). Measures of lateral dominance: Interrelationships and temporal
stability. Perceptual and Motor Skills, 50, 783-790.
Satz, P. (1972). Pathological left-handedness: An explanatory model. Cortex, 8, 121-135.
Satz, P. (1976). Cerebral dominance and reading disability: An old problem revisited. In R.
M. Knights & D. Bakker (Eds.), The neuropsychology of learning disorders (pp. 273-
294). Baltimore: University Park Press.
Satz, P., Bakker, D. J., Teunissen, J., Goebel, R., & van der Vlugt, H. (1975). Developmental
parameters of the ear asymmetry. A multivariate approach. Brain and Language, 2, 71-
85.
Satz, P., Rardin, D., & Ross, J. (1971). An evaluation of a theory of specific developmental
dyslexia. Child Development, 42, 2009-2021.
Shankweiler, D., & Studdert-Kennedy, M. (1975). A continuum of lateralization for speech
perception. Brain and Language, 2, 212-225.
Shucard, D. W., Shucard, J. L., & Thomas, D. G. (1977). Auditory evoked potentials as probes
of hemispheric differences in cognitive processing. Science, 197, 1295-1297.
Sparrow, S., & Satz, P. (1970). Dyslexia, laterality, and neuropsychological development. In
D. J. Bakker & P. Satz (Eds.). Specific reading disability: Advances in theory and method
(pp. 41-60). Rotterdam: Rotterdam University Press.
Sperry, R. W., Gazzaniga, M. S., & Bogen, J. H. (1969). Interhemispheric relationships: The
neocortical commissures: Syndromes of hemispheric disconnection. In P. Vinken & G.
W. Bruyn (Eds.), Handbook of clinical neurology (Vol. 4, pp. 273-290). New York:
Wiley.
Studdert-Kennedy, M., & Shankweiler, D. (1970). Hemispheric specialization for speech
perception. Journal of the Acoustical Society of America, 48, 579-594.
Summers, R. K., & Taylor, M. L. (1972). Cerebral speech dominance in language-disordered
and normal children. Cortex, 8, 224-232.
Tjossem, T. D., Hansen, T. J., & Ripley, H. S. (1961). An investigation of reading difficulty in
young children. Paper presented at the 117th annual meeting of the American Psycho-
logical Association, Chicago.
Tomlinson-Keasey, C., & Kelly, R. R. (1979). Is hemispheric specialization important to
scholastic achievement? Cortex, 15, 97-107.
Wada, J., Clark, R., & Hamm, A. (1975). Cerebral hemispheric asymmetry in humans. Ar-
chives of Neurology. 32, 239-246.
Walker, H. A., & Birch, H. G. (1970). Lateral preference and right-left awareness in schizo-
phrenic children. Journal of Nervous and Mental Disease, 15, 341-351.
Witelson, S. F. (1976). Early hemispheric specialization and interhemisphere plasticity: An
empirical and theoretical review. In S. Segalowitz & F. Gruber (Eds.)' Language and
development and neurologic theory (pp. 213-286). New York: Academic Press.
Witelson, S. F. (1977). Developmental dyslexia: Two right hemispheres and none left. Sci-
ence, 195,309-311.
Witelson, S. F., & Rabinovitch, M. S. (1972). Children's recall strategies in dichotic listening.
Journal of Experimental Child Psychology, 12, 106-113.
Woo, T. L., & Pearson, K. (1927). Dextrality and sinistrality. Biometrika, 19, 192-198.
Zangwill, O. C. (1960). Cerebral dominance and its relationship to psychological function.
London: Oliver & Boyd.
Zangwill, O. C. (1962). Dyslexia in relation to cerebral dominance. In J. Money (Ed.), Reading
disability (pp. 103-113). Baltimore: Johns Hopkins.
Zurif, E. F., & Carson, G. (1970). Dyslexia in relation to cerebral dominance and temporal
analysis. Neuropsychologia, 8, 351-361.
9
Infant and Early Childhood

Assessment
GLEN P. AYLWARD
THE STATUS OF NEUROPSYCHOLOGICAL ASSESSMENT IN

INFANCY AND EARLY CHILDHOOD
Neuropsychological assessment of infants and young children is a fledg-

ling area. This type of evaluation was previously accomplished by means
of a neurological examination performed by a pediatric neurologist or a
developmental examination administered by a developmental psychol-
ogist. There has not been complete concordance between the two assess-
ments since underlying neural function or maturation is not necessarily
reflected in developmental acquisitions. Because the aims of these assess-
ments and the information derived from each can overlap and differ,
boundaries have not been clearly defined and the area has been essen-
tially relegated to a "no-man's-land" status. Recently, general interest in
neuropsychological assessment has grown rapidly, so it is logical that this
interest should also extend to children 4 years of age and below. However,
despite the obvious need for theoretical frameworks and acceptable tech-
niques, few exist currently.
Many terms are used to describe infant and early childhood assess-
ment. These include neurological, psychological, developmental, neu-
romotor, neurodevelopmental, psychomotor, sensorimotor, neurobehav-
iOTal, and neuropsychological. The last term, however, is used rarely. The
reason for this lack of use may be related to the definitions of neuropsy-
chology that predominate. For example, neuropsychology is defined as
GLEN P. AYLWARD Division of Developmental and Behavioral Pediatrics. Departments

of Pediatrics and Psychiatry. Southern Illinois University School of Medicine. Springfield.
Illinois.
225
226 GLEN P. AYLWARD
the "behavioral expression of brain dysfunction" (Lezak, 1983), or "the

investigation of the role of individual brain systems in complex forms of
mental activity" (Luria, 1973). For various reasons, such definitions are
not appropriate in the case of the infant or young child.
Assessment of the infant and the young child is generally undertaken
to measure achievement in different functional areas; development is con-
sidered to reflect an increase in the number of abilities (Touwen, 1973).
Early in life, particularly in the neonatal period and early infancy, the
limited behavioral repertoire and the more direct, less complex rela-
tionship between neural function and overt behavior make it extremely
difficult to separate developmental (psychological) from neurological
functions. However, with increased age there is more divergence between
the two. It is likely that divergence is influenced by both maturation
(changes in underlying physical (neural) structure) and environmental
influences. Development is the result of the interaction between matura-
tion and environmental influences, although the infant's nervous system
must reach a critical level of maturation prior to effective utilization of
environmental input.
Neuropsychology of infancy and early childhood could best be de-
fined as the assessment of brain-behavior relationships in the context of
developmental change and maturation. Assessment of early neuropsy-
chological functions occurs against a backdrop of qualitative cognitive,
behavioral, and structural change; the functions that are assessed are more
variable and become increasingly more complex. Because the nervous
system of the infant and the young child (particularly the former) devel-
ops rapidly, examinations must be age-specific. Interpretation of findings
may be influenced by either a functional delay in development, neural
dysfunction, or both. Further, the frequently used actuarial approach may
not be as appropriate as more qualitative approaches, owing to the central
nervous system dynamics that occur during development. For these rea-
sons, early neuropsychological assessment differs markedly from assess-
ment of older individuals.
There are several uses for early neuropsychological assessment: im-
mediate diagnosis, change in status, prognosis, and research into early
brain-behavior relationships. The utility of immediate diagnosis for treat-
ment purposes is straightforward. Repeat assessments provide informa-
tion related to improvement or decline in an infant's status by provision
of baseline and subsequent data. Prognostic issues are particularly impor-
tant for parents, physicians, and educators. Finally, early assessment en-
ables investigators to gain knowledge about the developing nervous sys-
tem and to apply this knowledge in clinical situations.
The need for early neuropsychological assessment is increasing in
direct response to advances in neonatology and pediatrics. Early assess-
ment is particularly important in populations "at risk," from which many
children who will later manifest neuropsychological dysfunction are de-
EARL Y ASSESSMENT 227
rived (Kalverboer, Touwen, & Prechtl, 1973). Tjossem (1976) identified

three categories of risk: established, environmental, and biological. Estab-
lished risk involves medical disorders of a known etiology in which com-
promised developmental outcome is well documented (e.g., Down's syn-
drome). Environmental risk includes the quality of the mother-infant
interaction, opportunities for stimulation, and health care. The biological
risk category consists of infants and young children who have been ex-
posed to potentially noxious prenatal, perinatal, or postnatal events (i.e.,
those babies who are typically involved in follow-up studies).
Neuropsychological assessment over the first 4 years is particularly
important for this last category since progress in perinatal medicine has
resulted in a decrease in severe central nervous system damage in neo-
nates (e.g., Vohr & Hack, 1982). However, mild or moderate impairment is
now more frequent. This degree of dysfunction is often subtle but can
nonetheless prevent realization of a child's full potential. These high-
prevalencellow-severity disabilities are often precursors of subsequent
learning disabilities (Levine, 1983). The need for early neuropsycholo-
gical assessment is underscored further by the fact that these subtle diffi-
culties could easily be overlooked in a general neurological examination
or in developmental indices. Early neuropsychological evaluation also
could be applied to degenerative disorders, inborn errors of metabolism,
pervasive developmental disorders, acute trauma, oncology, and other
handicapping conditions in pediatric populations.
Finally, Public Law 94-142 (The Education for All Handicapped Chil-
dren Act, 1975), also has contributed to the increased need for detailed,
early neuropsychological assessment. Under this law, evaluation of chil-
dren with a broad range of developmental disabilities (emotional disor-
ders, mental retardation, physical disabilities, and sensory impairments)
is required. Early evaluation enables the placement of children in early
childhood education programs and also may help to specify areas needing
emphasis in birth-to-3 stimulation programs that have been legislated in
many states.
CONCEPTUAL ISSUES IN EARLY NEUROPSYCHOLOGICAL

ASSESSMENT
Differences in Assessing School-Age versus Younger Children

Assessment procedures developed to test school-age children or
adults are not applicable at younger ages (Bayley, 1969; Ulrey & Schnell,
1982). Reasons for this lack of applicability include behavioral state, tem-
perament, limitations of the test instruments, the need to consider both
qualitative and quantitative data, and the issue of noncompliance.
228 GLEN P. AYLWARD
Behavioral State
The behavioral state of the neonate is defined as the level of nervous
system arousal (Prechtl, 1977). States, originally developed by correlating
observed behavior with measures of physiological function, are generally
classified as quiet sleep, rapid eye movement sleep, quiet wakefulness,
active wakefulness, and crying. Behavioral states are critical because the
quality of elicited reflexes often varies as a function of state (Lenard, von
Bernuth, & Prechtl, 1968; Prechtl, Vlach, Lenard, & Grant, 1967). State
also influences habituation to external stimuli, muscle tone, and deep
tendon reflexes. Many neurobehavioral items are best administered in
specific behavioral states. States also afford insight into an infant's reac-
tivity and therefore can aid in assessment. Prechtl has included states in
the criteria for the diagnosis of several abnormal syndromes (Prechtl,
1977; Prechtl & Beintema, 1964). A predominance of certain states (e.g.,
crying) has been associated with subtle central nervous system dysfunc-
tion (Parmelee & Michaelis, 1971).
The influence of these behavioral states on assessment becomes less
important with increasing age. However, problems in physical discomfort
such as hunger, thirst, sleepiness, crying, and bowel and bladder disten-
tion continue to be important (Ulrey, 1982). Therefore, more flexibility in
neuropsychological testing is needed at these younger ages.
Temperament
Temperament is analogous to behavioral style, a biologically based
behavioral predisposition in spontaneous and reactive behavior (Chess &
Thomas, 1983). There are three major classifications of temperament. The
difficult child, estimated to comprise 10% of the general population, is
characterized by irregularity in biological function, negative withdrawal
responses to new situations, negative mood and intensity, and nonadap-
tability or slow adaptability to change. The slow-to-warm-up infant (15%)
typically displays negative responses of a mild intensity to new stimuli
and slow adaptability after repeated contact. In contrast, the easy child,
(40% of the population) adopts a positive response to new stimuli, adapts
quickly and easily to change, and has a predominantly positive mood.
These behavioral styles persist and can affect test scores of infants and
young children in various ways (see Hertzig, 1983).
Test Limitations/Qualitative and Quantitative Information

Reference norms, derived from apparently normal populations of in-
fants and children are typically used in the neuropsychological assess-
ment of handicapped and at-risk babies (Yang, 1979). As a result, it is
common for young, severely handicapped babies to be deemed untesta-
EARLY ASSESSMENT 229
ble. This is especially true among infants with severe motor impairment
because our current infant assessment instruments rely heavily on motor
responses. In neuropsychological assessment of young infants, the pres-
ence or absence of a behavior, and the qualitative manner in which a task
is carried out should be evaluated, with emphasis on the inclusion of
incidental observations. Many have argued that systematic assessment of
test-taking behavior should be included in the general evaluation of in-
fants and young children. This is evident in the development of the Bay-
ley (1969) Infant Behavior Record, which is used in addition to the Mental
and Psychomotor Developmental Indices.
There are other limitations as well. Hanson, Smith, and Hume (1984),
using the Griffiths Scales of Mental Development, question the reliability
of maternal report, the administration of certain items, and scoring crite-
ria. Discrepancies in scoring were found to be greatest at younger ages.
Disagreement in the scoring of what is observed may not be the issue; the
potential problem is in deciding whether to accept a behavior as the
child's optimum performance. The examiner's judgment of response qual-
ity (i.e., whether the response is the best of which the infant is capable) is
difficult to quantify, and traditional reliability measures do not take this
into account.
Noncompliance
This problem is reflected in the "nontestable" classification on the
Denver Developmental Screening Test (Frankenburg & Dodds, 1967).
Ounsted, Cockburn, and Moar (1983) reported lower global developmen-
tal scores in infants who were not cooperative, and noncompliant behav-
ior in 18% of their 4-year-olds precluded computation of a developmental
score. Many infants refused gross motor tasks, whereas those who were
minimally cooperative (enough for completion of a score) frequently dis-
played compromised visual-motor function. These investigators also re-
port an increased probability of subsequent problems in the noncoopera-
tive infants. It is possible that an unwillingness to attempt certain tasks
reflects an inability to perform these tasks, and this needs to be differenti-
ated in early neuropsychological assessment. There also is evidence that
abnormal infants behave differently on tests (Oski & Honig, 1978), and
behavior that is poorly adaptive and observed infrequently may indicate
abnormality and thus have diagnostic utility (Wolf & Lozoff, 1985).
Consistency of Dysfunction
Prediction
Early neurobehavioral diagnosis in severely impaired infants is fairly
straightforward (Davies, 1984). Prediction, however, is less certain. Pre-
230 GLEN P. AYLWARD
diction rates may account for as much as 49% of the variance between
early and later measures in significantly damaged infants, usually those
with developmental quotients less than 80 (Aylward & Kenny, 1979).
Similar predictive accuracy is reported in newborn infants who manifest
Prechtl's (1977) neurological syndromes (hyperexcitability, apathy, hemi-
syndrome). However, these prediction rates are for groups, not indi-
viduals, and there is much individual variability in cognitive, motor, and
neurological diagnoses over the first 3 years (Aylward, Gustafson, Ver-
hulst, & Colliver, 1987).
Diagnosis and longitudinal prediction are difficult in those infants
who are not overtly abnormal at an early age but whose performance is
suspect. Coolman et a1. (1985) found a 50% incidence of mild neuromotor
dysfunction in their high-risk nursery population, 75% of whom were
normal by age 2. Seven percent of the babies displayed moderate dysfunc-
tion, and half of these were normal by 2. In another investigation (Ayl-
ward et a1., 1987), using approximately 275 infants followed to age 3,
function was divided into cognitive, motor, and neurological spheres.
Seventy-one percent of the babies who were cognitively suspect at 40
weeks were normal by 3 years of age, 88% of the babies suspected to have
motor defects improved by age 3 whereas 93% of the infants defined as
neurologically suspect at 40 weeks had improved. Certain neu-
robehavioral conditions, such as hypotonia or head lag, are less persistent
over time than other neurological conditions (Bierman-van Eendenburg,
Jurgens-van der Zee, Olinga, Huisjes, & Touwen, 1981; Dubowitz et a1.,
1984; Hack, Caron, Rivers, & Fanaroff, 1983).
Central Nervous System Change

These data suggest that central nervous system dysfunction at birth is
not permanent since the majority of babies apparently recover (Touwen,
1978). Although there is a high false positive rate, these actually may be
"pseudo-false positives," which reflect reversible dysfunctions influ-
enced by maturation and development. The concept of developmental
discontinuity holds that structure to function relationships change. Neu-
ral development is not a linear progression in the number of elements and
interconnections; rather, there also is a retrogressive process. Prechtl
(1984) has identified the second month after birth as an age characterized
by major transitions in neural and sensory functions, whereas Amiel-
Tison (1982) suggests that major changes occur in the first 6 months.
Although there is much evidence to support the concept of plasticity of
the nervous system, there also is evidence that early injury causes perma-
nent qualitative change in neural organization and interconnections. This
may explain the so-called silent period (Amiel-Tison, 1982; Prechtl,
1977), in which dysfunction observed in the first few days of life often
disappears, only to reappear later.
Psychometric Concerns
Several methodological problems that may affect the reliability of
early neuropsychological assessment have been delineated (Aylward &
Kenny, 1979): the criterion measure used, the age at which testing is
administered (i.e., the keyage or the age of acquisition of new behaviors),
maturity, and strongly canalized sensory-motor development. Correction
for prematurity (Hunt, 1981; Siegel, 1983) may obscure subtle neuropsy-
chological findings by overcorrecting for gestational age. Any of these
factors could significantly affect neuropsychological status and therefore
must be considered.
Role of the Environment

Sameroff (1975) emphasizes a biological-environmental interaction
in his transactional model of development. This model was expanded in
the risk-route concept (Aylward & Kenny, 1979), in which outcome is
thought to be affected by three spheres of influence: medical/biological,
environmental/psychosocial, and behavioral/developmental. This model
has been applied in the assessment of consistency in cognitive, motor,
and neurological dysfunction (Aylward et al., 1987). Infants who were
cognitively normal at either 40 weeks or 9 months (corrected age) had a
20% chance of becoming cognitively abnormal by 36 months, yet those
who were motorically or neurologically normal had a 90% chance of
remaining so. When the influence of various background variables on
diagnostic consistency was assessed, consistency in cognitive function
from 9 to 36 months was influenced by environmental variables (center,
race, and a composite index of socioeconomic status). In contrast, motor
function from 9 to 36 months (which was most consistent) was influenced
only by the infant's gender. Consistency in neurological diagnosis was
influenced by a mix of several environmental and biological factors. Thus,
it appears that early assessments are adequate for diagnosis of motor and
neurological function, but that longer follow-up is necessary for cognitive
assessment, owing to the stronger environmental effects on this function.
Classification of Early Neuropsychological Findings

From the previous sections it is apparent that early neuropsychol-
ogical assessment is a hybrid, but yet unique, process that incorporates
techniques from neurological, clinical, and developmental approaches.
Therefore, it helps to list the various functions that may be measured or
observed in early assessment. Test instruments that are available to mea-
sure these functions are outlined in a subsequent section.
A proposed classification schema for early neuropsychological as-
sessment is found in Table 1. This schema is a synthesis of several classi-
232 GLEN P. AYLWARD
TABLE 1. Classification Schema for Early Neuropsychological

Assessment
I. Basic neurological functions (intactness)

Reflexes Left/right discrepancies
Asymmetries Motor inhibition
Stereotyped/lateralized postures Protective reflexes
Muscle tone Auditory orientation
Visual tracking/intactness
II. Receptive functions
Visual perception Spatial relations
Auditory/language Tactile processing
III. Expressive functions
A. Fine motor/oral motor B. Gross motor
Visual-motor integration Motility
Language/ speech Coordination
Fine motor-constructional
Visual-spatial orientation
Articulation
IV. Processing
A. Memory and learning Anticipatory behaviors
Temporal-sequential organization (auditory) Visual sequencing
Visual memory
Word retrieval
B. Thinking/reasoning
Cognitive-adaptive (problem solving) Number Concepts
Abstracting (verbal, nonverbal) Imitation
Seriation/classification Judgment
V. Mental activity
A. Attentional activities/level of consciousness
Conceptual tracking/concentration Alertness
Perseveration Distractibility
Behavioral states
B. Activity
Arousal
Cognitive
Motor
fications (e.g., Levine, 1983; Lezak, 1983) with applicability extended to

the neonatal period and early childhood. Since neuropsychological as-
sessment must be age-specific, all functions are not tested at each age.
Some functions (e.g., reflexes, muscle tone) are predominant during the
neonatal period, when behavior is less complex and more neurologically
oriented. Other functions, such as temporal-sequential organization or
verbal abstract abilities, are not testable until later, coinciding with in-
creased cortical development. There is much overlap between functions
and between classes of functions.
Basic Neurological Functions

This classification encompasses measurement of general functional
intactness. Early reflexes (Moro, Babinski, palmar grasp), asymmetries in
movement or muscle control (upper vs.lower extremities), and lateralized
or stereotyped postures (asymmetric tonic neck reflex, fisted hands) are
included. In addition, left-right discrepancies such as impaired midline
behaviors, hemiplegias, prematurely established handedness, muscle tone
(hypo- or hypertonia, head control), motor inhibition (mirroring, reflex
overflow), and visual and auditory intactness (conjugate gaze, nystagmus,
tracking skills, response to sounds) are involved. This classification also
includes protective reactions in response to change in the infant's body
orientation in space (e.g., parachute reactions). Keen observation skills are
critical because some functions can be evaluated in a present/absent man-
ner, whereas others are based on more qualitative, subjective judgment.
Receptive Functions
Receptive functions involve the entry of information into the central
processing system, utilizing sensation and perception. This class of func-
tion becomes more complex with age. Visual, auditory, and tactile abili-
ties are involved, but owing to testing instruments, visual and auditory
functions are emphasized more than tactile processes in early assessment.
Tactile function is assessed in the first several months of life in the form of
proprioceptive reflexes, grasping, and mouthing objects. Auditory dis-
crimination and verbal receptive skills also are involved in this functional
grouping.
Expressive Functions
Expressive functions refer to behaviors produced by the child and
observed by the examiner. The repertoire of behavior reflecting expressive
functions increases with age. Fine motor abilities include eye-hand coor-
dination, manipulating with the fingers, constructional abilities, and psy-
chomotor speed. Oral motor and language functions consist of what is
said and how it is articulated. Gross motor function involves motility,
coordination, ambulation, and gait. Deficits in this area are often defined
as apraxias, dysphasias, anomias, or dyskinesias.
Processing
Processing includes two components: memory/learning and think-
ing/reasoning. Memory/learning involves visual and auditory short-term
memory (registration and immediate memory) and long-term memory.
234 GLEN P. AYLWARD
Early functions include habituation, object permanency, stranger anxiety,

and recognition of self in a mirror. Time sequence relationships, such as
digit span and recall of objects presented visually, and word retrieval
(naming pictures or objects), are later memory/learning functions in the
young child.
Thinking/reasoning is often called higher-order processing, in which
the infant or child synthesizes information. These complex processes are
thought to provide better prognostic indicators of an infant's true poten-
tial, particularly in motorically impaired infants. In such cases, if spikes
of age-appropriate function in these areas are observed against a backdrop
of lower levels of motor and neurologic function, long-term mental defi-
ciencies may not be as likely. Problem solving, imitation, conceptual
grouping, body image concepts, counting and sorting, and performance
on form boards are later abilities that fall in this subcategory.
Mental Activity
This class of function consists of two components: attentional ac-
tivities/level of consciousness, and activity. Attentional activities are
heavily dependent upon developmental age since such skills are poorly
developed in younger infants and children. These functions can therefore
be diagnostically useful when attentional behavior is compared with what
is expected at a specific age. Many of these functions also determine how
an infant approaches test taking. Behavioral states are important early;
later, alertness becomes more crucial. Activity, whether characterized as
hypo- or hyperactivity, will affect test-taking behavior, but it is also of
diagnostic utility because diffuse neuropsychological dysfunction can be
reflected in either excessively high or low activity levels.
In summary, it is important to emphasize that the classes of function
in Table 1 are interrelated, and that the divisions are provided to form a
conceptual framework. For example, when an infant follows commands,
receptive functions (temporal-sequential) as well as expressive abilities
are involved. Similarly, on form boards, receptive functions (visual per-
ceptual), processing (thinking/reasoning), and mental activities (attention
and concentration) are employed. Nonetheless, careful analysis of data
obtained in early neuropsychological assessment enables the progressive
discrimination of areas of adequate function and those of deficit.
ASSESSMENT INSTRUMENTS
There are few, if any, tests that are designed specifically to provide
neuropsychological data for infants and young children. However, many
tests yield neuropsychological information but simply have not been
identified as doing so previously. Early neuropsychological assessment of

infants and young children can therefore be accomplished by drawing on
tests from neurology and developmental psychology.
Many of these tests are discussed subsequently. The listing found in
Table 2 is selective and not exhaustive. The test instruments that are
included are those that are used frequently by examiners of infants and
young children, and are specifically useful in children under the age of 4.
Other lists of developmental and cognitive tests are available (Franken-
burg, 1983; Ulrey, 1981).
In Table 2, five test instruments for neonates, five assessments for
infants, and four tests applicable to the assessment of young children are
listed. A brief description of the test, the information that is derived, and
comments about the strengths (+) and weaknesses (-) of each instrument
are provided.
Assessment instruments, and the functions assessed by each are
found in Table 3, which is a synthesis of Tables 1 and 2. The assessment
instruments are classified as newborn/neonatal, infancy, and toddler/ear-
ly childhood.
Newborn/N eonatal
Neuropsychological functions measured by assessment instruments
in this age range cluster in the neurological function/intactness, ex-
pressive (gross motor), and mental activity groupings (level of con-
sciousness, activity). All examinations are adequate to assess intactness,
although the examiner's additional observations are necessary in some.
Infancy
At this age, assessments cluster in the receptive, expressive, and pro-
cessing groupings. Sensorimotor skills are generally emphasized; these
include visual/motor integrative, fine motor-constructive, and gross
motor functions. Visual/perceptual abilities are also heavily weighted.
Inclusion of more neurological function items in assessment at this age is
necessary. The other major deficiency is found in the mental activity
grouping. In both cases, additional items and observations could be in-
cluded. Reflexes are evaluated only with the Milani-Comparetti exam at
this age; this suggests the advisability of incorporating this assessment
instrument with other more developmentally oriented techniques.
Early Childhood
As in the infancy age range, little emphasis is placed on the neu-
rological function and mental activity groupings. Gross motor skills are
TABLE 2. Assessment Instruments Used in Early Neuropsychological Testing
Assessment instrument Description Information Comment
Age: Newborn
Brazelton Neonatal Behav- 27 behavioral items (9-point How infant responds to ( +) Clinical instrument, provides good re-
ioral Assessment Scale scale), 20 elicited re- caretakers and environ- search information, recovery from stress;
(Brazelton, 1973) sponses (3-point scale) ment; states, habituation, ( -) some problems with data interpreta-
orientation (visual, audition and scaling, poor long-range predic-
tory), muscle tone, action
tivity; "best performance"
Prechtl Neurological Ex- Approximatley 52 items + Gross state of CNS reactivity (+) Extensivley studied, good standardiza-
amination (Prechtl, 1977; states, postures, spon- (motor activity, states) + tion, provides immediate diagnosis, prog-
Prechtl & Beintema, 1964) taneous activity specific lesional approach; nosis, those needing follow-up,
syndromes identified behavioral response descriptions; (-)
lengthy, perhaps too detailed
Brief Infant Neu- Reduced version of Revised Screen of CNS intactness (+) Easily administered, less time-consum-
robehavioral Optimality Prechtl Neurological Ex- (tonus, tendon reflexes, ing, similar sensitivity and specificity
Scale (Aylward, Verhulst, amination, 15 items scored spontaneous activity, some values as Prechtl, summary score; (-) not
& Colliver, 1985) in optimal/nonoptimal primitive reflexes) as much data on states, not as extensive
manner as Prechtl
Parmelee's Neurological Divided into two sections- Active and reflexive muscle (+) Brief-takes 10 minutes to administer,
Examination (Parmelee, Section I: state observa- tone, primary reflexes, summary score; (-) not as extensive as
Michaelis, Kopp, & Sig- tions, 20 items; Section II: arousal, and spontaneous Prechtl
man, 1974) crying, activity, tremor, behavior
etc.
Graham Behavior Test for Five scales: pain threshold, Originally designed to differ- ( +) Correlations between neonatal data and
Neonates (Graham, maturational level, visual entiate normal and brain- early infant data, revision had large sam-
Matarazzo, & Caldwell, responsiveness, irritability, injured newborns; motor ple; (-) long-range predictability fair,
1956; Rosenblith, 1974) and muscle tension rat- function (head control, used infrequently
ings; pain threshold crawl) tactile-adaptive
deleted in revision (cotton over nose), visual
tracking, auditory respon-
siveness, irritability rating
Age: Infancy
Gesell Developmental Items provided at each of Key ages: 4, 16, 28, and 40 ( +) Basis for most examinations of infancy,
Schedules (Knobloch & key ages, under five weeks; 12, 18, 36, and 48 good screening inventory; (-) not as well
Pasmanick, 1974; groups; can be scored by months; gross motor, fine standardized as other infant examina-
Knobloch, Stevens, & Mal- observation or history motor, adaptive, language, tions, ratio developmental quotient
one, 1980). personal-social
Bayley Scales of Infant De- Mental Developmental Index Sensorimotor skills, problem (+) Well standardized, good overall assess-
velopment (Bayley, 1969) (163 items), Psychomotor solving, verbal skills, and ment of child's strengths and weaknesses,
Developmental Index (81 areas outlined previously most widely used; (-) questionable pre-
items), Infant Behavior in Gesell dictability, very dependent on motor
Record function
Griffiths Mental Develop- Derived from Gesell, 498 Locomotor, personal-social, (+) Used predominantly in Great Britain,
mental Scale (Griffiths, items, most detailed in hearing and speech, eye rigorous training; (-) many unreliable
1954, 1970) first 2 years; five areas and hand coordination, items
with several items for each performance; designed to
month of age enable examiner to distin-
guish normal and abnor-
mal children
Cattell Infant Intelligence Items arranged on an age Many motor items at early (+) Smooth transition from infant to early
Scale (1940) scale, drawing from Gesell ages; later ages include childhood examination, good for testing
at early ages and Stanford- more cognitive and lan- older children who function at lower lev-
Binet from 22 months up- guage items; provides an els; (-) small standardization sample
wards; 2-36 months IQ score (294), IQ concept questionable, not used
frequently
Milani-Comparetti Neuro- Assesses spontaneous be- Neuromotor abnormalities- (+) Administered in 5-10 minutes, good
Developmental Screening havior, postural control, head and body control, screening device that can be used with
Examination (Milani-Com- evoked responses; chart movement, evoked primi- other examinations; (-) tilting reactions
paretti & Gidoni, 1967) provided with age norms tive reflexes, righting, par- need tilt board, which is often not readily
in terms of onset and dis- achute, and tilting available
appearance of responses reactions; enables detec-
0-24 months tion of cerebral palsy, de-
velopmental delay
(continued)
Assessment instrument Descri ption Information Comment
Age: Toddler/early childhood

Merrill-Palmer Scale of Verbal, perceptual-motor, Perceptual-motor function; (+) Can be used in language-delayed chil-
Mental Tests (Stutsman, and nonverbal items extended version-seman- dren, high interest level for children; (-)
1948); Extended Merrill- grouped into 6-month age tic production and evalua- norms outdated, heavily loaded with per-
Palmer Scales (1969) ranges; extended version tion, figural production ceptual performance items, should be
has 16 tasks-1 ' /z-6 years, and evaluation (verbal ex- used as a supplemental test, poor predic-
3-5 years pression, comprehension, tive validity
perceptual-motor, and
nonverbal skills)
Stanford-Binet (1960; Ter- Age 2-adult, provides IQ General intellectual abilities; (+) Used frequently, good predictive utility;
man & Merrill, 1972) score; grouped at half-year general comprehension, vi- (-) unequal distribution of verbal and
intervals from 2 to 5, year- sual-motor ability, arith- nonverbal items at different ages, heavy
ly thereafter metic reasoning, memory loading of perceptual/performance items
and concentration, vocabu- at early ages
lary, judgment and reason-
ing
Stanford -Binet-Revised Items of same type grouped Four areas assessed: verbal, (+) Differentiates between MR and LD, bet-
(Thorndike, Hagen, & Sat- into 15 tests-2-adult abstract/visual, quantita- ter reliability, better underlying test theo-
tler, 1986) tive reasoning, short-term ry; (-) new instrument, needs more
memory application
McCarthy Scales of Chil- Eighteen component tests di- Fi ve scales + general cogni- (+) Age equivalents available for subtests,
dren's Abilities (McCarthy, vided into five scales: ver- tive (summary of verbal, includes a gross motor component, best
1972) bal, perceptual- perceptual-performance, in the 3- to 5-year age range, can prorate;
performance, quantitative, and quantitative); provides ( -) LD children score lower on this test,
memory, and motor; pro- good overview of child's may give underestimate of child's true
vides a General Cognitive general function (verbal abilities, of limited utility for older chil-
Index, 2 ' /z_8 ' jz years and nonverbal) as well as dren
specific strengths and
weaknesses; early detec-
tion of LD
not assessed consistently, whereas receptive, expressive, and processing

functions are evaluated in depth. Because it requires a great deal of skill to
elicit reflexes at this age, observation of neurological function, as evident
in gross and fine motor behavior, becomes even more important.
Useful Findings
Neuropsychological indicators in the newborn period that are partic-
ularly useful as indicators of dysfunction include Prechtl's syndromes,
hypertonicity of the axial musculature and lower extremities, high ampli-
tude-low frequency tremors, nystagmus, clonus, reflex overflow, and ab-
errant state regulation. Isolated abnormal findings are generally not help-
ful. In addition, early hypotonia and head lag, subtle asymmetries,
mediocre visual tracking, and suspected motor or neurological diagnoses
are often transient. During infancy, persistence of primitive reflexes, cor-
tical thumb sign, consistent irritability, persistent dystonia, a very pre-
dominant hand preference, stereotyped, repetitive movements, and devel-
opmental quotients that are 2 or more standard deviations below average
indicate significant problems. Suspect motor or neurological diagnoses
are often variable, and most infants tend to improve. In early childhood,
many of the functions seen in older children become more apparent.
Neuropsychological assessment should focus particularly on emerging
temporal-sequential skills (but not digits backwards), concentration, word
retrieval difficulties, perseverative tendencies (motor and cognitive), ac-
tivity levels, and gross motor function. Visual-motor skills may be over-
emphasized during infancy and at this later time period.
DIRECTIONS
The emergence of a discipline of early developmental neuropsychol-

ogy is timely and meets a need that is expanding in relation to improve-
ments in pediatric health care. This discipline consists of a synthesis of
existing developmental and neurological techniques, but it is distinct
from developmental neurology or developmental neuropsychiatry. Early
developmental neuropsychology includes assessment from the neonatal
period to early childhood, an age span that is currently underserved.
It is likely that this chapter includes many concepts and descriptions
of assessment instruments that are not familiar to neuropsychologists ac-
customed to working with older populations. Those who are acquainted
with these instruments and concepts constitute a definite minority. There
are several reasons for this situation: Assessment in this group has been
provided by other disciplines, there are very limited training oppor-
tunities in traditional psychology programs, and a lack of acceptance by
TABLE 3. Assessment Instruments and Functions Assessed a
Assessment instrument
Newborn/neonatal Infancy Toddler/early childhood

Functions (;J
... ~
}fJ ~
~ .9. Q)
-8.8
0..-1 -+-'
~-E
;>,
00
;>,
.8 -;:::
- ~ '"";d~ S.:.::: .- Q) ._ OJ
... -0 ~
-00::;
OJ 00 '" ",..0 OJ ai ~ ... ... .L: 00
- 0 ~
- : o~
. ~ OJ
a003 ...o OJ
a3N {1 .~ ~ S i:l .L: ~ ;;, ::: ::: S
OJ :-;::
'" '" 0- OJ- ""'~ ~ ""' OJ ~ca
~ 05~'~Uj s"8 s '"
... OJ
00 ~ ~ u
:B... .- u
...'" ... ;:l '" '-' '-' 1:0
'" u'" ~ S "'1:0
1:0 (l.. ;:;0 '" OJ X 0::;0 '-' ~~ '"
U51:o .- ;gC/l
OJ
(l..z~ 8 U5
Z
I. Neurologic functions
Reflexes + + + + + +
Asymmetries + + +
Stereotyped/lateralized postures + + + + + + + +
Muscle tone + + + + + + + + + +
Visual tracking/intactness + + + + + + +
LlR discrepancy + + + + +
Motor inhibition +
Protective reflexes + + + + +
Auditory orientation + + + + + + +
II. Receptive functions
Visual perception + + + + + + + + + +
Auditory /language + + + + + + + +
Spatial relations + + + + + + + +
Tactile processing + + + + + + + +
III. Expressive functions
A. Fine motor/oral motor
Visual/motor integrative + + + + + + + +
Language/ speech + + + + + + + +
Fine motor constructive + + + + + + + +
Visual/spatial + + + + + + + +
Articulation + + + +
B. Gross motor
Motility + + + + + + + + + + + +
Coordination + + + + + + + + + + + +
IV. Processing
A. Memory
Temporal-sequential organization + + + + + + + + + +
(auditory)
Anticipatory behaviors + + + +
Visual memory + + + + + + + +
Visual sequencing + +
Word retrieval + + + + + + + +
B. Thinking/reasoning
Cognitive/adaptive problem + + + + + + + +
solving
Abstracting + + + + +
Seriation/classification + + + + +
Number concepts + + + + + + +
Imitation + + + + + + +
Judgment + + + + + + + +
V. Mental activity
A. Attentional/level of consciousness
Conceptual tracking/concentration + + +
Perseveration
Attention/ distractabili ty +
Behavioral states + + + +
Alertness + + + + +
B. Activity
Arousal + + +
Cognitive + + +
Motor + + + + + + + +
Irritability + + + + +
aA "+" indicates function is assessed by evaluation instrument; a "-" indicates function is not assessed directly; a "*,, means observations of this function can be made
but are not scored formally.
242 GLEN P. AYLWARD
other professionals, particularly those in medicine, is prevalent. A signifi-

cant component of neonatal and infant neurological assessment is behav-
ioral observation; psychologists are specialists in the observation of be-
havior. Therefore, a major thrust in early developmental neuropsychology
is to make examiners better versed in which behaviors to elicit and to
observe.
Early developmental neuropsychology requires specific training that
is not afforded in traditional clinical or developmental psychology train-
ing programs. Training should be allied closely to pediatrics, pediatric
neurology, neonatology, genetics, and interdisciplinary clinics. Reading
materials should be expanded to include texts such as that of the Commit-
tee on Joint Assessment of Prenatal and Perinatal Factors Associated with
Brain Disorders (Freeman, 1985). Examiners cannot learn assessment
techniques solely from manuals or books; hands-on experience is re-
quired under the direction of a variety of professionals. Unfortunately, as
is often the case with instruments such as the Bayley Scales, unqualified
examiners attempt to pass themselves off as competent in the use of a test.
In addition, and perhaps most important, a thorough understanding of
normal development is necessary so that the examiner can detect devia-
tions from normality. Concepts and techniques should be structured from
fetal development upwards; techniques applicable to adults or even older
children cannot be downscaled. Information that is helpful at later ages
may be useless in infants and young children, and vice versa.
In early neuropsychological assessment, damage, age, development,
and environment form a complex interactive matrix in which cause-
effect relationships and models are obscured. The previously mentioned
Risk Route concept (Aylward & Kenny, 1979), which takes many of these
factors into account, should be considered in any early neuropsychol-
ogical assessment. A major dilemma facing those involved in this type of
assessment is that the probability of detecting subtle damage is greater
with increased age; however, the longer the time between an insult and
the assessment, the greater is the influence of intervening variables, par-
ticularly the environment. Therefore, earlier assessment has a better po-
tential to elucidate potential cause-and-effect relationships, although
there are limitations on what can be assessed. At later ages, assessment
techniques are better, but cause-and-effect relationships are more mud-
dled, largely owing to environmental influences. A major area for future
research is the development of predictive models of dysfunction. The
optimum duration of follow-up, and selection of appropriate instruments
are two additional research topics that require attention.
There are many limitations in the test instruments that are available
currently since many were developed in the 1920s and 1930s (Gesell,
1928; Gesell & Amatruda, 1947). Other tests lack a unifying conceptual
base and instead contain unrelated items that children are required to
pass at a given age. Better methods are needed to help in scoring behaviors
that have not been scored previously. It may be more informative to focus
on how an infant or a young child achieves a criterion measure rather than
if the item is passed or not. Further, many functions are linked with test
items in a post hoc method (e.g., Stanford-Binet), rather than defining a
function and devising a task to assess it.
Selection of combinations of items, drawn from existing tests but
based on a conceptual framework of functions, is suggested as the best
approach. It is advisable to implement the basics of Prechtl's (1980) op-
timality concept in early neuropsychological assessment. Rather than
focusing on the normality or abnormality of a response, the optimality
concept provides a measure of deviation (or lack of deviation) from a
known optimal neuropsychological response. The problem of weighting
different items is circumvented because the optimal responses are
summed and divided by the total number of neuropsychological items
administered. Cutoff scores (e.g., 75% of the responses were optimal)
could be devised to assist in diagnosis. This technique has been used by
investigators during the neonatal and earl infancy period with good re-
sults (Aylward, Verhulst & Colliver, 1985; Stave & Ruvalo, 1980). This
method would be particularly helpful in infancy and early childhood
assessment, where combinations of examination instruments and tech-
niques need to be employed. The optimal response for each item could be
determined from previous data, and these criteria could then be as-
sembled into a workable assessment battery.
The need for an instrument that would identify infants and young
children who have abnormalities of posture, tone, and movement, as well
as those with developmental delays, has recently been emphasized. This
is because categorical results of developmental assessment instruments
often yield low sensitivity (Sciarillo, Brown, Robinson, Bennett, & Sells,
1986). We are currently applying this methodology with the Early Neuro-
psychological Optimality Rating Scales (ENORS). Different scales have
been developed in an a priori manner to reflect major developmental
acquisitions at several key ages (6, 12, 18, and 24 months). As an example,
the 12-month ENORS is found in the Appendix.
The scales are a synthesis of information outlined previously in Ta-
bles 1 and 3 and are designed to provide a more thorough assessment by
covering functions that would ordinarily be missed with the administra-
tion of a developmental or neurological examination in isolation. The
scales utilize components of the Bayley, the Milani-Comparetti, and neu-
rological indicators, as well as developmental milestones. Individual
items included in the ENORS are not necessarily new; basic reflexes and
many expressive and receptive functions are emphasized in pediatric
postgraduate training. Similarly, developmental milestones and the Bay-
ley developmental quotients are used by developmental psychologists.
244 GLEN P. AYLWARD
The standard format, the combination of a variety of items into a concep-

tual framework, and having scoring and interpretation based on the op-
timality concept are innovations. A short scoring manual to define op-
timal responses is available for each age scale, and this affords reliability
in scoring at different centers.
We are in the process of applying this concept to a sample of 611
babies enrolled in the National Heart, Lung and Blood Institute Antenatal
Steroid Study, approximately 70% of whom were followed to 3 years of
age. We plan to produce cutoff scores for various ages, based on more
long-term sensitivities and specificities, and items that are redundant will
be eliminated. This instrument will subsequently be applied to a smaller
sample (n = 100) of high-risk infants derived from our High Risk Nursery
Developmental Continuity Clinic. This approach is a fertile area for subse-
quent research.
In summary, there is a need for early developmental neuropsychol-
ogy, particularly in neonatal follow-up and multidisciplinary diagnostic
clinics. Acceptance of this discipline will be gained by the continued
demonstration of clinical and research competencies.
APPENDIX
Early Neuropsychological Optimality Rating Scale

