Theories of Neurosis

Michael Gossop
Theories
of Neurosis
With a Foreword by H. J. Eysenck
With 6 Figures
Springer-Verlag
Berlin Heidelberg New York 1981
Michael Gossop, B. A., Ph. D.

The Bethlem Royal Hospital
Drug Dependence Clinical Research
and Treatment Unit
Monks Orchard Road
Beckenham, Kent, BR3 3BX
Great Britain
ISBN 978-3-642-88475-7
ISBN 978-3-642-88473-3 (eBook)
DOI 10.1007/978-3-642-88473-3
Library of Congress Cataloging in Publication Data
Gossop, Michael, 1948 Theories of neurosis.
Bibliography: p.
Includes index.
1. Neuroses. I. Title. [DNLM: 1. Neuroses. 2. Psychological theory. OM 170 G68 It]
RC 530. G58 616.85'2 8()'-2521O
This work is subject to copyright. All rights are reserved, whether the whole or part of the
material is concerned, specifically those of translation, reprinting, re-use of illustrations,
broadcasting, reproduction by photocopying machine or similar means, and storage in data
banks. Under 54 of the Gennan Copyright Law where copies are made for other than
private use, a fee is payable to the publisher, the amount of the feet to be determined by
agreement with the publisher.
by Springer-Verlag Berlin Heidelberg 1981
Softcover reprint of the hardcover 1st edition 1981
The use of registered names, trademarks, etc. in this publication does not imply, even in the
absence of a specific statement, that such names are exempt from the relevant protective laws
and regulations and therefore free for general use.
2125/3140-543210
The absolutely central problem in neurosis . .. is a paradox - the paradox of behaviour which is at one and the
same time self-perpetuating and self-defeating.
O.H.Mowrer
The knowledge which we have that is objective and compelling has only limited extent and beyond this the individual's comprehension and attitude towards himself
make endless progression.
Karl Jaspers
(translated by J. Hoenig and M. W. Hamilton)
Foreword
In view of the practical importance of neurotic disorders (with

something like one-third of the population suffering such disturbances at some time of their lives) and the equally great
theoretical importance of types of behaviour that clearly seem
to contradict both common sense and the law of effect, one
might have expected that psychologists would develop consistent and testable theories of neurosis and that there would be
many textbooks outlining these theories and describing the
experiments done to test them. Oddly enough nothing of the
kind seems to have happened. There is a dearth of theories of
neurosis; those that do exist are not usually put in a readily
testable form, and the amount of research that has been done
in order to test these theories is nothing like as large as one
might have hoped. Nor are there many books setting out the
various theories, the arguments for and against and the empirical evidence; in fact, this may be the only book to have undertaken this task in the past 20 or 30 years.
It is fortunate that the author has succeeded in what is an
extremely difficult and complex task. He has examined issues
and theories dispassionately and impartially, has clarified the
contradictions inherent in most theories and has wisely refused
to come to any kind of final judgment about the adequacy of
the given theories. His aim has been to give a scholarly presentation of the most widely held theories and the evidence
favouring or faulting them; in this attempt he has admirably
succeeded. Only those who have tried to teach courses in this
field will know the difficulties that he must have encountered
and will be able to admire the ingenuity with which he has
surmounted them.
The book will thus be a godsend for academics who have to
teach courses in clinical or abnormal psychology, but it will
also be welcomed by psychiatrists, clinical practitioners in
psychology and indeed all whose business is with neurotic
patients. Unless they are so taken up with one particular
theory that they cannot believe that other theories too might
contribute something useful, they will find here a clear statement of alternatives and a critical discussion of both the
VIII
Foreword
advantages and disadvantages of different theories. The book

thus makes an important and indeed unique contribution to a
topic that has been probably even more confused than most in
the general field of psychology. The author cannot give us the
right answer, but at least he will help us to ask the right questions. That is an important contribution, and the book will
undoubtedly become a classic, to be revised and reprinted
many times as our knowledge advances. One's only regret can
be that it was not written many years ago!
November 1980
H. J. Eysenck
Contents
Introduction . . .. . .. . .. . . . . . .. . . . . ..
Part I. The Concept of Neurosis
1.
Historical Background
1.1
1.2
1.3
Neurosis.......
8
11
The Organic Model
The Psycho-dynamic Model and the Concept of
15
Intrapsychic Conflict . . . . . . . . . . .
Moral Treatment . . . . . . . . . . . .
18
Learning Theory and Behaviour Therapy
20
1.4
1.5
2.
2.1
2.2
2.2.1
2.2.2
2.2.3
The Neurotic Disorders

Psychiatric Diagnosis
The Neuroses
Anxiety Neurosis
Hysterical Neurosis
Hysterical Neurosis, Conversion and Dissociative
Types . . . . . . . . . . . . .
2.2.4 Phobic Neurosis . . . . . . . . . . . . . .
2.2.5 Obsessive-Compulsive Neurosis . . . . . .
2.2.6 Depressive Neurosis (Reactive Depression)
2.2.7 Neurasthenic Neurosis (Neurasthenia)
2.2.8 Hypochondriacal Neurosis . . . . . . . . .
2.2.9 Depersonalisation Neurosis (Depersonalisation
Syndrome) . . . . . .
2.2.10 Other Neuroses . . . . . .
2.3
Problems of Diagnosis . . .
2.4
Reliability and Consistency
2.5
Validity
2.6
Social Context . . . . . . .
26
26
29
29
29
29
30
30
30
30
31
31
31
31
34
35
36
3.
Neurosis, Normality and Psychosis
39
3.1
3.2
Normality and Abnormality

Neurosis and Psychosis
39
46
Contents
Part n. Theories of Neurosis
53
4.
Conditioning Theories of Neurosis
55
4.1
4.2
4.3
4.4
4.5
4.6
4.7
Classical Conditioning
Conflict and Frustration
Two-Stage Theory
Incubation . .
The Stimulus . . .
Preparedness
Therapeutic Implications of Conditioning
Theory . . . . . . . . . . . . . . . . . .
56
58
61
64
66
67
s.
Social and Cognitive Theories . . . . . . .
71
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
Dissatisfaction with Conditioning Theories

Modelling . . . . . . . .
Life Events and Neurosis
Labelling . . . . . . . . .
Attribution . . . . . . .
Cognitive Accounts of Depression
Social Anxiety and Social Competence
Further Implications for Therapy
6.
Psycho-analytic Theories of Neurosis
91
6.1
6.2
91
6.8.2
6.8.3
6.9
Psychological Determinism
Repression and the Unconscious
The Structure of the Mind . . .
Regression and Infantile Sexuality
The Actual Neuroses
The Psycho-neuroses
Little Hans . . . . . .
Neo-freudian Theories
Adler
Homey . . . . . . . .
Fromm . . . . . . . . .
The Evaluation of Psycho-analysis
7.
Trait Theories of Neurosis
. 109
7.1
Guilford
Cattell
Eysenck
Criticisms of Trait Theory
.110
111
114
118
6.3
6.4
6.5
6.6
6.7
6.8
6.8.1
7.2
7.3
7.4
68
71
74
76
78
81
83
85
87
92
94
96
97
99
100
101
101
102
104
105
Contents
XI
8.
The Physiology and Genetics of Neurosis
121
8.1
8.2
8.3
8.4
8.5
8.6
Physiology and Neurosis

Eysenck
Claridge .......
Gray .........
Genetics and Neurosis
Studies of Twins
121
125
127
127
128
129
9.
Oveniew
132
10.
References
139
11.
Author Index
155
12.
Subject Index
159
Acknowledgements
In preparing this book I was helped in different ways by many
of my friends and colleagues. I am particularly indebted to

Hans Eysenck, who pointed out the need for a book of this
sort and who provided constant help and encouragement during the time that I was struggling to write the book. I would
also like to thank Dr. J. Birley, Professor Dick Eiser, Dr.
Michael Feldman, Professor Walter Mischel, Professor O. H.
Mowrer and Dr. Gudrun Sartory for their comments upon
drafts of various chapters and for drawing my attention to
work of which I would otherwise have remained unaware.
Finally, I am grateful to Janet Chesson and Nan Allen for
being so patient with me and for typing out the manuscript so
meticulously.
Introduction
Among the most common psychological disorders are those that have traditionally been
included within the category of 'neurosis'. There are a number of difficulties (both
practical and theoretical) which prevent us from making any precise estimate of how
many people are affected by one or other of the neurotic disorders; but even the most
conservative estimates suggest that several million people must be incapacitated by them.
Some of the less restrained estimates have suggested that as many as 50% of the population may at some time be affected. Whichever figure one chooses to accept, it is clear that
neurotic behaviour cannot be regarded as something that is isolated from everyday life,
or that is unusual. Large numbers of people are seriously affected by the neurotic disorders, and few of us are entirely free from some minor 'neurotic' reaction. Theories of neurosis play an important part in our understanding of human activity and
should not be seen as relevant only to the behaviour of a small number of disturbed individuals.
The manifestations of the neurotic disorders have, of course, been obvious for centuries, though it is only comparatively recently that they have been formulated in the way
in which we now recognise them, and the theories that are presented in this book are all
of comparatively recent origin. The most venerable, and in a general sense, probably the
most influential theory of neurosis still in current use is Freudian psycho-analysis. Freud's
extensive contributions are dated from 1886 to his death in 1939, but his influence still
dominates psychiatric approaches to neurosis. It is an interesting footnote to the study of
neurosis that the concept of anxiety, which plays such a central role in most formulations
of the neuroses, is itself attributable to Freud's writings. In 1894 Freud described anxiety
neurosis as a syndrome distinct from neurasthenia, though the concept did not achieve
the popularity that it currently enjoys amongst psychologists until much later. Again it
was one of Freud's books, 'The Problem of Anxiety' (1936), that was largely responsible
for extending its usage.
Insofar as the various neurotic disorders have a common feature, this is usually thought
to be their involvement with anxiety; and in this respect some theories of neurosis are
largely theories of anxiety. In some disorders, anxiety is the central and dominant feature
- as in anxiety neurosis itself, where the anxiety appears as a generalised feeling of dread
which contaminates almost all aspects of the person's life. In other cases, the anxiety is
not free floating in this way, but is much more closely linked to specific objects or events,
as in certain phobic disorders. Several theories have also regarded the behaviour rituals
and compulsive thoughts that characterise the obsessional and compulsive states as being
a sort of defence against the painful experience of anxiety. Such other neuroses as the
conversion and dissociative disorders might appear to be less closely involved with anxiety, though anxiety is again involved as a central factor in the explanations offered by a
Introduction
number of different theories. Of the traditional neurotic disorders, the one that seems to
be least directly related to anxiety is neurotic or reactive depression.
The neuroses are of immense clinical and theoretical importance to psychology and
medicine. Because of this, it is unusual that there is no single source to which the student
can turn for a survey of the different theories. Any reader who wishes to familiarise
himself with a wider perspective upon the neuroses than that provided by a single theory
must refer to the original material. And whereas there is no substitute for reading the
original works of particular authors, over such a vast field as that covered by the neuroses, the enormity of the task must have deterred a great many readers from all but the
most cursory inspection of alternative accounts to their own. By presenting an outline of
each of the different theories of neurosis within a single volume, it is hoped that this book
will provide the reader with a better understanding both of the neurotic disorders themselves and of the relative strength and weakness of the different theories.
The material presented in the book is necessarily selective, and the problem of selection proved to be a general difficulty throughout the preparation of the book. It was often
difficult to decide what to include and what to leave out. This was a particular problem in
the case of the empirical work relating to the neuroses that has appeared in the journals.
There seemed to be limitless amounts of this sort of material, though its relevance to a
book of this sort was, in many cases, very difficult to assess. There were also problems of
selection in relationship to the theories. Some, like psycho-analysis and the conditioning
models of neurosis, demanded inclusion. Others made less compelling demands for inclusion. The post-Freudian psycho-analytic theories could have been given more attention
than they have in this book. The views of Melanie Klein, for instance, which have been so
influential amongst British analysts, receive virtually no consideration. On the whole, it
was felt that a discussion of the Freudian system itself (particularly when presented in
conjunction with the views of Horney, Fromm and Adler) was adequate to give the
reader an idea of the sort of explanation that psycho-analysts use in this context. I hope
that any omissions will be offset by the advantages of presenting the different theories
within a single volume.
The book deals with material which should be useful both to undergraduates with a
special interest in abnormal psychology and to post-graduate and qualified psychologists
and psychiatrists. Indeed, it is to be hoped that many o.thers who are professionally
involved with individuals suffering from neurotic difficulties will find the book useful
(though a certain minimum familiarity with the concepts of psychology is assumed).
The book is divided into two parts. Part I provides the framework within which the
theories are set. The first chapter provides an historical introduction to the concept of
neurosis itself and describes the emergence of the different formulations of the neurotic
disorders over the past two centuries. Chapters 2 and 3 extend this discussion. Chapter 2
deals with the way in which the specific neurotic disorders are currently formulated.
Discussions of this sort have usually been based upon the nosological systems of
psychiatry, and these are taken as the starting point for the present analysis. Chapter 2
also deals more widely with the problems inherent in different systems of classification
and the difficulties of diagnosis. The final chapter of Part I examines the relationship of
the concept of neurosis to two other general concepts which underpin any approach to
abnormal psychology - those of normality and psychosis.
The second (and major) part of the book presents the theories themselves. These come
in all shapes and sizes. In some cases they are recognisable as theories; in other cases they
Introduction
are more like a collection of studies linked together by a particular view of human
behaviour. Nonetheless, in their own way each provides a particular perspective upon the
sort of problem that is being presented, a particular formulation of the neurotic disorder
itself and a particular sort of explanation of that disorder.
The traditional approach to psychological disorders over the past 200 years or so has
been based upon medical assumptions. This medical (or disease) model leads us to think
in terms of mental illness, of the aetiology and course of illnesses and of 'patients' and
'therapists'; and whatever defects there are in such concepts, it is still difficult to avoid
their pervasive influence. Psycho-dynamic theories of neurosis represent one expression
(albeit an unusual one) of the medical model. In this case the pathology of intrapsychic
rather than physical, but nonetheless disordered unconscious psychological processes is
thought to cause the neuroses, in the same sort of way that the medical model traces
physical symptoms to an underlying physical disorder.
Personality theories have always had an appeal of their own in abnormal psychology.
Trait theory represents a different attempt to explain the neuroses in personality terms,
though the idea of a trait, like so many concepts in psychology, is somewhat elusive. Trait
theorists have sought to identify the patterns of behaviour that remain relatively consistent for each individual over a period of time and in different situations. It has been
suggested that trait theories are, therefore, descriptive rather than causal, though the
conclusions of trait theorists have been translated into causal explanations of behaviour.
In recent years there has been a revival of criticism directed against the most basic
assumptions of trait theory, namely those concerning the cross-situational consistency of
behaviour.
The work of several trait theorists has been closely linked to further assumptions about
the physiological basis of personality. Attempts to explain neurotic disorders in physiological and genetic terms are as old as the concept of neurosis itself. Ever since
Cullen's formulation, different psychologists and physicians have investigated these
aspects of the disorders, and the search for biological causes of neurosis could be seen as
another reflection of the medical tradition.
A quite different tradition, but one with an even longer ~istory, has sought to explain
behaviour by relating it to external, environmental factors. This position was first stated
as far back as 3000 years ago by Homer. The impact of such views upon current
psychologists, however, owes more to the work of Pavlov, Watson, Thorndike and Skinner in the early part of this century. The conditioning accounts of neurosis are still
vigorous, and the work of the behaviour therapists has had a profound impact upon
psychological medicine. But like each of the attempts to explain the neuroses, the conditioning theories face their own theoretical and conceptual problems. It is interesting to
note that among their most vocal critics have been other learning theorists who favour a
more molar, social and cognitive approach. Finally there are those sociological critics
who have argued that psychology and psychiatry have too often failed to take into
account those social forces which operate independently of the individual. Societal reactance (labelling) theory is one of the better-known attempts to explain the development
and maintenance of psychological disorders in such terms.
Each of these theories of neurosis carries with it certain implications for treatment. The
same set of behavioural symptoms look entirely different when seen as a maladaptive
learned response, as an expression of unconscious intrapsychic conflicts or as a function
of social expectations; and the ways in which the clinician might try to help the person
Introduction
who presents with such symptoms will depend upon his theoretical inclinations. The
neuroses are inaccessible except through some set of theoretical assumptions (albeit
implicit and unstated). For this reason, if for no other, it is important for the clinician to
maintain an interest in the theoretical, as well as the more obviously 'clinical', aspects of
his work. Some of the theories are closely related to specific treatment procedures or
approaches. Psycho-analysis, for instance, functions simultaneously as a theory of
neurosis and as a method of treatment. It is, however, beyond the scope of this book to
discuss in any detail the techniques of treatment that have been applied to the neurotic
disorders. It is even further beyond the scope of the book to attempt to evaluate the
effectiveness of such treatments. But since it proved impossible to avoid any discussion of
treatment issues, several chapters do contain a brief discussion of some of the clinical
implications of a particular theory.
Psychologists generally show a regrettable lack of interest in the history of their discipline. Sometimes this is based upon a naive view of research as a sort of cumulative
process which progressively increases our understanding of psychological problems.
Unfortunately, what were thought to be advances quite often turn out to be evasions; and
problems that were thought to have been solved re-emerge years later to puzzle other
generations of psychologists. The advantage of a wider, historical perspective is that it
can help to draw attention to the biases and omissions of current research, as well as
illuminating the reasons why certain other issues should have achieved a more central
status. As psychology becomes more and more specialised (which is itself an historical
rather than a logical development), it becomes increasingly difficult to maintain any
overall perspective upon certain problems in the face of this diversity. Specialised
research frequently produces more specialised research, and the larger issues which
stimulated the research in the first place become obscured. In an area such as that
covered by the term 'neurosis', so many different disciplines have made their own distinctive contributions that an historical perspective is especially useful to counteract the
dangers of treating one approach to the problem as if it were the only legitimate
approach. There are historical as well as scientific reasons for the concepts and theories
of psychology.
Part I
The Concept of Neurosis
1. Historical Background
Abnormal forms of human behaviour have existed for as long as records have been kept,
though the history of abnormal psychology is not necessarily related to that of medicine.
Mora (1967) has suggested that the earliest concepts of mental disorders were probably
influenced by universal beliefs in supernatural phenomena, and specifically by a belief in
ancestral spirits. For this reason the first models of abnormal behaviour were generally of
a religious nature, the mentally disturbed person being seen as powerful and supernormal. In such models it was usually assumed that medicine had no power over the 'possessed' individual, and even that it had no right to deal with such people (Zilboorg and
Henry 1941).
In the Neolithic Age, between about 10000 and 7000 B. c., certain diseases (for
instance convulsive disorders) appear to have been treated by trepanation of the skull
(Kaplan and Kaplan 1956). This operation was probably performed in order to allow the
evil spirit to escape through the hole in the head. Also, there are accounts in the Bible (I
Samuel) of King Saul's possession by evil spirits, and Herodotus describes how an insane
king was confined and eventually committed suicide. Hippocrates (born around the year
460 B. C.) viewed mental disorders from a physical standpoint and saw the brain as the
seat of such disorders. He proposed that the four elements and the four bodily humours
could provide a basis for the understanding of such phenomena. Diseases were seen in
terms bf imbalance in the elements and humours, and this view of pathology lived on
until the seventeenth century; indeed in a much modified form its influence is still with us
(see for instance Pavlov 1955; Eysenck 1957).
Among the achievements credited to Hippocrates are his emphasis on natural causes,
his astute clinical observations, his attempts at classifications and diagnosis, his focus on
the role of the brain, and his psychological view. All these achievements are consistent
with contemporary viewpoints (Ullman and Krasner 1969).
However, with the possible exception of Galen, who lived between around 130 A. D.
and 200 A. D., there were few developments until comparatively recently that add to our
understanding of mental disorders. Christian theology developed a view of mental life
which divorced it from any significant dependence upon the body, and mental disturbances were again seen as manifestations of supernatural agencies - generally possession
by demons. Those who were identified as insane within such a system were often treated
worse than criminals. From the point of view of the abnormal psychologist, one of the
distinguishing features of the Middle Ages is the dominant role which the clergy held in
the designation and treatment of mental disorders; and in particular the events that have
received most attention are the witch-hunts. These have been singled out as significant
historical events in the evolution of psychiatry, though for very different reasons, by such
writers as Szasz (1961), Schoeneman (1977) and Zilboorg and Henry (1941).
1.1 Neurosis
Thomas Sydenham (1624-1689) drew particular attention to the importance of differentiating between diseases and pioneered the idea of a specific pathology underlying
each disease. He was also one of the first physicians to draw attention to the symptoms
of neurosis. Although not specifically interested in disorders of the mind, he described
in detail some of the neurotic and hysterical symptoms and noted the frequency with
which they appeared in his patients. Until this time, those interested in abnormal psychology had tended to focus almost exclusively upon the more dramatic cases of psychosis.
The term 'neurosis' appears to have been first used by William Cullen in his Nosology,
which was published in 1769. In this work, he classified all diseases according to their
symptoms. The class 'neuroses' was used to refer to what we might to-day call neurological, psychosomatic, neurotic and psychotic disorders (Knoff 1970). They were diseases
for which there was no known localised pathology or fever, and they included such
diverse conditions as apoplexy, palsy, hypochondriasis, epilepsy, palpitation, asthma,
hysteria and hydrophobia (Cullen 1781).
According to Cullen, the concepts of nervous energy and of irritability of the nervous
system were of particular importance. In the course of time, many of Cullen's neuroses
were reclassified as neurological and systemic diseases, but the term retained much of the
notion of disordered nerve function without structural pathology for the next century.
The modern usage which denotes minor psychological disorders is in many ways diametrically opposed to Cullen's use of the term (Hunter and Macalpine 1970).
From Cullen, the official psychiatry of neuroses was mostly practised by neurologists,
and the term 'neurosis' referred to any somatic disorder of the nerves or of nerve
function. Such disorders would now be thought of in terms of neuropathy. Almost the last
commitment to the concept of neurosis as nervous disease was made by G. M. Beard,
who first described neurasthenia as a clinical entity (though the term was first introduced
by either Van Dusen or Maynes). Macmillan (1976) has pointed out how Freud derived
his own concept of neurasthenia from that of Beard, who regarded it as a disorder
characterised by general malaise, poor appetite, back pain, neuralgic pains, hypochondriasis, persistent headaches and similar complaints which occur in the absence of
anaemia or organic illness (Beard 1869). Beard suggested that the primary cause of this
condition was a distrubance of the balance between nerve waste and repair, which leads
to an unstable and weakened nerve force. In his original statements, Beard appears to
have believed that phosphorus deficiency was the major cause, though such suggestions
are omitted from his later theories.
At about the same time, Mesmer was extending the tradition of the 'magnetisers' in his
studies of hypnotism, and John Elliotson, a physician at St. Thomas' Hospital, London,
was later to advocate the use of hypnotism in the treatment of hysteria (hysterical
paralysis). Another important historical development came with Charcot's interest in this
phenomenon of hysteria. During the latter part of the nineteenth century, hysterical
disorders became a prominent phenomenon in psychiatry (though they showed a rapid
and mysterious decline after 1900). Hysteria had been considered a strange and incomprehensible disease for centuries. Early conceptions regarded it as a physical illness
confined to females, with its origin in the uterus; and a belief in the sexual psychogenesis
of hysteria still persists in various forms.
Neurosis
In the early 1850s, Briquet made one of the first systematic studies of hysteria, which
he came to regard as a neurosis of the brain with a strong hereditary component. Charcot
based his concept of hysteria on that of Briquet and extended his investigations of the
disorder. One of Charcot's great achievements was his investigation of traumatic paralyses, and in his studies between 1884-5 he showed the similarity between these conditions
and the hysterical paralyses and reproduced similar effects under hypnosis. In 1892 he
distinguished 'hysterical amnesia' from 'organic amnesia' and earned from the public the
curious title of Napoleon of the Neuroses (Ellenberger 1970).
Among Charcot's failings were his over-enthusiastic extension of neurological methods
into this field and his excessive concern with the specification of disease entities. Both of
these features were criticised by Janet (1895). Charcot was also one of the first to label
the hysterical patient as 'sick' in the medical sense of the term, a development which
some writers (e. g. Szasz 1961) consider to be as detrimental as it was significant, since
from this it followed that the status of such people was profoundly altered. Prior to this
reconceptualisation, any physician who considered an hysteric to be sick was being
fooled, whereas now the doctor could be commended for his diagnostic acumen in
spotting such hysterics.
One result of the work of Charcot (together with that of Bernheim) was the release of a
new impetus in psychiatry, and the neuroses were increasingly included within the
respectable domain of medicine. Among those who were profoundly influenced by their
work were Pierre Janet and Sigmund Freud.
Janet began his research with hysterical patients, but widened his interests to include
other neuroses. He attempted to bring a greater coherence to these disorders and prepared a theory of neuroses, grouping them according to their dynamics. These ideas are
summarised in his Neuroses (1909). He distinguished between two basic neurotic conditions - hysteria and psychasthenia, preferring the term 'psychasthenia' to 'neurasthenia'
in order to avoid the implications of neurological dysfunction.
His concept of hysteria was divided into two parts and differentiated the accidental or
contingent symptoms from the stigmata or basic symptoms. The accidental symptoms
were thought to depend upon subconscious fixed ideas, and the stigmata were due to
what Janet termed a narrowing of the field of consciousness. He rejected the purely
neurological theory as well as the theory that hysterical symptoms were faked, and,
following Charcot, regarded hysteria as a psychogenic disease.
Psychasthenia was also thought to include two levels of symptoms. The superficial
symptoms included psychasthenic crises, acute anxiety states and obsessional and phobic
disorders; and at the 'deeper' level, Janet described a basic disturbance in the individual's
relationship with reality. In both the hysterical and the psychasthenic neuroses, Janet
regarded the neurotic disposition as an important factor and described the inherited
constitutional weaknesses of the nervous system in neurotic patients. It was, however, his
dynamic theory that had the most significant influence upon current thought, mainly
through its impact upon psycho-analysis. Both Jung and Adler acknowledged Janet's
influence upon their work, and although Freud in his later work came to emphasise the
differences between his own system and that of Janet, there is good reason to believe that
Freud too owed a good deal to Janet's influence. This question of Janet's influence upon
psycho-analysis is discussed later in this chapter (see The Psycho-dynamic Model).
Jaspers' Allgemeine Psychopatho[ogie (General Psychopathology) first appeared in
1913 and has had a profound effect upon thinking in abnormal psychology, though it was
10
not translated into English until half a century later (by Hoenig and Hamilton in 1963).
One of Jaspers' explicit aims was to produce a systematized approach in terms of
methods in psychopathology and to build a common terminology. He distinguishes three
groups of disorders. Group I includes such diseases as cerebral trauma, cerebral tumours,
acute infections (meningitis for instance) and GPI (General paralysis of the insane) for
which there are known somatic events; in Group II we find the three major psychoses
and in Group III what he refers to as Psychopathien, the personality disorders (including
the neuroses, such as anxiety states, phobias, hysterical disorders, obsessive-compulsive
disorders and reactive depression. Indeed, Jaspers' concept of the neuroses includes all
the conditions included within the contemporary psychiatric systems of classification,
though he tends to make certain distinctions which are not currently seen - such as that
between psychasthenia and neurasthenia).
The first group of diseases satisfies the usual requirement of real disease entities, but it
is the mental and affective disorders represented by the psychoses and the personality
disorders which present the main problem in abnormal psychology. Jaspers suggests that
one must ... assume that many of these psychoses have a somatic base which one day will
be known (General Psychopathology 1972). Having made this point, Jaspers goes on to
propose the possible division of psychopathological events into two classes, Groups I and
III. Those that may be classified as organic disorders with somatic origins would comprise
one group, and those that are disorders of living would comprise the other. The psychoses
may be considered to be of rather uncertain status - hovering somewhere between
Groups I and III, in the absence of adequate or complete knowledge of their origins.
When they are better understood they might be expected to fall into either one or the
other of the two main categories. More recently, a similar proposal has been made by
Hans Eysenck (1975).
For the present purposes, it is Jaspers' ideas about the neuroses that are the most
interesting. Of the Group III disorders he says that classification poses very difficult
problems. Diagnostic efforts have got lost in the establishment of individual facts,
mechanisms, states and characteristics. However, he points out how important it is to
distinguish between psychological reactions and types of personality.
To illustrate the significance of this point, Jaspers refers to the work of Schultz, who
attempted to distinguish between the generally neurotic personality and the individual
neuroses. In some cases the neurotic symptoms may occur in relative isolation. In the
neurotic individual, on the other hand, it is sometimes found that a large number of
different neurotic features occur and that these differ both in quantity and quality and
may be constantly changing. Schultz gives the name exogenous neuroses to the
phenomena which are conditioned by external stimuli and which are capable of being
cured by environmental means. To the 'deeper' disorders which are a result of wellestablished personality features he gives the name nuclear or core neuroses. These nuclear or core neuroses are regarded as extremely difficult for the therapist to deal with,
being curable only very slowly through personality development. For this latter type of
disorder, Schultz suggests that Freudian or Jungian therapies may be appropriate. Unfortunately, as Jasper points out, these diagnostic categories were too reliant upon subsequent therapeutic outcome. If the neuroses responded to environmental manipulation
they were exogenous; if they did not, and more lengthy psychotherapeutic procedures
were needed, then this indicated the presence of nuclear neuroses based upon personality
disturbance. It is this circularity which contains the essential weakness of such a diagnos-
The Organic Model
11
tic scheme, and Jaspers suggests that in an area that is so vague, post hoc procedures of
this kind are of very limited value in terms of concrete results. "In the last resort probably
all neuroses point to a nuclear neurosis and even the severest neurosis will still allow
simple exogenous neuroses to arise" (General Psychopathology).
Jaspers argues that for the disorders of Group III no proper diagnosis can be made,
though it is possible to separate the disorders into a large enough number of descriptive
type-groupings. Apart from this, the best that can be achieved is an extensive analysis of
the case which deals with each of the specific personality, social and phenomenological
features. Since Cullen first coined the term 'neurosis', its meaning has been continually
changing. Marks (1973 a ) describes two different meanings. In its first connotation, it has
been used to describe syndromes like anxiety states, depression, phobic disorders, obsessional states and hysterical symptoms. But in its wider meaning, the term has been used
to describe persistent maladaptive behaviour, especially in an interpersonal context,
which is distressing to the person concerned.
The varying and imprecise definitions that have been offered have led to some serious
problems in psychological medicine. They have prevented direct comparisons of the
results of different research studies and caused disagreement between clinicians that have
been at least partly due to semantic confusion. Culpin (1962) expressed his exasperation
with the term, which he describes as" ... that final triumph of the nineteenth century ...
an embodiment of pseudo-physiology, unsound metaphysics, and moral condemnation".
But apart from developments in the concept of neurosis itself, other profound changes
were taking place in abnormal psychology during this period. These were also to have
important effects upon professional attitudes and approaches to be neurotic disorders.
1.2 The Organic Model

The earliest statements of this position were made by Hippocrates, the father of
medicine, who argued that certain mental disorders were diseases of the brain. A little
later, Lucretius in De Rerum Natura pursued the same theme, that mental disorders were
different from normal, healthy behaviour and were to be regarded in the same way as
physical diseases. From a contemporary viewpoint, the organic model (sometimes also
referred to as the medical model, though the two are not strictly synonymous) is usually
traced back to the eighteenth century. Since then it has occupied a dominant position in
psychiatry and has had a massive influence upon the development of research and clinical
practice in psychological medicine.
Wilhelm Griesinger's book, Pathology and Therapy of Mental Diseases, was published
in 1845 and soon became accepted as the most authoritative work in" its field. Griesinger
maintained that mental disorders could only be explained in terms of physical changes
which occurred in the nervous system, and he considered the brain and its pathology to
be the essential feature of all mental illnesses. For this reason he argued that it was futile
for the psychiatrist to pay too much attention to the problems of psychological description and classification, since such phenomena were of only secondary importance as
correlates of the illness.
Henry Maudsley (1835-1918), one of the most influential British psychiatrists of his
day, was also committed to the notion of insanity as a bodily disease. He differs from
earlier writers in his emphasis upon chemical determinants of psychological phenomena.
Lewis (1951) quotes Maudsley's views on this.
12
How different the world looks according as the liver is or is not acting well. Doubt, despair, even
suicide on the one hand; faith, hope, and life-love on the other hand; these are determined
respectively by some minute and subtile organic compound which has been either insufficiently or
sufficiently manipulated before its discharge into the bloodstream.
Maudsley appears also to have accepted that these 'diseases' were incurable, refusing for
instance to classify those patients who were well on discharge from hospital as 'recovered'.
At this time, the concept of neurosis was still at a comparatively early stage of development. Maudsley wrote of three neuroses - the epileptic, the insane and the criminal
neuroses - and saw these in terms of inherited predispositions. 'The sufferer from any
one of these neuroses represents an initial form of degeneracy ... and life to him will be a
hard struggle against the radical bias of his nature' (Maudsley 1899). This predisposition
he referred to as the insane temperament (Neurosis spasmodica or Neurosis insana), that
is, a lack of balance between the different nerve centres and their tendency to instability
and to disruptive action. The neurotic person could not be said to be mad, but manifested
a lesser degree of disturbance and was seen in terms of eccentricity or strangeness, and
this suggestion that neurosis represents a less extreme disturbance than the psychotic
disorders, and that one of the distinguishing features of the neuroses is that the individual
retains some contact with reality, has been an enduring theme of the various concepts of
neurosis. These issues are discussed at greater length later in the book. Maudsley also
suggested that the distinctions between the neuroses were vague and ill-defined, and that
the different nervous diseases were very closely related.
Another figure whose work has had a profound influence on the history of psychiatry is
Emil Kraeplin (1856-1926). The publication of his Psychiatrie and its subsequent translation into English provided a foundation for the current nosological systems of
psychiatry, and Slater and Roth (1974) have suggested that modem psychiatry begins
with Kraeplin. In his general approach, Kraeplin attempted to show that mental illnesses
are like physical illnesses, in that they run a typical and predictable course from their
onset to their conclusion. He placed great value on prognosis - the final outcome of the
illness. Unfortunately, this emphasis on prognosis carried with it the idea that certain
mental disorders had an unfavourable prognosis, as in the case of dementia praecox,
which Kraeplin regarded as an incurable condition. In this way, Kraeplin's views sometimes led to a rather pessimistic and predetermined view of certain disorders.
In this objective - to establish a conception of mental disorders as diseases in exactly
the same sense that disorders such as tuberculosis and pneumonia were diseases - Kraeplin was successful. A contemporary of Kraeplin argued that any classification which is
based upon clinical symptomatology is necessarily unsatisfactory, because the indications
of one form of disease frequently coincide with those of another. Equally, an aetiological
basis is unsatisfactory, since the fundamental causation is frequently unknown, and the
best approach is a classification dependent upon 'the morbid anatomy, for when the
pathology of a disease has once been recognised, the course can be predicted with some
certainty, and the treatment instituted rests upon a solid foundation' (Berkley 1901).
Kraeplin also succeeded in having mental diseases accepted as an appropriate area for
the intervention of medical science (Thomson 1968). Indeed psychiatry became more
completely integrated with medicine at this particular period than at any previous time.
In the editor's preface to the 1906 translation of Kraeplin's Lectures on Clinical
Psychiatry, Johnstone voiced his optimism about recent developments in which a possible
The Organic Model
13
physical basis for certain mental disorders had been proposed. Johnstone was referring
specifically to the work associating general paralysis of the insane with a bacillus. In 1897
a crucial experiment had been conducted by Richard von Krafft-Ebing, a Viennese
psychiatrist, which demonstrated the relationship between syphilis and paresis. The Wasserman test was devised in 1906; and rather later, in 1913, the presence of the spirochete
causing syphilis was discovered in the brain of paretic patients during post-mortem
examinations by Noguchi and Moore. This discovery of the cause of general paresis had a
considerable effect on the study of abnormal behaviour, in that it gave massive support to
the organic model. Since the physical basis of syphilis had been discovered, many
psychiatrists were reinforced in their belief that such a basis would eventually be found
for all forms of abnormal behaviour.
But whereas this event gave impetus to the organic model, the work of Griesinger and
Kraeplin played a large part in establishing the tradition. Zilboorg and Henry (1941)
have commented:
It was medicine which rose to conquer the field of psychiatry. But, curiously enough, medicine
was quite unwilling to accept, was even harshly opposed to, psychotherapy; it was the sphere of
diseases which it was anxious to capture, but it seemed unable to abandon the tradition of the
dissecting room, the apothecary, or the kitchen and insisted upon anatomy, drugs, and diet.
Medicine captured psychiatry, brought it into its scientific empire, and offered it rights of citizenship only on the condition that it learn the language and submit to the administration. Medicine
refused to have psychology admitted; any appearance of psychology was considered an intrusion
or an illegal importation (pp 354-355).
By the end of the nineteenth century and largely through Kraeplin's influence, much of
the confusion surrounding psychological disturbances had apparently been resolved. The
categorisation which clearly delineated different disease entities had been established,
and this system of classification remains the most potent influence upon present-day
psychiatric nosology.
Kraeplin's position is that of traditional medical practice; he saw psychiatric conditions
as diseases in the full sense of the term, each with a specific aetiology, pathology, symptomatology and course of its own. Kraeplin insisted, for instance, upon definite clinical
diagnoses. Diagnosis was deemed to be vital, since it gave the correct basis for the
prognosis of the illness. In his text Psychiatrie he comments:
What has convinced me more than anything else of the superiority of the clinical methods here
used over the traditional diagnostic methods is the certainty with which we are able, on the basis
of our concept of disease, to predict the future course of events, and the ease with which the
student can orient himself under this guidance in the difficult field of psychiatry (Kraeplin 1896).1
But his search for structural lesions in the brain or for biochemical dysfunction yielded no
satisfactory gains in understanding. Kraeplin's own inference in the face of his failure to
determine organic factors at the root of his illnesses was that their aetiology and pathology depended upon hereditary and constitutional factors of an unknown kind. However,
Kraeplin's influence may be said to have had its greatest effect in terms of his concept of
"Was mir aber die Dberlegenheit des hier befolgten klinischen Verfahrens iiber die herkommliche Diagnostik unzweifelhaft dargetan hat, das ist die Sicherheit, mit welcher wir auf Grund
unserer Krankheitsbegriffe den zukiinftigen Gang der Dinge vorauszusagen im Stande sind."
14
psychosis. He succeeded in establishing two new categories of 'functional psychoses' manic depressive illness and dementia praecox. The neurotic disorders, on the other
hand, received rather less attention. In his Lectures on Clinical Psychiatry (1906), for
instance, the only type of neurotic disorder that is discussed is hysteria. This failure to
differentiate the neuroses from the psychotic disorders represents one of the great
deficiencies of Kraeplin's work.
The various theories of psychopathology that have been proposed since Griesinger
have implicated almost all biological systems in the causation of mental ilIness. There are,
however, certain repetitive themes. Sheflen (1958) notes the tenacity of the conviction
that some lesion or somatic dysfunction causes mental illness, despite the general paucity
of empirical support for any such belief. One consequence of this belief is that the
pathology is frequently seen as a passive response to toxic metabolites, pathogens or
inherited effects: also brain dysfunction and mental symptoms are seen not as correlates,
but as related through cause and effect. Hunter (1973) has stated that an abnormal
mental state is equivalent to a physical sign of something going wrong in the brain and
that ultimately mental disorders wilI be shown to be caused by physical dysfunction. Like
Griesinger, Hunter also directs attention away from the psychological features of the
disorder, on the grounds that these psychological aspects do not constitute the disorder
itself, nor even its essence or determinant, but an epiphenomenon.
In recent years there has been some dispute about this organic approach to abnormal
psychology, and it has received a good deal of criticism (see for instance, Szasz 1961;
Eysenck 1961). But despite the fact that several psychiatrists have denied that the model
is stilI used, this view retains its enthusiastic advocates, and the majority of psychiatrists
probably implicitly accept the model. Lazare points out how few psychiatrists explicitly
identify the conceptual models that they use in their formulation of cases, and suggests
that one of the important models used is in fact the medical model. This regards psychiatric illnesses as diseases like any others. For each disease, it is assumed that eventually a
specific cause related to the functional anatomy of brain wilI be found. The physician
using the medical model concerns himself with the problems of aetiology, signs and
symptoms, differential diagnosis, treatment and prognosis. Knowledge of the particular
syndrome or disease is assumed to determine the treatment, and although the doctor
addresses his patients with proper medical respect, he keeps his distance so as to maintain
his objectivity (Lazare 1973).
Dietz (1977) accepts this definition of the medical model but suggests that it is no
longer the prevalent ideology among psychiatrists and has not been so probably since the
1960s. Despite this, there is stilI a considerable amount of support being offered. It is
claimed, for instance, that the primary identity of the psychiatrist is that of a physician,
and that the medical model is the most useful and appropriate for the practice of
psychiatry (Foulds 1955; Ludwig 1975; Engel 1972; Ludwig and Othner 1977; Sandifer
1977).
However, there is no necessary connection between the medical model and the organic
model, though the two usually go together. One psychiatrist who has stressed that the two
are different is Clare (1976), who asserts that the medical model is more broadly based
than the organic model, and that it takes into account both environmental and individual
factors. Nonetheless, Clare accepts the notion of underlying causes, which is common to
both. It is this feature of the two models that has most significance to this discussion of
abnormal psychology.
The Psycho-dynamic Model and the Concept of Intrapsychic Conflict
15
Although the underlying cause is usually inferred to be some sort of physical disease
entity, in some special cases the medical model is applied in an allegorical sense. The
most important example of this concerns Freudian systems of psycho-analysis. Psychoanalysis, like the literal, organic version of the medical model, relies upon this notion of
underlying causes. The abnormal behaviour itself is seen as the symptomatic expression
of the 'deeper' pathogenic condition, and 'treatment' should be geared to the elimination
of this underlying disorder. The 'superficial' behavioural problems are related to their
cause in the same way that elevation of fever or white corpuscle count are manifestations
of infection in the body (Stuart 1970). Craighead et al. (1976) refer to this as the quasimedical model of abnormal behaviour.
1.3 The Psycho-dynamic Model and the Concept of Intrapsychic Conflict

The psycho-dynamic model of abnormal behaviour assumes the existence of hypothetical
'psychological diseases'. Although these are not usually considered to be real entities,
they are assumed to function in a similar manner to the systemic or traumatic disorders of
the organic model. Neurotic symptoms are seen as the symbolic expression of some sort
of underlying conflict, and the symptoms themselves are assumed to have deeper
psychological significance (Blum 1966; Laughlin 1967).
The psycho-dynamic model can be traced back to Ancient Greece, and Ducey and
Simon (1975) have described the development of ideas about the determinants of mental
life from the time of Homer to that of Plato - from roughly the eighth to the fourth centuries B. C. The Platonic position contains a number of remarkable similarities to current
psycho-dynamic models of intrapsychic conflict, and Freud himself always acknowledged
his debt to Plato and the Greek tragedians. Plato, for instance, was one of the first to
present a unified concept of the sexual instinct, and both Plato and Freud taught the
original bisexuality of the human being and the sublimation of the sexual instinct. The
earlier Homeric position, on the other hand, has more in common with social learning
theory interpretations of neurosis. Table 1 presents some features of the two positions.
Any analogy between the positions of Homer and Plato and modern psychologists
needs to be treated with some caution, since there are also massive differences. Plato, for
instance, stressed that the emotions were to be subdued by reason, which differs radically
from Freud's views about the essential role of the emotions. Nonetheless, it is possible
that there is some direct link between the Platonic position and Freud. Freud was certainly interested in Ancient Greece. This is reflected in his choice of terms (Oedipus
complex, Thanatos, Electra complex), and Simon argues that Freud's assumptions about
mind, madness and the proper treatment for intrapsychic conflicts were influenced by
Plato (Simon 1972, 1973).
Because of the impact that Freud's work has had upon contemporary thought, there is
a tendency to equate the intrapsychic conflict model with Freudian psycho-analysis.
There are, however, other important models of this sort, and Freud's originality has been
somewhat overstated in accounts of the development of such ideas. The determinants of
mental illness that are suggested by the dynamic schools are based upon a concept of
mental energy which is not greatly different from earlier theories of nerve fluids. Beard's
concept of neurasthenia relied heavily upon the idea of energy distribution; and Mesmer
also thought of health as a product of the balance or imbalance of a universal physical
16
Table 1. Models of man in the writings of Homer and Plato. This compares the models of psychological disorder that are contained in the writings of Homer and Plato. The same issues are dealt with at
greater length by Ducey and Simon (1975), Simon and Weiner (1966), Simon (1972, 1973) and
Culpin (1962)
Homeric View
What we consider to be inner mental states are
represented in concrete observable behaviour.
Mental activity is not to be seen as intrinsically
private and inaccessible but as amenable to
observation.
The events of a person's life constitute his identity - though this does not imply a passive and
deterministic view of human nature.
.
Platonic View
Primary emphasis upon the psyche as a structure, a mental apparatus, the parts of which are
in conflict, each having its own interests, needs
and ways of functioning. The rational (Iogistikon) is in opposition to the appetitive
(epithumetikon ).
Emphasis upon 'Eros' as an important driving
force in human behaviour.
Identity is closely linked to family, friends and

social environment.
Concept of sanity as ideal functioning. Therapy

through self-knowledge, personality growth and
insight.
Mental disorders are caused by external events

and forces, not by internal conflict.
Emphasis upon internal life of the individual,

and use of introspection as a way of knowing.
Mental disturbances can be relieved by attempts

to reintegrate the person within the social
group.
Mental disturbances as a result of intrapsychic

conflict. Sickness of the psyche caused by the
dominance of primitive forces.
Therapy builds upon the relationship between
philosopher and student.
fluid and regarded hypnosis in terms of the circulation of this fluid in the body of the
hypnotised person or between the hypnotist and subject. Many of Freud's other ideas
have clear historical antecedents; Ellenberger (1970) and Marx and Hillix (1963) have
identified Leibniz, Bernheim, Charcot, Schopenhauer, Darwin, Fechner, Janet, Herbart
and Meyer among the many influential figures to propose ideas that Freud later included
within his own system.
Leibniz was the first to offer a psychological theory of the unconscious mind and to
distinguish degrees of consciousness. A century later Herbart extended these ideas and
introduced a dynamic conception of the conflict of unconscious forces as they struggle to
reach consciousness. Herbart thought of the threshold of consciousness in terms of the
conflict of a dynamic system of perceptions and memories, in which the stronger ones
force down or repress the weaker. The repressed representations then continue their
efforts to emerge into consciousness and often become associated with other representations in their struggle.
These concepts of conflict and repression which play such an important role in psychodynamic systems can also be traced back to Schopenhauer and Nietzsche. Schopenhauer
explained insanity as being due to The Will's opposition to repellent information and its
resistance to allowing this to become conscious; and Nietzsche too described repression
as an active, inhibitory force. Indeed, the philosophical works of Schopenhauer and
Nietzsche contain some of the closest parallels to psycho-analysis. Thomas Mann, who
read Schopenhauer before he became familiar with Freud's work, described psycho-
The Psycho-dynamic Model and the Concept of Intrapsychic Conflict
17
analysis as Schopenhauer translated from metaphysics into psychology. This can be seen
in the striking resemblances between The Will and Freud's later concept of the unconscious. The Will was characterised as a blind, irrational driving force which determined
man's behaviour, and Schopenhauer compared consciousness with the surface of the
earth, the interior of which remains unconscious and unknown to us: among the irrational
forces, the most powerful was the sexual instinct. The study of hypnotism had also led to
a concept of the duality of the human mind and, later, to the notion of multiple subpersonalities. In his book, The Double Ego, Dessoir presented a statement of the former
position - that the mind functioned at two separate levels, each of which had its own
distinct characteristics. Dessoir called these upper consciousness (OberbewuBtsein) and
under consciousness (UnterbewuBtsein). It was from this tradition that Janet derived his
concept of the subconscious.
The experimental investigations of perception that Fechner and Helmholtz were conducting around 1860, and their implications in terms of thresholds of consciousness, and
unconscious interference with conscious processes were also potent influences upon
Freud's thinking. Helmholtz (1866) coined the term unconscious interference to describe
the relationship between perceptions and past experiences; stimuli are perceived not
merely in terms of their physical properties but also in terms of their associations. Freud
frequently quoted Fechner and derived certain features of his concept of mental energy
and the pleasure principle from him. Fechner's comments that the difference between
waking and sleep states was less one of intensity of function, but was more like the
alternate presentation of the same mental activities on different theatre stages, have also
been said to have formed the basis for Freud's topology of the mind (Ellenberger 1970).
Pierre Janet was the first to use the term 'subconscious', and he was also the first to
publish a report of hysterical patients being cured through bringing their subconscious
fixed ideas into consciousness (Janet 1891). Marcelle, a young woman of 20, showed
difficulty in carrying out movements in which any sort of conscious decision was required
and also displayed amnesias. Janet classified these and other symptoms according to their
depth within the personality - from the most superficial, which he equated with posthypnotic suggestions, to the most profound, such as early traumatic events. Among the
events noted in treatment are the substitution of symptoms and the effects of catharsis.
Janet also discusses the gradual process of removing the superficial layers of delusions in
order to penetrate into the deeper, more basic disturbances. As these were worked
through, the patient showed considerable improvement. One of the general points made
by Janet in this case is that nothing is ever lost to the human mind, though it may be
hidden deep within the subsconscious.
In his earlier work (e. g. Breuer and Freud 1895), Freud acknowledged Janet's influence, but later he increasingly came to deny the similarities between their two positions.
In 1896, Freud called his own system 'psycho-analysis' in order to differentiate it from
Janet's 'psychological analysis', and Ellenberger (1970) suggests that Freud emphasised
the differences by giving a distorted account of Janet's views, asserting, for instance, that
Janet's theory of hysteria was based upon the notion of 'degeneration'. Among the
concepts and ideas that Freud shared with, or derived from, Janet, are the reality principle (Janet referred to this as the 'function of reality'), the role of repression (narrowing of
the field of consciousness) in the aetiology of hysterical symptoms and the emphasis upon
the nature of the patient-therapist relationship. But despite the similarities in their positions and the fact that Janet had made the earlier statement of some of the important
18
features of the new psycho-dynamic theory, the slow development of interest in his own
work in its academic framework contrasts with the remarkable popularity which Freud's
psycho-analysis acquired. Of the period between 1905-1910, Brill (1936) commented
that, 'Not only were the Freudian principles applied to the patients, but psychoanalysis
seemed to obsess everybody at the clinic.'
1.4 Moral Treatment