12 Months
Classification Optimal Nonoptimal
I. Basic Reflexes
Persistence of primitive reflexes (Moro, asymmetric No Yes
tonic neck, palmar/plantar grasps, etc.)
Asymmetries/marked L-R discrepancy (mirroring, ob- No Yes
vious hand preference)
Presence of protective reactions (parachute reactions) Yes No
Good head control Yes No
Hypotonia/hypertonia-extremities No Yes
Absence of stereotyped, repetitive movements and/or Yes No
cortical thumb sign
II. Receptive Functions
Appropriate responses to auditory stimuli (orients, Yes No
turns head)
Appropriate responses to visual stimuli (tracking, lack Yes No
of nystagmus, etc.)
Follows simple commands Yes No
III. Expressive Functions
A. Fine motor/oral motor
Neat pincer grasp Yes No
Presence of midline behaviors (hands together, trans- Yes No
fers, holds two objects simultaneously)
Radial-digital grasp Yes No
Babbles/appropriate sounds for age (dada/mama spe- Yes No
cifically, etc.)
Good eye-hand coordination Yes No
B. Gross motor
Age-appropriate gait/ambulation (no tiptoe walking, Yes No
no dragging of an extremity)
Uncoordinated movements No Yes
Psychomotor Developmental Yes No
Index 2: 90
IV. Processing
Appreciation of object permanency Yes No
Imitative abilities Yes No
Simple problem-solving skills Yes No
V. Mental Activity
Goal-directed behaviors Yes No
Attentive to procedures Yes No
Average level of activity for age Yes No
Persistant crying/irritability No Yes
Mental Developmental Index 2: 90 Yes No
Total Optimal
Total Nonoptimal
Total Optimal ( )
Neuropsychological Optimality Score =
Optimal ( ) + Nonoptimal (
246 GLEN P. AYLWARD
REFERENCES
Amiel-Tison, C. (1982). Neurologic signs, aetiology, and implications. In P. Stratton (Ed.),
Psychobiology of the human newborn (pp. 75-94). New York: Wiley.
Aylward, G. P., Dunteman, G., Hatcher, R. P., Gustafson, N. F., & Widemayer, S. (1985). The
SES-Composite Index: A tool for developmental outcome studies. Psychological Docu-
ments, 15(Ms. No. 2683).
Aylward, G. P., Gustafson, N., Verhulst, S. J., & Colliver, J. A. (1987). Consistency in the
diagnosis of cognitive, motor and neurologic function over the first three years. Journal
of Pediatric Psychology, 12, 77-98.
Aylward, G. P., & Kenny, T. J. (1979). Developmental follow-up: Inherent problems and a
conceptual model. Journal of Pediatric Psychology, 4, 331-343.
Aylward, G. P., Verhulst, S. J., & Colliver, J. A. (1985). Development of a brief infant neu-
robehavioral optimality scale: Longitudinal sensitivity and specificity. Developmental
Neuropsychology, 1, 265-276.
Bayley, N. (1969). Bayley Scales of Infant Development. New York: Psychological Corpora-
tion.
Bierman-van Eendenburg, M. E. C., Jurgens-van der Zee, A. D., Olinga, A. A., Huisjes, H. H.,
& Touwen, B. C. L. (1981). Predictive value of neonatal neurological examination. A
follow-up study at 18 months. Developmental Medicine and Child Neurology, 23, 296-
305.
Brazelton, T. B. (1973). Neonatal behavioral assessment scale. Clinics in Developmental
Medicine, No. 50. Philadelphia: Lippincott.
Cattell, P. (1940). The measurement of intelligence of infants and young children. New York:
Psychological Corp.
Chess, S., & Thomas, A. (1983). Dynamics of individual behavioral development. In M. D.
Levine, W. B. Carey, A. C. Crocker, & R. T. Gross. Developmental-behavioral pediatrics
(pp. 158-174). Philadelphia: W.B. Saunders.
Coolman, R. B., Bennett, F. C., Sells, C. J., Swanson, M. W., Andrews, M. S., & Robinson, N.
M. (1985). Neuromotor development of graduates of the neonatal intensive care unit:
Patterns encountered in the first two years of life. Journal of Developmental and Behav-
ioral Pediatrics, 6, 327-333.
Davies, P. A. (1984). Follow-up of low birthweight children. Archives of Disease in Child-
hood, 59, 794-797.
Dubowitz, L. M. S., Dubowitz, V., Palmer, P. G., Miller, G., Fawer, C. L., & Levine, M. l.
(1984). Correlation of neurologic assessment in the preterm newborn infant with out-
come at one year. Journal of Pediatrics, 105, 452-456.
Ellenberg, J. H., & Nelson, K. B. (1981). Early recognition of infants at high risk for cerebral
palsy: Examination at age four months. Developmental Medicine and Child Neurology,
14, 575-584.
Frankenburg, W. K. (1983). Infant and preschool developmental screening. In M. D. Levine,
W. B. Carey, A. C. Crocker, & R. T. Gross (Eds). Developmental-behavioral pediatrics
(pp. 927-937). Philadelphia: W. B. Saunders.
Frankenburg, W. K., & Dodds, J. B. (1967). The Denver Developmental Screening Test.
Journal of Pediatrics, 71, 181-191.
Freeman, J. (Ed). (1985). Prenatal and perinatal factors associated with brain disorders (NIH
Pub. No. 85-1149). Washington, DC: U.S. Department of Health and Human Services.
Gesell, A. (1928). Infancy and human growth. New York: Macmillan.
Gesell, A., & Amatruda, C. (1947). Developmental diagnosis (2nd ed.). New York: Hoeber.
Graham, F. K., Matarazzo, R. G., & Caldwell, B. M. (1956). Behavioral differences between
normal and traumatized newborns: II. Standardization, reliability and validity. Psycho-
logical Monographs, 70(21. Whole No. 428).
Griffiths, R. (1954). The abilities of babies. New York: McGraw-Hill.
Griffiths, R. (1970). The abilities of young children. A comprehensive system of mental

measurement for the first eight years of life. London: Child Development Research
Center.
Hack, M., Caron, B., Rivers, A., & Fanaroff, A. A. (1983). The very low birthweight infant:
The broader spectrum of morbidity during infancy and early childhood. Journal of
Developmental and Behavioral Pediatrics, 4, 243-249.
Hanson, R., Smith, J. A., & Hume, W. (1984). Some reasons for disagreement among scorers
of infant intelligence test items. Child: Care, Health, and Development, 10, 17-30.
Hertzig, M. E. (1983). Temperament and neurological status. In M. Rutter (Ed.). Developmen-
tal neuropsychiatry (pp. 164-180). New York: Guilford Press.
Hunt, J. V. (1981). Predicting intellectual disorders in childhood for preterm infants with
birthweights below 1501 gm. In S. 1. Friedman & M. Sigman (Eds.)' Preterm birth and
psychological development (pp. 329-352). New York: Academic Press.
Kalverboer, A. F., Touwen, B. C. 1., & Prechtl. H. F. R. (1973). Follow-up of infants at risk of
minor brain dysfunction. Annals of the New York Academy of Sciences, 205, 173-187.
Knobloch, H., & Pasamanick, B. (1974). Developmental diagnosis (3rd. ed.). New York:
Harper & Row.
Knobloch, H., Stevens, F., & Malone, A. F. (1980). Manual of developmental diagnosis. New
York: Harper & Row.
Lenard, H. G., von Bernuth, H., & Prechtl, H. F. R. (1968). Reflexes and their relationship to
behavioral state in the newborn. Acta Paediatrica Scandinavica, 57, 177-185.
Levine, M. D. (1983). The developmental assessment of the school age child. In M. D. Levine,
W. B. Carey, A. C. Crocker, & R. T. Gross (Eds.). Developmental-behavioral pediatrics
(pp. 938-947). Philadelphia: W.B. Saunders.
Lezak, M. D. (1983). Neuropsychological assessment (pp. 15-38). New York: Oxford Univer-
sity Press.
Luria, A. R. (1973). The working brain (pp. 19-42, 341-344). New York: Basic Books.
Psychological Corp.
Milani-Comparetti, A., & Gidoni, E. A. (1967). Routine developmental examination in nor-
mal and retarded children. Developmental Medicine and Child Neurology, 9, 631-638.
Oski, F. A., & Honig, A. S. (1978). The effects of therapy on the developmental scores of iron
deficient infants. Journal of Pediatrics, 92, 21-25.
Ounsted, M., Cockburn, J., & Moar, V. A. (1983). Developmental assessment at four years:
Are there any differences between children who do, or do not, cooperate? Archives of
Disease in Childhood, 58, 286-289.
Parmelee, A. H., & Michaelis, R. (1971). Neurological examination of the newborn. In J.
Hellmuth (Ed.). Exceptional infant 2: Studies in abnormalities (pp. 3-23). New York:
Brunner/Maze!'
Parmelee, A. H., Michaelis, R., Kopp, C. B., & Sigman, M. (1974). Newborn neurological
examination. UCLA Infant Studies Project.
Prechtl. H. F. R. (1968). Neurological findings in infants after pre- and parinatal complica-
tions. In J. H. P. Jonxis, H. K. A. Vissir, & J. A. Troelestra (Eds.). Aspects of praematurity
and dysmaturity. Leiden: Stenfert Kroese.
Prechtl. H. F. R. (1977). The neurological examination of the full-term newborn infant (2nd
ed.). Clinics in Developmental Medicine, No. 63. London: Heinemann.
Prechtl. H. F. R. (1980). The optimality concept. Early Human Development, 4, 201-205.
Prechtl. H. F. R. (1984). Continuity and change in early neural development. In H. F. R.
Prechtl (Ed.). Continuity of neural functions from prenatal to postnatal life (pp. 1-15).
Clinics in Developmental Medicine, No. 94. Philadelphia: Lippincott.
Prechtl. H. F. R., & Beintema, D. (1964). The neurological examination of the full-term
newborn infant. Little Club Clinics in Developmental Medicine, No. 12. London:
William Heinemann Medical Books.
248 GLEN P. AYLWARD
Prechtl, H. F. R., Vlach, V., Lenard, H. G., & Grant, D. (1967). Exteroceptive and tendon
reflexes in various behavioral states in the newborn infant. Biologia Neonatorum, 11,
159-175.
Public Law 94-142. (1975). Education for all handicapped children act. 94th Congress, S.6.
Rosenblith, J. F. (1974). Relations between neonatal behaviors and those at eight months.
Developmental Psychology, 10, 779-792.
Sameroff, A. J. (1975). Early influences on development: Fact or fancy? Merrill-Palmer Quar-
terly, 21, 267-294.
Sciarillo, W. G., Brown, M. M., Robinson, N. M., Bennett, F. c., & Sells, C. J. (1986). Effective-
ness of the Denver Developmental Screening Test with biologically vulnerable infants.
Journal of Developmental and Behavioral Pediatrics, 7, 77-83.
Siegel, L. S. (1983). Correction for prematurity and its consequences for the assessment of the
very low birth weight infant. Child Development, 54, 1176-1188.
Stave, V., & Ruvalo, C. (1980). Neurological development in very-low-birthweight infants:
Application of a standardized examination and Prechtl's optimality concept in routine
evaluations. Early Human Development, 4, 229-241.
Stutsman, R. (1948). Guide for administering the Merrill-Palmer Scale of mental tests. New
York: Harcourt Brace Jovanovich.
Taylor. D. j. (1984). Low birthweight and neurodevelopmental handicap. Clinics in Obstet-
rics and Gynecology, 11, 525-542.
Terman, 1., & Merrill, M. (1972). Stanford-Binet intelligence scale. Boston: Houghton
Mifflin.
Thorndike, R. 1., Hagen, E. P., & Sattler, J. M. (1986). Guide for administering and scoring the
fourth edition. Stanford-Binet intelligence scale. Chicago: Riverside.
Tjossem, T. (1976). Intervention strategies for high risk infants and young children. Bal-
timore: University Park Press.
Touwen, B. C. 1. (1973). The neurological development of the human infant. In D. Davis & j.
Dobbing (Eds.)' Scientific foundations of pediatrics. Leiden: University Press.
Touwen, B. C. 1. (1978). Variability and stereotyping in normal and deviant development. In
J. Apley (Ed.), Care of the handicapped child. Clinics in Developmental Medicine, No.
67 (pp. 99-110). London: Heinemannn.
Ulrey, G. (1981). Psychological evaluation. In W. K. Frankenburg, S. M. Thornton, & M. E.
Cohrs (Eds.), Pediatric developmental diagnosis (pp. 175-183). New York: Thieme-
Stratton.
Ulrey, G. (1982). Influences of infant behavior on assessment. In G. Ulrey & S. J. Rogers
(Eds.), Psychological assessment of handicapped infants and young children (pp. 14-
24). New York: Thieme-Stratton.
Ulrey, G., & Schnell, R. R. (1982). Introduction to assessing young children. In G. Ulrey & S. j.
Rogers (Eds.), Psychological assessment of handicapped infants and young children
(pp. 1-13). New York: Thieme-Stratton.
Vohr, B. R., & Hack, M. (1982). Developmental follow-up of low-birthweight infants. Pedi-
atric Clinics of North America, 29, 1441-1454.
Wolf, A. W., & Lozoff, B. (1985). A clinically interpretable method for analyzing the Bayley
Infant Behavior Record. Journal of Pediatric Psychology, 10, 199-214.
Yang, R. K. (1979). Early infant assessment: An overview. In J. D. Osofsky (Ed.), Handbook of
infant development (pp. 165-184). New York: Wiley.
10
Questions of Developmental
Neurolinguistic Assessment
MICHAEL A. CRARY, KY]TA K. S. VOELLER,
and NANCY]. HAAK
INTRODUCTION
It is obvious upon the most casual inspection that children develop lan-
guage gradually and that there is an apparent temporal sequence to the
acquisition of linguistic skills. A basic premise of this chapter is that the
order of events of language acquisition reflects, or more pointedly, is
dependent upon, a normal sequence of neurological maturation. This
seems to be an almost obvious supposition. Yet the clinical evaluation of
communicative abilities in children has not, to date, subscribed to a de-
velopmental neurolinguistic philosophy. In the following pages the ques-
tions of the utility of a developmental neurolinguistic approach to speech/
language assessment will be addressed. It would be preliminary to offer a
comprehensive model given present knowledge. Nonetheless, the major
purpose is to propose the possibility of such an approach and to identify
areas for additional study.
An oversimplified description of language development might in-
clude the sequence from babbling to single words, to word combinations,
to increasingly complex grammatic expressions. Additional descriptive
subdivisions might include the separation of comprehension of spoken
language from the ability to express verbal concepts. There are, however,
multiple skill areas that have potential influence upon a developing lan-
guage system. Included among these attributes are motor performance,
MICHAEL A. CRARY. KYJTA K. S. VOELLER. and NANCY J. HAAK Departments of

Communicative Disorders and Neurology. University of Florida. Gainesville. Florida.
249
250 MICHAEL A. CRARY et 01.
memory, and attending skills. The necessity to evaluate multiple behav-

iors is complicated further by the fact that the various components change
during the course of development. Both of these observations address the
need for a dynamic, multidimensional emphasis in any proposed model
of developmental neurolinguistic assessment.
The neurological portion of a neurolinguistic evaluation is no less
challenging than the linguistic portion. Present approaches to commu-
nicative assessment typically begin with descriptions of language behav-
iors, usually derived from an adult model of language. An attempt is then
made to connect these descriptions with underlying psychological func-
tions. The dynamic nature of the developing system(s) is usually over-
looked. In this chapter, developmental aspects of language behavior and
anatomical features of neurological maturation will be synthesized to for-
mulate a neurolinguistic systems model of language development. In this
respect, our proposed model of developmental neurolinguistic functions
will be based upon facts of neuroanatomical maturation and knowledge of
the relative contribution of various developing neuroanatomical systems
to the acquisition of linguistic and paralinguistic skills.
Present approaches to language assessment typically address a devel-
opmental performance-level model. In this approach, an individual's per-
formance is compared with normative data from groups of children in a
similar age range. It is not the intent of this chapter to advocate a massive
revision in test procedures, but rather to propose that the interpretation of
language test results be referenced to an underlying neuropsychological
model of linguistic functions. Such an approach would give equal impor-
tance to process and product. Language assessment from this perspective
would be an invaluable asset in the interpretation of other neuropsycho-
logical test results. Rather than assigning clinical significance to a set of
test scores derived solely from a normative basis, a pattern of results
would be identified in conjunction with a biological model as well. Using
this strategy, both functional and nonfunctional neuropsychological pro-
cesses would be identified. In short, it is proposed that a common explan-
atory, neurological model be developed that applies to language as well as
other psychological functions.
BASICS OF LANGUAGE DEVELOPMENT
In formulating an approach to developmental neurolinguistic assess-

ment, it is necessary first to specify the significant parameters of language
to be accounted for. An explosion of psycholinguistic research in recent
years has identified numerous aspects of linguistic development. There
have been, however, only a few longitudinal studies and these have evalu-
QUESTIONS OF NEUROLINGUISTIC ASSESSMENT 251
ated limited numbers of children (e.g., R. Brown, 1973). These studies

have described developmental changes that children demonstrate as they
acquire the rudimentary skills of verbal language. An additional body of
information has focused upon the period of language readiness or pre-
linguistic stage (e.g., Bates, Benigni, Bretherton, Camaioni, & Volterra,
1979; Halliday, 1975). Work by these and other authors has helped to fill
the developmental gap between birth and the formal use of verbal lan-
guage. From these descriptions of prelinguistic and linguistic develop-
ment, three general stages within the developmental continuum can be
constructed that include a pre linguistic stage, a stage of lexical expansion,
and a stage of grammatical expansion.
Prelinguistic Communication
The prelinguistic stage extends from birth to approximately 12
months. It is during this time that the child is developing rapidly in a
variety of skill areas, with the most obvious being motor skills, social
development, and finally the emergence of spoken language. Early behav-
iors lack intention and might be characterized as reflexive. At this stage,
the infant is able to communicate basic biological and social needs via a
rudimentary system of movements and vocalizations. Language, as the
adult recognizes it, has not yet emerged. However, language foundations
are being constructed during this time. For example, there is increasing
evidence that infants have a remarkable ability to perceive certain aspects
of human communication. Neonates will show excitation to selective peo-
ple, voices, or objects (Alegria & Noirot, 1978). Also Lieberman (1967) has
pointed out that children as young as 3 months will change the pitch of
the voice during vocal play depending on the adult with whom they are
interacting. Voice pitch is higher if the adult is a female rather than a
male. This would indicate that the infant has recognition of certain para-
linguistic parameters and is able to exert some control over vocal produc-
tions at this early age.
The child's repertoire of preintentional behaviors expands until at
about 8 months of age, when the first signs of intentional behaviors are
recognized (Bates, 1975). At this time the infant becomes more persistent
in communicative signaling. Behaviors give the adult a clear indication of
whether responses are appropriate or not, and vocal patterns become
increasingly varied. During this stage the interactive system of gestures,
facial expressions, and vocalizations becomes so rich that it has been
called a "protolanguage" (Halliday, 1975). This pattern of complex behav-
iors expands and leads to the onset of first words early in the second year
of life. The emergence of productions recognized by adults as meaningful
words signals the beginning of the lexical stage of language development.
Lexical Expansion Stage

For children and adults words are referents for objects or events oc-
curring in the environment. However, to suggest that a child's word has
the same implications as that of the adult would be misleading. From a
linguistic perspective, the stage of lexical growth is characterized by in-
creased semantic (meaning) and phonological knowledge (sound organi-
zation and production ability). Like the prelinguistic stage, this stage of
lexical expansion may be divided into two components. Bloom (1973)
and Moskowitz (1980) have suggested that the first 50 words represent
broad concepts that serve as independent units of language (holophrases).
These productions have been referred to as idioms, and it has been pro-
posed that they emerge outside of the existing phonological system. In
support of this contention is Leopold's (1947/1970) depiction of his
daughter's use of the word pretty. Hildegard used this word with adultlike
pronounciation before 1 year of age. By the age of 22 months her produc-
tion of the word had deteriorated to a point where it was consistent with
her general phonological ability. These "advanced forms" may represent
the automatic production of a semantic concept that has not yet been
integrated fully into the child's linguistic system.
Clark (1973) explains growth in semantic knowledge as a process of
semantic feature acquisition. This position states that, although a child
may use an adult word-form, the child does not possess a complete adult
representation of the meaning of that word. The process of acquiring
semantic knowledge is a result of the child assigning more features to
the lexical representation of the verbal concept. These "features" are pre-
sented as the child's representation of encoded perceptual information.
An example offered by Clark involves the word doggie. Initially, this label
may apply to any four-legged animal. As additional features are ac-
quired-e.g., hoofs, horns, says moo-the child begins to subdivide the
larger category or concept into units more closely aligned with adult
representation. Acquisition of semantic knowledge then is a function of
the child's analyzing the environment and applying increasingly complex
feature contrasts to a general set of verbal labels.
Phonological growth is an additional characteristic of this lexical
stage of language development. Phonologists also subdivide this stage.
Ingram (1976) discusses the phonology of the first 50 words and subse-
quently the phonology of the simple morpheme. The early stage of pho-
nological growth is limited and tends to coincide with initial vocabulary
acquisition. It is at this time that phonological growth may be subservient
to semantic growth (i.e., advanced forms). Jakobson (1968) described the
process of phonological growth as the acquisition of a system of feature
contrasts. The child learns contrasts between various sounds of language
on the basis of differences in meaning that result from using different
sounds. The description of phonological acquisition during this period is

similar to the semantic feature acquisition position presented by Clark
(1973). Perhaps at this early stage of verbal language growth the child's
task is to extract and analyze meaningful perceptions from the environ-
ment, and subsequently to attach linguistic significance to these percep-
tions in terms of semantic and phonological features.
Prosodic variation appears to be an additional device of linguistic
contrast during this period of development. As stated above, children are
responsive to prosodic aspects of spoken language as early as 3 months of
age. During the later part of the stage of lexical expansion children report-
edly use prosodic variation to signal linguistic distinctions. Menyuk and
Bernholtz (1969) reported that children between 18 and 20 months of age
used prosodic variation to signal statements, questions, or emphasis. This
variation appears to coincide with the emergence of multi word utter-
ances. Multiword expressions typically emerge at about 18 months of age;
however, single-word utterances are dominant in child speech until ap-
proximately 2 to 2% years of age.
The period of growth from 18 months to 2 years represents the second
division of the stage of lexical expansion. During this period linguistic
changes include a sudden growth in vocabulary, rapid phonological ex-
pansion, and the emergence of multiword utterances. The presence of
these growth spurts signals the onset of our next general stage of language
development-growth in grammatical complexity.
Grammatical Expansion Stage

The emergence of word combinations follows the single-word stage
and occurs at about 18 months of age. By 2 years of age the child is
demonstrating systematic and consistent increases in the length and com-
plexity of verbal productions. This begins the stage we refer to as the
grammatical expansion stage. Chronologically this is the longest stage of
language development. Limitations in syntactic knowledge have been
demonstrated in children ranging from 5 to 9 years (Chomsky, 1969).
During the years from 2 to 5 the child is demonstrating dramatic
growth in all aspects of grammar. Phonological productions become in-
creasingly complex and approximate adult performance by age 6 (Sander,
1972). Systematic increases in the length and complexity of spoken utter-
ances also are observed. R. Brown (1973) has designated five stages of
grammatical performance based upon mean length of utterance during
this period. His first stage is similar to the end of the lexical expansion
stage. The next stage is characterized by the emergence of grammatical
morphemes. Subsequent stages are characterized by increasing length and
linguistic complexity of spoken utterances. If one uses Brown's observa-
254 MICHAEL A. CRARY et a1.
tions as a guideline, the onset of the grammatic expansion stage tends to

be characterized by a significant increase in multiword utterances and the
emergence of morphological grammatical markers. Although the specifics
of this stage of grammatical expansion are beyond the scope of this chap-
ter, the central concepts include an increase in the sequential complexity
of word order manifested in increased length and syntactic complexity.
Also, at this time, the basic phonological system reaches a state of relative
maturity.
A COMMENT ON STYLES OF LANGUAGE PROCESSING
Although the order of acquisition of language structures and func-

tions follows a general pattern, there is considerable variability in the rate
and style of language learning among children. Variations in styles of
language learning have been dichotomized by terms such as "referential
vs. expressive" (Nelson, 1973) and "analytic vs. gestalt" (Peters, 1977,
1983). However, as Peters (1983) points out, these labels represent the
extreme ends of a continuum of learning strategies used during language
development. From a simplified perspective, gestalt/expressive children
tend to be more verbal, producing a greater number of multiword utter-
ances at an early age. Analyticlreferential children tend to be more con-
servative, typically speaking in one-word units. Bates (1979) has sug-
gested that the temporal order of emerging cognitive skills might
contribute to differences in language-learning strategies. Peters (1983) re-
views factors that she feels may influence the strategies used by a particu-
lar child and suggests that the child's "individual neurologic endow-
ment" affects the style of language learning (p. 21). The implication of
these observations is that variations in the organization of neurolinguistic
systems may relate to, or even predict, the style(s) of language learning
employed by an individual child.
RELA TION OF NEUROLOGICAL MATURATION TO

ACQUISITION OF LANGUAGE SKILLS
Neuroanatomic Maturation
There are at least two perspectives regarding maturation of human
cortex and the development of functional systems. Traditionally, a hier-
archial model of cortical development has been proposed. On the basis of
indices such as brain weight, dendritic proliferation, and myelination
(Altman, 1967; Conel 1939-1963; Flechsig, 1901; Yakovlev & Lecours,
1967), this model has proposed that the primary sensory and motor pro-
jection areas are the first to mature, followed in sequence by secondary

(projection-association) areas, and finally by tertiary (association-associa-
tion) areas.
A typical example of neurological maturation from this perspective is
the degree of dendritic proliferation noted in various anatomic sub-
systems. DeCrinis (cited by Lenneberg, 1967) reported different patterns
of dendritic growth for different cortical areas. The first areas to reach
dendrogenetic maturity are the primary projection areas of vision and
audition and the pre- and post-Rolandic strips for sensory and motor
functions. The next areas to mature are the secondary association areas
that receive input from the projection areas. These areas mature at approx-
imately 2 years of age. The final stage of maturation begins between the
ages of 2 and 4 years. During this time period, parts of the frontal and
parietal lobes begin to mature. As with the other cortical regions, primary
sensory and motor areas develop first, followed by their peripheral asso-
ciation areas. Regions of interarea association are the last to develop. The
primary projection areas develop independently and are connected only
when the final stages of this rapid dendrogenesis are complete. This pat-
tern also has been reported by Poliakov (1961).
Recent studies of brain growth in both primates and humans have
provided an alternative model for describing cortical development. Rakic,
Bourgeois, Eckenhoff, Zecevic, and Goldman-Rakic (1986) examined pre-
and postnatal synaptogenesis in five areas of macaque cerebral cortex.
Although they expected to identify differential growth patterns support-
ing a hierarchical model, they instead identified more homogeneous syn-
aptogenesis in all areas and layers of cortex examined. During the last 2
months of gestation, synaptic density increased at a rapid rate in all cor-
tical areas examined, reaching a level equivalent to that of the sexually
mature adult macaque. Synaptic density increased to levels much higher
than that of the adult during the first 4 postnatal months. Subsequently, a
reduction of synapses was noted that was rapid at first and then more
gradual. This study suggests that the macaque brain develops as a whole
rather than hierarchically and implies that functional subsystems may be
an outgrowth of synaptic "pruning" postnatally. Huttenlocher (1979) has
demonstrated a similar pattern of synaptic overproduction in the human
cortex. The tempting implication of these observations is that in both the
human child and the macaque, functional maturation may result from
synapse elimination facilitating the development of subsystems to medi-
ate more refined behaviors.
Jason Brown (1979) has suggested a similar model of neurological
maturation in explanation of language functions and disorders. He re-
views findings concerning neurological maturation in both human and
nonhuman species that "are in agreement with the concept of an evolu-
tionary and maturational progression from a diffuse to a focal organiza-
tion ... " (p. 143). He contends that focal cortical areas develop out of
generalized neocortex. This phylogenetic approach proposed by Brown
appears to share points of agreement with the recent findings of Rakic et
a1. (1986). The resulting model appears to be the converse of the more
traditional hierarchical model. However, both views recognize the post-
natal development of cortical functional subsystems. In this respect, the
two views have similar implications for language development. Both
would predict that early behaviors would reflect more diffuse/nonfocal
processes and that development of more refined/analytic behaviors
would result from maturation of specific subsystems. Thus, although
there are questions concerning the precise nature of neurological matura-
tion relevant to language development, the functional interpretations may
be similar at a simplified behavioral level.
Hemispheric Functions
Another question concerns the role of the right versus the left hemi-
sphere in the development of neurolinguistic functions. One approach to
this topic is to evaluate the effects of early unilateral lesions on subse-
quent language development. Until the early 1970s, it was accepted wide-
ly that left hemisphere lesions early in development had little impact on
subsequent language acquisition. This position was based on the concept
of plasticity or the equipotentiality of the two cerebral hemispheres to
process language in the young child. It was assumed that if a child sus-
tained left hemisphere damage, the right hemisphere would take over
these functions. Lenneberg (1967) stated that lateralization of language to
the left hemisphere was a function of development and that plasticity
continued until adolescence. However, the concept of plasticity rested on
two assumptions and several clinical observations that have not weath-
ered the test of time.
The first premise was that the immature brain had a remarkable re-
silience to early lesions (Kennard, 1938). Acceptance of the fact that spe-
cific early lesions resulted in impaired functioning at maturity in experi-
mental animals did not occur until the 1970s, following the work of
Goldman and colleagues (Goldman, 1971, Miller, Goldman, & Rosvold,
1973). Working with macaques, these investigators demonstrated that the
timing and location of focal brain lesions was critical in determining the
extent of behavioral deficits. Early prenatal lesions resulted in little func-
tional impairment whereas later lesions (infancy) tended to create pro-
found deficits. Also, the appearance of some behavioral deficits was tem-
porally distant from the lesion. For example, lesions of dorsolateral
frontal cortex did not result in any immediate impairment in a delayed
response task. However, at approximately 16 months of age, the time at
which this cortical area is believed to become functionally active in the
macaque, behavioral limitations became obvious and increased in sever-

ity as the animal matured. Lesion location interacted with timing in pro-
ducing functional impairments. Specifically, these investigators demon-
strated that subcortical lesions (caudate) produced in infancy resulted in
more severe deficits than cortical lesions occuring in the same time
period.
The second premise was that there is no anatomical difference be-
tween the cerebral hemispheres. Once it was established that there was a
difference in the size of the right and left planum temporale (Geschwind &
Levitsky, 1968), and that these asymmetries were present in the fetus and
infant (Chi, Dooling, & Gilles, 1977), the morphological basis of plasticity
was undermined. Physiological studies revealing hemispheric asymme-
tries for many tasks in young children provided further contradictions to
the concept of equipotentiality (Witelson, 1977).
Proponents of plasticity frequently relied upon Basser's (1962)
clinical observations of 102 hemiplegics. Basser reported that, following
left or right hemispherectomies, "speech was developed and maintained
in the intact hemisphere and in this respect the left and right hemispheres
were equipotential" (p. 451). However, Robinson (1981) has pointed out
that the majority of Basser's subjects were either retarded or uncoopera-
tive and untestable. Robinson concluded that "careful reading of his find-
ings in the 30 children whose hemiplegia occurred after the onset of
speech does show that those having a right hemiplegia had a greater
impairment in language than those with a left hemiplegia ... " (p. 380,
italics added). Kiessling, Denckla, and Carlton (1983) evaluated language
outcome in right and left hemiplegic children who had onset of hemi-
plegia in the pre- or perinatal period. The right hemiplegic group per-
formed poorly on tasks requiring higher-level syntactic awareness and
repetition of semantically coherent materials. Dennis and Whitaker's
studies (1975, 1976) of hemispherectomized infants also indicated that
subjects with left hemispherectomies had residual deficits in processing
linguistic information, especially when it was conveyed by syntactic
conventions.
From these reports, the impression is formed that the left hemisphere
has a special role in mediating certain language functions from birth on-
ward. However, lesions in the right hemisphere of very young infants may
cause residual language problems, and left hemisphere lesions in very
young children do not preclude language development. It should be
pointed out, however, that subtle linguistic deficits may persist in such
cases even when the onset occurs very early (Alajouanine & L'Hermitte,
1965; Woods & Carey, 1979). Vargha-Khadem, O'Gorman, and Watters
(1983) compared the residual effects of left versus right hemisphere le-
sions acquired at different ages (prenatal, birth to 5 years, and 5 to 14
years). They found no significant deficits in their language measures from
258 MICHAEL A. CRARY et al.
the right hemisphere group, but they did find a significant age effect
among subjects in the left hemisphere group. Children who incurred
damage later in life presented more severe residual deficits than children
who sustained early damage. Yet, even in those cases of prenatal left
hemisphere lesions, subtle linguistic deficits persisted.
The collective impression from reports such as those reviewed here is
that a modified form of plasticity may apply to the developing neu-
rolinguistic system. Although it appears that the left hemisphere has a
special function in language processing from birth, the right hemisphere
also is capable of linguistic processing, albeit not as efficiently. With
development, the left hemisphere becomes more dominant in the control
of linguistic functions. Given these observations, a better understanding
of developmental neurolinguistic functions will require information re-
garding the role of the right hemisphere in language processing and the
development of subsystems within the left hemisphere to process differ-
ent types of language tasks.
Contributions of the Right Hemisphere

Studies with adults have demonstrated a significant contribution
from the right hemisphere (RH) to language processing. Zaidel (1979,
1985) reviewed data from split-brain and lesion cases and summarized
many of the linguistic capabilities of the right hemisphere. In adults, the
RH has substantial semantic comprehension abilities. Notably, it pos-
sesses a rich lexical knowledge. The RH also has the ability for auditory
discrimination, though this skill deteriorates in noise. This pattern may
represent an extention of lexical identification. The RH also has the abil-
ity to perform rudimentary syntactic analyses. This ability is limited by
increases in the length (limited auditory-verbal short-term memory) and
syntactic complexity of verbal stimuli. Zaidel's conclusions indicated
that the RH contributes to the semantic analysis of language but has little
skill for syntactic or phonological aspects.
Zaidel (1985) pointed out that the adult RH has the linguistic ability
of a child between 3 and 6 years of age. This finding was taken to suggest
that the RH is highly active in language acquisition up to this point in
time but that the left hemisphere becomes more dominant thereafter. Con-
sidering the normal progression of language development, this apparent
shift in hemispheric "dominance" may be a phenomenon of the demands
of a maturing linguistic system in cooperation with specific neurological
growth or change. As the child enters the stage of grammatical expansion,
analytic and sequential processing demands of language increase dramat-
ically. Apparently, the left hemisphere is better suited to perform the
types of processing required by more complex grammatic language.
Additional language or paralanguage functions of the RH include
pragmatic aspects of language and prosodic abilities. Foldi, Cicone, and

Gardner (1983) reported that the adult with RH damage has reduced abili-
ties for non literal interpretation and reduced ability to comprehend nar-
ratives. In essence, the RH-damaged patient has adequate linguistic in-
terpretation but cannot use contextual information to extract indirect
meanings from communications. These investigators suggested that the
RH is important for the integration of cues to form a single concept (ge-
stalt). Production and comprehension of prosodic features also are associ-
ated with RH functioning. Ross (1981) described dysprosodias susequent
to RH damage in a manner similar to descriptions of aphasias resulting
from LH damage. Although there has been little systematic assessment of
prosodic abilities in children, Voeller (1986) has observed prosodic devia-
tions in 9 of 15 children who had sustained RH damage.
This brief summary of RH language functions suggests that the right
hemisphere may have an active role in language acquisition and subse-
quent language use. The semantic and prosodic abilities of the young
infant seem to parallel processing properties designated to the right hemi-
sphere. However, this observation raises questions of early left hemi-
sphere functions and potential developmental interactions between the
two hemispheres. Most present knowledge of left hemisphere neurolin-
guistic function comes from studies of adults who present language im-
pairment subsequent to hemispheric damage.
Left Hemisphere Contributions-Neurolinguistic Subsystems

In adult patients, lesions to different areas of the left hemisphere (LH)
are known to result in differential impairment to language processes. A
simple method of classifying lesion effects on language is with reference
to comprehension, expression, and repetition functions. For example, le-
sions involving lateral frontal-parietal cortex typically result in a marked
reduction in expressive abilities with a less severe reduction in com-
prehension (Broca aphasia). Damage to posterior temporal cortex typ-
ically devastates auditory comprehension but does not interfere with the
motoric aspects of speech production (Wernicke aphasia). Lesions that
spare perisylvian cortex also spare repetition skills (transcortical
aphasias). Whitaker (1971) has formalized observations in this respect
into a neurolinguistic representation of the Central Language System.
Syntactic/phonological skills are subserved via Broca's area, the arcuate
fasciculus, and the anterior portion of Wernicke's area. Semantic func-
tions are subserved by posterior Wernicke's area and portions of the pos-
terior temporal and inferior parietal lobes. Lexical abilities are diffuse
without a specific anatomic "home." However, work by Cappa, Cavotti,
and Vignolo (1981) has suggested the presence of a lexical-semantic area
in posterior temporal-parietal cortex.
In addition to the traditional cortical representations of adult lan-

guage functions, recent investigations have highlighted subcortical influ-
ences. Damage to the thalamus (Mohr, Watters, & Duncan, 1975) and the
capsular/striatal system (Naeser et a1., 1982) has been shown to produce
dysphasias in adults that parallel, but have characteristics distinct from,
cortical dysphasia. Although the linguistic significance of these cortical
and subcortical mechanisms is not completely understood, they do con-
tribute toward a framework from which to study neurolinguistic functions
in adults.
Unfortnuately, language symptoms reported in children with ac-
quired aphasia do not fit a predictable model like those of their adult
counterparts. This observation may be due, in part, to the lack of a specific
neurolinguistic assessment perspective in those studies that have evalu-
ated children with acquired aphasia. Reports of childhood aphasia typ-
ically depict language symptoms as initial muteness followed by non-
fluent, telegraphic speech with relatively spared comprehension. There
are few reports of posterior LH language symptoms such as paraphasias,
fluent speech, and reduced auditory comprehension (Satz & Bullard-
Bates, 1981). If auditory comprehension is impaired, it is typically tran-
sient (e.g., Hecaen, 1976), or present in children above 9 or 10 years of age
(VanDongen, Loonen, & VanDongen, 1985). These reported patterns of
linguistic impairment in childhood-acquired aphasia apparently are not
helpful in proposing an LH subsystems model of neurolinguistic func-
tions for children.
There is a separate, larger group of language-impaired children, how-
ever, who do present different profiles of linguistic impairment at an early
age. Examination of subtypes of development language disorder (DLD)
may be useful in formulating developmental neurolinguistic subsystems
of the LH. Many investigators have proposed different subtypes of DLD
from various theoretical positions. A major stumbling block to cross-study
interpretation is that different investigators have used different popula-
tions and different test protocols to classify subgroups of DLD children.
Yet there are often commonalities in the resultant subgroups. For exam-
ple, Aram and Nation (1975) used standardized clinical tests to examine
formulation (expression), repetition, and comprehension of phonological,
syntactic, and semantic aspects of language functions. They identified six
subgroups of children representing varying combinations of abilities and
deficits for the selected tasks. Rapin and Allen (1983) grouped children on
the basis of clinical observation, longitudinal results of observation, and
clinical testing into seven subgroups. Their criteria for grouping were
based upon observations of children's expressive language, interactive
behavior, and apparent comprehension. Despite dramatic differences in
the techniques of evaluation and data analysis, these two studies did
report commonalities in some of their subgroups. There appear to exist
robust subgroups of DLD children with impaired phonological and syn-
tactic abilities, but spared comprehension of spoken language. Also, there

are children who present a more global deficit of language functions in-
volving both comprehension and expression of all language functions. A
somewhat different perspective has been offered by Denckla (1981), who
studied language impairment in learning-disabled children. Denckla
found six subgroups of language impairment on the basis of evaluation of
vocabulary, digit span, memory, and naming tasks. Some of these sub-
types were topographically similar to those identified by Aram and Na-
tion (1975). Global impairments were noted as well as more selective
impairments of naming, repetition, and comprehension. Also, both of
these studies identified a subgroup of language-impaired children with
spared repetition ability.
Differences in subject selection, tests employed, and method of data
analysis preclude direct comparison of most studies. However, it is
important to note two facts. First, despite methodological differences,
various studies have identified similar profiles of language impairment in
subgroups of DLD children. Second, collectively, the various studies indi-
cate a range of possible language-impairment profiles that approximates
that seen in adult aphasia. This observation implicates the presence of
developing neurolinguistic subsystems in children who have reached the
stage of grammatical expansion. Available information appears to support
this implication. Frisch and Handler (1974) compared neuropsychologi-
cal profiles (Reitan-Indiana) among normal children and two groups of
language-impaired children. One of their language-impaired subgroups
presented profiles similar to those of adults with LH post-Rolandic le-
sions. The clinical language profiles of these children would be similar to
what Rapin and Allen (1983) call their "phonologic-syntactic syndrome."
Such reports support the existence of LH neurolinguistic subsystems in
children and suggest that once children reach the stage of grammatic
expansion they have the basic components of adult neurolinguistic sub-
systems. Subsequent neurolinguistic development may be a function of
neurological maturation of these subsystems, permitting the gradual in-
crease in grammatical performance that is observed in the child from 2
years onward. If this proves to be a valid assumption, then developmental
neurolinguistic assessment during this time must incorporate considera-
tions for the comprehension, repetition, and expression of phonological,
syntactic, and semantic aspects of language.
LANGUAGE ASSESSMENT AT PRESENT-WHAT'S COVERED,

WHAT'S NOT?
The first step in language assessment is to determine what the goal of
the evaluation is to be. Many assessment measures are descriptive in
nature and designed to assess the performance level of an individual
compared with standardized performance data. If the assessment goal is
to portray language strengths/deficits as part of a more global profile, then

standardized developmental protocols should be employed. If, on the
other hand, a major purpose of the assessment is to identify specific areas
of skill or impairment to be addressed in a treatment program, then stan-
dardized batteries only begin the task. The practitioner subsequently must
employ more qualitative protocols/strategies in order to identify specific
intervention targets. A complete assessment battery should include both
quantitative and qualitative evaluation. In the following pages a prelimi-
nary description is provided for how currently available clinical protocols
may be utilized in a developmental neurolinguistic assessment approach.
In evaluating proposed developmental neurolinguistic abilities, both
the functions and the content of language must be assessed. The following
review of currently available protocols is not meant to be exhaustive.
Rather, the intent is to present a representation of what is or is not readily
available to assess various language functions and content from a neu-
rolinguistic perspective. Obviously, assessment of prelinguistic commu-
nicative functions cannot be approached in the same manner as linguistic
abilities. As such, only brief commentary on prelinguistic assessment is
warranted.
Prelinguistic Assessment (0 to 12+ Months)

Communicative assessment of the prelinguistic child (both reflexive
and intentional) typically is accomplished through application of so-
called readiness scales (Thorum, 1981). These may include both parent
interview and direct observation of infant behaviors in a variety of low-
structure playlike situations. Such scales typically address a range of
developmental skill areas, including motor, self-help, emotional, cog-
nitive, and communicative. The general goal of such instruments is to
identify global developmental delays or, in the older infant, to assess
specific cognitive or language-readiness skills that are thought to be
important for subsequent growth in communicative ability. Examples of
these scales include the Receptive-Expressive Emergent Language Scale
(Bzoch & League, 1971) and the Denver Developmental Screening Test
(Frankenburg, Dodds, Fandal, Kazuk, & Cohrs, 1975). Specific commu-
nicative functions in the young child may be described using observa-
tional guidelines such as those published by Bates (1975) and Halliday
(1975). There are few procedures designed to assess specific neurolinguis-
tic functions. However, available research suggests directions for develop-
ment of new procedures. Given the early maturational lead of the right
hemisphere, such protocols may do well to address early communicative
functions associated with right hemisphere attributes. For example, face-
and voice-recognition procedures, and procedures for evaluating percep-
tion and possibly imitation of simple prosodic features, may be of benefit
in generating specific neurolinguistic predictions regarding the child's

developing linguistic system. These assessments would be most applica-
ble to the child who has reached the intentional stage of language devel-
opment. By using the global developmental surveys to sample a range of
developing behaviors, and with subsequent focus on specific commu-
nicative functions (possibly RH in nature), practitioners may find in-
creased ability to predict impending linguistic impairment in the pre-
linguistic infant.
Lexical Expansion Assessment (12 to 24+ Months)

As the title of this stage implies, assessment of children in this devel-
opmental range (12-24 months) should address semantic growth and im-
pending phonological development. As with the earlier stage, there are no
standardized tests of language abilities for this young age range. In this
respect, many of the readiness scales used for the prelinguistic child also
are employed here. Since children in the lexical expansion stage are ex-
pected to demonstrate emerging verbal skills, nonverbal scales of lan-
guage readiness will address only a portion of developing skills in these
children. However, several mutlidimensional inventories of early lan-
guage are available that "sample" a variety of receptive and expressive
language functions. Typically, these focus upon early aspects of social
communication. The goal of these protocols typically is to arrive at a
developmental performance level for receptive versus expressive lan-
guage functions.
At this stage of growth, it is suspected that the right hemisphere may
be actively involved in language processing. As such, an evaluation from
a neurolinguistic perspective should include detailed assessment of se-
mantic knowledge, pragmatic functions, and prosodic recognition and
production (possibly for both affective and linguistic prosody). Semantic
knowledge may be evaluated informally by observing the child's use of
single words or word combinations to express concepts. There are no
standard estimates of semantic knowledge; however, repeated evaluations
will provide the practitioner with information regarding growth in vocab-
ulary and the child's acquisition of semantic features. As word combina-
tions emerge, calculations of the average utterance length and vocabulary
growth estimation will provide useful information regarding semantic
growth. Pragmatic functions also may be evaluated by observing how the
child uses verbal language to manipulate the environment (e.g., request-
ing, commenting, directing). Although no standard assessments exist in
this respect, several authors have proposed formal assessments of prag-
matic functions of language (e.g., Bates, 1975; Halliday, 1975). Assess-
ment of prosodic abilities also would seem to be an important aspect of
the lexical expansion stage. Menyuk and Bernholtz (1969) reported that
children near the end of this stage used prosodic information to signal
grammatical distinctions in expressive language. Also, there has been
speculation that "normal" prosodic abilities in this age range are ex-
tremely important to subsequent grammatical growth (Crystal 1979; Shad-
den, Asp, Tonkovich, & Mason, 1980). However, there are no formal as-
sessment approaches for prosodic recognition or production at this stage
of development. Practitioners may be able to infer information regarding
prosodic abilities by evaluating how the child responds to different affec-
tive voices (happy, sad, angry) or by using linguistic prosody to signal
grammatical distinctions (questions, emphasis).
Another aspect of linguistic growth is the rapid expansion of pho-
nological abilities that occurs near the end of the stage of lexical expan-
sion. Although the child's sound inventory begins to expand during this
time, the majority of phonological growth coincides with syntactic devel-
opment during the stage of grammatic expansion.
Grammatical Expansion Assessment (2 to 4+ years)

The stage of grammatic expansion is marked by dramatic increases in
phonology and syntax along with continuing growth in semantic, prag-
matic, and prosodic abilities. As mentioned previously, a multitude of
tests and procedures are available for evaluation of phonological and
syntactic abilities. Semantic abilities in this stage are assessed primarily
in association with syntactic skills (it is difficult if not impossible to
separate these two aspects) and/or with reference to word knowledge and
word use. Given the rapid growth in verbal language skills during this
period, the practitioner also is confronted with consideration of para-
language variables that may influence a child's performance on linguistic
tasks. Major focus is given to motor, memory, and attending abilities. The
growth of grammatical functions is believed to signal the emergence of the
left hemisphere as the "dominant" language hemisphere. Within the left
hemisphere there appear to be functional subsystems serving the com-
prehension, expression (formulation), and repetition of the language con-
tent areas. From available knowledge, naming functions appear to be
bihemispheric, or at least not localized to a specific area.
Comprehension of Spoken Language

As discussed earlier, both the right and left hemispheres have the
ability to comprehend basic linguistic information. The RH, however, is
not as sophisticated as the LH, being limited by constraints on short-term
memory and specific grammatical knowledge. From this perspective, we
would expect that overreliance on RH comprehension strategies and/or
deficient LH comprehension abilities would result in language com-
prehension abilities that differ significantly from those of the normal
child. In this respect, it would be necessary to compare a child's com-

prehension of different language content areas. There are a variety of
standardized procedures that permit such comparison. For example, the
Test for Auditory Comprehension of Language-Revised (Carrow, 1985)
contains three subtests addressing word classes and relationships, gram-
matical morphemes, and elaborated sentences. It also is possible to evalu-
ate comprehension of various language content areas by using combina-
tions of tests and/or subtests from different protocols. For example, word
knowledge might be assessed by the Peabody Picture Vocabulary Test-
Revised (Dunn & Dunn, 1981), knowledge of semantic relations might be
assessed by the Processing Word Classes Subtests of the Clinical Evalua-
tion of Language Functions (Semel & Wiig, 1980), comprehension of spe-
cific syntactic structures might be assessed by the receptive portion of the
Northwestern Syntax Screening Test (Lee, 1971), and comprehension of
spoken instructions with increasing length and complexity might be as-
sessed by the Token Test for Children (DiSimoni, 1978). The important
issue to address is that the comprehension of spoken language may be
differentially impaired across language content areas. It is this pattern of
differential impairment that permits the practitioner to begin building an
explanatory profile of specific neurolinguistic impairment. The assess-
ment strategy is to employ standardized performance tests to ascertain the
extent of comprehension deficit, and subsequently to evaluate multiple
content areas to delineate the pattern of deficits.
Language Expression/Formulation
Expressive language functions would appear to be predominately
within the domain of the left hemisphere, even in the young child. As
discussed before, children with expressive language deficits in the pres-
ence of relatively good comprehension typically present with both pho-
nological and syntactic deficits. Again, there are multiple clinical pro-
cedures available to assess these language areas. Standardized
phonological assessments typically follow a developmental sound acquisi-
tion orientation, and test for the production of various speech sounds in
different positions within a word. Examples of common speech-articula-
tion tests include the Templin-Darley Tests of Articulation (Templin &
Darley, 1969) and the Goldman-Fristoe Test of Articulation (Goldman &
Fristoe, 1969). Recent application of linguistic theory has emphasized a
rule-governed or pattern analysis over the more traditional one-sound-at-a-
time approach. Although no standardized protocols exist for these analy-
ses, formal descriptive procedures are available (e.g., Hodson, 1980;
Shriberg & Kwiatkowski, 1980). Using both of these approaches permits the
practitioner to establish a developmental performance level for pho-
nological ability and to describe major error patterns.
There are numerous procedures available for the evaluation of syn-
tactic formulation. Specific grammatical morphemes are evaluated in pro-

cedures such as the Grammatic Closure Subtest of the Illinois Test of
Psycho linguistic Abilities (Kirk, McCarthy, & Kirk, 1968) and the Gram-
matic Completion Subtest of the Test of Language Development (Hammill
& Newcomer, 1982). Sentence-formulation skills are assessed in a variety
of methods by different protocols. The expressive portion of the North-
western Syntax Screening Test (Lee, 1971) evaluates syntactic construc-
tions of increasing complexity in a delayed imitation format. The Produc-
ing Model Sentences Subtest of the Clinical Evaluation of Language
Functions (Semel & Wiig, 1980) provides the child with a stimulus word
to be used in a sentence generated by the child. The Carrow Elicited
Language Inventory (Carrow, 1974) requires the child to imitate 52 model
utterances. Each of these procedures is an example of a performance level
assessment of syntactic formulation abilities. Beyond this quantitative
analysis, it might be beneficial to evaluate specific syntactic structures
used or misused by a particular child. A variety of formal descriptive
approaches are available. Miller's (1981) Simple Sentence Analysis pro-
vides a format for the analysis of syntactic structure used by children in a
conversational sample. Lee (1974) presented an attempt at quantifying a
descriptive analysis of conversational language samples in her Develop-
mental Sentence Scoring Procedure. A short and simple way to estimate
the grammatical complexity of an utterance is the calculation of mean
length of utterances (e.g., R. Brown, 1973).
The collective utilization of procedures evaluating different content
areas of language expression from both a performance level and a descrip-
tive perspective will facilitate a more complete depiction of any ex-
pressive language deficits noted in the child. As is the case with com-
prehension, it may be that the pattern of phonological and/or syntactic
errors has relevance to a neurolinguistic explanation of language deficits.
Unfortunately, this is a poorly investigated area of developmental lan-
guage impairment in children.
Repetition
Although many clinical protocols use repetition in one way or an-
other, the significance of strengths or deficits in this communicative func-
tion has not been explained fully. It would be tempting, for example, to
speculate that those children studied by Aram and Nation (1975) or
Denckla (1981) who demonstrated good repetition skills represented a
developmental transcortical neurolinguistic impairment. However, such
an assumption would be highly speculative. Use of repetition tasks has
been criticized from the perspective that such tasks overestimate (Marat-
sos & Kuczaj, 1974; Smith, 1970) or underestimate (Prutting, Gallagher, &
Mulac, 1975) language knowledge. It is possible that comparison of lan-
guage content areas expressed in repetition versus conversation may be a

differentiating factor in identifying operational neurolinguistic subsys-
tems in the child in the same way that this comparison is applied to adults
with acquired language deficits.
Naming
In both the child and the adult, naming is probably a bihemispheric
linguistic task. Thus, the identification of quantitative deficits on naming
tasks such as the Boston Naming Test (Kaplan, Goodglass, & Weintraub,
1983) or the Expressive One-Word Vocabulary Test (M. F. Gardner, 1979)
may not be useful in differentiating neurolinguistic deficits. Differential
patterns of naming impairment across tasks may be more informative in
identifying a pattern of selective neurolinguistic deficits.
Pragmatic Language Functions

The child's use of language functions in the social environment may
provide an indication of right hemisphere contributions to language defi-
cits. However, the evaluation of pragmatic functions is a relatively recent
focus of linguistic assessment in children, and few or no formal protocols
exist. Some authors (e.g., Thorum, 1981) consider the use of existing
behavioral scales that contain social and interpersonal measures (Le.,
Adaptive Behavior Scales, Nihira, Foster, Shelhas, & Leland, 1974). In
combination with assessment of other pragmatic functions, such as the
ability to extract inferential meanings, social irregularities may have
important diagnostic indications for neurolinguistic assessment. Wiig
(Wiig, 1982a, 1982b; Wiig & Bray, 1983) has developed a series of check-
lists for evaluating pragmatic language functions in children 3 years of age
and up. For younger children, observation of verbal interactions in a
natural setting is recommended. For older children, elicited tasks are
employed. Though useful, such checklists focus primarily upon social
communicative skills, with little emphasis on other RH pragmatic func-
tions (e.g., indirect meaning, narrative interpretation). Perhaps, as more
developmental information regarding pragmatic function/dysfunction is
available, more diverse and systematic assessment protocols will emerge.
Prosody
Assessment of prosodic abilities is an additional area that has re-
ceived limited attention in child populations. The Seashore Tests of Mu-
sical Ability (Seashore, Lewis, & Saetveit, 1960) evaluates the percep-
tion/discrimination of multiple prosodic features. This procedure is,
however, designed for children 9 years of age or older. There has been
recent clinical research with younger speech-language-impaired children

demonstrating deficits in production of prosodic features. Crary and Tall-
man (1985) found that children with severe phonological disorders, con-
sidered to represent verbal dyspraxia, produced prosodic patterns similar
to adults with speech dyspraxia secondary to LH damage. The recent
increase in prosodic studies with adult patients (Ross, 1981) may provide
neurolinguistic models and subsequent assessment procedures that will
make it possible to study prosodic abilities in language-impaired children
systematicall y.
Paralanguage Measures
Three paralanguage areas that may affect a child's performance on
language tasks include motor ability, memory, and attending skill. Motor
speech ability may affect growth in the child's phonological system.
Motor speech performance may be assessed by visual inspection of the
speech mechanism and by tests of oral-facial praxis and speech di-
adochokinesis. There is no formal protocol for assessing facial praxis in
the child. Most practitioners use modifications of procedures recom-
mended for adult patients. Motor speech abilities are typically assessed
with speech diadochokinetic tasks. In this regard, Fletcher (1978) has
published performance data from 6 years to 13 years, and Riley (Riley &
Riley, 1985) has provided limited normative data down to 4 years of age.
The child with reduced short-term memory ability may present im-
paired repetition functions as well as deficient performance in other lan-
guage areas. There are several memory assessment procedures available.
For example, subtests of the Detroit Test of Learning Aptitude (Hammill,
1985) evaluate auditory retention of unrelated words and oral directions.
In addition, digit span memory may be assessed using the appropriate
subtest from the Illinois Test of Psycho linguistic Ability (Kirk, et 01.,
1968).
Impaired attending ability has been postulated to contribute to re-
duced comprehension performance in certain brain-damaged language-
impaired children (Campbell & McNeil, 1985). Yet mechanisms of atten-
tion, their clinical assessment, and their potential contributions to appar-
ent language dysfunction have received limited investigation. Aside from
skilled observations regarding a child's attending ability, there are few
available tests that attempt to estimate the potential influence of attention
deficits on language performance. Two procedures that evaluate aspects
of "auditory attention" are the Goldman-Fristoe-Woodcock Selective At-
tention Test (Goldman, Fristoe, & Woodcock, 1974) and the Flowers-Cos-
tello Tests of Central Auditory Abilities (Flowers, Costello, & Small,
1970). Additional aspects of attention such as concentration or sustained
attention may be inferred from tasks such as digit span (forward and
backward) or other counting and spelling tasks (e.g., Lezak, 1983). Can-
cellation tasks, visual closure tasks, and selected subtests from the
Wechsler Intelligence Scale for Children-Revised (Arithmetic, Coding,
Mazes) also have been suggested for the evaluation of concentration
(Gardner, 1979). Gardner (1979) further describes tasks of auditory or
visual vigilance (sustained attention) that might be useful for evaluation
of attention in language-impaired children. Barkley (Chapter 6, this vol-
ume) provides a superb overview of the assessment of attention in
children.
Summary of Present Assessment Capabilities

Our brief review identified examples of available protocols or strat-
egies for evaluating language functions for the three general stages of
development. An attempt also was made to identify the gaps in available
clinical assessment protocols. At present, there is no battery of tests that
has been developed or organized from a developmental neurolinguistic
perspective. However, it is possible to form testable inferences about neu-
rolinguistic functions in language-impaired children through the use of
currently available protocols supplemented with specific clinical obser-
vations and "qualitative" assessment. A basic approach to assessment of
the language-impaired child should include use of standardized perfor-
mance tests supplemented by more descriptive procedures, particularly
in identified deficit areas. It is important not only to profile basic language
functions (comprehension, formulation, repetition, naming) but also to
profile abilities in various language content areas (phonology, syntax,
semantics, pragmatics). There are multiple clinical procedures available
for children across the developmental span in most aspects of language
assessment. However, significant gaps exist in the systematic assessment
of pragmatic and prosodic functions. Because these communicative as-
pects may rely heavily upon right hemisphere functions, it would seem
important to pursue clinical research leading to the development of valid
assessment protocols in these areas. Finally, in addition to a basic com-
munication assessment, it would be valuable to consider a variety of
paralanguage factors such as motor, memory, and attention abilities, that
may influence a child's linguistic performance.
GUIDELINES AND QUESTIONS REGARDING

DEVELOPMENTAL NEUROLINGUISTIC ASSESSMENT
In designing a neurolinguistic assessment approach, considerations

must be given not only to language functions (e.g., comprehension, nam-
ing) but also to language content (e.g., syntax, morphology, pragmatics).
Additionally, procedures evaluating paralanguage factors must be in-

cluded (e.g., motor speech, attending, memory). A simplified summation
of these factors is presented in Table 1. With a few notable exceptions
there are multiple clinical procedures available for evaluating these fac-
tors in the developing child. However, considerable diversity in the the-
oretical basis, method of assessment, test construction, and scoring pro-
cedures has been noted for tests within the same function/content area
and age range (Thorum, 1981). In addition, various psychometric con-
cerns have been expressed for many protocols, including lack of reported
predictive validity, interexaminer reliability, and test-retest reliability
(McCauley & Swisher, 1984; Wiig & Semel, 1984). Finally, most available
procedures test a specific language function, such as naming or auditory
comprehension. The few "batteries" that are available measure an array of
content and functions; however, the theoretical basis of measurement
may change from one task to the next (Spekman & Roth, 1984).
The fact that no developmental neurolinguistic assessment protocol
exists at present does not imply that such an approach to language assess-
ment in children cannot be developed. Rather, the absence of this approach
to language assessment appears to be the result of gaps in knowledge and
unanswered questions. In this chapter, consistencies between language
development and neurological maturation have been discussed. Further,
an attempt was made to explain growth in language functions with refer-
ence to the roles of the respective cerebral hemispheres and the develop-
ment of subsystems within the left hemisphere. Table 2 presents a summa-
ry model of neurolinguistic development. Although this model is
admittedly incomplete in its explanation of neurolinguistic functions, it
does permit a systematic approach to questions that must be addressed in
TABLE 1. Summary of Language Functions, Language Content Areas,

and Paralanguage Performance Factors
Paralanguage
Language performance
Language functions content factors
Comprehension Pragmatics Motor speech

Dysarthria
Dyspraxia
Expression (formulation) Semantics
Repetition Syntax Memory functions
Naming Morphology Attending abilities
Speech produc- { PhOnOlOgy}
tion and-
prosody Phonetics
TABLE 2. Depiction of Preliminary Model of Developmental Neurolinguistic Functions
Hemispheric Language
Age range Anatomic subsystem relations stage Language mode Representative skills
0-8 months Primary projection Primarily subcortical Prelinguistic Reflexive (pre intent) 1. ~R crying and cooing
(sensory and motor) control and/or dif- 2. Voice and face
fuse cortical sys- recognition
terns 3. Prosodic recognition and
limitation
8-12 months Primary projection Right> left Pre linguistic Intent 1. Increased intentional
communication
Growth in corpus (pragmatics)
callosum 2. Onset of first words
Growth in projection
association system
1-2 years Projection-association Right> left Lexical Semantic (gestalt) 1. Increased vocabulary
or 2. Preliminary phonological
Right = left development
3. Initial word combina-
tions
2-4+ years Association-associa- Left> right Grammatic Sequential! analytic 1. Phonological growth
tion 2. Syntactic growth
3. Continued growth in all
language areas
developing a more comprehensive model and subsequent clinical ap-

proaches.
Areas that must be addressed in subsequent research include the
consideration of normal aspects of language development, evaluation of
neurological dysfunction as it may relate to specific linguistic impair-
ment, and the development of valid clinical neurolinguistic assessment
strategies. With respect to normal aspects of language development, there
are obvious questions regarding relationships between structure and func-
tion at various points along the developmental continuum. Peters (1977,
1983) has asserted that different language-processing strategies can be
identified both within and across normal children. Furthermore, she
points out that children with language-learning difficulties may differ
from normal children in the repertoire of strategies useful to them. For
example, Weeks (1974) described a child who had limited strategies for
learning language but who was adept at using other learning strategies in
compensation of language limitations. These observations raise questions
pertaining to the issues surrounding neurological maturation and plas-
ticity. As was pointed out, the hierarchical and holistic views on cortical
maturation appear to reach similar ends in the development of functional
subsystems. However, the learning strategies available to the child prior
to the functional activation of these subsystems would be expected to vary
according to specific aspects of neurological maturation. Also, disruption
of the anticipated pattern of maturation at various points in development
may lead to different compensatory processes available to the child. To
address these issues adequately, focus must be given to the processes of
language development in relation to neurological maturation. Unfortu-
nately, a great deal of the work currently available addresses product and
not process.
In evaluating potential relationships between structure and function
in the developing central nervous system, it will be mandatory to have
accurate descriptions of structural alterations in those individuals with
language impairments. However, until recently, studies of acquired apha-
sia in children referenced hemispheric involvement by physical symp-
toms (i.e., right or left hemiplegia). There are several limitations to those
studies lacking adequate control for describing areas of brain damage. For
example, if the brain injury has occurred pre- or perinatally, diffuse
damage secondary to hypoxia is likely. Also, at any age, it will be difficult
to control for the possibility of bihemispheric lesions without objective
measurement procedures. Delineation of the extent and location of any
lesion will require at least application of one of the imaging techniques
currently available. Studies that undertake longitudinal evaluation of
children with identified brain lesions will be invaluable in defining de-
veloping neurolinguistic systems.
Related to the issue of lesion localization is the issue of cortical ver-
sus subcortical lesions. Knowledge of the differential effects of these le-

sions at different points in development would help to clarify the respec-
tive roles of cortical versus subcortical systems in language acquisition. It
is known that specific subcortical lesions may result in more severe func-
tional impairments than cortical lesions in the young macaque. Also,
language impairments subsequent to subcortical lesions have been re-
ported in adults (Naeser et al., 1982). Aram, Rose, Rekate, and Whitaker
(1983) have reported two instances of aphasia in older children secondary
to capsular/striatal lesions. Aside from these reports, little is known of the
potential roles of subcortical mechanisms in language development and
disorders.
The development of valid developmental neurolinguistic assessment
tools must be an outgrowth of future advances in the aforementioned
areas. Most language assessment protocols currently available are ori-
ented toward a developmental level approach. In general, these protocols
describe the end result (product) of language processes, but they do not
address the processes themselves. Some protocols incorporate tasks from
a particular psycholinguistic perspective. However, it is often unclear
how the various tasks fit into the specified model or whether the model
has useful clinical application. Practitioners attempting a neurolinguistic
interpretation from the results of currently available protocols will find a
combination of quantitive and qualitative procedures useful in formulat-
ing testable clinical hypotheses. Yet significant gaps in areas such as
prosodic and pragmatic functions in combination with the traditional
focus on product rather than process will limit the ability to make specific
conclusions regarding neurolinguistic abilities. To fill these gaps, re-
searchers and clinicians will have to address normal aspects of para-
language functions as they influence language performance in the devel-
oping child. Protocols that attempt to identify underlying processing
strategies used by children of different ages also will be helpful. Such
research must incorporate large numbers of children and be longitudinal,
not cross-sectional. Resulting explanations must be dynamic and multidi-
mensional to account for potential variability in the interaction among
developing cognitive processes (e.g., Bates, 1979).
A Concluding Example
Crary (1984) has attempted an initial neurolinguistic interpretation of
one group of developmental communication deficits. In evaluating rela-
tionships between motor and speech-language limitations, he has pro-
posed a continuum of impairments under the label of developmental
verbal dyspraxia. Children with more "executive" dysfunctions would
present difficulties in most motor and speech tasks, whereas children
with deficits more toward the "planning" end of the continuum would
present difficulties in sequential activity with relatively spared executive

functions. The proposed neurolinguistic basis for this deficit involves
subsystems of the left hemisphere with dysfunction in either frontotem-
poral regions, temporoparietal regions, or both. Though this position has
not been substantiated unequivocally, preliminary supportive evidence
has been generated.
The language profiles of these children demonstrated strengths in
semantic processing in both comprehension and expression tasks, but
pronounced deficits in phonology and syntax. This observation led to the
hypothesis that such children are functioning primarily in the lexical
stage of language development with better gestalt/simultaneous than ana-
lytic/sequential abilities. In evaluating motor ability, an experimental test
of facial mimicry revealed deficits associated with the sequential com-
plexity of the task. To evaluate the hypothesis that sequential processing
was impaired, six children were administered the Kaufman Assessment
Battery for Children (Kaufman & Kaufman, 1983). This procedure com-
pares performance on tasks designed to assess simultaneous versus se-
quential processing skills. The mean scale score (Mean = 10, Standard
Deviation = 3) for simultaneous subtests was 9.5 whereas that for sequen-
tial subtests was 8.2. This was not a great difference, but it was in the
predicted direction. Four of these children underwent central auditory
testing using dichotic procedures in an attempt to identify hemispheric
asymmetries. Two children demonstrated a strong left ear advantage (po-
tentialleft hemisphere deficit). The remaining two demonstrated right ear
advantages; however, their scores for the right ear (left hemisphere) were
outside of 2 standard deviations of expected normal performance.
Although these observations require replication with larger samples,
they support our hypothesis that some children with developmental ver-
bal dyspraxia have left hemisphere dysfunction specifically related to
sequential aspects of information planning. These cases may represent
children who are "stuck" between the lexical and grammatical stages of
linguistic development. Work must continue to address this hypothesis
and to formulate neurolinguistic arguments that might help to identify
functioning or dysfunctioning subsystems that are limiting the develop-
ment of the grammatical stage. It is obvious that linguistic assessment or
neuropsychological assessment in isolation would be limited in reaching
this goal. A collaboration of orientations is necessary.
REFERENCES
Alajouanine, T. H., & L'Hermitte, F. (1965). Acquired aphasia in children. Brain, BB, 635-
662.
Alegria, J., & Noirot, E. (197B). Neonate orientation behavior towards the human voice.
International Journal of Behavioral Development, 1, 291-312.
Altman, J. (1967). Postnatal growth and differentiation of the mammalian brain, with im-
plications for a morphological theory of memory. In C. Quarton, T. Melnechuk, & F.
Schmitt (Eds.), The neurosciences, New York: Rockefeller University Press.
Aram, D. M., & Nation, J. E. (1975). Patterns of language behavior in children with develop-
mental language disorders. Journal of Speech and Hearing Research, 18, 229-241.
Aram, D. M., Rose, D. F., Rekate, H. L., & Whitaker, H. A. (1983). Acquired capsular/striatal
aphasia in childhood. Archives of Neurology, 40, 614-617.
Basser, L. S. (1962). Hemiplegia of early onset and the faculty of speech with special refer-
ence to the effects of hemispherectomy. Brain, 85, 427-460.
Bates, E. (1975). Acquisition of performatives prior to speech. Merrill-Palmer Quarterly, 21,
205-226.
Bates, E. (1979). The biology of symbols: Some concluding thoughts. In E. Bates, L. Benigni,
I. Bretherton, L. Camaioni, & V. Volterra (Eds.), The emergence of symbols: Cognition
and communication in infancy. New York: Academic Press.
Bates, E., Benigni, L., Bretherton, I., Camaioni, L., & Volterra, V. (1979). The emergence of
symbols: Cognition and communication in infancy. New York: Academic Press.
Bloom, L. (1973). One word at a time: The use of single-word utterances before syntax. The
Hague: Mouton.
Brown, J. W. (1979). Language representation in the brain. In H. D. Steklis & M. J. Raliegh
(Eds.), Neurobiology of social communication in primates: An evolutionary perspective
(pp. 133-195). New York: Academic Press.
Brown, R (1973). A first language: The early stages. Cambridge: Harvard University Press.
Bzoch, K. R, & League, R (1971). Receptive-expressive emergent language scale. Gaines-
ville, FL: Tree of Life Press.
Campbell, T. F., & McNeil, M. R (1985). Effects of presentation rate and divided attention on
auditory comprehension in children with an acquired language disorder. Journal of
Speech and Hearing Research, 28, 513-520.
Cappa, S., Cavotti, G., & Vignolo, L. A. (1981). Phonemic and lexical errors in fluent aphasia:
Correlation with lesion site. Neuropsychologia, 19, 171-177.
Carrow, E. (1974). Carrow Elicited Language Inventory. Austin, TX: Learning Concepts.
Carrow, E. (1985). Test for Auditory Comprehension of Language-Revised. Boston: Teaching
Resources.
Chi, J. D., Dooling, E. C., & Gilles, F. H. (1977). Left-right asymmetries of the temporal speech
areas of the human fetus. Archives of Neurology, 34, 346-348.
Chomsky, C. (1969). The acquisition of syntax in children from 5 to 10. Cambridge, MA:
M.I.T. Press.
Clark, E. V. (1973). What's in a word: On the child's acquisition of semantics in his first
language. In T. Moore (Ed.), Cognitive development and the acquisition of language.
Conel, J. L. (1939-1963). Postnatal development of the human cerebral cortex (Vols. 1-6).
Cambridge: Harvard University Press.
Crary, M. A. (1984). A neurolinguistic perspective of developmental verbal dyspraxia. Com-
municative Disorders, 9, 33-49.
Crary, M. A., & Tallman, V. L. (1985, November). Production of propositional prosodic
features by speech-language normal and speech-language disordered children. Paper
presented at the American Speech-Language-Hearing Association Annual Convention,
Washington, DC.
Crystal, D. (1979). Prosodic development. In P. Fletcher & M. Garman (Eds.), Language
acquisition. Cambridge: Cambridge University Press.
Denckla, M. B. (1981). Minimal brain dysfunction and dyslexia: Beyond diagnosis by exclu-
sion. In M. E. Blair, I. Rapin, & M. Kinsbourne (Eds.), Child neurology. New York:
Spectrum.
Dennis, M., & Whitaker, H. (1975). Hemispheric equipotentiality and language acquisition.
In S. Segalowitz & F. Gruber (Eds.), Language development and neurological theory.

Dennis. M . & Whitaker. H. A. (1976). Language acquisition following hemi-decortication:
Linguistic superiority of the left over right hemisphere. Brain and Language. 3. 404-
433.
DiSimoni. F. (1978). The token test for children. Hingham. MA: Teaching Resources.
Dunn. L. M .. & Dunn. L. M. (1981). Peabody Picture Vocabulary Test-Revised. Circle Pines.
MN: American Guidance Service.
Flechsig. P. (1901). Developmental (myelogenetic) localization of the cortex in human sub-
jects. Lancet. October. 1027-1029.
Fletcher. S. G. (1978). The Fletcher Time-by-Count Test of Diadochokinetic Syllable Rate.
Tigard. C. C. Publications.
Flowers. A.. Costello. M .. & Small. V. (1970). Flowers-Costello Test of Central Auditory
Abilities. Dearborn. MI: Perceptual Learning Systems.
Foldi, N. S . Cicone. M .. & Gardner. H. (1983). Pragmatic aspects of communication in brain-
damaged patients. In S. J. Segalowitz (Ed.). Language functions and brain organization
(pp. 51-86). New York: Academic Press.
Frankenburg. W. K. Dodds. J. B.. Fandal, A. W .. Kazuk. E.. & Cohrs. M. (1975). Denver
Developmental Screening Test. Denver: LADOCA Project & Publishing Foundation.
Frisch. G.. & Handler. L. (1974). A neuropsychological investigation of functional disorders
of speech articulation. Journal of Speech and Hearing Research. 17. 432-445.
Gardner. M. F. (1979). The Expressive One-Word Vocabulary Test. Novato. CA: Academic
Therapy Publications.
Gardner. R. A. (1979). The objective diagnosis of minimal brain dysfunction. Cresskill. NJ:
Creative Therapeutics.
Geschwind. N .. & Levitsky. W. (1968). Human brain left-right asymmetries in temporal
speech region. Science. 161. 186-187.
Goldman. P. S. (1971). Functional development of the prefrontal cortex in the infrahuman
primate. In C. L. Ludlow & M. E. Doran-Quine (Eds.). The neurological bases of language
disorders in children: Methods and direction for research (NINCDS Monograph (No 22,
NIH Publication No. 79-440). Bethesda: U.S. Department of Health. Education and
Welfare.
Goldman. R.. & Fristoe. M. (1969). Goldman-Fristoe Test of Articulation. Circle Pines. MN:
American Guidance Service.
Goldman. R.. Fristoe. M . & Woodcock. R. (1974). The Goldman-Fristoe-Woodcock Test of
Auditory Discrimination. Circle Pines. MN: American Guidance Service.
Halliday. M. A. K. (1975). Learning how to mean: Explorations in the development of
language. London: Edward Arnold.
Hammill. D. D. (1985). Detroit Tests of Learning Aptitude (DTLA-2). Austin. TX: Pro-Ed.
Hammill. D. D.. & Newcomer. P. L. (1982). Test of Language Development. Austin. TX: Pro-
Ed.
Hecaen. H. (1976). Acquired aphasia in children and the ontogenesis of hemispheric func-
tional specialization. Brain and Language. 3. 114-134.
Hodson. B. (1980). The assessment of phonological processes. Danville. IL: Interstate.
Huttenlocher. P. R. (1979). Synaptic density in human frontal cortex: Developmental
changes and effects of aging. Brain Research. 163. 195-205.
Ingram. D. (1976). Phonological disability in children. New York: American Elsevier.
Jakobson. R. (1968). Child language. aphasia and phonological universals. The Hague:
Mouton. (Original work published. 1941)
Kaplan. E.. Goodglass. H .. & Weintraub. S. (1983). The Boston Naming Test. Philadelphia:
Lea & Febiger.
Kaufman. A. S . & Kaufman. N. L. (1983). Kaufman Assessment Battery for Children. Circle
Pines. MN: American Guidance Service.
Kennard, M. A. (1938). Reorganization of motor function in the cerebral cortex of monkeys

deprived of motor and premotor areas in infancy. Journal of Neurophysiology, 1,477-
496.
Kiessling. L. S., Denckla, M. B., & Carlton, M. (1983). Evidence for differential hemispheric
function in children with hemiplegic cerebral palsy. Developmental Medicine and
Kirk, S. A., McCarthy, J. J., & Kirk, W. D. (1968). Illinois Test of Psycholinguistic Ability (Rev.
ed.). Urbana: University of Illinois Press.
Lee, L. L. (1971). Northwestern Syntax Screening Test. Evanston, IL: Northwestern Univer-
sity Press.
Lee, L. L. (1974). Developmental sentence analysis. Evanston, IL: Northwestern University
Press.
Lenneberg, E. H. (1967). Biological foundations of language. New York: Wiley, M.LT. Press.
Leopold, W. F. (1947). Speech development of a billingual child: A linguists' record (Vo1. 2).
Sound learning in the first two years. Evanston, IL: Northwestern University Press.
(Reprinted by AMS press, New York, 1970)
Lezak, M. D. (1983). Neuropsychological assessment (2nd ed.). New York: Oxford University
Press.
Lieberman, P. (1967). Intonation, perception and language. Cambridge, MA: M.LT. Press.
Maratsos, M., & Kuczaj, S., (1974). Evidence from elicited imitation for productive compe-
tence in a grammatical system. Papers and Reports of Child Language Development.
Palo Alto, CA: Stanford University.
McCauley, R. J., & Swisher, L. (1984). Psychometric review of language and articulation tests
for preschool children. Journal of Speech and Hearing Disorders, 49, 34-42.
Menyuk, P., & Bernholts, N. (1969). Prosodic features and children's language. Quarterly
Progress Report of Research Laboratory of Electronics (Vol. 93, pp. 216-219). Cam-
bridge, MA: M.LT. Press.
Miller, E. A., Goldman, P. S., & Rosvold, H. E. (1973). Delayed recovery of function following
orbital prefrontal lesions in infant monkeys. Science, 182, 304-306.
Miller, J. (1981). Assessing language production in children. Baltimore: University Park
Press.
Mohr, J., Watters, W., & Duncan, G. (1975). Thalamic hemorrhage and aphasia. Brain and
Language, 2, 3-17.
Moskowitz, B. A. (1980). Idioms in phonology acquisition and phonological change. Journal
of Phonetics, 8, 69-83.
Naeser, M., Alexander, M., Helm-Estabrooks, N., Levine, H., Laughlin, S., & Geschwind, N.
(1982). Aphasia with predominantly subcortical lesion sites. Archives of Neurology, 39,
2-14.
Nelson, K. (1973). Structure and strategy in learning to talk. Monographs of the Society for
Research in Child Development, 38.
Nihira, N., Foster, R., Shelhas, M., & Leland, H. (1974). Adaptive behavior scales. Wash-
ington, DC: American Association of Mental Deficiency.
Peters, A. M. (1977). Language learning strategies: Does the whole equal the sum of the parts?
Language, 53, 560-573.
Peters, A. M. (1983). The units of language acquisition. Cambridge: Cambridge University
Press.
Poliakov, G. 1. (1961). Some results of research into the development of the neuronal struc-
ture of the cortical ends of the analyzers in man. Journal of Comparative Neurology, 117,
197.
Prutting, c., Gallagher, T., & Mulac, A. (1975). The expressive portion of the N.S.S.T. com-
pared to a spontaneous language sample. Journal of Speech and Hearing Disorders, 40,
40-49.
Rakic, R., Bourgeois, J. P., Eckenhoff, M. R., Zecevic, N., & Goldman-Rakic, P. S. (1986).
Concurrent overproduction of synapses in diverse regions of the primate cerebral cortex.

Science, 232, 232-235.
Rapin, I., & Allen, D. A. (1983). Developmental language disorders: Nosologic considera-
tions. In U. Kirk (Ed.), Neuropsychology of language, reading and spelling (pp. 213-
219). New York: Academic Press.
Riley, G., & Riley, J. (1985). Oral motor assessment and treatment: Improving syllable pro-
duction. Tigard, C.C. Publications.
Robinson, R O. (1981). Equal recovery in child and adult brain? Developmental Medicine
and Child Neurology. 23, 379-383.
Ross, E. D., (1981). The aprosodias. Archives of Neurology, 38, 561-569.
Sander, E. K. (1972). When are speech sounds learned. Journal of Speech and Hearing
Disorders, 37, 55-61.
Satz, P., & Bullard-Bates, C. (1981). Acquired aphasia in children. In M. T. Sarno (Ed.),
Acquired aphasia (pp. 399-426). New York: Academic Press.
Seashore, c., Lewis, D., & Saetveit, J. (1960). Seashore Measures of Musical Talents. New
York: Psychological Association.
Semel, E. M., & Wiig, E. H. (1980). Clinical evaluation of language functions. Columbus, OH:
Charles E. Merrill.
Shadden, B. B., Asp, C. W., Tonkovich, I. D., & Mason, D. (1980). Imitation of suprasegmental
patterns of five-year-old children with adequate and inadequate articulation. Journal of
Speech and Hearing Disorders, 45, 390-400.
Shriberg, L., & Kwiatkowski, J. (1980). Natural process analysis. New York: Wiley.
Smith, C. S. (1970). An experimental approach to children's linguistic competence. In J. R
Hayes (Ed.), Cognition and the development of language. New York: Wiley.
Spekman, N., & Roth, F. (1984). Clinical Evaluation of Language Functions (CELF) Diag-
nostic Battery: An analysis and critique. Journal of Speech and Hearing Disorders, 49,
97-100.
Templin, M., & Darley, F. (1969). Templin-Darley Tests of Articulation, (2nd ed.). Iowa City:
Bureau of Ed. Res. and Services, University of Iowa.
Thorum, A. R (1981). Language assessment instruments: Infancy through adulthood.
Springfield, IL: Charles C Thomas.
VanDongen, H. R, Loonen, M. C. B., VanDongen, K. J. (1985). Anatomical basis for acquired
fluent aphasia in children. Annals of Neurology, 17, 306-309.
Vargha-Khadem, F., O'Gorman, A., & Watters, G. V. (1983, February). Aphasia in children
with prenatal versus postnatal left hemisphere lesions: A clinical and CT scan study.
Paper presented at the International Neuropsychological Society Annual Convention,
Mexico City, Mexico.
Voeller, K. K. S., (1986). Right hemisphere deficit syndrome in children. American Journal
of Psychiatry, 143, 1004-1009.
Weeks, T. (1974). The slow speech development of a bright child. Lexington, MA: D. C.
Heath.
Whitaker, H. A. (1971). Neurolinguistics. In W. O. Dingwall (Ed.), A survey of linguistic
science. College Park, MD: Linguistic Program, University of Maryland.
Wiig, E. H. (1982a). Let's talk: Developing prosocial communication skills. Columbus, OH:
Charles E. Merrill.
Wiig, E. H., (1982b). Let's talk inventory for adolescents. Columbus, OH: Charles E. Merrill.
Wiig, E. H., & Bray, C. M. (1983). Let's talk for children. Columbus, OH: Charles E. Merrill.
Wiig, E. H., & Semel, E. (1984). Language assessment and intervention for the learning
disabled. Columbus, OH: Charles E. Merrill.
Witelson, S. F. (1977). Early hemisphere specialization and interhemisphere plasticity. In S.
J. Segalowitz & F. A. Gruber (Eds.), Language development and neurological theory (pp.
Woods, B. T., & Carey, S. (1979). Language deficits after apparent clinical recovery from
childhood aphasia. Annals of Neurology, 6, 405-409.
Yakovlev, P. 1., & Lecours, A. R. (1967). The myelogenetic cycles of regional maturation of
the brain. In A. Minkowski (Ed.), Regional development of the brain in early life (pp. 3-
70). Oxford: Blackwell.
Zaidel, E. (1979). The split and half brain as models of congenital language disability. In C. L.
Ludlow & M. E. Doran-Quine (Eds.), The neurological basis of language disorders in
children: Methods and directions for research (NINCDS Monograph 22, pp. 55-89).
Washington, DC: U.S. Government Printing Office.
Zaidel. E. (1985). Language in the right hemisphere. In D. F. Benson & E. Zaidel (Eds.), The
dual brain: Hemispheric specialization in humans. New York: Guilford Press.
11
Learning Disabilities Subtypes

Perspectives and Methodological Issues in
Clinical Assessment
GEORGE W. HYND, ROBERT T. CONNOR,
and NAOMI NIEVES
INTRODUCTION
In a relatively young industrial society such as ours, only about 100 years
have elapsed since an inability to learn, retain, and use a symbolic means
of communication fluently has been viewed as a learning disability. Prior
to the end of the 19th century, the acquisition of reading, arithmetical,
and other more demanding cognitive skills was not viewed as essential for
a productive life. These abilities are clearly essential today.
Considerable evolution has taken place in how learning disabilities
are conceptualized. Approximately a century ago they were considered
only insofar as reading abilities were affected. Today, serious deficits
cutting across many different domains such as arithmetic or spelling may
result in a diagnosis of learning disability. Even in the historically impor-
tant domain of reading, a variety of studies suggest the existence of many
subtypes or subpopulations of disabled readers. Further articulation of
these subtypes, more well-controlled research studies, and technologi-
cally sophisticated investigations that relate deficient regions of brain
activity (e.g., metabolic) to disordered cognitive ability will probably im-
pact on new conceptualizations of learning disabilities and their
assessment.
GEORGE W. HYND Departments of Educational Psychology and Psychology, University
of Georgia, Athens, Georgia, and Department of Neurology, Medical College of Georgia,
Augusta, Georgia. ROBERT T. CONNOR and NAOMI NIEVES Kennedy Institute,
Johns Hopkins School of Medicine, Baltimore, Maryland.
281
282 GEORGE W. HYND et 01.
Within this context, the purpose of this chapter is to provide some

integrative perspective as to how learning disabilities were viewed and
diagnosed historically, provide a current critique of subtyping meth-
odologies, and relate assessment practices in clinical child neuropsychol-
ogy to a neurolinguistic model of cognitive processes known to be impor-
tant in learning. It should be emphasized that the primary focus will be on
reading disabilities since most research efforts have been directed toward
this most common of all learning disabilities. In this fashion, the unique
clinical problems associated with neurodevelopmentallearning disorders
and their diagnosis may be addressed productively.
EARL Y CONCEPTUALIZATIONS OF LEARNING DISABILITIES
Clinical Case Reports and Observations

Early investigators were struck by the variable visual-perceptual defi-
ciencies seen in patients who had suffered neurological trauma, es-
pecially as it related to reading ability. Although the highly variable
assessment practices and reporting styles among these investigators were
acknowledged as problematic even at that time (Bastian, 1898), these
early reports showed an evolving realization of what behaviors charac-
terized learning disabilities.
For example, as early as 1877 Kussmaul used the term "word blind-
ness" to describe a patient who, although not blind, was unable to read
words. Subsequently, Hinshelwood and others (Bastian, 1898; Claiborne,
1906; Fisher, 1905; Hinshelwood, 1895, 1900; Jackson, 1906; Morgan,
1896; Stephenson, 1905) published similar cases involving word blind-
ness and related neurological disorders. The reporting of these early cases
was obviously important since these cases of acquired learning or behav-
ioral disabilities set the stage for the awareness that learning disabilities
could also manifest as a neurodevelopmental or congenital disorder.
Hinshelwood (1900) reported two cases of what he referred to as
"congenital" word blindness. The first, a boy of 11 years, had been unable
to identify highly familiar words after 4 years of instruction. His auditory
memory, however, was excellent; he had been able to learn his lessons
without reading the words. He was contrasted with a boy of 10 years who
could recognize familiar words but had difficulty with uncommon short
words (e.g., tub, rug). Additional case studies were rapidly forthcoming
and, in reviewing them, one can see the developing notions regarding
differential manifestations according to gender, processes involved, and
possible neuroanatomical correlates.
Fisher (1905), for instance, reported yet another case of a girl with
reading disabilities. After 2 years of instruction she persisted in mixing
SUBTYPES OF LEARNING DISABILITIES 283
her letters, displayed poor arithmetic skills, and misnamed musical notes.
Family history revealed that her uncle had reading problems as a young-
ster and still experienced difficulty spelling as an adult. Fisher described
this case as evidence of a familial condition of impairment in the left
angular gyrus; "the visual memory centre for words." These notions con-
cerning the possible congenital or genetic etiology continued to attract
support, as evidenced by Stephenson's (1905) case of congenital word
blindness affecting three generations.
Thus, by 1905 it was already reasonably well established that (1)
severe learning disabilities were manifested in children of normal intel-
ligence (although no standardized measure of IQ was yet generally avail-
able), (2) they affected a greater number of boys than girls, (3) the deficits
were manifested variably, impacting differentially on each child's ability
to read or perform some cognitive task, (4) the disability was not generally
responsive to traditional learning opportunities, (5) it seemed to have a
genetic component that expressed itself variably, (6) it could be diagnosed
using many different informal and formal clinical assessment practices,
and (7) it might be related to some neurodevelopmental pathology in the
region of the angular gyrus in the left hemisphere. With the benefit of
hindsight, it is remarkable how sophisticated and accurate were the obser-
vations made by these early clinicians.
What's in a Name?
By the early part of this century, many different terms had been
suggested in attempts to define this syndrome accurately. A representa-
tive few include amblyopia, amnesia visualis verbalis, bradylexia, cogeni-
tal alexia, congenital word blindness, and later, primary reading retarda-
tion, strephosymbolia, and developmental dyslexia (Drew, 1956; Hynd &
Cohen, 1983; Pirozzolo, 1979). Each of the investigators who introduced
these terms offered his or her own rationale.
However, it was Orton (1928) and his ideas regarding mixed cerebral
dominance that seemed to capture the attention of psychologists and edu-
cators alike. Historically, he was important, not so much because his
theories were well accepted but because he stimulated some five decades
of research into the relationship involving laterality, language abilities,
and reading acquisition. Overall, Orton estimated that 2 to 4% of the
school population had a significant reading deficit that merited special
training. After decades of research in which the relationships between
laterality and academic or cognitive performance have been examined, it
can be concluded firmly that Orton's (1928) ideas have not stood the test
of time. Reviews by Benton (1975), Kinsbourne and Hiscock (1981), and
Zangwill (1960) support this conclusion. His ideas do, however, continue
to attract attention (Geschwind & Behan, 1982). More recent efforts to
define specific learning disorders, such as severe reading disability, have,

as in the past, generated continued controversy and yet more definitions
and research.
DEFINING LEARNING DISABILITIES
It might well be asked, why so much attention on how learning dis-

abilities are defined? A multitude of reasons exist, of course, and include
the following: (1) so that the syndrome can be adequately differentiated
from other neurological or behavioral syndromes manifesting similar
symptoms, (2) for treatment reasons, i.e., differential diagnosis leads to
differential treatment, (3) for fiscal planning purposes (how one defines a
condition impacts directly on how large a clinical population is and the
resulting need for training of teachers and related services), and (4) agree-
ing, even temporarily, on the definition of a syndrome and how it is
diagnosed can aid researchers in their continued efforts to assess the
parameters of the syndrome and build further testable hypotheses as to its
etiology, manifestation, and treatment. Thus, in this vein, assessment is a
vital and dynamic component in clinically identifying children and fur-
thering, through research, our understanding of learning disabilities.
The Public Forum and PL 94-142

Public awareness of the potential impact of having a significant
number of children progress through school without profiting from their
experience led to a growing realization that special funds needed to be set
aside for special education efforts. The neurological perspective con-
tinued to be articulated (e.g., Strauss & Kephart, 1955; Strauss & Lehtinen,
1947), and through the efforts of parent and teacher groups (Doris, 1986;
McCarthy & McCarthy, 1969), a consensus definition, based considerably
on the World Federation of Neurology's definition (Critchley, 1970) and
that offered by Kirk and Bateman (1962), evolved. The result of these
efforts led in 1975 to the passage of a public law, PL 94-142, that affirmed
the rights of all handicapped children, including the learning-disabled, to
a public, appropriate, and least restrictive education.
Acknowledging the Neurological Etiology

The impact of PL 94-142's definition was indeed tremendous. With
the significant increase in the number of learning-disabled children re-
ferred, assessed, and diagnosed, there were concurrent increases in school
personnel (e.g., learning disabilities teachers, school psychologists). A
vast administrative network now existed for the identification and treat-
ment of children who experienced learning difficulties.
However, a great deal of controversy arose. One issue of particular

relevance was the notion that these disorders were of neurodevelopmen-
tal etiology. There were those who argued that learning disabilities were a
result of inadequate instruction or other nonneurological factors (e.g.,
Harris & Hodges, 1981; Ross, 1976; Smith, 1978).
Recognizing, however, that a large body of evidence argued for a
neurological etiology, the National Joint Committee for Learning Dis-
abilities (NJCLD) published a more complete definition, which, it hoped,
would address directly many concerns raised by critics of the definition
offered by PL 94-142. The NJCLD defined learning disabilities as "a gener-
ic term that refers to a heterogeneous group of disorders manifested by
significant difficulties in the acquisition and use of listening, speaking,
reading, writing, reasoning or mathematical abilities. These disorders are
intrinsic to the individual and presumed to be due to central nervous
system dysfunction" (Hammill, Leigh, McNutt, & Larsen, 1981).
In the context of the present discussion, the NJCLD committee appar-
ently agreed that "hard evidence of organicity did not have to be present
in order to diagnose a person as learning disabled, but that no person
should be labelled LD unless CNS dysfunction was the suspected cause"
(Hammill et al., 1981, p. 340).
It is in this particular regard and in the belief that learning disabilities
have a neurological etiology that the focus of this chapter is offered. To
understand fully the literature regarding the neuropsychological subtypes
of learning disabilities, a brief overview of over two decades of research
will be presented. Then, the supporting evidence as to the distribution of
neuropathology in the brains of learning-disabled subjects will be re-
viewed in an attempt to place the subtyping literature in some neu-
rological perspective. Finally, and most important, implications for clini-
cal neuropsychological assessment of these children will be offered.
MUL TIF ACTOR RESEARCH
Perspectives on the Single Factor Research