This tradition has been largely out of favour for some time. It was superceded by the
medical model, and in many ways the two conceptions of psychological disorders are
quite opposed. Whereas the medical model relies heavily upon the disease concept, with
its implications of physical and organic disorder, and therefore upon medical modes of
intervention, both Pinel and Tuke emphasised the rational aspects of the disturbed
person and looked to psychological treatments. Tuke (1813) explains his preference for
moral treatment, 'attaching as I do, little importance to pharmaceutic preparations'.
Over 150 years later, the same point has been restated - that too often psychiatrists
obscure the psychological and moral problems through their emphasis upon the physical
and physiochemical needs of their patients (Sederer 1977).
Historically, the moral treatments are often traced back to one specific event - the
'striking off of the chains' of the patients at the Bicetre Hospital in Paris which occurred
in 1793; and humanistic forms of psychiatric treatment flourished during the period
following the French and American Revolutions. In his classic work 'A Treatise on
Insanity' (1801), Pinel comments,
My experience authorises me to affirm, that there is no neccessary connection between the
specific character of insanity, and the nature of its exciting cause. Among the cases of periodical
mania, which I have seen and recorded in my journals, I find some which originated in a violent
but unfortunate passion, others in an ungovernable ambition for fame, power or glory. Many
succeeded to reverses of fortune; others were produced by devotional phrenzy; and others by an
enthusiastic patriotism, unchastened by the sober and steady influence of solid judgement. The
violence of maniacal paroxysms appears, likewise to be independent of the nature of the exciting
cause; or to depend, at least, much more upon the constitution of the individual, - upon the
different degrees of his physical and moral sensibility ...
Pinel developed a simplified classification system which included the neuroses (after
Cullen), though unlike Cullen's, Pinel's concept of hysteria and the neuroses seems to be
that their importance as nervous disorders (i. e. physical disorders) was of less significance than their status as moral (i. e. psychological) disorders. King (1958) has described
Pinel as providing a bridge between the nosologists of the eighteenth and twentieth
centuries. But Pinel's principal concern lay with his patients and their clinical description
rather than with rigid systems of classification.
Another individual whose name is associated with this approach to the treatment of the
mentally disturbed individual is William Tuke, the founder of The Retreat Hospital in
York. The decision to open a fund for the establishment of The Retreat almost coincided
with the actions of Pinel at the Bicetre in Paris, though whereas Pinel was motivated
largely by his scientific studies, the founders of The Retreat were swayed primarily by
humanitarian and religious motives. Samuel Tuke has stated the aims of the managers of
The Retreat. They were based upon the care and alleviation of the mental disorder 'by
Moral Treatment
19
judicious modes of management and moral treatment'. It was recognised that "Insane
persons generally possess a degree of control over their wayward propensities. Their
intellectual, active and moral powers are usually rather perverted than obliterated: and it
happens, not unfrequently that one faculty only is affected. The disorder is sometimes
still more partial, and can only be detected by erroneous views, on one particular topic"
(Samuel Tuke 1813).
In his description of the early history of The Retreat, Hunt (1932) points to certain
questions posed by the proponents of moral treatment. Firstly there was the question of
how the patients could be helped to control their disorder, and secondly 'by what means
the general comfort of the insane is promoted'.
Both Pinel and Tuke recognised that the abnormal behaviour found among the insane
was dependent to some degree upon the psychological and social environments in which
such people found themselves, and their view of the causes of madness emphasised
psychological factors. Indeed the term 'moral' has been regarded as synonymous with
'psychological' by Sederer (1977), who points out that the German phrase for this form
of treatment - psychologische Hilfsmittel- means 'psychological means of help'.
Moral treatment usually involved making the patient feel comfortable, arousing his
interest and forming a friendly relationship with him. Having established a relationship in
a friendly atmosphere, free from restraint, the patient was then encouraged to discuss his
troubles. Manual work was considered beneficial, and the patients' time was filled with
purposeful activities (Bockoven 1956). This form of treatment has similarities with several of the present-day psychotherapies. Proponents of dynamic psychotherapies may
find in the friendly relationships that were such an important aspect of moral treatment
something suggestive of 'positive transference'. Equally, the Skinnerian behaviour
therapist may see the similarities between the emphasis upon responsibility and appropriate behaviour of moral treatment and modern operant programmes. In a comparison of
the therapeutic orientation of Pinel, Tuke and others such as Vincenzo Chiarugi, who was
working along similar lines in Italy at this time, Mora (1967) shows that each of their
approaches emphasised the healthy part of the patient's behaviour and personality, and
Albee (1969) refers to moral treatment as a suitable model for contemporary psychotherapy in its emphasis upon the psychological strengths and skills of the patient, rather
than upon his 'sickness'. It is easy to recognise in these features certain aspects of
behaviour modification (Bandura 1969); there are also aspects of moral treatment that
are closely related to what we would today call occupational therapy and recreational
therapy, though any attempt to draw too close a comparison is bound to be misleading. It
is probably a mistake to regard moral treatment as a specific procedure. It was much
more of an attempt to create a generally favourable environment for the person, in which
spontaneous recovery might take place.
Although the direct influence of moral treatment upon modern psychiatry and clinical
psychology has been comparatively slight, the significance of these developments lies in
their expression of the humanistic tradition, with its emphasis upon the person and his
human dignity. In recent years there has been an increasing expression of dissatisfaction
among psychologists with mechanistic and reductionist approaches to the explanation of
human behaviour (Kagan 1967; Sanford 1965; Braginsky and Braginsky 1974; Giorgi
1970). More specifically there has also been a reaction to the inadequate and damaging
social environments that exist in many psychiatric hospitals. Goffman (1961) has argued
that the more progressive and medical institutions may be even more damaging than
20
purely custodial ones, and Rosenhan (1973) has given an account of the effects of such
institutions upon sane and 'normal' individuals.
In short, the demonstrations that moral treatment could provide a humane and a
therapeutically effective environment were largely ignored. The medical model that
replaced it relied upon
The authoritative scientific verdict that mental illness was due to incurable brain disease (which)
eliminated the belief, on which moral treatment was based, that patients were capable of responding as persons to sympathetic human understanding. Since no scientific treatment for mental
illness had yet been discovered, mental hospitals became mere receptacles for the incurable
(Bockoven 1956).
As such, the hospitals generally fell back upon custodial care, content to look after the
physical well-being of their patients and fulfilling the dual roles of control and protection.
1.5 Learning Theory and Behaviour Therapy

In contrast to the Cartesian view that man's ideas and behaviour were dependent upon
innate factors, John Locke, the English empiricist philosopher, proposed that the child's
mind be regarded as a tabula rase. Psychological phenomena were to be regarded as a
product of learning and experience. Although others had expressed this sort of view
before Locke (e. g. Aristotle), Locke laid greater emphasis upon acquired behaviour than
those before him.
Locke's views had particular influence on Ivan Sechenov, the founder of Russian
physiology, who in his Reflexes of the Brain (1863) presented the argument that all forms
of behaviour are reflexive in origin. Reflexes were themselves acquired through a process
of associative learning, and pointing to the role of repeated presentations of stimuli in the
formation of all kinds of acquired associations, Sechenov undoubtedly came close to
Pavlov's conception of conditioned reflexes. Pavlov read the works of Sechenov, and they
left a profound, life-long impression on his ideas.
The term 'conditioning' refers to two psychological procedures - classical conditioning
and instrumental conditioning. It is the first of these, classical or respondent conditioning,
which is derived from Pavlov's investigations. Pavlov was a physiologist: in his early
research he was concerned with the heart and the digestive organs, and in 1904 he won
the Nobel prize for his work on digestion. In his investigation of the digestive organs in
dogs, Pavlov noted that the sight or odour of food could provoke the secretion of
digestive juices. Pavlov used this as a vehicle for the investigation of the functions of the
cerebral cortex and was the first to use the procedures and concepts of conditioning,
conditioned and unconditioned stimuli, reinforcement, irradiation (stimulus generalisation) and extinction and spontaneous recovery (Gormezano and Moore 1969). In his
classical experiment on the conditioned reflex, Pavlov presented the sound of a bell at the
same time as food powder was given to the experimental animal. After a number of
simultaneous presentations the bell alone came to elicit the salivation. The essential
feature of classical conditioning is that an unconditioned stimulus (UeS), which reliably
produces an unconditioned response (UCR), is presented with a cbnditioned stimulus
(CS), which has been shown not to produce the UCR. By repeated presentations of the
es and ues (in a specified time sequence), the CS comes to produce a response similar
21

UCS - - - - -..I~UCR
[food powder]
[salivation etc]
Time-
CS
[bell]--_
---
UCS
[food powder]
- , UCR
[salivation etc J
CS~
[bell ]
CR
[ salivation etc ]
-similar to UCR
Fig. 1. The classical conditioning paradigm

(Pavlovian A conditioning)
to the VCR. This is the conditioned response (CR). This procedure is sometimes referred
to as Pavlovian A conditioning (Grant 1964) and is shown in diagrammatic presentation
in Fig. 1.
In stage 1, the simple unconditioned reflex is shown. In stage 2, repeated paired
presentations of the CS and ves lead to stage 3, in which the conditioned connection is
established: the es produces the CR, which is similar but not identical to the VCR. An
important feature of Pavlovian A conditioning is the consummatory response, in this case
the ingestion of the food powder.
Of greater relevance to any consideration of the neuroses is Pavlovian B conditioning.
In this case, events are not dependent upon the instrumental behaviour of the animal,
and after conditioning the CS appears to act as a partial substitute for the VCS; that is,
unlike Pavlovian A conditioning, the ves elicits a more complete UCR. Grant (1964)
suggests that this subclass of classical conditioning should be called Watsonian conditioning, after Watson and Rayner's (1920) study.
The interval between presentation of the CS and UCS is known to affect the rate of
conditioning, although the exact details of the relationship between the stimuli are not
fully understood. It has been suggested on the basis of the current empirical data that the
optimum interval between CS and UCS is of the order of 0.5 seconds (Gormezano and
Moore 1969); and it is also generally accepted that conditioning takes place most efficiently When the CS precedes the VCS. At one time Pavlov regarded backward conditioning (where the UCS precedes the CS) as an impossibility (Pavlov 1927). He later
22
revised this opinion, but noted that under such conditions the CR was at best weak and
unstable (Pavlov 1928 and Pavlov 1941).
Apart from his research into digestion and higher nervous activity in animals, Pavlov
was concerned with the problems of abnormal psychology. One event which stimulated
research in this field was the Leningrad flood of 1924. During the storms the animals had
to be rescued from the flood waters which had penetrated the laboratory buildings, and
although all the dogs were saved, they showed several unusual reactions. The conditioned
responses which had been established experimentally prior to the flood disappeared; it
was as if they had never been formed, and it was possible to restore them only very
gradually by means of numerous repetitions. But apart from this, one of the dogs had
acquired a strong aversion to the sound of the bell that had formerly been used in the
laboratory as a conditioned stimulus. Pavlov seems to have regarded this form of experimental neurosis as analogous in some ways to neurasthenic disorders in humans. Frolov,
who was one of Pavlov's students, writes that 'Pavlov came to the conclusion that the
powerful sound, precisely because of its strength ... reminded the dog of the flood ...
the dog had ceased to be able to bear strong stimuli, just as sick persons, neurotics, are
unable to bear them' (Frolov 1938).
At about the same time, J. B. Watson in America was developing a theory of behaviour
based upon this concept of the conditioned reflex. Despite its many similarities to Pavlov's work, Watson's work was done largely independently. A few years before Pavlov's
observations on the reaction of his dogs to the flooding of his laboratory, J. B. Watson
and Rosalie Rayner published an account of the same sort of conditioned emotional
reaction as described by Pavlov in the Leningrad flood (Watson and Rayner 1920). Using
an ll-month-old infant (Albert), Watson paired a neutral stimulus (a white rat) with a
fear-producing stimulus (a loud noise). After only seven pairings the infant showed a
completely conditioned fear response to the white rat. This conditioned response was
found to generalise to similar stimuli, for instance a white rabbit, a fur coat and a Santa
Claus mask. Watson and Rayner concluded that many of the phobias found in human
psychopathology are true conditioned emotional responses.
Watson is usually regarded as an extreme environmentalist who felt that man's complex behaviour patterns were entirely dependent upon learning, and in one of his bestknown pronouncements he asserted that, given any healthy infant, he could make the
child into a doctor, a lawyer, a thief or whatever he chose through conditioning. In view
of this, the comment of Watson and Rayner that 'One may possibly have to believe that
such persistence of early conditioned responses will be found only in persons who are
constitutionally inferior' is interesting in its suggestion that individual constitutional differences may have some effect on the development of neurotic responses. This aspect of
individual differences in responsiveness, however, was particularly emphasised by Pavlov, who recognised three basic properties of nervous function - strength, balance and
lability. On the basis of these, Pavlov used a typology similar to that of Hippocrates,
whose influence he acknowledges (Pavlov 1955).
Around the beginning of the 1920s Pavlov came to the conclusion that the 'strength' of
the nervous system plays a crucial role in the establishment of individual differences
(Nebylitsyn 1964): the basis for Pavlov's typology lay in his ideas concerning the extreme
reactivity and rapid exhaustability of cortical cells. As a result of repeated exposure to a
conditioned stimulus, the cells of the hemispheres were thought to pass into an inhibitory
state. Pavlov developed the theory that different animals exhibited different degrees of
23
reactivity to stimulation and spoke of the ease with which excitation or inhibition built up
within the nervous system as an index of its strength or weakness. The 'balance' within
the nervous system depended upon the equality or inequality of the two processes of
excitation and inhibition.
These properties were closely related to the sort of disorder that an animal or person
might develop. In those cases where there was an imbalance between the excitatory and
inhibitory processes, experimental neuroses could be produced with relative ease. The
excitatory type of animal loses much of its ability for inhibition and becomes unusually
excited and agitated, whereas the inhibitory type becomes excessively passive (Pavlov
1928). The properties of the nervous system are clearly identified as a predisposing factor
in the neurotic disorders.
Unlike Pavlov, who remained a physiologist in outlook, Watson was not prepared to
reduce classical conditioning to brain activity. Indeed, in his book, Behaviourism (1930),
he implicitly criticises Pavlov as one of those 'Psychologists (who) talk . .. quite voluably
about the formation of new pathways in the brain, as though there were a group of tiny
servants of Vulcan there who run through the nervous system with hammer and chisel
digging new tracks and deepening old ones'.
But Watson's contribution to abnormal psychology consists largely in his assertion of
the importance of learning. The psychologist's task was to discover the ways in which the
elemental conditioned reflexes combined to form complex behaviours. He believed that
the process of conditioning could account for many of the conflicts and behavioural
difficulties found in the mentally ill. Indeed, he also considered that conditioning could
lay the foundations for the onset of actual organic changes which could lead finally to
infections and lesions (Watson 1930).
One of the oldest beliefs about learning is that our behaviour is determined by its
effects. Jeremy Bentham's utilitarian emphasis upon pleasure and pain is one expression
of this belief and his ideas were subsequently reflected in several different psychological
theories. A restatement of the 'Pleasure principle' can be found for instance in Freudian
psycho-analysis.
The second sort of conditioning, operant conditioning, has its roots in the work of
Thorndike. Thorndike was influenced by the comparative tradition of Romanes and of
Lloyd Morgan, who was interested in the possibility that simple associative processes
might be found to account for the adaptive learnt behaviour of animals. His book,
Animal Intelligence, was a landmark in the development of psychology. It described a
series of experiments concerned with learning and memory. These experiments used a
wooden puzzle-box with an escape door held shut by a catch: this could be released by
pulling a loop hanging in the box (or by turning a catch in some models). The cats, when
placed in this box, showed various random movements which Thorndike refers to as trial
and error behaviour. Eventually the cat accidentally releases itself, and on subsequent
occasions there is an irregular but generally progressive decrease in the length of time
before the animal makes the correct response. It was on the basis of this work that
Thorndike (1911) formulated his law of effect.
24
The Law of Effect is that: of several responses made to the same situation, those which are
accompanied or closely followed by satisfaction to the animal will, other things being equal, be
more firmly connected with the situation, so that, when it recurs, they will be more likely to recur;
those which are accompanied or closely followed by discomfort to the animal will, other things
being equal, have their connections with that situation weakened, so thilt when it recurs, they will
be less likely to recur. The greater the satisfaction or discomfort, the greater the strengthening or
weilkening of the bond.
Watson was highly critical of what he regarded as the mentalistic language in this formulation, and he argued that Thorndike appeared to 'believe habit formation is implanted
by kind fairies' in his suggestion that pleasure stamps in the successful response and
displeasure stamps out the unsuccessful one (Watson 1930).
Thorndike was later to modify his views on this question of reward and punishment
when it became clear that punishment did not always have the clear-cut effect of stamping out the unsuccessful response. Instead, he began to lay greater stress on the effects of
reward in the modification of behaviour (Thorndike 1932). One of the best-known
experimental contradictions of the Law of Effect was offered by Muezinger (1934). In a
study of the effects of punishment on discrimination learning, one group of experimental
animals was rewarded for correct responses and punished for incorrect responses. This
led to a more rapid rate of learning than among the group which was merely rewarded for
making a correct choice. However, the results also showed that punishing the correct
response produced a more rapid rate of learning than reward alone. This is clearly
contrary to the predictions made by the Law of Effect.
Thorndike modified his views on punishment by suggesting that the effects of both
'satisfiers' and 'annoyers' upon learning were dependent upon what they caused the
animal to do. But this modification brings the position back towards an associative
learning model. Thorndike suggests that the future response to a situation is best predicted by the past associations; that is, stimuli which act during a response tend to evoke
that response on subsequent occasions (Guthrie 1952).
Thorndike's views had a strong influence upon theories of learning and directed them
towards the use of reward or reinforcement as a basic concept in learning. His ideas
provided a foundation for the work of Hull, Spence, Mowrer, Skinner and, more generally, for those psychologists who reject simple association in favour of the concept of
reinforcement. Clark Hull provided a synthesis of Pavlov's work on conditioned reflexes
and Thorndike's research on trial and error learning. Hull's hypothetico-deductive
theory of behaviour (Hull 1943, 1952) used intervening variables to extend the straightforward stimulus-response model into an S-O-R formula. Two of the most important
intervening variables were habit strength (SHR) and drive (D), and Hull laid particular
emphasis on reinforcement as the central principle of learning (Hull 1937). The essence
of reinforcement was drive reduction; any response which preceded drive reduction
would become associated with that drive. Hull argued that classical conditioning and trial
and error learning were both to be explained by this drive reduction principle and not by
different principies or laws (Hull 1943).
Another behaviourist of this period was Edward Tolman. Tolman differed from most
of his colleagues in his interest in behaviour at the molar rather than the molecular level
and in his emphasis upon the purposive character of behaviour (Tolman 1948). This
introduction of cognitive factors stood in marked contrast to the rather mechanistic
formulations of Watson and Pavlov, and in his claim that learning was a process of
25
acquiring knowledge, of learning the meaning of what leads to what, Tolman can be
regarded as one of the pioneers of the recent cognitive trend in abnormal psychology. In
his attempt to develop a 'purposive behaviourism' he again drew attention to an active,
intentional and purposive view of man. A common theme shared by the various cognitive
theories is that behaviour is not simply determined by physical input that impinges on the
senses, but rather depends upon the selection and transformation of information taken
from stimuli.
2. The Neurotic Disorders

Definitions are a kind of scratching and generally leave a
sore place more sore than before (Samuel Butler).
2.1 Psychiatric Diagnosis

A medical diagnosis is a definition. The term 'diagnosis' is derived from the Greek and
implies the distinguishing or discernment of a disease. But it implies other things, too.
Without a proper diagnosis in medicine it is impossible to arrive at a rational prognosis
and treatment of the disease (Warner 1952). Similarly, in psychiatry the diagnostic
classification of data is usually expected to go beyond the description of behaviour and to
make a statement about the past, present and future states of the patient (Frank 1975;
Slater and Roth 1974). By medical criteria, this emphasis upon the information that is to
be conveyed by the diagnosis is crucial, and the evaluation of the current diagnostic
system is largely dependent upon its success in this function.
Diagnosis need not be tied to any particular system of classification, though it is usually
identified with the symptomatological classification system derived from Kraeplin. The
system of classification that is used will depend to a large extent upon the theoretical bias
of the person involved. The phrases 'unresolved Oedipal complex' from psycho-analysis,
'manic depression' from orthodox psychiatric classification and 'introverted neurotic'
from trait theory are all diagnostic classifications (Caveny et al. 1955); though it must be
said that they represent very different sorts of diagnoses.
Some critics (notably the phenomenologists and existentialists) have taken the extreme
position of rejecting all systems of classification in psychological medicine. Classification
is seen not only as a useless exercise, but also as a potentially destructive barrier to
therapy (Rogers 1965; Menninger 1948). The terms 'neurosis' and 'psychosis' have been
described as wrong labels with wrong implications which are both obsolete and dangerous (Menninger, Mayman and Pruyser 1963). Anthony Clare (1976) offered a mordant
statement of this position which regards psychiatric classification as:
an utterly unproductive form of occupational therapy for those psychiatrists who have little
inclination to delve into the deeper recesses of their patients' minds, less stomach for what they
might find there, little aptitude for intuitive understanding, and no ability to empathize with or
imagine themselves into their patients' predicaments. A preoccupation with description and
classification is itself portrayed as a symptom, an abnormal psychological phenomenon whereby
psychiatrists defend themselves against their own inadequacy and fear by emphasizing the 'objectivity' inherent in their arid statistical and classificatory approaches and by hiding behind a
smokescreen of measurements and 'scientific' jargon.
Jellinek (1939) on the other hand has argued that there is no justification for abandoning classification in psychiatry, but that the system must be more securely based than it
has been in the past. He describes two needs that must be satisfied by classification, the
most important of which is that of organising the data according to 'homologous or
Psychiatric Diagnosis
27
analagous characteristics ... (which) reflect common origin'. This function of classifying
a set of objects or events into subclasses according to certain common characteristics has
also been stressed by Hempel (1961); and in psychological medicine it involves the
attempt to classify according to common signs, symptoms or other identifying characteristics. An adequate classification of the data is an almost indispensable requirement
for the development of theory, although the classification itself depends upon the theory
and theoretical assumptions. This latter point is of some significance since it implies that
the concept of neurosis itself is dependent upon theoretical assumptions (albeit
unstated).
In Kraeplin's classification system there were eight major categories. The system in use
between World Wars I and II contained 24 categories, and the 1952 nosology of the
American Psychiatric Association had almost 100 diagnoses. The second edition of the
Diagnostic and Statistical Manual of Mental Disorders (DSM II 1968) now contains over
100 different diagnostic entities. This system (DSM II) and the World Health Organization's Eighth Edition of the International Classification of Diseases (ICD 8), upon which
DSM II is based, represent the two established nosologies of psychiatry. (A third edition
of the Diagnostic and Statistical Manual is currently being prepared.) Among the
developments that have taken place in the American system (DSM II) is the general
elimination of the word 'reaction' as applied to the neuroses. This seems to imply a return
to an earlier and more classical Kraeplinian typology (though this was not the intention of
the APA Committee on nomenclature and statistics), and the uncertainties surrounding
the terminology of diagnosis reflected in this change indicate a fundamental confusion
about the nature of the various neurotic disorders (i. e. whether they are to be seen as
reactions or conditions). But despite the large number of revisions in the official systems,
no truly satisfactory system has emerged (Shakow 1965), and the ever-increasing complexity has not been accompanied by any real advance in the information contained in the
diagnostic terms themselves. Howells (1970) commented that
Developments in a field depend on a number of factors, but probably none so retards progress in
psychiatry today as the confusions of its nosology and, linked with it, the lack of agreement on
criteria for defining syndromes together with the imprecision of its nomenclature.
It might be said that the neurotic disorders can be approached from different theoretical
levels and the failure to take account of this has also tended to obscure many discussions
of the neuroses.
There is general agreement upon the need for some sort of classification system
(Eysenck 1960c; Maher 1966; Shakow 1965). Unless it is possible to isolate and define a
particular disorder or dysfunction, there is little chance of investigating its determinants.
But it is equally important to be able to define and measure the symptoms of a given
disorder before we attempt any classification, and the widespread devaluation of the
importance of symptoms has only further confused this already confused area.
In the most general sense classification systems are devices by which complex material
may be reduced to a simpler and more intelligible form through a process of ordering the
phenomena according to certain of their similarities. The classes chosen for this purpose
are therefore best seen as abstractions imposed from above by the scientist or diagnostician, and not as 'things' inherent in the real world. Nor is there only one system of
classification which is appropriate for a given set of phenomena. This depends upon the
purposes of the investigator.
28
In psychiatry there is considerable disagreement about the requirements which should

be met by the diagnosis. It has been suggested by psychiatrists who are committed to the
organic or medical model that diagnosis in psychiatry should be treated in exactly the
same way as in medicine. The disorders included within a particular category (for instance the neuroses) are expected to have characteristics in common, and these characteristics are in tum required to be different from those of other categories, for instance
the psychoses (Brill 1967). The psychiatric classification system also assumes that the
major clinical variables form distinctive patterns which appear as qualitatively discrete
and discontinuous disorders (Strauss 1975). Brill (1974) has defended this traditional
medical approach to psychiatric classification and lists six requirements of an acceptable
clinical classification. One of the points that he raises is that the diagnoses should meet
the requirements of validity and reliability. Clearly any adequate system of classification
must produce the same result each time it is applied to a specific problem, and it must
also measure what it was intended to measure.
These basic requirements of the medical diagnostic system are that the classifications
should not be merely descriptive, that the data should be divided into independent and
discrete categories, that the classifications should be reliable and that they should be
valid. But before going on to examine the adequacy of the psychiatric classification
system for the neuroses, it is necessary to describe how neurotic disorders are actually
classified. Most sources are in general agreement that anxiety is the chief characteristic of
neurosis. Surprisingly, the term 'anxiety' was hardly used in psychology and psychiatry
until the late 1930s, and the wide currency that the term now enjoys can be traced back
to its introduction by Freud as a translation of Angst (Sarbin 1964). The term itself,
however, has a deceptive simplicity. It is used by behaviour therapist, psycho-analysts,
psychiatrists and physiological psychologists, and its meaning changes each time. Cattell
and Scheier (1961) found that there were 120 different procedures used to assess anxiety. There is no single unequivocal definition of anxiety. As an inferred construct with
numerous different meanings, the ways in which the term is used will depend upon the
theoretical persuasion of the clinician or the experimenter.
The definition of anxiety offered by DSM II reflects the Freudian influence: Anxiety
' ... may be felt and expressed directly, or it may be controlled unconsciously and automatically by conversion, displacement and various other psychological mechanisms' (my
italics). This comment removes the statement from the realm of description and links it to
the inferences and assumptions of a particular model of neurosis. The failure to distinguish between observation and inference has been a common feature of psycho-dynamic
thinking and has led to unavoidable tautologies in their theoretical statements.!
Apart from laying emphasis upon anxiety as a chief characteristic of the neuroses,
DSM II describes the neurotic disorders as causing psychological distress which the
person seeks to avoid. The neuroses are also frequently said to differ from the psychoses
in the patient's awareness of his own disabilities, his lack of gross distortions of perception and his lack of severe personality disorganisation.
Ultimately it may be true, as Lakatos (1970) has argued, that there is no clear division between
observational and theoretical propositions. Nonetheless, it seems to me to be of considerable
importance to distinguish between such statements as 'this condition (hypochondriacal neurosis)
is dominated by preoccupation with the body and with fear of presumed diseases of various
organs' and 'symptoms (of hysterical neurosis) ... are symbolic of the underlying conflicts.'
The Neuroses
29
2.2 The Neuroses

2.2.1 Anxiety Neurosis
DSM II describes this neurosis as characterised by anxious over-concern which may
extend to panic, and which is frequently associated with somatic symptoms. Unlike
phobic neurosis, anxiety may occur under any circumstances, and it is not restricted to
particular situations, objects or events. It is necessary to distinguish this neurosis from
normal states of apprehension and fear which occur in realistically dangerous situations.
Individuals with anxiety neurosis may be thought of as experiencing directly the discomfort and distress generated by that anxiety, whereas other neurotics have developed
methods of redirecting the anxiety into certain symptoms which relieve the immediate
distress (Mackay 1975). Those who suffer from 'free-floating anxiety' tend to experience
a persistent tension, though this may occasionally be interrupted by acute crises in the
form of panic attacks.
2.2.2 Hysterical Neurosis
The psycho-dynamic model has also had an influence upon the formulation offered for
the hysterical reactions.
'This neurosis is characterized by an involuntary psychogenic loss or disorder of function. Symptoms characteristically begin and end suddenly in emotionally charged situations and are symbolic of the underlying conflicts. Often they can be modified by suggestion alone' (DSM II).
The suggestion that symptoms are symbolic and that they imply the existence of
underlying conflicts (presumably in a causal role) may be seen as a further commitment of
psychiatry to the psycho-dynamic position. In more general terms these comments show
the subtle interrelationship between concepts and theoretical assumptions.
2.2.3 Hysterical Neurosis, Conversion and Dissociative Types
In conversion, the special senses and voluntary nervous system are affected, and the
symptoms may take the form of blindness, deafness, anaesthesias, paraesthesias, paralysis, ataxias and sometimes seizures of a pseudo-epileptiform character. The pseudoneurological symptoms in particular tend to correspond to the lay notions of such disorders and are not based upon the anatomical organisation. They are frequently variable
and may dramatically disappear without apparent reason. The patient may show an
inappropriate lack of concern about his symptoms, and it is often found that these
provide secondary gains through winning the patient the sympathy of others, or by
relieving them of some unpleasant responsibility. DSM II warns that the conversion type
of hysterical neurosis must be distinguished from physiological disorders which are mediated by the autonomic nervous system, presumably on the grounds that these are not
under voluntary control, though this assumption may not be warranted (Miller 1969;
Kamiya 1969), and also from malingering which is done consciously.
It is worth drawing attention to the phrase 'secondary gains' in this statement. If it is
accepted as a definition, then it clearly carries a certain theoretical bias. The clear implication is that the conversion neurosis is caused by factors other than the social rewards
associated with it. Operant learning theorists might well object to this presupposition.
In dissociative neuroses, alterations may occur in the patient's state of consciousness or
in his identity, to produce such symptoms as amnesia, somnambulism, fugue and multiple
personality (DSM II).
30
2.2.4 Phobic Neurosis

The phobic neurosis is characterised by intense fear or dread of an object, event or
situation which the patient consciously recognises as presenting no real danger to him.
The apprehension may be experienced as fatigue, faintness, palpitations, sweating,
nausea and panic. DSM II includes within its definition of phobic neurosis certain
psycho-dynamic assumptions. The phobias are described as being attributed to fears
which have been displaced onto the phobic object from some other object of which the
patient is unaware. Many clinicians working within a behaviour therapy framework might
wish to dissociate themselves from this implication of the symbolic significance of phobic
fears.
2.2.5 Obsessive-Compulsive Neurosis
This disorder is characterised by the persistent intrusion of unwanted thoughts, urges or
actions that the patient feels he cannot control or prevent. The thoughts may consist of
single words or ideas, or more often they may be trains of thought. These mayor may not
be nonsensical. The attempt to dispel these experiences may lead to an inner struggle
which occupies all the person's concentration, and there can also be severe anxiety when
obsessive-compulsive activities are interfered with. The actions vary from simple movements to complex rituals such as compulsive repeated hand washing, and they are often
bizarre or eccentric. Sometimes the actions are seen as a means of warding off the threat
associated with morbid thoughts. Obsessive-compulsive phenomena are sometimes
found in endogenous depression, schizophrenia and some organic states (notably
encephalitis), and ICD-8 states that such instances should be excluded from this neurotic
category.
2.2.6 Depressive Neurosis (Reactive Depression)
In reactive depression the extreme dejection is seen as a response to some stressful event
or series of difficulties. DSM II again includes certain unnecessary assumptions about the
significance of internal conflicts in its definition. Reactive depression has usually been
distinguished from psychotic, endogenous depressive illness. ICD-8, for instance, notes
that symptoms which are usually found in affective psychosis, such as early-morning
waking, loss of weight, self-depreciation and marked retardation, are not typically found
in this state, though the distinction between the two conditions is far from clear (Kendell
1975).
2.2.7 Neurasthenic Neurosis (Neurasthenia)
This condition is characterised by complaints of chronic weakness, easy fatigability and,
sometimes, exhaustion. Unlike hysterical neurosis, the patient's complaints are genuinely
distressing to him and there is no evidence of secondary gain (DSM II).
Despite the agreement of DSM II and ICD-8 upon this neurotic disorder, it is one of
the most vague categories. DSM II suggests that it be differentiated from anxiety neurosis
by the nature of the predominant complaint and from reactive depression by the moderation of the depressive symptoms. The earlier DSM I referred to this condition as a
psychophysiological nervous system reaction. A further complication is that such neurotic states are virtually identical to the separate diagnosis of asthenic personality (included
within the personality disorder category). This behaviour pattern is characterised by easy
fatigability, low energy level, lack of enthusiasm, marked incapacity for enjoyment and
Problems of Diagnosis
31
over-sensitivity to physical and emotional stress (DSM II). DSM II also adds the instruction 'this disorder must be differentiated from "Neurasthenic Neurosis"', but no assistance is given to the diagnostician as to how this might be achieved. Apart from illustrating a further weakness inherent in the psychiatric nosology, the vagueness of the definitions offered for neurasthenic neurosis raises the doubt as to the value of maintaining a
separate diagnostic term for this disorder.
2.2.8 Hypochondriacal Neurosis
In this condition, the patient shows a persistent anxious preoccupation with his health,
often in the form of a fear of presumed diseases of various organs. The fears do not
achieve the same degree of intensity that is sometimes found in the delusions of psychotic
depression, but they persist despite reassurance or contrary evidence. Unlike hysterical
neurosis, there are no actual losses or distortions of function.
2.2.9 Depersonalisation Neurosis (Depersonalisation Syndrome)
This syndrome is dominated by feelings of unreality and of estrangement from the self,
body or surroundings - by an unpleasant feeling of strangeness or unreality. Sometimes
the person feels that his own actions have an automatic, robot-like quality.
2.2.10 Other Neuroses
This classification includes specific psycho-neurotic disorders that are not described elsewhere. DSM II mentions 'writer's cramp' as an example and warns the clinician against
using this category for patients who present with 'mixed neuroses', which it suggests
should be diagnosed according to their predominant symptom.
2.3 Problems of Diagnosis

The foregoing categories may be used as an initial statement of the different sorts of
neurotic disorders, but there are, necessarily, disagreements about the specific expressions of neurosis. This is partly due to the different theoretical assumptions made by
different authors. Although they are not explicitly included in the above categories, the
following disorders have also been regarded as neuroses: tics, stuttering, enuresis, various
psychosomatic disorders such as asthma, headaches dermatitis, insomnia, obesity,
alcoholism and drug dependence. At the other extreme, some psycho-analysts have
placed great emphasis upon the concept of character neurosis which may be found in the
absence of specific symptoms.
At the time of writing this book, the second edition of the Diagnostic and Statistical
Manual is being revised, and it seems likely that DSM III will dispense completely with
the general category of 'neurosis'. The various neurotic disorders will instead be reclassified under other headings. Neurotic depression, for instance, is to be subsumed under
'affective disorders'; phobias, obsessional and compulsive disorders, and generalised
anxiety disorders are each included within the 'anxiety disorders' section; hysterical and
conversion neuroses appear under 'somatoform disorders', and depersonalisation
neurosis is included in the 'dissociative disorders' category.2 The traditional category of
2 This is taken from the draft of Axes I and II of DSM III, available from January 16 1978.
32
'neurasthenic neurosis', which was included in DSM II and leD 8, has been omitted from
the revised American system (which may not be a bad thing in view of its vagueness).
But apart from abolishing the general category of 'neurosis', DSM III remains very
much a part of the psychiatric-medical tradition. If the specific diagnoses are required to
go beyond description and to make a statement of aetiology and prognosis (Warner
1952; Kendell 1975; Slater and Roth 1974), then the psychiatric nosology must be
regarded as a failure, since apart from a few psycho-dynamic assumptions about the
causes of neurotic disorders, the diagnoses are largely descriptive.
After a comprehensive survey of the research concerning psychiatric diagnoses, Frank
(1975) concludes that the evidence weighs heavily against the current system of classification used in psychiatry: the system is purely descriptive, and Frank argues that to be
useful, the diagnostic statements should provide information about the individual beyond
merely describing his symptoms. However, Frank is probably assuming that a classification system of this sort could in principle be built upon symptomatology. This is doubtful,
since invariant conjunctions of symptoms are not to be expected in psychiatric disorders,
and it is the persistent assumption that there is some deeper disease entity which underlies the symptom which has again produced this breakdown in the system. By asserting
that the diagnoses are 'purely descriptive', Frank overstates his case. Paradoxically, a
purely descriptive system could be of more value than the present one, in which diagnostic statements based upon aetiology alternate with others based upon symptom description. Zubin (1967) has complained about the puzzling failure of diagnoses to match with
symptoms in a system which is largely based upon overt symptomatology. This failure is
largely a result of the confusing mixture of aetiological and descriptive content of the
diagnoses themselves. The simultaneous use of these two different approaches in diagnosis increases the chaos in psychiatric classification, and Essen-Moller (1967) has advocated the complete separation of these aspects. Given the present state of knowledge (or
ignorance) concerning most psychiatric disorders, disagreements over aetiology are
inevitable. The function of science is both to describe and to explain events, and a
descriptive system which avoided unnecessary assumptions about aetiology could be
useful in its delineation and clarification of the problem to be explained.
The present system has other failings which are also due to the inclusion of alternate
statements of symptom description and aetiology in the diagnoses. It is not clear, for
instance, whether or not more than one diagnosis should be made. It is a fundamental
principle of most classification systems and a traditional aspiration of all branches of
medicine that each patient should be restricted to membership of a single diagnostic
category (Kendell 1975).
In DSM I, certain multiple diagnoses were forbidden (e. g. alcoholism could not be
made as a separate diagnosis when it was associated with an underlying disorder). The
special instructions for DSM II, on the other hand, state that multiple diagnoses are
permissible and cite as an example the fact that a patient with anxiety neurosis may
develop morphine addiction. But this example avoids the fundamental problem inherent
in multiple diagnosis. There are fewer difficulties raised by the use of multiple diagnoses
for different categories of disorders (such as neuroses and drug dependence) than by the
use of multiple diagnoses within a category.
Jaspers (1972) considered the difficulties raised by this problem and proposed a diagnostic hierarchy in which his Group I disorders took precedence over Group II disorders;
these in tum took precedence over Group III disorders. Nonetheless, Jaspers recognised
Problems of Diagnosis
33
that his proposal was unsatisfactory, particularly for the neuroses and personality disorders, and suggested that the use of categorical diagnoses might have to be abandoned for
a classification system in which every case may appear as often as the diagnostician likes.
Within such a system, 'all the possible kinds of data can be enumerated but we are no
longer enumerating diseases'.
More recently, Foulds restated the hierarchical principle. He suggested four classes of
personal illness - class 4, delusions of disintegration; class 3, integrated delusions (e. g. of
grandeur or persecution); class 2, neurotic symptoms (conversion, dissociative, phobic,
compulsive and ruminative symptoms); and class 1, dysthymic states (anxiety and depression) (Foulds and Bedford 1975). All members of class 4 should also fall into classes 3, 2
and 1; members of class 3, into classes 2 and 1, and members of class 2, also into class 1.
It is interesting that Foulds proposed the separation of anxiety and depression from the
other neurotic disorders on the grounds that the relationship between anxiety and depression and the conversion, dissociative, phobic, compulsive and ruminative symptoms is
different from that between pairs of these symptom groups (Foulds and Bedford 1976).
For this reason, anxiety and depression were allotted a lower position in the hierarchy.
Lesse (1970) used a similar hierarchical model in his study of the psychosomatic
disorders. Of 151 patients suffering from autonomic faciocephalagia (cluster headaches),
migraine headaches, essential hypertension, colitis, asthma and neurodermatitis, Lesse
found that psychosomatic disorders rarely exist as an isolated clinical entity. They usually
occur as part of a wider neurotic disorder and are usually secondary to anxiety. Such
other symptoms as phobias, hypochondriasis, obsessions and compulsions were generally
noted prior to the appearance of psychosomatic problems.
Another problem for the clinician who wishes to make a single differential diagnosis is
that in practice, anxiety and depression are often found together. One extensive study of
depression (Grinker et al. 1961) found five factors underlying diagnosed depressive
disorders. These were (a) hopelessness, sadness, helplessness and feelings of unworthiness; (b) a concern over material loss and conviction that the emotional state could be
remedied by external changes; (c) gUilt over some perceived wrongdoing and a wish to
make restitution; (d) envy, loneliness, martyred affliction and secondary gains through
provoking guilt in others; and (e) a free-floating anxiety factor. Grinker's investigation
showed that the anxiety factor was an important part of the clinical concept of depression. A diagnosis of depression was contingent not only upon depressed affect in the
patient, but also on the presence of anxiety; and in those cases in which there was only
minimal depressive affect, the anxiety factor greatly increased the probability of a 'depressed' diagnosis. The instructions of DSM II that the diagnostician 'should not lose sight of
the rule of parsimony and diagnose more conditions than are necessary to account for the
clinical picture' is unhelpful here. If each of the different neurotic disorders are important
features of the patient's difficulties, the psychiatrist can only list them all or limit himself,
on the basis of a subjective choice, to a single 'differential diagnosis'. Masserman and
Carmichael (1938) have complained about a similar problem. One hundred patients who
had been admitted to the psychiatry department of the University of Chicago Clinic were
re-examined after 1 year. Except for those individuals who showed a 'well-demarcated
anxiety syndrome', there was a wide variation in the diagnostic categories, 'indicating the
mixed character of nearly all neurotic and psychotic reactions'.
The tendency for mixed states to occur poses another major problem for any system
such as that of Kraeplin which relies upon categorical distinctions between the disorders
34
(Curran et aI., 1972). The same difficulty arises in those cases in which the overt symptomatology changes over time. In other areas of psychiatry the problem is even worse. In
drug addiction, for instance, the use of discrete categories for dependence upon opiates,
barbiturates, stimulants, hallucinogens or other drugs breaks down completely, since the
general pattern of drug taking is that of multiple drug abuse (Blumberg et ai. 1974;
Gossop and Connell 1975; Gossop 1978a). The diligent diagnostician must resolve the
mutually exclusive requirements of validity and of parsimony for himself; the system has
broken down.
A further difficulty arises if the diagnostician chooses to select from a large number of
potential diagnoses in order to present his own interpretation of the most salient features;
but despite the problems inherent in this introduction of subjective bias, this is exactly
what the psychiatrist is instructed to do. DSM II states that for patients with several
different neurotic disorders, the clinician should make a diagnosis according to the predominant symptom. The simplicity of the word 'predominant' has little in common with
the complexity of the selective perceptual and judgmental processes involved in the
actual choice of what constitutes a 'predominant symptom'. This introduction of subjectivity could be confidently predicted to lower the inter-rater reliability of the system.
2.4 Reliability and Consistency

Kreitman (1961) stressed the importance of reliability, since no set of observations can
be adequately incorporated within a body of knowledge unless they meet the basic
requirements of reliability. In the case of psychiatric diagnosis, reliability can be defined
either as the amount of agreement between different observers when examining the same
subjects, or as the agreement between an initial and a subsequent diagnosis. Zubin
(1967) refers to the former as 'agreement' and the latter as 'consistency'. Hunt et ai.
(1953) looked at the consistency of diagnoses over time and found that the average
agreement between psychiatrists on major categories (psychosis, neurosis, etc.) was only
54%; and for specific diagnoses within categories, the average agreement was 33%. For
the neuroses the problem was even worse. Only 24% of the category diagnoses of
neurosis were in agreement, and for specific neurotic disorders only 12% agreement was
achieved. One of the largest studies of diagnostic reliability was conducted by Norris
(1959), who reported the results of an investigation of over 6000 patients. The reliability
of the category diagnoses for various disorders between occasions were: functional psychoses, 89%; neurosis, 46%; and character disorder, 34%. No figures were given for
agreement levels achieved in the specific diagnoses of neurotic disorders.
In an investigation ofthe amount of agreement between raters, Kreitman et ai. (1961)
reported 52% agreement for the general diagnosis of neurosis and 28% for specific
diagnoses. Similar results are presented by Ash (1949), Beck (1962) and Kaelbling and
Volpe (1963).
Zubin (1967) has reviewed the evidence for the reliability of diagnoses, and states that
the level of agreement for the neuroses is low and variable. For the diagnosis of an
unspecified neurosis, the agreement varies between 16% and 56%, for reactive depression between 18% and 63% and for anxiety states between 27% and 55%. Schmidt and
Fonda (1956) concluded that some diagnoses reached satisfactory levels of reliability; on
the diagnoses of schizophrenia, for instance, the agreement level was between 73% and
Validity
35
95%, but they point to the unreliability of diagnoses on the personality disorders and the
neuroses, where agreement between psychiatrists was almost completely absent. These
overall levels of agreement and consistency are far too low for individual diagnoses of
neurotic disorders to be accepted as reliable.
The findings of Kreitman et al. (1961) suggest that the fault lies with the system itself,
and not with the psychiatrists. In their study they compared the agreement between the
psychiatrists on duration of illness, family history, previous illness, symptomatology and
other variables. On none of these were there significant inter-rater differences, and the
authors conclude that the low reliabilities were not therefore due to differences in the
application of diagnoses between psychiatrists. Ward et a!. (1962) investigated the
sources of disagreement in diagnosis, and noted nine basic problems. The most profound
source of error was due to the nosological system itself (62.5%), though they also
attribute 32.5% of the variance to the inconstant behaviour of the diagnostician. Only
5% of the variance could be attributed to the inconstant behaviour of the patient. Much
of this unreliability is probably directly due to the confusion between descriptive and
explanatory statements that are contained in the diagnostic system.
Wing and his colleagues (1974) have taken these objections to the psychiatric system
of diagnosis more seriously than most, though they still believe that 'psychiatric disease
theories can be useful in everyday clinical practice'. The Present State Examination, a
structured interview, was devised by these workers as an attempt to improve the reliability and validity of diagnoses, and a similar effort has been made by Goldberg (1972) with
the General Health Questionnaire.
2.S Validity
The whole problem of validity in this context is extremely complex, since there is no
ultimate criterion against which the diagnoses or measures can be validated.
Everitt et al. (1971) reported a sophisticated attempt to validate the traditional
psychiatric system of classification using cluster analysis on data derived from 250 English and 250 American patients. A total of 728 items of information was available for
each patient. The authors found that three distinct categories emerged as relatively
distinct clusters. These were the manic phase of manic depression, paranoid schizophrenia and psychotic depression. A fourth category, schizophrenia, was also found to
present a fairly distinct cluster. But in the case of the neurotic disorders, there was a
striking failure to find any sort of cluster. This failure was particularly significant for
neurotic depression, since there were more than 60 patients with this diagnosis in the two
series. A similar failure to obtain any neurotic cluster has also been reported by Pilowsky
et a!. (1969).
However, other studies by Derogatis and his colleagues suggest that the conventional
descriptions of the neurotic disorders may have some validity (Derogatis et al. 1970,
1971 a, 1971 b; Williams et a!. 1968). These studies found a high coincidence between
five clinical clusters for the neurotic disorders and the results of factor analyses. Using
large numbers of neurotic out-patients, all of whom displayed anxiety, they found factors
of:
1) Somatisation. This was characterised by feelings of heaviness in the limbs, numbness
in parts of the body, muscular soreness, weakness and pains in the chest.
36
2) Obsessive-Compulsive-Phobic. The items that showed strong positive loadings for

this factor were having to check and re-check actions, difficulty in remembering
things, difficulty in making decisions, difficulty in concentrating and worries about
carelessness.
3) Irritability-Over-sensitivity. This factor loaded on items such as feeling easily annoyed
by things and critical of others, outbursts of temper, feelings easily hurt and feelings of
inferiority.
4) Depression. This was characterised by thoughts of ending one's life, loss of sexual
interest and pleasure, general depressive mood and hopelessness about the future.
5) Anxiety. The anxiety items included feeling fearful, sensations of nervousness or
shakiness, suddenly feeling frightened for no obvious reason, sensations of one's heart
pounding and feeling tense or keyed-up.
Although these findings cannot be freely generalised beyond anxious neurotic patients,
the results provide some support for the validity of the descriptive classification of the
neurotic disorders. Derogatis et al. (1970, 1971a, 1971b) suggest that the factors may
represent primary or 'core' dimensions of the neurotic disorders.
These results and those of Everitt et al. (1971) suggest that the neurotic disorders
appear to merge into one another, and that no independent types of neurotic patients can
be distinguished. However, there is some evidence to support the validity of the neurotic
disorders as describing separate dimensions which underlie the symptoms and manifestations of neurotic psychopathology.
The evidence that has been presented suggests that the traditional system of classification is extremely unreliable and inconsistent in its application to the neurotic disorders.
This is probably due to the confounding of the descriptive and explanatory functions of
the diagnoses themselves; and it has been suggested that purely descriptive statements of
the nature of the neurotic disorders would at least have the advantage of increasing the
level of agreement upon what is to constitute the explicandum. The studies of Derogatis
and his colleagues provide some support for the validity of the traditional clinical descriptions of the neuroses. A further problem with the current psychiatric system is its use of a
categorical system to describe the neurotic disorders. Hine and Williams (1975) reach the
same conclusion about the inadequacy of this system of classification and propose that it
should be replaced by a dimensional system. As Hempel (1961) has pointed out, most
sciences start out with a typology based upon dichotomies, but frequently move on to
continuous, dimensional systems.
2.6 Social Context

Another reason for rejecting the categorical, disease-entity notion is that the social
context within which both patient and clinician operate has a profound influence upon
the diagnostic process and upon the neurotic disorder itself. Sandifer et al. (1969) presented a description of the psychopathology of a group of patients to 41 psychiatrists in
England and America. In their evaluation of the clinical material, the most striking
finding was that the psychiatrists of the United States reported almost twice as many
symptoms as those of the United Kingdom.
Meehl (1973), who is committed both to the notion of disease entities in functional
psychiatry and to the Kraeplinian nomenclature as the best system of classification, has
Social Context
37
implied that diagnoses are unreliable partly because of the 'careless and unskilled'
approach of some psychiatrists. The same sort of argument in defence of psychiatric
diagnosis has been used by Clare (1976) in reply to the findings presented by Rosenhan.
In his study, Rosenhan (1973) poses the basic question, 'How do we recognise
insanity?'
Do the salient characteristics that lead to diagnoses reside in the patients themselves or in the
environments and contexts in which observers find them? From Bleuler, through Kretchmer,
through the formulaters of the recently revised Diagnostic and Statistical Manual of the American
Psychiatric Association the belief has been strong that patients present symptoms, that the symptoms can be categorised, and, implicitly, that the sane are distinguishable from the insane.
Eight 'norma!' volunteers presented at 12 different psychiatric hospitals, complaining

that they had been hearing voices which spoke meaningless words such as 'hollow' and
'empty'. These symptoms were chosen because of their similarity to existential symptoms
and because of the absence of any reported existential psychosis in the literature. Apart
from this single symptom, and the use of a false name and occupation, all other characteristics were presented truthfully from the person's life history, and immediately upon
admission to a psychiatric ward, the pseudo-patient ceased to simulate any symptom of
abnormality. All but one of the individuals was admitted with a diagnosis of schizophrenia, and discharged as schizophrenic 'in remission' (that is, not as 'cured' or 'norma!'); the one patient not diagnosed as schizophrenic was diagnosed as manic depressive.
Despite the fact that 35 out of 118 patients voiced their suspicions that the pseudopatients were only faking, no psychiatrist or any member of staff detected this.
Both Clare (1976) and Wing (1973) take the position that this failure reflects the illinformed and slipshod approach of the diagnosticians, rather than any weakness inherent
in the system itself. It is interesting that this was the same excuse offered by a research
and teaching hospital in Rosenhan's report. Subsequently, this institution was informed
that one or more pseudo-patients would attempt to gain admission to the hospital over
the following 3 months. Since a certain amount of prestige and professional acumen were
at stake, the psychiatrists and staff involved could be expected to be more alert to the
subtleties of symptomatology. And indeed they were. They were confidentthat they had
detected 41 pseudo-patients (19 of whom were detected both by a psychiatrist and one
other staff member). Unfortunately, no pseUdo-patient presented himself to the hospital
during this period.
The full implications of this disturbing study are far from clear. The results do, however, point to the way that the meaning of behaviour depends upon the context in which
it occurs. It should, by now, be common knowledge that perceptions are not passive,
accurate representations of the real world, but are highly selective. The Gestalt school
showed how important the context was in its effect upon perception, and Sherif (1936)
described how our experience is organised around, and modified by, the social circumstances within which it occurs. Other studies have pointed to the distortions in person
perception that are dependent upon the psychological state of the perceiver. Kelley
(1950), for instance, showed how sensitive person perception can be to subtle verbal cues
which are interpreted in terms of the perceiver's implicit personality theory. The meaning
of the social interaction and the roles of the individuals involved are also potent influences upon the final perception (Kelley 1948), a finding which has special relevance for
the diagnostician. Katz et al. (1969) locked at the diagnostic disagreement among clini-
38
cians and suggested that this may be due in part to differences in perception between the
observers.
There can be little doubt that the current psychiatric classification system is inadequate. In its defence, Foulds (1955) has argued that its inadequacy does not prove its
unsuitability, and that in any case, it should not be replaced until a better substitute is
available. This chapter presents the case that a system which was based upon overt
symptomatology would have several advantages over the confused psychiatric nosology.
In particular, it would improve the chances of agreement between clinicians on the nature
of the neurotic disorders, and it would avoid the inappropriate mixture of description and
theory that has bedevilled most of the previous work in this area (see also Essen-Moller
1967; and Costello 1970).
3. Neurosis, Normality and Psychosis
3.1 Normality and Abnormality

There has been little explicit discussion of what constitutes mental health, and although it
is impossible to practise psychiatry or clinical psychology without some (usually implicit
and unstated) concept or assumptions about normality, few clinicians seem to have had
either the time or the inclination to give serious consideration to the concept of normality
that underlies their work. In order to recognise abnormality we must have some notion of
normality. The concept of normality is what philosophers would call parasitic; that is, it is
dependent upon the existence of some distinction between normality and abnormality.
But any attempt to specify what is meant by these terms leads immediately to considerable difficulties, and Aubrey Lewis has described the concept of mental health as 'invinciblyobscure' (Lewis 1958).
In the first place, the problem of psychological health or normality is closely related to
that of the classification of mental disorders. The traditional medical psychiatric
approach usually shows little interest in the question of normality per se and focuses
instead upon the specific disease entities (Barton 1959). Mental health is defined in terms
of the absence of any clear psychopathology. The influence of this tradition can be seen in
the omission of any discussion of the problem of normality from most psychiatric text
books (e. g. Slater and Roth 1969; Henderson and Batchelor 1962), though Arieti's
Handbook of American Psychiatry includes a chapter by Offer and Sabshin (1974). The
largely implicit psychiatric concept of normality appears to regard health as an almost
universal phenomenon, and one which is characterised by reasonable, rather than ideal,
levels of functioning. It is not clear whether the neurotic pathology that differentiates
health from illness is the overt symptomatology or some underlying personality factor. As
has been described, the diagnostic manuals tend to confuse elements of both in their
description of the neuroses, and this further compounds the problem, since the criteria
for recognising behavioural symptomatology are very different from those used to identify features of personality. Martin Roth doubts whether the concept of disease can be
defined other than in the broadest of terms in either psychiatry or in medicine. This
leaves an indistinct line of demarcation between health and disease, but it may still be
preferable to the spurious precision to be gained by definition in terms of lesions (Roth
1963). In the absence of any clear discussion of this problem from the medical-psychiatric perspective, the precise relationship between neurosis and health remains unclear.
To the psycho-analyst, freedom from symptoms is not an acceptable criterion of mental health, and psycho-analysts would also reject any definition based upon 'absence of
pathology'. Plaut (1960) has even suggested that under some circumstances an absence
of symptoms may indicate 'sterility and stagnation'. In psycho-analysis, 'health' and
40
Neurosis, Nonnality and Psychosis
'sickness' are used as attributes of individuals and not of behaviours (Hartmann 1960).
Psychologists and psychiatrists with interests in learning theory and behaviour modification, on the other hand, have tended to restrict the use of such terms as 'normal',
'healthy' and 'abnormal' to behaviour or to specific events.
The psycho-analysts, and in particular the Freudian psycho-analysts, have used a concept of normality which is based upon the notion of an unattainable optimal or ideal level
of functioning. Offer and Sabshin (1967) traced this position back to Sigmund Freud,
whose theory assumed the universality of unconscious conflicts, the Oedipus complex,
childhood repression and so forth (Freud 1900, 1905). Freud stated that everyone is
to some extent neurotic and that a normal ego, like normality in general, is an ideal fiction (Freud 1901, 1937). Levine (1942) has also suggested that normality is 'non-existent in a complete form', and can only be found 'as (a) relative and quantitative approximation'.
For the psycho-analyst, of course, the ideal fiction of normality refers to personality
and not to the subjective feelings of the patient or to the absence of overt symptoms.
Melanie Klein (1960) has provided one of the more detailed psycho-analytic descriptions
of normality, which she sees in terms of personality integration. Complete integration is
again seen to be an unattainable ideal. The important features of normality (in Kleinian
terms) are emotional maturity, the capacity to deal with conflicting emotions, and
strength of character, each of which derives from infantile experiences of gratification
and frustration, and the capacity to resolve the conflicts between one's own emotions,
and between the intrapsychic and external realities. Ernest Jones (1950) has been one of
the few psycho-analysts to emphasise the importance of social adjustment in any definition of normality, and he includes among his criteria of normality the personal satisfaction of the individual and his effectiveness in social and psychological functioning. The
behaviour of the normal person should be both personally satisfying and socially appropriate.
The influence of the psycho-analytic concept of 'normality as ideal' extends beyond
psycho-analysis into psychiatry and psychology. Maslow's notions about self-actualisation rely in large measure upon Freudian ideas of normality. Maslow and Mittelmann
(1951) suggest that psychological normality consists of adequate self-knowledge, integrated personality, ability to learn from experience and other similar criteria. The selfactualised or ideally normal person emerges as spontaneous, creative, with a sense of
humour and aware of, and accepting, both his own needs and those of others. J ahoda
(1958) reviewed the literature on mental health and provided a similar description of the
constituents of positive mental health. Like Maslow and Klein, Jahoda emphasises the
independence, growth and integration of the individual as well as the importance of selfattitudes. But despite the superficial plausibility of this approach to normality as ideal,
there are many problems involved in such definitions. In particular the specification of
what is to count as 'adequate', 'integrated', 'mature' or 'creative' depends upon culturebound and value-laden judgments, and these difficulties have not been discussed satisfactorily by the proponents of this perspective. Lewis (1958) referred dismissively to such
definitions as a 'clutter of words'. Nor do these lists of the supposed characteristics of
normality match the subtle ways in which the concept is actually used.
Unlike the medical perspective, which is usually based upon a categorical system, this
concept of 'normality as ideal' relies upon a straight-line continuum between normality
and abnormality. Menninger, Mayman and Pruyser (1963) proposed that specific classifi-
41
cations of psychological disorders should be abandoned in favour of a unidimensional

system in which the different manifestations of mental illness were to be seen as sequential stages of a single process. This, however, is a rather extreme position. Goldberg
(1972) suggested that the distribution of psychiatric symptoms in the general population
does not correspond to any dichotomy between 'cases' and 'normals', and that nonpsychotic disorders are best seen as being evenly distributed throughout the population.
This single dimension varies between 'severe disorder to a hypothetical normality, with
many steps between these end-points'.
A more statistically sophisticated notion of normality uses the normal distribution
curve as its basis. In this model, 'normal' refers to the middle range of the curve, and both
extremes represent the statistically abnormal; that is, they are abnormal because of their
infrequency. Offer and Sabshin (1966) discussed this perspective and noted that there
are thousands of biochemical tests which are based upon this principle. The statistical
perspective is also widely used in the field of psychometrics.
It is one of the more important implications of dimensional approaches to neurosis that
no clear dividing line can be drawn between states of mental health and those of neurosis.
Many clinical psychologists have been at pains to stress that in its development, its
maintenance and in the ways in which it may be altered, abnormal behaviour is no
different from normal behaviour. There is no intrinsic difference between the two. The
difference is to be found in the reactions of others (Ullmann and Krasner 1967). Accordingly, it is more productive to examine the ways in which a person develops any belief or
behaviour pattern rather than how he develops a false belief or a maladaptive pattern of
behaviour, since the principles underlying both are the same. Ullmann and Krasner's
Psychological Approach to Abnormal Behaviour (1969) expressed this position which
rejects the distinction between normality and abnormality as categories. Indeed, Ullmann
and Krasner suggest that the failure to arrive at any single clear definition of abnormality
is one of the strongest reasons for attempting to discuss in terms of normality those
behaviours that have traditionally been accepted as the subject matter of abnormal
psychology. To this extent, the dimensional perspective is often used in a descriptive way
as an attempt to avoid the use of such evaluative terms as 'normality'.
Among the strongest advocates of the dimensional approach has been Hans Eysenck
(1947 a, 1961, 1967). Instead of using a variable number of disease entities which differ
from psychiatrist to psychiatrist, Eysenck proposed the use of three dimensions (introversion/extraversion, neuroticism and psychoticism), along which any person may be ranged
and given a numerical score. The neuroticism and psychoticism factors (N and P) are the
two main independent factors in psychological medicine. Both define continua which
range between extreme disorder and normality, and there are no breaks or qualitative
differences which make it possible to classify people into separate groups (Eysenck
1961). The psychiatric diagnoses appear not as categories but as points of intersection
within the dimensional framework; a diagnosis represents a specific region in multifactor
space (Eysenck 1970a). Patients who also fit into that region would have certain
resemblances because of their similar position on the relevant dimensions, but these
would shade imperceptibly into other diagnoses.
Another personality theorist who has used this dimensional approach is Cattell. Unlike
Eysenck, who proposes a single neuroticism factor, Cattell's interpretation of neurosis is
multifactorial. Cattell suggests that this approach has its parallels in physical medicine
which is in transition from single-cause thinking to a multifactorial view of disease as a
42
dysfunction of the whole organism (Cattell and Scheier 1961). There is an equal emphasis upon the need to relate this quantative approach to personality description to a
quantitative description of environmental factors, and Cattell (1957) has suggested a
mathematical model for the typing and measurement of environmental events.
A purely statistical interpretation of normality still fails to avoid the pitfalls of valuejudgements, which seem to obtrude themselves into any classificatory system. A dimensional system which relies upon the normal distribution of scores faces problems of
evaluation concerned with the tails of the curve. It is difficult to think of genius as being
abnormal in the same way as idiocy, nor do we think of the extremely well-adjusted and
emotionally stable in the same terms as the maladjusted and disturbed person. At the
same time this model is not entirely satisfactory in its equation of normality with the
middle range of the distribution. This tends to regard conformity to group norms as
normality.
On the other hand, a few psychiatrists and psycho-analysts have taken a hard-line,
absolutist view of normality. Money-Kyrle (1957), for instance, dismisses the problems
of value-judgements and cultural relativity and lays stress upon insight as the crucial
factor. This sort of view, however, has few supporters and most writers have been
troubled by this problem of value-judgment. Jahoda (1958) has been one of the few
psychologists to face up to the problem of values, and she explicitly identified some of the
culture-bound values that have influenced concepts of normality (such as the positive
evaluation of personal independence and autonomy). Ultimately it seems unlikely that
any adequate definition of normality will be achieved which completely avoids this particular difficulty, and the best that can be hoped for is that the value judgements underlying clinical concepts of normality should be explicitly identified.
Goldberg (1972) suggested that a quantitative dimensional measure of psychiatric
disturbance provides a probability estimate of that person's being a psychiatric case.
Goldberg is caught in the dilemma of wishing to use a dimensional as well as a categorical
approach; instead of using a simple typology, he argued that some threshold point on the
continuum could be chosen to differentiate 'cases' from 'normals'. This faces the same
theoretical difficulties as a typological model of abnormality. There can be no clear
dividing line. Equally, the assumption that the severity of a psychological disorder can be
defined entirely in terms of the phenomena of psychiatric illness faces a number of
problems. Minor psychological symptoms are widely distributed in the general population, but the meaning of those symptoms is dependent upon how the individual sees,
evaluates and acts upon them, how others act upon them, and in what sort of social
situation they occur. It is probably true to say that these aspects of abnormal psychology
have been better appreciated by social psychologists and sociologists than by clinical
psychologists and psychiatrists.
In one of the classic experimental studies of the effects of cognition upon emotion,
Schachter and Singer (1962) demonstrated that emotional states are the result of an
interaction between cognitive and physiological factors. Some subjects were told that
they were to be given an injection of a vitamin but were actually given adrenalin, a drug
which causes autonomic arousal, increased heart rate, sweating, etc. Having received no
explanation for these effects, they provided their own interpretation of their experience,
based upon the circumstances in which they found themselves. When the subject was in
the company of an angry person they themselves felt anger, and when exposed to
euphoric behaviour they felt happy. Those subjects who were correctly informed of the
43
effects of the drug showed least emotional response to the behaviour of the experimental
confederate, since they already had an appropriate explanation of their state of arousal.
Schachter and Singer argue that the physiological arousal itself is indeterminate, and that
the cognition channels arousal into specific emotional states. Similar findings have also
been reported by Schachter and Wheeler (1962), and Valins (1966) has shown that the
subject's perception of his own state of arousal need not be correct in order to affect the
resulting emotion. After subjects had received misleading feedback of their own heart
rate when viewing certain photographs of nude females, they described a preference for
those photographs over others. Conversely, the individual's emotionality can also be
decreased by misattributing his naturally occurring emotions to a non-emotional source
(Nisbett and Schachter 1966).
This recent emphasis upon the individual's beliefs about his own behaviour has profound implications for the concepts of health and mental illness. It is necessary, for
instance, to re-interpret any division between normality and neurosis in the light of the
individual's cognitions, since self-attributions affect a whole range of emotional, attitudinal and behavioural responses. The perception of oneself as 'neurotic' carries implications quite distinct from the 'normal' self-perception, and to that extent it is possible to
suggest a discontinuity between normality and neurosis - though not in the way proposed
by the medical model.
The attribution theorists emphasise self-perceptions. Sociologists tend to place greater
emphasis upon the significance of the perceptions of others. Parsons (1952, 1964)
defines both health and illness in social terms. Health is defined as the optimum capacity
of a person for the effective performance of the roles and tasks for which he has been
socialised, and illness is also seen as a socially institutionalised role. Both are defined in
terms of the individual's participation within a social system. Parsons has pointed out that
it is not possible to regard all illnesses as 'natural phenomena' which merely happen to
people, though there are still some who attempt to reduce all illnesses to the physiological
and biological level and to look for therapies at that level. Parsons offered a statement of
how illness constitutes a social role rather than merely a 'condition', and his formulation
of the sick-role has been widely used by medical sociologists. There is also an increasing
awareness of the value of this concept among clinicians. Parsons discussed the social
implications of the sick-role, and in particular the exemptions from social responsibility
that are conferred upon the sick. These privileges and exemptions of the sick-role may
themselves become objects of secondary gain which the patient learns to value. However,
it is one condition of the legitimisation of illness as deviance that the sick individual
recognises the inherent undesirability of his sickness. To the extent that the neurotic
individual is seen by others to derive gratification from his position (as is also the case
with alcoholics and drug addicts) there exists a considerable tension between accepting or
rejecting the validity of the sick-role.
One of the most widely used sociological explanations of how individuals come to
occupy the sick-role is the societal reaction theory, or labelling theory. Labelling theory
makes a basic distinction between primary and secondary deviance. Primary deviance,
the fact of which causes a person to be labelled as deviant in the first place, has been said
to have at best only marginal implications for the psychic structure of the individual
(Lemert 1967); and secondary deviance consists of the ways in which the individual
adapts to the problems created by how others react to his primary deviance. Erikson has
emphasised that deviance is not to be seen as a property inherent in certain forms of
44
behaviour, but rather that it represents a property which is conferred upon those fonns
by society (Erikson 1957).
Societal reaction theorists are therefore primarily interested in the social responses of
others, arguing that the self-perceptions of the deviant individual depend upon the image
of themselves that they receive through the actions of others. Scheff (1966), who applied
labelling theory to mental illness, uses this concept to refer to the occupancy of a social
role and not to a state of personal distress, behavioural disorganisation or personality
disorder. Mental illness is to be seen as an ascribed status dependent upon factors
external to the individual.
The principal weaknesses of the societal reactance theory are that it has failed to
consider how neuroses (and mental illnesses in general) develop, and it has exaggerated
the amount of secondary deviance produced by treating someone who is mentally ill.
Gove (1970) examined the assumptions of labelling theory and found that a person's
behaviour determines the expectations of others to a greater extent than the reverse, and
that a substantial majority of hospitalised individuals present a considerable psychiatric
disturbance quite apart from any secondary deviance associated with their sick-role. Cole
and Lejeune (1972) also suggest that individuals who are unable to function adequately
in a social setting are likely to be motivated to define themselves as sick in order to
legitimise their self-defined failure.
This aspect of social adjustment has important implications for the clinical psychologist, and it has been regarded as a crucial feature of psychological health. Some clinicians have used social training programmes as a technique of therapy. Libennan
and Raskin (1971), for instance, were able successfully to treat a case of depression
through the modification of social behaviour, and Phillips and Zigler (1961) related social competence to the nature of the psychiatric difficulties experienced by different
patients. Similar findings were presented by Libet and Lewinsohn (1973), who found
that depressed patients were less socially skilled than other non-depressed individuals,
and social-skill deficits can be seen as a major feature of many psychiatric disorders
(Hersen and Eisler 1976). Gove (1970) accepts that the expectations and social responses of others may have an important effect, but denies that labelling theory offers
a sufficient explanation, because of its neglect of the individual's own attitudes and behaviour.
Many individuals who suffer from neurotic problems remain in the 'normal' population
and avoid contact with psychiatrists. To that extent they also avoid the stigma of the
'neurotic' label. It is, however, extremely difficult to make any assessment of the prevalence rates of neurosis in the normal population, since the inclusion of any particular case
within the 'neurotic' group will reflect as much about the criteria of the investigator as
about the difficulties of the person involved.
Essen-Moller (1956) interviewed 2500 inhabitants of a rural area of Sweden and
classified the population on a continuum ranging from definite mental illness through
personality abnonnality to nonnality. His estimate of life-time prevalence of various
disorders was 1.7% for psychosis and 5.2% for neurosis, with about 35% in the 'normal'
range. In a later study of the same population, Hagnell (1959, 1968) used a more
extensive interview with more overtly psychiatric content and suggested that the figures
were 1.7% for psychoses and 13.1 % for neuroses. It is interesting that whereas the
psychotic disorders were unaffected by the changes in criteria of abnonnality, the estimate of the prevalence of neuroses was more than doubled.
Nonnality and Abnonnality
45
Anthony Ryle (1967) conducted a survey of the prevalence rates of neurosis and
emotional disturbance among a group of 112 psychiatrically unselected London families.
A large proportion of both the adults and children in this group were found to show
'pathological' symptoms, and Ryle argues that those individuals who seek professional
help and advice for their neurotic difficulties represent only a fraction of those with such
problems. Unfortunately Ryle fails to make clear what were his notions of neurosis,
health and normality in this study, and there are many pitfalls involved in the use of
retrospective analysis of general practitioners' records for such information. Ryle does
use scores from the Cornell Medical Index as a more objective form of data, but here
again, Ryle denies that any useful distinction can be made between neurosis and neuroticism (i. e. between neurotic behaviour and personality predisposition). This may well be
a hazardous assumption to make. Whether or not the person who scores highly on
measures of neuroticism develops a neurotic disorder will depend upon more than his
predisposition. Clearly the degree of environmental stress to which he is exposed will
playa crucial role.
These weaknesses raise considerable doubts about Ryle's specific findings, though in
his general conclusion he may be correct. Another investigation of 46 medical practices
in London, for instance, also suggested that neurotic disorders may constitute a sizeable
problem for the general practitioner, and that as many as 14% of his consultations may
be concerned with such problems (Shepherd et al. 1966; Shepherd 1973). But although I
would not wish to deny the general suggestion in these studies that neurotic difficulties
are fairly common in the general population, I would like to draw attention to the
theoretical nature of the question. The attempt to reduce this issue of prevalence rates of
health and neurosis to an empirical question is misguided.
In their investigation of London medical students, Lucas et al. (1965) fall into this sort
of error. They conclude that about one-third of their sample showed psychological symptoms such as anxiety, tension and poor concentration. However, these psychological
'symptoms' were generally directly related to some specific situational stress, such as
examinations. To focus upon symptomatology to the exclusion of situational factors in
this way shows a fundamental misunderstanding of the nature of neurotic disorders. If it
is neurotic to respond to stress with anxiety, then we would all be neurotic, and it is
probable that such misunderstandings have led to conclusions such as 'anxiety neurosis is
the most frequent disorder of civilised life' (Weiss 1942). It remains one of the unfortunate consequences of the medical perspective that neurotic disorders are seen in terms of
identifiable illness to the exclusion of social and cognitive factors, and Shey (1971) has
drawn attention to iatrogenic anxiety, a problem that has been largely ignored by physicians because of their 'somatic' orientation.
The idea that neurotic disorders can be seen in terms of their objective severity is
misconceived; indeed, it is not clear what we could mean by such a notion as objective
severity in this instance. The 'severity' will depend upon the personal distress that the
individual suffers as a result of his neurotic behaviour: and the distress itself will in turn
depend both upon how the person regards his own behaviour and on how others see his
behaviour. If he comes to see his anxiety as a 'normal' response to a stressful situation, he
may well be able to cope with his anxiety and the situation quite adequately. If, on the
other hand, he or others decide to label the anxiety as 'neurotic' or as 'sick', then quite
different consequences will follow. He may firstly approach a doctor, in which case the
professional and personal criteria that the doctor uses to differentiate normality from
46
neurosis will detennine any reaction to the patient. If the doctor does regard the patient
as having a neurotic disorder, then the person may fall into the sick-role that has been
described by such sociologists as Parsons.
3.2 Neurosis and Psychosis

There are two fundamental issues to be dealt with here. Firstly, whether it is useful to
maintain any distinction between neurosis and psychosis; and if it is, whether they are to
be represented by different points on the same continuum, or in tenns of separate and
independent dimensions. Martin Roth (1963) has described how neurosis and psychosis
are usually differentiated in tenns of a number of characteristics, none of which is on its
own sufficient to produce a reliable and valid distinction. In neurosis, the disorder is
generally restricted in its effect upon the individual's functioning; the person retains
contact with reality and a certain amount of insight into his difficulties. When used to
refer to personality, the tenn is again used to suggest partial dysfunction in which some
facets of personality are unaffected. In addition, the neurotic behaviour can often be seen
as an exaggeration of ordinary emotional reactivity. Psychoses, on the other hand, have a
more global effect upon the individual's personality. All, or most, aspects of behaviour
are affected, and there is usually a dramatic reduction in the person's contact with reality.
Taylor (1966) expressed this in terms of insight - that psychotic patients lack insight into
the truth or falsity of their idiosyncratic beliefs, whereas neurotics generally do have
some insight.
Freud distinguished between neurosis and psychosis in the following way: 'Neurosis is
the result of a conflict between the ego and its id, whereas psychosis is the analogous
outcome of a similar disturbance in the relations between the ego and the external world'
(Neurosis and Psychosis). This brief statement provides an outline of the difference in the
genesis of the two disorders. However, Freud insisted that the aetiology of a neurosis and
of a psychosis is always the same. It consists in a frustration of primitive childhood wishes,
though it depends upon the nature of the intrapsychic solution as to whether the resultant
disorder proves to be neurotic or psychotic. If in resolving the conflict the ego retains its
dependence on external reality and attempts to overcome the demands of the id, the
outcome is neurosis: on the other hand, if the ego is itself overcome by the demands of
the id, it loses contact with reality with the consequence that the person becomes
psychotic.
Freud repeatedly emphasised the idea that those trends and conflicts that he described
were not to be seen as 'causes' of psychological disorders. These were a result of the
quantitative distribution of energies within the psychic system (Munroe 1957), and the
essential feature of psychotic conditions was the greater depth of the regression. It is this
central concept of regression which plays such an important part in Freud's theory of the
neuroses and psychoses. The type of disorder is dependent upon the organism's reaction
to conflict by regression. In the case of the psychotic disorders, the regression is more
severe, though this distinction between neurosis and psychosis is often blurred, since
neurotics may often show breaks with reality. In its simplest fonn, the Freudian view is
that neurosis can be conceived as an intermediate stage along a single dimension, with
nonnality at one end and psychosis at the other. 'Neuroses and psychoses are not separated by a hard and fast line, any more than health and neurosis' (1923, vol XIX, p. 204).
47
Wexler (1971), however, has suggested that this is representative only of Freud's
earlier position (e. g. Neuropsychoses of Defence 1894), and that Freud later adopted the
view that schizophrenia was to be sharply contrasted with neurosis. In his article, however, Wexler does not make it clear whether he is claiming that Freud rejected the
unidimensional model, or simply that he claimed that the disorders were a result of
different processes. The latter point is certainly true (see Neurosis and Psychosis), but
this has little bearing on the unidimensional model.
In an empirical study of 100 schizophrenic, neurotic and normal subjects, Bellak et al.
(1973) looked at the adequacy of 'ego function' in these three groups. Scores which were
based upon judgements of the person's characteristic level of functioning were presented
for reality testing, sense of reality, defensive functioning, autonomous functioning, etc. In
all, scores were presented for 11 ego functions, and in each case the psychotics were
significantly more disturbed than neurotics, and the neurotics more than normals. The
authors examined some of the ways in which their data might be distorted. There was a
slight tendency, for instance, for the ego-function ratings to correlate positively with
social class, educational level and IQ and negatively with age. The objection might be
raised that the concept of ego function is itself sufficiently vague to reflect more about the
rater's own notions of psychopathology than about the subject's behaviour. However,
this evidence if taken at face value is clearly supportive of the psycho-analytic position.
The mean values for the ego function scores are presented in Fig. 2. This illustrates the
single dimension and the overlap of the three conditions that is generally assumed by
psycho-analysts.
A number of psychiatrists have also taken this unidimensional position, though without
the psycho-analytic assumptions about psychosexual regression. Myerson (1936)
restated the case that neuroses span the bridge between normal mental states and certain
psychotic states, and this unidimensional perspective has prompted most psychiatric
interest in the area of depression. It has been suggested that because of the importance of
depressive disorders in psychological medicine, any failure to demonstrate two indepen-
normalO
neurotic ll.
schizophrenicO
(mean scores)
Normal
Neurotic
Schizophren i c
b
Fig. 2. The undimensional model
Ego - function scores

(range of scores)
48
Neurosis, Nonnality and Psychosis
dent factors for the depressive conditions would severely limit the value of the separate
concepts of neuroticism and psychoticism (Trouton and Maxwell 1956).
In the case of the depressive disorders, most psycho-analysts would accept the
unidimensional model. Fenichel (1945) wrote that the difference between neurotic and
psychotic depression is one of the depth of narcissistic regression. Mendelson (1974)
regards this as a fair summary of the views of both Freud and Rado. This sort of
unidimensional point of view has also received considerable support from such influential
psychiatrists as Lewis and Mapother in their discussions of neurotic depression.
Mapother (1925) denied that there was any fundamental difference between the nature
of neurotic and psychotic depressions. Lewis (1934) has also made this point about the
neurotic and psychotic types of depression (which he suggests are not to be regarded as
synonymous with reactive and endogenous depression). Lewis was impressed by the
difficulty of making any qualitative distinctions. The patient who appeared to show
reactive depression to a bereavement might have proved to have had a mood change
beforehand, and the patient overwhelmed by guilt and self-reproach at one stage could
later become angry and hostile. In an analysis of 70 cases admitted to a military hospital,
Tredgold (1941) also failed to differentiate two groups of depressives, since many
showed typically neurotic and typically psychotic clinical pictures at different stages of
their illness.
In a study of over 1000 unselected admissions to the Maudsley Hospital, Kendell
(1975) collected 60 different items of information relative to the distinction between
neurotic and psychotic depression for each patient and subjected this data to a discriminant function analysis. The distribution of scores was unimodal rather than bimodal, and
Kendell concluded that the depressive disorders are best regarded in terms of a single
continuum between the neurotic and psychotic states. Kendell acknowledged that the
neurotic and endogenous states of depression differ, but like Aubrey Lewis he regarded
the majority of cases as falling between the two extremes. When further analysis failed to
show any discrete cluster for neurotic depression (Everitt et al. 1971), the authors concluded that psychotic depression is a more firmly based concept than neurotic depression.
After earlier assertions of the validity of the distinction between neurotic and endogenous depressive states (Kiloh and Garside 1965), Kiloh appears to have changed his
opinion on this matter. On the basis of factor analytic results, Kiloh et al. (1972) have
argued that neurotic depression is a vague concept which 'as a distinct syndrome is an
artefact'. It is not clear whether these comments suggest that Kendell, Kiloh and their
colleagues are moving towards a position denying the validity of neurotic depression as a
disorder altogethe-r. Certainly Kiloh et al. argued that the diagnosis of neurotic depression is often made by excluding endogenous depression and that the disorder is likely to
be vague and heterogeneous.
Another study which provided support for the continuity position was made by Garmany (1958) using 525 depressed patients. He found that situational stress factors of
various kinds could be identified among 95% of the reactive depression and among 79%
of the endogenous cases. Similarly, constitutional predisposing factors could be found
among 70% of endogenous depressions and among 55% of the reactive cases. Garmany
suggests that these differences are not sufficient to justify any clear differentiation of the
two types and that the distinction is really one of degree, that is, between mild and severe
depressions.
49

Table 2. Different approaches to depression
One sort of depression

Two sorts of depression
Categorical model
Dimensional model
Mapother, Lewis
Kiloh et al.
Freud, Kendall
Eysenck
There are, however, two issues involved in this problem, and much confusion has
resulted from a failure to identify them. There is the problem of whether or not neurotic/
reactive and psychotic/endogenous depression should be conceptually separated, and of
the relative merits of categorical and dimensional frameworks. Four possible positions
can be adopted in the dispute, and these are shown in Table 2. The traditional psychiatric
position regards neurotic and psychotic depression as two distinct entities (see Kiloh and
his colleagues), and it was against this view that Mapother and Lewis first reacted. Freud
and Kendell differ from this view in their use of a dimensional framework. The twodimensional system is best exemplified in the work of Eysenck (1970), who argued that
Kendell's unidimensional factor solution is inadequate and severely limited as a
framework within which to classify the depressive disorders.
The unimodal hypothesis has also been strongly challenged by Carney et al. (1965),
who used similar methods to Kendell and reached a quite different conclusion. The
results of their factor analysis show that a bimodal factor accounted for a greater part of
the variance than a general factor, and that the various statistical analyses broadly
matched the clinical descriptions of neurotic and psychotic depression. Their results
could not be accounted for in terms of a single depressive disorder, though the factor
loadings suggested that the two disorders did have some features in common. This
suggests, therefore, that there are quantitative as well as qualitative differences between
the depressive conditions, but that the two dimensions are best regarded as independent
of each other.
Burt (1954) described factor analysis as the best statistical device for resolving the
problems of classification, and several psychologists have relied heavily upon factor
analytic methods in their work. Eysenck and Eysenck (1975) see factor analysis as a
necessary, but not a sufficient, method for isolating the main dimensions of personality.
The main weakness of this method is its dependence upon correlation, and Eysenck has
pointed to the need for personality research to be firmly related to the main body of
experimental psychology. Eysenck (1955) firmly rejects the discontinuity model of
psychological disorders: neurosis is to be seen as the extreme end of a normal-neurotic
continuum.
In order to test the one- and two-dimensional hypotheses about the relationship between neurosis and psychosis, Eysenck (1955) gave four tests (of visual acuity, object
recognition, mental speed and visual accommodation) to three groups of normal, neurotic and psychotic subjects. Eysenck argued that these tests had the advantage of being
based upon objective rather than interpretative data (their relevance to abnormal
psychology is discussed in the paper itself). Using a discriminant function analysis the
results supported the view that neurotic and psychotic disorders lie along different and
independent dimensions. Loevinger (1955) criticised Eysenck's conclusion that this evidence is contrary to the Freudian unidimensional model on the grounds that the measures chosen have no direct relevance to the particular dimension proposed by Freud.
50
S. B. G. Eysenck has discussed this issue in the context of 'mixed states', in which it is
not clear whether the patient has a neurotic or a psychotic disorder. The clinical textbook definitions tend to focus upon the presence of hallucinations, delusions and cognitive deterioration as evidence of psychosis. Unfortunately, such typical syndromes are
rarely encountered in practice. The unidimensional view that neurotic and psychotic
disorders differ only in terms of their severity suggests that the difficulty in such cases is
due to the patient's presenting 'borderline' symptoms. That is, the severity of the
patient's disorder is placed on the continuum between two clear-cut disorders, and therefore includes features of both. As the severity of the disorder increases, it is likely to show
the psychotic features more clearly until these have displaced the neurotic symptomatology. If the disorder becomes less severe, the neurotic symptoms will predominate as the
patient moves back towards normality. The two-dimensional perspective, on the other
hand, accepts the independence of neuroses and psychoses. It is, therefore, possible that
the 'mixed-state' cases are simultaneously suffering from both neurotic and psychotic
disorders, and that each may vary in severity independently of the other.
In her study (S. B. G. Eysenck 1956) three groups of normal, neurotic and psychotic
subjects completed a number of psychological tests. On several of the tests the results of
the simple analyses of variance appear to support the unidimensional model; the normals
performed best, the psychotics, worst, and the neurotics were in between. On other tests,
however, the neurotics were the most disordered group (in terms of their psycho-galvanic
response to stressful stimuli, and in the number of abnormal responses to the Maudsley
Medical Questionnaire). On the manual dexterity test, on which the psychotics performed worst, Eysenck suggested that the impairment in performance for the two abnormal groups may be a result of different processes. Psychotics may be slow because of their
mental retardation, whereas the neurotics may be adversely affected by their anxiety. In
the discriminant function analysis, two latent routes were found, each of which were
significant beyond the 0.1 % level: it was impossible to account for the test responses of
the three groups in terms of only one dimension. Eysenck sees these findings as strongly
indicative of the independence of neurosis and psychosis.
Eysenck uses a single, general neuroticism factor and locates the various neurotic
disorders within a three-dimensional personality framework. Cattell, who also based his
system on factor analysis, regards neurosis as a more complex multi-factor concept:
anxiety is an important part of neurosis, but it is only one of several dimensions making
up the higher-order factor. The Derogatis studies suggest that five basic factors may
constitute the core dimensions of neurosis (Derogatis et al. 1970, 1971a, 1971b), and
O'Connor (1953) obtained seven factors from an analysis of the symptom clustering of
over 300 male neurotic out-patients: again, no general factor of neurosis emerged.
Cattell and Scheier (1961) suggest that Eysenck and the psycho-analysts (an unlikely
combination!) both oversimplify neurosis by presenting it as a uni-factor concept.
Psycho-analysis accounts for neurosis in terms of regression, and Eysenck through a
largely constitutional neuroticism factor. Cattell argues that neither in terms of firstorder, nor second-order factors do neurotics differ from normals on only one general
factor. To some extent this conflict between single-factor and multi-factor concepts of
neurosis reflects differences in the use of factor analytic methods and differences in the
type of input data used. It also has much to do with the theoretical persuasions of the
investigators.
51
Cattell has given close consideration to the question of whether neurosis is functionally
separable from psychosis. Among the evidence that suggests that the two general types of
disorder may usefully be differentiated, Cattell and Scheier (1961) point to the larger
hereditary component that seems demonstrable for the psychotic disorders. Also, considering how unreliable specific psychiatric diagnoses can be, there is a relative absence of
confusion of the neurotic and psychotic classes of diagnosis.
In a series of studies, Cattell contrasted the questionnaire data (16PF) obtained from
480 clinically judged psychotics and 201 neurotics. The questionnaire profiles of the two
groups are different, and when compared with normal subjects, the factors that distinguish neurotics from normal groups are not the same as those which differentiate between
psychotics and normals (Cattell and Scheier 1961). Nor did the factors that differentiated
between normal and neurotic groups distinguish between normals and psychotics. The
bulk of the MMPI research data also points to important differences between neurotic
and psychotic groups (Gough 1946; Meehl 1946; Rosen 1962; Silver and Sines 1961),
and Cattell, like Eysenck, concluded that the neurotic and psychotic dimensions should
be regarded as separate and independent.
Part II
Theories of Neurosis
4. Conditioning Theories of Neurosis
If it is accepted that the modification of the central nervous system through experience is
the basis of most, if not all, human and animal behaviour, it should be clear why modern
learning theory plays an important role in psychology. Learning and conditioning are
pervasive determinants of behaviour, and insofar as the neurotic reactions are also
learned reactions, they too are dependent upon the laws of learning (Eysenck 1959).
Eysenck places great stress on the importance of experimental studies of learning and
conditioning, and of learning theory for an understanding of the neurotic disorders.
Joseph Wolpe, another figure whose name is closely associated with the classical conditioning model, has made the same point (Wolpe 1970).
Both Eysenck and Wolpe would accept the following general definition of neurotic
symptomatology, that it is a persisting pattern of learned behaviour which (for some
reason) is maladaptive. Neurotic behaviour fails to achieve its objective and leaves the
person dissatisfied and frustrated by his failure. Wolpe tends to be more prescriptive in
his definitions by emphasising anxiety, including within his (1970) definition of the
neurotic disorders that such behaviour be acquired in anxiety-generating situations; and
suggesting that the core of neurotic behaviour lies not in any particular behaviour or
motor activity, but in anxiety itseH, which is usually the central constituent of neurosis
and is always present in the causal situations (Wolpe 1958).
Like so many other concepts in abnormal psychology, that of anxiety is evasive; Wolpe
defines it as the responses predominantly of the autonomic nervous system with which the
individual characteristically responds to painful or noxious stimuli. Among its most common manifestations are rapid perspiration, rapid pulse rate, increased blood pressure and
a fairly general elevation of muscle tension. A similar specification of the characteristics
of neurotic behaviour was made by Eysenck (1975b). Neurotic behaviour is (a) learned
and not due to injury or lesions; (b) maladaptive: and (c) involves strong emotions,
particularly anxiety. Sidman (1964) has suggested that anxiety is an almost ubiquitous
phenomenon and that not all anxiety is neurotic. When faced by a realistic threat, anxiety
becomes a normal response. It has been suggested that fear is a more appropriate term to
use in this context. Anxiety is seen as neurotic when it appears in the absence of any
objective danger. Eysenck and Rachman (1965), however, distinguish between those
neurotic reactions such as phobias, anxiety states, obsessional and compulsive disorders,
in which anxiety plays a large part, and other disorders which are the result of a failure of
conditioning, and in which anxiety plays little part.
56
4.1 Classical Conditioning

The early classical conditioning model of neurosis proposed by Watson (Watson and
Rayner 1920) seems to have equated the neurotic disorder with (Pavlovian) conditioned
emotional responses. However, this model in its simplest form has had comparatively
little influence, possibly because of the failure of later investigators to replicate Watson's
findings (e. g. Bregman 1934), but also because of certain fairly obvious inherent shortcomings. Eysenck (1976) proposed an updated version of a conditioning theory of
neurosis described as the third conditioning model, of which Watson's was the first, and
Mowrer's, the second. But before dealing with the weaknesses of Watson's simple conditioning model and Mowrer's modified version, there are certain issues which can be
interpreted in the light of a comparatively straightforward classical conditioning model.
In the case of nocturnal enuresis, which has been regarded as a manifestation of
profound underlying emotional disturbances, passive-aggressive feelings towards the
parents, or as a substitute form of repressed sexuality, Mowrer's (1938) simple conditioning-deficit model was refreshingly straightforward. In this case there is an apparent
failure to connect enlargement of the bladder and the beginning of urination with the
appropriate response of waking up and going to the toilet. Mowrer (1976) described how
during the late 1930s he became convinced that Pavlovian conditioning could provide an
account of the development of nocturnal enuresis and a model for its treatment. The
process of conditioning which accounts for the normal waking response is shown in Fig. 3.
The conditioning establishes a connection between distention of the bladder and the
person waking up and inhibiting the urination. In the enuretic individual there is a failure
to wake up and prevent urination. Mowrer's answer to this problem was the by now wellknown pad-and-bell apparatus, which enables the establishment of the missing learned
connection between distention of the bladder and waking.
More ambitiously, Eysenck (1960b, 1977) has analysed delinquency and antisocial
behaviour in similar terms, arguing that conscience may be regarded as a conditioned
anxiety response to certain types of situations and actions. The child who behaves in
some socially undesirable way, by lying, stealing, being aggressive or whatever, is frequently punished: this leads to pain or fear, with the associated autonomic reactions, and
these become associated with the type of situation or type of action which led to the
punishment. Similar sequences of events, through the process of stimulus generalisation,
establish a conditioned response to a class of antisocial responses, and this generalisation
is usually aided by verbal labelling. In the case of the delinquent child, there has been a
Discomfort produced _ _'"!!!"'_...~ Relaxation of urethral

by bladder distention
""'- __ ~i:rj~d urination
r---- --
.
.
Awakening
stimulus
Fig. 3
Waking response and

contraction of sphincter
Classical Conditioning
57
failure of the conditioning process such that the child has not acquired the internal
restraints necessary to prevent him from behaving badly. In fact, Eysenck's explanation
of such phenomena is closely linked to a separate theory of individual differences, in
which extraverts show weaker conditioned responses than introverted subjects. This is
discussed at greater length in Chap. 7.
The reduction of conscience to generalised conditioned avoidance responses has met
with considerable resistance, though there are sufficient experimental demonstrations,
both with children and animals, to establish that behaviour can be guided in this manner.
Solomon et al. (1968), for instance, conducted an experiment with dogs in which the
animals were conditioned to avoid an initially preferred food and to choose an initially
non-preferred food. The dogs which had been trained with a short delay of punishment
quickly learned to confine their eating to the directed food. Delayed punishment produced a less effective conditioned restraint, and the behaviour of the animals was furtive
and unreliable. Solomon regarded these experimental procedures as directly comparable
to those involved in childhood socialisation, and Mowrer (1950) emphasised that neurotic disorders could best be regarded as examples of under-learning, and both enuresis and
the failure to acquire moral restraints were interpreted in this light.
Eysenck and Rachman (1965), however, have also drawn attention to a rather different problem in which the neurotic behaviour may be seen as the result of positive,
appetitive conditioning to stimuli contrary to the social, moral or legal requirements of a
particular country. Fetishistic behaviour and homosexuality are cited as examples of this
sort of behaviour, in which erotic behaviour patterns have been conditioned to particular
stimuli, possibly through orgasm, but possibly by other means. This category of 'neurotic'
disorders is surrounded by much controversy, since any decision to include such
behaviour within the realm of abnormal psychology cannot be divorced from morality
and value-judgment. In the case of homosexuality, for instance, the homosexuality itself
may have only minor implications for the personality development of the individual. It is
the attitudes of others towards this behaviour that are most likely to create the stresses
and anxieties which affect personality and social integration. Many adult homosexuals
strongly deny that they have any need for psychological help and are insulted by the offer
of 'treatment' (Schofield 1965). Siegelman (1978) found that homosexuals scored higher
on measures of neuroticism than heterosexual subjects, though this may be related more
to feminity (in both sexes) than to sexual preference itself, and Adelman (1977), who
gave the MMPI to groups of professionally employed lesbians and heterosexual women,
found no evidence of higher levels of neuroticism among the lesbians, though this group
did feel more socially alienated than the heterosexual women.
The complexity of the notion of neurosis, incorporating as it does not only individual
symptomatology and subjective distress, but also self-perceptions and the social reactions
of others, has already been discussed in earlier chapters, and this is not the place to
restate these issues. However, there are few areas in abnormal psychology which pose
quite so many problems as this of learned preferences (particularly sexual) which also call
forth social disapproval. One mayor may not choose to regard such behaviour as appropriate material for psychotherapy, though the individual may experience considerable
personal distress because of his homosexual, fetishistic, pedophiliac (or whatever) preferences. Although the persistence of such forms of sexual behaviour may lead to socially
aversive consequences which may in some cases outweigh their rewarding consequences,
one would wish to avoid equating behaviour which receives social disapprobation with
58
neurosis as psychopathology, though Flew (1978) has noted a similar recurrent error in
discussions of criminality.
But quite apart from the problem of how one chooses to categorise particular sexual
preferences, there is some evidence that classical conditioning procedures can produce
positive sexual responses. Rachman and Hodgson (1968) were able to condition a
number of male subjects to give a sexual response to previously neutral stimuli (photographs of knee-length boots) in the manner of fetishistic attraction, and Beech et al.
(1971) were able to use this sort of classical conditioning paradigm in the successful
treatment of a young man with deviant sexual tastes. The simple conditioning model,
however, is inadequate here, since in the typical fetishist the individual's sexual desire
for, and responses to, the fetishistic object are stronger than to the sexual object itself.
Indeed, the fetishists who see a clinician are likely to require the fetishistic object as an
indispensable stimulus for the occurrence of any response to the primary sexual object.
These aspects of fetishism are not explained by the simple conditioning theory.
There are also problems for a simple conditioning theory in accounting for the more
traditional neurotic reactions such as anxiety states, obsessional and compulsive disorders
and phobic reactions. In the first place, the limitation of unconditioned stimuli to pain
and fear presupposes that neuroses develop as a result of traumatic single trial conditioning or multiple sub-traumatic conditioning. Except in a few exceptional cases, for instance in some war neuroses or in simple phobic reactions, it is comparatively rare to find
such 'natural' fear producers at the beginning of a neurosis (Eysenck 1975b, 1976).
4.2 Conflict and Frustration