Decades of research investigating unifactorial conceptualizations of
the primary perceptual or cognitive deficit underlying developmental
dyslexia led down a nonproductive path (Hynd & Cohen, 1983). Single
factor research efforts focused on visual perceptual deficits (e.g., Lyle,
1969; Lyle & Goyen, 1968, 1975) and particularly their importance in the
early acquisition of fluent reading skills (Jansky & deHirsch, 1972;
Rourke, 1976; Satz, Friel, & Rudegeair, 1974; Silver & Hagen, 1971). Dis-
satisfaction with this perspective and nonsupportive evidence provided
by Liberman, Shankweiler, Orlando, Harris, and Brett (1971) and Fischer,
Liberman, and Shankweiler (1978) led to a general recognition that the
286 GEORGE W. HYND et a1.
deficits children typically evidenced in these visual-perceptual studies

actually resulted from linguistic intrusion errors. These, perhaps, were
reflective of deficits in verbal mediation, not visual perception.
Early Subtyping Studies

Concurrent with the popular efforts of many investigators in the
1960s to define the one single deficit underlying learning disorders in
children was a growing awareness by some investigators that there existed
two, if not more, subtypes of disabled readers.
Kinsbourne and Warrington (1963) clearly should receive the credit
for providing the first of some 30 or more published studies in which
subtypes have been identified. This study is important for several reasons:
First, it has historical significance; second, it clearly represents those
approaches that might be considered more "clinical" in nature; and third,
unlike many of the multivariate studies published today, that of Kins-
bourne and Warrington had a clearly defined hypothesis to examine.
Employing only 13 subjects referred to them because of a history of
reading or writing problems, Kinsbourne and Warrington (1963) divided
these subjects into two groups. One group of subjects (n = 6) had at least a
20-point difference in Verbal and Performance IQ (PIQ > VIQ). The sec-
ond group (n = 7) were characterized as having Wechsler IQs in the
reverse direction (VIQ> PIQ). Both groups of subjects received a clinical
neurological examination, an achievement test, and tests of finger differ-
entiation. They predicted that if, in fact, "reading backwardness" was due
to multiple causes, then their two groups of subjects, divided primarily
along a verbal-visual perceptual continuum, might be differentiated on
other dependent measures. To this end they found that the subjects who
had superior perceptual skills did indeed have marked language dis-
abilities, whereas those characterized as having normal language skills
but poor visual-perceptual abilities were remarkably like children with a
developmental Gerstmann syndrome (finger agnosia, dysgraphia, dys-
caiculia, and right-left orientation problems). They suggested that "inso-
far as the acquisition of reading and writing skill is a complex procedure,
involving a variety of cerebral functions, it is not surprising to find that
retarded development of one or other of the functions subserved by the
cerebral hemispheres may delay this acquisition, and do so in different
ways, depending upon the exact nature of the function which is insuffi-
ciently developed" (p. 153).
Thus, the stage was set for further investigations. Bannatyne (1966)
and particularly Bateman (1968) provided further evidence that at least
two, if not three or more, subtypes existed, as suggested by Kinsbourne
and Warrington (1963). It remained, however, for Boder (1973) to add to
Kinsbourne and Warrington's (1963) findings. She reasoned that reading
and spelling were interdependent functions, and thus, by examining read-

ing and spelling patterns, one could elicit a more clear conceptualization
as to the nature of the cerebral dysfunction characteristic of developmen-
tal dyslexics. Using a diagnostic reading-spelling task she had developed
earlier (1971), Boder (1973) administered it to 107 children from 8 to 16
years of age. All were diagnosed as developmental dyslexic. Primarily on
the basis of her clinical impressions, she identified three groups of dis-
abled readers.
First, there were those who had difficulty with the phonetic aspects
of reading and read words globally using visual gestalt perception. Ac-
cording to Boder, this subtype accounted for approximately 63% of all
disabled readers in her sample. The second subgroup Boder identified
were the dyseidetic dyslexic or those disabled readers who read la-
boriously and had a poor visual memory for words. These dyslexics, she
believed, accounted for some 9% of the disabled readers in her sample.
Finally, a third identifiable pattern of reading-spelling errors emerged in
which the children, about 22% of her sample, were severely handicapped,
sharing the traits of both the dysphonetic and the dyseidetic dyslexic.
This subgroup she referred to as her mixed type, and it was believed that
these disabled readers had the poorest educational prognosis.
These two earlier studies have been highlighted because they both
adequately represent the contribution that clinical neurology made in
delineating the behavioral manifestations of subtypes of disabled learn-
ers. In each case, these investigators applied clinical-experimental mea-
sures to differentiate the unique behavioral manifestations of each sub-
type. However, potential issues related to instrument reliability and
validity clearly were not a focus of concern.
There quickly followed a number of statistically sophisticated studies.
Some, such as Rourke, Young, and Flewelling (1971) were more empirical
replications of Kinsbourne and Warrington's (1963) basic study, whereas
others (e.g., Mattis, French, & Rapin, 1975) examined related issues per-
taining to the presumed degree of neurological involvement. Many of these
studies and others that followed are summarized in Table 1.
Some Directions in Subtyping Research: Multivariate

Classification Approaches
By the mid 1970s several equally productive lines of research regard-
ing subtypes of learning disabilities were being pursued. Generally, there
were those studies that sought, through multivariate techniques, to identi-
fy the deficient neuropsychological processes that characterized the sub-
types. For example, the study by Petrauskas and Rourke (1979) used a
large sample of retarded readers and, applying factor analysis to a large
data set, found that three reliable subtypes existed. The first group of
TABLE 1. A Chronology of the Major Subtyping Studies Conducted 1963-1986 a ,b
Age range
Author(s) N Population Sex SES Matched (years) Tests Type of study Analysis General findings
Kinsbourne and 13 Nonverbal> verbal by M = 62% N/R No 8 to 31 1. WISC-R or W AIS Clinical impressions Qualitative P > V clinical tesing
Warrington at least 20 points F ~ 38% 2. Schonell Reading revealed signifi-
(1963) Verbal> nonverbal by and Spelling language
at least 20 points 3. Tests of Finger impairment.
Differentiation V > P failed tests of
finger differentiation
and had difficulties
with arithmetic.
Called Group 1, the
language-retardation
group.
Called Group 2, the
Gerstmann Group.
Ingram, Mason, 82 Dyslexics M = 80% NIR No 7 to 15.11 1. Stanford-Binet Frequencies Chi-square Specifics had a signifi-
and Blackburn 1. Specific dyslexics F = 20% 2. Schonell Reading, cant higher propor-
(1970) = 62 Spelling, tion of boys than
2. General dyslexics Arithmetic girls.
= 20 3. Goodenough Generals had signifi-
Draw-a-Man cantly more neu-
4. Neuropsychologi- rological findings.
cal Assessment Specifics had signifi-
(EEG, lateral cantly more au-
dominance) diophonic reading
5. Physical exam errors.
Rourke, Young, 90 LD 3 groups M ~ 80% N/R No 9 to 14 1. PPVT Differential perfor- Multiple 1-way HV-LP group was sig-
and Flewelling 1. PIQ> VIQ F = 20% 2. WRAT mance of LD types ANOVAs nificantly better on
(1971) 2. PIQ = VIQ 3. Reitan's Category those tasks involv-
3. PIQ < VIQ Test ing verbal, language,
4. Aphasia Screening and auditory per-
Test ceptual skills.
5. Speech Perception LP-HV group signifi-
Test Rhythm Test cantly better on
Trails A and B tasks involving visu-
al-perceptual skills.
Naidoo (1972) 98 Dyslexics All males Middle class Age + type 8 to 12.11 1. WISC Differential perfor- Cluster analysis Dyslexics had a sig-
1. 56 reading of school 2. Reading, Spelling mance of LD types nificantly greater
deficits tests family history of
2. 42 spelling 3. Perceptual tasks reading problems
deficits (auditory, visual, than controls. No
motor) subtypes of dyslexia
4. Social Adjustment were identified. Evi-
Scales dence suggests that
sequencing dis-
ability may underlie
reading and spelling
retardation.
Bader (1973) 107 Children who met the M ~ 86% N/R 8 to 16 Boder Reading-Spell- Clinical impression Qualitative 3 subtypes
world federation F = 14% ing Pattern Test 1. Dysphonetic
definition of devel- (63%)
opmental dyslexia. 2. dyseidetic (9%)
3. Mixed (22%)
Undetermined (6%)
Nelson and War- Study 1 N/R No 8 to 14 1. WISe Error analysis Multiple Hests Degree of VIQ decre-
rington (1974) Reading and Spell- 2. Schonell Reading ment is much more
ing deficits~2 years and Spelling strongly associated
39 1. More general Yer- M ~ 85% with degree of read-
bal retardation by F:::: 15% ing retardation than
15 points (WISC) with spelling retaf-
82 2. No general verbal M = 76% dation.
retardation F ~ 24%
Study 2 N/R No Same as Same as above Error analysis Multiple t-tests Spelling and reading
17 1. Reading < 2 M ~ 88% above retardates had sig
years retarded, F = 12% nificantly lower VIQ
spelling> 2 than spelling-only
years retarded retardates.
"Abbreviations in this table include (N/R) not reported, WISC-R (Wechsler Intelligence Scale for Children-Revised), WAIS (Wechsler Adult Intelligence Scale), PPVT
(Peabody Picture Vocabulary Test), WRAT (Wide Range Achievement Test), PIQ (Performance IQ), VIQ (Verbal IQ), FSIQ (Full Scale IQ), LP (Low Performance IQ), SPA
(Spatial factor), CON (Conceptual factor), SEQ (Sequential factor), FTNW (Finger Tip Number Writing), LNNB-CR (Luria-Nebraska Neuropsychological Battery-Chil-
dren's Revsion), VMI (Developmental Test of Visual Motor Integration), PlAT (Peabody Individual Achievement Test), K-ABC (Kaufman Assessment Battery for
Children), P (WISC Performance IQ), V (WISC Verbal IQ), M (males), F (females), EEG (Electroencephalogram), ANOV A (Analysis of Variance), LD (Learning-Disabled), X
(Mean), NLD (Non-Learning-Disabled), R-L (Right-Left), WISC (Wechsler Intelligence Scale for Children), EMR (Educable Mentally Retarded), SES (Sociometric Status),
ANCOVA (Analysis of Covariance), MANOVA (Multivariate Analysis of Variance), CAT (California Achievement Test), SLDR (Specific Learning Disabled Resource), CBI
(Classroom Behavior Inventory), SCAN (Schedule of Classroom Activity Norms), SEARCH (Systematic Evaluation of Reading Criteria for High Risk).
bIt should be noted that subtypes (auditory, visual, auditory-visual, emotional, and pedagogic) of reading disability were proposed earlier than 1963. For example,
Gjessing (1953) noted those listed above a decade earlier than Kinsbourne and Warrington's (1963) report. However, the validating evidence for the existence of these
subtypes has only been the focus of researchers during the past 25 years.
(continued)
Age range
54 2. Reading> 2 M = 83% Speiling and reading

years retarded, F = 17% retardates made
speiling > 2 more phonetic er-
years retarded rors than did speil-
ing-only retardates.
Mattis, French 113 3 groups Not spec- NIR No 8 to 18 1. WAIS or WISe Comparisons of Multiple Brain-damaged readers
and Rapin N = 31 brain-damaged ified 2. Benton Test of Vi- group perfor- ANOVAs PIQ significantly
(1975) reader sual Retention mance with Dun- lower than either
N = 53 brain-damaged 3. Ravens PIogres- can's new dyslexic group.
dyslexic sive Matrices multiple Brain-damaged readers
N = 29 Non-brain- 4. Oral language (4) range test significantly better
damaged dyslexic 5. WRAT than one or both
6. Motor tasks (3) dyslexic groups on
verbal labeling, lan-
guage comp., and
imitative speech.
Similarity between the
developmental dys-
lexic and the brain-
damaged dyslexic.
Tben defined 3 pat-
terns and assigned
criteria. Accounted
for 90% of dyslex-
ics.
Doehring and 65 Reading and learning M = 91% NIR No 8 to 17 31 Tests: Differential perfor- Q Factor Group R~3-factor
Hoshko (1977) problems Group R, F= 9% 9 Reading mance of LD types Analysis for solution
N = 34 reading 8 Visual scanning Group R, Group M-3-factor
problems Group M, M = 68% 7 Visual matching Group M, solution
N = 31 mixed prob- F = 32% 7 Auditory visual and com- Combinecl-4-factor
lerns matching bined. solution
Group R
1. Slow oral word
reading
2. Slow aud-vis. let-
ter association
3. Slow aud-vis. as-
sociation of
words and sym-
bols.
Smith, Coleman, 208 Learning-disabled M ~ 76% N/R No 6.3 to WISC-R (no digit Differential perfor- Repeated mea- Total sample-SPA>
Dokecki, and 1. N ~ 132 IQ > 75 F = 24% 12.1 span given) mance on the sures CON> SEQ and ac-
Davis (1977) VIQ& PIQ '" 90 WISC-R ANOVA, qui red knowledge
2. N ~ 50 not meet- Newman- EMR-spatial > than all
ing above criteria Keuis others.
3. N ~ 26 EMR
FSIQ < 75
Omenn and We- 21 Dyslexics-3 groups N Both, but #s N/R No School- 1. Slingerland Differential perfor- Chi-square Evoked potential rec-
ber (1978) = 7 auditory prob- not re- aged, 2. WISC mance of LD types ording-no signifi-
lems, N = 11 visual ported but not 3. Spelling tests cant differences
problems, N ~ 3 re- 4. Visual and Audi- when comparing left
mixed. ported tory Evoked and right parietal re-
Potentials sponses. No asym-
5. Family history metries were seen
6. Academic records within subtypes,
7. Handedness handedness, or
spelling error types
for auditory or visu~
al evoked response.
Visual dyslexics
tend to have fam-
ilies with visual
dyslexia and audi-
tory dyslexics have
families with audi-
tory dyslexia.
Rourke and 45 Learning-disabled-3 M ~ 87% N/R No 9 to 14 1. WRAT Comparison of Multiple I-way Group 3 > 2 on all
Finlayson groups F = 13% 2. WISC group perfor- ANOVAs verbal and percep-
(1978) 1. Low in reading, 3. PPVT mances with Tukey tual measures (ex-
math, spelling 4. Halstead-Wepman comparisons cept auditory
2. Low in reading, Aphasia Screening closure)
spelling. Math Test Group 3 VIQ > PIQ
ok. 5, Vocabulary Recog- PIQ Group 2 > 3
3. Low in math, nition Task Group 1 & 2 similar
reading, and Group 1 VIQ < PIQ
spelling ok.
(continued)
Age range
Sweeney and 48 Groups (3) M = 63% Middle class Age + 4th grade 1. Auditory tests Differential perfor- Multiple Normals significantly
Rourke (1978) 4th- and 8th-graders F = 37% wIse 9.9 to 2. Verbal tests mance of LD types ANOVAs greater than pho-
1. Nonphonetic re- PIQ 10.1 3. Reading and control with New- netically inaccurate
tarded spellers 4. Visual Closure man-Keuls spellers only at
2. Phonetic retarded test comparison older age level on
spellers 8th grade all but the simplest
3. Control 13.3 to of auditory discrimi-
13.4 nation and immedi-
ate auditory~verbal
memory measures,
Doehring, 158 3 subtypes N/A N/R Age + Sex 8 to 17 31 Tests Differential perfor- 1. Q~factor Q factor analysis repli~
Hoshko. and Reading problems (n 9 Reading mance of LD and 2. Cluster anal- cated same 3 sub-
Bryans (1979) ~ 34) 8 Visual Scanning controls ysis types when original
Mixed problems (n 7 Visual Matching data pooled with
~ 31) 7 Auditory-Visual normal readers.
Combination of Matching Same subtypes re-
reading and mixed mained well defined
problems (n ~ 62) when reexamined
Controls (n ~ 31) using cluster analy-
sis approach.
Fisk and Rourke 264 Learning-disabled M = 81% N/R No 9 to 14.9 1. WRAT Differential perfor- Q~factor analy- 54% of subjects were
(1979) 3 groups F = 16% 2. Reitan's Neuro- mance of LD types sis classified into sub-
9-10 years (n ~ psychological Bat- on neuropsycho~ types that were
100) tery logical battery "replicated" across 2
11-12 years (n ~ or more age levels.
100) Type A-Poor auditory
13-14 years (n ~ verbal processing
64) and finger localiza-
tion deficit.
Type B-Linguistic
deficit.
Type C-Poor perfor-
mance in fingertip
number writing per-
ception (may be a
variant of Subtype
A).
Petrauskas and 160 2 groups M = 86% N/R No 7 to 8.9 1. WRAT Reading Differential perfor- Pearson Prod- 3 reliable subtypes
Rourke (1979) N = 133 retarded F = 14% 2. Reitan 6 skill mance of normal uct-Moment found. 50% of sub-
readers areas: Tactile, Mo- and retarded read- Correlations ject sample could be
N = 27 normal readers tor, Visual-Spatial, ers Factor analysis reliably classified.
Auditory-Verbal, using iterated Type I-Most lan-
Abstract-Concep- principal guage disturbances.
tua!, and Sequenc- axes solution Type 2-Sequencing
ing deficits. Finger
Agnosia.
Type 3-Conceptual,
motoric. and verbal
expressive deficits.
Pirozzolo (1979) 24 Right-handed All males N/R Age + Full X=ll.1 1. 16 three-letter Differential perfor- ANOVA with Auditory-linguistic
3 groups Scale IQ range stimulus words mance of LD and Newman- group showed no
1. Visual dyslexics not re- 2. Eye movement controls Keuls com- lateral asymmetries
2. Auditory-linguis- ported monitor parisons for word recognition
tic dyslexics 3. Gray Oral Reading when compared
3. Normal readers Test passages with normals.
Visual-spatial group
had significant
deficits recognizing
words presented
parafoveally, faster
saccadic leftward
eye movement, and
more return sweep
reading inaccuracies
when compared
with normals.
Lyon and Wat- 150 100 learning-disabled M = 90% Middle class Age + 11 to 12.5 10 tests Differential perfor- Multiple I-way 6 homogeneous sub-
son (1981) resource F = 10% WISC-R 1. Auditory tests mance of LD and ANOVA groups for raw and
50 controls FSIQ 2. Token tests controls Cluster Analy- standard scores, ac-
3. Raven's Col- sis *(only on counted for 94% of
oured Progres- LD subjects) SLDR subjects.
sive Matrices Multivariate Discriminant analysis
4. VMI discriminant between 6 sub-
5. PlAT (2) analysis groups yielded 2
6. Memory for ScheWi post discriminant
designs hoc pairwise functions.
7. Naming Test comparisons 1. Language and vi-
(continued)
Age range
sual-motor inte-
gration function
2. An orthogonal vi-
sual-motor inte-
gration
dimension.
Satz and Morris 325 5th-grade white boys All males N/R No X=l1 WRAT Differential perlor- 2 cluster anaIy- 5 subtypes
(1981) in Florida Longitu- years PPVT mance of white ses-first one 1. Impaired on both
dinal Project (range Neuropsychological 5th-graders to identify language
not re- Battery learning dis- variables.
ported) SES ability group, 2. Impaired on ver-
second to bal fluency only.
subtype the 3. Impaired on lan-
learning-dis- guage and per-
abled group ceptual measures.
4. Impaired on per-
ceptual only.
5. No impairment.
Thompson 83 Spelling and reading- N/R N/R No 3 age 1. WISC-R Error analysis l-way 3 subgroups were de-
(1982) disabled groups: 2. Raven's ANaVAs fined as basis of
Average IQ 8 to 10.11 3. British Ability with Tukey reading and spelling
3 age groups (n = Scales multiple errors. A greater
29) 4. Neal's Analysis of comparisons number of children
11 to Reading Ability for a repeat~ with auditory~lin~
13.11 ed-measures guistic deficits than
(n = design visuo-spatial at all
29) ages. Visuo-spatial
14 to group scored less
16.11 well on speed of in~
(n = formation process~
24) ing, block design
level and block
design power.
Nolan, Ham- 36 a. Normal-WRAT ;::= M = 78% N/R No 7 to 13 WISC-R, FTNW, Fin- Determine if LD I-way ANaVA Partial evidence for
meke, and 40 percentile F = 22% ger Agnosia, subtypes would Scheff;; AN- notion of unique
Barkley (1983) b. Reading-Spelling s LNNB-CR, WRA T exhibit unique CaVA neuropsychological
20 percentile neuropsychologi~ profiles. The expres-
Math" 40 cal profiles sive speech, writing.
percentile and reading scales
c, Math s 20 percen- of LNNB-CR differ-
tile, Reading-Spell- entia ted the poor
ing 2: 40 percentile reading/spelling
group from the other
2 groups. Did not
differentiate the
math deficient
group.
Watson, Goldgar, 65 Children who were re- M = 77% N/R No 7 to 14.11 6 skill areas: Differential perfor- Cluster analysis 3 clusters
and Ryschon ferred for LD F = 23% Reading/ Spelling mance of children with a. visual processing
(1983) Visual Processing referred for LD K-means iter- deficit
Language ative parti- b. generalized
Memory tianing language disorder
Perceptual method c. minimal deficit
Organization, subtype
VMJlBehavior Clusters relatively het-
erogeneous and may
have limited clinical
utility.
Meacham and 50 4 groups CAT scores M = 54% N/R No 7.1 to 9.2 CAT Differential perfoT- t tests, Q factor, 4 subtypes for
Fisher (1984) Normal readers (n = F ~ 46% SEARCH mance of LD types discriminant kindergarten.
20) WRAT and controls analysis 3 subtypes for second
Reading-disabled in grade.
Special Ed. (n ~ 10) There was little sta-
Reading/Language dis- bility in the sub-
abled in self-con- types from
tained LD (n = 10) kindergarten to sec-
Scores < 35%ile on and grade.
CAT-No remedia-
tion [n ~ 10)
Fletcher (1985) 87 Controls (16), Reading- M = 58% N/R No 7.6 to 12 WISC-R, WRAT, Ver- Differential Perfor- MANOVA, Relative to controls,
Spelling LD (10), F = 42% (X ~ bal and Nonverbal mance of LD ANOVA, Arithmetic-Spelling
Reading-Spelling- 9.94) Selective Remind- achievement sub- post hoc LD subjects had dif-
Arithmetic LD (38), ing Task types on verbal comparisons ficulty on the non-
Spelling-Arithmetic and nonverbal se- (Tukey) verbal task, the
LD (10), and Arith- lective reminding reading, spelling
metic LD (13) task. Influence of children had diffi-
IQ also examined. culty on the verbal
(continued)
Age range
memory task. and

Reading-spelling-
arithmetic LD sub-
types did poorly on
both verbal and
nonverbal tasks. IQ
is not a significant
factor. Supports dif-
ferentiation of LD
Children on the
basis of patterns of
achievement.
Hooper and 117 5 groups M = 55% Middle class Age, sex, 8.4 to K-ABC Differential diag- MANOVA, Sequential processing
Hynd (1985) 30 Normal Readers F = 45% geo- 12.5 Boder nosis of LD sub- ANOVA, (K-ABC) was able to
32 Nonspecifics graphic types Duncan Mul- distinguish between
30 Dysphonics location tiple range normals and
5 Dyseidetics with harmon- dyslexics.
20 Alexics ic n, discrim- Simultaneous process-
iuant iog component did
analysis not discriminate
among subtypes ex-
cept for matrix anal-
ogies subtest.
Snow, Cohen, 106 LD M = 83% N/R No X = 10.4 WISC-R Differential diag- Cluster analysis A 6-subgroup solution
and Holliman 7 black subjects F = 17% nosis of LD sub- using factor was considered
(1985) 99 white subjects types scores de- most appropriate
rived from discussed in terms
the WISC-R. of cognitive pat-
A hierarchi- terns.
cal approach
was utilized.
Snow and Hynd 100 LD M = 85% N/R No 8 to 12 LINNB-CR, WISC-R Differential diag- Q factor analy- 72% subjects classified
(1985) 79 resource F = 15% (X = WRAT nosis of LD sub- sis. ScreeTest into distinct sub-
21 self-contained 10) t;"es MANOVA groups, but all 3
and t tests groups probably rep-
resented variation of
language-disordered
LD.
Speece, McKin- 129 LD (n ~ 63) a. LD All levels Sex + race LD-X~ CBI Differential diag- Hierarchical More than 1/3 of LDs
ney, and Ap- Normal achievers M ~ 76% repre- 7.2 SCAN nosis of LD based cluster Anal- did not exhibit mal-
pelbaum (n ~ 66) F ~ 24% sented NLD-X on ratings of ysis, Ward's adaptive behaviors.
(1985) b. M ~ 77% ~ 7.2 classroom be- Minimum 7 subgroups of LD
F ~ 23% haviors variance. (85% of non-LD fell
split sam- into 2 clusters) were
piing replica- identified.
tion and
forecasting,
MANOVAs
Breen (1986) 90 LD-3 groups M ~ 65% N/R No a. X ~ Woodcock-Johnson Differential diag- t tests, 1-way Most subtests and
a. 30 Math> F ~ 34% 9.5 Psycho.-Ed. Battery nosis of LD sub- ANOVA, clusters differed sig-
Reading b. X ~ types multiple re- nificantly for at least
b. 30 Reading> 9.6 gression and one comparison.
Math c. X ~ discriminant Groups 1 and 2
c. 30 Math ~ 9.9 function, were comparable
Reading Tukey com- and tended to score
parisons higher than Group
3. 64% of LD chil-
dren were correctly
classified by sub-
types. Inconclusive
results in terms of
stable cognitive pat-
terns.
Morris, 200 Nonclinical normal All males N/R No X ~ 5.6 13 measures of Longitudinal 1. Outlier iden- 5 developmental
Blashfield, and and reading-dis- in neuropsychological tification subtypes
Satz (1986) abled males, tested kinder- and cognitive techniques. Type A. deficient
at kindergarten, sec garten skills. Factor anal 2. Longitudinal verbal skills and in
ond, and fifth ysis of this battery cluster creasing strength on
grades, from the yielded two major analysis visual perceptual
Florida Longitudinal factors, which 3. Internal val- motor skills with
Project these 8 measures idatian of age.
represented. useful Type B. increasing
1-4, verbal/concep solutions deficits in perfor-
tual 4. External val- mance, especially in
1. PPVT idation of verbal-conceptual
(continued)
Age range
2. Verbal Fluency useful solu- skills with age.

3. Similarities tions across Type C. below aver-
4. Dichotic Listen- nine do- age on all tests and
ing 5-8, sensor- mains of ratings.
imotor/percep- variables not Type D. average to
tual used for above average on all
5. Recognition-Dis- classifica- tests and ratings.
crimination Test lion (MAN- Type E. above aver-
6. Embedded OVA + Chi- age on all tests and
Figures Square) ratings.
7. Beery-Buktenica
Developmental
Test of Visual-
Motor
Integration
8. Auditory-Visual
Integration Test
Spreen and Haaf 335 LD referrals Not reported N/R Age, Sex a. X~ PlAT Longitudinal study Hierarchical Group 1-6 subtypes
(1986) a. 63-referral group SES 10.14 WISC to look at stability cluster anaIy- Group 2-8 subtypes
b. 96-IQ> 79 b. X~ SAIS of LD subtypes sis, with (similar to Group 1)
c. 170-Adult sub- 9.91 Sentence across age Ward mini- Group 3-9 subtypes
jects (including c. X~ Rep., Categories mum vari- visuospatial and
46 con trois) 24.28 Purdue Pegboard, allee, Cluster graphomotor were
R-L orient. Cubing Crite- identified but not
fion, K linguistic subtype.
means itera-
tive parti-
tioning, Chi-
Square.
ANOVA,
MANOVA
children suffered language disturbances. The second group of subjects

evidenced sequencing deficits and, like Kinsbourne and Warrington's
(1963) Gerstmann group, finger agnosia. The third group of subjects was
identified as suffering conceptual, motoric, and verbal-expressive defi-
cits. This study was well conceptualized but, as with so many other multi-
variate studies (e.g., Fisk & Rourke, 1979; Omenn & Weber, 1978), either
many potentially important variables were not reported or the study suf-
fered from other flaws that only compromised the results. For example, in
not one of Rourke's or his colleagues' studies (Fisk & Rourke, 1979;
Rourke & Finlayson, 1978; Sweeney & Rourke, 1978) was any mention
made regarding the socioeconomic level of the subjects. This is clearly a
significant oversight since considerable evidence suggests this to be an
important variable with regard to neuropsychological abilities (e.g., Reyn-
olds & Gutkin, 1979). Further, in studies where previously existing data
bases were employed, rarely was an effort made to control for potentially
important variables such as SES, IQ, or gender.
In a considerably more sophisticated study in which some of these
issues were addressed, Lyon and Watson (1981) examined 100 reading-
disabled (diagnosed according to federal guidelines [PL 94-142] em-
ploying measures of reading recognition and comprehension) and 50 nor-
mal readers matched according to school and chronological age. The nor-
mal readers were used primarily to establish local norms for the test
variables used. In an effort to identify subtypes, raw scores and standard-
ized scores of the reading-disabled children (using local norms) were
cluster-analyzed separately. Six homogeneous subgroups for raw and
standard scores were obtained, accounting for 94% of the reading-dis-
abled subjects. Thus, rather compelling evidence from this particular
study suggested considerably more discrete subtypes than previously
reported.
Other studies (e.g., Satz & Morris, 1981; Watson, Goldgar, & Ryschon,
1983) also have contributed considerably in providing convincing evi-
dence that (1) more than two subtypes exist, and (2) each subtype can be
characterized by reliable cognitive profiles. However, these studies do not
contribute greatly to our understanding of the linguistic systems involved
in fluent or dysfluent reading. A number of these studies have employed
neuropsychological tests from the Halstead-Reitan Neuropsychological
Battery (e.g., Fisk & Rourke, 1979; Petrauskas & Rourke, 1979; Rourke &
Finlayson, 1978; Rourke et a1., 1971). Considerable evidence suggests that
this test battery is heavily loaded with tasks that assess visual-motor abili-
ties (Crockett, Klonoff, & Bjerring, 1969), and, worse yet, Seidenberg, Gior-
dani, Berent, and Boll (1983) provide evidence that nearly half (6 out of
14) of these measures correlate significantly with IQ. Thus, these tests
reveal little beyond the IQ results, and what extra is contributed has little
to do with semantic-linguistic processes that typically characterize the
disabled learner.
Thus, many of the studies cited in Table 1 are by and large athe-
oretical (rarely is any specific rationale given for tests employed). A typ-
ical multivariate study (in Table 1) utilizes variables that are distantly
related to actual reading or achievement. These variables then are clus-
tered to yield "subtypes" that often have little significance in helping us
understand why these children cannot learn normally.
Satz and Morris (1981), Taylor, Fletcher, and Satz (1982), and others
(e.g., Coltheart, 1980) have argued cogently that subtypes should be de-
fined in terms of actual patterns of deficient learning or reading behaviors
and then studied neuropsychologically. This type of approach makes
sense for three reasons.
First and foremost, grouping or subtyping children by the actual pat-
terns of errors they make in learning should, through careful error analy-
sis, eventually reveal many of the potential subprocesses directly in-
volved in learning within any specific domain. Thus, neurocognitive
theory will be advanced. Second, it should seem obvious that by assessing
actual processes involved in fluent reading, the assessment achieves more
ecological validity. Thus, the derived subtypes may actually be observed
in the clinic setting. Third, and probably most important, if neuropsychol-
ogists can delineate competently the actual neurocognitive, especially
neurolinguistic, behaviors that characterize these subtypes, direct treat-
ment implications may result.
In fact, Fletcher (1985) provides an excellent example of how such
research efforts should be conducted, He examined verbal and nonverbal
memory (storage and retrieval) using a selective reminding procedure
among four groups of learning-disabled children. These groups were
those with (1) reading-spelling disabilities, (2) reading-spelling-arith-
metic disabilities, (3) spelling-arithmetic disabilities, and (4) arithmetic-
only disabilities. Consistent with his hypotheses, he found that arith-
metic- and spelling-arithmetic-disabled children did poorer on the non-
verbal tasks whereas the reading-spelling group did worse on the verbal
task. Those with the reading-spelling-arithmetic disabilities performed
poorly on the storage and retrieval tasks tapping both verbal and nonver-
bal memory. Thus, Fletcher concluded that there does seem to be validity
in subtyping disabled learners on the basis of patterns of academic
achievement. With this study in mind, let us now return our attention to
reading disorders where some initial progress along this dimension is
being made.
Neurolinguistic Subtyping
Frustration with the fact that decades of research on "subtypes" of
reading disorders have failed to increase significantly our theoretical un-
derstanidng of the linguistic processes by which children access printed
material has led to efforts in subtyping these children according to the

kinds of reading errors they make. Based on a neuroanatomical-neu-
rolinguistic model of reading (see Hynd & Hynd, 1984, for a review), a
number of subtypes may exist in which deficient reading processes can be
tied directly to hypothesized neuroanatomical correlates. Basically, sever-
al subtypes have emerged.
Deloche, Andreewsky and Desi (1982) have described what they have
referred to as the surface dyslexic. The surface dyslexic reads by well-
established phonological rules, processes short words better than long
words, and has difficulty in semantic access. Whenever words are un-
familiar or irregular, reading becomes difficult. These dyslexics have dif-
ficulty with the visual aspects of word recognition, and whereas they have
good phonological skills, their comprehension is seriously affected.
Another subtype that has been identified through careful error analy-
ses of reading error patterns is what has been referred to as the deep
dyslexic. Deep dyslexics are able to read familiar words well, particularly
nouns, but seem to make many semantic paralexias in oral reading (Fried-
man & Perlman, 1982). They may, for example, see the word mutton and
say sheep. It has been proposed that the deep dyslexic suffers an anomia,
and thus, the semantic category is accessed correctly but the incorrect
visual-semantic association occurs, resulting in the paralexic response.
These readers seem to rely on imageability, concreteness, and word fre-
quency in their reading behavior.
Other neurolinguistic subtypes have been proposed (Marshall & New-
combe, 1973, 1980; Warrington, 1981); however, these studies are rela-
tively few in number, they focus on small numbers of children, and most
of the validating evidence as to correlated neuroanatomical structures
comes from adult brain-injured patients. Thus, the results from this line of
investigation might still be considered as preliminary insofar as identify-
ing subtypes of reading-disabled children is concerned. However, it is
clear that this may prove to be a much more fruitful avenue for subtyping
research in the areas of reading, spelling, and mathematical disabilities
than those studies that have unfortunately typified much of the efforts of
researchers to date.
NEUROANATOMICAL-LINGUISTIC PERSPECTIVES
The Wernicke-Geschwind Model

Geschwind (1974) has had a dramatic impact on the development of
our understanding of the neurolinguistic-neuroanatomical relationships
involved in speech, language, and reading (Mayeux & Kandel, 1985).
Largely, his ideas, as well as those advanced by Luria (1980) and others
(e.g., Sevush, 1983; Whitaker, 1976), represent an extension of the neu-

rolinguistic model of speech and language first formally advanced by
Wernicke (1874). This model is of particular relevance, both theoretically
and clinically, in that most learning-disabled children have language-
related disorders of one kind or another.
One issue should be dealt with at this point. First, it has long been
assumed that the brains of children either are organized differently from
those of adults, such that most functions have bilateral representation, or
that the brain, although organized similarly to that of an adult, has a
remarkable capacity for recovery or reorganization (Sandoval & Haapenen,
1981). We assume, however, that the failure of most of the research to
document developmental trends in lateralized function (Hynd, Obrzut,
Weed, & Hynd, 1979) argues strongly that children's brains are at the least,
organized similarly to those of adults and that, with developmental learn-
ing problems, the organizational pattern remains as normally preordained
but operates less efficiently (Hynd & Willis, 1988; Kinsbourne & Hiscock,
1981). Thus, we would view Denckla's (1973) suggestion as conservative
when she proposed that "as localization in the traditional sense is not a
reasonable goal, utilization of analogies and critical differences between
childhood and adult syndromes sharing similar complaints brings us
closer to a clinical classification scheme" (p. 449) for childhood learning
disorders.
It is proposed here that validation of cortical dysfunction in the
brains of learning-disabled children in the regions predicted by the Wer-
nicke-Geschwind model argues that the organizational pattern of these
children's brains is similar to that of adults (Duffy, Denckla, Bartels, &
Sandini, 1980). Owing to neurodevelopmental abnormalities in these cor-
tical regions (Drake, 1968; Galaburda & Kemper, 1979; Galaburda, Sher-
man, Rosen, Aboitiz, & Geschwind, 1985), normal cognitive functioning is
impaired significantly.
Basically, in terms of the functional system involved in reading, as an
example, it would appear that there is a widespread network of subcor-
tical and cortical zones that interact or contribute to fluent reading. Figure
1 illustrates this proposed functional system. For instance, reading aloud
a word would involve the image projected to the retina being transferred
via the lateral geniculate nucleus to the primary visual cortex (Brod-
mann's area 17). Low-level feature analysis occurs in the visual associa-
tion cortex (Brodmann's areas 18 and 19), with imageable words pro-
cessed better in the right visual association cortex and letter-strings
favored in the left visual association cortex. Input from the visual cortex is
shared via commissural fibers from the right visual cortex, and via intra-
hemispheric fibers from the left visual cortex with the angular gyrus
(Brodmann's region 39) at the juncture of the left temporal, parietal, and
occipital regions. Cross-modal integration (e.g., grapheme-phoneme cor-
Left Right
(ANTERIOR)
Broca's
Area
Arcuate
Corpus
'Com,ph,,",;o" _-.~-----1- Callosum
Wernicke's
Area - - . "
Angular
Gyrus ---'~
Medial
Occipital
Area
(POSTERIOR)
FIGURE 1 A neurolinguistic model of the functional system believed to contribute signifi-

cantly to fluent reading. (From Hynd & Hynd, 1984, with permission.)
respondence) is thought to result here. This information is then shared

with Wernicke's region (Brodmann's area 22) in the left superior posterior
temporal cortex. Here auditory-linguistic images are associated with the
visual-phonemic stimuli and the word is recognized and comprehended.
In oral reading, this comprehended word would be shared with Broca's
region via the arcuate fasciculus, where the motor-grammatical images
necessary for expressive speech and articulation reside. From here the
facial and motor musculature for speech (Brodmann's area 4) would be
activated (Hynd & Hynd, 1984; Mayeux & Kandel, 1985; Patton, Sundsten,
Crill, & Swanson, 1976).
Knowledge of this model and linguistic processes involved in speech
allows one to make accurate predictions about lesion sites involved in
aphasia. Also, it is clear that while most linguistic functions are indeed
lateralized to the left cerebral cortex, other associated components may be
differentially associated with the right hemisphere. Ross (1981), for exam-
ple, has provided convincing evidence that disturbances of the affective
components of language (e.g., intonation or prosody, emotional gesturing,
prosodic comprehension, and comprehension of emotional gesturing) are
associated meaningfully with the right cerebral hemisphere. As yet, the
validation and localization of these functions in children's brains remains
a challenge for clinical child neuropsychologists.
Clinical Implications for Process-Based Assessment
From the perspective presented in the preceding two sections on the

neurolinguistic systems believed to exist in disabled readers, it should be
clear that many, if not most, of the studies summarized in Table 1 relate
poorly to any actual processes involved in dysfluent reading. Concep-
tually, it makes more sense to begin one's organization of a neuropsycho-
logical test battery to differentiate subtypes of learning disability by in-
cluding measures that assess those processes known to be important in
the neurolinguistic system involved in reading or achievement in some
other domain.
The neurolinguistic model discussed projects that performance on
tasks assessing the automatization of basic subskills directly involved in
actual performance will distinguish subtypes of learning-disabled chil-
dren (Sternberg & Wagner, 1982). Therefore, assessment of these children
during letter and word recognition tasks, especially on speeded classifica-
tion tasks (Hayes, Hynd, & Wisenbaker, 1986) makes sense. It also makes
conceptual sense to assess not only bisensory perceptual and memory
tasks where cross-modal integration is involved (e.g., Obrzut, 1979) but
various components of comprehension as well. For instance, different
subtypes of learning disability may emerge when different components of
comprehension (word, passage, contextual) are assessed. Some other as-
pects of the neurolinguistic processes thought to be important in reading
would also include repetition, fluency, syntactic-pragmatic abilities, and
naming. The differentiation of subtypes according to variable perfor-
mance on tasks assessing these abilities should tell us more about the
actual pattern of disabilities and abilities that characterize these learners.
For the clinician, what does this perspective suggest? First, when one
considers the neuropathological studies of the brains of dyslexic indi-
viduals (Galaburda & Kemper, 1979; Galaburda et 01., 1985), it seems
reasonable that we should expect much psychometric variability among
the subtypes. The distribution of the neurodevelopmental abnormalities
(focal dysplasias, disordered cortical layering, polymicrogyria), although
clustering in the left perisylvian region, is completely distinctive in each
case. Thus, it makes conceptual sense that each particular case seen by
clinicians should have a qualitatively different porfile of neurolinguistic
abilities. Also, it suggests that clinical neuropsychologists should consid-
er refocusing their assessment strategies away from largely redundant
neuropsychological batteries (e.g., Luria Nebraska Neuropsychological
Battery-Children's Revision; Halstead-Reitan Batteries; Hynd, Snow, &
Becker, 1986) and include a wide array of functional skills with more
direct measures of actual processes reflected in learning-disabled chil-
dren. A conceptual model for such an assessment exists (Hynd & Cohen,
1983; Obrzut, 1981).
CONCEPTUAL FRAMEWORK FOR CLINICAL EVALUATION
Obrzut (1981) proposed that neuropsychological assessment of chil-

dren generally should follow a modification of Johnson and Myklebust's
(1967) hierarchy of information processing. As modified further by Hynd
and Cohen (1983), this conceptual hierarchy may lead to a more orga-
nized, and thus systematic, method for clinical appraisal of children,
particularly those with learning disorders. Table 2 presents this concep-
tual framework.
This framework merely presents a way in which assessment pro-
cedures may be related more efficiently to systems important to adequate
functioning in children. It also suggests that by carefully selecting various
measures, one may assess across the domains of information processing
efficiently and in a manner more directly tied to a referral question. Thus,
the assessment becomes less time-consuming than traditional neuropsy-
chological batteries, less redundant (see Hynd et al., 1986), and more
germaine to the referral question. It also should be pointed out that one
must be familiar with a great number of assessment procedures across
these domains.
Specifically relevant to the topic of this chapter, however, is the
question: What does this hierarchy have to do with diagnosing subtypes
of learning disabilities? In light of our knowledge of the research reviewed
previously, it can be stated that most learning disabilities are diagnosed
on the basis of a significant discrepancy between measured intelligence
and academic achievement. Thus, in the context of this hierarchy, tests of
cognitive ability and general achievement should be administered to de-
termine (1) if there is such a discrepancy and (2) in what academic area.
Further relying on our knowledge of the literature, and presuming
that the disability is documented in the area of reading, it seems reason-
able on the basis of Kinsbourne and Warrington's (1963) study to select
tests of finger agnosia, laterality, and visual-motor ability. Other research
(Petrauskas & Rourke, 1979) suggests a subtype with motoric dysfunction.
Thus, the proposed battery also should include measures of motor matu-
rity, such as tests of fine and gross motor coordination, and dysmetria.
Clearly, though, the vast majority of the research in this area suggests
that the majority of children with reading disorders have difficulty with
auditory-linguistic processing (Pirozzolo, 1979), phonemic segmentation
(Liberman, Shankweiler, Liberman, Fowler, & Fischer, 1977), verbal flu-
ency, and comprehension. Their reading errors are often marked by fre-
quent paralexic errors as well (semantic, visual, and auditory) (e.g., Fried-
man & Perlman, 1982). Thus, the evaluation should include tasks of these
sorts drawn from the conceptual hierarchy which in turn are directly tied
to the neuroanatomic ally based neurolinguistic model presented in Fig-
ure 1.
TABLE 2. Conceptual Hierarchy and Some Associated Tests/Evaluation Procedures
I. Sensation and Sensory Recognition

Acuity Lateral Dominance-Motor Clinical Measures
Visual Grip Strength Test for Phonetic Sounds (nonsense
Auditory Edinburgh Inventory words)
Recognition Finger Tapping Test for Vowel Principles (nonsense
Finger Agnosia IV. Psycholinguistic words)
Finger-Tip Number Writing Screening Measures Syllabication (nonsense words)
Tactile Form Recognition Aphasia Screening Test Oral Reading Passages (graded)
II. Perception Fluency Test Try-Outs-Diagnostic Teaching
Auditory Peabody Picture Vocabulary Test Phonemic Segmentation
Speech Sounds Perception Test Formal Assessment Flash Vocabulary
Seashore Rhythm Test Illinois Test of Psycho linguistic Ability VI. Cognitive-Intellectual
Visual Boston Aphasia Examination Category Test
Bender Gestalt Test Language Asymmetries Raven's Coloured Matrices Test
Beery Visual Motor Integration test Dichotic listening Kaufman Assessment Battery for Children
Benton Visual Retention Test Visual half-field tasks (K-ABC)
Tactile-Kinesthetic Time-sharing tasks Wechsler Intelligence Scale for Children-
Tactual Performance Test V. Academic Revised (WISC-R)
III. Motor Comprehensive Battery
Cerebellar Screening Peabody Individual Achievement Test
Tandem Walking Woodcock-Johnson Psychoed ucational
Tests for Dysarthria Battery
Fine Motor Coordination Reading Tests
Tests for Nystagmus Durrell Analysis of Reading Difficulties
Gates-McKillop Reading Mastery Tests
Woodcock Reading Mastery Test
Consequently, the neuropsychological battery should (1) be con-

structed around the tasks that the research literature suggests these chil-
dren do poorly on, (2) be tied to a valid neurolinguistic model, (3) be
flexible such that depending on what the referral suggests, the battery can
be individually constructed, and (4) allow the clinician to relate perfor-
mance back to the subtyping literature such that a diagnosis consistent
with this literature, if possible, can be derived.
To carry our example to its final conclusion, suppose we selected an
intelligence test (WISC-R), a general achievement battery (PlAT), tests of
finger agnosia, laterality, motor functioning, tests for phonetic sounds,
vowel principles, and syllabication, an oral reading test in which we
would note the kind and type of errors, as well as tests for word knowl-
edge and passage comprehension (Woodcock Reading Mastery Tests). Let
us suppose we found that our reading-disabled child had good skills in
phonological coding, had poor comprehension, was able to read short
words better than long words, but had considerable difficulty in com-
prehension. No motor or tactile-kinesthetic deficits were noted. It was
found, however, that he did poorly on visual-perceptual tasks. This
clinical description fits well with what has been referred to by Deloche et
01. (1982) as the surface dyslexic. Thus, through a carefully selected test
battery, a knowledge of the subtyping literature, and an understanding of
the neurolinguistic model of reading, a consistent set of data may well
emerge. Further, with such a diagnosis, specific intervention strategies
may be particularly useful (see Hynd, 1986).
CONCLUSIONS
There has indeed been a long history, first in neurology and more
recently in psychology, in attempting to describe the manifestations of
learning disabilities in children. What first struck the early clinicians was
the similarities between these children's performance and that of brain-
damaged adults. Clearly, these disorders have a neurodevelopmental ori-
gin remarkably consistent with what was originally proposed by neu-
rologists nearly 100 years ago. Currently accepted definitions reflect this
perspective.
Concurrent with the involvement of psychologists in research with
these children, there has been a move away from the benefits that careful
clinical-observational research provides (e.g., Boder, 1973; Kinsbourne &
Warrington, 1963) toward a largely atheoretical multivariate approach.
Tests are often included in neuropsychological batteries for no apparent
reason, existing populations previously diagnosed as being disabled are
often employed without any screening as to the validity of the initial
diagnosis, and the resulting subtypes are often characterized by profiles
across variables that frequently correlate poorly, if at all, to actual defi-

ciencies on academic tasks.
It is clearly time to use these statistical tools, which indeed are
powerful, to evaluate theory, rather than construct post hoc explanations
that have little ecological validity. Clinically, subtypes should be identi-
fied according to actual deficient academic processes, evaluated using a
theoretically driven neuropsychological battery. For researchers, groups
of children should then be examined statistically as to how the subtypes
vary on neurolinguistic and neurocognitive variables deemed important
in cognitive maturation. Only in this fashion will researchers in this area
provide results that are meaningful to the clinicians who must work with
these children.
REFERENCES
Bannatyne, A. (1966). The etiology of dyslexia and the color phonics system. In J. Money
(Ed.), The disabled reader: Education of the dyslexic child. Baltimore: Johns Hopkins
Press.
Bastian, H. C. (1898). Aphasia and other speech defects. London: H. K. Lewis.
Bateman, B. (1968). Interpretation of the 1961 Illinois Test of Psycholinguistic abilities.
Seattle: Special Child Publications.
Benton, A. L. (1975). Developmental dyslexia: Neurological aspects. In W. J. Friedlander
(Ed.), Advances in neurology (Vol. 2). New York: Raven Press.
reading patterns. Developmental Medicine and Child Neurology, 15,663-687.
Breen, M. J. (1986). Cognitive patterns of learning disability subtypes as measured by the
Woodcock-Johnson Psychoeducational Battery. Journal of Learning Disabilities, 19,86-
90.
Claiborne, J. H. (1906). Types of congenital symbol amblyopia. Journal of the American
Coltheart, M. (1980). Deep dyslexia: A review of the syndrome. In M. Coltheart, K. Patterson,
& J. C. Marshall (Eds.), Deep dyslexia. Boston: Routledge & Kegan Paul.
Critchley, M. (1970). The dysexic child. London: William Heinemann Medical Books.
Crockett, D., Klonoff, H., & Bjerring, J. (1969). Factor analysis of neuropsychological tests.
Deloche, G., Andreewsky, E., & Desi, M. (1982). Surface dyslexia: A case report. Brain and
Language, 15, 12-31.
Denckla, M. B. (1973). Research needs in learning disabilities: A neurologist's point of view.
Journal of Learning Disabilities, 6, 44-50.
Doehring, D. G., & Hoshko, 1. M. (1977). Classification of reading problems by the Q-tech-
nique of factor analysis. Cortex, 13, 281-294.
Doehring, D. G., Hoshko, 1. M., & Bryans, B. N. (1979). Statistical classification of children
with reading problems. Journal of Clinical Neuropsychology, 1, 5-16.
Doris, J. (1986). Learning disabilities. In S. J. Ceci (Ed.), Handbook of cognitive, social, and
neuropsychological aspects of learning disabilities (pp. 3-54). Hillsdale, NJ: Erlbaum.
Drake, W. E. (1968). Clinical and pathological findings in a child with a developmental
learning disability. Journal of Learning Disabilities, 1,486-502.
Drew, A. L. (1956). A neurological appraisal of familial congenital word blindness. Brain, 79,
440-460.
Duffy, F. H., Denckla, M. B., Bartels, P. H., & Sandini, G. (1980). Dyslexia: Regional dif-
412-420.
Federal Register (1976). Education of handicapped children and incentive grants programs
(Vol. 41, p. 46977). Bethesda, MD: U.S. Department of Health, Education and Welfare.
Fisher, J. H. (1905). Case of congenital word-blindness (Inability to learn to read). Ophthal-
mic Review, 24, 315-318.
Fisher, W. F., Lieberman, 1. Y., & Shankeiler, D. (1978) Reading reversal and developmental
dyslexia: A further study. Cortex, 14, 496-510.
Fisk, J. 1., & Rourke, B. P. (1979). Identification of subtypes of learning disabled children at
three age levels: A neuropsychological, multivariate approach. Journal of Clinical
Neuropsychology, 1,289-310.
Fletcher, J. M. (1985). Memory for verbal and nonverbal stimuli in learning disability sub-
groups: Analysis by selective reminding. Journal of Experimental Child Psychology, 40,
244-259.
Friedman, R. B., & Perlman, M. B. (1982). Underlying causes of semantic paralexias in a
patient with deep dyslexia. Neuropsychologia, 20, 559-568.
Galaburda, A. M., & Kemper, T. 1. (1979). Cytoarchitectonic abnormalities in developmental
rology, 18, 222-233.
Geschwind, N. (1974). Anatomical foundations of language and dominance. In C. 1. Ludlow
& M. E. Doran-Quine (Eds.), The neurological basis of language disorders in children:
Methods and direction for research (NIH Publication No. 79-440). Bethesda, MD: U.S.
Department of Health, Education, and Welfare.
Geschwind, N., & Behan, P. O. (1982). Left handedness: Association with immune disease,
migraine, and developmental learning disorders. Proceedings of the National Academy
of Sciences, 79, 5097-5100.
Harris, T. 1., & Hodges, R. E. (Eds.). (1981). A dictionary of reading and related terms.
Newark, NJ: International Reading Association.
Hayes, F. B., Hynd, G. W., & Wisenbaker, J. (1986). Learning disabled and normally achiev-
ing college students on reaction time and speeded classification tasks. Journal of Educa-
tional Psychology, 78, 39-43.
Hinshelwood, J. (1895). Word-blindness and visual memory. Lancet, 2, 1564-1570.
Hinshelwood, J. (1900). Congenital word-blindness. Lancet, 1, 1506-1508.
Hooper, S. R., & Hynd, G. W. (1985). Differential diagnosis of subtypes of developmental
dyslexia with the Kaufman Assessment Battery for Children (K-ABC). Journal of
Hynd, C. (1986). Educational intervention in children with developmental learning disor-
ders. In J. E. Obrzut & G. W. Hynd (Eds.)' Child neuropsychology: Clinical practice (pp.
Hynd, G. W., & Cohen, M. (1983). Dyslexia: Neuropsychological theory, research, and
clinical differentiation. New York: Grune and Stratton.
Hynd, G. W., & Hynd, C. R. (1984). Dyslexia: Neuroanatomicallneurolinguistic perspectives.
Reading Research Quarterly, 19,482-498.
Hynd, G. W., Obrzut, J. E., Weed, W., & Hynd, C. R. (1979). Development of cerebral domi-
nance: Dichotic listening asymmetry in normal and learning-disabled children. Journal
of Experimental Child Psychology, 28, 445-454.
Hynd, G. W., Snow, J. H., & Becker, M. G. (1986). Neuropsychological assessment in clinical
child psychology. In B. B. Lahey & A. Kazdin (Eds.), Advances in clinical child neuro-
psychology (Vol. 9). New York: Plenum.
Hynd. G. W., & Willis, W. G. (1988). Pediatric neuropsychology. Orlando, FL: Grune and
Stratton.
Ingram, T. S., Mason, A. W., & Blackburn, I. (1970). A retrospective study of 82 children with
reading disability. Developmental Medicine and Child Neurology, 12, 271-281.
Jansky, J., & deHirsch, K. (1972). Preventing reading failure-Prediction, diagnosis, and
intervention. New York: Harper & Row.
Jackson, E. (1906). Developmental alexia (congenital word blindness). American Journal of
Medical Science, 131,843-849.
Johnson, D. J., & Myklebust, H. R (1967). Learning disabilities: Educational principles and
practices. New York: Grune and Stratton.
Kinsbroune, M., & Hiscock, M. (1981). Cerebrallateralization and cognitive development:
Conceptual and methodological issues. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsy-
chological assessment and the school-age child: Issues and procedures. New York:
Grune and Stratton.
Kinsbourne, M., & Warrington, E. K. (1963). Developmental factors in reading and writing
backwardness. British Journal of Psychology, 54, 145-156.
Kirk, S. A., & Bateman, B. (1962). Diagnosis and remediation of learning disabilities. Excep-
tional Children, 29(2), 73-78.
Kussmaul, A. (1877). Disturbance of speech. Cyclopedia of Practical Medicine, 14, 581.
Liberman, I. Y., & Shankweiler, D., Orlando, L. Harris, K. S., & Bertt, F. B. (1971). Letter
confusion and reversal of sequence in the beginning reader: Implications for Orton's
theory of developmental dyslexia, Cortex, 7, 127-142.
Liberman, I. Y., Shankweiler, D., Liberman, A. M., Fowler, C., & Fischer, W. F. (1977).
Phonetic segmentation and recoding in the beginning reader. In A. Reber & D. Scar-
borough (Eds.), Toward a psychology of reading. Hillsdale, NJ: Erlbaum.
Luria, A. R (1980). Higher cortical functions in man (2nd ed.). New York: Basic Books.
Lyle, J. G. (1969). Reading retardation and reversal tendency: A factorial study. Child Devel-
opment, 40, 833-843.
Lyle, J. G., & Goyen, J. (1968). Visual recognition developmental lag and stephosymbolia in
reading retardation. Journal of Abnormal Psychology, 73, 25-29.
Lyle, J. G., & Goyen, J. (1975). Effects of speed of exposure and difficulty of discrimination on
visual recognition of retarded readers. Journal of Abnormal Psychology, 8, 673-676.
Lyon, R, & Watson, B. (1981). Empirically derived subgroups of learning disabled readers:
Diagnostic characteristics. Journal of Learning Disabilities, 14, 256-261.
Marshall, J. 1., & Newcombe, F. (1973). Patterns of paralexia: A psycho linguistic approach.
Journal of Psycholinguistic Research, 2, 175-197.
Marshall, J. C., & Newcombe, F. (1980). The conceptual status of deep dyslexia: A historical
perspective. In M. Coltheart, K. Patterson, & J. C. Marshall (Eds.), Deep dyslexia (pp. 1-
21). Boston: Routledge & Kegan Paul.
Mattis, S., French, J. H.,& Rapin, I. (1975). Dyslexia in children and young adults: Three
independent neuropsychological syndromes. Developmental Medicine and Child Neu-
rology, 17, 150-163.
Mayeux, R, & Kandel, E. R (1985). Natural language, disorders of language, and other
localizable disorders of cognitive functioning. In E. R Kandel & J. H. Schwartz (Eds.),
Principles of neural science (2nd ed., pp. 688-703). New York: Elsevier.
McCarthy, J. M. (1975). Children with learning disabilities. In J. J. Galagher (Ed.), The
application of child development research to exceptional children. Reston: Council for
Exceptional Children.
McCarthy, J. J., & McCarthy, J. F. (1969). Learning disabilities. Boston: Allyn Bacon.
Meacham, M. 1., & Fisher, G. 1. (1984). The identification and stability of subtypes of
disabled readers. International Journal of Clinical Neuropsychology, 4, 269-274.
Morgan, W. P. (1896). A case of congenital word-blindness. British Medical Journal, 2, 1378.
Morris, R, Blashfield, R, & Satz, P. (1986). Developmental classification of reading-disabled
children. Journal of Clinical and Experimental Neuropsychology, 8, 371-392.
Naidoo, S. (1972). Specific dyslexia. New York: Wiley.

Nelson, H. E., & Warrington, E. K. (1974). Developmental spelling retardation and its relation
to other cognitive abilities. British Journal of Psychology, 65, 265-274.
Nolan, D. R, Hammeke, T. A., & Barkley, R. A. (1983). A comparison of the patterns of the
neuropsychological performance in two groups of learning disabled children. Journal of
Clinical Child Psychology, 12,22-27.
Obrzut, J. E. (1979). Dichotic listening and bisensory memory skills in qualitatively diverse
dyslexic readers. Journal of Learning Disabilities, 12(5), 24-33.
Obrzut, J. E. (1981). Neuropsychological assessment in the schools. School Psychology Re-
view, 10, 331-342.
Omenn, G. S., & Weber, B. A. (1978). Dyslexia: Search for phenotypic and genetic hetero-
geneity. American Journal of Medical Genetics, 1,333-342.
Orton, S. T. (1928). Specific reading disability-Strephosymbolia. Journal of the American
Patton, H. D., Sundsten, J. W., Crill, W. E., & Swanson, P. D. (1976). An introduction to basic
neurology. Philadelphia: Saunders.
Petrauskas, R, & Rourke, B. P. (1979). Identification of subgroups of retarded readers: A neuro-
psychological multivariate approach. Journal of Clinical Neuropsychology, 1, 17-37.
Pirozzolo, F. J. (1979). The neuropsychology of developmental reading disorders. New York:
Praeger Press.
Reynolds, C., & Gutkin, T. B. (1979). Predicting premorbid intellectual status of children
using demographic data. Clinical Neuropsychology, 1, 36-38.
Ross, A. O. (1976). Psychological aspects of learning disabilities and reading disorders. New
York: McGraw-Hill.
Ross, E. D. (1981). The aprosodias. Archives of Neurology, 38, 561-569.
Rourke, B. D. (1976). Issues in the neuropsychological assessment of children with learning
disabilities. Canadian Psychological Review, 17, 89-102.
Rourke, B. P., & Finlayson, M. A. J. (1978). Neuropsychological significance of variations in
patterns of academic performance: Verbal and visual-spatial abilities. Journal of Abnor-
mal Child Psychology, 6, 121-133.
Rourke, B. T. Young, G. C., & Flewelling, R W. (1971). The relationships between WISC
verbal-performance discrepancies and selected verbal, auditory-perceptual, visual-per-
ceptual, and problem-solving abilities in children with learning disabilities. Journal of
Clinical Psychology, 27(4), 475-479.
Sandoval, J., & Haapanen, R M. (1981). A critical commentary on neuropsychology in the
schools: Are we ready? School Psychology Review, 10, 381-388.
Satz, P., Friel, J., & Rudegeair, F. (1974). Differential changes in the acquisition of develop-
mental skills in children who later became dyslexic. In D. G. Stein, J. J. Rosen, & N.
Butters (Eds.), Plasticity and recovery of function in the central nervous system. New
Satz, P., & Morris, R (1981). Learning disability subtypes: A review. In F. J. Pirozzolo & M. C.
Wittrock (Eds.), Neuropsychological and cognitive processes in reading (pp. 109-141).
Sevush, S. (1983, February). The neurolinguistics of reading: Anatomic and neurologic
correlates. Paper presented at the annual meeting of the International Neuropsychologi-
cal Society, Mexico City.
Silver, A., & Hagen, R (1971). Memory and attention factors in specific learning disabilities.
Journal of Learning Disabilities, 4, 94-106.
Smith, F. (1978). Reading without nonsense. New York: Teachers College Press.
Smith, M. D., Coleman, J. M., Dokecki, P. R, & Davis, E. E. (1977). Recategorized WISC-R
scores of learning disabled children. Journal of Learning Disabilities, 10, 444-449.
Snow, J. H., Cohen, M., & Holliman, W. B. (1985). Learning disability subgroups using
cluster analysis of the WISC-R Journal of Psychoeducational Assessment, 4, 391-397.
Snow, J. H., & Hynd, G. W. (1985). A multivariate investigation of the Luria-Nebraska Neuro-
psychological Battery-Children's Revision with learning disabled children. Journal of
Psychoeducational Assessment, 3, 101-109.
Speece, D. L., McKinney, J. D., & Appelbaum, M. 1. (1985). Classification and validation of
behavioral subtypes of learning disabled children. Journal of Educational Psychology,
77, 67-77.
Spreen, 0., & Haaf, R G. (1986). Empirically derived learning disability subtypes: A replica-
tion attempt and longitudinal patterns over 15 years. Journal of Learning Disabilities,
19, 170-180.
Stephenson, S. (1905). Six cases of congenital word blindness affecting three generations of
one family. Ophthalmoscope,S, 482-484.
Sternberg, R, & Wagner, R K. (1982). Automatization failure in learning disabilities. Topics
in Learning and Learning Disabilities, 2, 1-11.
Strauss, A. A., & Kephart, N. C. (1955). Psychopathology and education of the brain-injured
child (Vol. 2). New York: Grune and Stratton.
Strauss, A. A., & Lehtinen, 1. E. (1947). Psychopathology and education of the brain-injured
child (Vol. 1). New York: Grune and Stratton.
Sweeney, J. E., & Rourke, B. A. (1978). Neuropsychological significance of phonetically
accurate and phonetically inaccurate spelling errors in younger and older retarded
spellers. Brain and Language, 6, 212-225.
Taylor, H. G., Fletcher, J. M., & Satz, P. (1982). Component processes in reading disabilities:
Neuropsychological investigation of distinct subskill deficits. In R N. Malatesha & P. G.
Aaron (Eds.), Reading disorders: Varieties and treatments (pp. 121-147). New York:
Academic Press.
Thomson, M. E. (1982). The assessment of children with specific reading difficulties (dyslex-
ia) using the British Ability Scales. British Journal of Psychology, 73, 461-478.
U.S. Department of Education, Office of Special Education and Rehabilitative Services
(1983). Fifth annual report to Congress on the implementation of PL 94-142: The Educa-
tion of All Handicapped Children's Act.
Warrington, E. K. (1981). Concrete word dyslexia. British Journal of Psychology, 72, 175-
196.
Watson, B. U., & Goldgar, D. E., & Ryschon, K. 1. (1983). Subtypes of reading disability.
Journal of Clinical Neuropsychology,S, 377-399.
Wernicke, C. (1874). Der aphasiche symptemkomplex. Breslaw: Cohn and Weigert.
Whitaker, H. A. (1976). Neurobiology of language. In E. R Carterette & M. P. Friedman (Eds.),
Handbook of perception (Vol. 7). New York: Academic Press.
Zangwill, D. 1. (1960). Cerebral dominance and its relation to psychological function. Edin-
burgh: Oliver & Boyd.
12
The Prediction of Learning

Disabilities in the Preschool Child
A N europsychological Perspective
STEPHEN R. HOOPER
INTRODUCTION
Although neuropsychological measures have not been developed specifi-

cally for the preschool child, the early identification and prediction of
learning problems are major challenges for the child neuropsychologist.
Working with such young children presents any array of problems, in-
cluding measurement difficulties, interpretive issues, and developmental
questions. This latter concern is particularly important given that even a
group of children having the same chronological age, gender, and ethnic
background will show a wide range of "normal" developmental variation
(Wolff, 1981). Assessment methods will need to be guided by a neu-
rodevelopmental theoretical framework if the study of brain-behavior
relationships in this age group is to be advanced.
Assuming that an assessment technology can be advanced that has
strong neurodevelopmental underpinnings, what services or treatment
linkages will be available to meet the needs of learning-disabled children
and their families? The Education for All Handicapped Children Act (PL
94-142) mandated that a free and appropriate education be provided to all
handicapped children from birth to 21 years of age, but it has been only
recently that infant and preschool services have come to the forefront
with respect to public school services. Funding issues notwithstanding, it
STEPHEN R. HOOPER Department of Psychiatry, University of North Carolina School of

Medicine, and Clinical Center for the Study of Development and Learning, University of
North Carolina, Chapel Hill, North Carolina.
313
314 STEPHEN R. HOOPER
is no accident that this area has not been advanced sooner owing to its
significant complexity.
Further compounding the difficulties confronted in the preschool
population, some investigators have taken the stance that disability la-
bels, such as learning disability, should be used sparingly with young
children, if at all, owing to the negative expectations that can arise (Al-
gozzine, Mercer, & Countermine, 1977; Foster, Schmidt, & Sabatino, 1976;
Foster & Ysseldyke, 1976). For example, preschoolers experiencing only a
temporary developmental delay may be misdiagnosed, and those working
with the child may begin to hold negative expectations of the child's
potential. This could create secondary problems of adjustment for such
children, confusion in their families, and self-fulfilling prophecies in the
form of actual learning difficulties. Given these concerns, it will be impor-
tant for developmentally appropriate, reliable, and valid neuropsycholog-
ical measures to be developed for this population. Such tools will prove
useful for the early identification and prediction of learning problems,
and in obtaining accurate descriptive accounts of a preschool child's
strengths and weaknesses in academic readiness.
This chapter will discuss the prediction of learning problems in the
preschool child from a neuropsychological perspective. Initially, the im-
portance of this area to child neuropsychology is presented, and this is
followed by a review of the literature addressing this question. Given
Aylward's discussion of the application of a neuropsychological perspec-
tive in infancy and early childhood (Chapter 9, this volume), this chapter
will be confined to children from approximately age 3 to age 5. Finally,
the literature is integrated to underscore important directions for neuro-
psychologists working with preschool children and to shed some light on
issues relevant to the early prediction of learning disabilities.
THE IMPORTANCE OF EARLY PREDICTION OF LEARNING

DISABILITIES
There are several major reasons that make the early identification and
prediction of learning disabilities important to the field of child neuro-
psychology. First, the extent of the problem is such that it cannot be
ignored by professionals working with preschoolers. Second, these chil-
dren appear to be at greater risk for a continuing downward spiral of
learning and perhaps social-emotional difficulties. However, it is reason-
able to assume that early identification and intervention can contribute to
eliminating or lessening the severity of later learning, emotional, and
behavioral difficulties that might arise. In conjunction with this view-
point, evidence also has been presented that timely, early intervention
actually may capitalize on sensitive neurodevelopmental periods during
the preschool years, thus maximizing the possibilities for a child to devel-
PREDICTION 315
op adaptive coping strategies. These various reasons for early identifica-

tion will be discussed in turn.
Extent of the Problem

Learning disabilities represent one of the major areas of study for the
child neuropsychologist (Rourke, 1983). However, advances in the field of
learning disabilities have been hindered by various definitional issues.
More generally, Behr and Gallagher (1981) have proposed that a more
flexible definition of what constitutes a handicap is needed for preschool
children with special needs, one that involves not only the extent of a
developmental variation but the type of variation. With respect to learn-
ing disabilities, recent efforts to clarify definitions have conceptualized
learning disabilities as a heterogeneous group of neurologically based
disorders (Hammill, Leigh, McNutt, & Larsen, 1981). Obviously, the diffi-
culties encountered in arriving at an accepted definition have contributed
to added problems in the understanding of learning disabilities and their
epidemiology.
Despite these definitional difficulties, the study of learning dis-
abilities has seen tremendous growth, particularly over the past 25 years.
This growth has been fueled by legislative action as well as by the in-
creased survival rates of many medically high-risk infants. Although not
surprising, the bulk of this study has occurred with the school-age child
and has not addressed the needs of the preschool population as intensely.
However, figures from the U. S. Department of Education (1984) revealed
that children with learning disabilities represented about 8.3% of all pre-
school special education recipients, making them second only to speech-
impaired students in terms of categorical dysfunction. When all children
receiving special education services are considered, learning-disabled
children become the number one handicapped group receiving services in
the public schools (Doris, 1986). These figures support the need for appro-
priate identification and prediction procedures for learning disabilities in
the preschool population.
Given the extent of this problem, the National Joint Committee on
Learning Disabilities (NJCLD, 1986) has put forth a list of recommenda-
tions for addressing the needs of preschool children. Specifically, the
NJCLD outlined personnel training recommendations, testing bias con-
cerns, and family needs. Suggestions also addressed concerns regarding
the need for systematic identification programs, appropriate assessment
technology and procedures, development of early intervention programs
and, generally, the need for more research to gain a better understanding
of learning dysfunction in preschoolers. These concerns are consistent
with a child neuropsychological perspective and they reflect issues rele-
vant to the accurate prediction of learning disabilities in the preschool
child.
Minimizing Educational Failure

A second major reason why this area is important to child neuropsy-
chology is the potential that a comprehensive neuropsychological evalua-
tion has for guiding interventions aimed at arresting continued academic
decline. The hypothesis is that the earlier a child showing learning im-
pediments can be identified, and the more detailed the profile descrip-
tion, the greater is the possibility that an intensive intervention program
will be able to address the child's needs in an adequate manner.
Support for this thinking has been mixed, but a significant positive
trend has been asserted. White (1986), in a meta-analysis of the early
intervention literature, concluded that early intervention has definite
positive effects for most children, but that there was no evidence of supe-
riority for anyone kind of intervention program. In another meta-analysis,
Casto and Mastropieri (1986) supported these findings and further noted
that children participating in the longer, more intensive programs showed
the most improvement. However, these investigators found inconclusive
evidence regarding the benefits of various factors, including early pro-
gram initiation (i.e., there actually was a surprising trend for better prog-
nosis to be associated with later entry), the importance of parental in-
volvement, and the degree of program structure (although there was a
trend that favored more structured programs). '
Despite these generally positive findings, Schonhaut and Satz (1983)
found equivocal evidence for the notion that early identification and
treatment of learning disabilities improves prognosis. In their review of
follow-up studies in the learning disability area, they noted that the aca-
demic prognosis is generally poor for children with early learning prob-
lems, and that many learning-disabled children continue to experience
learning difficulties as adults, even if they are occupationally or educa-
tionally successful.
Although more research is needed in this area, it appears that early
intervention programs do have a positive effect for most learning-disabled
children. By providing a detailed profile of a preschooler'S cognitive,
sensorimotor, and preacademic strengths and weaknesses, a comprehen-
sive neuropsychological evaluation could provide a solid foundation for
the development of an intervention plan designed to address the child's
specific educational needs. Conceivably, this should contribute to lessen-
ing the possibility of continued educational failure.
Minimizing Social-Emotional Difficulties

Bryan and Bryan (1986) asserted that learning-disabled children are
at greater risk for social adjustment difficulties than their normal achiev-
ing peers. Other researchers have suggested that these children are at
PREDICTION 317
greater risk than normal learners for developing secondary emotion-

al/behavioral disorders and dropping out of school (Rutter, Tizard, Yule,
Graham, & Whitmore, 1976; Spreen, 1978). Children with learning defi-
cits also have been shown to have a higher incidence of delinquency and
psychopathology, although these data remain strictly correlational and
are not as yet conclusive (Schonhaut & Satz, 1983). Nonetheless, the logic
for early identification of learning problems remains the same. If a child's
specific learning problems can be addressed comprehensively and inten-
sively at an early age, the better are the chances that the child will develop
a more positive self-image and show a healthy personality development.
Social-emotional difficulties can exert a pervasive effect on a child's
functioning. Where these difficulties are sparked by learning problems, it
would be important for early identification procedures to address a
child's learning needs accurately to minimize the potential for social-
emotional difficulties that may arise from repeated failure and frustration.
Certain kinds of neuropsychological profiles might carry a greater risk for
developing psychopathology, and planning for these difficulties may help
the child's adjustment. In this regard, some evidence has surfaced in the
school-age population showing potential relationships between particular
patterns of behavioral maladjustment and specific neuropsychological
profiles (Nussbaum & Bigler, 1986; Porter & Rourke, 1985; Speece, McKin-
ney, & Appelbaum, 1985; Strang & Rourke, 1985). Although these rela-
tionships have not been replicated in a preschool population, these find-
ings do underscore the potential contributions that a neuropsychological
perspective may have for identifying children at risk for social-emotional
difficulties.
Effects of Environmental Influences

A fourth important reason for child neuropsychologists to be in-
volved in the early identification of learning problems comes from the
evidence suggesting that children may be more sensitive to positive
change during early developmental periods. Currently, there are two ma-
jor lines of thinking on this issue.
One viewpoint suggests that relatively immature brain regions may
subserve functions that typically would be managed by other brain re-
gions when these are disrupted by an insult (i.e., neural reorganization).
The second line of thinking views the less crystallized brain systems in
the younger child as being more susceptible to environmental influence
and, subsequently, positive behavioral change (Rourke, Bakker, Fisk, &
Strang, 1983). Evidence for these suppositions has surfaced from both
animal (Bennett, 1976; Coss & Globus, 1978; Walsh, 1981) and human
research (Bakker, Moerland, & Goekoop-Hoefkens, 1981; Zihl, 1981). Re-
gardless of the processes involved in the acquisition or reacquisition of
function, the implication that optimal learning periods exist provides a

potential therapeutic window for the child neuropsychologist who is
working with the preschool child (Rourke et a1., 1983).
Summary
The above discussion highlighted the importance of learning dis-
ability prediction to the child neuropsychologist. The extent of the prob-
lem, the possibility of minimizing learning and emotional difficulties,
and the potential for making a valuable contribution to the treatment
programming of children at risk for learning disabilities all argue for more
active efforts with preschoolers on the part of child neuropsychologists.
The next section will provide a review of the current status of the field
with respect to research on the early prediction of learning problems.
THE PREDICTION OF LEARNING DISABILITIES: CURRENT

STATUS
There have been hundreds of studies that have addressed the predic-
tion of learning problems during the preschool years. For over two dec-
ades researchers have attempted to find the "best" predictor, or groups of
predictors, for identifying the preschool child with a learning problem.
Needless to say, this search has turned up a wide variety of "effective"
predictors, numerous assessment strategies and, characteristic of the
learning disability field in general, findings that are contradictory in
nature. These contradictory data have served to demonstrate the complex-
ity of the prediction process. This section will discuss several recent
literature reviews related to the preschool prediction of learning dis-
abilities. Several exemplary studies also will be presented that have made
major contributions to the early prediction efforts.
Recent Reviews of the Literature

To date several major reviews of the literature in this area have been
put forth. Most of these have focused on predicting learning problems (de
Hirsch, Jansky, & Langford, 1966; Horn & Packard, 1985; Jansky, 1978;
Mercer, Algozzine, & Trifiletti, 1979; Simner, 1983), and others have ad-
dressed the prediction of learning in the preschool child more broadly
(Barrett, 1965; Tramontana, Hooper, & Selzer, 1988). Examining the more
recent reviews reveals an inconsistent set of conclusions with respect to
the early identification literature.
Mercer et a1. (1979) reviewed early prediction studies extending from
1966 to 1977 that employed single and multiple predictor procedures.
PREDICTION 319
Although 70 studies were reviewed, only 15 were deemed appropriate for

their matrix analysis of classification hit rates. For single measures, such
as the Reading Subtest of the Wide Range Achievement Test, the Bender
Visual-Motor Gestalt Test, and an unusual birth history, Mercer et al.
found a median hit rate of 73%. For multiple instrument batteries (e.g., de
Hirsch Index) the median overall hit rate was slightly increased to 79%.
Mercer et a1. noted that teacher ratings tended to predict later learning
difficulties best (Le., approximately 84%). especially if the teachers were
provided with checklists containing items directly related to academic
learning.
These reviewers also stated that the literature supported the inclu-
sion of preacademic skills (e.g., letter recognition) in the preschool assess-
ment since these tasks accurately identified at-risk learners more often
than other predictor measures. Further, they noted that the timing of the
initial screening was potentially important, with administration during
the winter or spring of the kindergarten year having predictive advantages
over testing during the fall (e.g., allowing for the child's adjustment to
formal schooling and increased time for teacher observation). However,
the conclusions of this review were difficult to evaluate because no de-
scriptions were provided of the studies included, thus leaving any meth-
odological concerns unknown but crucial elements.
Simner (1983) discussed the potential warning signs of school failure
in the kindergarten child. Selectively reviewing about 18 studies con-
ducted from 1966 to 1982, Simner concluded that gross or fine motor
coordination, peer acceptance, basic language skills, and drawing/ copy-
ing errors were not effective predictors of later learning problems. Howev-
er, attention/distractibility. verbal skills, and printing errors did seem to
be effective predictors of later school failure. Despite these findings, Sim-
ner's review was limited in that it tended to focus on univariate predic-
tors. Global measures (e.g., IQ) and test batteries were omitted from con-
sideration with respect to their comparative effectiveness. Other
important issues also were neglected, such as multivariate prediction,
variation as a function of the academic criterion predicted, time of assess-
ment, prediction interval, and demographic factors.
In a well-designed meta-analysis, Horn and Packard (1985) con-
ducted a review of 58 studies predicting reading achievement during the
elementary school years. Most of the studies included in the meta-analy-
sis were conducted between 1960 and 1980. They found that the best
single predictors of reading achievement in grades 1 and 3 were measures
of general cognitive functioning, language, attention/distractibility, and
internalizing behavioral problems. Horn and Packard's review suggested
that measures tapping sensory-perceptual functions and soft neurological
signs (e.g., gross motor skills) were less effective predictors of later read-
ing problems.
Horn and Packard were cautious to note that predicting school suc-
cess or failure is a complex process, and that it is unreasonable to think
that a single predictor could account for all of the variance in school
achievement. They noted that even the best early predictors accounted for
less than 36% of the variance in elementary school reading achievement.
Nonetheless, as with the two previous reviews, the focus of the meta-
analysis was on univariate prediction, with only token consideration
being given to how various measures might be combined effectively to
increase prediction of learning difficulties. Also, like many of the early
identification and prediction studies, the Horn and Packard review
focused solely on reading problems and did not address other academic
areas. Further, the studies included in their review were not limited to
pre-first-grade prediction efforts. Finally, although the meta-analysis was
well done, this procedure likely masked the vast heterogeneity across
studies due to the nature of the technique (i.e., the need to compare the
investigations on a common metric) (Wilson & Rachman, 1983).
More recently, Tramontana et a1. (1988) provided a comprehensive
examination of the preschool prediction of later academic achievement.
This review focused on 74 studies conducted since approximately the
inception of PL 94-142 and included only studies in which the predictor
measures were administered prior to first grade. In general, effective pre-
dictors spanned a wide range of variables, which included cognitive,
verbal, sensory-perceptual, perceptual-motor, behavioral, and demo-
graphic domains. Further, the authors noted that predictive relationships
tended to vary according to the specific criterion variable being assessed.
Measures of letter naming, general cognitive ability, language, visual-
motor skills, and finger localization were found to be accurate predictors
of reading in grades 1 through 3. However, these findings can only be
viewed as preliminary because very few of the studies examined reading
in terms of its component parts (e.g., recognition, comprehension). The
one study that did examine reading more precisely found a differential
predictive pattern, with IQ being the best predictor of reading vocabulary
and the Bender-Gestalt being the best predictor of reading comprehension
(Wallbrown, Engin, Wallbrown, & Blaha, 1975). To complicate matters,
developmental factors seemed to contribute to an apparent shift in the
predictive power of a set of predictors. Satz, Taylor, Friel, and Fletcher
(1978) demonstrated this by noting the importance of visual perceptual
factors in the prediction of early reading recognition, but the relatively
greater importance of verbal abilities as children progressed through ele-
mentary school.
The Tramontana et a1. review found that math was predicted about as
well as reading, although it was not studied as extensively. Effective pre-
dictors included IQ, visual-motor skills, memory, attention, and auditory
comprehension of language. As with the reading criterion, selective com-
PREDICTION 321
ponents of mathematics rarely were considered. None of the studies in-

cluded in the review devoted primary emphasis to the preschool predic-
tion of mathematics achievement.
The Tramontana et al. review concluded that the research to date had
succeeded in providing a general set of preschool measures that allow for
a gross discrimination of children falling at the extremes of academic
ability during the early elementary school years. However, these measures
lack sensitivity in the identification of the child who initially appears to
be functioning normally but later develops a learning impediment.
Exemplary Studies
Although the early identification literature extends back as far as the
1920s (Smith, 1928), the interest in this area exploded during the 1960s.
Despite the large number of studies in this area, there have been only a
few exemplary works that have set the stage for current endeavors, partic-
ularly from a neuropsychological vantage point.
One of the first longitudinal efforts in the early prediction of reading
problems was initiated by de Hirsch et al. (1966). These investigators
examined the effectiveness of a battery of tests containing linguistic and
perceptual measures in the prediction of later reading achievement. Chil-
dren were tested at the end of their kindergarten year and re-evaluated at
the end of second grade. Jansky and de Hirsch (1972) reported that their
Screening Index accurately identified 79% of children with reading prob-
lems, although there was a large number of false-positive prediction errors
(i.e., 25%). Specific variables that were most predictive of second-grade
reading problems included letter naming, picture naming, word match-
ing, copying, and sentence memory. Generally, these authors stressed the
importance of language factors (e.g., verbal retrieval) as major predictors
of reading success.
Silver and Hagin (1972) also presented data from one of the first
longitudinal investigations on early identification. As early as 1960, Sil-
ver and Hagin postulated that the perceptual dysfunction typically ob-
served in reading disability was an extension of deficits in spatial and
temporal organization (e.g., deficits in visual discrimination, auditory se-
quencing problems, right-left confusion, finger agnosia). These investiga-
tors speculated that these deficits were associated with a lack of cerebral
dominance for language and/or they resulted from poorly modulated in-
terhemispheric transfer of information.
Silver and Hagin evaluated beginning first-grade students with a bat-
tery measuring neurological, psychiatric, psychological, social, and
achievement areas. A factor analysis of their battery was performed, and
five factors emerged that accounted for approximately 61% of the total
prediction variance. These factors were Auditory Association, Visual-
Neurological, Psychiatric Impairment, Chronological Age, and General

Intelligence. Working from their theoretical position, measures from the
top two factors, Auditory Association and Visual-Neurological, were com-
bined to form one predictive battery, the SEARCH (Silver & Hagin, 1976).
Not only has this screening battery been successful in predicting reading
problems at the end of second grade, but it has been the diagnostic foun-
dation for a prescriptive intervention program, TEACH (Hagin, Silver, &
Kreeger, 1976). Although the Silver and Hagin work was conducted with
entry-level first-graders, their investigations were noteworthy for specify-
ing underlying factors of early identification instrumentation, establish-
ing a theoretical perspective involving neurocognitive processing abili-
ties, and demonstrating efforts to validate a systematic intervention
approach based upon specific assessment technology.
A third early identification and prediction project, the Florida Longi-
tudinal Project (Satz et a1., 1978), represented one of the most comprehen-
sive longitudinal efforts in the early identification literature. Satz and co-
workers conceptualized that specific precursors to reading were related to
maturational factors affecting readiness. Reading problems were de-
scribed as initially reflecting delays in the early development of sensory-
perceptual processes and, subsequently, delays in conceptual-linguistic
skills. Generally, these delays were postulated to be related, in part, to
lagged maturation of the cerebral cortex (Satz & van Nostrand, 1973).
Satz and colleagues employed a broad-based, multivariate approach
to prediction using multiple follow-up points and numerous cross-valida-
tion studies. Their project was based on an original sample of 497 white
male kindergarten students in the Alachua County Public School System
in Florida. The predictor battery consisted of 22 measures, 16 of which
were grouped empirically into three factors. These included factors mea-
suring skills in the Sensory-Motor-Perceptual, Conceptual-Verbal, and
Verbal-Cultural domains. Taken together, these factors accounted for ap-
proximately 60% of the variance in reading prediction. Criterion mea-
sures included tests of reading recognition and teacher ratings of reading.
These investigators found about an 84% correct classification rate, with
finger localization, recognition-discrimination tasks, and alphabet recita-
tion being the best predictors of reading outcomes in second grade. At the
end of grade 5, a 76% correct classification rate was obtained, with the
Peabody Picture Vocabulary Test, the Developmental Test of Visual-
Motor Integration, alphabet recitation, and finger localization emerging as
the best predictors.
However, when these results were scrutinized more closely, it was
noted that although these measures were quite good at predicting the
severely disabled and the superior readers, the correct classification rates
fell significantly for the middle ranges of functioning. False negative rates
PREDICTION 323
(i.e., children who were functioning adequately in kindergarten, but who

later "grew into" deficits) ranged from about 5% in grade 1 to about 26% in
grade 5. This is disappointing, given that this was based on one of the best
sets of screening procedures that the field has to offer at the present time.
Despite these difficulties, the Florida Longitudinal Project represents one
of the most extensive efforts to date in the early identification and predic-
tion of learning problems.
In addition to the three studies above, two other efforts deserve men-
tion. Stevenson, Parker, Wilkinson, Hegion, and Fish (1976) conducted a
well-designed study and follow-up (Stevenson & Newman, 1986) in iso-
lating a differential set of predictors for reading and mathematics. Using a
comprehensive battery of cognitive, perceptual, memory, and academic
tasks, these investigators found the best kindergarten predictors of read-
ing to be letter naming, visual-auditory paired associates, Horst reversals,
and verbal categorization. The best predictors of math included verbal
recall, visual-auditory paired associates, perceptual learning, and coding.
This pattern of predictors was relatively stable for first-, second-, third-,
fifth-, and tenth-grade follow-up points. Not only is this study important
for describing a differential pattern of predictors for reading and mathe-
matics, but it also has one of the longest follow-up intervals reported to
date.
Finally, the last study to be discussed is important largely because of
its direct application of standard neuropsychological methods. Although
several studies have employed a neuropsychological perspective (e.g., the
Florida Longitudinal Project), Teeter (1985) used an established neuro-
psychological battery to predict first-grade achievement. Using the Mc-
Carthy Scales of Children's Abilities and portions of the Reitan-Indiana
Neuropsychological Battery, Teeter predicted first-grade achievement
with 57% and 61% accuracy, respectively. Generally, Teeter found the
Reitan-Indiana variables to be slightly more predictive than the McCarthy
for spelling, reading, and mathematics scores on standardized achieve-
ment testing.
Summary
There is an extensive literature dealing with studies that were de-
signed to obtain the best predictor, or set of predictors, for later learning
difficulties. However, it appears that this has been too simplistic a goal,
given the complexities of the learning process and its underlying neuro-
psychological mechanisms. The child neuropsychologist investigating
the early identification and prediction of learning disabilities during the
preschool years will need to address a more complicated series of ques-
tions. For example, a more sophisticated question might be: What com-
bination of measures, administered during specific times, are effective in

predicting specific outcomes for specific types of children at specific
points in their subsequent development? Future efforts in the early identi-
fication area should address such questions more thoroughly.
Previous work has laid the foundation for exploring the complexities
of this area, but these complexities are rapidly beginning to confront
practitioners in the preschool domain. Recent legislation (Le., PL 99-457,
the Education of the Handicapped Act Amendments of 1986) directly
addressing some of the service needs of the preschool population also will
drive increased efforts in this area. The literature does suggest the pres-
ence of broad neuropsychological constructs that could be incorporated
into a predictive preschool battery, but their specific predictive accuracy
will require further study. Also, the field would benefit from a greater
reliance on neurodevelopmental theory in the development of neuropsy-
chological assessment technology.
ISSUES AND DIRECTIONS
Although there have been efforts to apply neuropsychological assess-

ment methods to the preschool prediction of learning disabilities, the
bulk of the literature is not unique to the field of neuropsychology. In fact,
aside from the use of the Reitan-Indiana Neuropsychological Battery with
5-year-old children, even the instrumentation is not exclusive to the field.
In this regard there are few, if any, instruments specifically designed to
assess brain-behavior relationship in this population. Consequently,
then, what are the advantages of using a neuropsychological assessment
model over more traditional psychological or psychoeducational proce-
dures?
The advantages do not seem to be associated with particular instru-
mentation, although further development certainly is needed with respect
to this, but more with the perspective that can be employed in planning
measurement strategies and organizing assessment findings around spe-
cific neuropsychological constructs. The literature has begun to provide a
comprehensive list of broad-band neuropsychological constructs for this
population, but the assessment of these constructs in a developmental
fashion will require more than just the use of age-appropriate measures
having adequate normative bases. Any instrumentation must be sensitive
to how a child's abilities are increasing and changing (e.g., flexibility in
problem solving) and provide information relative to the normal and aber-
rant development of specific behaviors over time (e.g., academic precur-
sors )-even if a child is at a satisfactory level. The assessment also must
be grounded within a theoretical framework consistent with what is
known about preschool brain-behavior relationships.
PREDICTION 325
Neurodevelopmental Theory
It will be crucial for work in the area of early prediction to be guided
by theoretical conceptualizations. Knowing that a specific predictor-cri-
terion relationship exists is not enough; it also is necessary to know why
such a relationship exists. Although the efforts of Luria (1980), Rourke et a1.
(1983), Satz et a1. (1978), Silver and Hagin (1972) and, more recently,
Levine (Blackman, Levine, Markowitz, & Aufseeser, 1983; Levine &
Schneider, 1982) have set the stage for employing a neurodevelopmental
theoretical perspective, it is surprising that this has not been developed
more extensively. This is especially the case with learning disabilities,
given that current definitions of learning disabilities require at least the
suspicion of underlying neuropathology (Hammill et a1., 1981). Some work
already has been advanced noting neuroanatomical (Hier, LeMay, Rosen-
berger, & Perlo, 1978) and neurophysiological (Duffy, Denckla, Bartels, &
Sandini, 1980) anomalies as correlates of learning disabilities. These
anomalies have been implicated in contributing to the academic deficits in
learning-disabled children as well as underlying their deviant information
processing.
Despite the promise of these findings, certain aspects of neurodevel-
opment are better substantiated and should be incorporated into a neuro-
psychological perspective with this population. For example, it has been
argued that cerebral lateralization of function does not "develop," but
that it is present at birth (Kinsbourne & Hiscock, 1978). Indirect evidence
for this supposition has been provided by Gilbert and Climan (1974),
Ingram (1975), and Marcotte and LaBarba (1985) using dichotic listening
paradigms with preschoolers as young as 2% years of age. Anatomical
asymmetries, favoring the left hemisphere, also have been reported in the
neonate (Witelson & Pallie, 1973). These data suggest that left hemisphere
language lateralization exists from the outset, and thus set the stage for the
interpretation of hemispheric specialization and functional systems in the
preschooler.
At present there are at least two major neurodevelopmental concep-
tualizations that attempt to account, at least in part, for the later emer-
gence of a learning disability. The first model proposes a maturational lag
in cortical functioning (Satz & van Nostrand, 1973), and the other suggests
that the area of deficit is always present, but that deficiencies are man-
ifested only when specific demands challenge the processing capabilities
of the involved brain region (Rourke et a1., 1983). This latter position also
is consistent with the neurodevelopmental model espoused by Luria
(1980).
The model put forth by Satz (Satz & van Nostrand, 1973) provided
much of the theoretical guidance for the Florida Longitudinal Project
(Satz et a1., 1978), which investigated the precursors to later reading prob-
lems. This theory predicted delays in brain maturation that differentially