Conflict is one factor that has been suggested as a link between exposure to punishment
and neurosis, and Dollard and Miller (1950) regarded conflict as the primary cause of the
neurotic's unhappiness.
Pavlov (1927) first reported that experimental neuroses could be induced in animals,
and classical conditioning theorists have placed considerable weight upon such experimental results. In Pavlov's 'circle and ellipse' experiment, the animal was conditioned to
associate the circle with food and the elliptical stimulus with no reward. Having established these conditioned responses, the shape of the ellipse was gradually approximated
to that of the circle. When the ratio of the semi-axes of the ellipse reached 9: 8, the close
resemblance of the ellipse to the circle produced a dramatic effect upon the animal's
behaviour. The dog, which had previously been a quiet animal, began to squeal and
struggle in the apparatus, and when it was taken into the experimental room it would
bark violently. Wolpe (1970) referred to this as a state of conflict produced by the strong
simultaneous activation of opposite action tendencies - the evocation and inhibition of
the alimentary responses. This conflict led to a high level of autonomic arousal which
became conditioned to the environmental stimuli of the laboratory. Other investigators
were later to use similar methods involving such forms of conflict to produce experimental neuroses. Jacobsen et al. (1935), for instance, obtained experimental neuroses in
chimpanzees by presenting discrimination tasks too difficult for the animal, which
became angry and emotionally aroused by its errors. There are, of course, many difficulties involved in the attempt to translate these findings into the human neuroses, though
the essential issue is not whether the external behaviour is in all respects the same in
Conflict and Frustration
59
man and animal, but whether the mechanisms controlling the behaviour are the same or
not (Hebb 1947). Wolpe (1952) examined the similarities between the two and argued
that experimental and clinical neuroses are parallel phenomena.
The use of the term 'conflict' in these accounts must not be misunderstood; it has no
relationship to the psycho-dynamic notion of 'unconscious conflict' between the ego and
the id (or super-ego and id), as proposed by Freud. The source of anxiety in this case is
quite different. Pavlov, for instance, regarded conflict as the simultaneous elicitation of
positive and negative conditioned responses which he related to the physiology of the
central nervous system. A more behavioural discussion of the same problem was provided by Neal Miller, whose gradient model of conflict was derived from the work of Hull
(1932). Miller (1959) suggested that the nearer the subject is to a goal-object, the
stronger is the tendency to approach that object: equally, the tendency to avoid a feared
stimulus increases as the subject gets nearer to it. In addition, Miller found it necessary to
postulate that the strength of avoidance increases more rapidly with proximity to the goal
than does that of approach.
More recently, Miller's proposals have been challenged by Hearst (1969), whose
experiments showed that the gradients of approach and avoidance depended upon the
specific parameters of the behavioural task. By careful manipulation of the experimental
condition, Hearst was able to produce either a steep or a flat approach gradient and
either a steep or a flat avoidance gradient. Hearst suggested that there was probably no
profound difference in the nature of approach and avoidance behaviour as originally
suggested by Miller, and that research might concentrate more profitably upon the particular parameters that flatten or steepen gradients, regardless of approach or avoidance.
This does not, however, lessen the conceptual value of the distinction between
approach and avoidance. There are various types of conflict that can be proposed; in
their simplest form, these are approach-approach, avoidance-avoidance and approachavoidance conflict. Teasdale (1974) has suggested that the avoidance-avoidance paradigm can be most profitably applied to the obsessional-compulsive disorders. The obsessional-compulsive patient is seen as being caught in the conflict of whether to perform his
rituals or not. If he does, he avoids the feared aversive consequences which might follow
non-performance; but the performance itself is hypothesised to lead to other aversive
consequences, hence the conflict. Teasdale's analysis, however, assumes that the rituals
are avoidance responses which save the person from aversive consequences, which may
be probable, but leaves unexplained the fact that obsessionals may sometimes stop performing their rituals or even deliberately engage in acts of contamination (Beech and
Perigault 1974).
Also, Teasdale suggests that the oscillation between two alternative responses that is
characteristic of laboratory avoidance-avoidance conflict, has been described in clinical
accounts of the behaviour of obsessional-compulsive patients who were in doubt as to
whether or not to perform their rituals (e. g. Mather 1970). It is not clear, however, why
the patient should sometimes perform the ritual and sometimes avoid doing so. The
suggestion that the ritual is performed because the overall aversiveness of the consequences of performance of the ritual is less than that following non-performance (Teasdale 1974) is unhelpful because it falls so quickly into tautology.
Others have suggested that approach-avoidance conflict may provide a more useful
model for the analysis of the neurotic disorders (e. g. Maher 1966). The approachavoidance conflict is a form of passive avoidance learning, in which the animal is simul-
60
taneously attracted to and repelled by the same stimulus. Miller asserted that conflict
generates drive, and being part of the Hullian tradition, maintained Hull's uni-process
view of learning, looking to drive reduction as the source of reinforcement.
Gray (1971) applied Miller's analysis to sexual deviations, suggesting that such
phenomena as voyeurism and fetishism could be seen as examples of the conflict between
heterosexual approach tendencies and avoidance tendencies, and that such psychosomatic disorders as stomach ulcers can be caused by conflict and helplessness. Gray also
referred to the interesting series of studies discussed by Azrin and Holz (1966). In one of
their experiments (Hoiz and Azrin 1961), these authors punished the responses of the
subject animal during the period of reinforcement. The result of this procedure was to
reverse completely the usual effects of punishment - the rate of responding increased in
the presence of punishment and decreased in its absence. Punishment when followed by
reward can acquire secondary reinforcing properties. This finding is extremely interesting
and may well have a certain bearing upon such human disorders as masochism, in which a
person becomes aroused by things which most of us would regard as painful or humiliating. Gray (1971) has also pointed to the way in which masochism may be explained in
terms of the pairing of reward and punishment such that punishment becomes a sign of
reward.
Dollard and Miller (1950) presented one of the more explicit early statements of the
effects of conflict upon neurosis. In the case of approach-avoidance conflict, Dollard and
Miller saw this as the sort of conflict which usually occurred between the responses of
trying to think about or remember something and the response of repression; equally it
could be seen as the sort of conflict which occurred when the neurotic is prevented by his
inhibitions from achieving his goal. These authors also regard the neurotic patient as a
specially selected case with unusually strong avoidance tendencies. Any attempts to
increase the patient's motivation to achieve his thwarted goals in life are likely only to
increase the level of conflict and therefore his misery, and Dollard and Miller suggested
that the therapist should begin his work by reducing the fears which cause the avoidance.
Most behaviour therapists would probably accept that conflict may play an important
role in the development of experimental and clinical neuroses through its aversive properties (Chesser 1976). It is not, however, accepted as a necessary cause of anxiety
neuroses, and Wolpe (1958) has argued that traumatic conditioning is a more likely
sufficient cause of such neurotic reactions.
Amsel (1962) assigned a similar function to frustration as a source of drive. Hull's
theory failed to acknowledge that the omission of a reward constituted any special sort of
event for the subject, and Gleitman et al. (1954) traced some of the difficulties which
follow from this. Working within a neo-Hullian framework, Amsel described frustrative
non-reward as having many of the properties of punishment, and operating in many
respects like fear. Indeed, he suggested that the avoidance of non-reward was a more
powerful factor in discrimination learning than approach to reward.
Gray (1971) has taken a similar position, arguing that the omission of an expected
reward has effects which are very similar to, and may even be identical with, the effects of
punishment. In both its functional and physiological effects, frustration (in this sense) can
be equated with fear. It is impossible to demonstrate the aversive qualities of non-reward
directly, since the termination of non-reward is simply the attainment of reward, but it
can be shown that stimuli which have been associated with frustrative non-reward are
aversive (Adelman and Maatsch 1956). It is also possible to substitute direct physical
Two-Stage Theory
61
punishment for frustrative non-reward and obtain very similar results (Brown and
Wagner 1964).
Early accounts of conditioned fear responses emphasised physical pain or fear as the
UCR in the conditioning process (cf. Watson). Eysenck (1975b) has argued that this is
too restrictive and invoked the notion of frustrative non-reward as a much more likely
variable. Actual physical pain and fear are not often easy to find when one tries to
determine the characteristics of the situations from which the neurotic reaction
developed, though they may be found in the extreme circumstances leading to war
neuroses. It is also possible that some simple phobias may be associated with such factors.
But with these exceptions, traumatic single trial conditioning, or multiple, sub-traumatic
conditioning, with pain or fear as the UCR, cannot provide a satisfactory explanation for
the development of most neurotic disorders. Frustrative non-reward has greater plausibility as a UCR in this context.
The use of this concept of frustrative non-reward, however, introduces a cognitive
element into the discussion. The absence of reward cannot have specific aversive properties unless the organism expects to be rewarded. Eysenck (1979) denied that this makes
any fundamental difference to the non-cognitive status of the conditioning theory, on the
grounds that cognition can itself be explained in terms of the principles of conditioning.
This is hotly disputed by the cognitive theorists (see Chap. 5).
4.3 Two-Stage Theory

One of the greatest obstacles for learning theory in attempting to explain how neuroses
develop is the phenomenon of extinction. Unreinforced conditioned reactions extinguish
rapidly, yet neurotic reactions appear to contradict this fundamental principle of learning
theory. Mowrer (1950) provided one of the clearer statements of this 'neurotic paradox':
Common sense holds that a normal, sensible man, or even a beast to the limits of his intelligence,
will weigh and balance the consequences of his acts: if the net effect is favourable, the action
producing it will be perpetuated; and if the net effect is unfavourable, the action producing it will
be inhibited, abandoned. In neurosis, however, one sees actions which have predominantly
unfavourable consequences; yet they persist over a period of months, years, or a lifetime.
It is because neurotic behaviour runs contrary to the predictions of Thorndike's law of
effect and Skinner's principles of reinforcement that a special theory is required.

Mowrer and Ullman (1945) tackled this neurotic paradox of behaviour which persisted, despite the fact that it was more punished than rewarded, and suggested that
Hull's (1932, 1943) notion of gradients of reinforcement could be used here. If a
response is followed by two consequences, one of which is rewarding and the other
punishing, this may lead to persistent non-integrative behaviour. If a weak reward is
received before a strong punishment, behaviour may persist despite the greater punishment: equally, if the reward occurs later, behaviour may be inhibited despite its being
more rewarding than punishing. In Mowrer's view, neurotic behaviour is best seen as a
learning deficit rather than as a case of overlearning, and an explanation of neurosis
should deal with the problem of why the neurotic fails to learn, and not with why he fails
to unlearn. Like the ostrich, the neurotic is seen as dealing directly with his anxiety by
62
avoiding exposing himself to reality, or at least by delaying the consequences of his

actions in the real world.
Mowrer's two-stage theory of neurosis proposed that a neutral stimulus comes to elicit
anxiety as a result of classical conditioning. This learning is then protected from extinctioQ. by the second stage, in which an instrumental avoidance response leads to anxiety
reduction and sets up a conditioned avoidance reaction to the CS. The remarkable
persistence of the neurotic behaviour can therefore be accounted for in terms of the
immediate anxiety reduction consequences of the avoidance, against the de/ayednegative
effects of such behaviour. In psychiatric terms, this behaviour can be interpreted as the
avoidance of 'reality testing' (Eysenck 1976).
Strictly speaking, Miller's theory was a one-stage theory rather than a two-stage theory
like that of Mowrer. Miller regarded learning very much in Hullian terms, as a process of
drive reduction, and thought that the original fear response was itself learned on the basis
of fear reduction. Gray (1975) has reviewed the evidence in support of this position and
concluded that it is incompatible with the weight of empirical evidence. What both
theorists agree upon is that the person or animal in an active avoidance situation does not
learn to avoid the punishment but to terminate the stimuli which are followed by punishment. Both, therefore, rely heavily upon the motivating properties of anxiety Ifear reduction; (Miller [1948] showed that conditioned fear stimuli were indeed aversive). The fear
and anxiety are seen to motivate and reinforce behaviour that avoids or prevents the
recurrence of the aversive unconditioned stimulus (Mowrer 1939).
The unusual circumstances of warfare which produced 'war neuroses' have been cited
by several authors in support of their argument. Dollard and Miller (1950) referred to
the case of a pilot who developed a phobic fear of aeroplanes as a result of his traumatic
experiences in combat. The fear resulting from the initial trauma of combat became
associated with the aeroplane and generalised to other aeroplanes: this fear motivated his
avoidance, which in turn was reinforced by the reduction in the strength of his fear. In the
case of the war neuroses, the trauma of combat was usually counteracted by a process of
enforced extinction in the front line. By removing the person to a base hospital, the
conditioned fear response would be protected from extinction in the same way that the
avoidance response protects the first stage of learning. As early as 1926, and contrary to the predominantly psycho-dynamic orientation of military psychiatry at that
time, Fenton wrote that 'shell-shock' could be understood in terms of the laws of habit
formation.
Among the more common neurotic disorders, Mowrer (1969) has discussed the
psycho-neurotic defences (for example, deception and denial) as punishment-avoidance
strategies. In particular, he examined the ways in which the subject can perform a
forbidden act without getting punished or refrain from performing a required response
and still avoid punishment. Under these circumstances, the person is likely to experience
conflict due to the opposing requirements of the situation. Lying and deceit are ways of
reducing the original conflict, though they entail the problems of a further conflict between the guilt resulting from the deceit and the fear of punishment which would occur if
the deception were admitted. In passive avoidance situations, Mowrer uses the term
'guilt' to refer to the fear experienced between the performance of the forbidden act and
the occurrence of punishment. This distinguishes it from temptation fear. It is not clear if
guilt is a relevant feature of active avoidance, since this can be explained quite satisfactorily by the simpler concept of 'fear'.
Two-Stage Theory
63
Two features of neurosis which require explanation are the occurrence of spontaneous
recovery in some cases and the massive resistance to extinction in others. Eysenck and
Rachman (1965) used a two-stage theory (though they refer to it as a three-stage theory)
to explain both. Spontaneous recovery, which they describe as the most obvious, most
impressive and least frequently acknowledged aspect of neurotic disorders, is explained
as being due to classical conditioning processes only. That is, the first stage of conditioning establishes a learned fear through the association of a neutral es with the fearproducing DeS, and the es alone subsequently elicits the original maladaptive emotional
response. However, the conditioned autonomic responses extinguish as the person
encounters further instances in which the es presentation is not reinforced. Denker
(1946) looked at 500 severe cases of neurosis and found that 70% recovered within 2
years without any form of psychotherapy. After 5 years, over 90% had recovered. To this
extent, the classical conditioning account can explain why a substantial majority of
neurotics show spontaneous recovery. In order to explain why some cases of neurosis are
apparently so resistant to extinction, Eysenck and Rachman (1965) found it necessary to
utilise the second stage of instrumental learning. The active avoidance response protects
the original learned fear from extinction and is itself reinforced through anxiety reduction.
Dollard and Miller (1950) suggested that compulsive behaviour and hysterical disorders could also be explained by the two-stage theory, and Eysenck and Rachman (1965)
argued that all the dysthymic neuroses (anxiety states, phobias, obsessional and compulsive disorders etc.) could be explained in this way. More recently, however, both Eysenck
(1968, 1975b, 1976) and Rachman (1976) have recognised the deficiencies of this sort of
account and have proposed a number of revisions and extensions.
Among the deficiencies of this theory was its inability to explain the extreme persistence of active avoidance behaviour. Solomon and Wynne (1953) used an adaptation of
the Miller-Mowrer shuttle-box to investigate traumatic avoidance learning. Their dogs
were trained to avoid an electric shock by responding to a signal preceding the shock.
During the escape phase of learning, the animals received shocks, and the latency of the
escape response (jumping to the other side of the shuttle-box) gradually decreased. After
the onset of the successful avoidance response the latencies continued to decrease,
although the animal was no longer being exposed to the noxious DeS. Solomon and
Wynne (1954) also described how a few intense shocks during the acquisition of avoidance could lead to hundreds of successful avoidance responses which occurred with fewer
and fewer overt signs of anxiety. But according to an unmodified two-stage theory, as the
fear of the warning signal declines (this is described by Solomon and Wynne 1953) the
avoidance response should also decline, and in the absence of further reconditioning the
successful avoidance response should extinguish. Solomon and Wynne dealt with these
difficulties by introducing the principle of 'anxiety conservation', which referred to a
complex interrelationship between the strength of the anxiety reactions and the instrumental avoidance responses. In those cases where the subject is exposed to a particularly
strong DeS which produces intense anxiety, Solomon and Wynne also suggested that the
classical conditioning of the emotional response may be partially irreversible.
Several authors have stressed the relative independence of fear and avoidance (e. g.
Gray 1971, 1975; Rachman and Hodgson 1974), and Gray has used the notion of 'safety
signals' to explain the persistence of active avoidance behaviour. In this way, the fear
established in the initial stage could extinguish and still leave the avoidance behaviour
64
intact. The safety signals possess reinforcing properties established by classical conditioning (Rescorla 1969) and are as effective as warning signals in maintaining active avoidance (Bolles 1970; Bolles and Grossen 1969). In this way, successful avoidance
behaviour could be ascribed to reinforcement provided by the accompanying safety
signals even when there was no conditioned stimulus for the organism to terminate.
Nonetheless, this modification of the Mowrer two-stage theory is not entirely satisfactory, since it is difficult to see how avoidance behaviour could ever spontaneously extinguish under such conditions.!
4.4 Incubation
Wolpe (1958) drew attention to a number of experimental findings that present other
difficulties for any learning theory. These show that in many cases not only does the
unreinforced CS fail to extinguish, but it actually produces more and more anxiety (the
conditioned response) with each presentation of the CS. Eysenck (1976) also discussed
this phenomenon, which even more than the failure of extinction seems to constitute the
central paradox of the neuroses. Eysenck (1975b, 1976) has argued that the law of
extinction as originally stated is in need of major revision, and has suggested that the
presentation of an unreinforced CS (symbolised as CS) which has been associated with a
pain, fear or anxiety-producing UCS may have two separate and contradictory effects.
The CS may produce (a) extinction of the CR and (b) enhancement of the CR.
The specific outcome in any case will depend upon the strength of the UCR, the
duration of the CS and the emotionality of the individual. Grant (1964) distinguished
between Pavlovian A conditioning, which is exemplified by Pavlov's bell-salivation
experiments, and Pavlovian B conditioning. This distinction has not perhaps received the
attention that it deserves, since its implications are of particular importance for a conditioning theory of neurosis. Eysenck (1979) has argued that neurosis may only be understood in terms of Pavlovian B conditioning.
This can be illustrated by an experiment in which the experimental animal is given
repeated injections of morphine which lead to nausea, increased salivation and vomiting.
After repeated injections the animal begins to show nausea, vomiting etc. when
approached by the experimenter (Pavlov 1927). In Pavlovian B conditioning the UCS is
not contingent upon the subject's behaviour, is less dependent for its effect upon the
subject's motivational state, and the CS acts as a partial substitute for the UCS. In
addition the UCS elicits the complete UCR.
Eysenck (1968) suggested that anxiety has certain peculiar properties which distinguish it from other types of conditioned responses (with the possible exception of sexual
CRs). Anxiety has secondary drive properties and is also a source of secondary reinforcement (the primary drive involved acting through pain or frustrative non-reward). 'Neutral stimuli associated with pain give rise to anxiety-fear responses ... and the proprioceptive consequences of these learned responses produce the drive stimuli (SD) that
serve the secondary drive' (Eysenck 1975b). The conditioned pain reaction resembles
the unconditioned pain reaction in that both have the drive-producing and reinforcing
For the determined reader who wishes to immerse himself in the intricacies of two-stage theories
of avoidance learning, Gray's (1975) discussion can be recommended.
65
Incubation
properties of primary drives. In Eysenck's account it is essential to distinguish between

CRs that do have drive properties and those that do not. Conditioned stimuli which do
not produce drives are subject to the classical law of extinction. Those CSs that do
produce drives lead to an enhancement or incubation of the CR. This is because when
anxiety (or another response with drive properties) is being conditioned, it is not true to
say that the CS is not being reinforced in the absence of the VCS. After conditioning, the
CS itself produces fear/anxiety.
The CS and CS both come to signal danger and punishment. Eysenck terms these
consequences NRs - for nocive or noxious responses, to distinguish them from VCRs and
CRs. Each NR which follows a CS represents a reinforcement and increases the habit
strength associating stimulus and response. This incubation effect in the absence of the
VCS and VCR is shown in Fig. 4. Although it may be a much weaker form of reinforce"
ment than that accompanying the VCS, the presence of the NR would theoretically lead
to incubation. The conditioning has in this case established a positive feedback cycle in
which the CR (NR) provides reinforcement for the CS. More accurately one might say
that it is not the CR itself, but the response-produced stimuli, the fear/anxiety, that acts
as reinforcer (Eysenck 1979). An electric shock is followed by pain; the CS is followed by
fear. Shock + CS is followed by fear and pain (NR). This combined NR is more unpleasant than either on its own and therefore has greater reinforcing properties. The CS is
followed by fear, which although less reinforcing than pain and fear, may still be sufficiently reinforcing to offset the extinction and lead to an incubation effect.
As a clinical illustration of this process, Eysenck (1975) presents two cases - one of
insomnia and one of sexual impotence. In the case of a man who finds himself impotent
on one occasion due to drink, fatigue or illness, the CSs associated with this occasion
produce anxiety as a CR. On the next occasion, these CRs (or NRs) follow the CS and
incubation of CR
250
~
~
CS
200
Q)
:;
III
III
n.
150
"0
0
0
:n
.E
100
Q)
III
0
~
.E
50
Habituation of UCR
50
75
100
Triais
Fig. 4. The incubation effect. The increase in blood pressure as a function of repeated exposure to
the ues (habituation) or to the eS-only. Presentation of the es alone led to an incubation of the
eR (Eysenck 1975b). I am grateful to Professor Eysenck for allowing me to use this illustration
66
cause reciprocal inhibition of the sexual responses. This failure increases the strength of
the anxiety eRs and sets up the conditions for the positive feedback to establish the
impotence without any repetition of the original es-ues presentation.
As well as its value in explaining features of the dysthymic neuroses, the incubation
effect can also be applied to neurotic disorders of the second kind. In fetishism, for
instance, the sexual responses to the fetishistic object are often stronger than the primary
sexual object. A simple conditioning theory cannot account for this. However, sexual
responses, like anxiety, possess drive properties, and the incubation paradigm may be
used to explain the development of fetishistic behaviour in the absence of a es-ues
pairing.
Eysenck (1976) has also pointed out that the duration of eS-only exposure is an
important factor, and that short periods of exposure to the es on its own are likely to
produce incubation, whereas long periods of exposure to the eS-only produce extinction.
This has clear implications for response-prevention or 'flooding' therapies, since the use
of short es exposures may have an adverse effect. In one instance, a phobic patient who
had been exposed to a number of brief (2-min) sessions of eS-only presentations of her
phobic stimuli reported that her fear increased during treatment (Rachman 1966).
4.5 The Stimulus

We have largely taken the concept of the 'stimulus' for granted, and few learning theorists have examined the problems associated with this notion. One of the first difficulties
with the conditioned stimulus is that in real life the es is very rarely comparable to the
single stimuli which are presented in the usual laboratory experiments; instead, it is much
more likely to consist of a variety of interrelated events and objects. In the laboratory the
subject is usually exposed to a single clear-cut stimulus which is highly salient because of
the controlled conditions, but in real life, classical conditioning theory cannot specify
which aspect or aspects of the environment will take on the role of the es. Early
conditioning theory seems to have taken for granted that all potential stimuli were
equally associable, but Seligman (1970) has challenged this assumption (as is discussed
later in this chapter).
Eysenck (1975b) acknowledged that single, specific conditioned stimuli are rarely
found in real-life conditioning, that single-trial traumatic conditioning is exceptional, and
that even multiple sub-traumatic conditioning with a single es is not often observed in
the genesis of human neuroses. Eysenck goes on to state that there is no reason why this
should invalidate a conditioning theory of neurosis, though the problem clearly becomes
far more complex and difficult when there is a multiplicity of conditioned stimuli.
In those cases of neurosis which are characterised by 'free-floating anxiety', it has
sometimes been suggested that the conditioning theory is inadequate because of the
absence of a specified conditioned stimulus. Wolpe (1967 a), however, has argued that
anxiety responses can be conditioned to more or less omnipresent properties of the
environment which cause the person to become persistently restless and anxious, and
suggests that such amorphous features as light, light and shade contrasts, spatiality and
even the passage of time can function as conditioned stimuli in classical conditioning.
Under such conditions, the patient appears to show a generalised agoraphobia or 'freefloating anxiety', when in fact they are still responding to a es. Wolpe regards the term
Preparedness
67
'free-floating anxiety' as a misnomer and has offered 'pervasive anxiety' as a more

appropriate description. Wolpe views neurotic behaviour as directed by external stimulation and supports his suggestion that anxiety has specific stimulus sources even in cases of
pervasive anxiety disorders by citing clinical evidence.
This sort of analysis of the more generalised neurotic disorders in terms of specific,
external conditioned stimuli by Wolpe is interesting, but in danger of becoming tautologous. It is not clear how one could ever provide an example of an anxiety-based neurotic
disorder that was not controlled by specific conditioned stimuli if Wolpe is prepared to
allow this role to such vague notions as spatiality or the passage of time. The passage of
time is, in any case, a highly abstract concept which cannot be said to exist in any real
sense except in cognitive terms.
Classical conditioning theory has always been somewhat ambiguous about the role of
cognitive factors in determining behaviour. One might infer from the work of S-R theorists that external stimuli have a direct impact upon the organism, and that cognitive
factors are unimportant. Hearst (1969) stated that, 'the delivery of noxious stimuli to an
organism may affect not only the future probability of various responses by the subject,
but also the extent to which different stimuli in the external environment control the
subject's behaviour.' Conditioning and cognitive theorists have usually resisted explanations couched in terms of the other's concepts and theories, though there have been
attempts to reconcile the two in recent years (e. g. Seligman and Johnston 1973).
4.6 Preparedness
A further modification to the simple classical conditioning account has been proposed by
Rachman and Seligman (1976) and Eysenck (1976). This involved the concept of preparedness.
In a review of the data from traditional learning paradigms, Seligman (1970) showed
that the assumption of equivalence - that all events are equally associable and obey
common laws - made by general process learning theory is incorrect. In classical conditioning, it is not true that any CS and UCS can be associated with equal facility, nor that
the laws governing the rates of acquisition, extinction, spontaneous recovery, etc. are the
same for all CSs and UCSs. Nor, in instrumental conditioning, is it the case that the
choice of response and reinforcer is a matter of relative indifference. The organism
appears to show a preparedness to associate certain stimuli with relative ease, whereas
others may be associated only with the greatest difficulty (contra-preparedness).
In one experiment (Garcia and Koelling 1966), rats were exposed to X-rays whenever
they drank saccharin-flavoured water. At the same time, they were also exposed to
flashing lights and to a noise. The X-irradiation makes the rats sick after a delay of about
an hour. When the rats were retested, they had acquired an aversion to the taste of
saccharin, but not to the visual and auditory stimuli to which they were also exposed
during the experimental sessions. There is a host of experiments which show that rats will
readily learn certain responses, such as bar-pressing for food and running to avoid shock,
but it is difficult to train rats to press bars to avoid shock (D'Amato and Schiff 1964).
Seligman (1970) pointed out the biological utility of the taste-nausea preparedness, and
the same can also be said of the preparedness of certain avoidance or appetitive
responses. As in other areas (e. g. Maturana et al. 1960; Gossop, 1974), the animal
68
seems prepared to behave most appropriately to stimuli which have clear biological and
evolutionary significance for it.
Seligman (1971) has applied this sort of analysis to phobias and pointed out that in
general, phobias comprise a relatively limited set of objects. Agoraphobia, fear of certain
animals and insects, fear of heights and fear of the dark are common. Fears associated
with electricity, hammers or cars are extremely rare, although they are just as likely to be
linked with traumatic events. Seligman argued that phobias are highly prepared to be
learnt by humans and resistant to extinction: they are instances of prepared conditioning.
All stimuli are not equally likely to be transformed into fear signals (Rachman and
Seligman 1976), and the preparedness of stimuli is related to their biological significance.
Among the clinical implications of this, it has been suggested that unprepared phobias are
likely to be highly intractable to behaviour therapy, whereas behaviour therapy may be
the treatment of choice for prepared phobias (Rachman and Seligman 1976).
4.7 Therapeutic Implications of Conditioning Theory

It will become apparent as the various accounts of neurosis are presented in this book,
that different theories have different areas of strength and weakness. Conditioning
theory is particularly strong in explaining specific anxiety-linked neurotic symptoms (e. g.
phobias and obsessions) and rather weak in dealing with more generalised disorders such
as depression.
One therapeutic principle that underlies much of behaviour therapy is that of counterconditioning, in which the acquisition and strengthening of a new response incompatible
with the previous response leads to the elimination of the earlier response (Meyer and
Chesser 1970). It has been suggested that anxiety can be directly deconditioned by
associating it with an incompatible response, and Wolpe has emphasised the importance
of reciprocal inhibition in this context. If a response which inhibits anxiety can be made to
occur in the presence of anxiety-evoking stimuli, it will weaken the connection between
those stimuli and the anxiety (Wolpe 1958, 1970). It has even been suggested that despite
the theoretical persuasion of the therapist, effective psychotherapy is nearly always based
upon reciprocal inhibition of neurotic anxiety responses (Wolpe 1960, 1967b).
Watson and Rayner (1920) suggested that Albert could have been treated by attempts
to 'recondition' him in this way, using simultaneous pairings of the fear-producing CS and
stimulation of the erogenous zones. Mary Cover-Jones (1924) first demonstrated that
neurotic fears in children could be eliminated by pairing a feared CS with pleasant
stimuli, and the behaviour therapy literature is now full of accounts confirming and
extending the values of treatments based on this procedure. One of the best known
treatments is systematic desensitisation based upon relaxation. 2
Intense muscle relaxation effects are antagonistic to the autonomic arousal associated
with anxiety. By presenting a graduated hierarchy of anxiety-provoking stimuli paired
with relaxation, the anxiety responses become inhibited. Lazarus and Rachman (1960)
suggested that this procedure should be most readily applicable to disorders involving
neurotic reactions to concrete and definable stimuli, and in which specific rather than
2 The reader is referred to Wolpe (1958) and Wolpe et al. (1973) for a full description of the
precise procedures involved in systematic desensitisation.
Therapeutic Implications of Conditioning Theory
69
free-floating anxiety is present. In Wolpe's (1970) view, however, neurotic responses are
stimulus-specific even in the case of 'free-floating anxiety' in which anxiety responses
have been conditioned to 'more or less omnipresent properties of the environment' such
as light, noise or the passage of time (Wolpe 1967a). The term 'pervasive anxiety' is
suggested as a better alternative, since in Wolpe's view there is no clear distinction
between specific anxiety-provoking stimuli and stimuli to pervasive anxiety. It is, however, generally accepted that patients with multiple fears and extensive social and emotional difficulties are not as good candidates for systematic desensitisation (Wolpe et al.
1973).
In his review of the 75 outcome studies of systematic desensitisation published prior to
1968, Paul (1966) found that the results clearly showed its effectiveness as a treatment
for a range of anxiety-based neurotic disorders (including specific phobias as well as
agoraphobia, claustrophobia, frigidity and impotence). There are, however, problems in
explaining the effects of desensitisation. Benjamin et al. (1972) found that relaxation was
a redundant component in the total procedure, and Yates (1975) suggested that every
one of the specific components of systematic desensitisation could be removed without
leading to therapeutic failure: like the Cheshire cat, desensitisation is left with only its
smile. Although Wolpe has emphasised the role of reciprocal inhibition in desensitisation, the effects may equally be explained in terms of counter-conditioning, extinction or
habituation. For this reason, although it appears that desensitisation is effective as a
treatment for certain neurotic disorders, the critical factors have not been identified, and
the theoretical explanations of the effects are conflicting.
It has been suggested, for instance, that exposure to the anxiety-provoking stimuli may
be of more importance than relaxation (Marks 1974). Flooding, in which the patient is
exposed to intense imaginary or real-life, anxiety-provoking stimuli, is, in terms of
stimulus intensity, at the opposite extreme to desensitisation; yet Boulougouris et al.
(1971) found that in the treatment of phobic patients, flooding (in fantasy) was superior
to desensitisation on clinical and physiological measures, though both treatments led to
improvement. Marks et ai. (1975) treated 20 patients with chronic obsessive-compulsive
rituals by in-vivo exposure with self-imposed response prevention, and they found a
significant improvement in compulsive behaviour after 3 weeks, which was maintained at
follow-up. Other studies by Rachman et ai. (1971) and Hodgson et ai. (1972) also
suggested that obsessional neuroses improved significantly more with exposure than with
relaxation treatments, and flooding may be a more appropriate treatment for
agoraphobia than desensitisation. Habituation was suggested as a possible critical feature
of the flooding, modelling and relaxation treatments (Rachman et al. 1971), since the
patients were prevented from avoiding exposure to the anxiety-provoking stimuli. Similarly, response-prevention techni<J!les have been found to be effective with obsessionalcompulsive disorders (Meyer and Levy 1970; Levy and Meyer 1971). In this case, the
patient is again deliberately exposed to the anxiety-provoking stimuli and the process of
extinction is encouraged by the therapist preventing the compulsive thoughts or rituals.
Aversion therapy is one of the best-known, and most controversial, forms of behaviour
therapy. It has been used mainly in the treatment of approach responses which cause the
patient distress (and which are frequently also the subject of social disapproval). Among
these disorders are alcoholism, drug addiction and various sexual disorders such as fetishism, transvestism, voyeurism and, more controversially, homosexuality (e. g. Marks and
Gelder 1967; Hallam et al. 1972). In theory, the treatment works by the association of a
70
conditioned aversion with the undesired habit either by presenting a noxious stimulus
when the act is performed, or by pairing a noxious stimulus with the stimuli that elicit the
behaviour. Again, this eliminates the undesirable behaviour by establishing an incompatible response.
The two sorts of aversive stimulation that have been used in almost all studies are
electric shock and various chemical agents. The use of drugs has a number of drawbacks
associated with it. The drugs may have a delayed and unpredictable effect, and emetic
drugs are obviously messy and unpleasant for the therapist and nursing staff as well as for
the patient, and Rachman (1965) has advocated the use of electrical methods. This
enables a more precise control of the timing, intensity and duration of the stimulus, and
allows the use of partial reinforcement schedules.
Although some interpretations of aversion therapy have relied upon anticipatory
avoidance conditioning as the effective underlying principle (Feldman and MacCulloch
1971), there are good reasons to suppose that classical conditioning might be the critical
factor (Rachman and Teasdale 1969). Certainly chemical forms of aversion therapy may
be explained quite satisfactorily in terms of classical conditioning. In this case, the repeated association between particular stimuli and the VCR of nausea provoked by an
emetic drug establishes a conditioned nauseous reaction to the undesirable activity.
However, there are considerable difficulties involved in explaining electrical aversion
therapy in this way.
According to classical conditioning theory, the pairing of CS with shock should lead to
the CS producing a conditioned response of pain/anxiety. In fact, there is comparatively
little evidence to support this interpretation. Although Hallam and Rachman (1972)
found changes in cardiac responsiveness to sexually deviant stimuli after aversion
therapy, other studies found little alteration in physiological function: nor were patients
typically found to report feelings of subjective anxiety. Some patients described feelings
of indifference to the CS after treatment (Marks and Gelder 1967), and others experienced repulsion rather than anxiety (Hallam et al. 1972). Aversion therapy, like other
behaviour therapies, appears to work fairly well; it compares favourably with other
treatments for the same disorders, for instance sexual deviations (c. f. Feldman 1966).
But the mechanisms by which this and the other behaviour therapies achieve their effect
are still not completely understood.
5. Social and Cognitive Theories
In his analysis of the dynamics of scientific investigation, Kuhn (1962) described how
science is guided by paradigms. A paradigm has a guidance function, providing an intellectual framework which determines the scientist's views about his subject matter. In
psychology, the paradigm or framework within which one operates influences which
aspects of human functioning are studied most thoroughly and which are neglected.
Psychologists who choose to exclude the capacity for self-direction from their view of
human nature will tend to restrict their research focus to external sources of stimulation
and control. But although such external influences do have a powerful effect upon human
behaviour, it has been argued that the limitation of research to external influences leads
to an incomplete and deformed psychological theory (Bandura 1977).
5.1 Dissatisfaction with Conditioning Theories

In recent years there has been a noticeable upsurge of interest in social and cognitive
explanations of human behaviour. Kanfer and Phillips (1970) regard this as a reflection
of several factors, including an increasing dissatisfaction with learning theories based
upon the S-R conditioning model. The human capacity for speech and thought, for
instance, is said to have effects upon behaviour which by-pass simple conditioning, and
S-R theories have been particularly criticised for their inability to explain the acquisition
of novel responses, language, thought and social development.
Pavlov provided the first conditioning interpretation of verbal behaviour, using the
concept of a second signalling system which bridged the gap between perceptual and
symbolic stimuli, and more recently, others have maintained this view of speech as a set
of conditioned responses acquired by the child as a result of his social relationships (Luria
1961). By means of speech and thought, the child enlarges his experience and acquires
new modes of behaviour and new ways of actively organising his mental activities; a vital
feature of speech (and thought) is that it enables man to go beyond the bounds of his
physical capacities. But despite the efforts of Luria and others (e. g. Osgood 1953;
Berlyne 1954) to analyse speech, thought and knowledge in S-R terms, the cognitive and
social theorists have usually rejected this approach. There are, for instance, many problems involved in a linguistic analysis which uses words as its basic units, since communication is typically based upon sentences rather than isolated word-units, and the meaning of
a sentence is not the same as that of the summated meanings of the component words.
There is further dissatisfaction with the lack of any clear and universally accepted
definition of such central concepts as stimulus and response, and this creates an area of
confusion in the arguments between S-R theories and cognitive theories. Wolpe's (1967)
72
Social and Cognitive Theories
suggestion that abstractions such as 'the passage of time' can function as stimuli is
difficult to reconcile with the traditional behavioural emphasis upon external and observable stimuli, responses and reinforcement contingencies. The use of such terms to refer to
vague and abstract features of the naturalistic setting outside the laboratory often presents considerable semantic difficulties, and Mahoney et al. (1974) argue that the terminology of conditioning is often used metaphorically in these cases. In Wolpe's attempt
to explain pervasive anxiety by reference to the passage of time as a CS, it is easier to
accept the cognitive interpretation that the human organism responds primarily to cognitive representations of its environment rather than to the environment itself (Mahoney
1977).
Traditionally, conditioning theorists have suggested that conditioning is an automatic
and unconscious process (Thorndike 1932; Dollard and Miller 1950; Wolpe 1978).
Brewer (1974) has strongly challenged this view and argued that conditioning (in human
subjects) is the result of higher mental processes and not vice versa. Similarly, Bandura
(1969, 1974) suggested that in the classical conditioning paradigm the CS comes to
produce emotional responses not because it has been invested with emotional properties,
but because it elicits emotionally arousing thoughts. Conditioned reactions are largely
self-activated on the basis of learnt expectations rather than automatically evoked. The
emotional response is cognitively mediated rather than an automatic response to the CS,
and from this viewpoint the performance of responses that have been seen to be punished
in others can produce anticipatory self-arousal without requiring any emotional response
to have been conditioned initially to the CS.
Brewer (1974) asserted that the events in conditioning are a result of the subject's
becoming aware of the CS-UCS relationship (in classical conditioning) or of the reinforcement contingency in operant conditioning. Recent investigations have confirmed
the early findings of Cook and Harris (1937) that instructions alone are sufficient to elicit
and to extinguish the galvanic skin response. Katz et al. (1971) found that simply informing their subjects of the CS-UCS relationship established a conditioned response just as
efficiently as actual exposure to CS-UCS parings. Subjects also showed extinction of
GSRs within two trials after being told that no further shocks would occur, and the
authors regard these results as indicative of cognitive mediation processes.
Conditioning theory and cognitive theory also make opposed predictions about the
outcome of experiments in which the subject is exposed to a standard conditioning
procedure, but is deceived as to the nature of the CS-UCS relationship or the reinforcement contingency. Brewer (1974) cites five GSR studies which have used this 'masking'
technique, and in each case the results show that there is no conditioning when the
subject is unaware of the CS-UCS relationship.
Humphreys' (1939) classic study showed that intermittent CS-UCS pairings produced
a conditioned response which was more resistant to extinction than the 100% pairings,
and this result was explained in terms of expectancies; i. e. the intermittent reinforcement
group took longer to become aware of the changed CS-UCS relationship during the
extinction phase. On the other hand, Katz et al. (1971) found that subjects who had been
exposed to variable levels of shock took very much longer than a constant-level 100%
shock group to extinguish, even though both groups had been clearly informed that no
more shocks would occur. The expectancy hypothesis is faced with some difficulties in
explaining this result, and the cognitive position has not been immune from counterattack by S-R theorists.
Dissatisfaction with Conditioning Theories
73
Wolpe (1978) denied that cognitive variables function within a realm of activity independent from other sorts of stimuli. In Wolpe's view, cognition is ruled by the same
lawfulness as other forms of behaviour, and his concept of imagery relies upon conditioning. The image is a conditioned perception, and thoughts are responses whether they are
perceptions or imaginings. As such, they can be conditioned to other thoughts and to
responses in other categories. Wolpe therefore includes cognitive variables within his
system, but reclassifies them within a stimulus-response framework. In direct contrast to
Brewer, Wolpe, who sees cognitive responses as a subset of behavioural responses,
asserted that 'learning takes place automatically' (Wolpe 1978), and that neurotic anxiety is a response to definable aspects of the environment (Wolpe 1967b). Beech and
Perigault (1974) use a similar sort of argument in their discussion of obsessional-compulsive neurosis, where they suggest that recurrent, intrusive and anxiety-evoking thoughts
are usually elicited by environmental stimuli. The mental events have themselves
acquired the properties of phobic stimuli in the same way as external stimuli (i. e. through
conditioning).
With regard to the nature of psychological determinants, in Lewin's (1935) wellknown formulation, behaviour was described as a function of personal and environmental variables: B = f (p. E). Bandura, however, questioned this on the grounds that personal
and environmental factors do not operate independently; rather they determine each
other. It is largely through their own actions that people produce the environmental
conditions which surround them; and these in tum affect their behaviour. This relationship is described as reciprocal determinism. (The term 'reciprocal' is used here to mean
mutual interaction between events, rather than in the more restricted sense of similar or
opposite counter-reactions.) 'From the social learning perspective, psychological functioning is a continuous reciprocal interaction between the personal, behavioural, and
environmental determinants' (Bandura 1977). In contrast to Lewin's formula, Bandura
represents this as follows:
The relative strength of these factors varies with changing circumstances. At certain
times, environmental factors will exert a powerful influence upon behaviour, whereas at
other times personal factors will have the greatest impact.
The implications of this for the abnormal and clinical psychologist extend to the findings of Rosenhan (1973) about the effects of the social context upon the perception of
'abnormality'. In Rosenhan's study, eight pseudo-patients gained admission to 12
psychiatric hospitals in America by complaining of hearing voices: after admission they
ceased simulating this symptom and behaved normally. All were diagnosed as either
having a schizophrenic or a manic-depressive psychosis, and in none of the cases was the
diagnosis reversed. Rosenhan found that the tendency to over-diagnose illness could be
reversed by appropriate instructions, so that the psychiatric staff recognised pseudopatients among the genuine patient group. In Rosenhan's view, the social context and the
meanings attributed to the actions of the person labelled mentally ill are so powerful that
they determine our perceptions of mental illness. In contrast to intrapsychic interpreta-
74
tions of neurosis, the social perspective extends the concept beyond the individual into
his environment. It is recognised that the psychologist, psychiatrist and other members of
staff playa large part in determining the regularities which are supposedly seen in mental
illness (Ullmann and Krasner 1967).
The neurotic disorders cannot be reduced to objective, natural phenomena. Even if
one were to regard the neuroses as illnesses in the full sense of the term (and few
psychologists would accept this formulation), this remains true. From the psychological
perspective, and particularly from a social and cognitive viewpoint, a neurotic disorder is
best seen not merely as a behavioural response, but as a form of social action.
5.2 Modelling
Watson regarded the neurotic disorders as conditioned emotional responses produced by
Pavlovian conditioning, and in Mowrer's two-stage theory, the initial conditioned fear
response was explained in terms of classical conditioning. Eysenck (1976) has suggested
that fear responses may be acquired by other means, for instance, through modelling or
imitation learning, but S-R theorists have largely excluded these processes from their
detailed analyses.
One of the first descriptions of the acquisition of fear as a result of social observation is
that of Mary Cover Jones (1924), and more recently Marks (1969) described how fears
may be acquired by this form of vicarious learning (variously called social learning, notrial learning, vicarious conditioning, observational learning, modelling and imitation).
Rachman (1977) continued to emphasise the importance of conditioning effects in the
acquisition of fear and anxiety, but acknowledged the influence of two other determinants, both of which are indirect processes. The first, vicarious acquisition, relates to the
work of Bandura and others on modelling. Rachman also suggested that fears may be
acquired through the transmission of information or through instruction. Despite an
almost complete absence of studies on this effect, it is clear that this sort of teaching is an
inherent part of child-rearing. Parents, siblings and friends are all involved in this sort of
interaction with children from their earliest years.
Church (1959) found that rats which had been shocked after observing other animals
being shocked showed higher levels of fear to the pain of others than a control group
which had received the same amount of aversive stimulation but un associated with the
pain responses of other animals. The same result has been obtained with other species,
and Berger (1962) found that human subjects who observed another person being
shocked would themselves show an emotional response to the initially neutral conditioned stimulus. Both Berger and Church discuss their results in terms of conditioning
processes. Berger suggests that the observer's perceptions of the ues and the performer's UCR are cues to the performer's unconditioned emotional response, and that this
unconditioned response as perceived by the observer determines the oberserver's emotional response. This emphasis upon perception implies that the observer's emotional
response may be based upon both correct and incorrect attributions.
Bandura (1971) assigns a more prominent role to cognition in this context. Modelling
processes are said to operate principally through their informative function. The observer
acquires mainly symbolic representations of the modelled events rather than specific S-R
associations, and the observer functions as an active agent who transforms, classifies and
Modelling
75
organises the modelling stimuli into easily remembered schemes. Bandura suggests that
the four components in observational learning are attentional processes, retention,
motor-reproduction processes, and reinforcement and motivational effects. Vicarious
aversive conditioning, for instance, is directly related to the degree of psychological stress
in the observer (Bandura and Rosenthal 1966). There is, however, a point at which the
level of arousal becomes so high that emotionality and vicarious conditioning become
inversely related, and Bandura and Rosenthal interpreted their results in terms of an
inverted-U relationship between arousal and vicarious conditioning.
Other investigations have shown that behavioural inhibitions can be acquired through
observing aversive stimuli administered to performing subjects. In one study (Bandura
1965), a group of boys and girls (aged 3 11z to 6 years) watched a film of a model being
aggressive towards a doll. In the model-rewarded condition, a second person complimented the model upon his behaviour and gave him such rewards as sweets and soft
drinks. In the model-punished condition, the reinforcing agent strongly challenged the
model's aggressive behaviour and physically punished him. The results showed that
observing the reinforcing consequences for the model had a significant effect upon the
behaviour that the children produced spontaneously after watching the film. There was
also a highly significant difference between boys and girls. The boys developed more
imitative aggression than the girls after exposure to an aggressive male model, though
when offered a positive incentive to enact the aggressive behaviour, this sex difference
practically disappeared. Bandura suggests that the sex differences in overt aggressive
behaviour may primarily reflect a willingness to exhibit aggression, rather than any
learning deficit. These findings support the view that responses may be inhibited or
disinhibited as a result of observing the consequences of another person's behaviour.
There is evidence that phobic fears can be passed on from parents to children (Bandura
1969; Marks 1969). One young woman whose fear of spiders presented a marked social
handicap appeared to have learned this from her family, in which a grandmother, mother,
aunt and sister were all extremely afraid of spiders. The patient remembered that there
was a bell in the house, and whenever a spider was seen the alarm was sounded and
everyone ran screaming from the house.
In a study of 100 children with neurotic disorders, Britton (1969) looked at the
incidence and severity of psychiatric disorders in their mothers. The children were allocated to three major groups according to their symptoms - neurotic disorders (anxiety,
phobias, compulsions and mood disturbances), conduct disorders (lying, stealing etc.) and
mixed disorders, in which both sorts of disturbance were present. Britton found no
relationship between the presence or severity of psychiatric disorder in the mother and
the type of disorder in the child. Only one symptom - separation anxiety - was related to
the degree of maternal psychopathology. However, it was found that the children of
neurotic and depressed mothers were most likely to show predominantly neurotic symptoms or combined neurotic and conduct disorders. Sixteen of the 21 children with neurotic or depressed mothers showed high levels of anxiety, whereas pathological anxiety
was unusual in the children of mothers with personality disorders. There was also a
significantly higher incidence of phobias and mood disorders among the children of
neurotic and depressed mothers.
There is some disagreement between S-R theorists and social-cognitive theorists about
the nature of the neurotic disorders and about anxiety itself. Although Wolpe (1970) has
largely equated anxiety with autonomic arousal, this is a restrictive use of the concept.
76
Anxiety is best seen as a multidimensional concept which denotes a complex pattern of

responses characterised by subjective apprehension, behavioural avoidance and autonomic arousal. These three components show some degree of independence (Rachman
and Hodgson 1974), and a number of hypotheses relating to the notion of synchrony and
desynchrony between the experiential, biological and behavioural components have been
proposed by Hodgson and Rachman (1974); in particular they suggested that synchrony
or desynchrony between the response systems is a function of the intensity of emotional
arousal, level of demand, therapeutic technique and the specific physiological system
which is being measured. Sartory et al. (1977) looked at these hypotheses and found
some empirical support for them, though their results were in part inconclusive.
Borkovec (1976) also considered anxiety to involve any or all of these three separate
but interacting response components, and there are nine possible relationships between
them. Borkovec's analysis is interesting in that it specifies the sort of relationship which
exists between the cognitive and physiological components of anxiety. When the immediate anxiety response involves a strong physiological component, cognitive factors will be
relatively ineffective: simple manipulations of the cognitive and behavioural components
will be more effective in those cases in which there is only a weak physiological component. However, Borkovec-has pointed out how cognitive avoidance responses can maintain the physiological components of anxiety, and how many different cognitive manipulations (e. g. false feedback, misattribution, demand/expectancy effects and attention
focusing) contribute towards the maintenance of anxiety.
5.3 Life Events and Neurosis