affected skills that were in primary ascendancy at different chronological
ages. Skills that develop during the preschool years, such as visual-per-
ceptual abilities and cross-modal sensory integration, were speculated to
be delayed because of immaturity in the cortical regions associated with
those skills and, consequently, to interfere with early aspects of reading
acquisition. Although these visual-perceptual deficiencies eventually
would resolve, skills that develop ontogenetically later in childhood and
at a slower rate (such as linguistic skills and formal operations) would be
delayed in immature older children and would continue hindering read-
ing development. Although this theory is noteworthy for its incorporation
of neurodevelopmental processes in the understanding of reading diffi-
culties, the concept of maturational delays in cortical development does
not address several important issues adequately in the prediction of learn-
ing-disabled children.
First, if the underlying premise is that of developmental delays in
cortical maturation, then a child should be expected eventually to "catch
up" with chronological-aged peers, or at least to narrow the gap in abil-
ity-achievement discrepancies over time. Present research would suggest
that this is not the case (Schonhaut & Satz, 1983). Although learning-
disabled children do show gains in their skills and abilities, they rarely
demonstrate an adequate level of academic functioning, and this is incon-
sistent with the concept of neuromaturational delay.
Second, if a child's specific learning problems are due to develop-
mental delays, then one should expect a relatively normal (albeit slow)
pattern of skill acquisition. However, evidence has surfaced that suggests
that not only are these children delayed in their skill acquisition, but they
manifest aberrant information processing styles as well (Aram & Nation,
1975). A model of neuromaturationallag does little to explain the deviant
processing abilities evidenced by many learning-disabled children. It
does not reflect the number and kinds of deficits that a child can exhibit at
a particular age.
Finally, the model does not provide any explanation for why many
"normal" children appear to "grow into" a deficit during later childhood.
Indeed, the data from the Florida Longitudinal Project support this con-
tention, in that by fifth grade the false-negative rate (Le., children initially
described as "normal" in kindergarten) climbed from 5% in grade 1 to
about 26% by grade 5. Despite the difficulties inherent with a neu-
romaturational lag hypothesis, it has provided insights into some of the
kindergarten precursors to reading.
A second major position, proposed by Rourke et 01. (1983), suggests
that the deficits seen in learning-disabled children are not due to neu-
romaturational delays but are always present and remain "silent" until
the affected region or functional system is challenged by environmental
PREDICTION 327
demands. It is at this time that the selective deficits associated with the
brain lesion or structural anomaly are manifested. This theory would
begin to account for the observation that a large percentage of "normal"
functioning preschoolers (about 26% in the Florida Longitudinal Project)
later show learning impediments. This theoretical position also would
not limit the identification of early deficits mainly to visual-perceptual
abilities and, consequently, would be more consistent with recent hetero-
geneous conceptualizations of learning disabilities than a neuromatura-
tional-delay model.
The implications for neuropsychological assessment based on this
theoretical model provide guidance in the search for neurodevelopmental
precursors to learning problems. It broadens the number of potential
neuropsychological constructs that may be relevant to the preschool child
and would provide an explanation for the large number of "effective"
predictors that have been asserted. It also would explain why a measure
seemingly unrelated to academics, such as finger localization, has been
consistently predictive of later reading difficulties (Lindgren, 1978; Satz
et 01., 1978).
Specifically, finger localization is felt to represent a measure of par-
ietallobe functioning (Reitan & Davison, 1974). Structures directlyassoci-
ated with the parietal lobe, such as the angular gyrus, have been directly
linked to reading skills (Gaddes, 1985). These tertiary cortical regions are
slower to mature and hierarchically are dependent upon intact function-
ing of the modality-specific primary and secondary cortical zones (Luria,
1980). Finger localization, which is a secondary-level parietal function,
may forecast a child's readiness to engage in tertiary-level functions, such
as reading, although little is known about its predictive value for specific
aspects of reading dysfunction or perhaps other kinds of academic deficits
(e.g., math).
Given the neurodevelopmental perspective offered by Rourke et a1.
(1983), it will be important for the field to begin to evolve new instrumen-
tation and procedures. This goes beyond establishing normative data for a
set of tasks and extends into describing more accurately the neurocog-
nitive underpinnings of learning and achievement precursors. Although
debates will continue regarding when a child's brain becomes fully func-
tional, it is clear that there is a developmental sequence in brain develop-
ment that can be tapped by behavioral tasks during the preschool years. It
is suspected that the more accurately these underlying neurodevelopmen-
tal processes can be assessed, the better the prediction of later learning
and behavioral outcomes will become.
For example, with respect to the development of behaviors associated
with the frontal lobes, Luria (1959) demonstrated that 3-year-old children
typically are unable to inhibit their behaviors using overt verbal media-
tion, but that by age 4 this would emerge and later evolve into internal
verbal regulation of behavior. This was replicated in a well-designed

study by Tinsley and Waters (1982). Similar studies have begun to exam-
ine normal developmental expectations for other neurocognitive func-
tions. Using Luria's (1980) model of neurodevelopment, PassIer, Isaac,
and Hynd (1986) and Becker, Isaac, and Hynd (1988) found evidence for a
developmental progression of frontal lobe development in subjects rang-
ing from about 5 to 12 years of age. Heverly, Isaac, and Hynd (1986) noted
a similar progression in normal children with respect to tactile-visual
discrimination functions associated with parietal lobe development.
These kinds of theoretically driven neurobehavioral tasks hold promise
for improving the predictability of later academic difficulties and the
description of specific brain-behavior relationships in this population.
These efforts are noteworthy with respect to understanding the neu-
rodevelopmental sequences of various cognitive functions, but they will
require further refinements if indeed they are to be applied clinically to
the preschool child. Evidence with respect to their predictive validity for
later academic functioning also will need to be demonstrated. This latter
point will be important not only for predicting the extremes of functional
outcomes but for charting their false-positive and false-negative predic-
tion rates.
Finally, with respect to neurodevelopmental theory, it is important to
remember that learning problems are not totally inherent to the child.
Other variables, such as socioeconomic status, family climate, and nur-
sery school experiences, should be evaluated in tandem with specific
neurodevelopmental variables so as to better gauge their interactive com-
plexities. Although highly important, neurodevelopmental variables only
account for a portion of the variance of later academic successes and
difficulties.
Neuropsychological Constructs and Preschool Prediction

Wilson (1986) conjectured that child neuropsychology is about 25
years behind adult neuropsychology, and perhaps this gap becomes even
wider when one considers the status of the neuropsychological assess-
ment of the infant, toddler, and preschooler. Although predicting learning
outcomes is not a simple matter, and no universal set of techniques has
been established, there are important clues that have emerged from the
existing prediction literature.
The evidence generated to date would not indicate an overall prefer-
ence for linguistic abilities, perceptual skills, or any other broad classifi-
cation of functions in the prediction of learning problems. However, con-
sistent with the mainstream of neuropsychological testing, assessment of
a wide array of functional domains would appear most useful for the
PREDICTION 329
clinician at this time. The addition of maturationally sensitive tasks also

would provide the foundation for determining predictive brain-behavior
relationships in preschool children from a neurodevelopmental perspec-
tive.
In this regard there have been several general (Deysach, 1986; Hart-
lage & Telzrow, 1982; Wilson, 1986) and specific (Korkman, 1987; Reitan
& Davison, 1974; Satz & Fletcher, 1982) clinical neuropsychological as-
sessment models proposed for use with the preschooler. Structured neu-
rodevelopmental examinations also have been developed that purport to
measure a broad range of functional areas in this population (Blackman et
a1., 1983; Levine & Scheider, 1982). Generally, these models espouse the
assessment of abilities covering auditory processing and language, memo-
ry, preacademic skills, adaptive behavior, behavioral ratings, motor func-
tions, and sensory-perceptual abilities. It is perhaps the measurement of
these latter two functional domains that distinguishes neuropsychologi-
cal assessment from other types of testing. However, with the exception of
the study by Teeter (1985) and the work of Satz in the Florida Longitudi-
nal Project, there is no evidence regarding the validity of any of these
neuropsychological assessment models in predicting later learning
problems.
For example, how well do Reitan's four methods of inference (Le.,
level of performance, pattern of performance, pathognomonic signs,
left/right differences) apply to the preschool population, particularly
given this population's increased performance inconsistencies? Is one of
these methods of inference more related to accurate prediction than an-
other? Although models currently exist for the neuropsychological assess-
ment of the preschool child, their applicability in predicting later learning
difficulties remains to be addressed empirically.
Factor-analytic work with preschool assessment batteries has pro-
duced factors that are remarkably similar to the areas of functioning gen-
erally included in a neuropsychological assessment. Jansky (1970) identi-
fied five factors in her factor analysis of 19 kindergarten tests. These
included Oral Language (A and B), Pattern Matching, Pattern Memory,
and Visual-Motor Integration. Silver and Hagin (1972) also identified five
factors from their preschool battery that included Auditory Association,
Visual-Neurological, Psychiatric Impairment, Chronological Age, and
General Intelligence. The Silver and Hagin factors are notable for their
psychiatric factor suggesting that there also may be social-emotional pre-
cursors to later academic problems. Satz et a1. (1978) identified three
factors in their work, which largely assessed skills in the Snesory-Motor-
Perceptual, Conceptual-Verbal, and Verbal-Cultural domains. These stud-
ies have found precursors of later learning difficulties, and the inclusion
of skills assessed by these factors would prove useful for a preschool
neuropsychological battery designed for predicting learning outcomes.
Taken together, the clinical and empirical studies support the devel-
opment of a predictive neuropsychological assessment model covering a
wide array of lower as well as higher cognitive functions. It is suggested
that such a battery include specific neurodevelopmental tasks (e.g., Pass-
Ier et al. 1986) that will provide behavioral indicators of brain develop-
ment and, ultimately, clues to better prediction of learning problems.
Needless to say, a preschool neuropsychological battery designed to pre-
dict later learning difficulties also should include a comprehensive medi-
cal, family, and developmental history. This will contribute to delineating
social-cultural difficulties as well as possible brain-based learning deficits
in the assessment and prediction of learning disabilities.
Related Issues in the Prediction of Learning Disabilities

In addition to constructing a neuropsychological assessment battery
designed to predict later learning impediments, several related issues
should be addressed. First, the literature has focused largely on reading as
the criterion variable. The need to extend predictive relationships to other
academic criteria is obvious. Further, it will be important for predictive
relationships to examine specific components of a criterion area such as
reading comprehension. The few studies that have addressed these con-
cerns revealed a differential pattern of results, not only among reading
components but between reading and mathematics (Stevenson et a1.,
1976; Wallbrown et a1., 1975). Efforts in the further refinement of criterion
variables also will provide evidence with respect to the neurocognitive
underpinnings of specific component skills and perhaps contribute to
improved treatment linkages.
A second issue is related to when the initial assessment or screening
should occur. Is there an optimal time for this evaluation? At present, the
available literature suggests that late kindergarten is the optimal time for
the initial assessment to occur. In many of the studies reviewed by Tramon-
tana et a1. (1988), this time period minimized the interval between predic-
tor and criterion measurements. More important, however, it seems that
late kindergarten is optimal largely because of the increased stability of
prerequisite skills, abilities, and response tendencies in the preschooler.
The later assessment time also allows the child to become more comfort-
able with the kindergarten setting, thus minimizing the effects of adjust-
ment to the novel preschool situation. However, the timing of initial
screening may be limited by current assessment technology. Using instru-
mentation developed around neurodevelopmental factors might allow for
an earlier, yet reliable, prediction of learning disabilities. This question
will become even more important as legislation begins to require the early
identification of handicapped youngsters (i.e., PL 99-457).
Another issue to be addressed is the need for the child neuropsychol-
PREDICTION 331
ogist to employ multiple points for follow-up testing. The work by Satz
and colleagues amply illustrates how many children could be missed if
multiple assessment points were not instituted. There are many children
who appear to "grow into" learning problems despite functioning within
a normal range on a kindergarten screening battery. It would be useful for
the child neuropsychologist to institute additional planned follow-up
points so as to monitor the children who fall in the middle ranges of
functioning during the preschool years. This would serve to identify chil-
dren who may be primed to become immersed in an insidious academic
decline secondary to neurodevelopmental anomalies, particularly if such
testing were designed to occur simultaneously with the shifting pattern of
academic demands. This type of ongoing assessment/screening process
recognizes the dynamic and complex aspects of learning. It would allow
for the incorporation of external variables into the assessment process,
such as teacher variables, classroom behaviors, psychosocial issues (e.g.,
family), and socioeconomic factors. With the development of more so-
phisticated methods for the neuropsychological assessment of the pre-
school child, it also may prove fruitful for tracking the topographical
continuity of various learner subtype patterns with respect to their
evolvement, prognosis, and treatment.
It is surprising, given the contradictory predictive findings currently
generated and the interactive complexities of the learning process itself,
that predictors of specific learning disability subtypes have not emerged.
Rourke (1983) noted that the subtyping of learning disabilities may be the
most pressing issue at this time in the learning disability field, and its
downward extension to preschoolers is only logical. This will surface
undoubtedly once assessment technology and neurodevelopmental un-
derpinnings are better integrated, and their predictive accuracy substanti-
ated.
CONCLUSIONS
This chapter has attempted to establish the importance of viewing the

early identification of learning disabilities within a neuropsychological
perspective. From the literature reviewed, it seems that the complex is-
sues regarding the prediction of learning problems during the preschool
years have only begun to be addressed. Noteworthy efforts by Satz et a1.
(1978), Silver and Hagin (1972), and others have set the stage for future
research in this area. However, it is clear that the field needs to receive
theoretical direction and, given the recent conceptualization of learning
disabilities as having a neurological basis, a neuropsychological perspec-
tive appears logical. In addition, a neuropsychological perspective would
be consistent with the concerns recently outlined for the learning-dis-
abled preschooler by the NJCLD (1986), particularly with respect to the

need for reliable diagnosis and prediction.
Currently, clinicians and researchers should assess a wide variety of
abilities, including sensory and motor functions, within the context of
theoretically driven assessment model. Investigators also should be striv-
ing to develop new instrumentation as well as refining traditional meth-
ods so as to increase our ability to delineate brain-behavior relationships
in the preschool population. Criterion measures should be more focused
and refined as well. The prediction of learning disabilities during the
preschool years is complex and, consequently, any neuropsychological
assessment strategy must be dynamic and multifaceted. If learning and
learning disabilities were to be viewed in a more dynamic fashion, sys-
tematic and aggressive follow-up testing of all children could be pursued
in an effort to minimize the number of children who seemingly "grow
into" a deficit (false-negatives) as well as those who initially show deficits
but later "catch up" (false-positives). This also will permit the optimal
time for performing initial predictive assessments to be determined more
accurately, and perhaps contribute to lowering the age of initial assess-
ment. Addressing these concerns will permit investigators to identify fu-
ture learning problems more effectively.
REFERENCES
Algozzine. B., Mercer, C. D., & Countermine, T. (1977). The effects of labels and behavior on
teacher expectations Exceptional Children, 44, 131-132.
Aram, D. M., & Nation, J. E. (1975). Patterns of language behavior in children with develop-
mental language disorders. Journal of Speech and Hearing Research, 18, 229-241.
Bakker, D. J., Moerland, R, & Goekoop-Hoefkens, M. (1981). Effects of hemisphere-specific
stimulation on the reading performance of dyslexic boys: A pilot study. Journal of
Barrett, T. C. (1965). Relationship between measures of prereading visual discrimination and
first grade reading achievement: Review of the literature. Reading Research Quarterly,
1, 51-76.
Becker, M. G., Isaac, W., & Hynd, G. W. (1988). Neuropsychological development of non-
verbal behaviors attributed to frontal lobe functioning. Developmental Neuropsy-
chology
Behr, S., & Gallagher, J. J. (1981). Alternative administrative strategies for young handi-
capped children: A policy analysis. Journal of the Division for Early Childhood, 2, 113-
122.
Bennett, E. R (1976). Cerebral effects of differential experience and training. In M. R. Rosen-
zweig & E. L. Bennett (Eds.), Neural mechanisms of learning and memory (pp. 279-
287). Cambridge, MA: M.l.T. Press.
Blackman, J. A., Levine, M. D., Markowitz, M. T., & Aufseeser, C. L. (1983). The pediatric
extended examination at three: A system for diagnostic clarification of problematic
three-year-olds. Developmental and Behavioral Pediatrics, 4, 143-150.
Bryan, T., & Bryan, J. (1986). Understanding learning disabilities (3rd ed.). Palo Alto:
Mayfield.
PREDICTION 333
Casto, G., & Mastropieri, M. A. (1986). The efficacy of early intervention programs: A meta-
analysis. Exceptional Children, 52, 417-424.
Coss, R. G., & Globus, A. (1978). Spine stems on tectal interneurons in jewel fish are short-
ened by social stimulation. Science, 200, 787-790.
de Hirsch, K., Jansky, J., & Langford, W. (1966). Predicting reading failure. New York: Harper
& Row.
Deysach, R. E. (1986). The role of neuropsychological assessment in the comprehensive
evaluation of preschool-age children. School Psychology Review, 15, 233-244.
Doris, J. (1986). Learning disabilities. In S. J. Ceci (Ed.), Handbook of cognitive, social, and
neuropsychological aspects of learning disabilities (pp. 3-54). Hillsdale, NJ: Erlbaum.
Duane, D. D. (1979). Toward a definition of dyslexia: A summary of views. Bulletin of the
Orton Society, 29, 56-64.
ferences in brain electrical activity by topographical mapping. Annals of Neurology, 7,
412-420.
Foster, G., Schmidt, C., & Sabatino, D. (1976). Teachers' expectancies and the label "learning
disabilities." Journal of Learning Disabilities, 9, 58-61.
Foster, G., & Ysseldyke, J. (1976). Expectancy and halo effects as a result of artificially
induced teacher bias. Contemporary Educational Psychology, 1,37-45.
Gaddes, W. (1985). Learning disabilities and brain function: A neuropsychological approach
(2nd ed.). New York: Springer-Verlag .
Gilbert, J. H. V., & CHman, I. (1974). Dichotic studies in two and three year olds: A prelimi-
nary report. Speech Communication Seminar, Stockholm (Vol. 2). Upsala: Almquist &
Wiksell.
Hagin, R. A., Silver, A. A., & Kreeger, H. (1976). Teach. New York: Walker Educational Book
Corporation.
Hammill, D. D., Leigh, J. E., McNutt, G., & Larsen, L. C. (1981). A new definition of learning
Hartlage, L. C., & Telzrow, C. F. (1982). Neuropsychological assessment. In K. Paget & B.
Bracken (Eds.), Psychoeducational assessment of preschool children (pp. 295-320).
New York: Grune and Stratton.
Heverly, L. L., Isaac, W., & Hynd, G. W. (1986). Neurodevelopmental and racial differences
in tactile-visual (cross-modal) discrimination in normal black and white children. Ar-
chives of Clinical Neuropsychology, 1, 139-145.
Hier, D. B., LeMay, M., Rosenberger, P. B., & Perlo, V. P. (1978). Developmental dyslexia:
Evidence for a subgroup with a reversal of cerebral asymmetry. Archives of Neurology,
35, 90-92.
Horn, W. F., & Packard, T. (1985). Early identification of learning problems: A meta-analysis.
Journal of Educational Psychology, 77, 597-607.
Ingram, D. (1975). Cerebral speech lateralization in young children. Neuropsychologia, 13,
103-105.
Jansky, J. J. (1970). The contribution of certain kindergarten abilities to second grade reading
and spelling achievement. Unpublished doctoral dissertation, Columbia University,
New York.
Jansky, J. J. (1978). A critical review of "Some developmental and predictive precursors of
reading disabilities." In A. L. Benton & D. Pearl (Eds.), Dyslexia. An appraisal of current
knowledge (pp. 377-394). New York: Oxford University Press.
Jansky, J. J., & de Hirsch, K. (1972). Preventing reading failure-prediction, diagnosis, inter-
vention. New York: Harper & Row.
Kinsbourne, M., & Hiscock, M. (1977). Does cerebral dominance develop? In S. J. Segalowitz
& F. A. Gruber (Eds.), Language development and neurological theory (pp. 171-191).
Korkman, M. E. (1987, February). NEPSY-A neuropsychological test battery for children,
based on Luria's investigation. Paper presented at the Fifteenth Annual Meeting of the
International Neuropsychological Society, Washington, DC.
Levine, M. D., & Schneider, E. A. (1982). Pediatric Examination of Educational Readiness.
Examiner's manual. Cambridge, MA: Educators Publishing Service.
Lindgren, S. D. (1978). Finger localization and the prediction of reading disability. Cortex,
14, 87-101.
Luria, A. R. (1959). The directive function of speech in development and dissolution. Word,
15,341-352.
Luria, A. R. (1980). Higher cortical functions in man (2nd ed.). New York: Basic Books.
Marcotte, A. C., & LaBarba, R. C. (1985). Cerebrallateralization for speech in deaf and normal
children. Brain and Language, 26, 244-258.
Mercer, C. D., Algozzine, B., & Trifiletti, J. J. (1979). Early identification: Issues and consid-
erations. Exceptional Children, 46, 52-54.
National Joint Committee on Learning Disabilities, The. (1986, February). A position paper
of the National Joint Committee on Learning Disabilities. Baltimore, MD: Orton Dyslex-
ia Society.
Nussbaum, N. L., & Bigler, E. D. (1986). Neuropsychological and behavioral profiles of
empirically derived subgroups of learning disabled children. International Journal of
Passier, M., Issac, W., & Hynd, G. W. (1986). Neuropsychological development of behavior
attributed to frontal lobe functioning in children. Developmental Neuropsychology, 1,
349-370.
Porter, J. E., & Rourke, B. P. (1985). Socioemotional functioning of learning-disabled chil-
dren: A subtypal analysis of personality patterns. In B. P. Rourke (Ed.), Neuropsychol-
ogy of learning disabilities: Essentials of subtype analysis (pp. 257-280). New York:
Guilford Press.
Reitan, R. M., & Davison, L. A. (1974). Clinical neuropsychology: Current status and applica-
tions. New York: Wiley.
Rourke, B. P. (1983). Outstanding issues in research on learning disabilities. In M. Rutter
(Ed.), Developmental neuropsychiatry (pp. 564-574). New York: Guilford Press.
Rourke, B. P., Bakker, D. J., Fisk, J. L., & Strang, J. D. (1983). Child neuropsychology. New
Wight studies, 1964-1974. Psychological Medicine, 6, 313-332.
Satz, P., & Fletcher, J. M. (1982). Florida Kindergarten Screening Battery. Odessa, FL: Psy-
chological Assessment Resources.
precursors of reading disabilities: A six year follow-up. In A. L. Benton & D. Pearl (Eds.),
Dyslexia: An appraisal of current knowledge (pp. 315-347). New York: Oxford Univer-
sity Press.
Satz, P., & van Nostrand, G. K. (1973). Developmental dyslexia: An evaluation of a theory. In
P. Satz & J. Ross (Eds.), The disabled learner: Early detection and intervention (pp. 121-
148). Rotterdam: Rotterdam University Press.
Schonhaut, S., & Satz, P. (1983). Prognosis for children with learning disabilities: A review
of follow-up studies. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 542-563).
New York: Guilford Press.
Silver, A. A., & Hagin, R. A. (1960). Specific reading disability: Delineation of the syndrome
and relationship to cerebral dominance. Comparative Psychiatry, 1, 126-134.
Silver, A. A., & Hagin, R. A. (1972). Profile of a first grade: A basis for preventive psychiatry.
Journal of the American Academy of Child Psychiatry, 11, 645-674.
Silver, A. A., & Hagin, R. A. (1976). Search. New York: Walker Educational Book Corpora-
tion.
PREDICTION 335
Simner, M. 1. (1983). The warning signs of school failure: An updated profile of the at-risk
kindergarten child. Topics in Early Childhood Special Education, 17, 17.
Smith, N. B. (1928). Matching ability as a factor in first grade reading. Journal of Educational
Speece, D. 1., McKinney, J. D., & Appelbaum, M. 1. (1985). Classification and validation of
behavioral subtypes of learning disabled children. Journal of Educational Psychology,
77, 67-77.
Spreen, O. (1978). Learning-disabled children growing up (Finale report to Health and
Welfare Canada, Health Programs Branch). Ottawa: Health and Welfare Canada.
Stevenson, H. W., & Newman, R. S. (1986). Long-term prediction of achievement and at-
titudes in mathematics and reading. Child Development, 57, 646-659.
Stevenson, H. W., Parker, T., Wilkinson, A., Hegion, A., & Fish, E. (1976). Longitudinal study
of individual differences in cognitive development and scholastic achievement. Journal
of Educational Psychology, 68, 377-400.
Strang, J. D., & Rourke, B. P. (1985). Adaptive behavior of children who exhibit specific
arithmetic disabilities and associated neuropsychological abilities and deficits. In B. P.
Rourke (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis
(pp. 302-328). New York: Guilford Press.
Teeter, P. A. (1985). Neurodevelopmental investigation of academic achievement: A report
of years 1 and 2 of a longitudinal study. Journal of Consulting and Clinical Psychology,
53, 709-717.
Tinsley, V. S., & Waters, H. S. (1982). The development of verbal control over motor behav-
ior: A replication and extension of Luria's findings. Child Development, 53, 746-753.
Tramontana, M. G., Hooper, S. R., & Selzer, S. C. (1988). Research on the preschool predic-
tion of later academic achievement. Developmental Review, 8, 89-147.
U. S. Department of Education, Division of Educational Services. (1984). Sixth annual report
to Congress on the implementation of Public Law 94-142: The Education for All Handi-
capped Children Act. Washington, DC: Author.
Wallbrown, J. D., Engin, A. W., Wallbrown, F. H., & Blaha, J. (1975). The prediction of first
grade reading achievement with selected perceptual-cognitive tests. Psychology in the
Schools, 12, 140-149.
Walsh, R. (1981). Towards an ecology of brain. Lancaster, PA: MIP Press.
White, K. R. (1986). Efficacy of early intervention. Journal of Special Education, 19, 401-
416.
Wilson, B. C. (1986). An approach to the neuropsychological assessment of the preschool
child with developmental deficits. In S. B. Filskov & T. J. Boll (Eds.), Handbook of
clinical neuropsychology (Vol. 2, pp. 121-171). New York: Wiley.
Wilson, G. T., & Rachman, S. J. (1983). Meta-analysis and the evaluation of psychotherapy
outcome: Limitations and liabilities. Journal of Consulting and Clinical Psychology, 51,
54-64.
Witelson, S. F., & Pallie, W. (1973). Left hemisphere specialization for language in the
newborn: Neuroanatomical evidence of asymmetry. Brain, 96, 671-696.
Wolff, P. H. (1981). Normal variation in human maturation. In K. J. Connolly and H. F. R.
Prectl (eds.), Maturation and development: Biological and psychological perspectives.
Clinics in Developmental Medicine, 77-78. Philadelphia: Lippincott.
Zihl, J. (1981). Recovery of visual functions in patients with cerebral blindness. Experimen-
tal Brain Research, 44, 159-169.
13
Electrophysiological Assessment in
Learning Disabilities
GRANT L. MORRIS, JOEL LEVY,
and FRANCIS J. PIROZZOLO
INTRODUCTION
This chapter will explore some of the nonbehavioral neurodiagnostic as-

sessment techniques that were introduced in Chapter 3 of this volume.
These techniques include neuroimaging techniques, such as comput-
erized tomography (CT) and magnetic resonance imaging (MRI), elec-
troencephalography, and evoked potentials. To illustrate the potential of
these tehcniques for assessment in child neuropsychology we have
chosen to focus specifically upon learning disabilities (LD).
The subtle and uncertain symptoms that characterize learning dis-
abilities provide a challenge to both conventional psychometric assess-
ment techniques and those nonbehavioral techniques that we will be
reviewing in this chapter. As new neuropsychological assessment tools
are introduced, often one of the first problems with which they are tested
is LD. For this reason, and despite the relatively short time that some of
these techniques have been around, considerable research already exists
using both neuroimaging and electrophysiology in the assessment of
learning disabilities.
Learning disabilities typically are not associated with any known
neuropathology or trauma. However, LD symptoms often appear con-
sistent with localized dysfunction in at least one brain system. Conse-
GRANT L. MORRIS. Department of Psychology, University of Northern Colorado, Greeley,

Colorado. JOEL LEVY Department of Psychology, Texas Institute for Rehabilitation
and Research, Houston, Texas FRANCIS J. PIROZZOLO Department of Neurology,
Baylor College of Medicine, and Houston VA Medical Center, Houston, Texas.
337
338 GRANT 1. MORRIS et 01.
quently, learning disabilities have long been suspected as having a neu-

rological basis. This presumed etiology was not made explicit in the
definition of LD found in Public Law 94-142. The definition of LD subse-
quently published by the National Joint Committee for Learning Dis-
abilities addressed this question: "These disorders are intrinsic to the
individual and presumed to be due to central nervous system dysfunc-
tion" (Hammill, Leigh, McNutt, & Larsen, 1981). At the same time, indi-
vidual learning-disabled children show such a variety of specific behav-
ioral anomalies that assessment often ends with cataloging the problem
behaviors rather than attempting to discover the underlying neurological
dysfunction.
Some of the purposes potentially served by electro physiological and
neuroimaging techniques in the assessment of LD include the following
(also see John, 1977; Otto et a1., 1984):
1. Providing insight into the neurological bases of LD. Numerous

models of aberrant brain function in LD point to problems in spe-
cific systems or regions of the brain. Will there be support from
electrophysiological and neuroimaging findings, or will the sus-
pected multiple etiologies of LD be confirmed by neurological cor-
relates as variable as the behavioral dysfunctions themselves?
2. Differential diagnosis of learning disabilities with neurological
involvement. Being able to confirm or exclude organic involve-
ment in a child's learning difficulties has clear implication for
therapy.
3. Early detection. This may lead to therapeutic intervention at the
onset or even prior to formal schooling rather than after the learn-
ing disability has manifested itself behaviorally.
4. Culture-fair assessment. This is a promise that follows from the
presumption that neuroimaging and electrophysiology are mea-
sures of brain status and functioning that are not heavily affected
by the child's cultural experience and aohievement.
5. Assessment of the course of the disorder and effect of therapy. An
assessment can be made to the extent that the disorder is the result
of, or at least correlated with, neurological status.
Anatomical Correlates of Learning Disorders

The roots of theorizing about anatomical bases of learning disorders
began with the observations of Broca (1861) and Wernicke (1874) on apha-
sia. Later, Hinshelwood (1895) found areas of cerebral cortex to be
damaged in patients with acquired reading disorders. However, it was left
for Morgan (1896) to suggest that these areas probably were under-
ELECTROPHYSIOLOGICAL ASSESSMENT 339
developed in children with developmental reading difficulties. Fisher

(1910) proposed specifically that underdevelopment of the left angular
gyrus might be responsible for developmental dyslexia. Orton (1937) sub-
sequently hypothesized that dyslexia in children might result from a "re-
versal of cerebral dominance." Little further investigation into the pres-
ence of a neurostructural abnormality associated with learning disorders
was carried out until the 1960s. In 1968, Geschwind and Levitsky called
attention to a statistically normal hemispheric asymmetry in the temporal
lobe, just posterior to the primary auditory projection zone, an area re-
ferred to as the planum temporale. This area is roughly equivalent to
Wernicke's area. Geschwind and Levitsky found the planum temp orale to
be larger on the left than on the right in autopsied adults. Witelson and
Pallie (1973) confirmed this asymmetry in young children from autopsy
specimens. Neuroimaging markers of this asymmetry have been reported
by LeMay and Kido (1978) and LeMay (1982) using computed tomogra-
phy (CT) scans, and by Kertesz, Black, Polk, and Howell (1986) using
magnetic resonance imaging (MRI).
This asymmetry appears to be a normal structural marker for domi-
nance. Kertesz reported that these anatomical measures classify subjects
for handedness with better than 95% accuracy. Departures from this
asymmetry may be markers for structural anomalies associated with
learning disorder, and as such, they may constitute modern support for
Orton's (1937) original premise. Several autopsy reports (Drake, 1968;
Galaburda & Kemper, 1979; Galaburda, Sherman, Rosen, Aboitiz, &
Geschwind, 1985) on learning-disabled patients have described devia-
tions from the normal left-greater-than-right asymmetry of the planum
temp orale and, in some cases, cytoarchitectonic anomalies. They have
found either that the planum temp orale was equal in size across hemi-
spheres or that there was a reversal of the asymmetry, with the right side
being larger. Hier, LeMay, Rosenberg, and Perlo (1978) examined the CT
images from 24 developmental dyslexics and found that 42% had a rever-
sal of the expected asymmetry. This neuroanatomical finding also corre-
lated with lower mean verbal IQ. Symmetry or reversal-of-expected asym-
metry has been observed in at least three subsequent studies. Denckla,
LeMay, and Chapman (1985), on the other hand, found only 20% of CT
scans to be abnormal, despite the presence of neurobehavioral evidence of
lateralized dysfunction.
Taken together, these studies do support a neuroanatomical basis for
learning disabilities. The ability to correlate structural abnormalities with
neuropsychological assessment will enrich our understanding of struc-
ture-function relationships and perhaps lead to new directions for assess-
ment and intervention in these disorders. What is needed is a comfort-
able, safe, precise, and preferably noninvasive method for neuroimaging.
Some of the currently available methods are discussed below.
340 GRANT L. MORRIS et a1.
Neuroimaging Methods
CAT Scans
The now conventional CAT or CT scan is an X-ray technique that
technically should be referred to as X-ray transmission tomography. This
is because CT only stands for computed tomography, and a tomographic
image can be derived from different sources, including X-ray absorption
by tissue, the emission of internally administered radioactive substances,
or magnetic resonance.
In X-ray transmission tomography, the patient's head is placed in a
large doughnut-shaped housing. Inside this housing, a movable X-ray
emitter is positioned, with a detector aligned to oppose it on the other side
of the ring housing. This configuration revolves around the patient's head
in small stepwise increments. It also moves up and down the long axis of
the patient's body to make images every centimeter from the top of the
head to the base of the skull. As the scanner moves around the head,
measurements of the amount of X-radiation absorbed by the tissue be-
tween emitter and detector are made and converted to numerical density
data. Each point in the head is scanned from at least two directions. The
image is based on the principle that the cerebrospinal fluid absorbs less
radiation than brain white matter, which absorbs less than gray matter,
which in turn absorbs about l/zoth of the amount that bone does. Flowing
blood absorbs an amount that overlaps both gray and white matter ranges;
therefore, to visualize intracranial circulation or defects in the venous
system (e.g., hemorrhage or thrombosis), a radiation-opaque contrast me-
dium is usually injected intravenously.
The radiation absorption values are stored in a computer, which re-
constructs an image of the brain and skull after the scans have been
completed. It is important to remember that the scan is not like a direct
visual X-ray film but is the trans axial absorption data translated into a TV-
type picture. The most recent scanners can resolve about a 1 mm X 1 mm
area of tissue. The technique involves approximately the same dose of
radiation as a standard skull radiograph. Inaccuracies or artifacts in the
image are sometimes produced by the beam hitting bony structures and
producing a shadow. However, these shadows actually may hide small
brain lesions. Other inaccuracies include failure to distinguish some
types of hemorrhages or tumors from normal brain tissue. However, this is
not typically a consideration in brain volumetric studies such as would be
proposed for an LD evaluation. Risks associated with this procedure in-
clude allergic reactions that some people experience to the contrast me-
dia. The exposure to radiation and the infusion of contrast material make
X-ray transmission tomography perhaps the most invasive neuroimaging
procedure.
ELECTRO PHYSIOLOGICAL ASSESSMENT 341
Positron Emission Tomography

Positron emission tomographs, or PET scans, also utilize radiation,
but they derive this energy source from radioactively labeled compounds
that are injected into the bloodstream. Rather than imaging only static
structures, PET scans image the dynamic, physiological function by de-
tecting radioactively labeled glucose usage in the tissue. The standard
compound used is fluorodeoxyglucose. This compound competes with
the normal sugar in blood that the brain needs to fuel its activity. It thus
concentrates in brain areas that are undergoing increased energy utiliza-
tion. This compound is a positron-emitting source that, upon encounter-
ing an electron, annihilates it with a release of electromagnetic energy.
This energy is detected with arrays of sensors that translate these data into
an image that reflects the localization and concentration of the radioactive
material in the brain.
Once again, this is not a direct radiation-exposed film image, but a
two-dimensional reconstruction of the energy data by means of computer
processing. The advantage of this technique is its visualization of energy
usage in the brain, which can be correlated with cognitive or motor ac-
tivity. The disadvantages are the high cost of producing and using the
radioactive compounds, the necessity of introducing these compounds
into the bloodstream, the length of time necessary to transport and utilize
these compounds in the tissue under study (which requires a sustained
cognitive or motor activity because the scan is not sensitive to transient
motor or cognitive events), and the limits of spatial resolution of the scan.
To date, PET represents strictly a research rather than a diagnostic
tool for the evaluation of learning disorders. Future investigators may be
able to use PET to diagnose LD subtypes on the basis of functional data
profiles, even in the absence of clear structural anomalies reported by CT.
Regional Cerebral Blood Flow

In much the same way that positron emission tomography is accom-
plished, regional cerebral blood flow (rCEF) detects blood-borne com-
pounds that emit radiation. The principle here is that areas of the brain
that are involved in intensive specific cognitive function require in-
creased blood flow to deliver nutrients for that functioning. The com-
pound frequently used in this technique is 133-Xenon, which typically is
inhaled. Sensors detect the radiation from this gas in the blood as it flows
through different regions in the brain. The gas then quickly diffuses out of
the body via exhalation from the lungs. The major advantages of this
technique are the speed of scanning and the correlation of blood flow with
cognitive activity. The disadvantages of the method are the limited resolu-
tion of the image and the use of a radioactive material, although the 133-
Xenon gas is thought to be relatively innocuous.
342 GRANT L. MORRIS et 01.
Like PET, blood flow studies remain at the investigational stage for
assessment of childhood cognitive disabilities. Normative studies on in-
ter- and intrahemispheric blood flow have been reported (Hynd, Hynd,
Sullivan, & Kingsbury, 1987), and limited data are now available on hemi-
spheric asymmetry. Despite the lower resolution for rCBF, increased left
hemisphere blood flow has been correlated with verbal test performance,
whereas performance on a nonverbal perceptual test was found to be
associated with increased right frontal and parietal flow. In one of the first
blood flow studies on dyslexic children, Hynd et a1. (1987) found that
there was less cerebral blood flow in the dyslexic groups, for both resting
and activation states, than for matched controls. Increased blood flow
during reading occurred in the left perisylvian and right parietooccipital
regions for the normal controls.
As the technology advances and resolution of structures becomes
more refined, this methodology offers the same promise as PET for diag-
nosis of subtypes of learning disabilities because it traces active function
and directly relates it to structure. Unlike PET, however, the ease of using
inhalable tracer compounds appears to be more comfortable and suitable
for young children.
Magnetic Resonance Imaging

Nuclear magnetic resonance imaging is a relatively new technique for
obtaining images of internal body structures. This method is now more
commonly known as magnetic resonance imaging (MRI) to dispel any
misconception that it involves subjecting patients to nuclear or ionizing
radiation. In MRI the body is exposed to a strong, encompassing magnetic
field. Atoms with odd numbers of protons, neutrons, or both will align
themselves like bar magnets when placed in a magnetic field, although
the atomic nucleus continues to spin like a top, with some wobble. This
wobble has a certain characteristic frequency, the "magnetic resonance"
frequency, which depends on the magnetic strength and mass of the nu-
cleus. When the nucleus is then pulsed with more magnetic energy, via
radio waves at the same resonant frequency, the nucleus "rings." The
radio wave pulse is tuned to the frequency of the atom under investiga-
tion, usually hydrogen because it is part of most chemical structures
found in the body, and the pulse times are adjusted to maximize the signal
from the nuclei. The "ringing" can be detected with receiving coils.
Two terms commonly involved in MRI are T1 and T2. These denote
the times when (a) the nucleus dissipates the ringing with respect to the
direction of alignment to the external, large magnetic field (T1), and (b)
the nucleus stops ringing coherently with the other hydrogen nuclei
around it (T2). These two types of resonance also are called spin-lattice
and spin-spin, respectively. Another term used frequently is inversion
recovery. This is when the radio wave pulse is directed through the tissue
from the opposite direction immediately after the initial pulse so as to

enhance the "ringing" even further. This method provides the sharpest
detail for differentiating between gray and white matter.
Very striking images that rival anatomical drawings for detail can be
obtained from MRI. Tumors, hemorrhages, ischemic regions, and de-
myelination sites all stand out more distinctly than with X-ray tomogra-
phy because MRI capitalizes on the subtle differences in chemical struc-
ture between these types of tissue. MRI has the added advantages that no
radiation is applied to the body, no contrast material is injected, and scans
may be done frequently without radiation risk. Further, structures do not
cause shadow artifiacts as they do in CT scans. No known risk to health or
cognitive function has been associated with exposure to magnetic fields of
the strength used in MRI. The only disadvantages now known are the
current expense of the procedure, the long times required for the scanning
during which the body must remain motionless or distortion artifact oc-
curs, and reports of claustrophobia from having to spend such long times
in the scanner.
Recent advances in this method allow for tuning of the device to
different molecules so that information on metabolism can be detected,
much like the PET scan but without the radiation involved. In our labora-
tory, we have undertaken a study of six children, referred to us from
speech pathologists, diagnosed with developmental dysphasia (Pirozzolo
et al. 1985). These children underwent neuropsychological testing, au-
diometric assessment, and neurostructural imaging by MRI. The children,
ranging from the low-average to the high-average range of intellectual
functioning, demonstrated decreased performance in verbally based cog-
nitive functions (e.g., vocabulary and comprehension), cognitive set shift-
ing, and arithmetic achievement. Abnormalities were found in the audito-
ry evoked responses (in the middle latency or late vertex wave forms)
consistent with central auditory dysfunction. Results of MRI scanning in
five of the six children revealed reversal of temporal structural asymmetry
consistent with CT and microscopic cytoarchitectural findings previously
reported. MRI provides a safe and precise means for identifying these
structural anomalies.
ELECTROPHYSIOLOGICAL TECHNIQUES AND FINDINGS
Electroencephalography
Basic Description
Electroencephalography (EEG) is a technique for amplifying and re-
cording the variations in the electrical potentials detectable on the scalp.
Exclusive of artifact generated by muscle activity (electromyographic ar-
tifact) in the vicinity of scalp electrodes, these potentials have their ori-
gins within the brain. This cortical activity is measured either as the
difference in potential seen at two scalp electrodes (bipolar recording) or
the activity at a single site referred to some other part of the body pre-
sumed to be electrically inactive insofar as EEG is concerned (referential
recording). Bipolar recordings are especially useful in localizing abnor-
mal EEG activity, e.g., associated with a tumor site. Referential recordings
are of less use in localizing abnormal EEG activity, but they are sensitive
to EEG activity whose origin lies at some distance from the active elec-
trode (so-called far-field recording). The signal itself is a complex wave
form that appears to have an unstable frequency and amplitude. However,
under constant conditions of recording site and arousal, a stable dominant
frequency can be seen. The dominant frequency of EEG may vary from 0.5
to more than 50 Hz. Most EEG amplitudes lie within a range of 5 to 50
microvolts. Much of the research employs a band-pass classification
scheme that labels specific frequency ranges. These typically include del-
ta (0.5-4 Hz), theta (5-7 Hz), alpha (8-12 Hz), beta (13-30 Hz), and
gamma (30-50 Hz).
A "montage" is a description of a specific set of electrode locations.
Until the establishment of the standard International 10/20 system of
electrode placement (Jasper, 1958), variations in electrode placements
confounded attempts at replications and comparisons among laboratories.
The letters of the 10/20 system refer to the major area of the brain over
Front
Al~
Lef:1J
Inion
Top of head
FIGURE 1. International 10/20 system of electrode placement.

which the electrode is placed: occipital (0), parietal (P), central (C), front-
al (F), and temporal (T). The subscript associated with each letter denotes
laterality, with odd numbers indicating left hemisphere placements, and
even numbers placement over the right hemisphere. The magnitude of the
number indicates the degree of displacement from the midline of the
brain, which is designated as zero (z). Although this convention is widely
followed, particular experiments may use different placements. However,
these unique placements typically are described in terms of their relation
to the 10/20 system. Figure 1 illustrates the 10/20 system for electrode
placement.
EEG Analysis Methods

Traditionally, analysis of the frequency and amplitude composition
of a paper EEG record was accomplished by visual inspection. This type
of summary analysis is typically accompanied by the examiner's report of
any anomalous EEG patterns or wave forms, such as the classic "spike
and dome" pattern associated with a petit mal epilepsy. Such laborious
manual methods of analysis must, of necessity, use only a small fraction
of the information in an EEG record, and consequently they are heavily
dependent upon the training and experience of the individual examiner.
Advances through Computerized EEG Data Analysis. In recent years

EEG methodology has been refined by the use of computer-based analytic
techniques. These techniques aim for the spectral decomposition of the
complex EEG signal through techniques such as fast-Fourier transforma-
tions (FFTs) or digital filtering, and the comparison of EEG activity among
various electrode sites through techniques such as coherence analysis.
The results from these efforts at measuring EEG from various electrode
sites and under different behavioral conditions in LD children have been
promising (see John, 1977).
The analogue voltage changes of the EEG are recorded on magnetic
tape for later analysis, or fed on line into an analogue-to-digital (AD)
converter. This device converts the continually varying voltages into a
series of numbers whose values are directly proportional to the instantan-
eous EEG voltages. The output of the AD converter is then fed into a
computer that can store or analyze these numeric equivalents of the origi-
nal EEG. Fourier transformations and digital filtering quantify the propor-
tion of the total power of the EEG wave form that lies within prespecified
ranges of frequencies or band passes.
Spectral analysis is the result of Fourier transformations or digital
filtering as described above. The frequency and power composition of
spontaneous EEG can be defined as a function of electrode site, task, and
diagnostic group of the participant. Further analyses include measure-
346 GRANT 1. MORRIS et a1.
ment of coherence and asymmetry of power among various electrode

sites. Coherence is a reflection of the similarity of EEG being produced at
the two electrode sites being compared. As such, coherence is essentially
the correlation of the instantaneous EEG voltage being produced at these
two sites. Asymmetry is usually expressed as the ratio of the EEG power
occurring at two sites within a particular frequency band pass. Typically,
homotopic electrode sites are used in coherence and asymmetry analysis.
Homotopic sites are the bilateral companion sites such as 0 1 and O 2 or T 3
and T 4
Differences in methodology among studies often are responsible for
the conflicting results in electrophysiological research:
... the study of EEG and AEP (averaged evoked potentials) characteristics of
learning-disabled children has not yielded consistent results over the last 30
years. This variability has arisen from problems both in the behavioral specifi-
cation of what set of variables characterizes learning disabilities. and from the
absence of a uniform system for acquisition and quantification of electro-
physiological data. It is not surprising that inconsistent and unreliable pro-
tocols in either domain will negate any attempts at brain-behavior correlation.
regardless of which domain is inconsistent or unreliable. and regardless of how
carefully the statistical analysis is carried out. (Kaye. John. Ahn. & Prichep.
1981. p. 16)
The variability in method for acquiring and analyzing electrophysio-

logical data from LD children, combined with the large amount of data
generated by these techniques, has nearly overwhelmed our capacity to
comprehend and assimilate the results. Consequently, the most helpful
approaches with learning-disabled children have been those that either
have offered methods for standardizing data collection protocols, such as
the neurometrics approach (John et 01., 1977; Kaye et 01., 1981) or have
developed new methods for graphically summarizing and presenting the
data, such as the brain electrical activity mapping (BEAM) technique
(Duffy, Denckla, Bartels, & Sandini, 1980; Duffy, Denckla, Bartels, Sand-
ini, & Kiessling, 1980).
The neurometrics approach consists of a battery of electrophysiologi-
cal measures, including spontaneous EEG along with auditory,
somatosensory and visual evoked potentials. Evoked potential stimuli
range from the rather simple, such as taps or flashing panels of light, to
those that are intended as significant probes ("challenges") of brain func-
tion, such as different geometric shapes and letters. This effort also is
commendable for the establishment of large normative data from normal
and special populations, including learning-disabled children.
The BEAM technique deals with the problem of extensive tables
filled with the numerical results of power or coherence analysis by repre-
senting these data as varying colors or textures that represent value as a
function of cortical area on a graphical representation of the head. (An
example of this technique of data presentation can be found in Chapter 3).
EEG in LD Children
Results of Analysis of EEG through Visual Inspection. In 1978, John
Hughes summarized work over the preceding two decades that looked at
EEG in hospitalized nondyslexic children as well as dyslexic children.
These studies used visual inspection to detect significant anomalies in the
EEG records. Ten of the reviewed studies involved a total of over 900
non dyslexic subjects. Across all of these studies, nearly half of the non-
dyslexic subjects were identified as having significant EEG anomalies. In
another ten studies during the same time period, a slightly smaller pro-
portion of dyslexic children were identified as having abnormal EEGs.
Clearly, the probability of false negatives and false positives was high.
Conners (1978), critiquing the studies covered in the Hughes review,
noted that the diagnostic category of subjects was often known to the EEG
reader, and thus the reliability of the EEG evaluation was questionable.
Conners found that the definition of dyslexia varied across studies, and
rarely was any attempt made to screen out possible associated behavioral
or neurological conditions. He observed that experimental procedures,
such as matching control and experimental subjects, were often poor or
absent, especially in the earlier studies. As experimental method im-
proved in the more recently reviewed studies, significantly less abnormal-
ity tended to be reported. He found that there was no consistent rela-
tionship across studies in terms of the locus or type of EEG abnormality,
and that there was often an inverse relationship between the degree of
behavioral impairment and the degree of severity of EEG abnormality. In
those few studies that employed a follow-up, the EEG contributed nothing
to the prediction of outcome.
Until that time the studies analyzing EEG through visual inspection
seemed to offer little consistent evidence to support a general conclusion
of an association between EEG abnormality and dyslexia. Although many
of the problems raised in the Hughes review have not been systematically
addressed by more recent studies, the use of automatic data analysis has
at least begun to address the problems presented by the visual inspection
technique.
Computerized EEG Findings in LD Children. Sklar, Hanley, and Sim-

mons (1973) conducted one of the first computerized EEG studies with
dyslexic children. In the place of visual inspection of paper EEG records,
the data were collected using frequency-modulation (FM) recorders and
magnetic tape. Later, these data were digitized and then read into the
minicomputer for Fourier analysis. This permitted reliable quantification
of the proportion of the power in the raw EEG within any frequency band.
Twelve dyslexic and 13 normal children matched for age and sex
participated. EEGs were recorded from both groups during various mental
tasks and rest situations. The spectral vectors derived from the Fourier
348 GRANT L. MORRIS et al.
transformations were used as predictor variables in a stepwise discrimi-

nant analysis. This analysis permitted the identification of the variables
most disparate between the two groups. The most distinctive differences
appeared in the parietooccipital region during the resting-with-eyes-
closed condition. The dyslexic subjects, on the average, had more energy
in the 3- to 7-Hz (theta) band and the 16- to 32-Hz (beta) band, whereas the
normals exhibited the greatest energy within the 9- to 14-Hz (alpha) band.
However, during the reading tasks, the spectral disparity between the two
groups was reversed at 16 to 32 Hz, with the normals tending to show
greater energy in this band.
During reading, the coherences of broad spectrum EEG among vari-
ous scalp leads were the most prominent discriminating feature. Within
the same hemisphere coherences between leads were higher in dyslexics
than in normals. However, between symmetrical regions across the mid-
line of the head (i.e., between homotopic electrode sites) coherence tend-
ed to be higher in normals. Thus, Sklar et a1. were able to differentiate
between dyslexic and normal age- and sex-matched children on the basis
of spectral estimates and coherence of their EEGs. However, their findings
did not fit well into any model that might predict cortical immaturity in
the dyslexic group, as reflected by general slowing of EEG, or a lack of
hemispheric specialization, which would predict less, rather than greater,
interhemispheric coherence in the normal group.
Rebert, Wexler, and Sproul (1978) undertook" an EEG study to deter-
mine if the left and right hemispheres of LD children responded differ-
ently to changing cognitive demands. The children included 11 residents
of a school for neurologically handicapped who were diagnosed as having
dyslexia. The other 11 participants were characterized as severely handi-
capped in oral language and were designated as dysphasic. Unfortunately,
normal control subjects were not available. The bilateral electrode place-
ments were at locations that were intended to overlie the angular gyrus in
the parietal region, and the planum temp orale in the temporal region. The
average and peak powers within the theta and alpha bands were com-
puted for the monitored sites. Subjects were tested in five conditions,
which included eyes closed-relaxed, eyes opened-relaxed, reading by
child, reading to child with child's eyes open, and while performing a
drawing task.
In the absence of an appropriate control group, the authors were
unable to assess the EEG maturation level of the participatns, but they felt
that there was a great deal of theta activity, and that the theta band was
most responsive to experimental manipulations. The Matousek and Pe-
tersen (1973) age-based norms clearly suggest that theta power declines
throughout childhood and adolescence, and that its greater-than-normal
presence in any age group is indicative of neurological immaturity.
Changes in EEG asymmetry among certain recording conditions dis-
tinguished the two diagnostic groups. A complex task, power, and diag-
nostic group relationship (as reflected in right and left parietal theta) was
most pronounced at the angular gyrus. A significant interaction with
groups led to the conclusion that the dyslexics had more left than right
power during reading-by-child that became less marked during drawing.
A discriminant analysis on theta power correctly classified 20 of the 22
children. An abundance of left theta during eyes open-relaxed and a slight
decline of that frequency during reading is consistent with Witelson's
(1977) supposition that dyslexics have a dysfunctional left hemisphere.
Fein et a1. (1983) recorded resting eyes-closed and eyes-open EEG in
normal and dyslexic 10- to 12-year-old boys. The EEG was recorded bilat-
erally from the central, parietal, and midtemporal regions. Two segments
of EEG were recorded which were separated by approximately a 4- to 5-
hour interval. Test-retest reliabilities were computed for each band and
for the entire spectrum, and reliabilities were assessed separately for each
condition and for absolute power and relative power. High reliability was
observed in the normal reading group. Reliabilities for the dyslexic group
were slightly lower but still were judged acceptable. No systematic EEG
differences were reported between the two groups. However, the authors
concluded that excellent reliability of both absolute and relative power
for the passive eyes-closed and eyes-open conditions can be obtained
from dyslexic children, and that this supports the utility of EEG power
spectra as a reliable index of brain functioning in studies of normal and
learning-disabled children.
The BEAM recording technique has been used to study a group of
pure dyslexic subjects and their age-matched controls (Duffy, Denckla,
Bartels, & Sandini, 1980; Duffy, Denckla, Bartels, Sandini, & Kiessling,
1980). These researchers found significant differences in patterns of EEG
activity between the two groups. Discrete brain areas that have been clas-
sically associated with speech and reading were identified in the dyslex-
ics as being significantly different from the control subjects. The dyslexic
children exhibited EEG anomalies in Broca's area and in the region of
intersection of the left temporal and parietal lobes, including the angular
gyrus and Wernicke's area.
Kaye et a1. (1981), using the neurometrics battery, evaluated the dif-
ferences in EEG and average evoked potential (AEP) characteristics of 50
children who were of normal IQ but who were underachievers either in
verbal academic performance, in arithmetic performance, or in both
(mixed). These underachievers were compared with matched controls
who were adequately achieving. They built the EEG portion of their neu-
rometrics battery around electrode sites and eyes-closed EEG relative
power age-related norms published by Matousek and Petersen (1973). The
EEG was recorded from four bilateral, bipolar locations, which included
the central, temporal, parietooccipital, and frontotemporal regions. Rela-
tive spectral EEG densities were derived for the delta, theta, alpha, and
beta frequency bands while eyes were closed. Age-normed and Z-trans-
formed measures of delta, theta, alpha, and beta wave shape coherence
and amplitude symmetry were available for each region.
In comparison with the matched controls, the poor verbal achievers
showed significant deviations only in temporal coherence and symmetry.
The poor arithmetic achievers exhibited marked deviations in several
bilateral regions across several of the frequency bands. The mixed under-
achievers showed yet another pattern of differences in comparison with
the controls. Their pattern included a larger number of left hemisphere
frontotemporal, temporal, and parietooccipital differences, and marked
frontotemporal incoherence and asymmetry. Kaye et a1. were successful
at replicating these findings with a larger sample of 215 children. These
highly variable patterns of group-specific EEG anomalies support the pre-
sumption of multiple etiologies of learning disability. Further, they also
may form the basis for significant discrimination between LD and normal
subjects, and among the subcatagories of LD.
In an earlier presentation of LD data using this approach, John (1977)
discussed the problem of unreliable, anecdotal, or psychometric diag-
nosis of learning disability when attempting to determine the elec-
trophysiological correlates of this disorder. Based upon the presumption
of accurate anecdotal diagnosis, or "prediagnosis" made by the teacher or
referring clinician, discriminant analysis of LD and normal children was
conducted using either a battery of psychometric instruments or the EEG
data collected as part of the neurometrics battery. Instruments found to
contribute to the multiple discriminant function included subscales of the
Wechsler Intelligence Scale for Children, the Wide Range Intelligence
Test, and the Peabody Picture Vocabulary Test. Overall accuracy of classi-
fication of normal and LD children based on the psychometric battery was
71 %. Selected EEG measures from the ueurometric battery yielded a 77%
accuracy in classification.
At that point, John had classifications of all subjects into normal or
LD categories based upon three independent criteria: the original classifi-
cations or "prediagnoses," the assignment to group by the psychometric
battery, and the group assignment by the EEG discriminant function. By
comparing the concordance of classification by the three schemes for each
child, John proposed some details of various etiologies. Children who
were classified as normal by prediagnosis and EEG measures, but classi-
fied as LD by psychometric criteria, were presumed normal; in these
cases, the psychometric-based LD classification was viewed as the result
of cultural factors or bias in the psychometric instruments. These ac-
counted for 5% of his sample. An additional 6.5% of his sample were
classified as normal by prediagnosis and psychometrics but were classi-
fied as LD by the neurometrics EEG. Although none of the children in his
study was initially prediagnosed as normal and subsequently classified as

LD by both psychometrics and neurometric EEG, a total of 11.5% of pre-
diagnosed normal children were classified as LD by one of the two bat-
teries. The 13% of his sample who were prediagnosed as LD, but classified
as normal by the other two measures, were presumed to be neurologically
normal, with their learning difficulties stemming from emotional prob-
lems. The 9% who were classified as LD by the pre diagnostic and EEG
criteria, but normal by psychometric criteria, were presumed to be LD and
to have been missed on the psychometric criteria because of the narrow
scope of functions assessed.
The neurometrics approach has seen mixed reviews. Yingling, Galin,
Fein, Peltzman, and Davenport (1986) reported that the neurometrics EEG
tests, supplemented with a subset of evoked potential tests, were not able
to discriminate dyslexics from controls. This failure was attributed to the
stringent screening applied to both groups in the study, designed to en-
sure that all subjects were of normal intellectual, neurological, sensory,
and emotional status. They concluded that pure dyslexia is not associated
with neurological anomalies detectable by the neurometric battery, and
that earlier indications to the contrary were artifactual.
Lubar et a1. (1985) conducted a study with boys classified as learning-
disabled according to the Tennessee state definition of this disorder. A
bipolar electrode montage was used bilaterally in the frontal-temporal,
temporal, and occipital-parietal regions. The EEG was recorded in each of
seven different conditions, which included baseline (relaxed, eyes open),
two levels of reading, two levels of math, and hard and easy puzzles. The
fast-Fourier transformation analysis produced percent and power values
of the EEG broken down into seven frequency bands of 4-Hz width from 0
to 28 Hz. It was found that learning-disabled children had more power in
the theta band (4-8 Hz), the high-thetallow-alpha bands (6-10 Hz), and in
the high-beta band (20-28 Hz).
The theta and high-theta low-alpha bands generally distinguished
normal from LD children, but other EEG frequencies did not. An EEG
recorded during the puzzles task provided the greatest discrimination
between the diagnostic groups. For the measure at the frontal-temporal
locations F7 -T5 and F8-T6 the analysis identified 79.2% of the normals
and 75.6% of the LDs correctly. For central locations the discrimination
was lower. Correct classification in the study was 68.8% for the normals
and 64.4% for the LD children. An EEG from the occipital locations cor-
rectly classified the normals 71% of the time and the LDs 75% of the time.
These results indicated that the frontal-temporal data provided slightly
better group discrimination than posterior data, and that significant slow-
ing of EEG frequency seemed to characterize the LD subjects. The neu-
rometric data, by contrast, suggested that EEG slowing in the occipital
region is most discriminative of LD (John, 1977).
Summary
Recent reanalysis of the Matousek and Petersen (1973) developmen-
tal EEG changes in normals further confirms that there is an age x cortical
area x frequency interaction, with the modal EEG frequency moving pro-
gressively upward with age. This frequency-shift process (Le., matura-
tion) occurs earliest and most rapidly in the occipital region, but it is
considerably delayed in the frontal-temporal region in normals. At all
electrode sites the pattern of change seen within the EEG frequency spec-
trum is a progressive decline of delta and theta power acco~panied by an
increase in alpha and beta between birth and age 19 to 20 years (W. H.
Hudspeth, personal communication, June 1987). Alpha is the dominant
frequency in more developmentally mature individuals when resting
with eyes closed. In normals, alpha has replaced delta-theta as the domi-
nant frequency band in the pareitooccipital region by age 8 years.
Much of the research reviewed in this section is consistent with the
concept of cortical immaturity in LD as revealed by a "slowing" of EEG
activity. This is most frequently seen as higher-than-normal theta band
power for a given age. This finding is most stable during the recording of
eyes-closed spontaneous EEG in the posterior region (Lubar et a1., 1985;
Rebert et a1., 1978; Sklar et a1., 1973). The Lubar et a1. (1985) findings also
suggested that, while engaged in an activity, the theta EEG from the front-
al region permits considerable classification accuracy of LDs.
Rebert et a1. (1978) and Duffy, Denckla, Bartels, and Sandini (1980)
reported specific lateralized differences in LDs in the areas of the left
temporal-parietal cortex associated with language and reading skills, par-
ticularly during tasks that would activate these areas. Although the fact of
general cortical EEG "slowing" seems established, the more intriguing
findings of lateralized temporoparietal anomalies in LD children await
replication and further definition.
Evoked Potentials
Basic Description
An evoked potential (EP) is the electrical indication of the brain's
reception of, and response to, an external stimulus. Typically, EPs are not
visible during the ongoing EEG recording. This is because of the relatively
low amplitudes of most EPs (0.1-20 microvolts) and their presence with-
in normal spontaneous EEG from throughout the brain and other artifacts
(such as electromyographic activity) that may be as much as several hun-
dred microvolts in amplitude. Separation of the buried EP wave forms
from these unwanted artifacts is accomplished by computer signal averag-
ing. It is assumed that the electrical response of the brain to the stimulus
(the EP) always comes at the same interval of time after the stimulus, and
thus has a constant latency. It also is assumed that the other electrical
activities present are not coupled to the stimulus and may be considered
random. Because these assumptions appear to be valid, computers (signal
averagers) can be used to isolate the evoked potential from the temporally
random background activity. Stimuli are given repetitively, and the com-
puter averages the new data acquired after each stimulus (random EEG
plus EP) with the averaged results from previous stimuli stored in its
memory. This averaging process is continued until the random EEG and
other artifacts have been reduced to an acceptably low level, and the
desired evoked potential has "grown" by virtue of the summation of its
constant temporal characteristics relative to the eliciting stimulus over
many trials (Chiappa, 1983). When averaged in this manner, visual, au-
ditory, and somatosensory stimuli are seen to evoke a series of discrete
electrophysiological responses or "peaks" of characteristic latency for a
given set of stimulus parameters. In their early (short latency) stages, these
peaks reflect the progress of the stimulus-evoked neural response through
the series of synaptic junctions within the anatomical pathway of the
stimulated modality. As the latency of these peaks extends past 100 msec,
and their loci shift from peripheral or brain stem sites to the cortex, the
peaks become more variable in latency and amplitude, more diffuse in
terms of anatomic origin, and more sensitive to cognitive state and
attention.
Exogenous Evoked Responses. Up to a latency of approximately 100

msec after the presentation of the eliciting stimulus, there is a relatively
clear linkage between the characteristics of the evoked response and the
specific sensory pathway. Further, most researchers are reasonably com-
fortable with assigning the origin of the various components of evoked
responses to specific anatomical regions along the sensory pathway fol-
lowed by that specific modality. These qualities of the short latency com-
ponents of an evoked response have led to their being characterized as
"exogenous" (John, 1963). They are highly stimulus/modality/anatomic
pathway-specific and independent of factors such as state of conscious-
ness and attention. The clinical utility of these early components lies in
their ability to reflect the integrity and functioning of the neuroanatomical
pathway they follow. "Changes in latency or amplitude, or the lack of
distinct waves, reveal the malfunctioning of that receptor, pathway, or
brain area which the particular component represents" (Rockstroh, Elbert,
Birbaumer, & Lutzenberger, 1982, p. 3). Although both visual and audito-
ry stimuli have been used to study middle- and long-latency evoked re-
sponses in dyslexic children, exogenous evoked response work with this
population has employed auditory stimuli almost exclusively.
BSER. Brain stem evoked responses (BSER) are classified as ex-

ogenous evoked potentials. BSERs are the series of polyphasic voltage
changes that occur within the first 10 msec after the presentation of a
brief, discrete auditory stimulus. The measured potentials are time-locked

to the presentation of a stimulus and require computer averaging to ob-
serve clearly the small discrete electrical responses from the brain. More
specifically, the evoked electrical activity is often recorded from a set of
four electrodes placed externally on the subject's scalp. One electrode is
located on the vertex (reference site), one electrode is located on each of
the subject's earlobes (active site), and the fourth electrode (ground) is
placed on the forehead. The electrophysiological response can be evoked
by simple auditory signals (e.g., clicks) or more complex auditory stimuli
MS EEG SER IES

10
~o '1J[
ACOUSTIC
IT
COC HLEAR S ;PRtORi
ml LA1ERAl jN~ERIOR MEO I A ~
l:m
IAUDITORY
ERVE '.1 .. NUCLEI OllVAR t /lE MN ISCUS COLllCUlUS GENICULATE I RACIATIONS
(MEDULLA) cOVP. E)I (PO~S)
I (M IDBRA.IN) tTH " lMUS) (l HALA.MO
I
,po S) / CORTICA L)
OUlEIit
sv".'"
01.1 ..... ""
CO UPLEJI
I
t
I!
I
Nu"C~I:U5
! ~ :~~~:~U5
: OJ
"'(DIAl
C; t;NI CU,,&l(
lOOT
VUtTlitAL DOill5"~
COCKL{ ". COCt4lt ,."
III UCUUS NUCLEUS.
FIGURE 2. Far-field recording of auditory brain stern responses latencies measured in human
subjects. Proposed functional-anatomical correlations.
such as phonemes. This is amplified for each test run, with the response
to the auditory stimulus being averaged over trials. Stimuli may be pre-
sented either monaurally or binaurally.
BSERs are thought to be the far-field reflection of sequential electrical
events at successively higher levels in the brain stem auditory pathway.
Depth recordings and lesion experiments in the feline suggest that the
approximate neural generators of waves I, II, III, V, and VI of the BSER are
the acoustic nerve, cochlear nuclei, superior olives, inferior colliculi, and
medial geniculate nuclei, respectively (Picton, Hillyard, Krausz, & Galam-
bos, 1974). The origin of other components of the polyphasic BSER are
less clear, but some authors speculate that late components have their
origins in the thalamic nuclei and auditory cortex (Stockard & Rossiter,
1977). However, Chiappa (1983) concluded that the origins of peaks V, VI,
and VII are still very much in doubt, and because waves VI and VII are
unreliable in appearance, they are not used in clinical interpretation.
These proposed functional-anatomical associations are illustrated in Fig-
ure 2.
Until recently, BSERs were evoked by bilateral auditory stimulation.
This practice has been criticized on the clinical grounds that a lateralized
dysfunction might be masked through blending the abnormal BSER from
the affected side with the BSER from the unaffected side (Chiappa, 1983).
Consequently, it is now standard practice to elicit BSERs with monaural
stimulation. This practice also opens the door to tlie discovery of any
asymmetry, normal or anomalous, that may exist in an individual's mon-
aural BSERs. An asymmetry would be any significant difference in the
latency or amplitude of the corresponding peaks in the BSERs of the two
ears. To date, only a few attempts at specifically assessing BSER symmetry
have been made. These reveal a controversy over the existence of symmetry
in neurologically normal persons (Chiappa, Gladstone, & Young, 1979;
Levine & McGaffigan, 1983) and characteristic asymmetry in patients with
various neurological diseases (Chiappa & Norwood, 1977) and in stutterers
(Decker & Howe, 1981). The limited research looking at evoked response
symmetry in LD populations has focused upon abnormalities in lateral
asymmetry in the cerebral cortex rather than in the brain stem.
Endogenous Evoked Responses. As the components in the evoked

response reach increasingly longer latencies (50-500 msec), both their
dependence upon the eliciting stimulus and independence from the pre-
vailing state of consciousness or cognitive state undergo progressive re-
versal. "Endogenous" components, then, appear to reflect perceptual,
cognitive, and perhaps anticipatory motor actions of the central nervous
system (Rockstroh et al., 1982). "Endogenous ERPs" (event-related poten-
tials) "are frequently triggered by external stimulus events, but their
waveform and timing are determined by the particular cognitive pro-
cesses activated by the stimulus rather than by its modality or physical

properties" (Hillyard & Woods, 1979, p. 346).
The method for recording middle and late components of the visual
and auditory evoked potentials is similar to that used for the short-latency
or exogenous components, except that the interval of EEG sampling fol-
lowing the presentation of the eliciting stimulus may be as long as 600
msec. The conventions for labeling the various components use a P to
designate positive voltage shifts at the active recording site relative to the
reference site, and an N for negative voltage shifts. Some researchers
differentiate among components of the same polarity by specifying the
typical latency in milliseconds at which this component occurs. For ex-
ample, P 300 refers to a positive peak (usually measured at the vertex) that
is typically seen around 300 msec after the presentation of the eliciting
stimulus. Other researchers follow the convention of sequentially num-
bering the various major positive and negative peaks in the order in which
they occur temporally. For example, Fried, Tanguay, Boder, Doubleday,
and Greensite (1981) identified their auditory evoked component P1 as a
positive component with a latency of approximately 80 msec, N1 as a
negative component with a latency of around 120 msec, and P2 as a
positive component peaking at around 180 msec. The latency, amplitude,
and occurrence of these components vary significantly depending upon
modality of the stimulus, concurrent task, attention, and perceptual pro-
cessing dysfunction. A review of recent research with middle and late
components appears in Donchin (1984), and the interested reader is re-
ferred there for further discussion. Figure 3 presents an example of an
endogenous ERP.
Latency
I
I
I
I
I
I
I
I Po
Stimulus
t _ - - - - - 400-600 msec. - - - - - - - 1
FIGURE 3. Endogenous (middle- and late-latency) evoked potential components.

Probe-Elicited Event-Related Potentials. Galin and Ellis (1975) devel-

oped the probe-ERP paradigm. Their intent was to combine the major
advantages of the EEG and ERP methods. In their original procedure, task-
irrelevant visual stimuli were presented at 3-second intervals while sub-
jects performed writing and block design tasks. These tasks were selected
to produce left and right hemisphere activation, respectively. They noted
that probe-ERP asymmetry, like EEG alpha, was task-dependent. They
observed differential suppression of the probe-elicited response in the
hemisphere currently activated by the specific task in which the subject
was engaged. They concluded that a brain region is less responsive to the
probe stimulus when that region is engaged by the concurrent task. John-
ston (1982) also presented data showing that the amplitude of modality-
specific components (e.g., auditory N100 and visual PlOO) is most attenu-
ated by a task in the same modality.
Exogenous Evoked Potentials in LD Children

Although the literature on LD and BSERs contains a limited number
of studies, the application of BSERs to assist in the diagnosis of LD ap-
pears to be promising. Lenhardt (1981) presented a single-case study
using audiometry, dichotic listening, and BSERs. Click stimuli were pre-
sented monaurally to both ears. A standard vertex referred to the right and
left ear montage was employed. At a variety of click intensities and in-
terstimulus intervals the subject displayed missing ipsilateral and con-
tralateral waves II and VII when stimuli were presented to the right ear.
Greenblatt, Bar, Zappulla, and Hughes (1983) described a child with a
specific auditory learning disorder and central auditory dysfunction. Pure
tone testing revealed hearing to be within normal limits bilaterally, and
spectral analysis on the child's EEG was normal when referenced to age-
specific norms (Matousek & Petersen, 1973). Visual evoked responses
were found to be symmetrically distributed throughout the cortex and
within normal limits. Endogenous auditory evoked responses were well
defined in the vertex leads and were symmetrical for the central and
temporal areas. Monaural BSERs were recorded from each ear ipsilateral
and contralateral to the side of stimulation and referenced to the vertex.
Clicks were delivered at 60 dB above threshold. BSER latencies for ip-
silateral peaks I and V and for contralateral peak V were within normal
limits for their laboratory. However, BSERs showed an absence of the
contralateral peak III for both left and right ear stimulation.
The authors acknowledge that absence or attenuation of the con-
tralateral peak III does occur in normals. However, they had not pre-
viously observed an instance in normals or other populations where the
contralateral peak III was absent on stimulation of both ears, as demon-
strated in this case. Unfortunately, marked individual variations on the
occurrence and symmetry of bilateral peaks seem to be common. Conse-

quently, the report of such a variation in a single subject must await
replication before any assessment significance can be attributed to this
finding.
Piggott and Anderson (1983) compared two groups of 10 children
matched for age and sex on BSER patterns. One child in each pair had
evidence of a central language disturbance as determined by neuropsy-
chological testing. All subjects had hearing and IQ scores that fell in the
normal range. The results indicated that the children with central lan-
guage disturbances showed longer latencies and interwave transmission
time than did the controls. The segments in which the differences reached
significance did not extend past wave III. The findings indicated that
BSERs were useful in demonstrating that children with central language
disabilities may have auditory stimuli transmission disturbances below
the level of the olivary nuclei. This suggests that some presumed central
disturbances may actually have their origin, if not their entire basis, in
dysfunction in the early stages of the peripheral sensory input system.
Unfortunately, these rather clear-cut findings have been difficult to
replicate. Tait, Roush, and Johns (1983) compared absolute and interwave
BSER latencies between school-defined LD children and an age-matched
group of normal children. BSERs to monaural 70-dB clicks were recorded
from the mastoid of the test ear referenced to the vertex. Left and right ears
were alternated between subjects during data collection. Findings from
this study did not demonstrate any evidence of abnormality in auditory
BSERs of LD children. Furthermore, psychoeducational performance as
measured by the Woodcock-Johnson Psychoeducational Battery (Wood-
cock & Johnson, 1977) was not related to BSER latency in these children.
The authors concluded that conventional BSER test procedures em-
ploying monaural click stimuli do not appear to provide an effective
means of differentiating LD groups from age-matched normal controls. In
a follow-up study with another sample of LD and normal children, Roush
and Tait (1984) again did not find auditory BSERs to be significantly
different between the two groups.
Summary. Exogenous BSERs have not yet provided any useful diag-
nostic schema for the various subtypes of LD, nor have they been shown
to be consistently useful in a broad pathognomic sense (i.e., being able to
separate normal from LD children). This may well reflect the lack of
consistent attention paid to the selection and classification of subjects in
these studies. Nevertheless, the occurrence of anomalous BSERs in this
population as reported in several studies should encourage further re-
search using this technique. Unlike middle- and late-latency evoked re-
sponses, BSERs are a mature clinical tecnhique in audiology and neu-
rology, and extensive norms across developmental periods and for
numerous disorders exist (Chiappa, 1983). Learning disabilities appear to

have multiple etiologies, and no reason exists at present for excluding
peripheral sensory involvement in these disorders. In the evaluation of a
child, BSERs have, at the very least, a role to play in permitting the
assessment or exclusion of peripheral auditory processing difficulty as
contributing to the problem (see Obrzut, Morris, Wilson, Lord & Caraveo,
1987).
Endogenous Evoked Potentials in LD Children

Cohen and Breslin (1984) compared normal readers and dyslexic
children on the hemisphere-specific endogenous components of their vi-
sual evoked responses. The stimuli were three- or four-letter words and a
blank light flash. Data were collected from 10 lead positions, which in-
cluded bilateral frontal, central, parietal, occipital, and midtemporal. For
the two stimulus types, the normal group had a faster P2 occipital area
response latency than the dyslexic group. Across all recorded areas, the
P3 responses to the flash stimuli tended to have longer latencies and were
less positive in the dyslexic group than in the average reading group. In
normals, the activity of the right and left hemispheres was dissimilar
during reading. However, the interhemispheric activity was quite similar
during word reading in dyslexic children. The authors concluded that
significant group differences existed in the way that the two hemispheres
processed visual information.
The goal of the research conducted by Holcomb, Ackerman, and Dyk-
man (1985) was to determine whether specific dyslexic subgroup dif-
ferences were apparent in late ERP components, especially P3, recorded
from multiple electrode sites. In addition to P3, Pc (a slow positive com-
ponent that peaks between 700 and 1100 msec) was discovered to be
sensitive to diagnostic group differences in this study. Several different
visual stimuli were presented to 24 reading-disabled children and 24
normal controls. Electrodes were attached Fpz' F z' Cz ' Pz' and 0z' a mid-
line referential montage that is not sensitive to lateralized asymmetries.
The reading-disabled children were found to have significantly smaller
overall bilateral P3s than age-matched controls. Further, the reading-dis-
abled children showed reduced P3 and Pc amplitudes to three-letter
words in contrast to nonalphanumeric symbols, whereas controls had
equivalent amplitudes for these two stimulus types. P3 latency data re-
vealed further differences between the two groups. Controls had earlier P3
peaks than the reading-disabled group and made faster behavioral re-
sponses to the presented stimuli.
Researchers in the area of endogenous EP components in dyslexia
have been, as a group, more sensitive to the need to distinguish among
subcategories of dyslexia. In regard to this distinction, Pirozzolo, Dunn,
and Zetusky (1983) stated that the evidence compiled during the last two
decades has rendered untenable the hypothesis that dyslexia represents a
single neuropsychological or educational entity. Further, contemporary
evidence overwhelmingly supports the existence of at least two subtypes
of developmental dyslexia. These findings are the result of several clinical
studies employing differential diagnosis of the cognitive deficits observed
in large populations of reading-disabled children (see Chapter 12, this
volume).
Fried et a1. (1981) concurred that treating the dyslexic population as
homogeneous has been a major problem. Their classification included a
subgroup composed of individuals having serious impairments in audito-
ry-verbal decoding, auditory sequencing, and symbol manipulation (Le.,
dysphonetic), whereas the other subgroup showed deficits in visuospatial
organization and visual-motor performance (Le., dyseidetic). Dysphonetic
dyslexics have great difficulty in reading and spelling words phonet-
ically, which may be conceptualized as a defect in auditory processing of
speech sounds. In contrast, dyseidetic children have good auditory per-
ception and memory. Their handicap is primarily deficient visual pro-
cessing. The alexic dyslexics are considered to be a mixture of both dys-
phonetic and dyseidetic types.
In their investigation Fried et al. (1981) applied ERP techniques to
study auditory information processing in the left and right hemispheres of
dyslexic subjects classified into subgroups. Dyslexic subjects were 2 or
more years delayed in reading and spelling as measured by the Wide
Range Achievement Test (Jastak, Bijiou, & Jastak, 1965). Through Boder's
(1973) Diagnostic Screening Test for Developmental Dyslexia, five of the
dyslexic subjects were classified as dysphonetic, six as dyseidetic, and
two as alexic. Dyslexic subjects were age- and sex-matched with normal
controls, all of whom were right-handed. Electrodes were placed at left
and right frontal locations (F 7' F8) and temporoparietal sites (midway
between T 5 and C3 , and midway between T 6 and C4 , and labeled W 1 and
W 2' respectively). Recording was referential. Auditory evoked potential
stimuli were voiced words Do and Go, and strummed chords A7 and D7.
The word and musical chord stimuli were delivered to each subject bin-
aurally at 70 dB.
The ERP wave form lying between 50 and 400 msec after stimulus
onset was chosen as the most likely time in which endogenous ERP ac-
tivity relating to stimulus processing would occur. The normal and dys-
eidetic subjects exhibited the expected larger ratio ERP wave form ampli-
tudes to words over chords in the left hemisphere and a differentially
smaller ratio in the right hemisphere. However, the dysphonetic subjects
did not exhibit this ERP asymmetry to stimulus type by hemisphere. "The
lack of greater word-musical-chord ERP waveform differences over the
left hemisphere in the dysphonetic group suggests that the left hemi-
sphere of dysphonetic dyslexics may not have a fully developed capacity
to process auditory information in a normal manner" (Fried et aI., 1981, p.
20).
Auditory and visual evoked potential measures also were used in the
Kaye et al. (1981) Neurometrics study reported above with children who
were underachieving in verbal, arithmetic, or mixed areas. Auditory
evoked potentials were evaluated for signal-to-noise ratio, wave shape
coherence, and amplitude symmetry. They reported major differences
among the underachieving groups in comparison with the matched nor-
mals and, with respect to each other, on the grand averages of both audito-
ry and visual evoked responses. In comparison to the AEP peak latency
values observed in the controls, the poor verbal achievers showed pri-
marily scattered left hemisphere latency deviations, whereas the poor
arithmetic achievers showed scattered left but consistent right hemi-
spheric latency deviations. The mixed (arithmetic and verbal) under-
achievers showed almost no right hemisphere problems, but consistent
left hemisphere latency deviations. Although these diagnostic group x
cortical area deviations do not fit easily into conventional models of
hemispheric or regional functional specialization, they do support the
importance of these diagnostic subtypes by showing their linkage to func-
tional cortical anomalies.
An attempt has been reported to remediate dysphonetic and dys-
eidetic subgroup specific asymmetry differences. Training involved prac-
tice at a task designed to appeal to the hemisphere in which in ERP
anomalies occurred (Bakker & Vinke, 1985). They concluded that their
training reduced these hemisphere-specific asymmetry anomalies, and
that this reduction was correlated with changes in measures of reading
accuracy and speed.
Summary. A number of investigators have found that differences ex-

ist between normal children and those with LD. However, it is becoming
increasingly evident that it is unwise to think of LD children as having
homogeneous disabilities. The subtyping of LD allows for the analysis of
specific disabilities in terms of endogenous evoked potentials. The re-
search thus far allows for the conclusion that ERPs can discriminate
among LD subtypes; however, it is premature to utilize these data for
routine clinical assessment. A major effort at evaluating infants for ERPs
in response to phonetic stimuli and then following those children through
early schooling to validate the use of ERP in the early prediction of devel-
opmental language difficulties is under way (Molfese & Molfese, 1986).
The preliminary results of this major longitudinal effort are very encour-
aging and appear consonant with respect to ERP findings to date.
Probe-Elicited Event-Related Potentials in LO Children

Johnstone et a1. (1984) found a difference between dyslexic and con-
trol children in regional response to an irrelevant visual probe during
reading. While the children performed complex verbal and spatial tasks,
light flashes irrelevant to their task were presented and subsequent visual-
event related potentials (VERPs) were recorded. VERPs were recorded
from central, parietal, midtemporal, and linked-ears leads referred to ver-
tex. Although the dyslexics showed a significant amplitude decrease in
the 250- to 350-msec range while reading difficult material, the normal
readers did not. The authors believed this to be the result of a complex
change in latency of the different components of the VERP that normally
occurs during this interval. Therefore, a complete description of VERP
effects in dyslexics must include not only regional differences but also a
description of the latency of individual components for the VERP within a
specific region.
Shucard, Cummins, and McGee (1984) used another variation on the
probe-ERP method by presenting irrelevant tones to normal and disabled
readers while they were engaged in two reading-related visual tasks. One
task (sounding out letters) was designed to activate the left hemisphere
differentially, whereas the other (matching letter patterns) was intended
to activate the right. The pattern of auditory-event-related potentials
(AERPs) was the opposite for the two groups. During both tasks, lower
right hemisphere response amplitude was seen in the disabled readers in
comparison with the normals. The left hemisphere amplitude also was
significantly higher than that of the normals during the task requiring
sounding out letters. The authors concluded that the opposing pattern of
asymmetries in the two groups suggested that the tasks require different
cortical processes in these two populations.
Summary. In terms of hemisphere relevant probes, special tech-

niques are necessary to find differences between normals and LD chil-
dren. Specifically, it is suggested that the analysis of data should include
latency components as well as regional differences. Further, and perhaps
more important, there appear to be differential cortical process require-
ments between LDs and normals.
CONCLUSIONS
Most studies have used only a small portion of the range of neu-
roimaging and electrophysiological techniques we have surveyed. Over-
all, the results of these studies are promising, with many of them finding
structural or electrophysiological anomalies in the parietooccipital region
where receptive speech systems are involved. Several studies also have
reported global electro physiological anomalies, especially in modal EEG
frequency. Relatively few of these studies have been sensitive to the need
to distinguish among the various subtypes of learning disability. Those
that were seem to have found considerable accuracy of classification into
those subtypes based upon the functional and structural neurological
evidence.
The most successful efforts have been those using a battery of elec-
trophysiological measures combined with clear classification of subjects
according to useful subcatagories of learning disability, and involving a
healthy sample size (e.g., Kaye et aI., 1981). These desirable features of
method are not unique to this area of research, and they have long been
recognized as sine qua non in the establishment of the research founda-
tion for any useful clinical assessment tool.
Computerized methods of neuroimaging and electrophysiology are
easily able to generate massive quantities of data. It is a temptation, when
applying new methods, to employ an atheoretical or "shotgun" approach,
rather than restricting the sphere of observation. Over a decade of comput-
erized investigation of EEGs and EPs in learning-disabled children has
yielded a large body of provocative, but scattered and inconsistent, re-
sults. An effort at consolidation would be helpful. This consolidation
needs to take the form of systematic theory and model testing, simul-
taneously employing as many of the complementary methods as possible.
Researchers committing themselves to a well-defined-and therefore
risky-explanation of the neurological basis of learning disabilities may
be confident, along with Sir Francis Bacon, that truth arises more readily
from error than from confusion.
The techniques described in this chapter hold all of the allure and
fascination of a possible technological breakthrough for a difficult assess-
ment problem. To be significant, however, neuroimaging and electro-
physiological procedures must demonstrate that they offset their high cost
and demand for specialized training. They must yield useful diagnostic
information about learning disabilities beyond that more cheaply and
easily gathered by conventional psychometric and clinical methods.
REFERENCES
Bakker, D. J., & Vinke, J. (1985). Effects of hemisphere-specific stimulation on brain activity
and reading in dyslexics. Journal of Clinical and Experimental Neuropsychology, 7,
505-525.
reading-spelling patterns. Developmental Medicine and Child Neurology, 15,663-687.
Broca, P. (1861). Remarques sur Ie siege de la factule du language articule, suivi d'un
364 GRANT L. MORRIS et aJ.
observation d'amphemie. Bulletin de la Societe Anatomique de Paris, Series 2, 6, 330-

357.
Chiappa, K. H. (Ed.). (1983). Evoked potentials in clinical medicine. New York: Raven Press.
Chiappa, K. H., Gladstone, J., & Young, R. (1979). Brain stem auditory evoked responses:
Studies of waveform variations in 50 normal human subjects. Archives of Nuerology,
36, 81-87.
Chiappa, K. H., & Norwood, A. E. (1977). Brainstem auditory evoked responses in clinical
neurology: Utility and neuropathological correlates. Electroencephalography and
Clinical Neurophysiology, 43, 518.
Cohen, J., & Breslin, P. W. (1984). Visual evoked responses in dyslexic children. Annals of
the New York Academy of Sciences, 425, 338-343.
Conners, C. K. (1978). Critical review of "Electroencephalographic and neurophysiological
studies in dyslexia." In A. L. Benton & D. Pearl, (Eds.), Dyslexia: An appraisal of current
knowledge (pp. 253-261). New York: Oxford University Press.
Decker, T. N., & Howe, S. W. (1981). Auditory tract asymmetry in brain stem electrical
responses during binaural stimulation. Journal of the Acoustical Society of America, 69,
1084-1090.
Denckla, M. B., LeMay, M., & Chapman, C. A. (1985). Few CT scan abnormalities found even
in neurologically impaired learning disabled children. Journal of Learning Disabilities,
IB, 132-135.
Donchin, E. (1984). Cognitive psychophysiology: Event-related potentials and the study of
cognition (Vol 1). Hillsdale: NJ: Erlbaum.
Drake, W. E. (1968). Clinical and pathological findings in a child with a developmental
learning disability. Journal of Learning Disabilities, 1, 4B6-502.
412-430.
Duffy, F. H., Denckla, M. B., Bartels, P. H., Sandini, G., & Kiessling, L. F. (1980). Dyslexia:
Automated diagnosis by computerized classification of brain electrical activity. Annals
of Neurology, 7, 421-428.
Fein, G., Galin, D., Johnstone, J., Yingling, C. D., Marcus, M., & Kiersch, M. E. (1983).
Electroencephalogram power spectra in normal and dyslexic children: 1. Reliability
during passive conditions. Electroencephalography and Clinical Neurophysiology, 55,
399-405.
Fisher, J. H. (1910). Congenital word blindness: Inability to learn to read. Transactions of the
Ophthalmological Society, 3D, 216-225.
Fried, I., Tanguay, P. E., Boder, E., Doubleday, C., & Greensite, M. (1981). Developmental
dyslexia: Electrophysiological evidence of clinical subgroups. Brain and Language, 12,
14-22.
Galaburda, A. M., & Kemper, T. L. (1979). Cytoarchitectonic abnormalities in developmental
rology, IB, 222-233.
Galin, D., & Ellis, R. (1975). Asymmetry in evoked potentials as an index of lateralized
cognitive processes: Relation to EEG alpha asymmetry. Neuropsychologia, 13, 45-50.
Geschwind, N., & Levitsky, W. (1968). Human brain: Left-right asymmetries in temporal
speech region. Science, 161, 186-187.
Greenblatt, E. R., Bar, A., Zappulla, R. A., & Hughes, D. A. (1983). Learning disability
assessed through audio logic and physiologic measures: A case study. Journal of Com-
munication Disorders, 16, 309-313.
Hier, D. B., LeMay, M., Rosenberg, P. B., & Perlo, V. P. (1978). Developmental dyslexia:
Evidence for a subgroup with reverse asymmetry. Archives of Neurology, 35, 90-92.
Hillyard, S., & Woods, D. (1979). Electrophysiological analysis of human brain function. In
M. Gazzaniga (Ed.), Handbook of behavioral neurobiology (Vol. 2, pp. 345-378). New
York: Plenum Press.
Hinshelwood, J. (1895). Word-blindness and visual memory. Lancet, 2, 1564-1570.
Holcomb, P., Ackerman, P. T., & Dykman, R. A. (1985). ERPs: Attention and reading deficits.
Psychophysiology, 22, 656-667.
Hughes, J. R. (1978). Electroencephalographic and neurophysiological studies in dyslexia. In
A. L. Benton & D. Pearl (Eds.). Dyslexia-An appraisal of current knowledge (pp. 205-
240). New York: Oxford University Press.
Hynd, G., Hynd, C., Sullivan, H., & Kingsbury, T. (1987). Regional cerebral blood flow (rCBF)
in developmental dyslexia: Activation during reading in a surface and deep dyslexic.
Journal of Reading Disabilities, 20, 294-300.
Jasper, H. H. (1958). The ten-twenty electrode system of the International Federation. Elec-
troencephalography and Clinical Neurophysiology, 10, 371-375.
Jastak, J., Bijiou, S. W., & Jastak, S. R. (1965). Wide Range Achievement Test. Wilmington,
DE: Guidance Associates.
John, E. R. (1963). Neural mechanisms of decision making. In W. S. Fields & W. Abbott
(Eds.). Information storage and neural control (pp. 243-282). Springfield, IL: Charles C
Thomas.
John, E. R. (1977). Neurometrics: Clinical applications of quantitative neurophysiology.
Hillsdale, NJ: Erlbaum.
John, E. R., Karmel, B. Z., Corning, W. c., Easton, P., Brown, D., Ahn, H., John, M., Harmony,
T., Prichep, L., Toro, A., Gerson, I., Bartlett, F., Thatcher, R., Kaye, H., Valdes, P., &
Schwartz, E. (1977). Neurometrics. Science, 196, 1393-1410.
Johnston, J. (1982). Probe event-related potentials during language processing in children: A
comparison of resource allocation and stimulus set models of attention. Unpublished
doctoral dissertation, University of California, San Francisco.
Johnstone, J., Galin, D., Fein, G., Yingling, C., Herron, J., & Marcus, M. (1984). Regional brain
activity in dyslexic and control children during reading tasks: Visual probe event-
related potentials. Brain and Language, 21, 233-254.
Kaye, H., & John, E. R., Ahn, H., & Prichep, L. (1981). Neurometric evaluation of learning
disabled children. International Journal of Neuroscience, 13, 15-25.
Kertesz, A., Black, S. E., Polk, M., & Howell, J. (1986). Cerebral asymmetries on magnetic
resonance imaging. Cortex, 22, 117-127.
LeMay, M. (1982). Morphological aspects of human brain asymmetry: An evolutionary per-
spective. Trends in Neurosciences, 5, 273-275.
LeMay, M., & Kido, D. K. (1978). Asymmetries of the cerebral hemispheres on computed
tomograms. Journal of Computer Assisted Tomography, 2, 471-476.
Lenhardt, M. L. (1981). Childhood central auditory processing disorder with brain stem
evoked response verification. Archives of Otolaryngology, 107, 623-625.
Levine, R. A., & McGaffigan, P. M., (1983). Right-left asymmetries in the human brain stem:
Auditory evoked potentials. Electroencephalography and Clinical Neurophysiology,
55, 532-537.
Lubar,J. F., Bianchini, K. J., Calhoun, W. H., Lambert, E. W., Brody, Z. H., & Shabsin, H. S.
(1985). Spectral analysis of EEG differences between children with and without learning
disabilities. Journal of Learning Disabilities, 18, 403-408.
Matousek, M., & Petersen, I. (Eds.). (1973). Automation of clinical electroencephalography.
New York: Raven Press.
Molfese, D. L., & Molfese, V. J. (1986). Psychophysiological indices of early cognitive pro-
cesses and their relationship to language. In J. E. Obrzut & G. W. Hynd (Eds.). Child
neuropsychology (Vol. 1, pp. 95-115). New York: Academic Press.
Morgan, W. P. (1896). A case of congenital word-blindness. British Medical Journal, 2, 1378.