Psychological theories of neurosis, and particularly the conditioning theories, have
tended to concentrate rather exclusively upon the individual and his specific learning
experiences, to the neglect of the social circumstances in which the individual lives. Yet
there is a well-documented relationship, for instance, between social class and the diagnosis of psychiatric disorder. In a series of studies in America, Hollingshead and Redlich
(1958) found an inverse relationship between social class and the prevalence of psychosis
(or the prevalence of diagnoses of psychosis, if one wishes to maintain an attitude of
scepticism towards the validity of psychiatric diagnosis). The lower the social class, the
higher was the prevalence of psychotic disorders; Hollingshead and Redlich found that
the incidence of psychosis in social class 5 was three times higher than in social class 4.
These authors also reported a strong direct relationship between class status and the
distribution of treated neuroses in the population. The effect of social class varied for
each diagnostic group. 'Character neuroses' were most likely to be diagnosed in class 1
and 2 patients; and for depressive neuroses the rate declined almost in a straight line
from class 1 to class 5. Antisocial reactions were most prevalent among class 3 patients,
and hysterical reactions, unlike the other neurotic disorders, showed an inverse relationship with class position (being more prominent in class 5 patients). Hollingshead and
Redlich described their concept of neurosis as 'a condition of subjective malaise and
disturbed social interaction'. In addition, they regarded neurosis not only as a state of
mind of the sufferer, but also as a result of the patient's interaction with the therapist, and
suggested that the class positions of both therapist and patient may affect the resultant
diagnosis.
Life Events and Neurosis
77
It is likely that an important feature of this interaction is the patient's attitudes toward
the 'neurotic' label. He may accept or reject this attribution according to his own or other
people's appraisal of his behaviour and personality, and the criteria for assumption of the
neurotic role probably differ from class to class. Hollingshead and Redlich's data, for
instance, suggest that there are differences between the readiness of the various social
classes to accept the 'patient' self-concept which depend on (a) degree of discomfort, (b)
whether one's body hurts or functions poorly, (c) being unable to work properly, (d)
interpersonal difficulties and (e) difficulties with the law. In this way, a neurotic disorder
is an important part of each person's definition of his class position in society and
expresses an attitude towards the way he expects to be able to function.
There is also a sizeable body of research which has been concerned with the relationship between socially stressful life events and mental health. The concept of social stress
is rather vague, and Coates et al. (1976) have described four different ways in which the
term is used:
1) Stress as background; this includes the sort of living conditions and social supports
that are available to an individual.
2) Stress as the pace of life, including the pressure and intensity of daily life. When stress
is interpreted in these terms, life events are significantly related to high levels of
anxiety, concern over one's health and personal distress (Coates et al. 1969). Other
definitions of social stress have included 3) Stress as changes in the pace of life.
4) Stress as acceleration in the rate of living.
These sorts of stress also have some effect on mental health. It is well known that
emigration, bereavement, personal illness and natural disasters can have an adverse
effect on the mental health of some individuals.
One of the more sophisticated studies of the effects of social stress and life events upon
psychological health is that of Brown and Harris (1978). In contrast to Hollingshead and
Redlich's study, their own investigation of depression in women showed that in London,
psychiatric disorder, and depression in particular, was more common amongst working
class women. Brown and Harris emphasised the view of depression as a social phenomenon and, in explaining their findings, used a model similar to that in Fig. 5. This describes
how a depressive reaction results from the interaction of social class, protective and
vulnerability factors and specific stressful events (the provoking agents). The provoking
agents determine when the depression occurs, and the vulnerability factors determine
whether or not these agents will have an effect. In addition, Brown and Harris proposed
that a class of symptom-formation factors (not shown in Fig. 5) also operate to determine
the severity and form of the depressive disorder. A major loss in the past, for instance,
particularly in childhood or adolescence, is seen as the most potent factor influencing the
severity of depression; it was suggested that the loss of a mother before age 11 significantly increased the risk of depression in adulthood, whereas the same loss between ages
11 and 17 had no influence.
Although many studies have cited correlational findings which suggest that social class
and the prevalence of certain psychological disorders may be related in some way, there
has been little explicit discussion of the nature of this relationship. Brown and Harris's
study is valuable because of the way in which it attempts this sort of analysis. As well as
social class, they took into account five 'life stages' related to the age of the women and
the age of their youngest child. Among the working-class subjects, the highest rate of
78
Grief
Specific
hopelessness
Resolution ----<-r-:-A-vo-:-id-:-a-n-c-el
Resolution
of
depression
- - - - L___...l
Depression
F1g.5
psychiatric disorder (the majority of these being cases of depression) was found among
women with a child of under 6 years. Middle-class women showed no differences in terms
of life stage, and there was no class difference in the risk of becoming depressed among
women without children. When they looked at the different types of life events which
happened to their different groups, Brown and Harris found that severe household
events were more likely to occur amongst women with children and that for these events
there was a class difference; working-class women were three times more likely to have at
least one severe household event. Among the events included in this category, Brown
and Harris cite the following: being given a month's notice at a long-standing job,
husband losing his job, son in trouble with police, husband sent to prison, being
threatened with eviction, etc.
It was also found that the difficulties faced by working-class women tended to last
longer than for middle-class subjects. The results showed that among women with children who have had a severe event or a major difficulty, 31 % of the working-class women
developed depression compared with 8% of middle-class women.
Among their depressed subjects, Brown and Harris found that almost all had a severe
event or some other provoking factor. However, only one-fifth of those with such provoking agents broke down. In order to explain this, the authors invoked the notion of
vulnerability and protective factors. One of the most significant protective factors was the
presence of an intimate and confiding relationship with a friend or relative (usually a
husband or boy-friend). Again, it was suggested that working-class women were more
vulnerable than middle-class women; as well as being less likely to have a husband as a
confidant, working-class women were also more likely to have lost a mother and to have
three or more children younger than 14 at home.
5.4 Labelling
Although Brown and Harris are sociologists, their work incorporates a number of important psychological and psychiatric concepts as well as the social and sociological ones.
The S-R theorists have generally defined neurosis in terms of maladaptive behaviour;
Labelling
79
they have not, however, been particularly concerned to explore this aspect of behaviour
in any depth. Indeed, one sociological criticism of behaviour modification is that in
practice it often operates in the same way as the medical model, as an individual system
of mental disorder. Scheff (1966) argued that the techniques of behaviour modification
are oriented towards changing the patient's psychological system rather than the interpersonal or social system of which he is a member, and this tends to isolate the symptom
from its context. Although Scheff's criticism may be appropriate in some cases, it is
probably over-stated. It is not clear, for instance, why it should be necessary for a
behaviour therapist to change the social environment of a patient with fears of contamination and obsessional-compulsive rituals, though in some cases the treatment of obsessions may be facilitated if it is extended into the patient's home environment and involves
the patient's relatives (Marks 1973 a-b). In some cases of neurosis, the disorder may be
an individual one: in other cases, the therapist ignores the social context at his own, and
at his patient's, peril. As illustrations of this one might cite drug addiction, alcoholism,
fetishism, homosexuality, reactive depression and probably many agoraphobic and hysterical disorders. In sociological terms, such behaviours may be regarded as deviance and
as the violation of social norms which leads to negative social sanctions; in these terms an
important function of the social model of mental disorder is to provide a framework
which avoids the questionable assumptions of inherent psychopathology in all cases of
neurosis.
In their assertion that mental illness may be seen as an ascribed social role, the
sociological critics do have some psychological company. Ullmann and Krasner (1969)
have described a social role in the terminology of operant conditioning, as a category of
instrumental acts which are reinforced in a given situation. Specific roles, such as that of
father, lecturer or patient, include those elements of behaviour which are common to
certain specific types of situation, and the individual behaves in the way which is most
likely to produce some reward. The performance of a particular role depends upon the
individual's past experiences and his ability to perform the role. Neurotic behaviour is
maladaptive in its failure to meet the expectations held by the patient and/or by others
for someone in his role.
In an investigation of women receiving state welfare payments, Cole and Lejeune
(1977) found that 45% of their subjects complained of 'nerves', and that defining one's
health as poor was associated with emotional complaints. The authors suggested that
many welfare mothers regarded their welfare status as a kind of personal failure and were
prone to adopt the sick -role as a justification of their failure. Those women who regarded
themselves as inadequate wives and mothers were particularly likely to define their
health as poor. These results demonstrate the important interaction of the individual's
self-perceptions (possibly acting as the protective factors suggested by Brown and Harris)
and their social circumstances.
Few psychologists, and even fewer psychiatrists, would be prepared to go quite so far
as the societal-reactance theorists and assert that the maladaptive behaviour which
causes a person to become labelled as 'neurotic' has little relationship to the personal and
social attributes of the individual. Lemert (1967), however, has suggested that primary
deviance (in this case, the neurotic behaviour itself) 'has only marginal implications for
the psychic structure of the individual.' This is essentially the same position as that of
Scheff (1966), who also regarded mental illness as an ascribed status dependent mainly
upon conditions external to the individual.
80
Others have found a different pattern. Yarrow et al. (1955) looked at the ways in
which wives came to regard their husbands as mentally ill. They found that the wives
showed a marked resistance to interpreting the husband's behaviour as 'ill' rather than
'normal', and it was only when his behaviour exceeded tolerable limits that the wife
sought medical help. Even at this point the wife would often resist seeing the behaviour
as illness. Gove (1970) cites a number of studies which reach this same conclusion - that
the psychological disturbance occurs prior to the process of labelling rather than as a
consequence of it.
According to the societal-reaction theory, the experience of being caught and labelled
as a deviant is a crucial stage in the development of deviant behaviour, and this is largely
related to the person's position in the social structure. The greater an individual's social
and economic resources, the greater his chances of being able to deal successfully with
others and therefore avoid being channelled into a socially deviant role. In the case of the
psychological disorders, this implies that the higher an individual's social status, the more
likely he is to avoid treatment and particularly in-patient treatment. This is broadly in
agreement with Hollingshead and Redlich's (1958) data on psychotic disorders, and the
results of Brown and Harris's (1978) study on depression could also be interpreted in this
light. On the other hand, in an investigation of 258 in-patients which examined the
predictions of societal reaction theory, Gove and Howell (1974) found that the resources
of the individual generally appear to facilitate entrance into psychiatric treatment, including in-patient treatment. This is contrary to what one would expect on the basis of the
societal-reaction account. However, Gove and Howell are careful to point out that their
results do not imply that the processes suggested by the societal-reaction account do not
exist, but that they are not usually the most salient effects.
Gove (1975) has also challenged the suggestion that those individuals who have been
labelled 'mentally ill' have undergone a profound and generally irreversible change.
Rosenhan (1973) referred to the labelling process in order to account for the ways in
which psychiatric patients are perceived by staff. Rosenhan suggested that the psychiatric
label profoundly colours the perception of other people towards the patient, and that the
distortion is often great enough to lead the staff to overlook or misinterpret normal
behaviour. It is true that the public stereotype of mental illness is often inaccurate and
distorted. The psychiatric patient is often regarded with fear, distrust and dislike, and this
negative stereotype of the mentally ill is probably shared by many professionals. But
although this view of the mentally ill is negative and suggestive of serious discrimination,
data on the effects of social stigmatisation after being treated in a psychiatric hospital are
fairly sparse. Gove and Fain (1973) found no evidence of deterioration in the occupational status, social relationships or financial situation of a group of patients after hospitalisation. If anything, their results suggested a slight improvement. Most of the
patients themselves saw their stay in hospital as having been beneficial and felt that their
ability to handle problems had improved: only a small number felt that the stigma
associated with hospitalisation had proved a serious problem for them. This study suggests that when dealing with someone who has been mentally ill, most people do not
seriously discriminate against them. Although labelling processes may have some effect,
societal reaction theorists have over-emphasised the importance of such factors in the
explanation of psychological disorder, and parts of the societal-reaction account of mental illness seem to be substantially incorrect. However, its greatest value to psychological
medicine may be that it draws attention to a much neglected facet of psychological
Attribution
81
disorders and helps to correct the predominantly individualistic bias of most psychologists and psychiatrists.
5.5 Attribution
One area of research in social and cognitive psychology that is directly relevant to any
theory of neurosis is that concerning attribution. The term 'interpersonal attribution'
refers to the process or processes by which we come to attribute various dispositions,
motives, intentions, abilities and responsibilities to one another (and indeed to one's own
behaviour). Attributions are a sort of explanation of why someone acted as he did and
assume that people are active agents, capable, at least in principle, of free and intentional
action and of determining their own behaviour (Eiser 1978).
In their classic experiment, Schachter and Singer (1962) showed how an individual's
emotions were dependent both upon physiological arousal and upon environmental cues
which might account for the arousal. After receiving an injection of adrenalin or a
placebo injection of saline solution, one-third of the subjects were correctly informed of
the drug effects, one-third were incorrectly informed, and the third group received no
information. When the subjects then encountered another person who acted in an angry
or euphoric manner, the subjects who had received the injection of adrenalin and who
had no clear explanation as to their own arousal were most likely to be influenced by
their social circumstances. They tended to behave in a similar manner to the experimental confederate. The subjects who had been correctly informed of the reasons for their
aroused state were able to attribute this to the drug effect and were uninfluenced by the
emotional behaviour of the actor. Subsequent research showed that a person's emotionality could be decreased by falsely attributing an emotional state to a non-emotional
source (Nisbett and Schachter 1966).
The implications of this research suggest that emotions, and specifically pathological
emotional states, may be significantly affected by the attributions that a person makes
about the causes of his neurotic symptomatology. Storms and Nisbett (1970) looked at
the way in which attribution processes could affect the emotional behaviour of insomniacs. Subjects were given placebo tablets; one group was told that the drugs would lead
to insomniac effects, i. e. that they would cause alertness, increased heart" rate and
increased body temperature. The other group was informed that their drugs were sedatives. The authors hypothesised that the subjects who had received the 'stimulant' drug
would be able to reattribute their sleeplessness to the drug and therefore would experience less emotional arousal and fall asleep more quickly. The other group, having
received a supposedly sedative drug, might have been expected to attribute their sleeplessness to personal causes and to experience greater insomnia. (This is, incidentally,
quite the opposite prediction to that made on the basis of a 'suggestibility' model.) The
results showed that subjects' attributions about their symptoms influenced the subsequent intensity of their emotional states. Those subjects who had taken 'stimulants'
reported a decrease in the time that it took them to fall asleep, whereas the 'sedative'
group took longer.
Storms and Nisbett also interviewed the subjects after the study and found that the
subjects' accounts of their insomnia were consistent with an attributional model. It was
notable that each of the subjects had adopted the label of 'insomniac' and believed that
82
this carried with it the further implications of some special psychological condition. Many
also described their difficulties in falling asleep in the following terms. After going to bed
they would begin to look for signs of insomnia, such as feeling too hot or being easily
disturbed and upset by noise; having noticed these, they would feel anxious and begin to
imagine the consequences the next day of another sleepless night. This would increase
their anxieties and keep them awake, perhaps speculating on the underlying causes of
their difficulties, which would in turn increase their level of anxiety even further.
This description has close parallels to the anxiety-feedback cycle which was described
by Eysenck (1975b) as a possible cause of such disorders as impotence and insomnia,
though the explanation offered by Eysenck is quite different. Storms and Nisbett use a
more explicitly cognitive and attributional framework to account for their results, the key
elements being self-attributions of pathology or inadequacy, and increased anxiety. Like
Eysenck, they too suggest that their analysis may apply to such disorders as male impotence, stuttering, extreme shyness and excessive physical awkwardness. Valins and Nisbett (1971) have extended this analysis to other neurotic disorders, such as depression,
homosexual panic and phobic disorders. In many of these cases the person becomes
aware of their own problems at the symptomatic level before seeking professional advice
from a therapist. In doing so, they often reach self-diagnoses of abnormality or inadequacy that can set up the vicious cycle of increased anxiety and increased difficulties.
Such self-attributions are probably found in most neurotic patients and may present a
severe obstacle to therapeutic change.
Storms and McCaul (1976) discussed the evidence for this two-stage process. Their
review suggested that the level of experienced anxiety is directly related both to the
individual's attributions of control over his environment and to situational stress factors.
To the extent that he feels helpless or out of control, he is likely to experience considerable anxiety, and these self-cognitions can interact with external factors to produce
magnified increases in anxiety. In insomnia and male impotence, there is evidence that
excessive arousal of the sympathetic nervous system is directly related to the problem
(e. g. Monroe 1967; Masters and Johnson 1970). In other cases, however, the relationship between the anxiety and the disorder is likely to be cognitively mediated.
In an investigation of the effects of attributional processes upon stuttering, Storms and
McCaul (1976) found that normal speakers who were led to make a dispositional attribution for their speech faults (pauses, repetitions, stammers, etc.) found it more difficult to
perform a verbal task under conditions of stress than other subjects who were given a
situational explanation of their speech faults. In this case, the results support the suggestion that dispositional attributions interact with higher levels of stress to produce an
exacerbation of stammering. Negative self-attributions may also play an important role in
producing the more severe speech difficulties that lead individuals to seek therapeutic
assistance (Storms and McCaul 1976).
In therapy, it has been suggested that individuals who have developed dispositional
explanations of their behaviour may be helped if the therapist helps them to substitute
situational attributions. This distinction between intrinsic and extrinsic causes of
behaviour is of fundamental importance to the application of attribution theory to clinical
practice. Valins and Nisbett (1972) discussed some of the ways in which attribution
processes effect the development and treatment of emotional disorders, but as a comparatively recent research topic, there ic clearly a need for further investigations of what
appears to be a promising area for clinical research.
Cognitive Accounts of Depression
83
5.6 Cognitive Accounts of Depression

One of the unfortunate consequences of the medical model is that psychological disorders have tended to be seen as distinct and different from normal forms of behaviour. So
whereas social psychology has investigated the relationship between cognition and emotion (e. g. Schachter and Singer 1962), psychiatry has frequently preserved the concept of
depression as a primary affective disorder in which cognitive factors play little part. Any
cognitive abnormalities in depressive disorders have usually been seen as the inevitable
result of affective distortions, possibly due to the disorganisation that can follow intense
emotions.
Beck (1971) pointed out how this notion of the primacy of affect over cognition in
psychopathology is contrary to that which applies for normal subjects. Here, the way a
person sees his situation determines his affective state, and Beck has argued that the
direction of the reaction is exactly the same in the depressive disorders. The difference
between the normal and abnormal reactions is to be found in the degree of correspondence between the cognition and reality. In the abnormal condition, there is a distorted
cognitive representation of reality, and in the case of the depressed patient, the cognitions show a systematic error (biased against the person's own interests).
The thought content of neurotic depressive patients often centres around what Beck
calls 'the cognitive triad' - a negative self-image, a negative view of life's experiences and
a nihilistic view of the future. The typical emotions of depression, sadness, disappointment and apathy follow from the sense of loss, negative expectations and helplessness
(Beck 1971). A similar discussion has been offered by Melges and Bowlby (1969), who
stressed the central role played by attitudes in many psychological disorders and in
particular by feelings of hopelessness and the anticipation of failure. Melges and Bowlby
suggested that a key process in depression is the belief that available plans of action can
no longer achieve established goals.
Seligman has also described a theory of depression based upon the notion of learned
helplessness, in which cognitions of helplessness operate as the basic cause of depression
(Seligman 1975). Seligman's account is an important development in the psychology of
depression, since it is the first full-scale treatment of the issue which starts from an
experimental rather a clinical point of view. Although both are cognitive accounts of
depression, Seligman's theory differs from that of Beck in its denial of the value of
distinguishing between cognition and emotion in this context. In depression there is an
interdependence of feelings and thought; a person does not feel depressed without
depressing thoughts, nor does he have depressing thoughts without feeling depressed.
Seligman pointed out how the subject in a Pavlovian conditioning experiment is helpless; his responses have no effect upon the es and ues, whereas in operant conditioning,
the subject's voluntary response controls the outcome. Overmier and Seligman (1967)
found that dogs which had been exposed to inescapable shock in a Pavlovian harness
were unable to learn an escape/avoidance response in a shuttle box as effectively as
animals without this experience of inescapable shock. Typically, the animals exposed to
uncontrollable shock did not escape from the shock in the shuttle box, but passively
accepted the shock. When the experimental animals had first received escapable shock in
the shuttle box, then inescapable shock in the harness, they did not react passively to
shock when they were again placed in the shuttle box, as did animals which had first
received inescapable shock in the shuttle box (Seligman and Maier 1967). Seligman
84
suggests that when an organism has had an initial experience of uncontrollable trauma, its
motivation to respond to later trauma is likely to wane.
In a study with rats, Seligman and Meyer (1970) found that animals which received
unpredictable high-intensity shocks when bar-pressing for food reduced their response
levels to about 10% of the pre-shock rate, and that this suppression appeared to be
permanent. At the end of a 70-day period, no recovery was evident. Animals which were
exposed to low levels of shock, or to predictable shock, did not show this same suppression. It was also found in this experiment that stomach and intestinal ulceration were
significantly more likely to occur in the fearful (response-suppression) animals. Seligman
has used a safety-signal hypothesis to explain these results - that after experiences of
trauma, an organism will remain in fear unless it is in the presence of a signal which
reliably predicts safety. In this formulation, fear is a joint function of the intensity of
shock and the probability that shock will occur.
Seligman's results with respect to stomach ulcers and uncontrollability are quite different from those of Brady. In what became a classic experiment, Brady et al. (1958)
divided a group of monkeys into pairs. Each animal received a shock unless one of the
pair, the 'executive' monkey, pressed a lever which delayed the shock for both animals.
After a period of weeks, several 'executive' animals died, and analysis showed extensive
gastro-intestinal lesions with ulceration in the experimental animals but not in the controls. These results were interpreted to show that it was the decision-making animals who
developed the psychosomatic disorders, and not the animals who received uncontrollable
shock. However, there is a serious methodological drawback to Brady's study. This
concerns selection of animals for the experimental and control conditions. Brady's
animals were not randomly assigned to the 'executive' and helplessness conditions.
During the initial session, those animals which first showed avoidance lever-pressing to
delay the shock were chosen to be the 'executive' animal. It is possible that these animals learned avoidance responses more quickly because they were more emotional,
and therefore more susceptible to stomach ulcers, in which case, Brady's result would
be an artifact of his experimental design - a possibility which is admitted by the authors.
Seligman's helplessness theory of depression is intended to account for three facets of
the disorder; disturbances of motivation, emotion and cognition. In its simplest form, it
suggests that when an organism is faced by an outcome which is independent of its
responses, as for instance in classical conditioning, it learns that the outcome is independent of the responses made (Seligman 1975). Put in these terms, this seems so obvious as
to be hardly worth stating. However, a non-cognitive learning theory faces some problems in accounting for learnt expectancies about the probability of any given outcome.
Seligman's theory starts with information about the response-outcome contingency, then
moves to a cognitive representation of the contingency and finally to the behavioural
response. When the outcome is independent of any response the person makes, Seligman
predicts that this will establish a belief or expectation about the inefficacy of responding
and difficulty in learning that success can follow responding. When the uncontrollable
outcome is also traumatic, such conditions will produce heightened anxiety followed by
depression.
There have been objections to the learned helplessness model on both empirical and
theoretical grounds, and Abramson et al. (1978) proposed a reformulated account that
incorporates the individual's attributions. The person first learns that his responses and
Social Anxiety and Social Competence
85
certain outcomes are independent and then makes an attribution about the cause of these
events.
In one of the Glass and Singer studies (1972), subjects were exposed to a loud noise.
Those subjects who were able to tum off this noise performed better in a problem-solving
task than others who were unable to avoid the noise. However, another group of subjects
was told that they could tum the noise off if they wanted to, but that the experimenters
preferred that they did not do so. In fact, these subjects were unable to stop the noise, but
believed that they could control it. Under these conditions, they performed as well as the
subjects who actually did tum off the noise. Actual controllability and actual uncontrollability can produce identical expectations; it is the cognitive representation of the world
that is the crucial determinant of helplessness. It is also possible (and clinically, it may be
quite common) for a person to believe incorrectly that environmental events are uncontrollable when they could be controlled. A person may for several reasons be unable to
find an appropriate coping response for a particular situation and therefore believe that
he is unable to control events.
But despite the introduction of attribution into the model, it remains vulnerable. There
are a number of conceptual difficulties still inherent in the theory. Buchwald et al. (1978)
criticised the indiscriminate use of the notion of learned helplessness to refer to observed interference with performance, to specific deficits that might have caused the
interference, and to the belief that responses and their outcomes are independent.
When this ambiguous usage is linked to the complex and heterogeneous collection
of signs and symptoms that comprise clinical depression, the possibilities of confusion
multiply.
On empirical grounds there are other difficulties. Of the three experiments of Willis
and Blaney (1978), none gave clear support to the predictions made by the helplessness
model. Other studies found that some subjects react to conditions of uncontrollability by
renewed determination and effort. This apparently paradoxical facilitation effect was
reported by Hanusa and Schulz (1977), and by Wortman et al. (1976). Indeed, these
studies found the facilitation effect in those conditions that the revised helplessness
model predicts would lead to the strongest deficits (i. e. where subjects were led to
attribute their failure to their own lack of ability). Wortman and Dintzer (1978) point out
that unless it is possible to specify in advance that certain attributions will be followed by
specific consequences, the model falls into circularity.
5.7 Social Anxiety and Social Competence

Neither the ICD-8 nor DSM-II classification systems differentiate between the varieties
of phobic disorder that occur. Marks (1970), however, distinguished between
agoraphobic disorders, social phobias, situational and animal phobias, and the revised
American system DSM-III is to include social phobias within the general category of
anxiety disorders. Social phobias are said to be characterised by a fear of situations in
which the person is exposed to possible scrutiny by others and in which he might behave
badly. The person with a social phobia is likely to show marked anticipatory anxiety if
required to enter such situations and to avoid them. The diagnostic criteria for social
phobia are primarily the avoidance of specific social situations, and the distress which is a
result of that avoidance.
86
One of the difficulties with this concept of social phobia is that the phobic disorders are
essentially linked to particular objects or events, whereas in this case the social fears are
often so generalised that it is difficult to specify what is acting as the socially phobic
stimulus. To this extent, the term 'social phobia' may be misleading, and in the present
discussion the term 'social anxiety' is preferred as a descriptive label, since it contains
less surplus meaning. Of course the term 'phobia' itself varies in its meaning: Eysenck and Rachman (1965) have used it as if it were synonymous with conditioned
emotional response, whereas in Freudian theory it implies displacement. In Marks'
(1970) view, phobias are intense fears which are out of proportion to the apparent
stimulus, which cannot be reasoned away, and which lead to avoidance of the feared
stimulus.
In one discussion of severe social anxiety (Nichols 1974), 12 separate aspects of this
disorder were described. The first was a sensitivity to, and fearfulness of, disapproval and
criticism. It is an important feature of Zajonc's (1965) notion of social facilitation that
the presence of others, in the role of observers or of co-participants, leads to a state of
generalised arousal. More recent research suggests that the presence of others is, in itself,
largely irrelevant; it is the possibility of being evaluated by them that produces the
apprehension (Cottrell et al. 1968). This effect is most marked when the audience is seen
as likely to be destructively critical and least marked when the audience is seen as a
source of help. In other words, the effect of other people depends largely upon whether
they are seen as safety-signals or threat-signals. As yet there is no clear statement of the
precise conditions under which audiences produce increased or decreased levels of anxiety, though Nichols (1974) suggested that the highly socially anxious person may tend to
see criticism where in reality there is none. Among the other features described by
Nichols are low self-evaluation, anticipatory fears of failure in social situations, rigid
concepts of appropriate social behaviour and fears of being seen by others as behaving
oddly. These cognitive aspects of severe social anxiety have been largely neglected by
psychological accounts of the phobic and neurotic disorders.
The increasing interest in the psychology of interpersonal behaviour (e. g. Argyle
1969) has prompted a greater interest among clinical psychologists in social-training
techniques as a form of therapy, and Mahoney (1974) has described a shift in behaviourtreatment research from a focus upon discrete, situation-specific responses and problemspecific procedures towards a more general coping-skills model. Social incompetence has
usually been equated with the degree of subjective dissatisfaction felt by an individual in
the performance of required social roles, which leads to an avoidance of these roles.
Argyle et al. (1974) suggested that some forms of mental disorder are caused or exacerbated by lack of social competence and can be cured or alleviated by social-skills training
programmes.
Depressed individuals may show such deficits as poor verbal behaviour (of a flat,
passive and expressionless kind), lack of initiative in conversation and little interest in
maintaining friendships or an active social life. Anxious patients, on the other hand, often
show their anxiety in rapid, breathy speech which seems to come out under pressure, by a
tense posture and jerky and poorly co-ordinated gestures. Among the other social-skills
deficits that are sometimes found among individuals with neurotic disorders are disturbed
perceptions of their own and other people's actions, and aggressive and hostile reactions
to others. Patients with more severe disorders, such as schizophrenia, may be correspondingly more disturbed in their social behaviour.
87
Until recently there was comparatively little interest in, or knowledge about, the
elements and processes of social behaviour, and even less was known about how these
might fail. The social-skills model proposed by Trower et al. (1978) assumes that skills
are acquired through various forms of learning, such as imitation, reinforcement and
instruction, and through exposure to skilled models, and the authors include the following non-verbal elements in their discussion of social skills: facial expression, bodily
gestures, gaze, spacial behaviour, non-verbal aspects of speech, bodily contact and physical appearance. The social functions of speech include giving instructions or directions
which are intended to influence the behaviour of others directly, asking questions either
to elicit an appropriate reply or to initiate an encounter, and offering comments or
suggestions. Speech is also used to express emotions or attitudes, and it is used in
informal chat in which little information is exchanged. Some verbal behaviour has a
formal function, such as giving speeches or engaging in a socially routine interaction
(greetings, farewells, thanks, etc.), and it is also possible to look at speech in terms of the
latent and implicit messages which it contains.
Social-skills training is, necessarily, a complex package of interventions which is likely
to vary from therapist to therapist, and even from occasion to occasion. In its simplest
form, the training might begin with an assessment procedure designed to identify the
specific types of social difficulties faced by the individual, who is then encouraged gradually to tryout unaccustomed forms of behaviour. This often begins with careful observation of the circumstances of particular social situations, cautious interaction in the role, at
first of a listener and then in a more active participation as a speaker. With increasing
confidence the client is encouraged to try out more ambitious interactions by the use of
assertive and rewarding strategies.
Because of the complexity both of the training procedures and of the social difficulties
being treated, there have been considerable difficulties in conducting adequate evaluative research studies of these sorts of intervention. Wolpe (1958) reported that assertiveness training was an effective treatment for a group of out-patients with neurotic difficulties, and Hallam (1974) used social-skills training together with an extinction schedule to
eliminate the performance of obsessional rituals. In a study of 20 patients, Trower et al.
(1978) found that after ten individual sessions there were significant reductions in
symptoms of anxiety, depression and social anxiety: the authors also suggested that
social-skills training was superior to desensitisation procedures in dealing with specific,
rather than general, difficulties. Marzillier (1978) concluded that skills training can
improve behavioural skills, that the effects of training can generalise to real life and
can be of lasting benefit, and that such training programmes are of clinical benefit to the
patient.
5.8 Further Implications for Therapy

Meichenbaum (1976, 1977) has adopted a therapeutic policy which is based both on the
cognitive work of Beck, Kelley, Singer and others, and upon more conventional
behaviour-therapy research. This cognitive-behaviour modification assumes that any distinction between a purely behavioural versus a cognitive therapeutic programme is misleading and mistaken, and it has been suggested that the common basis of therapeutic
88
change, regardless of the persuasions of the therapist, lies in the alteration of the internal
dialogue in which the person engages (Meichenbaum 1976).
It is one of Meichenbaum's criticisms of clinical research that not enough attention has
been paid to what happens in therapy prior to the beginning of specific treatments (e. g. a
behaviour-modification programme). In some ways this is a similar complaint to that of
Ramsay (1975), who discussed the question of the client's motivation for treatment. In a
study of individuals who suffered from phobic disorders, Ramsay found that only a low
percentage chose to enter therapy for their problems, and of these, many were probably
only marginally motivated. In his actual clinical practise, Ramsay pointed out that despite
the effectiveness of behaviour therapy as a treatment for phobias, much of his time was
spent not in using the techniques themselves, but in bolstering the motivation of the
patient to stay in treatment so that the techniques could be given a chance to work. This
issue of the patient's motivation to co-operate with the therapist is of crucial importance
in most areas of psychological medicine. Clinicians will recognise the problem from their
own experience, yet it has received hardly any mention in the literature, and as a concept,
it has often been misused. In the treatment of drug dependence, for instance, therapeutic
failures have been excused on the grounds of the patient's lack of motivation rather than
the inadequacy of the treatment. But despite the dangers of tautologous thinking
involved in this misuse of the term, motivation for treatment can be rescued from circularity if it is related to the patient's perceptions and expectations prior to treatment
(Gossop 1978c).
One theme of cognitive-behaviour modification is that the original S-R conceptualisation of behaviour therapy over-emphasised the importance of environmental events
(antecedents and consequences), to the neglect of the individual's perceptions and evaluations of these events. Meichenbaum and Cameron (1974) reviewed the empirical evidence relating to this proposition; they concluded that when standard behaviour-therapy
procedures were augmented with self-instructional components designed to help modify
the client's cognitions, the treatment effects were more powerful, showed greater
generalisation and lasted longer.
Meichenbaum's (1976) discussion of cognitive factors in behaviour-modification programmes examined the role of the client's expectations in therapy. It is clearly important
that both client and therapist should have a common view of what they are doing. In this
conceptualisation phase of treatment, an attempt is made to have the client change his
perceptions, attributions and sense of control or helplessness about his presenting difficulties. It has been suggested that faulty self-evaluations are at the root of many adult
neurotic disorders. One of the best-known statements of this position is Albert Ellis'
rational emotive therapy (1957, 1962). Ellis proposed that a major core of emotional
disturbances is a result of the individual's preoccupation with what others think of him.
The aims of the therapist are centred upon making the client aware of his negative selfstatements and of the anxiety-provoking and self-defeating nature of this sort of thinking.
Meichenbaum (1976) has also emphasised the therapeutic significance of the patient's
internal dialogue about his own difficulties and has attempted to modify this internal
dialogue. Meichenbaum has described this as a three-stage process. In stage 1 the client is
taught to become an observer of his own behaviour; this is then used as a stimulus for the
person to emit different cognitions and behaviours; finally, the internal dialogue that the
person has about his newly acquired behaviour determines whether or not the change will
be maintained and will generalise. Both Ellis and Meichenbaum are primarily interested
89
in the therapeutic implications of this position, and it has been suggested that whether or
not the patient actually engaged in this sort of internal dialogue prior to therapy is less
important than his willingness to view his behaviour as if it were affected by his selfstatements and modifiable by them (Meichenbaum 1977). To this extent, the self-verbalisation hypothesis is more appropriately regarded as a therapeutic strategy than an explanation of the development of emotional disturbances.
The research into modelling effects has also had an impact upon treatment. It has been
shown that behavioural inhibitions and conditioned emotional responses can be acquired
by observers after watching the actions of another person and the consequences of those
actions for the model (e. g. Bandura 1977; Berger 1962). It has also been suggested that
established avoidance responses can be extinguished vicariously. Bandura et al. (1967)
found that children who had observed a confident model gradually perform increasingly
fear-provoking acts showed a marked and persistent reduction in their own avoidance
behaviour.
Rachman (1972) looked at the clinical potential of the research findings on observationallearning and concluded that the fear/anxiety reducing value of therapeutic modelling has been convincingly demonstrated. Symbolic modelling (in which the observer
watches, but has no direct contact with either the therapist or the situation) produced a
significant and lasting reduction of fear, and participant modelling increased the extent of
the fear reduction. Modelling is not, however, a necessary condition for therapeutic
change, since some methods produce successful results in the absence of any modelling
experiences.
It is not clear, however, what the basis for these results is. One possibility is that the
modelling conveys information to the observer about the probable outcomes of close
interactions with feared stimuli (in this case, dogs). On the other hand, the reduction in
avoidance behaviour could be due to extinction processes following the non-occurrence
of the expected aversive consequences. Bandura (1969, 1977) suggested that the information derived from observing a model can be stored and used later as symbolic cues to
overt behaviour.
Rachman (1972) suggested that modelling, desensitisation and flooding may each be
based upon some common underlying process, possibly extinction, since in each form of
treatment, the patient is exposed for extended periods of time to the fear-provoking
stimuli. It is also possible that habituation of reinforcement effects could underlie these
procedures, though it is difficult to explain the effectiveness of symbolic modelling in
terms of reinforcement effects, since no response is made during the modelling period
which could be reinforced.
The theoretical basis for the effects of modelling are not clear. Nonetheless, the practical effectiveness of modelling procedures has established their place in the clinical work
of many therapists, and it seems likely that modelling will come to play an increasingly
important role in the treatment of a wide range of neurotic and social disorders.
The therapeutic implications of the learned-helplessness model of depression have
been presented by Abramson et al. (1978). The therapist might assist the depressed
patient by helping them: (1) to change the probability of aversive events (e. g. by the use
of social agencies to achieve rehousing or job-placement); (2) to re-evaluate the outcome
itself (e. g. by substituting realistic goals for unrealistic ones); (3) to change their expectation from one of helplessness and uncontrollability to controllability (e. g. through socialskills training); and (4) to change their attributions of success and failure. Abram-
90
son et aI. suggest that the person's attributions may be changed through discussion,
though there is little evidence that cognitions can be changed quite so simply. There are,
however, several studies that suggest that cognitive treatments for depression can be
more effective than either anti-depressant medication (Rush et aI. 1977) or behaviour
therapy (Shaw 1977). In view of the current popularity of this model, further studies can
be confidently expected to extend our understanding of its therapeutic potential.
6. Psycho-analytic Theories of Neurosis
Psycho-analysis - that interesting but peculiar and arcane doctrine: it has received more
publicity and it has had more impact upon child-rearing, psychiatric training, art and
literature than any other psychological theory, yet in any precise form it manages to
remain obscure. Of all the theories presented in this book, it is perhaps the furthest from
common-sense psychology, and yet its impact upon the layman has been so great that it is
often mistakenly thought to constitute the whole of psychology.
It is difficult to provide any complete account of the psycho-analytic theories of
neurosis within a chapter of this length. Even if one equated psycho-analysis with Freud's
own work, it would be impossible to achieve such an aim. Freud's collected works alone
run to 24 volumes in their English translation, and within these works he distinguished
between several different meanings of the concept of neurosis and developed several
different theories to explain the neurotic disorders.
Indeed, because Freud said so many different things at different times as his theory
evolved, it has been suggested that only an historical account of the development of his
ideas can provide a proper framework for understanding Freudian psychology (Nagera
1969).
When the various analytic factions that dissociated themselves (in varying degrees)
from Freud's views are also taken into account, the problem assumes an even more
daunting aspect. For these reasons the present chapter offers a necessarily selective
review of Freud's writings on the neuroses, with emphasis upon certain of the key issues
in his work. The chapter deals mainly with the Freudian system, since this provides a
prototype of the psycho-analytic approach, though a brief description is also offered of
some of the theoretical criticisms and modifications of classical Freudian theory made by
other influential analysts.
6.1 Psychological Determinism

One of the most general aspects of Freud's attempt to understand human psychology was
the tenet that human behaviour is never 'accidental'. Whereas other psychologists might
have believed that the small mistakes of everyday life, dreams, jokes and so forth might
ultimately be dependent on a complex interaction of certain laws of behaviour, they
tended to regard these phenomena as essentially transient, trivial and not worthy of
investigation. Freud on the other hand emphasised that all behaviour was goal-directed,
even though the goals might be hidden (unconscious), and that events of this sort could
have a special significance.
92
In their early studies of hysteria, Breuer and Freud noticed that the peculiar behaviour
of their patient made sense in terms of an event from her past that she had apparently
forgotten. Her symptom was a re-enactment of the past, and when the memory of the
incident which had provoked the symptom was brought into consciousness, that symptom
immediately and permanently disappeared (1893, vol II, P 6).1 This insight quite literally
made sense of the symptom, which had previously appeared to have a certain randomness, and it also pointed to the continuity of personality. Most psycho-dynamic theorists,
and the Freudians in particular, look for the causes of adult neuroses in the anxieties of
childhood and in the progressive sequence of unconscious conflicts and defences against
these feelings.
Psycho-analysis is quite explicit about the secondary role of neurotic symptoms. A
neurosis may be defined as a psychogenic condition in which the symptoms are the
symbolic expression of an underlying intrapsychic conflict whose origins lie in the individual's childhood history (Laplanche and Pontalis 1973). The close resemblance between this psycho-analytic definition and the conventional psychiatric definitions offered
in the nosologies DSM 2 and ICD 8 should be apparent, and this testifies to the profound
impact that psycho-analysis has had upon psychiatry. The specific points of influence
include a distinction between the superficial symptom and the underlying pathological
process, an acceptance of the meaning or symbolic significance of the symptoms and of
their relationship to an unconscious conflict. In view of the suggestion made in Chap. 2
that psychological medicine would have much to gain by separating the descriptive from
the aetiological components of its definitions, it is worth noting that such a suggestion
runs counter to the psycho-analytic tradition. Laplanche and Pontalis (ibid) pointed out
that the task of trying to define and understand what is meant by the concept of neurosis
tends to become indistinguishable from the psycho-analytic theory itself.
Many behaviour therapists have equated the neurotic disorder with the behavioural
symptomatology. Eysenck (1959), for instance, stated that 'there is no neurosis underlying the symptom, but merely the symptom itself'. To the analyst, on the other hand, the
neurosis cannot be equated with symptomatology. A person may be neurotic but show
none of the symptoms of phobia, depression, hysteria and so on. The personality disturbances underlying the disorder are the crucial feature of neurosis, and in contrast to the
views expressed by behaviour therapists, psycho-analysis suggests that 'the cure of a
symptom does not necessarily mean the cure of a neurosis' (Horney 1937). Neurotic
symptoms are like the eruptions from a volcano, while the pathogenic conflict, like the
volcano, is hidden deep within the individual.
6.2 Repression and the Unconscious

The idea of unconscious mental processes is absolutely central to the psycho-analytic
view of psychology. Indeed, it is this emphasis upon the role of unconscious mental
processes as determinants of behaviour, as much as any other single factor, that distinguishes psycho-analysis from other psychological theories.
References to the Standard Edition of Freud's Complete Psychological Works are given in this
form - date of original publication, and volume and page numbers of the Standard Edition.
Repression and the Unconscious
93
The tenn 'unconscious' can be used in several different ways. In its least problematic
sense it can be used descriptively to refer to the contents of the mind that are not present
in the field of consciousness at any given moment. Its other usages are more important;
the unconscious is also a topographical (see below) and a dynamic concept, which was
derived from Freud's clinical work. He became convinced that the mind could not be
equated with consciousness. There seemed to be contents and forces which, although
unconscious, remained active in their effects, and which had a particular significance in
the detennination of neurotic symptoms.
Breuer and Freud, for instance, pointed to the persistence of early memories; though
these memories appeared not to persist at a conscious level, nonetheless they had a
powerful effect upon the individual's behaviour and detennined the symptomatic expression of neurotic disorder - in this case hysteria (1893, vol II). It is an essential feature of
neurotic symptoms that the individual has the experience of something strange and
unintelligible happening to him. It may that he has involuntary movements or strange
changes in his bodily functions and sensations, as in hysteria; or an overwhelming and
inexplicable change of mood, as in the depressive disorders; or the need to act and think
in a bizarre manner, as in the case of the obsessional and compulsive disorders. In each of
these cases the symptoms appear to break into the personality from some unknown or
unconscious source.
Ramsay's (1975) complaints about the comparative lack of attention that behaviour
therapists have devoted to the question of the patient's willingness to accept the requirements of the treatment regime (see Chap. 5) stand in contrast to the work of the analysts
for whom the concept of resistance is of crucial importance. In the Introductory Lectures,
Freud describes the significance that this aspect of the patient's behaviour had upon his
own ideas. 'The patient, who is suffering so much from his symptoms ... who is ready to
undertake so many sacrifices in time, money, effort and self discipline ... this same
patient puts up a struggle in the interest of his illness against the person who is helping
him. How improbable such an assertion must sound' (1917, vol XVI, pp 286-287).
Freud regarded this resistance to treatment as a manifestation of the same forces that
kept the pathogenic ideas out of the patient's conscious awareness, and the concept of
resistance was recognised by Freud as one of the corner-stones of psycho-analysis. It has
particular importance in treatment, and the analyst usually aims to overcome the resistance by interpretation. This is 'the ultimate and decisive instrument' in psycho-analysis;
all other interventions are of subsidiary importance (Greenson 1967). The interpretation
is specifically intended to make the patient conscious of the meaning of an unconscious
psychic event.
After his collaboration with Breuer using hypnotism, Freud became increasingly dissatisfied with his results. It occured to him that although hypnosis made a certain amount
of material available for the analysis, it kept the resistance hidden from the therapist and
therefore established an impenetrable barrier beyond which the analyst could not go.
Freud was so impressed by this that he dated the beginning of 'psycho-analysis proper' to
the time at which he dispensed with the use of hypnosis.
The resistance was seen as evidence of powerful underlying forces which opposed any
alteration in the patient's condition, the same forces, in fact, which caused the condition
(1917, vol XVI, p293). There is therefore a close link between the concepts of resistance, repression and the unconscious, and Freud suggested that the concept of the
unconscious was itself derived from the theory of repression (1923, vol XIX, pIS). An
94
idea or memory may, because it is unacceptable to the ego for some reason, become
repressed; it remains within the mind - outside conscious awareness, but still operative,
and under certain circumstances it may re-emerge into consciousness. Repression is the
purposeful exclusion from consciousness of wishes, ideas or impulses which are unacceptable to the ego; generally these are of a sexual or aggressive nature. In this sense,
repression may be regarded as a specific defence mechanism, and it is one that is especially common in conversion hysteria.
In his refusal to equate the mind with the world of conscious experience, Freud was to
a large extent following the tradition established by others in Germany. Herbart, Leibniz,
Helmholtz and Fechner had each worked with the idea of the unconscious, but it was for
Freud to develop a systematic account of the operation of unconscious forces and to
apply this to the neuroses. He wrote of hysteria that the patient's mind was 'full of active
yet unconscious ideas; all ... symptoms proceed from such ideas. It is in fact the most
striking character of the hysterical mind to be ruled by them' (1912, vol XII, P 262). The
unconscious impulses were not below the threshold of consciousness because they lacked
sufficient intensity to maintain themselves in the conscious mind; indeed, in Freud's view
the opposite was often the case. It was the unconscious needs and impulses that were
frequently the most potent psychological determinants; these remained unconscious only
because they were actively excluded from consciousness.
The concept of defence appeared very early in Freud's writings, but for a middle period
of almost 30 years it was replaced by the similar notion of repression. In his later work he
returned to the notion of defence, and in The Problem of Anxiety (1936) he developed
the idea that far from being the result of repression, anxiety was the cause of repression.
6.3 The Structure of the Mind

In Freudian theory, the structure or topography of the mind includes a number of specific
subsystems, each of which has different functions and characteristics. In the New Introductory Lectures, Freud referred to this as 'the anatomy of the mental personality'. In the
earlier of his two topographical systems, Freud distinguished between the Conscious,
Pre-conscious and Unconscious systems: he was later to reject this formulation. In The
Ego and The Id (1923), Freud proposed a new personality structure, in which the three
subsystems of Id, Ego and Super-ego comprised the whole personality.
The id (das Es) is the instictive part of the personality; its contents are unconscious and
it is generally represented as a reservoir of instinctual energy or drive-force. 2 It includes
the sexual instincts, the somatic origins of which can be traced to the oral, anal and
genital zones of the body, and it also includes the aggressive instinct. Freud insisted that
the instinctual drives were based upon the bodily structure of the organism and rejected
the idea of occasional, transitory human instincts, which were so popular at the time (cf.
William James and William McDougall). At first, Freud would tolerate no dissent from
his thesis that the conflict underlying neurotic disorders was centred upon sexuality.
Largely as a result of Adler's ideas, he later admitted that non-sexual factors might lead
2 Although it is often translated as 'instinct', the word 'Trieb' carries the implications of irrestible
pressure rather than simply of direction to a predetermined goal. The Freudian concept of Trieb
implies pressure that is relatively indeterminate and is usually better translated as 'drive'.
The Structure of the Mind
95
to an unconscious conflict. His notion of an aggressive and destructive instinct (Thanatos)

was an attempt to incorporate a non-sexual instinct into his theory, though the inclusion
of aggression as one of the basic Freudian drives runs into some difficulties because of its
lack of any clear somatic foundation. For this reason it remains the subject of considerable controversy, and even Fenischel, otherwise one of the most loyal and orthodox
Freudians, found the Thanatos notion confused and implausible. Melanie Klein, on the
other hand, gave a central position to aggression in her own theory and attributed to it
many of the properties that Freud reserved for the libido.
The id is primitive and disorganised; it presses for release of its instinctual tension
according to the Pleasure Principle, and these needs and drives press for immediate
gratification. The ego is similar to what each of us means when he refers to the 'self': it
compromises those aspects of the personality which are responsible for perceiving,
understanding, reasoning, feeling and choosing (Hendrick 1958). Whereas the id
includes those mental attributes formerly ascribed to the System Unconscious, the ego
incorporates the attributes which were previously included in the System Conscious. 'The
ego represents what may be called reason and commonsense, in contrast to the id, which
contains the passions' (1923, vol XIX, P 25); Freud also used the metaphor of a horse
and its rider for the id and ego. The ego is absent in the infant and is an achievement of
maturation in the adult: Freud described it as a 'precipitate' of the experiences of the
individual in his encounters with the external world. It is the ego that develops the role of
reality testing, and as the id serves the Pleasure Principle, so the ego serves the Reality
Principle. The functions of the ego are partly synthetic insofar as they operate to maintain
the organism as a whole and to reconcile the primitive demands of the id and the moral
mandates of the super-ego.
The third agency, the super-ego, is a specialised part of the ego. Its nearest equivalent
in common language is the conscience, though Freud's concept of the super-ego differs
from this in several important ways. Most importantly, Freud recognised that censorship
of the super-ego could be unconscious. The obsessional neurotic 'behaves as if he were
dominated by a sense of guilt, of which, however, he knows nothing, so that we must call
it an unconscious sense of guilt, in spite of the apparent contradiction in terms' (1907,
vol IX, P 123).
In Freudian theory, the super-ego is based on the resolution of the Oedipus complex
when the child stops trying to satisfy his Oedipal wishes and internalises the parental
prohibitions. It is at this point that Freud suggests that a fundamental difference between
the sexes occurs. In the boy, the Oedipus complex and the threat of castration are closely
interwoven, and the super-ego is built upon their resolution. For girls it is suggested that
the castration complex prepares for the Oedipus complex instead of destroying it, and
that the complex is not resolved until comparatively late, and then only incompletely. 3 At
a later stage, the super-ego is refined by exposure to other social and cultural factors education, religion, morality and so forth, but it remains essentially an internalisation of
the parental prohibitions. Freud also explicitly stated that the super-ego is not based
upon an identification with the parents themselves but upon an identification with the
parental agency - that is, with the parents' super-ego (1933, vol XXII, P 64).
3 In Freud's view, women acquired a less effective super-ego than men (1933, vol XXII, P 129).
96
6.4 Regression and Infantile Semality

Much of the early criticism of Freud's ideas was fuelled by moral indignation. It was
thought that his views were irreligious and amoral, and his demonstration of the sexual
impulses of children was especially shocking to many. Shortly after the publication of the
Studies in Hysteria, Breuer withdrew from his collaboration with Freud, and one of the
principal reasons for this was probably their disagreement over the significance of sexuality in the causation of neuroses.
In his clinical practice, Freud consistently found that infantile sexual impulses persisted
in an unconscious but active form and exerted a powerful effect upon the adult neuroses.
He showed that the unconscious sexual wishes of the adult which gave rise to the psychoneurotic disorders were directly related to the infantile period of development during the
first 5 years of life. The significance of this period is that it constitutes the foundations of
adult sexuality, and that every adult who is frustrated or thwarted in his sexual behaviour
regresses to a more primitive stage of sexual functioning in order to gain satisfaction.
Freud described the neurotic as quite literally living in the past (1917, vol XVI, P 365).
The neurotic symptoms provide a substitute satisfaction for the individual who is faced by
frustration; they permit a regression of the libido to earlier and happier times when the
libido was able to gain satisfaction. The symptom or symptom-complex repeats this
earlier infantile kind of satisfaction, distorted by the censorship arising from the present
conflict and mingled with ailments from the precipitating causes of the disorder.
The libido is a biological concept used to refer to the energy of the sexual instinct - the
driving force of the personality. In his Collected Papers (1924), Freud spoke of it in
electro-dynamic or hydro-dynamic terms and described it as being like 'a fluid electrical
current', 'something which is capable of increase, decrease, displacement or discharge'.
This energy is channelled and directed during the innate, biologically-determined phases
of childhood development. Karl Abraham, one of Freud's chief disciples, developed the
libido theory in a series of papers, and it was Abraham's work that most influenced
Freud's thinking in this area; as a result of his researches, Freud came to regard libidinal
energy as organised differently during the different stages of infantile development. In
the first phase, the oral phase, the libido is bound mainly to the mouth, and the sources of
pleasure for the infant are his own body and the breast that feeds him. Because of the
inability to differentiate internal and external sources of stimulation, even the mother's
breast is conceived of as belonging to the ego.
During the second developmental stage, which occurs approximately between the ages
of 2 and 4, Freud suggested that the focus of libidinal interest begins to shift from the oral
to the anal region, and the infant's pleasures become associated with the retention and
expulsion of faeces. Unlike the comparatively passive oral phase, the anal phase is
directed towards the external world and relates to the parents in terms of defiance,
submission and feelings of power. It is at this time that the symbolic significance of giving
and withholding is established.
In the third, phallic phase of libidinal development, the libidinal gratifications of both
sexes become centred upon the genital organs, and it is at about this time that the little
girl begins to develop feelings of 'penis-envy'.
It is during the phallic phase that the Oedipus complex occurs. This plays a basic role in
the individual's personality development: indeed the neurotic's failure to resolve the
Oedipus complex (in which the libido is attached to the image of the parent figures)
The Actual Neuroses
97
fonns the nucleus of the psycho-neuroses (1925, vol XX, p 55). For the boy the resolution of his Oedipal feelings is connected with the fear of castration: for girls, Freud
suggested a different biologically-detennined course of development. The girl is resentful
of her lack of a penis; but at the same time, because of this, she is not affected by fears of
castration and is likely to spend a longer period in the Oedipal phase.
These phases of infantile sexual development are followed by a latency period, which
persists until increased levels of libido at puberty revive sexual interests. It is during the
latency period that there is an intensification of repression leading to an amnesia of the
earlier years. After puberty, the boy maintains his penis-interest and continues along the
same developmental course as before; but the girl must now become aware of her vagina,
and therefore, Freud suggested, of her feminine and passive role, and must renounce her
interest in the clitoris.
These various phases overlap and interact in the course of development, and it is
important to note that no phase is entirely given up. Freud presents the analogy of an
advancing anny which leaves garrisons on its route to send forward supplies and to
provide a line of retreat in the event of being checked and driven back. The tendency of
the forces to occupy and fortify one of the earlier positions is called fixation: the retreat
of the force to this point under adverse circumstances is the process of regression. The
neurotic disorder is prompted by the current difficulties in the patient's adult environment, but the disorder itself reflects the regression. 'The aetiology of the (neuroses) ... is
to be looked for in the individual's developmental history ... neuroses are acquired only
in early childhood ... even though their symptoms may not make their appearance until
much later' (1938, vol XXIII, pp 156 and 185). In this, the neurotic disorder is a form of
repetition. In an earlier work, Freud had described hysterics as suffering mainly from
their reminiscences, but he later modified his views on this, since the repetition of the
neurotic, far from being synonymous with memory, is actually an alternative to, and a
substitute for, reminiscence. The neurotic repeats instead of remembering (1914, vol
XII, P 151).
6.5 The Actual Neuroses

Freud made a clear distinction between the actual neuroses and the psycho-neuroses.
The fonner were a comparatively simple form of neurosis, the direct result of incomplete
sexual satisfaction, whereas the psycho-neuroses were more complicated and were
caused by a conflict between the ego and the,sexual instincts. The tenn 'actual neurosis'
first appeared in Freud's writings in 1898 to denote anxiety neurosis and neurasthenia,
though he was later to include various psychosomatic and hypochondriacal disorders
within this category.
The actual neuroses differ from the psycho-neuroses in several respects. In the first
place the actual neuroses are not caused by unresolved infantile conflicts, but have their
origins in the subject's present circumstances: (the adjective 'actual' is used here in the
sense of temporal actuality). In both types of neurosis the causes are sexual, but in the
actual neuroses the mechanism is somatic rather than psychological. 'The neuroses [are]
... without exception disturbances of the sexual function, the so-called "actual neuroses"
being the direct toxic expression of such disturbances and the psycho-neuroses their
mental expression' (1925, vol XX, P 25).
98
In the case of anxiety neurosis, the cause may be found in the non-discharge of sexual
excitation which results in 'a deflection of somatic sexual excitation ... and the consequent abnormal employment of that excitation' (1895, vol III, P 108), and in neurasthenia, in some sort of incomplete sexual satisfaction such as masturbation or spontaneous emission. This differs from the earlier account of neurasthenia given by Beard, in
which sexual factors were only one of many determinants. The symptoms of the actual
neuroses also differ from those of the psycho-neurotic disorders in that they are a direct
outcome of the absence or inadequacy of sexual satisfaction, rather than a symbolic
expression of some intrapsychic conflict.
The theoretical significance of this distinction is that the actual neuroses cannot be
treated by psycho-analysis, because their symptoms lack any symbolic meaning that can
be brought into the patient's consciousness. Freud maintained that these disorders were
to be explained in neuro-chemical rather than psychological terms. 'I cannot regard the
genesis of the symptoms (of the actual neuroses) ... as anything but toxic' (1912,
vol XII, P 248).
Freud appears to have accepted that 'abnormal' sexual practices had a biological
significance independent of cultural definitions and that they had a direct and straightforward causal influence upon the actual neuroses. He described how manual masturbation
led to one sort of actual neurosis, whereas another form of actual neurosis would immediately appear in place of the first if the patient changed to some other form of abnormal
and incomplete sexual behaviour (1917, vol XVI, P 386). Nonetheless, the symptom of
an actual neurosis can be the first stage of a psycho-neurotic symptom. Freud suggested
that although the pure cases of anxiety neurosis were usually the most marked, the
symptoms of anxiety more often occurred at the same time as symptoms of neurasthenia,
hysteria or depression, and in his earlier writings he often used the term 'mixed neurosis'
to refer to the combination of different types of neurosis. Mixed neurosis implied not
merely a complicated mixture of the two sorts of neurotic symptomatology, but also 'the
admixture of several specific aetiologies' (1895, volIII, p 113).
Freud's explanation of the actual neuroses was entirely consistent with his libido
theory. The libidinal energy was directed towards its release in orgasm and if deprived of
this satisfaction, the energy remained in the system pressing for discharge. In this way it
became converted into anxiety. Although anxiety is now regarded as a central concept in
understanding neurosis, this is a comparatively recent development. Prior to the 1920s
anxiety was regarded as of only incidental importance, and in his early work Freud
consigned anxiety to the category of actual neuroses and explained it as a direct physiological reaction to sexual frustration. He never entirely abandoned this position, but in
The Problem of Anxiety he introduced an entirely different account of anxiety, in which it
was seen as a signal of some internal, intrapsychic threat. In this second formulation, the
neurosis developed as an attempt to cope with anxiety.
Freud's interest in the actual neuroses was somewhat eclipsed by his later concentration upon the psycho-neuroses, and since few other analysts have shown much interest in
this topic it has been rather overlooked. Nonetheless, in a retrospective look at his
formulation of the actual neuroses, Freud suggested that although his ideas might have
been only a rough outline of a complicated subject, they remained fundamentally sound
(1925, vol XX, P 26).
99
The Psycho-neuroses
6.6 The Psycho-neuroses

The concept of psycho-neurosis appears very early in Freud's writings. In The Neuropsychoses of Defence (1894, vol III) he presented a psychological theory of phobias,
obsessions and conversion hysteria. The theory of repression (or defence) first appeared
in this paper, and Freud was to describe this as the 'cornerstone on which the whole
structure of psychoanalysis rests. It is the most essential part of it' (1914, vol XIV, P 16).
Freud divided the psycho-neuroses into two categories, the transference neuroses
(phobias, obsessions and conversion hysteria), in which the patient responds to frustration by investing libido in fantasised objects, and the narcissistic neuroses. Table 3 shows
the Freudian concepts of neurosis in relation to current psychiatric usage. Of course, it is
not possible to achieve a precise matching, both because of historical changes in the
meanings of the concepts themselves and because of the ambiguities which are inherent
in both systems. The Freudian distinction between the actual neuroses and the psychoneuroses is one that is not made in the psychiatric nosologies, and the concept of psychoneurosis strays beyond the conventional limits of 'neurosis' into that of 'psychosis'. The
concept of narcissistic neurosis has also disappeared from present-day usage. Freud used
it to refer to those cases in which the libido had been withdrawn from the outside world
and redirected towards the ego itself. Initially, the term had been used by Freud in an
attempt to extend psychoanalytic principles to the psychoses, though he was later to
limit its use to certain depressive conditions, and it most nearly corresponds to what
psychiatrists would now call manic-depressive psychosis.
Freud used the concept of repression to contrast the psycho-neuroses (more precisely,
the transference neuroses) with the actual neuroses. The symptoms of the psycho-neuroses were regarded as a symbolic expression of unresolved infantile conflicts. The conflict is caused by frustration which deprives the libido of satisfaction, causing it to seek
new sources of release in regression. If these regressions arouse no objection from the
ego, the libido will achieve some real, though perverse, form of satisfaction - as in the
sexual perversions. But despite this important difference, Freud noted that the sexual
perversions shared with the neuroses features of regression, fixation and repression, and
suggested that they too were accessible to treatment by psycho-analysis (1905, vol VII,
P 232).
Table 3. The Freudian concepts of neurosis
Actual neurosis
Freudian
concepts
Psychoneurosis
Transference neuroses
Narcissistic neuroses
Narcissistic
Paraphrenias
neuroses
Obsessional-compulsive,
phobic and
hysterical
neuroses
Manic-depressive
psychosis
Current
psychiatric
usage
Anxiety neurosis,
hypochondriacal
neurosis,
neurasthenia
Paranoid
schizophrenia
100
It is an essential feature of the neurotic conflict, on the other hand, that the new objects
and paths to libidinal satisfaction should be unacceptable to the ego, which opposes the
new attempt to gain satisfaction. 'One part of the personality champions certain wishes
while another part opposes them and fends them off. Without such a conflict there is no
neurosis' (1917, vol XVI, P 349). The frustrated libidinal impulses must seek some indirect way of finding a new form of satisfaction, and this results in the formation of neurotic
symptoms. In all cases the pathogenic conflict is between the ego and sexuality, and the
ego actively initiates defensive processes in response to the danger. This frequently takes
the form of repression, though the ego may also use other defence mechanisms to protect
itself from conscious awareness of unacceptable impulses: these include projection, reaction formation, denial, displacement, undoing and rationalisation. In each of these cases,
the neurotic remains in a state of basic conflict, since he must keep himself unconscious of
the defences that he is using and of the true significance of the unacceptable elements.
Because of this, repression has a paradigmatic importance in Freudian theory, and Freud
referred to the other defences as 'surrogates of repression' (1926, vol XX, p 119).
6.7 Little Hans

In his writings, Freud made considerable use of his own clinical material, both as evidence of the practical effectiveness of psycho-analysis and as illustrative material for its
theoretical principles. One of the best-known cases was that of Little Hans, first published under the title 'Analysis of a Phobia in a Five Year Old Boy' (1909, vol X,
pp 5-149). Although atypical in some ways, the case is significant because it provides the
first example of a child analysis. It also confirmed Freud's views about the sexuality of
childhood and has been used by analysts as a basic substantiation of psycho-analytic
theory. In this case Freud was able to 'observe directly' the sexual impulses and wishes
that are so deeply hidden in adults.
Little Hans4 had developed strong fears of being bitten by a horse, which developed
after he had seen a horse fall down in the street one day. It is a central task of the analysis
to establish the symbolic meaning of the horse. Hans' problem is described in terms of his
Oedipal desires for his mother, with whom he was allowed to sleep in his father's absence. Freud suggests that Hans developed a wish that his father should be permanently
absent (i. e. be dead) so that Hans could replace him in his mother's bed: because of his
love for his father, his Oedipal desires and the death wishes were denied direct expression. Hans' first statement of his phobia and his fear of being bitten represented his fears
of castration by his father for his Oedipal longings. The repressed libidinal energy had
found release in some new form, showing itself as anxiety.
It is as a means of coping with this anxiety that the symptom arises. The mind erects
'mental barriers in the nature of precautions, inhibitions, or prohibitions; and it is these
defensive structures that appear to us in the form of phobias' (1909, volX, p 117).
4 Because of the limitations of space in this chapter, the account of Little Hans is necessarily
abbreviated and cannot convey a true impression of Freud's analytic writing. The reader who
wishes more than the skeletal outline offered here might refer to the original work: for a critical
discussion of this work the reader might consider Wolpe and Rachman (1960).
101
Neurotic symptoms take the form of a compromise which disguise the original impulse at
the same time as allowing it some form of indirect expression. Hans' fear of horses was
actually quite complex. Behind his fear of being bitten was a deeper fear of horses falling
down. The symbolic meaning of a falling horse is presented by Freud as the father dying,
as well as the mother in childbirth. (When Hans was 31/2 years old, the mother had given
birth to a baby sister, and this event also plays a large part in the analysis.) The symptom
operates to hide this from Hans by transferring the anxiety from its true, but unacceptable, source to the overt symptom. At the same time the compromise solution that the
symptom represents permits some part of his repressed desire to achieve release, for
example in his inability to leave the house and therefore to leave his mother.
As a result of bringing to light the repressed complexes, Freud reports that Hans lost
his fear of horses and developed a better relationship with his father. Some years later
Freud met Hans when he was a young man and he had retained his therapeutic gains and
was free from any neurotic problems.
6.8 Neo-freudian Theories

Many critics of psycho-analysis have under-estimated both the complexity of psychoanalytic theory itself, and more particularly, the diversity of psycho-analytic viewpoints.
There can be little doubt that in his emphasis upon constitutional and organic determinants, Freud showed a serious neglect of the important role played by social and cultural
factors. He frequently mistook cultural for biological-instinctual phenomena, and this
was the focus of concern for several of the neo-Freudians. One of the basic issues that
divides the schools of psycho-analysis is Freud's libido theory, and particularly the role of
infantile sexuality in personality formation and in the determination of adult neuroses.
Jung's departure from orthodox Freudian theory was based upon his desire to avoid what
he saw as an undue emphasis upon infantile sexuality and the Oedipus Complex. The
non-libido theorists (including Homey, Adler, Fromm and Sullivan) also reacted against
the biological-drive formulation and emphasised the self and its social and interpersonal
experiences as the crucial determinants of behaviour.
6.S.1 Adler
In 1911, Alfred Adler was the first to break away from Freud's inner circle. His theory of
'Individual Psychology' was based upon the social and interpersonal experiences of the
child. Although these factors are also of obvious importance to the Freudian account,
they do not occupy the primary role that Adler granted them.
The Individual Psychology account of the neuroses (and of human behaviour in general) is based upon two main principles: (a) inferiority feelings which are the result of
man's physical endowments or his social situation and (b) his strivings for mastery,
control or power in an attempt to compensate for these feelings. In the case of the
neurotic, the attempted compensation is exaggerated and unrealistic, and it leads to the
person's becoming self-centred rather than outwardly-directed towards the social and
physical world.
In his early studies, Adler investigated the psychological techniques by which individuals learned to cope with 'organ inferiorities' such as bodily defects. A frequently quoted
example is that of Demosthenes, who became a great orator in compensation for a
102
childhood speech defect: other stutterers might use the same defect as an excuse to
defend their self-esteem against the threat of failure. Similarly, every normal child starts
life from a position of helplessness, then strives increasingly to control his sense of
inferiority before the environment. The point at which the individual feels most inferior is
determined by specific organ inferiorities, and by social and interpersonal factors. It is
precisely at this point of weakness that the compensatory strivings for superiority are
most marked (Adler 1946). The drive towards power, prestige, superiority or perfection
plays a central part in Adler's theory, and the style of life by which the child learns to
cope with his feelings carries through to adulthood.
Like Freud, Adler recognised the first 5 years of life as crucial for personality structure,
but he believed that human behaviour was to be explained in terms of present ambitions,
hopes and strivings towards the future, rather than by biological drives or the unconscious imprint of past experiences. Adler was unimpressed by the Freudian concepts of
the unconscious and repression. Although the patient might be unaware of their inner
life, this was not due to repressions, unconscious processes or resistances. This denial of
basic Freudian concepts led many analysts to regard Adler's system not as false, but as
superficial, dealing with the secondary gains of neurosis rather than the primary causes
(Hendrick 1958).
In the neurotic, the individual's goals are unreal and egocentric, and are primarily an
attempt to attract the attention of other people. 'The entire picture of the neurosis as well
as all its symptoms are influenced by, nay, even wholly provoked by an imaginary fictitious goal' (Adler 1918); this goal is the universal drive to power, which is unusally
marked in the neurotic. The neurotic shows 'no single trait which cannot likewise be
demonstrated in the healthy individual' (ibid), although the neurotic character is incapable of adjusting to reality because it is focused upon an impossible and fictitious goal.
Adler introduced the concepts of 'sibling rivalry' and 'inferiority complex', and the
accessibility of these notions to people outside psycho-analysis led to their rapid and
widespread acceptance. Indeed, they have now passed into common usage. But despite
its initial widespread acceptance at the time, Adler's Individual Psychology as a system is
now practically extinct, although his ideas had a marked influence upon Karen Horney's
work.
6.8.2 Horney
Karen Horney was also one of Freud's faithful disciples, but after emigrating to New
York she made a radical break with him. Her definition of psycho-analysis as 'certain
basic trends of thought concerning the role of unconscious processes and the ways in
which they find expression, and ... a form of therapeutic treatment that brings these
processes to awareness' (Horney 1937), although rather general, is fairly orthodox; but
like Adler's theory, which had a marked influence on her work, Horney rejected the
libido theory and the structural approach (id, ego, superego). Unlike Adler, for whom
anxiety was a comparatively minor concept, Horney gave a central role to anxiety in her
theory. It is anxiety and the individual's attempts to defend himself against feelings of
anxiety that lie at the heart of the neurotic disorders, and for Horney, an investigation
into the causes of anxiety was an investigation into the causes of the neuroses.
One of the basic causes of anxiety was 'inevitably a lack of genuine warmth and
affection in childhood' (Horney 1937). Coupled with the fundamental helplessness of the
child, this leads to hostility, which in turn creates anxiety, since the child is unable to give
103
free expression to these hostile feelings. Horney refers to this attitude as 'the basic
anxiety; it is inseparably interwoven with a basic hostility' (ibid), and it is from these two
opposing forces that the neurotic conflict arises. The simultaneous need to approach
others and to avoid or to oppose them, leads to the neurotic attitudes of submissiveness,
withdrawal or aggressiveness. Because of her approach, she was comparatively uninterested in specific symptoms and in the underlying personality determinants of the
neurotic behaviour. Instead, she attaches absolute importance to the interpersonal and
cultural situations in which the neurosis finds expression (Wyss 1966).
She recognised that psycho-analysis as a sytem has always been in danger of stagnation
through deference to Freud's own achievements. Horney stressed the cultural implications of the concept of neurosis. What might be considered abnormal in our own culture,
or in our own time, might be regarded as appropriate in other times and places. Both
Fromm and Horney pointed out how Freud's ideas were restricted by his own cultural
background. Fromm (1960) wrote that 'Freud was so imbued with the spirit of his culture
that he could not go beyond certain limits which were set by it. These very limits became
limitations for his understanding ... of the sick individual . . . of the normal individual
... of the irrational phenomena operating in social life.' Horney (1936) was concerned
with the possibility that neuroses may be moulded by socio-cultural processes 'in essentially the same way as 'normal' character formation', and that the neurotic conflict was
not simply one of coping with instinctual intrapsychic drives, but one of adapting to the
conflicting demands that society imposes upon its members.
Freud's conclusion that women are more jealous than men was based upon his view
that the anatomical differences between the sexes inevitably leads girls to be envious of
boys. This penis envy is later translated into an envy of other women's relationships with
men (i. e. their possession of other men). Other analysts have given even more emphasis
to this anatomically-based penis envy in women. Helen Deutsch, for instance, believed
that the psychology of women is primarily determined by their attempts to compensate
for their anatomical defect (Deutsch 1944).
Horney, however, has always been one of the most articulate critics of this aspect of
Freudian theory. She concluded that there were no basic psychological differences between the sexes, and insofar as penis envy exists, it is not biologically-, but culturallydetermined, and is more likely to represent an envy of the masculine social advantages
than an envy of his anatomical advantages. Freud's views on this matter represented an
over-generalisation from a culturally limited viewpoint. She cited anthropological evidence of the wide variations in the expression of jealousy by men and women that exist in
different societies, and suggested that Freud's pronouncements about the specific guilts,
competitiveness, sibling rivalry, etc. were based upon a restricted view of human nature.
The Freudian over-valuation of biological determinants to the neglect of cultural factors
was one of the greatest defects in his work. 'Freud's disregard of cultural factors not only
leads to false generalisations, but to a large extent blocks an understanding of the real
forces which motivate our attitudes and actions' (Horney 1937).
All analysts, including Freud, accept that repressed hostility is a frequent cause of
anxiety. But whereas Freud explains this in terms of his instinct theory, Horney sees the
hostility in terms of the individual's experiences with his socio-cultural environment. In
her own formulation, the essence of the neurotic condition is basic anxiety, by which is
meant 'the feeling (originating in childhood) of being isolated and helpless in a potentially hostile world' (Horney 1946). It is the pattern of responses by which the person
104
learns to cope with this culturally-determined basic anxiety (Homey refers to these as
safety devices) that constitutes the symptoms of the neurotic disorder.
The ego protects itself against basic anxiety by erecting defences, and in Homey's view,
these neurotic defences themselves can cause further anxiety, since they may be in
conflict with each other. Thompson (1950) illustrates this by the case of a child who has
developed powerful needs for success and recognition, but later experience leads him to
despise and devalue such needs. This produces the conflict between secret ambitions and
the need to appear outwardly modest and inconspicuous, and as a result of this conflict
the person becomes an under-achiever, unable to assert himself. This is intolerable to the
ambitious part of his personality, and as a result he must either experience an increase in
anxiety or add new defences against that anxiety, such as the belief that others are
obstructing his progress.
6.8.3 Fromm
Like Homey, Erich Fromm was opposed to the biological emphasis of Freud's work: as
with all psycho-analysts, Fromm looks to the child's experiences within the family as the
crucial determinants of personality development, but whereas Horney substituted an
alternative socio-cultural account of the determination of drives, Fromm criticised
Freud's theory of drives from a different perspective. Unlike Homey, whose explanation
of neurosis relies upon the notion of basic anxiety, Fromm argued that there is a fundamental and crucial distinction between the biological drives of animals and the human
drives. Whereas animal drives are predetermined and limited in their operation, human
behaviour is flexible. Man is, in principle, free to choose whether or not he will obey his
drives, and Fromm's basic premise is that of individuation, man's attempt to achieve selfrealisation and freedom from his biological nature. This freedom from biological restraint
imposes its own demands upon all individuals. Individuation is not an easy process, and it
is man's freedom from his biological nature that permits the formation of neurosis. For
the neurotic, the threat of freedom proves intolerable. In the tension between the
impulse towards free individuation and the security of primary (infantile) emotional ties,
the neurotic opts for security rather than freedom: in his attempts to avoid feelings of
isolation and loneliness the neurotic uses mechanisms of escape. These neurotic solutions
take various forms. The person may, for instance, adopt a pattern of blind conformity to
cultural expectations (automaton conformity), and much of Fromm's work is concerned
with the issue of conformity and authoritarianism (e. g. Fear of Freedom 1960). But these
neurotic attempts to solve the problem of isolation must fail because man cannot reunite
himself with the world in the way that was possible before the development of individuality.
Fromm's theory suffers from a number of drawbacks, not the least of which is its
somewhat vague and 'philosophical' style. Another problem with his account of neurosis
(as with those of Adler and Homey) is that although it introduces social factors into the
discussion, it reduces their psychological significance to an over-simplified form. The
phenomena of neurosis are too complex and varied to be easily explained solely in terms
of such concepts as the need for mastery, basic anxiety or fear of individuation. This
weakness of the neo-Freudian theories of neurosis is linked to their neglect of the specific
symptoms of the neurotic disorders. These neo-Freudian accounts focus upon character
neurosis, a term which is widely used, but has no exact meaning. This approach transcends the distinction between neuroses with symptoms and those without symptoms by
105
its overall emphasis upon personality organisation, but degrades the importance of symptoms in the pursuit of intrapsychic events. Admittedly, the clinical investigation of
neurosis was not one of Fromm's principal aims, but Adler and Homey also neglect this
problem of symptoms and attempt to conceptualise and explain neurosis in general terms
of personality adjustment.
6.9 The Evaluation of Psycho-analysis

On the fundamentally important question of 'evidence', there is good reason to believe
that many psycho-analysts, and Freud in particular, hold a mistaken view of their own
role. Freud wrote of his critics that 'the main obstacle to agreement lay in the fact that my
opponents regarded psycho-analysis as a product of my speculative imagination and were
unwilling to believe in the long, patient and unbiased 5 work which had gone into its
making' (1925, vol XX, p50).
The first objection to this view that the analyst's clinical observations could be
unbiased is a general one. Perceptions occur in a framework of beliefs, expectations,
values and so forth: the analyst's perceptions of the patient are not veridical and unbiased
in the way that Freud assumed. Beyond this, the complexity of the patient-doctor interaction is underestimated. Truax (1966) found that even the supposedly non-directive interviews of Rogers contained their own implicit reinforcements and directives, of which the
therapist was unaware. Whether he likes it or not, the therapist influences the patient's
responses.
Freud was also ambivalent about the analyst's role in treatment. Contrary to his own
directive that the analyst should be 'impenetrable to the patient' and 'like a mirror',
Freud also stated that 'during the analysis [the patient] had to be told many things that he
could not say himself, that he had to be presented with thoughts which he had so far
shown no signs of possessing, and that his attention had to be turned in the direction from
which his father was expecting something to come' (1909, vol X, P 104). Elsewhere he
stated that he would 'stand obstinately by my suspicions till I had overcome the patients'
disingenuousness and compelled them to confirm my views' (1917, vol XVI, P 386). The
analysis therefore has many of the characteristics of education or indoctrination, in which
the structure of psycho-analysis is imposed upon the patient. This leads to further difficulties associated with the process of interpretation, to which analysts attach so much
weight as a special sort of evidence. The patient's acceptance of the interpretation provides no evidence that the interpretation itself was an accurate or 'true' identification of
unconscious material. Farrell (1967) characterised the process of interpretation as principally a means of transforming the patient into a 'Freudian-type' of person. In short, the
analyst is likely to discover whatever he looks for.
The uncertainties surrounding the psycho-analyst's use of clinical observation and
interpretation render them a poor vehicle for gathering evidence about the validity of
psycho-analytic theory. Freud was mistaken in his view that his theory was supported by
such a 'wealth of reliable observations ... [that it was] independent of experimental
verification' (Freud's comments to Rosenzweig, cited in Postman 1962). There are,
5 My italics.
106
however, many problems inherent in any attempt to put Freud's theoretical statements to
the test. Farrell (1951, 1961) stated that 'psycho-analytic theory is qua theory, unbelievably bad', and that certain aspects of the theory are untestable.
Karl Popper's dissatisfaction with the psycho-analytic theories of Freud and Adler was
rooted in the apparent paradox that their explanatory power was too great. The theories
were able to explain practically everything that happened, and conversely, whatever
happened seemed to verify the theory. It is certainly true that Freud and most of his
colleagues believed that psycho-analytic theory was constantly being verified by their
clinical observations, and in one sense there are no eventualities that could contradict
them. This is their greatest flaw: they are not falsifiable.
Popper (1963) presented the criterion of testability or falsifiability, as a means of
distinguishing between the statements of the empirical sciences and those of pseudoscience. Another expression of this is that a scientific theory must be capable of learning
from its mistakes. Those propositions which we cannot (in principle) put to the test and
disconfirm belong to the realm of religion or mythology. At the same time this criterion
of falsifiability must not be confused with the issue of meaningfulness or of truth: because
a statement is untestable does not mean that it is necessarily nonsensical or untrue.
This view of science has been important and influential, and although there are
philosophical problems inherent in this position (d. Kuhn 1970), it is generally agreed
that theoretical propositions should be amenable to some sort of stringent test. Harzen
and Miles (1961) argued that this demand for evidence, although difficult to state in a
way which is itself free from theoretical objections, remains valid. The problem that
psycho-analysis must then face in one form or another is this - what could count as
satisfactory evidence for rejecting particular aspects of the theory?
The Pleasure Principle, for instance, is crucial to Freudian theory (repression and
unconscious processes depend upon it), yet there is no way of putting it to the test. The
reduction of tension within the psychic system is a vague notion which escapes empirical
scrutiny, because there is no indication of what form this tension takes, or how it might be
measured. The nearest principles from experimental psychology, the drive-reduction
hypothesis and the Law of Effect, have themselves been rejected because they are either
tautologous or incorrect (Eysenck 1976). Another illustration of the tendency of psychoanalytical theory to distance itself from the possibility of empirical investigation can be
found in Freud's statements about the distribution of libido. 'All through the subject's life
his ego remains the great reservoir of his libido, from which object-cathexes are sent out
and into which libido can stream back again from the objects' (1925, vol XX, P 56). It is
not clear what criteria we could adopt to test the truth or accuracy of such a statement.
Similarly, psycho-analysis proposes that obsessional-neurosis and anal-eroticism are
related, in the sense that obsessional traits are defence mechanisms against anal-eroticism that has been repressed in early childhood by over-zealous toilet training. Kline
(1968) made an empirical study of this issue using four questionnaire measures of obsessionality and the Blacky Pictures projective test. His results showed a significant positive
correlation between obsessionality as measured on three of the objective tests and the
measures of anality on the projective test. Kline suggests that his findings broadly support
the Freudian theory. Yet in the same paper he notes that the Freudian theory links
anality to fixation at the retentive stage, and that 'perhaps, there should be a negative
correlation with the [anal] expulsive score'. His own results show four positive correlations, one of which was statistically significant. Because of this vagueness of the theory, it
107
is frequently difficult (as in this case) to know whether or not results confirm its predictions or not.
The same state of affairs can be found in many other areas of the psycho-analytic
theory of neurosis. Either the theory itself is untestable, or its predictions are equivocal.
Stein and Ottenberg (1958) attempted to test the hypothesis that asthmatic attacks
represent a psychosomatic defence against smells which would activate unresolved childhood conflicts. The authors claim that their results support the Freudian theory because
dirty and unpleasant (anal) smells are more likely to induce attacks than pleasant smells.
But despite the authors' conclusions, the most common single cause of asthmatic attacks
was perfume (ten subjects). This contrasts with only two subjects who mentioned the
smell of faeces as a precipitant. One might also take issue with the inclusion of paint,
horses and smoke in the category of 'unclean' smells. The results are further weakened by
the authors' failure to consider alternative explanations. The asthma may, for instance,
be a result of hypersensitivity to certain chemicals, which causes the restriction of air
passages to avoid taking in substances to which the person is allergic. Yet again one is
faced by the problem that the theory does not specify precisely enough under what
conditions the psychosomatic reaction should occur.
The term 'psycho-analysis' has two quite distinct meanings - Freud referred to it as a
science of unconscious mental processes, and as a particular method of treating
psychological disorders. In its second sense Freud argued that psycho-analysis is a more
effective treatment for the psycho-neuroses (hysteria, phobias and the obsessional states)
than any other method of treatment (1926, vol XX, P 264).
In trying to evaluate the effectiveness of psycho-analytic treatment, the sheer length of
the course of treatment presents one of the first difficulties. Heilbrunn (1963) presented
the results of his own treatment of patients over a 15-yearperiod and concluded that only
38% of the patient who had been seen for between 301 and 1350 sessions showed
improvement. Other studies have cited higher rates of improvement, but only after
selecting out the patients who terminated their treatment prematurely (e. g. Knight
1941): this sort of data manipulation is obviously suspect.
In the evaluation of the effectiveness of psycho-analysis, most authors have neglected
the phenomenon of spontaneous recovery. Denker (1946) and Wallace and Whyte
(1959) suggested that as many as 70% of neurotics show spontaneous remission of their
symptoms without psychotherapy, though this may not be equally true of all the varieties
of neurosis. Meyer and Crisp (1970) suggested that obsessional neurosis does not typically show this sort of remission and is often resistant to treatment.
If one accepts the two-thirds spontaneous recovery rate, this leads to some problems
for proponents of psycho-analysis, since their own treatment success rates are lower than
this. Stuart's (1970) review presented rates of between 24 and 54% improvement for
psycho-analytic treatment, and suggested on the basis of these estimates of the rate of
spontaneous recovery that psycho-analysis may adversely affect the patient's chances of
recovery. This conclusion is probably unwarranted on the basis of any overall estimate of
the rate of spontaneous remission for neurosis. Different neuroses appear to show different rates of recovery (Subotnik 1972), and it is not clear that the patients treated by
psycho-analysis and those in the spontaneous recovery studies were directly comparable:
the processes of patient selection are still inadequately understood. Nonetheless, even
the most conservative estimates of spontaneous recovery suggest that between 20% and
40% of neurotics may recover without treatment (Paul 1966; Saslow and Peters 1956),
108
and any evaluation of treatment effectiveness should control for this effect. Few psychoanalytic studies have done so.
A further difficulty is raised by the use of different criteria of improvement in the
studies of spontaneous recovery and in psycho-analytic treatment. Because the analytic
definition of neurosis goes beyond the symptoms to unconscious personality factors,
many psycho-analysts would not accept the simple remission of specific symptoms as
evidence of recovery. The phenotypic concepts of behaviour therapy, which are concerned with overt behaviour, and the genotypic concepts of psycho-analysis, which deal
with unobservable psychic forces, are not necessarily independent, but in practice it is
often impossible to translate from one system into the other.
It is difficult to reconcile the facts of symptomatic treatment or spontaneous recovery
with psycho-analytic theory, since the disappearance of the substitute satisfactions
offered by the neurotic symptom should lead to a re-emergence of the libido in some
other form of expression. According to Freudian theory, all neurotic symptoms are
secondary and superficial. They are symbolic of some unconscious process and are not in
themselves the essence of the neurosis. Obsessional ideas, for instance, are invariably the
expression of transformed self-reproaches originating in some sexual act that was performed with pleasure in childhood and which have re-emerged from repression. In the
treatment of neurosis, the analyst aims to make the patient aware of the unconscious
meaning of the symptom, and it has been widely assumed that symptomatic treatment of
the neuroses is therefore a waste of time. Since the symptom is merely the superficial
expression of underlying conflicts and anxieties, the removal of the symptom should
leave the basic neurosis unchanged and likely to find some alternative form of expression.
On this issue the analysts and behaviour therapists make quite different predictions.
Whereas Freud stated that 'it is impossible to effect the cure of a phobia (and even in
certain circumstances dangerous to attempt to do so)' (1909, vol X, P 117) by forcing the
patient to confront his phobic stimulus, behaviour therapists have done precisely this. In
some cases the exposure has been gradual, as in systematic desensitisation, in others
symbolic, as in covert or imaginal sensitisation; in other cases the patient has been faced
with in vivo exposure, as in flooding. Each of these symptomatic treatments leads to
improvement in phobic disorders without evidence of symptom substitution (Marks
1974). Similarly, obsessional-compulsive disorders improve as a result of enforced ritualprevention (Levy and Meyer 1971): again, no symptom substitution occurred and there
was an additional gain in that the patients experienced less anxiety and depression after
the treatment. Mowrer (1976) has described how the direct treatment of enuresis by
conditioning methods led to not one single incidence of symptom substitution as predicted by Freudian theory.
Blanchard and Hersen (1976) attempted to explain how psycho-analysts and
behaviour therapists have failed to resolve their disagreement over such a fundamental
issue as whether or not symptom substitution occurs by suggesting that proponents of
each theory have generalised beyond their data. Psycho-analysts, for instance, have been
especially interested in hysteria; in which there is an element of secondary gain. This
might be expected to lead to symptom substitution (or to relapse) if the treatment
consisted merely of symptom removal or symptom suppresion. Behaviour therapists, on
the other hand, have been more concerned with the phobic, obsessional-compulsive and
anxiety disorders. These may be seen as avoidance behaviours which are maintained by
the anxiety-reducing properties of the symptomatic behaviours. In such cases, the evidence is much clearer. Symptom substitution in the psycho-analytic sense does not occur.
7. Trait Theories of Neurosis
We cannot understand behaviour simply as a response to the external environment.

People seem to act with a degree of consistency in different situations as if they were
predisposed towards certain sorts of behaviour. Some people react with anxiety to the
slightest threat, whereas others remain free from anxiety under the most stressful conditions. In those people who are psychologically disturbed, it is even more likely that their
behaviour will be seen as a result ~f distinctive individual or per;onality predispositions.
When someone reacts with terror to the sight of a dog, or feels an overwhelming compulsion to wash their hands every time they touch some object in their environment, or
believes that they have a fatal illness whenever they have the slightest ache or pain, then
it is likely that we will seek to explain their unusual behaviour in personality terms.
This chapter will deal mainly with the personality theories of two psychologists with
very different attitudes towards psycho-analysis - Hans Eysenck and Raymond Cattell.
Eysenck has consistently and vigorously opposed the formulations of psycho-analysis;
Cattell, on the other hand, has been more sympathetic to such views, and many areas of
his work reflect the influence of psycho-analysis. Eysenck and Cattell both share a
commitment to the validity of a trait-description approach to the neuroses. Unfortunately, it is not at all clear what is to count as a definition of a personality trait, and
despite the fact that Allport identified about 18000 trait names in the English language,
there remains considerable disagreement between psychologists as to what is meant by
the term. In the most general sense, a trait is a collection of responses bound together by
shared characteristics which permit them to be described by a common term. Trait
psychology assumes a certain consistency of behaviour across situations, and traits are
seen as relatively stable and enduring predispositions acquired through learning, or
through constitutional or genetic effects, and which exert fairly generalised effects on
behaviour.
Eysenck (1972 b) has argued that trait psychology does not provide a causal analysis of
behaviour, but rather a descriptive one; as such it attempts to identify the underlying
structures of superficially chaotic and complex forms of behaviour. With respect to the
causal and descriptive functions of personality theory, Eysenck could be accused of
inviting misunderstanding by the way that his writings tend to switch from the causal to
the descriptive level - usually without explicit comment. Cattell, on the other hand, does
allow a causal role to trait factors, though he also regards the descriptive function of trait
theory as important, arguing that psychologists have paid far too little attention to the
basic conceptual foundations upon which their discipline is based. As a result of this
neglect, research psychologists are inclined 'to resemble a mob rather than a disciplined
army, preferring the looting of more or less immediately useful data to the overcoming of
central issues of principle or method' (Cattell 1946).
110
The concept of a trait is intimately related to that of correlation, and the statistical
procedure underlying the work of the trait theorists is factor analysis. Charles Spearman
first discussed factor theory around the tum of the century, though the matrix theory
formulation of factor analysis did not really emerge until Thurstone's work in the 1930s.
Factor analysis is, incidentally, one of the few scientific methodologies that psychology
has been able to offer to other disciplines. The single most distinctive feature of factor
analysis is its capacity to reduce a complex mass of data to a more simple and more
orderly form. In the simplest design, one might measure a sample of people on a number
of variables. The scores are then intercorrelated in every possible pairing to form a
correlation matrix. Factor analysis is a technique which helps us to see whether there is
some underlying pattern of relationships enabling us to reduce or to rearrange the data in
such a way that a small set of factors can be extracted. Spearman's initial work on
factorial methods was based upon the assumption that whenever two or more abilities are
correlated, they can be regarded as depending upon a common factor. The factors that
emerge may be taken as the source variables which account for the observed correlations
in the data. The advantage of this procedure is obvious if there are a large number of
initial variables. The intercorrelations among 100 variables would yield almost 5000
simple correlation coefficients, but could probably be reduced quite effectively to about
10 factors.
Factor analysis has an especially useful place in the early stages of research when basic
concepts are still lacking. Until the important variables have been identified, the design
of experiments remains largely a matter of trial and error, and there is little hope of
working out any precise laws about the relationships between variables. Cattell (1952)
has stressed this role of factor analysis in finding the independently-acting influences
before moving on to experimentation, and factor theorists in general have emphasised
the importance of factor analysis as a technique for the identification of the inherent
organisation or structure in complex sets of data.
7.1 Guilford
Probably the first attempt to apply factor analytic methods to personality questionnaire
data was that of Guilford, whose work influenced both Cattell and Eysenck. The study of
Guilford and Guilford (1934) showed no single factor common to all test items, but a
number of common factors emerged. These were tentatively identified as (a) social
introversion-extraversion, (b) emotional sensitivity, (c) impulsiveness and (d) interest in
self. This analysis was later repeated using more refined factor analytic techniques, and
three main factors were found (Guilford and Guilford 1936). These were labelled S for
social shyness, E for emotional immaturity or emotional dependency, and M for masculinity. Later research isolated further factors, and on the basis of a very large factor
analysis Guilford and Zimmerman (1956) confirmed all the 13 factors hypothesised by
Guilford.
Guilford's main factors:
G. General Activity. Fast-moving, energetic, drive for activity, feelings of adequacy and
self -confidence, impulsiveness.
A. Ascendance. Standing up for one's rights, social leadership, liking to be conspicuous,
taking the initiative.
Cattell
111
M. Masculinity versus Femininity. Has masculine interests, not emotionally excitable

or expressive, not easily aroused to fear or disgust, somewhat lacking in sympathy.
I. Confidence versus Inferiority Feelings. Feelings of acceptance, social poise, versus
self-centred, discontent, fear of social contacts, ideas of reference, guilt and restlessness.
N. Calmness versus Nervousness. Nervous, easily tired, restless, tense, easily annoyed
or depressed.
S. Sociability. Enjoys friends and social events, takes the role of leader, gregarious.
T. Reflectiveness. Enjoys serious thinking and introspection. Meditative and daydreaming.
D. Depression. Emotional depression, physical tiredness, anxious, emotionally labile
and lonely.
Cl Emotionality. Emotions easily aroused and perseverating. Emotionally immature,
wanting sympathy and fault-finding towards others.
R. Restraint. Not spontaneous or carefree, serious and avoids action and excitement.
O. Objectivity. Realistic, objective, not suspicious or hypersensitive.
Ag. Agreeableness. Friendly and compliant. Low scorers show resistance to control,
contempt for others, suspiciousness, doubts about the honesty of others, hostility and
aggression.
Co. Co-operativeness. Low scorers are fault-finding, suspicious, hostile, self-pitying
and egocentric.
There is a considerable degree of overlap between these primary traits; they are not
separate and independent. The correlation between N (Calmness) and D (Depression),
for instance, is -0.44. Guilford rejected the use of an oblique rotational solution which
makes it impossible to carry out a further factoring of the intercorrelations between the
primary factors. The main factors include several which might be expected to relate to the
neurotic disorders. The most relevant factors would perhaps be Depression (D) and
Nervousness (N), though Emotionality (C 1) might also relate to some of the anxiety
neuroses, and Restraint (R) could be expected to relate to obsessional and compulsive
disorders. There is, however, no general, higher-order factor of neuroticism in Guilford's
factor solution.
7.2 Cattell
Factor analytic studies of personality traits are largely based upon questionnaire data.
Like Guilford, Raymond Cattell made extensive use of questionnaire data, but he also
supplemented his investigations with data derived from objective tests (T-data), such as
blood pressure and GSR measures, and from behaviour ratings taken from everyday life
(life-record data or L-data). Although questionnaire data (Q-data) provide much of the
basis for trait factor analytic studies, Cattell makes clear the limitations of such information. Personality questionnaires have a valid place in research, but the ease with which
they can be abused has led to a seemingly endless stream of trivial articles: 'As a serious
approach to discovering the structure of personality it [the excessive reliance upon personality questionnaires] represents the nadir of scientific inventiveness and subtlety'
(Cattell 1946, p 341). Questionnaire data may be seen as existing in a world of its own,
and there has been comparatively little research aimed at establishing the relationship of
questionnaire data to overt behaviour.
112
Table 4. Cattell's first-order factors which discriminate between clinically judged neurotics and
normals
Factor
Neurotic loading
04
High levels of tension and frustration