Obrzut, J. E., Morris, G. L., Wilson, S. L., Lord, J. M., & Caraveo, L. E. (1987). Brain stem
evoked response in the assessment of learning disabilities. International Journal of
Neuroscience, 32, 811-823.
Orton, S. T. (1937). Reading, writing and speech problems in children. London: Chapman &
Hall.
Otto, D., Karrer, R., Halliday, R., Horst, R. L., Klorman, R., Squires, N., Thatcher, R. W.,
Fenelon, B., & Lelord, G. (1984). Developmental aspects of event-related potentials.
Annals of the New York Academy of Sciences, 425, 319-337.
Picton, T. W., Hillyard, S. A., Krausz, H. I., & Galambos, R. (1974). Human Auditory evoked
potentials: I. Evaluation of components. Electroencephalography and Clinical Neu-
rophysiology, 36, 179-190.
Piggott, L. R., & Anderson, T. (1983). Brain stem auditory evoked responses in children with
central language disturbance. Journal of the American Academy of Child Psychiatry, 22,
535-540.
Pirozzolo, F. J., Dunn, K., & Zetusky, W. (1983). Physiological approaches to subtypes of
developmental reading disability. Topics in Learning and Learning Disabilities, 3, 40-
47.
Pirozzolo, F. J., Jerger, J., Jerger, S., Morris, G., Levy, J., Goldman, A., & Handel, S. (1985).
Neuropsychological electro physiological and NMR studies of developmental aphasia.
International Neuropsychological Society Bulletin, 15, 16.
Rebert, C. A., Wexler, B. N., & Sproul, A. (1978). EEG asymmetry in educationally handi-
capped children. Electroencephalography and Clinical Neurophysiology, 45, 436-442.
Rockstroh, B., Elbert, T., Birbaumer, N., & Lutzenberger, W. (1982). Slow brain potentials
and behavior. Baltimore: Urban and Schwarzenberg.
Roush, J., & Tait, C. A. (1984). Binaural fusion, masking level differences and auditory brain
stem responses in children with language-learning disabilities. Ear and Hearing, 5, 37-
41.
Shucard, D. W., Cummins, K. R., & McGee, M. G. (1984). Event-related brain potentials
differentiate normal and disabled readers. Brain and Language, 21, 318-334.
Sklar, B., Hanley, J., & Simmons, W. W. (1973). A computer analysis of EEG spectral sig-
natures from normal and dyslexic children. IEEE Transactions on Bio-Medical En-
gineering, 20, 20-26.
Stockard, J. J., & Rossiter, V. S. (1977). Clinical and pathologic correlates of brain stern
auditory response mechanisms. Neurology, 27, 316-325.
Tait, c., Roush, J., & Johns, J. (1983). Normal ABR's in children classified as learning dis-
abled. Journal of Auditory Research, 13, 56-62.
Wernicke, C. (1874). Der aphasiche symptemkomplex. Breslaw: Cohn and Weigert.
Witelson, S. F. (1977). Developmental dyslexia: Two right hemispheres and none left. Sci-
enc~ 195, 309-311.
Witelson, S. F., & Pallie, W. (1973). Left hemisphere specialization for language in the
newborn: Neuroanatomical evidence of asymmetry. Brain, 96, 641-646.
Woodcock, R. W., & Johnson, M. B. (1977). Woodcock-Johnson Psycho-Educational Battery.
Hingham, MA: Teaching Resources Corporation.
Yingling, C. D., Galin, D., Fein, G., Peltzman, D., & Davenport, L. (1986). Neurometrics does
not detect "pure" dyslexics. Electroencephalography and Clinical Neurophysiology,
63, 426-430.
IV
Comment
14
Problems and Prospects in Child

MICHAEL G. TRAMONTANA
The chapters in this volume have addressed a variety of issues pertaining

to the neuropsychological assessment of children. These included a con-
sideration of the implications for assessment that stem from our current
knowledge of neurodevelopmental disorders (Risser and Edgell), the
unique contributions of neuropsychological assessment in evaluating the
brain-impaired child (Bigler), the relative merits of different modes of
neuropsychological inference (Willis), as well as the general issue of link-
ing assessment and treatment from a neuropsychological perspective
(Lyon, Moats, and Flynn). Special consideration was given to the assess-
ment of key aspects of child neuropsychological functioning, including
attention (Barkley), memory (Boyd), functional laterality (Gray and Dean),
and language (Crary, Voeller, and Haak). There also was special emphasis
on infant and early childhood assessment (Aylward)-an area that has
received relatively little attention in child neuropsychology, and on the
assessment of learning disabilities-an area that, by contrast, has been the
subject of intensive neuropsychological inquiry. Learning disabilities
were addressed specifically in terms of the analysis of subtypes (Hynd,
Connor, and Nieves), the early identification of developmental precursors
(Hooper), and the application of electro physiological methods in assess-
ment (Morris, Levy, and Priozzolo). Many important topics were covered,
and a synthesis of this material in terms of major themes and recom-
mended directions is now in order.
Perhaps the single most prominent theme throughout this volume has
been the emphasis on assessing the component processes that underlie
MICHAEL G. TRAMONTANA Bradley Hospital and Department of Psychiatry and

Human Behavior, Brown University, East Providence, Rhode Island.
369
370 MICHAEL G. TRAMONTANA
complex functions in children. A global construct like attention was con-

ceptualized in terms of more specific skills and processes, including
arousal and alertness, selective or focused attention, sustained attention
or vigilance, and span of apprehension, as well as the search strategies or
rules used while attending. Deficits in attention were seen as taking a
variety of forms including distractibility, impulsivity, and, less com-
monly, hemi-inattention or neglect. Likewise, memory was viewed as a
multifactorial construct, with key dimensions involving episodic versus
semantic and strategic versus nonstrategic memory processes. Neurolin-
guistic assessment was seen as requiring not only a differentiation of basic
language functions (comprehension, formulation, repetition, naming) but
also various aspects of language content (phonology, syntax, semantics,
pragmatics), as well as prosodic features and paralanguage factors (motor
speech, attention, memory). There also was an emphasis on the multifac-
torial assessment of functional laterality, and the importance of differ-
entiating specific patterns of lateral preference according to coordinated
systems of cerebral function. To varying degrees, the specific component
processes underlying a complex function were conceptualized as being
mediated by different neural substrates, vulnerable to disruption by dif-
ferent etiologies, following different developmental courses, and possibly
responsive to different forms of intervention. Their differentiation in as-
sessment was seen as an essential prerequisite to the identification of
more homogeneous patterns or subtypes of disability.
There is no question that crude, undifferentiated constructs such as
attention or memory are of little value, and that progress in child neuro-
psychology will depend greatly on the ability to provide a more precise,
multidimensional specification of these aspects of a child's functioning.
However, there is a critical question concerning the proper level of analy-
sis, and the point beyond which the specification of component processes
ceases to have clinical utility. As Rourke, Fisk, and Strang (1986) have
pointed out, the use of homogeneous, narrow-band tests-although inter-
nally consistent and permitting a precise assessment of a particular abil-
ity-may be of little use in assessing the child's capacity to meet complex
everyday demands. This, instead, requires an appreciation for the com-
plex interplay of abilities and their integrated operation in performing
meaningful tasks. A reductionistic approach, if carried to an extreme,
would yield only a fragmented view of the functions in question and have
little meaning in relation to everyday functioning. Thus, in examining
component processes, the emphasis must be on balancing the goals for
specificity with considerations of ecological validity.
Likewise, efforts to identify subtypes of functional disability must be
grounded in a meaningful differentiation of the ways in which various
classes of disabled children actually perform. Important lessons in this
regard can be learned from the neuropsychological research on learning
PROBLEMS AND PROSPECTS 371
disabilities, the area in which the study of subtypes is most advanced. As

pointed out by Hynd et a1. in Chapter 11, much of this work has been
characterized by the use of multivariate statistical methods for the pur-
pose of distinguishing subgroups of learning-disabled children on the
basis of their profiles of neuropsychological performance. However, this
often has been done in a post hoc fashion, without a clear rationale for the
inclusion of various tests in a neuropsychological battery, and with many
of these correlating poorly, if at all, with the actual areas of academic
deficiency in question. Moreover, academic deficiencies typically have
been defined rather crudely (based, for example, on the presence of an
achievement lag in a general area like reading), without a more precise
specification of the actual deficiencies involved. Hynd et a1. are quite
right in arguing that it would make better conceptual sense to base the
definition of subtypes on the identification of actual error patterns in
academic processes, and then to evaluate these using a theoretically driv-
en test battery to assess those neuropsychological functions known to be
important to the academic processes involved. Such an approach empha-
sizes the use of statistical methods to evaluate theory, rather than to con-
struct subtypes of learning disability that may have little ecological valid-
ity. It also emphasizes the assessment of neuropsychological processes
that have greater direct relevance to learning disabilities, thereby promot-
ing a stronger foundation for meaningful linkages between assessment
and treatment.
Lyon et al. (Chapter 5) have provided a thoughtful analysis of the
various issues involved in bridging the gap between neuropsychological
assessment and treatment. At present, little is known regarding the pre-
scriptive significance of different patterns of neuropsychological data that
would be pertinent in the actual selection or design of treatments. This
certainly is a critical frontier for child neuropsychology that is greatly in
need of systematic inquiry that could link assessment and treatment more
directly. Whatever directions these efforts take, it is imperative that treat-
ment prescriptions be considered within the wider context of the child's
psychosocial environment. Brain-behavior relationships never operate in
isolation but instead are influenced by environmental factors that serve to
compound or mitigate any observed functional deficits. For example, the
presence of behavioral disturbance in seizure-disordered boys has been
found to depend not only on IQ and seizure-related variables but on the
degree of parental cohesion with respect to issues of childrearing (Curley,
Delaney, Mattson, Holmes, & O'Leary, 1987). Also, as discussed by Bark-
ley in Chapter 6, environmental contingencies have been shown to play
an important role in the manifestation of attention deficits. If one ignores
these, and focuses exclusively on the task parameters that influence atten-
tion, any attempts at intervention outside of the testing situation may be
completely ineffective. All of this underscores the value of a bio-
psychosocial perspective in the development of treatment-oriented ap-

proaches to neuropsychological assessment. This is quite consistent with
the major emphasis given to issues of ecological validity by various au-
thors in this volume.
Another point of emphasis has been the need for a coherent frame-
work of developmental brain-behavior relationships to help guide the
assessment process and the evaluation of assessment results. As noted in
Chapter 1, this requires an understanding of how brain functions develop
normally, as well as under various pathological conditions. Such a frame-
work would indicate not only what areas of function should be assessed
in evaluating the effects of childhood brain injury but how these should
be assessed at different points in development. It requires, quite frankly, a
more complete and dynamic understanding of neuropsychological devel-
opment than currently exists.
This is not to say that existing knowledge in this area is useless in
guiding assessment, because this clearly is not the case. The fact is that
more is known about the neuropsychology of the developing brain than is
reflected in current assessment practices. There is a complex interplay of
factors that can influence the developmental patterns of children who
suffer early brain damage (Chelune & Edwards, 1981), and systematic
knowledge regarding the developmental significance of different types of
early lesions has begun to emerge (e.g., Dennis, 1985). Risser and Edgell
(Chapter 2) have provided an excellent overview of much of the work in
this area, focusing especially on insults to normal brain development
occurring through the first 2 years of life. Although much remains to be
learned, there are important implications for how neurodevelopmental
disorders are conceptualized and assessed from what already is known.
For example, there is little basis for an unqualified acceptance of
Kennard's (1940) principle concerning the enhanced potential for (re)ac-
quisition of function after injury that supposedly characterizes the imma-
ture brain. There apparently are definite limits to the extent and type of
(re)organization of function that can occur, even for brain injuries sus-
tained very early in life. The available evidence suggests that such injuries
often are associated with some degree of deficit and less-than-optimal
functioning, provided that these are assessed appropriately. Crary et a1.
(Chapter 10) addressed this point in relation to the specialized role that
the left cerebral hemisphere plays in the development of language func-
tions right from birth. Early unilateral injury to the left hemisphere may
result in a takeover of language functions by the right hemisphere, but
usually at some expense to nonverbal abilities (Bigler & Naugle, 1984).
Moreover, although deficits in language functions may not be apparent
when assessed simply in terms of Verbal IQ, a careful neurolinguistic
assessment will reveal subtle inefficiencies in various aspects of higher-
level language processing (Kiessling, Denckla, & Carlton, 1983).
The overall impression from reports such as these is that the young
child who sustains brain injury may benefit to some extent from cerebral
plasticity. However, any compensatory organization of function is apt to
be ineffecient because it would tend to go against the grain of the intrinsic
avenues for neural specialization that normally would unfold. Conse-
quently, residual deficits of some significance generally should be ex-
pected. These sometimes can be missed unless the assessment strategy is
geared to revealing the more subtle features of neuropsychological func-
tioning normally mediated by the injured brain region(s).
Existing evidence also would call into question the purported role of
maturational delays in the manifestation of neurodevelopmental disor-
ders. The concept of maturational lag usually has been invoked to account
for the fact that some children, including those with learning disabilities,
may show significant difficulties in neuropsychological functioning in
the absence of a documented history of brain injury. Presumably, these
difficulties simply may reflect a slower rate of maturation of an otherwise
normal brain (Satz, Taylor, Friel, & Fletcher, 1978). One problem with this
idea is that there is no evidence that the brains of children with significant
performance lags are immature or "unfinished" with respect to mor-
phological features (Rodier, 1984). Rather, investigations such as the de-
tailed cytoarchitectonic studies of the brains of dyslexics at autopsy (e.g.,
Galaburda & Kemper, 1979) indicate that brain morphology is truly abnor-
mal. Hynd et a1. (Chapter 11) have interpreted this work to suggest that
abnormalities in cell migration during later periods of fetal development
(resulting in assorted defects of the left perisylvian region such as focal
dysplasias, disordered cortical layering, and polymicrogyria) constitute
the neurodevelopmental basis for many learning disabilities.
Another problem with the concept of maturational lag is that it im-
plies that the child with functional delays eventually will "catch up" and
exhibit normal abilities. However, this generally is not the case, as shown
in the follow-up literature on children with learning disabilities (Schon-
haut & Satz, 1983). Although disabled readers may improve over time,
they typically continue to show inefficiencies in their reading skills (Rut-
ter, Tizard, Yule, Graham, & Whitmore, 1976). Moreover, with the excep-
tion of the position taken by Boyd (Chapter 7), the general consensus
among the authors in this volume is that the processing strategies utilized
by disabled learners cannot be compared to those of a younger child.
Their functioning is not simply delayed but deviant. These children are
delayed in the sense that their inefficiencies result in lower performance
levels on age-normed tests, but the kinds of processing strategies that they
utilize are uncharacteristic of normal functioning at any age.
Thus, it appears that the concept of maturational lag has little mean-
ing in current conceptualizations of neurodevelopmental disorders. Use
of the term should be limited to those instances in which there is a genu-
ine delay in the normal timing or rate of skill acquisition without parallel
evidence of aberrations in existing processing abilities. This requires the
differentiation of process and product in assessment, and also has impor-
tant implications for how developmental precursors to learning problems
are identified. If functions are not organized differently in learning-dis-
abled children but instead operate less efficiently (as suggested by Hynd
et 01.), then learning disabilities should be forecast by dysfunction involv-
ing the brain regions that normally are preordained to mediate academic
skills. Hooper (Chapter 12) has discussed the challenge that this perspec-
tive poses for early identification research, given that deficits of this type
ordinarily are considered to be "silent" until relatively late in the pre-
school period.
Some concepts in child neuropsychology, like neuromaturationallag,
grew out of an era when it generally was not possible to obtain precise
information regarding brain structure and physiology in children with
functional handicaps. Just as inadequacies in neuropsychological assess-
ment sometimes resulted in an overestimation of recovery of function or
cerebral plasticity in brain-injured children, so too inadequacies in diag-
nostic tools for assessing brain pathology may have led to false (or at least
premature) conclusions concerning the apparent absence of brain anoma-
lies in certain neurodevelopmental disorders. Earlier researchers did not
have the benefit of the powerful methods now available for brain imaging
and the in vivo study of dynamic neurophysiological processes, which
were discussed in detail by Bigler (Chapter 3) and Morris et 01. (Chapter
13). The availability of this new generation of neurodiagnostic tools prom-
ises to have an unprecedented impact on the advancement of knowledge
in child neuropsychology, especially with respect to the neural side of
brain-behavior relationships. Full advantage should be taken of the new
avenues for investigation that these methods have opened, but without
losing sight of the unique contributions that child neuropsychology can
make in assessing and conceptualizing the behavioral aspects of various
brain anomalies. A clearer understanding of developmental brain-behav-
ior relationships cannot be obtained solely through technical advances in
diagnosing brain pathology but will depend also on parallel advances in
the development of more precise and ecologically valid tools for assessing
neuropsychological functioning.
At this point, there are a number of serious limitations associated
with all-purpose or omnibus approaches to the neuropsychological as-
sessment of children. Neuropsychological assessment is becoming in-
creasingly specialized and geared more toward the evaluation of compo-
nent processes and their impact on a child's everyday functioning. There
is a growing appreciation for the need to tailor the assessment process, as
well as the constructs that are assessed, according to the clinical questions
and specific populations involved. An assessment strategy that is quite
useful in one context may be useless in another. Also, new knowledge in
developmental neuropsychology is evolving rapidly, and undoubtedly

will have an ongoing impact on how the assessment process is conducted
and conceptualized. An omnibus test battery would tend to lack the speci-
ficity and flexibility necessary in meeting the varied and ever-widening
objectives of assessment in child neuropsychology.
However, as noted in Chapter 1, a fixed or standard battery of tests
does have the advantage of helping to assure a consistently broad sam-
pling of neuropsychological functioning that is not biased by either refer-
ral complaints or the child's initial presentation. There is a need for a core
set of test procedures that could provide a consistent frame of reference in
making screening decisions regarding those areas of function that require
a more in-depth, process-oriented assessment through a flexible selection
of tests. The use of a core battery also promotes a greater comparability of
finding across different times, patient groups, and research settings. The
trouble is that there currently is little agreement within the field as to
what procedures should constitute core components in a comprehensive
neuropsychological assessment. Existing examples of standardized test
batteries for children take much too long to administer to be practical for
the purpose of routine neuropsychological screening. Moreover, the infor-
mation that they yield is often redundant with the results of intelligence
testing (Tramontana, Klee, & Boyd, 1984).
What is needed is a core battery of procedures, tailored to children of
different ages, which is brief but nonetheless spans key dimensions of
neuropsychological functioning. A more coordinated effort in this direc-
tion among different groups of investigators would be a very positive step
for the field, and it is long overdue. Such an effort would permit a much
wider-scale standardization of various measures and a more rapid ac-
cumulation of validation data than could be achieved by any single group
of investigators working alone. Divergent methods naturally would con-
tinue to be used, both in exploring specific questions and as a matter of
personal preference, but it is critical for there to be some consensus on the
core features of assessment that could serve as general standards for prac-
tice. These obviously would be subject to periodic review and revision as
new knowledge unfolds and as better instruments become available. This
kind of collaboration would not be easy to achieve, and indeed may be
unrealistic, but its potential benefits should be recognized and promoted
actively by the professional organizations within the field. This ultimate-
ly would be more constructive than all of the rhetoric and energy devoted
to debates concerning the purported superiority of one assessment ap-
proach or another, often with no data to support the claims.
A major objective of this volume has been to emphasize the need for
further developments in various aspects of child neuropsychological as-
sessment. The coverage was not exhaustive but was focused instead on
certain key areas, including assessment of the younger child. The intent
has been to balance the appraisal of present shortcomings with the outlin-
ing of worthwhile directions for future work. Innovations and refinements

certainly are needed, but progress could be achieved in the meantime by
drawing selectively upon some of the methods already available. The
contributing authors in this volume have gone "out on a limb," so to
speak, in providing informed opinions as to what methods appear to hold
particular promise in assessing various aspects of neuropsychological
functioning. These recommendations undoubtedly will require change as
new knowledge in developmental neuropsychology is obtained, but the
judicious use of existing tools will be necessary in order for the knowl-
edge base to grow. Much like its subject of inquiry, the field of child
neuropsychology is growing, differentiating, and undergoing dynamic
change. It will be essential that our assessment methods both promote and
keep pace with the exciting developments that lie ahead.
REFERENCES
Bigler, E. D., & Naugle, R. 1. (1984). Case studies in cerebral plasticity. International Journal
of Clinical Neuropsychology, 7, 12-23.
Chelune, G. J., & Edwards, P. (1981). Early brain lesions: Ontogenetic-environmental consid-
Curley, A. D., Delaney, R. C., Mattson, R. H., Holmes, G. 1., & O'Leary, K. D. (1987). Determi-
nants of behavioral disturbance in boys with seizures. Paper presented at the 95th
Annual Convention of the American Psychological Association, New York.
Dennis, M. (1985). Intelligence after early brain injury: 1. Predicting IQ scores from medical
variables. Journal of Clinical and Experimental Neuropsychology, 7, 526-554.
Galaburda, A. M., & Kemper, T. 1. (1979). Cytoarchitectonic abnormalities in developmental
Kennard, M. A. (1940). Relation of age to motor impairment in man and subhuman primates.
Archives of Neurology and Psychiatry, 44, 377-397.
Kiessling, 1. S., Denckla, M. B., & Carlton, M. (1983). Evidence for differential hemispheric
function in children with hemiplegic cerebral palsy. Developmental Medicine and
Rodier, P. M. (1984). Exogenous sources of malformations in development. In E. S. Gollin
(Ed.), Malformations of development: Biological and psychological sources and conse-
quences (pp. 287-313). New York: Academic Press.
Rourke, B. P., Fisk, J. 1., & Strang, J. D. (1986). Neuropsychological assessment of children: A
treatment-oriented approach. New York: Guilford Press.
Wight studies, 1964-1974. Psychological Medicine, 6, 313-332.
precursors of reading disabilities: A six year follow-up. In A. Benton & D. Pearl (Eds.),
Dyslexia: An appraisal of current knowledge (pp. 313-348). New York: Oxford Press.
Schonhaut, S., & Satz, P. (1983). Prognosis for children with learning disabilities: A review
of follow-up studies. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 542-563).
New York: Guilford Press.
Tramontana, M. G., Klee, S. H., & Boyd, T. A. (1984). WISC-R interrelationships with the
Halstead-Reitan and Children's Luria Neuropsychological Batteries. International Jour-
nal of Clinical Neuropsychology, 6, 1-8.
Index
Academic 'achievement Attention (Cont)

learning disability and, 316, 349 behavior rating scales, 153-159
psychological assessment, 84 cancellation tasks, 163-164
See also Learning disability Child Behavior Checklist (CBCL), 156
Actuarial models, 97-102 Children's Embedded Figures Test
base-rate considerations in, 98-100 (CEFT), 164
clinical models contrasted, 94-96 components of, 148-149
methods, 97-98 Conners rating scales, 154-155
stability and generalizability of rules Continuous Performance Tasks (CPT),
in, 101-102 161-163
validity of criteria in, 100-101 defined, 145-146, 148
ADD-H Comprehensive Teacher Rating developmental issues, 150-151
Scale (ACTeRS), 159 direct observational measures of, 167-
Adults 169
brain-behavior differences with chil- Direct Reinforcement of Latency (DRL)
dren, 4 tasks, 165-166
Halstead-Reitan Neuropsychological Edelbrock Child Assessment Profile
Battery and, 6, 10 (CAP), 156-157, 158
learning disabilities, 302 Freedom From Distractibility (FFD)
memory assessment, 177 Factor (WISC-R), 167
treatment linkages, 116-117 Goldman-Fristoe-Woodcock (GFW) Se-
Age level lective Attention Test, 166
age specificity and, 31 importance of, 146-148
attention and, 150 Laboratory Measures, 159-167
brain development and, 51-53 Matching Familiar Figures Test
conceptual issues and, 227-234 (MFFT), 165
Halstead Neuropsychological Test Bat- Mazes, 164-165
tery, 9-10 measures of, 147-148
language development and, 254-261 Preschool Behavior Questionnaire
learning disabilities, 326-328 (PBQl, 158-159
pediatric neurological exam, 68 psychometric tests/laboratory measures
See also Infancy and early childhood of, 159-167
Age specificity, 31 Reaction Time (RT) tasks, 160-161
Attention, 145-176 Revised Behavior Problem Checklist
ADD-H Comprehensive Teacher Rating (RBPC) , 157
Scale (ACTeRS), 159 Span of apprehension test, 166
assessment of, 153-169 Attention deficit disorder, 146. See also
behavioral conceptualization of, 151- Attention
152 Auditory-Verbal Learning Test, 183, 184
377
378 INDEX
Bakker Research Program, 131-132 Central nervous system

Base-rate considerations, 98-100 infancy, 230
Bayesian analysis, 106 language skills, 254-261
BEAM. See Brain Electrical Activity Map- See also Brain
ping (BEAM) technique Child Assessment Profile (CAP), 156-
Behavioral state, 228 157, 158
Behavior rating scales, 153-159 See also Edelbrock Child Assessment
Bender Visual-Motor Gestalt test, 84 Profile (CAP)
Benton Revised Visual Retention Test, Child Behavior Checklist (CBCL), 156
183, 184 Child neuropsychological assessment.
Birth See Neuropsychological assessment
brain development, 47 Children's Embedded Figures Test
brain injury, 56 (CEFT),164
Boder Test of Reading-Spelling Patterns Classification systems, 231-232
(BTRSP) Clinical models, 102-107
developmental issues, 136 actuarial models contrasted, 94-96
treatment linkages, 133 debiasing techniques, 105-106
Boston Process Approach, 18 design and decision rules, 106-107
Brain interpretive strategies, 103-104
infancy, 230 Cluster analysis, 97, 98
language skills, 254-261 Cognition, 233, 234
laterality assessment, 205-223 Cognitive approach. See Functional pro-
learning disabilities, 301-304 file approach
learning disabilities (electrophysiolog- Coloured Progressive Matrices (CPM), 85
ical assessment), 337-366 Communication. See Neurolinguistic as-
Brain-behavior relationship sessment; Language development
child-adult differences, 4' Computed tomography, 337, 340-341.
infancy, 226 See also Neuroradiological tests
treatment linkages, 115-116, 118 Conners rating scales, 154-155
Brain damage Continuous Performance Tasks (CPT),
treatment linkages, 113 161-163
See also Localization Cranial nerve function, 69
Brain development, 4, 41-65
abnormalities in, 51-56
neuropsychological implications of, Data-diagnosis contingency, 94-95
56-59 Debiasing techniques, 105-106
overview of, 41-42 Denman Neuropsychological Memory
principles of, 42-51 Scale, 183, 184
treatment linkages, 118 Detroit Tests of Learning Aptitude
Brain Electrical Activity Mapping (DTLA), 179, 181
(BEAM) technique Developmental issues
described, 82 attention, 150-151
learning disabilities, 346, 349 language skills, 254-261
Brain injury localization. See Brain learning disabilities, 325-328
damage; Localization memory, 185-195
Brain stem evoked potentials, 353-355. psychological assessment, 84-85
See also Evoked potentials treatment linkages, 118, 121-122, 136-
137
See also Brain development
Cancellation tasks, 163-164 Developmental neurolinguistic
Category Test, 118 assessment.
CAT scan, 340 See Neurolinguistic assessment
INDEX 379
Developmental Test of Visual-Motor Electroencephalography (Cont.)

Integration, 85 neuropsychological assessment, 80-81,
Diagnostic models, 94-96 82, 83
data consideration in, 95-96 See also Evoked potentials
data-diagnosis contingency, 94-95 Electrophysiological assessment, 80-83,
data importance in, 96 343
Dichotic listening laterality assessment, 210-211
Bakker Research Program, 131-132 learning disabilities, 337-366
laterality assessment, 207-209 Embryonic development, 42-51
Differentiation (neural), 44-47 Endogenous evoked responses
Direct observation described, 355-356
attention assessment, 167-169 learning disabilities, 359-361
infancy, 233 See also Evoked responses
Direct Reinforcement of Latency (DRL) Environment
tasks, 165-166 infancy, 231
Discriminant analysis, 97 learning disabilities, 317-318
Draw-A-Person test, 84 Evoked potentials, 352-362
DSM-III, 106-107 described, 352-357
Dynamic assessment, 137-138 electroencephalography, 346, 349
Dynamic phase, 8 learning disabilities, 337, 357-362
Dyslexia neuropsychological assessment, 81-82
laterality assessment, 216 See also Electroencephalography
See also Learning disability; Reading Exogenous evoked responses
disability described, 353
learning disabilities, 357-359
See also Evoked potentials
Early childhood. See Infancy and early Expressive functions, 233
childhood
Eclectic test batteries
critique of, 19-20 Fast Fourier Transformation, 345
described, 17 Fetal development, 42-51
Ecological validity, 8 Fixed-battery approaches, 9. See also en-
Edelbrock Child Assessment Profile tries under names of specific test
(CAP), 156-157, 158 batteries
See also Child Assessment Profile Flynn Research Program, 132-135
Edinburgh Handedness Inventory, 212 Freedom From Distractibility (FFD) factor
Education for All Handicapped Children (WISC-RJ, 167
Act, 4 Functional laterality. See Laterality
infancy and, 227 assessment
learning disabilities and, 284, 313, 338 Functional profile approach, 6-8
EEG. See Electroencephalography Gait disturbance, 68

Electrodiagnostic tests, 80-83. See also Genetic disease, 55-56
Electrophysiological assessment Gestational age, 45
Electroencephalography, 343-352 Goldman-Fristoe-Woodcock (GFW) Selec-
analysis methods in, 345-346 tive Attention Test, 166
described, 343-345 Grammatical expansion stage, 253-254,
fast-Fourier Transformation (FFTJ, 345 264-269
laterality assessment, 210-211

learning disability, 337, 347-352 Halstead-Reitan Neuropsychological Test
montage, 334 Battery
380 INDEX
Halstead-Reitan Neuropsychological Test Laterality assessment (Cont.)

Battery (Cont.) lateral preference measures, 211-
critique of, 12-13 214
described, 9-13 learning disabilities and, 215-218
historical perspective on, 6 methods of, 206-215
Luria-Nebraska compared, 14, 15 unimanual performance measures,
memory assessment, 179, 182 214-215
treatment linkages, 119-122 visual half-field procedures, 209
Handedness, 211-214. See also Laterality Lateral preference measures, 211-214
assessment Learning
Head injury, 22-23 infancy, 233-234
Hemiinattention, 149 memory and, 177, 178, 188-192
See also Intelligence testing; Memory
Learning disability, 281-312
Impulsivity assessment applications, 26-27
attention, 148 Bakker Research Program, 131-132
cancellation tasks, 163 conceptual framework for evaluation
Infancy and early childhood, 225-248 of,305-307
assessment instruments, 234-239 definitions, 284-285
conceptual issues in, 227-234 Flynn Research Program, 132-135
current status, 225-227 Halstead-Reitan Neuropsychological
future directions, 239-244 Test Battery, 11
Information-processing models, 186-188 historical perspective on, 7, 282-284
Intelligence testing laterality assessment and, 215-218
Halstead-Reitan Neuropsychological Lyon Research Program, 125-131
Test Battery, 12 multifactor research in, 285-301
learning disabilities, 286 neuroanatomic ai-linguistic perspectives
localization, 11 on, 301-304
memory, 179 overview of, 281-282
neuropsychological assessment com- prediction of, 313-335
pared, 86-87 remediation, 124
psychological assessment, 84 treatment linkages, 114
systemic illness, 23-24 Learning disability (electrophysiological
treatment linkages, 117-118 assessment), 337-366
anatomical correlates in, 338-339
CAT scans, 340
Kaufman Assessment Battery for Chil-
electroencephalography, 343-352
dren,87
evoked potentials, 352-362
memory assessment, 181
magnetic resonance imaging, 342-
treatment linkages, 122-123
343
overview of, 337-338
Language development, 250-254 positron emission tomography, 341
grammatical expansion stage, 253-254 regional cerebral blood flow, 341-342
lexical expansion stage, 252-253 Learning disability prediction, 313-335
prelinguistic communication, 251 current status of, 318-324
See also Neurolinguistic assessment importance of, 314-318
Language-impaired children, 215-218 issues and directions in, 324-331
Laterality assessment, 205-223 neurodevelopmental theory and, 325-
brain, 205-206 328
clinical implications, 218-219 overview of, 313-314
dichotic listening methods, 207-209 preschool prediction, 328-330
electrophysiological measures of, 210- related issues in, 330-331
211 Levels-of-processing model, 187-188
INDEX 381
Lexical expansion Neonate, 235. See also Infancy and early

assessment of, 263-264 childhood
described, 252-253 N euroanatomical-linguistic perspective
Localization, 10-11 (learning disabilities), 301-304
Luria-Nebraska Neuropsychological Test Neuroimaging techniques. See Neu-
Battery, 9 roradiological tests; entries under
critique of, 15-16 specific procedures
described, 13-16 Neurolinguistic assessment, 249-279
memory assessment, 179, 182 current status of, 261-269
qualitative approaches, 18 guidelines/questions regarding, 269-
treatment linkages, 122 274
Lyon Research Program, 125-131 language development, 250-254
language processing styles, 254
maturational factors in, 254-251
overview of, 249-250
Magnetic resonance imaging
See also Language development
learning disabilities, 337, 342-343
Neurolinguistic subtyping, 300-301
See also Neuroradiological tests
Neurological disorder, 21-23
Matching Familiar Figures Test (MFFT),
Neuropsychological assessment
165
actuarial models of, 97-102
Mazes, 164-165
applications of, 21-28
McCarthy Scales, 179, 181
attention, 153-169
Memory, 177-204
brain development, 41-65
clinical relevance of models of, 199-
clinical models, 102-107
200
conceptual/practical issues in, 28-32
current status of assessment of, 179-
current interest in, 3-5
185
diagnostic models of, 94-96
developmental models, 185-195
eclectic test batteries, 17
format suggestions for battery assess-
electrodiagnostic tests, 80-83
ing, 195-196
fixed-battery approaches, 9
infancy, 233-234
Halstead-Reitan Neuropsychological
information-processing models of, 186-
Battery, 9-13
188
historical trends in, 5-9
interactionist models of, 188-195
infancy and early childhood, 225-248
task type suggestions for battery assess-
intelligence testing compared to, 86-
ing, 196-199
87
tests for assessment of, 180-183
laterality assessment, 205-223
Memory for Designs Test, 183, 184
learning disabilities, 26-27 (See also
Mental activity (infancy), 234
Learning disability)
Migration (neural), 43-44
Luria-Nebraska Neuropsychological
Montage, 334
Battery-Children's Revision, 13-16
Motor function, 68-69
memory, 177-204
Multiaxial system (DSM-III-R), 106-107
neurolinguistic assessment, 249-279
Multiple regression, 97
neurological disorders, 21-23
Multistore model (memory), 186-187
neuroradiological tests, 75-80
Muscle tone, 68-69
pediatric neurological exam, 68, 75
Myelination, 47-49
problems/prospects in, 369-376
process-oriented approaches, 18-19
psychiatric disorders, 24-26
Naming, 267 psychological assessment and, 84-87
National Joint Committee for Learning qualitative approaches, 17-18
Disabilities (NJCLD), 285, 315, 338 rehabilitation of function, 27-28
Neglect. See Hemiinattention special-purpose measures, 20-21
382 INDEX
Neuropsychological assessment (Cont.) Radiology. See Neuroradiological tests

systemic illness, 23-24 Reaction time tests, 160-161
treatment linkages with, 113-142 Reading disability
Noncompliance, 229 Bakker Research Program, 131-132
Flynn Research Program, 132-135
historical perspective on, 282-284
Observational measures. See Direct Lyon Research Program, 125-131
observation See also Learning disability
Oncology, 24 Receptive functions, 233
Reflexes, 233
Regional cerebral blood flow, 341-342
Paralanguage measures, 268-269 Rehabilitation
Peabody Individual Achievement Test assessment applications, 27-28
(PlAT), 126-127 treatment linkages, 116
Pediatric neurological exam See also Treatment linkages
anomalous physical development Reitan Evaluation of Hemispheric Abili-
markers, 70-73 ties and Brain Improvement Training
cranial nerve function, 69 (REHAB IT) , 120-121
findings in, 70 Remediation
"hard" versus "soft" findings in, 73- responses to, 124
74 treatment linkages, 116
mental status characteristics, 74 See also Treatment linkages
motor function, 68-69 Repetition, 266-267
neuropsychological assessment and, 68, Revised Behavior Problem Checklist
75 (RBPC), 157
sensory-perceptual function, 69-70
station/gait, 68
Personality, 84 Sensory-perceptual function, 69-70
Positron emission tomography, 341 Sex differences, 146
Prediction Single-test approach, 5-6
infancy, 229-230 Social functioning
learning disabilities, 313-335 attention, 150-151
Prelinguistic assessment, 251, 262-263 learning disabilities, 316-317
Preschool Behavior Questionnaire (PBQ], Span of apprehension
158-159 attention assessment, 166
Probe-elicited event-related potentials defined, 149
described, 357 Special-purpose measures, 20-21
learning disabilities, 362 Stanford-Binet test, 179, 180
Processing, 233-234 Statistical methods, 97-102
Process-oriented approaches critique of, Stimulants, 152
19-20 Sustained attention
described, 18-19 defined,148-149
Proliferation (neural), 43-44 developmental issues, 150
Prosody, 267-268 Systemic illness, 23-24
Psychiatric disorders, 24-26
Psychological assessment, 84-87
Psychometric tests, 159-167 Temperament, 228
Public Law 94-142. See Education for All Test battery/lesion-specification stage, 6
Handicapped Children Act Tetrahedral model (memory), 188-192
Trauma, 282
Treatment linkages, 113-142
Qualitative approaches assessment and, 113-142
critique of, 19-20 Bakker Research Program, 131-312
described, 17-18 classification r~search needs, 138-139
INDEX 383
Treatment linkages (Cont.) Validity

developmental issues, 136-137 actuarial models, 100-101
dynamic issues, 137-138 cancellation tasks, 163
Flynn Research Program, 132-135 Visual half-field procedures, 209
Halstead-Reitan Neuropsychological
Test Batteries, 119-122
Kaufman Assessment Battery for Chil-
Wechsler Intelligence Scale for Children-
dren (K-ABC), 122-123
Luria-Nebraska Battery-Children's Re- Revised, 118
Freedom From Distractibility factor,
vision, 122
Lyon Research Program, 125-131 167
memory assessment, 179, 180
models for, 119
overview of, 113-142 neuropsychological assessment, 86-87
systemic illness, 23-24
professional preparation and experi-
ence, 135-136 Wechsler Memory Scale, 178
Wernicke-Geschwind model, 301-304
purposes/measurement characteristics,
115-119
studies in, 123-124
Yale Neuropsychoeducational Assess-
ment Scales-Stigmata schedule, 70-
Unimanual performance measures, 214- 73
215

Assessment Issues in Child Neuropsychology

Uploaded by

Copyright:

Available Formats

Assessment Issues in Child Neuropsychology

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Assessment Issues in Child Neuropsychology

Uploaded by

Copyright:

Available Formats

Assessment Issues in

Editorial Advisory Board:

ASSESSMENT ISSUES IN CHILD NEUROPSYCHOLOGY

HANDBOOK OF CLINICAL CHILD NEUROPSYCHOLOGY

MEDICAL NEUROPSYCHOLOGY: The Impact of Disease on Behavior

NEUROPSYCHOLOGICAL FUNCTION AND BRAIN IMAGING

Springer Science+Business Media, LLC

Assessment issues in child neuropsychology / edited by Michael G. Tramontana and

1988 Springer Science+Business Media New York

All rights reserved

No part of this book may be reproduced, stored in a retrieval system, or transmitted

GLEN P. AYLWARD, Division of Developmental and Behavioral Pedi-

Development and Learning, University of North Carolina, Chapel

Neuropsychology has its roots in clinical neurology. Reading case de-

basic repertoire should be examined carefully in the light of developmen-

especially in gunshot and shrapnel victims. As outlined by Risser and

Recent years have witnessed the emergence of child neuropsychology as a

help outline constructive ways in which progress can be achieved. It is

always. We also owe a special note of thanks to Maryanne Tramontana for

Part II: GENERAL ISSUES IN CHILD NEUROPSYCHOLOGICAL

Principles of Neural Development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

Base-Rate Considerations ................................... 98

PART III: SPECIAL TOPICS IN ASSESSMENT

Importance of Attention in Child Neuropsychology. . . . . . . . . . .. 146

A Concluding Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 273

Summary .................................................. 323

Part IV: COMMENT

Index ........................................................ 377

There has been a tremendous growth of interest in child neuropsychology

MICHAEL G. TRAMONTANA' Bradley Hospital and Department of Psychiatry and

HISTORICAL TRENDS IN ASSESSMENT

The history of assessment in child neuropsychology can be dis-

regard the continued reliance on this approach as constituting mal-

has referred to as the cognitive phase of neuropsychological assessment.

limited the specificity of neuropsychological description, and quite possi-

one area in which the current status of neuropsychological assessment

CURRENT APPROACHES IN CHILD NEUROPSYCHOLOGICAL

Halstead-Reitan Neuropsychological Battery (HRNB)

Children (ages 5 to 8) and the Halstead Neuropsychological Test Battery

secondary pathological changes may serve to produce more generalized

on the selection of tests sensitive to brain damage. Although the battery

Luria-Nebraska Neuropsychological Battery-Children's Revision

populations, including children with learning disabilities (Geary, Jen-

factors (Tramontana, Hooper, Curley, & Nardolillo, 1988). This is surpris-

Critique. A good deal of research on the LNNB-CR has been generated

nant validity achieved by the WISC-R alone in children manifesting

Other General Approaches

qualitative approaches, and (3) process-oriented approaches. There are

Eclectic Test Batteries

pothesis-testing model of neuropsychological assessment designed spe-

Some measures have been validated on the basis of their prediction of

As was apparent in the discussion of historical trends in assessment,

netic disorders (e.g., Turner's syndrome), structural abnormalities of the

the most important areas to be monitored. The child's social-emotional

Conversely, there is a relatively high rate of neuropsychological dys-

important to apply more conservative rules for diagnosing neuropsycho-

rophysiological factors, as well as relevant aptitude x treatment interac-

Rehabilitation and Assessing Recovery of Function

the importance of monitoring developmental progress by documenting

CONCEPTUAL AND PRACTICAL ISSUES