Prone to feelings of guilt, worry and remorse
Depressed and lethargic
Emotional sensitivity, tender-minded, unable to face emergencies
Low ego strength, emotional maladjustment and dissatisfaction
Suspicious, jealous, wary of others, rigid defences
Dependent and submissive
Easily threatened, timid, withdrawn
F
I
C
L
E
H
In the analysis of data derived from different sources, Cattell obtained much the same
results with Q- and L-data, but found no simple one-to-one relationship between the
factors derived from Q- and L-data and those from T-data. As a result, Cattell's system is
extremely complex, since it contains primary factor solutions for all three types of data, as
well as second-order factor solutions. The primary trait factors (derived from Q- and Ldata) which relate to such aspects of neurosis as excessive worrying, irrational emotionality, feeling of inadequacy, hypochondria, low self-image, phobic, compulsive and somatic
symptoms are shown in Table 4. Each of these factors discriminates significantly between
clinically judged neurotics and normal subjects. As in Guilford's system, there is no single
neuroticism factor, and Cattell has stressed the need for a multifactor theory to cover the
various manifestations of neurosis (Cattell and Scheier 1961).
Second-order factors are broader dimensions which often correspond more closely to
common clinical concepts and permit discussion in terms of fewer categories. Cattell
recognised these higher-order factors as representing broader and more general features
of personality structure, but because of this he saw them as failing to account for as much
of the variance in specific behavioural events. The second-order factors that can be used
to discriminate between neurotics and normal subjects are F (Q) II, Anxiety, (clinically
judged neurotics are very much more anxious than normals) and F (Q) IV, Resignation
(neurotics are more likely to score low on measures of dominance and assertiveness). The
two remaining second-order factors approach, but just fail to achieve, statistically significant levels in discriminating between neurotic and normal groups. These are F (Q) I,
Invia-Exvio (clinically judged neurotics tend to be more introverted than normal subjects) and F (Q) III, Pathemia (neurotics show more evidence of emotional immaturity
and irrational hypersensitivity). As with the primary factors, Cattell finds no single factor
at the second-order level which is adequate to distinguish between neurotic and normal
groups; each of the factors is in some way essential to the definition of the neurotic
condition.
In the analysis of behavioural test (T -) data a number of investigations showed a single
anxiety-related factor at the first-order leyel. This contrasts with the several different
primary factors of tbis sort that emerged from the analysis of Q- and L-data (Cattell
1957b). The single, general anxiety factor (UI24), which emerged from the behavioural
and physiological measures, corresponds closely to the clinician's unitary concept and has
been identified as free-floating manifest anxiety. It includes such diverse manifestations
Cattell
113
of anxiety as irritability, emotionality, self-criticism, tremor and reduced salivary secretion. The concept of neurosis has traditionally been supposed to be closest to anxiety, and
neurotics do score highest on UI 24. Vsing this factor as a marker, Scheier (1972) ranked
some 20 clinical groups comprising over 2000 cases. The highest scores were recorded for
anxiety states (90th percentile), reactive depression (85th percentile), obsessive-compulsive neuroses (75th percentile) and conversion reaction (70th percentile). The anxiety
scores for psychosomatic disorders (in males) were much lower (50th percentile).
Nonetheless, neurotics also score highly on five or six other factors in Cattell's system,
and neurosis cannot be regarded as synonymous with VI 24: it is a composite of several
factors, of which VI 24 is only one. Cattell (1957b) found a seperate factor of neuroticism which is quite distinct from the general anxiety factor and which also discriminates
between neurotic and normal groups. This factor (VI 23) is characterised predominantly
in terms of lack of ego strength. This neurotic debility is distinct from the actual neurotic
behaviour which is determined partly by specific environmental stressors and by the preneurotic personality. Cattell refers to this factor as 'the essential ego weakness' which
shows itself in a varity of defence mechanisms, emotional instability and feelings of
inadequacy (Cattell op. cit.).
In the case of the anxiety factor, there remains the problem that analyses based upon
L- and Q-data fail to show any general primary factor of the sort that emerged from
analyses of T-data. However, the second-factor analyses established the higher-order
anxiety factor, F (Q) II (Cattell 1957b; Cattell and Scheier 1961). This factor is characterised by weak ego-strength (H-), suspicion (L+), gUilt-proneness (0+), low selfsentiment (Q3 -) and tenseness (Q4+). This higher-order Anxiety versus Dynamic
Integration factor may be identified with the general anxiety factor (VI 24) derived from
T-data. When Q- and T-measures have been used together, the second-order Q- and the
first-order T-pattem have emerged in the same factor. The most highly loaded primary
factor in this second-order questionnaire factor is Q3 ( - ), which may also be regarded as
the essential component of the higher-order factor. Q3 has been interpreted through its
content and associations as an expression of the strength with which the individual strives
to behave in accordance with a realistic and consciously approved self-image, and Cattell
(1957b) regards this factor as an indication of how anxiety is the result of a failure of the
dynamic integration of personality and behaviour.
It is interesting that Cattell sees the high loading of ergic tension (Q4) on the anxiety
factor as supportive of Freud's (1936) suggestion that anxiety should be proportional to
id pressure; 'Anxiety is ergic tension out of control' (Cattell 1957b). He also discusses
the relationship between low ego strength (C-) and anxiety in terms of Freudian theory,
though it is by no means clear what sort of relationship exists between the two. It is
possible that the weak ego responds to threat by generating anxiety, or that high levels of
anxiety, especially during early periods of development, lead to an impoverished selfimage. However, the three factor psychosexual theory of personality proposed by Freud
(oral, anal and phallic characters) bears little ressemblance to Cattell's system, and
neither Cattell's primary-, nor his secondary-factor solutions can be said to provide direct
support for Freudian theory (Cattell and Kline 1977).
114
7.3 Eysenck
In contrast to both Guilford and Cattell, Eysenck has always stressed the importance of a
general factor of neuroticism. Eysenck (1967) suggests that the widespread use of the
general concept of 'neurosis' reflects a universal recognition that there is something
common to each of the specific neurotic disorders, and that it is this common 'something'
that emerges from the factor analyses as neuroticism. Earlier concepts of this kind were
described by McDougall in his 'self-regarding sentiment', and both Janet's 'misere
psychologique' and Pavlov's notion of 'strength of nervous functioning' have some
resemblance to Eysenck's neuroticism factor (Eysenck 1947a, b). Eysenck refers to the
higher-order factors in his system as 'types', by which he means concentrations of correlated traits (Eysenck 1970b). As such, the difference between the concepts of type and
trait lies not in the continuity or the discontinuity of the hypothesised variable, nor in its
form of distribution, but in the greater inclusiveness of the type concept. Unfortunately,
the notion of types continues to carry implications of discontinuity, and for the present
discussion we will refer instead to higher-order trait factors. Eysenck has made extensive
use of hierarchical models to illustrate the different levels of generality involved in his
personality system, and a hierarchical model of the neuroticism factor is shown in Fig. 6.
Although Guilford and Cattell's trait factors clearly have some relationship to the
neuroses, they were more interested in primary factors than general higher-order trait
factors (Eysenck's type factors). As a result, their multifactorial questionnaires are difficult to relate directly to the clinical phenomena of neuroses. Eysenck, however, has
attempted to do this and developed a series of questionnaires. The first such inventory,
The Maudsley Medical Questionnaire, was succeeded by the Maudsley Personality
Inventory (MPI) , the Eysenck Personality Inventory (EPI) and, most recently, the
Eysenck Personality Questionnaire (EPQ).
Eysenck's studies (1947a, 1970) have offered substantial support for the view that
there are two separate and independent dimensions of neuroticism and psychoticism,
both of which could be regarded as continuous with normality. For this reason, it was
necessary to modify the earlier questionnaires to include a measure of psychoticism (P) as
well as the measures of neuroticism (N) and extraversion (E) that were contained in the
MPI and EPI. This was done in the EPQ (Eysenck and Eysenck 1975), which contains
measures of N, P and E. Each of these scales provides a measure of an underlying
personality trait which is assumed to be present to some degree in everyone (in normal
groups the terms emotionality and tough-mindedness can be substituted for the psychiatric terms neuroticism and psychoticism).
Guilford, Cattell and Eysenck all rely heavily upon factor-analytic methods in their
personality research, though factor analysis plays a lesser role in Eysenck's work. To him
it remains an introductory tool which can be used in the primary, descriptive function of
scientific enquiry, but which needs to be integrated with the main body of experimental
and theoretical psychology. Eysenck has been emphatic about the importance of linking
empirical research to a theoretical framework.
Despite the apparent differences in their work, there are a number of important areas
of agreement in the factor solutions of these three workers, though these only appear in
the higher-order factors. Guilford and Cattell have argued that a profile of many primary
personality factors has more practical value than a small number of higher-order factors.
On this point they disagree with Eysenck. Cattell, for instance, stated that first-order
Fig. 6. The structure of Eysenck's concept of neuroticism
Specific
response
level
Habitual
response
level
Primary factor
(Trait) level
Higher order
factor (Type)
tTl
......
......
Ut
~::
116
factors (and in particular the first-order factors derived from objective test data) provide
at least as much information as any of the higher-order factors (Cattell and Scheier
1961). Eysenck, on the other hand, has argued that it is the higher order factors that have
most psychological significance, and that little, if any, information is lost from the primary factors (Eysenck 1972). In addition, Eysenck has suggested that primary factors are
not always replicable across the sexes, whereas the higher-order factors are replicable in
this way.
The higher-order neuroticism factor that emerges from the questionnaires of all three
authors is virtually identical, though the primary-factor solutions have comparatively
little in common. The most comprehensive discussion of these issues (Eysenck and
Eysenck 1969) is based upon a massive factor-analytic study of the questionnaire items
and factors selected by the three authors. The analysis was conducted independently for
the male and female samples. At the highest-order factor level, the two factors of
extraversion and neuroticism were found in the male sample. In the female group, three
separate factors emerged. The first two were clearly extraversion and neuroticism,
whereas the third was difficult to identify and may have been largely an artifact. The
higher-order neuroticism factor that emerged from combined analysis of the Guilford,
Cattell and Eysenck factors contained all the items that had been identified with this
factor in the three separate analyses, and at the higher-order factor level, the same
factors emerged from the different sets of data. The primary factors, on the other hand,
were not replicable from the questionnaire data of one investigator to another.
Adcock (1965) concluded that the disagreement between Cattell and Eysenck was
based largely on differences of methodology regarding the number of factors to be
extracted, and especially with regard to the question of rotation. Eysenck's centroid
factor of neuroticism corresponds to Cattell's important second-order factor of anxiety.
In this light, the disagreement between the two authors is more apparent than real, and
Cattell's factor solutions may be regarded as supplementary rather than opposed to those
of Eysenck.
Eysenck (1960c) proposed that his system, whereby the various neurotic groups are
located within a two-dimensional framework (N X E) be used as a more reliable alternative to the categorical system of classification currently used in psychiatric diagnosis.
Within this dimensional system, it is suggested that all neurotics score highly on the
personality trait of neuroticism or emotionality. This N factor would appear to be essentially an emotional sensitivity which makes the person vulnerable to stress. The high N
scorer is described as being an anxious, worrying person, prone to moods and depressions. He is likely to sleep poorly and suffer from various psychosomatic disorders. The
dominant characteristic is an over-emotional disposition which responds strongly to a
wide range of stimuli and which takes some time to return to normal. These strong
emotional reactions may interfere with an appropriate adjustment to the person's environment and lead to irrational or rigid forms of behaviour. The second dimension (E)
may be used to differentiate between different neurotic groups. Those whose central
nervous system is innately predisposed towards excitation rather than inhibition will
develop introverted patterns of behaviour, and patients suffering from anxiety states,
reactive depression, phobias, obsessions or compulsive disorders tend to have high introversion scores: these are the dysthymic disorders. A combination of high N, high E, on
the other hand, is likely to lead to 'psychopathic' and hysterical disorders. This involvement of extraversion with neuroticism was suggested earlier by Carl Jung, who proposed
Eysenck
117
that in the case of a neurotic breakdown, extraverts were predisposed towards hysteria
and psychasthenia. Eysenck is fond of tracing this notion even further back to Galen's
doctrine of the four humours, particularly in the form given to it by Immanuel Kant (e. g.
Eysenck 1957). Within this system, the dysthymic neurotic suffering from anxiety, depression, obsessions, or phobias would be expected to fall into the melancholic quadrant,
with the hysterical disorders in the choleric quadrant.
It would be unwarranted, however, to assume that trait scores provide a direct measure
of personality structure (Eysenck 1967). This is well illustrated by the results of Ingham's
(1966) follow-up study. A group of 119 dysthymic and hysterical neurotic patients were
retested 3 years after leaving hospital. The mean N scores of. patients showing most
improvement had declined, and the N scores for patients who showed the least improvement were almost unchanged. It was also found that the mean E scores of the most
improved group had increased (again, those of the least-improved group were substantially the same). These findings suggest that trait-factor scores reflect changes in the
symptomatic status of the patients and should be used only as indirect measures of the
underlying personality structure.
Conditioning. theory plays a central role in Eysenck's theory of neurosis, and it is
predicted that introverts should condition more readily than extraverts. As a result,
introverted subjects might be expected to acquire the conditioned fears and anxieties
characteristic of the dysthymic states more easily than other people. The extraverted
neurotic, on the other hand, is predicted to be vulnerable to hysterical disorders and to
disorders of the second kind, such as nocturnal enuresis and the antisocial or
psychopathic disorders (see Chap. 4). In these cases, it is suggested that there has been a
failure of socially appropriate conditioning. The evidence on this point, however, is not
entirely clear. Subjects who score high on N and low on E do appear to be more prone to
dysthymic disorders, but hysterics, although high on N, are no more extraverted than
normal groups (Eysenck 1967). It is possible that an excessive reliance on questionnaire
measures may give a misleading impression, and Eysenck and Claridge (1962) conducted
an investigation of normal subjects, dysthymic and hysterical neurotics, using four objective laboratory tests (sedation threshhold, two choice reaction task measures, and spiral
after-effects) as well as questionnaire measures. A multiple discriminant function analysis was carried out, and two significant latent routes corresponding to N and E were
extracted. The hysterics were the most extraverted group and the dysthymics the most
introverted; on the neuroticism or emotionality variate, the normal subjects were the
least neurotic.
The relationship of high N, low E to the dysthymic neuroses relates to the work on
'manifest anxiety', since the Manifest Anxiety Scale (Spence 1964; Taylor 1953) is
loaded on both of Eysenck's dimensions of introversion and neuroticism (Eysenck 1965).
Therefore, both Eysenck and Spence would agree (though for different reasons) that
those individuals scoring high on this trait of manifest anxiety could be expected to form
conditioned responses more readily than low scorers. Gray (1970b) has plotted the
manifest anxiety scale against Eysenck's factors and suggests that the MAS covers the
range from stable extraversion to introverted neuroticism (i. e. it indicates the most rapid
increase in susceptibility to punishment).
Although Eysenck's theory works quite well for the dysthymic neuroses, the prediction
that individuals who behave in a criminal or psychopathic manner should score highly on
E and N must be regarded as unconfirmed. Although a number of studies (summarised in
118
Eysenck and Rachman 1965; Eysenck 1967) did find that criminals score high on both
these dimensions, other studies have failed to find any evidence that such individuals are
likely to be extraverted. Hoghughi and Forrest (1970) reviewed a number of studies of
young offenders which consistently found them to be more introverted than the population norms or than selected control groups. Other studies of drug addicts found comparatively low E scores, even among addicts with high levels of antisocial and criminal
behaviour in areas unrelated to drug abuse (Gossop 1978a, b; Gossop and Kristjansson
1977). With respect to antisocial forms of behaviour and criminality, therefore,
Eysenck's predictions have not been unequivocally supported by the research findings.
7.4 Criticisms of Trait Theory

It is one of the most ancient psychological beliefs that there are pervasive consistencies in
an individual's behaviour that show themselves across different situations. This assumption has been strongly challenged by several psychologists over the years. The earliest
challenge came from the Hartshorne and May studies of moral behaviour in children
(Hartshorne and May 1928; Hartshorne et al. 1929), which concluded that such traits as
deception, helpfulness, co-operativeness, persistence and sell-control were better seen as
groups of specific habits rather than as general traits. In recent years, there has been a
revival of criticism against the assumption of consistency in personality theory. Mischel
(1968, 1977) has been one of the major figures in this debate and has argued that
individuals show far less cross-situational consistency in their behaviour than has been
assumed by trait theorists (and by personality theorists in general). Mischel suggested
that the correlation coefficients of around + 0.30 that have been commonly found for the
consistency of behaviour across different situations reflect true behavioural variability
rather imperfect measurement techniques. Correlations of this size account for less than
10% of the variance, and Mischel suggests that their value for making any predictive
statements about behaviour is severely limited. This is a particular problem for the
clinician who is primarily concerned with individual assessment and treatment issues.
However, such correlations are fundamentally ambiguous. They can be used to support
the trait theory position or to point to the specificity of behaviour in different situations.
Mischel's criticisms have been mistakenly seen as the expression of a hard-line
environmentalist position which denies the existence of trait factors and attempts to
replace them with situational and environmental factors as the basic units of study. This is
not the case. Nor do trait theorists dismiss situational factors. Endler and Hunt (1969),
for instance, differentiated between situationally-evoked anxiety and the trait of anxiousness. The S-R Inventories of traits were explicitly designed to separate situations from
the modes of response which serve as indicators of reactions, states or traits. Trait anxiety
may be regarded as the typical or chronic level of anxiety. Cattell described trait anxiety
as a relatively permanent characteristic of individuals, and momentary state anxiety as a
condition which varies from day to day or from moment to moment and which reflects
the changes that occur in the person's environment (Cattell and Scheier 1961; Cattell and
Kline 1977).
Objectively, state anxiety can be characterised by loadings on such physiological variables as high systolic blood pressure, raised body temperature, low basal skin resistance,
increased heart rate, and in Q-data, state anxiety correlates with feelings of inability to
Criticisms of Trait Theory
119
cope. Cattell and Kline (1977) also offer an empirical distinction between trait and state
anxiety in terms of the lower scores on protension (L), guilt (0) and responsiveness to
threat (H) in state anxiety, though they point to some of the difficulties in maintaining
this distinction between the two; (e. g. how long can a state last before it should be
regarded as a trait?). Spielberger, who focused conceptual attention on this distinction
between traits and states, reported correlations of +0.44 and +0.67 between his A-trait
and A-state scales (Spielberger et al. 1968), and it has been suggested that both might fall
largely within Cattell's VI 24 factor: that is, they could be regarded as functionally
equivalent (Scheier 1972). There is a tendency in Cattell's work to incorporate environmentally-dependent states within the personality sphere, as when he argues for the
existence of state-liability traits corresponding to each particular state (Cattell and Kline
1977). This tendency to overrate the importance of traits at the expense of environmental factors has been ~me of the prime complaints of the social-cognitive learning theorists.
Nevertheless, it would be equally mistaken to overrate situational factors or to set up
pseudo-issues concerning the relative importance of persons versus situations. In one
study of anxiety (Endler and Hunt 1966), it was found that neither individual differences
nor situational factors alone contributed more than 4% to 6% of the total variance. The
specific types of behavioural response, on the other hand, accounted for as much as 25%
of the variance, and almost one-third of the variance came from simple interaction effects
(subjects X situations, subjects X modes of response and situations X modes of response). The authors suggest that anxiousness is idyosyncratically organised, and that the
validity coefficients for questionnaire measures of anxiety would probably be raised by
specifying the sort of situations in which response measures of anxiety are to be reported.
Another modification of the trait theory approach is that of Bern, who suggests that the
low correlations between trait measures and behaviour, as in the Hartshorne and May
studies, show that people are not consistent in the same way as each other, but not that
people are inconsistent with themselves. Trait research will only provide evidence of
cross-situational consistency if the individuals in the research share the investigator's
view of the equivalence of behaviours and situations. Those individuals who identify
themselves as being consistent with regard to a particular trait (e. g. anxiousness, introversion, friendliness) will be more consistent in the way they behave in different situations than others who see themselves as highly variable (Bern and Allen 1974). This
cognitive reinterpretation of trait theory has certain similarities to George Kelly's (1955)
personal construct theory, in which the individual is allowed to generate his own trait-like
constructs to characterise himself and his social world. The value of this self-attribution
approach to the neuroses has been shown in the recent study of depression by Lewinsohn
et al. (1979).
Few (if any) psychologists would deny that behaviour is dependent upon a complex
interaction of factors. Trait theorists as well as social-cognitive theorists agree on this
point. No trait theorist has asserted that such complex forms of behaviour as the neurotic
disorders are a result of personality factors alone. Cattell and Kline (1977) specify an
equation, which, whatever its other limitations, includes stimulus, response and situational factors, together with trait factors as multiple determinants of behaviour. They also
take account of cognitive factors (viz. the individual's perceptions). To this extent, the
disagreement between the social-cognitive and trait theorists is essentially one of the
relative utility of trait and cognitive formulations in the clinical context (Certainly this
seems to have been one of Mischel's primary concerns in Personality and Assessment).
120
The events that occur in the outside world and the phenomena that are studied by
psychologists can be categorised and interpreted in many alternative ways. The number
of alternatives is limited only by the investigator's inventiveness. The choice between the
available alternatives is guided not by some absolute criterion of truth or reality but by
assessing their usefulness, which in tum depends upon the investigator's purposes. This
question of the utility of trait theory and social-cognitive approaches to clinical problems
has been consistently stressed by Mischel (1968,1973,1979). Traits are useful as characterisations of what people are like 'in general', and as such they may show considerable
reliability and stability over a period of time. At lower and more specific behavioural
levels, however, categorisations tend to show the discriminativeness of behaViour in
relation to different situations and give greater weight to the within-individual variance in
behaviour than to measures of central tendency (Mischel 1979).
For the particular purposes of the clinician, it has been suggested that generalised traits
may be too crude and gross as units for the analysis of complex behaviour, and that noncognitive personality dispositions have far less effect upon behaviour than both trait and
psycho-dynamic personality theorists have assumed (Mischel 1969). Trait theory formulations may even have helped to obscure the situational-specificity of behaviour in the
attempt to demonstrate cross-situational generality. Ajzen and Fishbein (1977) have
shown that the best predictors of a given behaviour are likely to be tailored to that
specific behaviour. Equally, for clinical purposes (e. g. in setting up an individual treatment programme) an analysis of specific behaviours as they occur in a specific context is
likely to be more useful than knowledge of the individual's position on one or more trait
dimensions. Against this, it has been argued that trait theory is not intended to be used
for the prediction of specific behaviours, but for wider scientific purposes of personality
description.
A less tangible benefit of these social-cognitive criticisms is that they may help to
correct the mechanistic view of human behaviour that is sometimes associated with other
theories. Explanatory frameworks that are based upon stimulus-response theory, trait
theory or psycho-analytic theory sometimes appear to imply that human action is the
passive product of processes or forces over which the individual has no control. This can
be seriously misleading. The child's ability to delay gratification, for instance, is clearly
related to the cognitive strategies that he or she uses: the person's interpretation of his
environment has a powerful effect upon the way that the environment will influence him
(Mischel 1973). Even in areas such as drug addiction, where the individual appears to act
under compulsion and might therefore be expected to have less control over his own
behaviour, the person's attitudes, beliefs and expectations remain important factors
(Eiser 1979; Gossop 1979).
8. The Physiology and Genetics of Neurosis
8.1 Physiology and Neurosis

Emotional states may be gauged in terms of the subjective experiences of the individual,
his behavioural responses and his physiological reactions. Fear, for instance, has been
described as an amalgam of apprehension, autonomic disturbances and avoidance
behaviour (Rachman 1974); and these three components are now known to show some
degree of independence (Rachman and Hodgson 1974). The physiological components
of neurosis have always been important, both to our understanding of the neuroses and as
a pointer to the physiological bases of personality.
The principal feature of the emotional states that distinguishes them from other mental
phenomena is the extent to which autonomic arousal is an integral part of the emotions
(Sartory and Lader 1980). Most psychophysiological responses associated with the emotions originate in the autonomic nervous system (ANS). This is divided into the sympathetic and para-sympathetic systems. The brain centre most importantly involved in
the activation and regulation of these systems is the hypothalamus: the posterior
hypothalamus primarily activates the sympathetic system, and the anterior
hypothalamus, the para-sympathetic system. Strong sympathetic arousal is often indicated by increased heart rate and breathing, increased systolic blood pressure, dryness in
the mouth and increased sweating.
In addition to producing a general sympathetic response, activation of the posterior
hypothalamus leads to the release of hormones which stimulate the pituitary gland to
secrete adrenocorticotrophic hormone (ACfH) into the blood stream: this hormone in
tum causes the adrenal gland to secrete adrenocortical hormones (ACH). These latter
hormones appear to help the organism in its response to stress, and the interaction of
these two glands is generally referred to as the pituitary-adrenal axis.
When exposed to a wide range of stressful stimuli (whether emotional or physical) for a
prolonged period of time, the pituitary-adrenal axis responds with certain distinctive
changes. The sensitivity of the sympathetic-adrenomedullary system to certain
psychological stimuli was investigated by W. B. Cannon during the early part of the
century and this led to the 'emergency function' theory of adrenomedullary activity.
More recently, Fahrenberg (1977) suggested that high levels of ACH excretion may
indicate an alarm reaction as well as a constitutional hyper-reactivity of neuro-endocrine
mechanisms; and low excretion levels may point to stability as well as exhaustion. The
interpretation of such measures will depend upon knowledge of other factors.
Hans Selye's study of the body's responses to stress led to the formulation of the
General Adaptation Syndrome (GAS) (Selye 1957). This emphasised the general, nonspecific nature of the reactions. Selye suggested that many morphological, functional and
122
biochemical changes are common to the stress reaction, irrespective of the specific eliciting stimulus. The three phases of the GAS were the alarm reaction, in which the body's
defences were mobilised, the stage of resistance, in which the body attempts to adapt to
the stress and finally, the stage of exhaustion. This stress syndrome was seen as
pathogenic, and Selye (1950) noted the striking similarity between the GAS and the
combat neuroses observed in soldiers during prolonged exposure to the stresses of warfare. Under extreme conditions of this sort, a whole range of psychosomatic disorders
may emerge, including disturbances of cardiac rhythm and blood pressure, excessive
sweating, bodily aches and pains, dizziness and fainting, paraesthesias, loss of libido or
impotence in the male, insomnia at night coupled with sleepiness during the day, morbid
fears, depression and compulsive thinking. Selye suggested that disorders of this sort are
fairly typical of the body's response to any prolonged systemic stress.
Despite the early excitement about the GAS, it has received much less attention in
recent years. It seems increasingly likely that the body's physiological reactions to stress
are not so general as Selye suggested, and the notion of stress is itself surrounded by a
number of conceptual problems. Lader (1975), for instance, suggested that 'the concept
of stress is so diffuse as to be not merely meaningless but even misleading.'
There have been other interesting studies of this problem of how the adrenomedullary
system responds to stressful stimuli. Weiss et al. (1976), for instance, proposed an alternative explanation for the learned helplessness findings. Under experimental conditions
in which they could not control shock, animals showed more severe gastric lesions,
greater loss of body weight, higher plasma steroid levels and more fearfulness than
controls (Weiss 1968). Overmier and Seligman (1967) stressed the escape-avoidance
learning deficit of animals exposed to inescapable shock, but argued that there was no
necessary connection between the stress-induced physiological changes and the
behavioural (avoidance) deficit. In contrast to Seligman's learned helplessness explanation of depression, Weiss proposed that the observed behavioural deficits were directly
related to physiological changes.
In the Overmier and Seligman (1967) study it was found that the escape-avoidance
deficit which was present at 24-hour retesting had disappeared after 48 hours. This rapid
dissipation of the deficit is not typical of learned responses, and Weiss suggested that the
results were caused by some physiological imbalance which corrects itself with time.
After a series of experimental investigations, Weiss et al. (1976) proposed that endogenous noradrenaline is differentially affected under conditions of escapable and inescapable shock, with excapable shock producing higher levels. More pertinent to Weiss's case
were the findings that experimental conditions which seemed to involve helplessness
without any reduction of noradrenergic activity in the brain led to no escape-avoidance
deficit, and that conditions which did not involve helplessness but which did reduce
noradrenaline levels led to a similar deficit to that observed by Overmier and Seligman.
Inescapable shock appears to be an especially potent condition, since even a single
exposure was found to cause marked depletion of noradrenaline concentrations in the
hypothalamic telencephalic regions of the brain. In a further investigation, Weiss and his
colleagues administered a monoamine oxidase inhibitor to the experimental animals with
the purpose of preventing the degradation of monoamines consequent upon inescapable
shock. Such animals showed no behavioural deficit after the experimental session,
whereas controls who had not received the MAO inhibitor showed the usual escapeavoidance deficit.
Physiology and Neurosis
123
The authors concluded that the depressive phenomena described by Overmier and
Seligman (1967) and by Seligman and Maier (1967) were not a result of learned helplessness as such, but of a disturbance of central catecholamines. In particular, Weiss's explanation emphasised the role of noradrenaline depletion as the principal determinant,
though other neurotransmittors may also be involved.
Psychophysiology is concerned with those physiological activities which underlie or
relate to psychological functions, and a considerable amount of research has been concerned with the somatic concomitants of the neurotic disorders, and in particular of
anxiety. In an investigation of the psychophysiological correlates of morbid anxiety,
Lader and Wing (1966) found that the basal skin conductance level and the number of
spontaneous fluctuations in skin conductance were both higher in patients than in normal
subjects at rest: spontaneous fluctuations were also more frequent for the anxious
patients throughout the experiment. Other studies found significantly greater forearm
blood flow among anxious patients at rest than in normal subjects (Kelly 1966; Kelly and
Walter 1968), and Howe (1958) found palmar skin resistance levels at rest were lower
(increased sweating) among patients with chronic anxiety states. Other findings include
elevation of basal heart rate and blood pressure among anxious subjects (Lader and Wing
1966; Goldstein 1964). These results fit in with Malmo's (1957) suggestion that anxious
patients are over-aroused.
It has been widely assumed that neurotics (particularly anxiety neurotics) are more
'emotional' and therefore more autonomically labile than normal groups. However, the
results of studies in which subjects have been presented with stimulation have yielded
directly conflicting results. One group of investigators found that highly anxious subjects
were more reactive than normal subjects. Goldstein (1964), for instance, found that
systolic and diastolic blood pressure, respiration rate and heart rate were all higher in
anxious patients than controls, both at rest and in response to stress. Similarly, Malmo
and Shagass (1952) investigated the changes in blood pressure in a group of psychoneurotic patients under stress. Their results suggest that blood pressure is elevated in
neurotics when they are under stress, and that this may indicate a greater and more
prolonged change of blood pressure in response to stress among neurotics than in normal
subjects.
Wing (1964), on the other hand, presented a difficult task to a group of patients
suffering from anxiety states and measured skin conductance, pulse rate and EMG (muscle tension). The task produced short-lived physiological disturbances in both patients
and controls. On all three measures the patients' mean levels were higher than those of
the controls before, during and after the task, though the controls showed more responsiveness to stimulation than the patients, particularly in the case of the EMG. Lader and
Wing (1966) found that the psychogalvanic reflex was smaller in patients than for controls in response to stimulation, and Kelly and Walter (1968) also found that anxious
patients showed less physiological response to stressful stimulation than controls: normal
subjects showed the greatest percentage increases in forearm blood flow, heart rate and
self-rated anxiety when compared with a variety of neurotic and other psychiatric groups.
These authors suggest that it is not the neurotic's ability to feel anxious which is abnormal, but their inability to relax in the absence of threatening stimuli. In any case, the
assumption that anxiety neuroses reflect an underlying emotional hypersensitivity would
appear to be an over-simplification. Lader and Wing (1966) have argued that autonomic
reactivity is inversely related to the basal levels of activity in the sympathetic nervous
124
system, and that powerful stimuli tend to reduce the differences between anxious patients
and normals.
A recent overview of the psychophysiological literature (Sartory and Lader 1980)
pointed to the following conclusions: anxiety states, and neuroses characterised by fearfulness and anxiety, seem to involve increased heart rate and increased forearm blood
flow. Studies of muscle tension have led to inconsistent findings, possibly because of
differences in the experimental methodologies, and different indices of electrodermal
activity have also produced different results. Skin conductance level itself is a comparatively crude measure which only differentiates extreme groups, and skin resistance
response is a complicated measure which probably represents an orienting reaction.
Measures of spontaneous fluctuations in skin conductance may be a more valuable index
of anxiety.
In other areas there is greater agreement in the experimental findings. It is clear that
~nxio\ls patients take longer than normal both to adapt to changes in the experimental
conditions and to habituate to stimuli. This has been found in psychophysiological studies
of blood pressure (Malmo and Shagass 1952), GSR (Lader and Wing 1966) and EMG
(Davis et al. 1954).
One reason for some of the conflicting results may lie in the use of inadequately
defined patient groups such as 'neurotics'. This category covers such a wide range of
disorders that a more precise specification of the patient characteristics is necessary before the experimental results can be properly interpreted. Lader (1967, 1969)
has cautioned against the assumption that neurosis has any functional unity, and the
investigation of psychophysiological parameters has been quite valuable in this context.
In a study of 90 patients with various psychoneurotic disorders, Lader (1967) measured changes in skin conductance levels during sound stimulation, rate of habituation
and spontaneous fluctuations. Patients with specific phobias were differentiated from
those with other affective disorders (including patients with social phobias and
agoraphobia). The specific phobics habituated rapidly and had fewer spontaneous fluctuations (lower arousal), and in this respect they were most like normal subjects. Among
the other patient groups, levels of arousal were higher than normal, as would be expected
for anxious subjects, though it is interesting that increased arousal was also found in the
agoraphobic and socially phobic patients who were not actually in anxiety-provoking
situations. In an investigation of the clinical implications of this finding, Lader et al.
(1967) found that specific phobics differed from social phobics and agoraphobics on
clinical grounds, with earlier age of onset of symptoms and lower debility. The specific
phobics showed the best response to desensitisation both in terms of reduction of clinical
symptoms and in improvement in social functioning, and the authors conclude that
autonomic measures can be a valuable means of predicting treatment outcome among
phobic patients. Kelly and Walter (1968) also found that chronically anxious patients
could be distinguished from controls by measures of basal forearm blood flow. On this
measure the mean score of the anxious patients was more than twice as high as that of
normal subjects, and higher than all other diagnostic groups in the study (agitated and
non-agitated depressions, schizophrenia, obsessional neurosis, phobic states, hysteria,
personality disorder and depersonalisation disorders). Both Kelly and Walter (1968) and
Lader (1967) suggested that phobic patients have the closest resemblance to normal
controls, although different physiological measures were used in the two studies. In the
Eysenck
125
Kelly and Walter results, patients with phobic and chronic anxiety states differed on
almost every measure used.
Another study of 240 patients with symptoms of neurotic anxiety or depression was
conducted by Frith et al. (1978). This showed a close relationship between symptoms of
anxiety and depression (both self- and observer-rated). On physiological measures, however, the two groups could be differentiated - the anxiety neurotics taking longer than
the depressed neurotics to habituate to a novel stimulus. When variously treated with
amitriptyline (a tricyclic antidepressant), diazepam (an anxiolytic), both drugs together,
or a placebo, Frith and his colleagues found that amitriptyline gave greater benefit to the
depressed patients than the other treatments. Unexpectedly, the antidepressant also gave
greatest benefit to the anxious patients, though diazepam did produce some improvement among the anxiety cases in terms of their physiological response: their electrodermal habituation times increased. This suggests that diazepam may alter the peripheral
signs of anxiety without having any necessary effect upon the more complex subjective
and behavioural manifestations.
On the whole, the autonomic accompaniments of depression have been less well
researched than those associated with anxiety. This is partly because depression cannot
be experimentally induced in the same way as anxiety. A further problem is that depression seldom occurs without at least some anxiety, and this may itself produce some of the
observed physiological changes. Shagass and Jones (1958) found that the sedation
threshold was low in psychotic depression and high in neurotic depression. This finding
supports the distinction between the two types of depressive disorder discussed earlier in
the book. However, because barbiturate plasma levels after a given dose tend to be
inversely related to muscle blood flow (which has already be shown to be associated with
anxiety level), sedation threshold may be as much a measure of the state of the peripheral
vascular system as of any central process (Lader and Noble 1975). Similar difficulties
arise in the case of EMG and GSR recordings, both of which are known to be affected by
anxiety.
8.2 Eysenck
Eysenck's theory of the neuroses requires two descriptive dimensions and probably two
causal processes, both of which are closely linked to physiological and genetic factors.
The dimension of neuroticism is conceived in terms of individual differences in the
reactivity and lability of the autonomic nervous system, and the autonomic lability itself is
seen as an inherited characteristic of the organism. Eysenck and Rachman (1965) suggest
that some people are 'innately predisposed to respond more strongly, more lastingly, and
more quickly with their autonomic nervous system' when presented with stressful stimuli.
This notion that neurotic patients (particularly those with anxiety neuroses) might
suffer from some sort of autonomic lability or over-reactivity is not new. Many psychiatrists and psychologists with a physiological training have subscribed to this sort of view,
and Cattell like Eysenck favours two distinct concepts of physiological arousal. The first
is a general autonomic-endocrine activation which corresponds most closely to the anxiety trait factor UI 24 and to the anxiety state factor (SUI 9). This state anxiety factor is
associated with raised heart and respiration rates, high systolic blood pressure, low resting skin resistance, low serum cholinesterase and high plasma 17-ketosteroids (Fahren-
126
berg 1977). Cattell also includes cortical arousal in his system. This corresponds mainly
to his VI 22 factor and closely resembles the concept of cortical arousal, which plays such
a central role in Eysenck's theory.
Eysenck's trait factor dimension, introversion-extraversion, is linked to the processes
of inhibition and excitation. Indeed, it is this dimension of introversion-extraversion and
the closely related concepts of inhibition and excitation that have received most attention
in Eysenck's work. (Inhibition and excitation are used here to refer to properties of the
central nervous system, and not to aspects of behaviour.) Extraverts are said to be
characterised by a predominance of inhibitory activity, and introverts by a predominance
of excitatory forces (Eysenck 1957). Eysenck formulates these effects in terms of reactive
inhibition - a concept which can be traced back through the work of Clark Hull to that of
Pavlov. In those people with an autonomic predisposition towards neurosis, it is this
balance of inhibitory and excitatory forces that determines the development of specific
neurotic symptoms. Like the autonomic lability which is the physiological substrate of N,
this balance of forces is also regarded as constitutionally inherited.
Individual differences in neuroticism are related to differential thresholds of arousal in
the visceral brain, and behavioural differences with respect to extraversion are related to
differential thresholds in various parts of the ascending reticular activating system
(ARAS). It is suggested that cortical arousal can be produced in two ways. It may be a
result of sensory stimulation or of higher mental processes, in which case there need be
no arousal of the viceral brain, or it may be produced by emotion. In this latter case, there
is both cortical and autonomic arousal. Autonomic activation and cortical arousal are,
therefore, partially independent, and Eysenck cautions against any assumption that such
measures of cortical arousal as EEG, GSR or EMG can be used as direct indices of
emotional involvement.
Eysenck provides support for his theory with evidence derived from a wide range of
experimental techniques, including sedation threshold, spiral after-effect and vigilance.
Extensive reviews of these findings are offered both by Eysenck (1967) and by Claridge
(1967). With respect to sedation threshold, for instance, Eysenck's theory predicts that
extraverts (and extraverted neurotics such as hysterics, psychopaths and personality disorders) would show low thresholds, and that introverts (and dysthymic neurotics) would
show high thresholds. This result was obtained by Shagass and Jones (1958) in a
neurophysiological investigation of different diagnostic groups. Patients with hysteria had
the lowest sedation threshold, followed by patients diagnosed as suffering from mixed
neurosis and anxiety hysteria (phobic disorders). The highest threshold was obtained for
patients with anxiety states, followed by neurotic depressives.
Similar supportive evidence has been obtained using the spiral after-effect. Duration of
the after-effect should be less in extraverts than in introverts because of the higher levels
of reactive inhibition in the former group, and although there have been a few failures to
confirm this effect, most studies did find the predicted differences between introverted
and extraverted subjects. Claridge (1967), however, has argued that Eysenck's excitation-inhibition hypothesis, insofar as it is linked to the single dimension of introversionextraversion (or to a single underlying causal process), is inadequate. Instead, Claridge
suggests that two dimensions are necessary to account for the results of experimental and
physiological studies with neurotic groups.
Gray
127
8.3 Claridge
Claridge's theory may be regarded as a reinterpretation of Eysenck's theory in arousal
terms. Whereas Eysenck placed excitation and inhibition on a single dimension and
described them as reciprocally related, Claridge suggests that two dimensions are
required and that these are congruent in activity. Claridge (1960) suggests that Eysenck's
introversion-extraversion and neuroticism dimensions are not functionally independent,
and that both may be regarded as dimensions of arousal. On a number of objective tests,
the performance of hysterics and dysthymics failed to match their scores on one of
Eysenck's personality questionnaires (the MPI) (Claridge 1967). In a principal components analysis of the objective test data (e. g. sedation threshold and spiral after-effect)
and of the questionnaire measures of extraversion and neuroticism, two correlated factors emerged; both were related to arousal. On the basis of these results, Claridge
proposed a theory based upon two functionally related arousal mechanisms. The first is
identified as a tonic arousal system. This maintains the individual's gross level of arousal.
The second is described as the arousal modulating system, and its functions are to regulate the level of activity in the tonic arousal system and to integrate stimulus input into
both systems. The two systems may be seen as related in such a way as to provide a
negative feedback loop regulating the level of arousal.
Oaridge, like Eysenck, is mainly concerned to explain the differences between hysteria
and the dysthymic neuroses. But unlike Eysenck, Claridge proposed that where there are
high levels of activity in the arousal modulating system, this leads simultaneously to high
levels of inhibition in the arousal modulating system to match the high levels of activity in
the tonic arousal system. In the dysthymic neuroses, high tonic arousal is matched by high
levels of inhibition in the arousal modulating system. In the hysterical and psychopathic
neuroses, the reverse is true. Tonic arousal and (inhibitory) arousal modulation are both
at a low level.
One advantage of this theory is that it can account for the failure to find high E scores
in hysterical and psychopathic groups (McGuire et al. 1963; Gossop and Kristjansson
1977; Hoghughi and Forrest 1970). On the other hand, Oaridge's two dimensions and
the hypothesised tonic arousal and arousal modulating systems are comparatively difficult to reconcile with physiological findings (as Claridge acknowledges). They operate
at a conceptual and theoretical level rather than at the level of neuro-anatomy or physiology.
8.4 Gray
Gray (1972) has also proposed a psychophysiological theory of personality, which may
be seen as an extension and modification of Eysenck's (1957,1967) theory of introversion-extraversion. Instead of the ascending reticular activating system, Gray suggests that
a more extensive system acts as the physiological substrate of introversion. This consists
of the ARAS together with the medial septal area, the hippocampus and the orbital
frontal cortex and their interconnections.
The barbiturate drugs and alcohol both antagonise the activity of the physiological
system which Gray labels the punishment mechanism, without affecting the reward
mechanism. Gray argues that the ARAS is unlikely to be the chief site of action for this
sensitivity to specific types of reinforcement (i. e. reward or punishment), and the hip-
128
pocampus and medial septal area are suggested as an alternative site of action. The
behavioural effects of the barbiturates and alcohol seem to be due to an action on the
septal mechanism for production of the hippocampal theta rhythm (Gray 1970a). More
precisely, amobarbital appears to affect behaviour by antagonising the normal theta
rhythm response to punishment and frustrative non-reward (Gray 1972). Whereas
Eysenck (1957) links introversion to increased cortical arousal in the ARAS, Gray
regards the level of introversion as determined by the amount of activity in the septohippocampal system. This operates as a negative feedback loop consisting of the orbital
frontal cortex, the medial septal area and hippocampus as well as the ARAS.
Eysenck's theory places considerable weight upon the notion of conditionability as a
general personality feature. This is attributed to the relatively high level of arousal in
introverts who are regarded as over-socialized: (by socialization Eysenck primarily
means the acquisition of conditioned fear responses; e. g. Eysenck 1960b). In addition,
Eysenck uses the notion of general conditionability to account for the important
behavioural and psychiatric differences between the dysthymic, and the hysterical and
psychopathic disorders. However, it seems that introverts condition better than
extraverts only under certain conditions. In general, introverts seem to condition better
under conditions of under-arousal; whereas under conditions of over-arousal, extraverts
perform comparatively well (Eysenck 1967).
There is a considerable body of evidence showing systematic differences in arousal
between introverts and extraverts. However, after Seligman's (1970) discussion of the
inadequacy of the general process view of learning, there is increasing doubt that conditionability can be regarded as a unitary concept. Since much of Eysenck's evidence
rests upon the results of eye-blink conditioning studies, his theory would be considerably
weakened by studies demonstrating the specificity of conditioned connections.
Gray (1970b) rejected Eysenck's conditionability interpretation of introversion and
replaced the notion of conditionability (thought to be greater in introverts than
extraverts) with that of susceptibility to punishment and frustrative non-reward. This is
also supposed to be greater among introverted subjects. Because the introvert is more
highly aroused, Gray (1972, 1976) regards him as more susceptible to punishment.
Conversely, the extravert is described as being 'bad at fear' (Gray 1971). Gray has used
this to explain, for instance, the psychopathic features of criminality and antisocial
behaviour in the absence of anxiety or guilt, and the recidivism that accompanies
psychopathic behaviour (Gray 1970 b). Psychopathic behaviour is seen as the tendency of
the extraverted neurotic to seek reward without fear of punishment, and the dysthymic
symptoms of the introverted neurotic can usually be seen as a clear expression of fear.
Even in the case of the obsessional who performs his rituals in a state of apparent calm,
any interference with the obsessions is likely to cause overt and sometimes intense fear.
On this point, Gray's theory runs into the same difficulties as that of Eysenck (see
previous chapter): individuals who behave in a criminal or psychopathic manner do not
consistently show high levels of extraversion.
8.S Genetics and Neurosis

Virtually all psychologists recognise that individual differences are a salient feature of
any behavioural analysis. On the question of whether or not there may be a genetic
component underlying the observed individual differences, however, there is consider-
Studies of Twins
129
able (and often acrimonious) debate, though there has always been an interest in genetic
determinants of the neurotic disorders. Maudsley (1899) commented: 'There is a destiny
made for each man by his inheritance; he is the necessary organic consequent of certain
organic antecedents', and Freud, in his General Theory of the Neuroses (1917), stressed
the inherited component of libidinal development and referred to the hereditary sexual
constitution. More recently, Eysenck & Cattell have been concerned with the genetic
determinants of personality, and most psychologists with a biological training would be
prepared to accept that a predisposition towards nervousness could be inherited.
The genotype (or the biological inheritance transmitted by genes) cannot be measured
directly, and the phenotype (or observable characteristics of an individual) is not itself
inherited. However, by selecting animals according to phenotype and mating like with
like, it is possible to produce phenotypic differences. Broadhurst's selective breeding for
extremes of emotionality in the rat led to the establishment of the Maudsley reactive and
non-reactive strains (Broadhurst 1978). This emotionality factor may be comparable to
the autonomic lability which has been suggested as an underlying determinant of human
neuroticism (Eysenck 1967).
8.6 Studies of Twins

One of the most useful vehicles for the investigation of genetic effects is the study of
twins. Its basic rationale was first propounded by Galton in 1875, and twin studies have
been regarded as a sort of natural experiment allowing the differentiation of genetic ahd
environmental influences. Identical (monozygotic) twins are born with identical genotypes, whereas in genetic terms, fraternal (dizygotic) twins are no more alike than ordinary
siblings. Estimates of genetic effects are sometimes based upon the difference in correlation between identical and same-sex fraternal twins, though this method is not without its
limitations, because of the possibility that environmental effects for identical twins may
differ in subtle ways from those for other twins. Nonetheless, in studies of a particular
trait or illness,-results pointing to higher concordance rates in identical than in fraternal
twins are suggestive of some underlying genetic factor. Another type of investigation has
looked at the similarities between identical twins who have been reared apart.
Rutter et al. (1963) carried out a longitudinal study of the development from birth of
56 siblings and twins, in which several behaviour patterns (the authors refer to these as
'primary reaction patterns') were investigated. Evidence of genetic influence was
obtained for most of these primary reaction patterns, though it is interesting that those
categories in which the evidence for a genetic basis was strongest were also those showing
the greatest instability over time. Among these, was the approach-withdrawal pattern,
which may well be related to fearfulness, and which has been found to be a precursor of
later childhood behaviour disorders (Thomas et al. 1968). Conversely, Rutterfound that
those patterns showing the greatest stability gave the least indication of genetic determination.
On autonomic measures, identical twins have been found to be more alike on measures
of heart rate than fraternal twins (Vandenberg et al. 1965) and there is also a high degree
of similarity between identical twins in terms of palmar skin resistance (Block 1967).
Lader and Wing (1966) compared 11 pairs of monozygotic and 11 pairs of dizygotic
twins on a number of physiological measures which differentiated between anxiety states
130
and normals. On measures of the habituation of electrodermal response, a correlation of

+0.75 was obtained for MZ, and +0.13 for DZ twins; the spontaneous GSR fluctuations towards the end of the experimental session correlated +0.68 for MZ twins and
-0.02 for DZ twins; and pulse rate correlated +0.78 for MZ and -0.38 for DZ twins.
The negative correlations obtained for the DZ group are puzzling. If a genetic factor is
operative, correlations for the DZ group should be lower than for MZ subjects but they
should still be positive. It is possible that this result is a reflection of error variance caused
by the small groups used in the study.
This does not, however, alter the other findings, and the authors conclude that habituation of GSRs, the number of spontaneous fluctuations and pulse rate are subject to
genetic influences. Another study of 84 male twins (Rust 1974) also reported a substantial genetic effect (of more than 60%) underlying electrodermal and heart rate basal
measures, electrodermal response and spontaneous activity. This investigation failed to
find any genetic component for habituation of electrodermal response, as in the Lader
and Wing study.
Other studies of twins have used personality questionnaire data. On Cattell's High
School Personality Questionnaire the results from over 100 MZ and 100 DZ pairs of
twins suggest that there is greater resemblance between identical than fraternal twins on
the General Neuroticism versus Ego-Strength Factor. Cattell and Scheier (1961) found
that their main first-order anxiety factors were determined primarily by environmental
rather than genetic factors, and in a study using the MMPI and HSPQ, Gottesman (1963)
found comparatively low heritability indices for hypochondriasis and hysteria. This study
did, however, report a substantial inherited component for social introversion (0.71),
psychopathic deviation (0.50) and depression (0.45). The Introversion-Extraversion factor has consistently been found to have a considerable genetic component (Slater and
Shields 1969). The results for neuroticism have been less consistent, though Shields
(1962) found correlations of +0.38 on neuroticism for MZ twins reared together and
+0.53 for twins reared apart. The correlation between N scores for DZ twins reared
together was much lower (+0.11). An interesting feature of these results is that MZ
twins reared apart appeared to be more similar on measure of neuroticism and extraversion than MZ twins reared together. It is not at all clear why this result was obtained,
though it has also been reported in other investigations (e. g. Wilde 1964).
Clinical studies are undoubtedly hampered by the low reliability and validity of
psychiatric diagnoses, but several interesting studies have been conducted in this area. If
genetic factors contribute towards the determination of certain neurotic disorders, one
might expect the relatives of neurotics to be more likely to show similar disorders. Brown
(1942) looked at the incidence of neurotic disorders among the parents and siblings of
anxiety neurotics, hysterics and obsessional neurotics. The incidence of neurosis among
close relatives was found to be much higher than would be expected by chance, and there
was marked variation in the tendency of first-degree relatives to share a similar diagnosis
to the proband. For anxiety states 15.1 % of the relatives were given the same diagnosis;
for hysteria the rate was 11.2%, and 6.9% for obsessional neurosis. Unfortunately,
there are massive problems involved in interpreting the results in studies of this sort.
No correlation can be made for closeness of kinship and the expected rates for a
given disorder, nor is there any control for environmental determinants. In addition, it is difficult to know whether or not experimenter bias may have distorted the
results.
Studies of Twins
131
Slater and Shields (1969) reported a study of 146 cases who had been diagnosed as
suffering from a neurotic or personality disorder and who had a twin partner. Of those
individuals diagnosed as suffering from an anxiety state, 41 % of the MZ twin partners
were assigned the same diagnosis, against only 4 % of the DZ twin partners. For the other
neuroses (mostly reactive depressions), none of the twin partners received the same
diagnosis. These results point towards some degree of genetic specificity for the anxiety
states, but give no suggestion of genetic determination for the other neuroses. Neurotic
depression, for instance, appears to have no specific genetic component, though there is
some evidence that genetic factors do operate in endogenous or psychotic forms of
depression.
In a review of ten studies of twins involving 1264 pairs of twins, Pollin (1976) noted
that the concordance rates for neurosis between MZ twins were reliably found to be
about twice as great as those for DZ twins. This increased tendency for both MZ twins to
show signs of a neurotic disorder, although marked, was much less than the concordance
rates found for schizophrenia. In this latter case, where one twin had a schizophrenic
disorder, the MZ partner was 31/2-6 times more likely to have the same disorder than the
DZ partner. These results suggest that some genetic component operates in the neuroses,
but that there is a stronger genetic factor in the aetiology of schizophrenia.
The gist of these findings is that genetic factors appear to be involved in the determination of those neuroses which are primarily characterised by anxiety, but not in other
neurotic disorders. It should not need to be restated that this does not imply that genetic
factors may therefore be invoked as 'causes' of anxiety neuroses, or that environmental
factors are of lesser significance.
It would be entirely futile to return to this sort of genetics versus environmental
dichotomy, and several investigators have begun to express doubts about the utility of
this general approach to the problem. Caspari (1967), for instance, suggests that questions of whether or not a given behaviour has a genetic component, or even how much of
the variance can be attributed to genetic and environmental factors, may be misleading.
A more appropriate formulation (especially for the clinical researcher) might ask how
genetic individuality will express itself under varying environmental conditions. The
majority of genetic studies have been concerned with estimates of how much a particular
trait is inherited rather than with questions of what is inherited. Little is known about
this, except that various polygenic systems seem to be involved, and it remains to be
discovered how the genes exert their effect.
9. Overview
The concept of neurosis incorporates a group of disorders which are in many respects
poorly defined and delineated. Nonetheless they are of great theoretical as well as clinical
significance. If one accepts the estimate that as many as 13% or 14% of the general
population may suffer from some form of neurotic disorder (Hagnell 1968; Shepherd
1973), it is clear that many millions of people are incapacitated to a considerable extent
by the neuroses. Cattell and Scheier (1961) make an even higher estimate - that
25%-57% of the normal population will at some time be affected by a neurotic condition.
The American Psychiatric Association, in its recent attempt to rationalise the idea of
neurosis, has abolished the term. There is no longer anyone in America who can be said
to suffer from neurosis. However, the disorders that were previously subsumed within the
general category of neurosis now re-appear in their separate and newly constituted roles
in such subcategories as Anxiety Disorders (including phobias and obsessions),
Somatoform Disorders (including conversion symptoms), Dissassociative Disorders
(including amnesia and depersonalisation) and Post-Traumatic Reactive Disorders.
When faced by the ambiguities and contradictions inherent in the current use of the term,
it is understandable that the APA should have been tempted to abandon the general
concept of neurosis. Concepts in science are not, however, replaced by executive decision
(even when they are obviously unsatisfactory), but because they have been superseded by
other more useful concepts.
It has been apparent for some time that the unitary concept of neurosis has quite
obvious weaknesses. It lumps together disorders which are quite different in subjective
experience, in their physiological concomitants and in their behavioural expression. One
of the least useful features of the traditional concept of neurosis was the way in which it
confused the descriptive and theoretical levels of discourse. Neurosis exists as a higher
order concept which covers a variety of separate disorders with diverse symptomatic
expressions, and it contains the implicit assumption that each of the disorders are connected in some way. This assumption operates in an insidious manner because it fails to
specify the nature of the connection between the disorders.
It is important not to confuse the descriptive stage of research, in which we attempt to
establish useful concepts, with the explanatory stage, in which we attempt to build
theories to explain the phenomena covered by these concepts and bring together apparently diverse phenomena. It is essential to systems of classification that they should group
phenomena according to their common features (Hempel 1961). But the fact that the
neuroses share certain symptomatic similarities does not mean that they are necessarily
the result of the same determinants. Equally, the symptomatic differences need not imply
that different determinants were involved. The symptomatic expression of particu-
Overview
133
lar disorders exists at a descriptive level. The supposed connections between those disorders which are the essence of the higher-order concept of neurosis exist at a theoretical
level. By making a more explicit separation of the neurotic disorders, DSM III may
facilitate the observation and description of the separate disorders. But this will only be
true if the classification system is treated as a descriptive system dealing with the symptomatic expression of the various neurotic conditions. This is unlikely to have been the
intention of the APA, since the medical tradition requires diagnoses to go beyond
description to make further statements about the causes of specific disorders as well as
about their prognosis (Warner 1952; Kendell 1975).
It is improbable that each specific neurotic disorder will be found to have its own
distinct determinants which set it apart from the other disorders. Most theories agree that
anxiety is the central facet of neurosis. Freud's view after 1936 was that all of the psychoneuroses represented an attempt to avoid the conscious experience of anxiety. Neurotic
anxiety was itself a response to some intrapsychic conflict: repression and the other
mechanisms of this sort were motivated by the attempt to escape the conscious experience of this painful affect.
Although Eysenck's trait theory differs in most respects from Freud's system, this also
regards the neuroses as sharing some common factor. Again this is closely linked to
anxiety, but in this case through the proposed higher-order factor of neuroticism. All
neurotics are said to score highly on this autonomic lability factor. Neuroticism is essentially a form of emotional sensitivity which makes the person unusally vulnerable to
stress. Cattell's trait theory takes a different position. Again, anxiety is seen as the central
feature of neurosis, but anxiety and neuroticism are clearly separated. Indeed, in Cattell's
system, a range of factors are proposed, each of which is necessary to the understanding
of neurosis. This multifactor theory stands in contrast to Eysenck's system, since neither
at the first-order nor at the second-order level do neurotic patients differ significantly
from normal subjects on only one central factor of neuroticism.
Anxiety, which is generally regarded as a central component of neurosis, may be
usefully divided into its physiological, behavioural and cognitive components. Until
recently, the emphasis upon behavioural and physiological features obscured the importance of the individual's cognitions. One of the most promising recent developments in
abnormal psychology has been the interest in cognitive processes. A common weakness
of several theories of neurosis has been their neglect or exclusion of the individual's
attitudes, beliefs and intentions. It is now becoming increasingly apparent that the way in
which the individual perceives himself, the objects and events in his environment and the
meaning that he attaches to them are extremely important to our understanding of
human behaviour. The classic series of experiments by Schachter and Singer (1962)
showed how emotional states depend upon their meaning for the individual, and other
investigators have shown how attribution processes can affect the neuroses. Storms and
Nisbett (1970) used an attributional model to explain the results of their study on
insomnia, and Valins and Nisbett (1971) extended this analysis to such other neurotic
disorders as the phobias, depression and homosexual panic.
The challenge facing any complete account of neurosis, however, is that of the neurotic
paradox. The neurotic seems to behave in a manner which leads to predominantly
unfavourable consequences, and which even the neurotic individual himself may see as
irrational and undesirable. Yet he seems unable to maintain any change from his neurotic
patterns of behaviour. A cognitive account of neurosis faces special difficulties in
134
Overview
attempting to explain this paradox in cognitive tenns. Learning theorists such as Mowrer
(1950) have explained it by reference to the way in which the neurotic avoids exposure to
the fear-producing situation. This protects the conditioned fear response from extinction.
To this account, Eysenck (1976) added the idea that anxiety has drive properties and
therefore leads to an incubation effect in which the unreinforced conditioned response
retains its strength, or even becomes more powerful.
Freud was also well aware of this paradox. From the beginning of his psychological
studies he had been fascinated by the behaviour of the neurotic individual. This seemed
to be inexplicable in terms of common sense, in terms of the patient's conscious wishes
and in tenns of the current state of medical and scientific knowledge. The psycho-analytic
solution is unique in its proposal of three major provinces of the mind in which various
dynamic forces operate. The presence of intrapsychic conflict, which is itself detennined
in a highly complex way by the structure of personality, establishes neurotic anxiety, and
in the conflict of drives, the neurotic symptoms appear as a compromise solution. All
neurotic reactions function primarly as a defence against the threat posed by this irrational anxiety. Freud's attempt to explain the neurotic paradox consisted of his attempt to
show that the psycho-neuroses operated according to the laws of unconscious mental
forces.
The neurotic paradox has been most thoroughly investigated with respect to the anxiety disorders, but depression is also surrounded by paradox. There is usually a large
discrepancy between the individual's perceptions of himself and the objective facts, and,
as with the anxiety disorders, depressive behaviour also seems to contradict the principle
that the person might be expected to act in such a way as to increase his sources of
satisfaction and to minimise his pain.
The complex mixture of symptoms found among neurotic depressives also argues
against an atomistic approach which too readily rejects the higher-order concept of
neurosis in favour of discrete categories. It is quite clear, for instance, that neurotic
depression regularly occurs together with anxiety and with other neurotic traits and
symptoms. Grinker et al. (1961) found that free-floating anxiety was an important part of
the clinical concept of depression, and the diagnosis of neurotic depression may be
dependent upon the patient's also showing signs of anxiety. In those cases in which the
patient is only slightly depressed, the presence of anxiety greatly increases the probability
that they will receive a diagnosis of depression.
Several authors have related the neuroses to the individual's social circumstances,
including his socio-economic status. Brown and Harris (1978), for instance, found that
depression was more common among working-class women and offered an explanation
in terms of how certain social factors act as provoking agents in depression. Other factors
have a protective function, and yet others make the person more vulnerable to particular
stressful events. There has also been considerable interest, particularly among sociologists, in the way in which neurosis may be seen as an ascribed social role. Despite its
weaknesses, this view has been a useful antidote to the intra-individual approaches of
most psychological accounts, and it has drawn attention to the way in which the social
meaning of neurosis plays a part in the determination of the neuroses. Parson's concept
of the sick-role is one of the best-known examples of this sort, and even such an individual-oriented system as psycho-analysis has acknowledged that the sick-role (though it
is not referred to in such terms) can playa part in maintaining neurotic behaviour. Freud
(1933, vol XXII, P 142) for instance has commented that 'The ego will seek to tum even
Overview
135
illness to its advantage'. In Freudian theory, however, this is clearly described as a

secondary gain which may add to the primary and unconscious gains provided by the
neurotic symptoms. In itself it is not thought to be sufficient to account for the development of the symptoms, and the relationship of the individual's self-perceptions to his
social environment are not dealt with at any length.
Of the neurotic disorders in which there is often a strong element of secondary gain,
and to which the sick-role may well be applicable, hysteria stands out. This is one of the
most curious of the neuroses, and it occupies a unique place in the history of psychological medicine. It was one of the first of the neurotic disorders to be described, and the
work of Briquet, Charcot, Janet and Freud in this area has undoubtedly had a profound
impact upon current thinking in abnormal psychology. Nonetheless, like the concept of
neurosis itself, hysteria remains a vague and indefinite sort of term.
Partly because of this, partly because of its pejorative implications, and partly because
the incidence of hysteria in the general population seems to have declined over the past
century, hysteria is not often made as a specific formal diagnosis in current psychiatric
practice. Slater (1965) has argued that the concept of hysteria is little more than an
historical curiosity which owes more to neurology than to psychiatry or psychology; it
does not refer to any specific syndrome, and its use generally reflects little more than a
conviction that the wide range of symptoms displayed by such patients have no physical
or medical foundation. Slater suggested that both on theoretical as well as on practical
grounds the term should be avoided, since it is little more than a disguise for ignorance.
Guze (1975), on the other hand, has argued strongly in favour of a system of classification based upon the medical model in which hysteria (Briquet's syndrome) is seen as a
valid clinical entity which follows a predictable course.
Descriptions of hysteria still occasionally appear in the literature, as in Moss and
McEvedy's (1966) paper on an epidemic of overbreathing which affected a group of 550
English schoolchildren. The authors relate this to personality factors, the affected children being described as high scorers on Eysenck's Neuroticism and Extraversion factors.
Among the other recent reports of hysteria is that of Claridge (1967), who pointed to the
higher incidence of hysterical symptoms, including even gross conversion reactions,
among military populations. Claridge's explanation is also heavily reliant upon a physiologically based theory of personality.
Another personality theory of hysteria, albeit of a very different type, is the Freudian
account. This has had a powerful influence upon psychiatry, as can be seen in the
nosological systems currently in use (see Chap. 2). At the same time, it is unlikely that
more than a minority of psychiatrists or psychologists would accept the details of Freudian theory whereby libidinal energy is converted into somatic disturbance. In the end,
even Freud accepted that his account was unsatisfactory. 'One really does not know much
that can be said about these symptoms ... what the course is of the particular obscurity
which surrounds symptom formation in conversion hysteria we are unable to guess'
(Freud 1936). Nonetheless, this Freudian explanation of hysteria, whatever its defects, is
more than Slater's diagnosis by exclusion.
One consequence of the Freudian view is that hysterical symptoms are frequently
interpreted as a symbolic form of communication. Szasz (1961) has suggested that the
hysterical patient is merely using the language of illness (albeit in an inappropriate way)
to communicate with potential helpers. This sort of view led to an explanation of the high
incidence of hysteria among military populations, which is entirely different from the
136
Overview
physiological and medical accounts. Rabkin (1964) suggested that the conversion symptoms shown by army recruits reflect a fundamental disagreement between the doctor and
patient about the nature of the patient's problems. When the civilian approaches his
physician, he may describe his problem either in medical terms or in terms of his problems of living. In the army, Rabkin argues that the nature of the doctor-patient relationship is fundamentally different, and that this changed relationship encourages the
development of hysterical disorders as a means of coping with the stresses of army life.
Unlike fullblown hysterical reactions (such as gross conversion symptoms), which seem
to be comparatively rare, excessive concern about one's health or bodily functions is a
rather more common problem. In general, it is also less likely to incapacitate the individual. However, in recent years there has been a reluctance to use the diagnosis of
hypochondriacal neurosis, even in more severe cases. Although this category is contained
in the European and American nosological systems ISD-8 and DSM II, such hypochondriacal concern is more often regarded as a manifestation of another more general
anxiety disorder or of some personality trait.
Depressive reactions are one of the most common problems in the general population.
Because of this, it is surprising that they have received so much less research attention
than the neuroses characterised by anxiety and avoidance behaviours. Depression
remains one of the more problematic disorders, and there is still considerable disagreement even over such basic issues as whether or not we should distinguish between
neurotic and psychotic forms of depression (though there is considerable evidence to
support such a distinction). There is further doubt about whether this distinction is
equivalent to that between reactive and endogenous depression, or whether the two
supposedly distinct sorts of disorder differ only in the severity of their symptoms (as the
classical psycho-analysts maintain). The issue is confused even more by other workers
who prefer to distinguish between primary and secondary depression or between agitated
and retarded depressive states. Apart from their theoretical significance, such issues
affect the ways in which the clinician approaches the problems being presented by the
patient, and frequently they have direct implications for treatment. Patients diagnosed as
neurotic or reactive depressives, for instance, do not usually respond well to ECf, and
the presence of such other neurotic features as hypochondriasis, emotional lability and
neurotic traits also seems to be associated with a poor response to 'medical' interventions
of this sort (Mendels 1965; Carney et al. 1965).
As a research topic, depression seems to have had a particular attraction for cognitive
theorists. Beck and Seligman each proposed cognitive accounts of depression which are
interesting in their own right and which have also been useful in helping to correct the
traditional psychiatric view of depression as a primary affective disorder. Beck (1967),
for instance, suggested that the disturbances of the depressive patient can be seen in
terms of three major cognitive patterns that lead the person to view himself, his world
and his future in an idiosyncratic manner. The individual who has incorporated patterns
of negative expectation is predisposed towards a depressive disorder, and when exposed
to certain sorts of stressful events and experiences is more likely to become clinically
depressed. As the primary triad of negative expectations become progressively more
dominant, such cognitions lead to the other phenomena that are associated with the
depressed state (depressed mood, suicidal wishes, behavioural inertia, etc.). The affective
responses are therefore determined by the ways in which the individual structures his
experiences.
Overview
137
This is broadly similar to Seligman's notion of learned helplessness as the crucial determinant of depression, though Seligman's account relies more heavily upon experimental
evidence and upon the formulations of learning theory. Seligman noted that it is an
essential feature of Pavlovian conditioning that the subject is helpless to control the
outcome of his own responses. Overmier and Seligman (1967), for instance, showed that
exposure to inescapable shock leads to escape/avoidance learning deficits. Seligman's
theory suggests that the subject acquires the expectation that their responses and the
outcome of their responses are independent. When the uncontrollable outcome is also
traumatic, this produces increased anxiety followed by depression.
There are a number of difficulties with this account, both conceptual and empirical, but
it remains a stimulating view of depression and has encouraged a considerable amount of
research into this disorder. One of the alternative accounts to the cognitive emphasis of
the learned helplessness model is that endogenous noradrenaline is differentially affected
under conditions of controllable and uncontrollable shock (Weiss et al. 1976). Physiological and psycho-physiological studies have also contributed to our understanding of the
mechanisms of the other neurotic disorders. Lader's (1967) results differentiated between different diagnostic groups of neurotic patients in terms of physiological measures.
Patients with specific phobias, for instance, were differentiated from patients with other
affective disorders, and it has been suggested both by Lader (1967) and by Kelly and
Walter (1968) that phobic patients bear a closer ressemblance to normal groups than to
other neurotic patients in terms of their physiological responses. Results of this sort also
argue against a too-ready acceptance of the reclassification of DSM III in which the
phobic disorders are defined as anxiety disorders. As a result they are included in the
same category as panic attacks which are characterised by intense and widespread autonomic arousal, and with more chronic, generalised anxiety disorders in which motor tension and autonomic hyperactivity are defining characteristics. In the study of Kelly.and
Walter (1968), on the other hand, patients with phobic disorders differed from chronically anxious patients on almost every physiological measure that was used in the investigation.
There is always a temptation at this point to reach some conclusion to the effect that
the best of each of the theories might be combined to produce a single comprehensive
theory which would be more powerful than any of its components. There are, unfortunately, several objections which prevent the realisation of this pious hope. In the first
place, the theories are often stated in such imprecise terms that it is difficult to see at
which points they are in agreement and at which other points they make different predictions. For this reason alone it would be difficult to attempt any true synthesis. A further
difficulty arises from the way in which the different accounts provide a general orientation or perspective upon the phenomena in which we are interested. Each theory operates as a selective filter upon our approach to the problem. An unfortunate complication
of this has been the tendency for certain psychologists to display a sort of theoretical
imperialism. Psychologists who work in medical settings will probably recognise this in
the over-enthusiasm which some physicians display in their attempt to define social and
psychological problems in medical terms. This same tendency is seen in those psychologists who are committed so completely to a particular theoretical perspective that they
not only exclude alternative explanations of the neuroses, but that they also exclude
alternative formulations of the problem. This results in disagreements of the most elusive
kind - in which the proponents of different theories are unable to agree even upon such
138
Overview
fundamental issues as what is to be explained. What a psycho-analyst understands by a

neurosis may have little in common with what a conditioning theorist understands by the
same term.
At present, no single theory has achieved general acceptance, and none can legitimately claim to offer a satisfactory explanation of each of the neurotic disorders (as
contained in the psychiatric nosologies). There is a tendency for psychologists to be
defensive about this lack of any agreed theory of neurosis. Such theoretical diversity need
not, however, be regarded as a failure. Controversy may be the result of a lack of
empirical information about a particular topic, and it may therefore have the useful
function of stimulating research. It may also lead to a clarification of issues through the
way in which the criticisms of opposing theories expose the assumptions underlying a
particular theoretical approach. On the other hand, controversy can lead to a hardening
of positions through the emotional heat generated in the course of debate: theoretical
disputes can often produce a polarisation of views in which the various protagonists are
forced into extreme positions that they do not really hold. A further danger is that
definitions of a problem can be restricted in such a way as to fit the theory more exactly.
But whatever the disadvantages of controversy, it is unlikely that any theoretical
resolution will take place in the near future. In an area of this sort where our present level
of understanding is still rudimentary, provided that each of the theories pays attention to
its own weaknesses as well as to the strengths of its rivals, it may ultimately be more
productive to tolerate or even to encourage controversy than to enforce some premature
synthesis.
10. References
Abramson LY, Seligman MEP, Teasdale JD (1978) Learned helplessness in humans: critique and
reformulation. J Abnorm Psychol 87: 49-74
Adcock CJ (1965) A comparison of the concepts of Cattell and Eysenck. Br J Educ Psychol 35:
90-97
Adelman HM, Maatsch JL (1956) Learning and extinction based upon frustration, food reward, and
exploratory tendency. J Exp Psychol 52: 311-315
Adelman MR (1977) A comparison of professionally employed lesbians and heterosexual women
on the MMPI. Abnorm Sex Behav 6: 193-201
Adler A (1918) The neurotic constitution. Kegan Paul & Trench Trubner, New York
Adler A (1946) The practice and theory of individual psychology. Kegan Paul & Trench Trubner,
London
Ajzen I, Fishbein M (1977) Attitude - behaviour relations: a theoretical analysis and review of
empirical research. Psychol Bull 84: 888-928
Albee GW (1969) Emerging concepts of mental illness and models of treatment: the psychological
point of view. Am J Psychiatry 125: 870-76
Amsel A (1962) Frustrative nonreward in partial reinforcement and discrimination learning: some
recent history and a theoretical extension. Psychol Rev 69: 306-328
Argyle M, Trower P, Bryant B (1974) Explorations in the treatment of personality disorders and
neuroses by social skills training. Br J Med Psychol 47: 63-72
Argyle M (1969) Social interaction. Methuen, London
Ash P (1949) The reliability of psychiatric diagnoses. J Abnorm Soc Psychol 44: 272-276
Azrin NH, Holz WC (1966) Punishment. In: Konig WK (ed) Operant behaviour: areas of research
and application. Appleton-Century-Crofts, New York
Bandura A (1965) Influence of models reinforcement contingencies on the acquisition of imitative
responses. J Pers Soc Psychol 1: 589-595
Bandura A (1969) Principles of behaviour modification. Holt, Rinehart & Winston, New York
Bandura A (1971) Psychological modelling: conflicting theories. A1dine & Atherton, Chicago
Bandura A (1974) Behaviour theory and the models of man. Am Psychol29: 859-869
Bandura A (1977) Social learning theory. Prentice-Hall, Englewood Cliffs, NJ
Bandura A, Rosenthal TL (1966) Vicarious classical conditioning as a function of arousal level. J
Pers Soc Psychol 3: 54-62
Bandura A, Grusec JE, Menlove FL (1967) Vicarious extinction of avoidance behaviour. J Pers Soc
Psychol 5: 16-23
Barton WE (1959) Viewpoint of a clinician. In: J ahoda M (ed) Current concepts of positive mental
health. Basic Books, New York
Beard GM (1869) Neurasthenia or nervous exhaustion. Boston Med Surg J 3: 159-271
Beck AT (1962) Reliability of psychiatric diagnoses: 1. a critique of systematic studies. Am J
Psychiatry 119: 210-216
Beck AT (1967) Depression: clinical, experimental and theoretical aspects. Harper & Row, New
York
Beck AT (1971) Cognition, affect and psychopathology. Arch Gen Psychiatry 24: 495-500
Beech HR, Watts F, Poole AD (1971) Classical conditioning of a sexual deviation: a preliminary
note. Behav Ther 2: 400-402
Beech HR, Perigault J (1974) Toward a theory of obsessional disorder. In: Beech HR (ed) Obsessional states. Methuen, London
140
References
Bellak L, Hurvich M, Gediman HK (1973) Ego functions in schizophrenics, neurotics, and normals.
Wiley, New York
Bern DJ, Allen A (1974) On predicting some of the people some of the time: the search for crosssituational consistences in behaviour. Psychol Rev 81: 506-520
Benjamin S, Marks 1M, Huson J (1972) Active muscular relaxation in desensitisation of phobic
patients. Psychol Med 2: 381-390
Berger SM (1962) Conditioning through vicarious instigation. Psychol Rev 69: 450-466
Berkley HJ (1901) A treatise on mental diseases. Kimpton, London
Berlyne DE (1954) Knowledge and stimulus-response psychology. Psychol Rev 61: 245
Blanchard EB, Hersen M (1976) Behavioural treatment of hysterical neurosis: symptom substitution and symptom return reconsidered. Psychiatry 39: 118-129
Block JD (1967) Monozygotic twin similarity' in multiple psychophysiologic parameters and
measures. In: Wortis J (ed) Recent advances in biological psychiatry, vol 9. Plenum Press, New
York
Blum GS (1966) Psychodynamics: the science of unconscious mental forces. Wadsworth, Belmont,
Ca.
Blumberg HH, Cohen SD, Dronfield BE, Mordecai EA, Roberts JC, Hawks D (1974) British
opiate users I. People approaching London drug treatment centres. Int J Addict 9: 1-23
Bockoven JS (1956) Moral treatment in American psychiatry. J Nerv Ment Dis 3: 124,167-194,
292-321
Bockoven JS (1963) Moral treatment in American psychiatry. Springer, Vienna, New York
Bolles RC, Grossen NE (1969) Effects of an informational stimulus on the acquisition of avoidance
behaviour in rats. J comp physiol psychol 68: 90-99
Bolles RC (1970) Species-specific defence reactions and avoidance learning. Psychol Rev 77: 32-48
Borkovec TD (1976) Physiological and cognitive proceS!ieS in the regulation of anxiety. In: Schwartz
GE, Shapiro D (eds) Consciousness and self-regulation, vol L Plenum, New York
Boulougouris JC, Marks 1M, Marset P (1971) Superiority of flooding (implosion) of desensitisation
for reducing pathological fear. Behav Res Ther 9: 7-16
Brady JV, Porter RW, Conrad DG, Mason JW (1958) Avoidance behaviour and the development
of gastroduodenal ulcers. J Exp Anal Behav 1: 69-72
Braginsky DD, Braginsky BM (1974) Mainstream psychology. A critique. Holt, Rinehart & Winston, New York
Bregman E (1934) An attempt to rectify the emotional attitudes of infants by the conditioned
response technique. J Genet Psychol45: 169-198
Breuer J, Freud S (1895) Studies on hysteria. Standard edition of the complete works of Sigmund
Freud, vol II. Hogarth, London
Brewer WF (1974) There is no convincing evidence for operant or classical conditioning in adult
humans. In: Weimer WB, Palermo DS (eds) Cognition and the symbolic processes. Wiley, New
York
Brill AA (1936) Translator's introduction to C. G. Jung. The psychology of dementia praecox. Nerv
Ment Dis Publishing Co., New York
Brill H (1967) Oassification in psychiatry. In: Freedman AM, Kaplan HI (eds) Comprehensive
textbook of psychiatry. Williams & Wilkins, Baltimore
Brill H (1974) Oassification and nomenclature of psychiatric conditions. In: Arieti S (ed) American
handbook of psychiatry, vol I. Basic Books, New York
Britton RS (1969) Psychiatric disorders in the mothers of disturbed children. J Child Psychol
Psychiatry 10: 245
Broadhurst PL (1978) Drugs and the inheritance of behaviour. Plenum, London
Brown FW (1942) Heredity in the psychoneuroses. Proc R Soc Med 35: 785-790
Brown GW, Harris T (1978) Social origins of depression. Tavistock, London
Brown RT, Wagner AR (1964) Resistance to punishment and extinction following training with
shock or nonreinforcement. J Exp Psychol 68: 503-507
Buchwald AM, Coyne JC, Cole CSA (1978) A critical evaluation of the learned helplessness model
of depression. J Abnorm Psychol 87: 180-193
Burt C (1954) The assessment of personality. J Ment Sci toO, 1-28
Carney MWP, Roth M, Garside RF (1965) The diagnosis of depressive syndromes and the prediction of ECT response. Br J Psychiatry 111: 659-674
References
141
Caspari EW (1967) Behavioural consequences of genetic polymorphism in man: a summary. In:
Genetic diversity and human behaviour. Spuhler IN (ed) Wenner-Gren Foundation Inc
Cattell RB (1946) Description and measurement of personality. World Books, New York
Cattell RB (1950) Personality: a systematic theoretical and factual study. McGraw-Hili, New York
Cattell RB (1952) Factor analysis. Greenwood, Westport
Cattell RB (1957a) Personality and motivation structure and measurement. World Books, New
York
Cattell RB (1957b) The conceptual and test distinction of neuroticism and anxiety. Jain Psychol
13: 221-233
Cattell RB, Kline P (1977) The scientific analysis of personality and motivation. Academic Press,
New York
Cattell RB, Scheier IH (1961) Neuroticism and anxiety. Ronald, New York
Caveny EL, Wittson CL, Hunt WA, Herrman RS (1955) Psychiatric diagnosis, its nature and
function. J Nerv Ment Dis 121: 367-373
Chesser ES (1976) Behaviour therapy: recent trends and current practice. Br J Psychiatry 129:
289-307
Church RM (1959) Emotional reactions of rats to the pain of others. J Comp Physiol Psychol52:
132-134
Qare A (1976) Psychiatry in dissent. Tavistock, London
Qaridge GS (1960) The excitation-inhibition balance in neurotics. In: Eysenck HJ (ed) Experiments in personality. Routledge & Kegan Paul, London
aaridge GS (1967) Personality and arousal. Pergamon, Oxford
Coates DB, Moyer S, Wellman B (1969) Yorklea study: symptoms, problems and life events. Can J
Public Health 60: 471-481
Coates D, Moyer S, Wellman B (1969) The Yorklea study of urban mental health: symptoms,
problems and life events. In: Boydell C, Grindstaff C, Whitehead P (eds) Deviant behaviour and
societal reaction. Holt, Rinehart & Winston, Toronto
Coates DB, Moyer S, Kendall L, Howat MG (1976) Life-event changes and mental health. In:
Sarason IG, Spielberger CD (eds) Stress and anxiety, vol 3. Wiley, New York
Cole S, Lejeune R (1972) Illness and the legitimation of failure. Am Sociol Rev 37: 347-356
Cook SW, Harris RE (1937) Verbal conditioning of the galvanic skin response. J Exp Psychol21:
202-210
Costello CG (1970) aassification and psychopathology. In: Costello CG (ed) Symptoms of
psychopathology. Wiley, New York
Cottrell NB, Wack DL, Sekerak GJ, Rittle RH (1968) Social facilitation of dominant responses
by the presence of an audience and the mere presence of others. J Pers Soc Psychol 9:
245-250
Craighead WE, Kazdin AE, Mahoney M (1976) Behaviour modification, Houghton Mifflin, Boston, Ma.
Cullen W (1781) First lines on the practice of physic, vol 2. Creech, Edinburgh
Culpin M (1962) The conception of nervous disorder. Br J Med Psychol 35: 73-80
Curran D, Partridge M, Storey P (1972) Psychological medicine: an introduction to psychiatry.
Churchill Livingstone, Edinburgh London
D'Amato MR, Schiff J (1964) Long-term discriminated avoidance performance in the rat. J Comp
Physiol Psychol 57: 123-126
Davis JF, Malmo RB, Shagass C (1954) Electromyographic reaction to strong auditory stimulation
in psychiatric patients. Can J Psychol 8: 177-186
Denker R (1946) Results of the treatment of psychoneuroses by the general practitioner. NJ State J
Med 46: 2164-2166
Derogatis LR, Lipman RS, Covi L, Rickels K, Uhlenhuth EH (1970) Dimensions of outpatient
neurotic pathology: a comparison of clinical versus an empirical assessment. J Consult Clin
Psychol 34: 164-171
Derogatis LR, Covi L, Lipman RS, Davis DM, Rickels K (1971 a) Social class and race as mediator
variables in neurotic symptomatology. Arch Gen Psychiatry 25: 31-40
Derogatis LR, Lipman RS, Covi L, Rickels K (1971 b) Neurotic symptom dimensions. Arch Gen
Deutsch H (1944) Psychology of women. Grune & Stratton, New York, London
142
References
Dietz PE (1977) Social discrediting of psychiatry: the protasis of legal disenfranchisement. Am J

Dollard J, Miller NE (1950) Personality and psychotherapy. McGraw-Hili, New York
DSM II (1968) Diagnostic and statistical manual of mental disorders, 2nd edn. Am Psychiatr Assoc,
Washington, D.C.
DSM III (1978) Diagnostic and statistical manual of mental disorders, 3rd edn. Am Psychiatr Assoc,
Washington, D.C.
DSM III (1978) Draft of the diagnostic and statistical manual of mental disorders. Third edition.
Am Psychiatr Assoc, Washington
Ducey C, Simon B (1975) Ancient Greece and Rome. In: Howells JG (ed) World history of
psychiatry. Bailliere Tindall, London
Eiser JR (1979) Addiction as attribution. Paper presented to European Association of Experimental Social Psychology, Cumbria, England
Eiser JR (1978) Interpersonal attribution. In: Tajfel H, Fraser C (eds) Introducing social psychology. Penguin Books, New York
Eiser JR (1979) Addiction as attribution. In: Eiser JR (ed) Social psychology and behavioural
medicine. Wiley, London
Ellenberger HF (1970) The discovery of the unconscious. Allen Lane, London
Ellenberger HF (1974) Psychiatry from ancient to modern times. In: Arieti S (ed) American
handbook of psychiatry. Basic Books, New York, vol I.
Ellis A (1962) Reason and emotion in psychotherapy. Lyle Stuart, New York
Endler NS, Hunt J McV (1966) Sources of behavioural variance as measured by the S-R inventory
of anxiousness. Psychol Bull 65: 336-346
Endler NS, Hunt J McV (1969) Generalisability of contributions from sources of variance in the S-R
inventories of anxiousness. J Pers 37: 1-20
Engel GL (1972) Is psychiatry failing in its responsibility to medicine? Am J Psychiatry 128:
1561-1563
Erikson K (1957) Patient role and social uncertainty: a dilemma of the mentally ill. Psychiatry 20:
263-274
Essen-Moller E (1956) Individual traits and morbidity in a Swedish rural population. Acta Psychiatr
Scand [Suppl] 100
Essen-Moller E (1967) On classification of mental disorders. Acta Psychiatr Scand 37: 119-126
Everitt BS, Gourlay AJ, Kendall RE (1971) An attempt at validation of traditional psychiatric
syndromes by cluster analysis. Br J Psychiatry 119: 399-412
Eysenck HJ (1947a) Dimensions of personality. Routledge & Kegan Paul, London
Eysenck HJ (1947b) Screening out the neurotic. Lancet 1: 530
Eysenck HJ (1955) Psychiatric diagnosis as a psychological and statistical problem. Psychol Rep 1:
3-17
Eysenck HJ (1957) The dynamics of anxiety and hysteria. Routledge & Kegan Paul, London
Eysenck HJ (1959) Learning theory and behaviour therapy. J Ment Sci 105: 61-75
Eysenck HJ (1960a) Modern learning theory. In: Eysenck HJ (ed) Behaviour therapy and the
neuroses. Pergamon, London
Eysenck HJ (1960b) Symposium: the development of moral values in children. Br J Educ Psychol
30: 11-21
Eysenck HJ (1960c) A rational system of diagnosis and therapy in mental illness. In: Progress in
clinical psychology, volIV. Grune & Stratton, New York, London
Eysenck HJ (1961) Classification and the problem of diagnosis. In: Eysenck HJ (ed) Handbook of
abnormal psychology. Basic Books, New York
Eysenck HJ, Qaridge G (1962) The position of hysterics and dysthymics in a two-dimensional
framework of personality description. J abnorm soc psychol 69: 46-55
Eysenck HJ (1965) Extraversion and the acquisition of eyeblink and GSR conditioned responses.
Psychol Bull 63: 258-270
Eysenck HJ (1967) The biological basis of personality. Thomas, Springfield, III.
Eysenck HJ (1968) A theory of the incubation of anxiety/fear responses. Behav Res Ther 6:
309-321
Eysenck HJ (1970a) The classification of depressive illness. Br J Psychiatry 117: 241-250
Eysenck HJ (1970b) The structure of human personality. Methuen, London
References
143
Eysenck HJ (1972a) Primaries or second-order factors: a critical consideration of Cattell's 16PF

battery. Br J Soc Clin Psychol 11: 265-269
Eysenck HJ (1972b) Traits. In: Eysenck HJ, Meili WAR (eds) Encyclopedia of psychology, vol 3.
Search, London
Eysenck HJ (1975 a) The future of psychiatry. Methuen, London
Eysenck HJ (1975b) Anxiety and the natural history of neurosis. In: Spielberger C, Sarason I (eds)
Stress and anxiety, voll. Hemisphere, New York
Eysenck HJ (1976) The learning theory model of neurosis - a new approach. Behav Res Ther 14:
251-267
Eysenck HJ (1977) Crime and personality. Routledge & Kegan Paul, London
Eysenck HJ (1979) The conditioning model of neurosis. In: Behavioural and brain science, 2:
155-199
Eysenck HJ, Eysenck SBG (1969) Personality structure and measurement. Routledge & Kegan
Paul, London
Eysenck HJ, Eysenck SBG (1975) Manual of the Eysenck personality questionnaire. Hodder &
Stoughton, London
Eysenck HJ, Rachman S (1965) The causes and cures of neurosis. Routledge & Kegan Paul,
London
Eysenck SBG (1956) Neurosis and psychosis: an experimental analysis. J Ment Sci 102: 517529
Fahrenberg J (1977) Physiological concepts in personality research. In: Cattell RB, Dreger RM
(eds) Handbook of modem personality theory. Wiley, New York
Farrell BA (1951) The scientific testing of psychoanalytic findings and theory. Br J Med Psychol 24:
35-51
Farrell BA (1961) On the character of psychoanalytic discourse. Br J Med Psychol 34: 7-21
Farrell BA (1967) The criteria for a psychoanalytic interpretation. In: Gustafson DF (ed) Essays in
philosophical psychology. Macmillan, London
Feldman MP (1966) Aversion therapy for sexual deviations: a critical review. Psychol Bull 65:
65-69
Feldman MP, MacCulloch MJ (1971) Homosexual behaviour: therapy and assessment. Pergamon,
Oxford
FenischelO (1945) The psychoanalytic theory of neurosis. Norton, New York
Fenton N (1926) Shell shock and its aftermath. Mosby, St. Louis
Flew A (1978) Mental disease. J Med Ethics 4: 89-90
Foulds GA (1955) The reliability of psychiatric, and the validity of psychological diagnoses. J Ment
Sci 101: 851-862
Foulds GA, Bedford A (1975) Hierarchy of classes of personal illness. Psychol Med 5: 181-192
Foulds GA, Bedford A (1976) The relationship between anxiety-depression and the neuroses. Br J
Frank G (1975) Psychiatric diagnosis: a review of research. Pergamon, New York
Freud S (1893-1895) Studies on hysteria. Standard edition of the complete works of Sigmund
Freud, vol II. Hogarth, London
Freud S (1893-1899) Early psychoanalytic publications. Standard edition of the complete works of
Sigmund Freud, vol III. Hogarth, London
Freud S (1900) The interpretation of dreams. Allen & Unwin, London
901-1905) A case of hysteria, three essays on sexuality and other works. Standard edition of the
complete works of Sigmund Freud, vol VII. Hogarth, London
Freud S (1905) Three essays on the theory of sexuality. Basic Books, New York
Freud S (1906-1908) Jensen's 'Gravida', Standard edition of the complete works of Sigmund
Freud, vol IX. Hogarth, London
Freud S (1909) The cases of 'Little Hans' and the 'Rat Man'. Standard edition of the complete
works of Sigmund Freud, vol X. Hogarth, London
Freud S (1911-1913) Case history of Schreber, papers on technique. Standard edition of the
complete works of Sigmund Freud, vol XII. Hogarth, London
Freud S (1914-1916) A history of the psychoanalytic movement, papers on metapsychology and
other works. Standard edition of the complete works of Sigmund Freud, vol XIV. Hogarth,
London
144
References
Freud S (IFreud S (1917) Introductory lectures on psycho-analysis. Standard edition of the complete works of Sigmund Freud, vol XVI. Hogarth, London
Freud S (1923-1925) The ego and the id and other works. Standard edition of the complete works
of Sigmund Freud, vol XIX. Hogarth, London
Freud S (1924) Collected papers, vol II. Hogarth, London
Freud S (1925-1926) An autobiographical study, inhibitions symptoms and anxiety, lay analysis and
other works. Standard edition of the complete works of Sigmund Freud, vol XX. Hogarth,
London
Freud S (1932-1936) New introductory lectures. Standard edition of the complete works of Sigmund Freud, vol XXII. Hogarth, London
Freud S (1936) The problem of anxiety. Norton, New York
Freud S (1937-1939) Moses and monotheism, an outline of psychoanalysis and other works. Standard edition of the complete works of Sigmund Freud, vol XXIII. Hogarth, London
Frith CD, Johnstone EC, Owens D, McPherson K (1978) The phenomenological, pharmacological
and psychophysiological characteristics of neurotic outpatients to whom no diagnosis has been
applied. Proc 11th C.I.N.P. Conference, Vienna
Frolov yP (1938) Pavlov and his school. Kegan Paul, London
Fromm E (1960) The fear of freedom. Routledge & Kegan Paul, London
Garcia J, Koelling R (1966) Relation of cue to consequence in avoidance learning. Psychosom Sci 4:
123-124
Garmany G (1958) Depressive states: their aetiology and treatment. Br Med J 2: 341-344
Giorgi A (1970) Psychology as a human science. Harper & Row, New York
Glass, DC, Singer JE (1972) Urban stress: experiments on noise and urban stressors. Academic
Press, New York
Gleitman H, Nachmias J, Neisser U (1954) The S-R reinforcement theory of extinction. Psychol
Rev 61: 23-33
Goffman E (1961) Asylums: essays on the social situation of mental patients and other inmates.
Anchor, New York
Goldberg DP (1972) The detection of psychiatric illness by questionnaire. Maudsley Monographs.
Oxford University Press, London
Goldstein m (1964) Physiological responses in anxious women patients. A study of autonomic
activity and muscle tension. Arch Gen Psychiatry 10: 382-388
Gormezano I, Moore JW (1969) Qassical conditioning. In: Marx MM (ed) Learning processes.
Macmillan, London
Gossop MR (1974) Movement variables and the subsequent following response of the domestic
chick. Anim Behav 22: 982-986
Gossop MR (1978 a) A comparative study of oral and intravenous drug dependent patients on three
dimensions of personality. Int J Addict 13: 135-142
Gossop MR (1978 b) Drug dependence, crime and personality among female addicts. Drug Alcohol
Depend 3: 359-364
Gossop MR (1978c) Drug dependence: a study of the relationship between motivational, cognitive,
social and historical factors, and treatment variables. J Nerv Ment Dis 166: 44-50
Gossop MR (1979) Drug dependence: a reappraisal. In: Oborne DJ, Gruneberg MM, Eiser JR
(eds) Research in psychology and medicine, vol 2. Academic Press, London
Gossop MR, Connell PH (1975) Attitudes of oral and intravenous multiple drug users toward drugs
of abuse. Int J Addict 10: 459-472
Gossop MR, Kristjansson I (1977) Crime and personality: a comparison of convicted and nonconvicted drug dependent males. Br J Crimino117: 264-273
Gottesman II (1963) Heritability of personality: a demonstration. Psychol Monogr 77/9
Gough HG (1946) Diagnostic patterns on the MMPI. J Clin Psychol2: 23-37
Gove WR (1970) Societal reaction as an explanation of mental illness: an evaluation. Am Sociol
Rev 35: 873-884
Gove WR (1975) Labelling and mental illness: a critique. In: Gove WR (ed) The labelling of
deviance. Wiley, London
Gove WR, Fain T (1973) The stigma of mental hospitalisation. Arch Gen Psychiatry 28: 494-500
Gove WR, Howell P (1974) Individual resources and mental hospitalisation: a comparison and
evaluation of the societal reaction and psychiatric perspectives. Am Sociol Rev 39: 86-100
References
145
Grant DA (1964) Classical and operant conditioning. In: Melton AW (ed) Categories of human
learning. Academic Press, New York
Gray I (1970a) Sodium amobarbital, the hippocampal theta rhythm and the partial reinforcement
extinction effect. Psychol Rev 77: 465-480
Gray I (1970b) The psychophysiological basis of introversion-extraversion. Behav Res Ther 8:
249-266
Gray I (1971) The psychology of fear and stress. World University Library, London
Gray IA (1972) The psychophysiological basis of introversion-extraversion: a modification of
Eysenck's theory. In: Biological bases of individual behaviour. Nebylitsyn VD & Gray IA (eds)
Academic Press, New York
Gray I (1973) Causal theories of personality and how to test them. In: Royce IR (ed) Multivariate
analysis and psychological theory. Academic Press, London
Gray IA (1975) Elements of a two-process theory of learning. Academic Press, London
Gray IA (1976) The neuropsychology of anxiety. In: Sarason IG, Spielberger CD (eds) Stress and
anxiety, vol 3. Wiley, New York
Greenson RR (1967) The technique and practice of psychoanalysis. International University Press,
New York
Griesinger W (1876) Die Pathologie und Therapie der psychischen Krankheiten. Wreden, Brunswik
Grinker RR, Miller I, Sabshin M, Nunn R, Nunally IC (1961) The phenomena of depressions.
Harper & Row, New York
Guilford IP, Guilford RB (1934) An analysis of the factors in a typical test of introversion-extroversion. I Abnorm Soc Psychol 28: 377-399
Guilford IP, Guilford RB (1936) Personality factors S, E, and M, and their measurement. I Psychol
2: 109-127
Guilford IP, Zimmerman WS (1956) Fourteen dimensions of temperament. Psychol Monogr 70110
Guthrie ER (1952) The psychology of learning. Harper, New York
Guze SB (1975) The validity and significance of the clinical diagnosis of hysteria (Briquet's syndrome). Am I Psychiatry 132: 138-141
Hagnell 0 (1959) Neuroses and other nervous disturbances in a population living in a rural
area of Southern Sweden, investigated in 1947 and 1957. Acta Psychiatr Scand [Suppl] 136:
214-219
Hagnell 0 (1968) A Swedish epidemiological study: the Lundby project. Soc Psychiatry 3: 75-77
Hallam RS (1974) Extinction of ruminations: a case study. Behav Ther 5: 565-568
Hallam RS, Rachman S (1972) Some effects of aversion therapy on patients with sexual disorders.
Behav Res Ther 10: 171-180
Hallam RS, Rachman S, Falkowski W (1972) Subjective, attitudinal and physiological effects of
electrical aversion therapy. Behav Res Ther 10: 1-13
Hanusa BH, Schulz R (1977) Attributional mediators of learned helplessness. I Pers Soc Psychol
35: 602-611
Hartmann H (1960) Toward a concept of mental health. Br I Med Psychol 33: 243-248
Hartshorne H, May MA (1928) Studies in the nature of character, vol 1. Studies in deceit. Macmillan, New York
Hartshorne H, May MA, Shuttleworth FK (1929) Studies in the nature of character, vol 3. Studies
in the organisation of character. Macmillan, New York
Harzem P, Miles TR (1978) Conceptual issues in operant psychology. Wiley, London
Hearst E (1969) Aversive conditioning and external stimulus control. In: Campbell BA, Church
RM (eds) Punishment and aversive behaviour. Appleton-Century-Crofts, New York
Hebb DO (1947) Spontaneous neurosis in chimpanzees: theoretical relations with clinical and
experimental phenomena. Psychosom Med 9: 3-16
Heilbrunn (1963) Results with psychoanalytic therapy. Am I Psychother 17: 427-435
Helmholtz H von (1866) Handbuch der physiologischen Optik. Hamburg, Leipzig
Hempel CG (1961) Introduction to the problems of taxonomy. In: Zubin I (ed) Field studies in the
mental disorders. Grune & Stratton, New York, London
Henderson D, Batchelor IRC (1962) Textbook of psychiatry. Oxford University Press, London
Hendrick I (1958) Facts and theories of psychoanalysis. Dell, New York
Hersen M, Eisler RM (1976) Social skills training. In: Craighead WE, Kazdin AB, Mahoney MJ
(eds) Behaviour modification. Houghton Mifflin, Boston
146
References
Hine FR, Williams RB (1975) Dimensional diagnosis and the medical student's grasp of psychiatry.
Arch Gen Psychiatry 32: 525-528
Hodgson R, Rachman S, Marks I (1972) The treatment of chronic obsessive-compulsive neurosis:
follow-up and further findings. Behav Res Ther 10: 181-189
Hoghughi MS, Forrest AR (1970) Eysenck's theory of criminality. Br J Criminol 10: 240-254
Hollingshead AB, Redlich FC (1958) Social class and mental illness: a community study. Wiley,
New York
Holz WC, Azrin NH (1961) Discriminative properties of punishment. J Exp Anal Behav 4:
225-232
Homey K (1936) Culture and neurosis. Am Sociol Rev 1: 221-230
Homey K (1937) The neurotic personality of our time. Kegan Paul & Trench Trubner, London
Homey K (1946) Our inner conflicts. Kegan Paul & Trench Trubner, London
Howe ES (1958) GSR conditioning in anxiety states, normals and chronic functional schizophrenic
subjects. J Abnorm Soc Psychol 56: 183-189
Howells JG (1970) Nosology of psychiatry. Society of Clinical Psychiatrists, Ipswich
Hull CL (1932) The goal gradient hypothesis and maze learning. Psychol Rev 39: 25-43
Hull CL (19)7) Mind, mechanism, and adaptive behaviour. Psychol Rev 44: 1-32
Hull CL (1943) Principles of behaviour: an introduction to behaviour theory. Appleton-CenturyCrofts, New York
Hull CL (1952) A behaviour system. Yale University Press, New l'laven
Humphreys LG (1939) The effect of random alternation of reinforcement on the acquisition and
extinction of conditioned eyelid reactions. J Exp Psychol 25: 141-158
Hunt HC (1932) A retired habituation. A history of The Retreat. York, Lewis, London
Hunt WA, Wittson CL, Hunt CB (1953) A theoretical and practical analysis of the diagnostic
process. In: Hoch PH, Zubin J (eds) Current problems in psychiatric diagnosis. Grune & Stratton,
New York, London
Hunter R (1973) Psychiatry and neurology: psychosyndrome or brain disease. Proc R Soc Med 66:
359-364
Hunter R, Macalpine I (1970) Three hundred years of psychiatry. 1535-1860. Oxford University
Press, London
Ingham JG (1966) Changes in MPI scores in neurotic patients: a three year follow-up. Br J
ICD-8 (1967) Manual of the international statistical classification of diseases, injuries, and causes of
death. WHO, Geneva
Jacobsen CF, Wolfe JB, Jackson TA (1935) An experimental analysis of the frontal association
areas in primates. J Nerv Ment Dis 82: 1-14
Jahoda M (1958) Current concepts of positive mental health. Basic Books, New York
Janet P (1891) Etude sur un cas d'aboulie et d'idees fixes. Rev Philos 31: 258-287, 382-407
Janet P (1895) J. M. Charcot, son oeuvre psychologique. Rev Philos 39: 569-604
Janet P (1909) Les nevroses. Flammarion, Paris
Jaspers K (1972) General psychopathology, 4th edn. Manchester University Press, Manchester
Jellinek EM (1939) Some principles of psychiatric classification. Psychiatry 2: 161-165
Jones MC (1924) A laboratory study of fear. Pedagog Sem 31: 308-315
Jones E (1950) The concept of a normal mind. In: Halmos P, Iliffe A (eds) Philosophical Library,
New York, Readings in general psychology.
Jones MC (1975) A 1924 pioneer looks at behaviour therapy. J Behav Ther Exp Psychiatry 6:
181-187
Kaelbling R, Volpe PA (1963) Constancy of psychiatric diagnoses in readmission. Compr Psychiatry
4: 29-39
Kagan J (1967) On the need for relativism. Am Psychol 22: 131-142
Kamiya J (1969) Operant control of the EEG rhythm and some of its reported effects on consciousness. In: Tart cr (ed) Altered states of consciousness. Wiley, New York
Kanfer FR, Saslow G (1967) Behavioural analysis: an alternative to diagnostic classification. In:
Millon T (ed) Theories of psychopathology. Saunders, Philadelphia London
Kanfer FR, Phillips JS (1970) Learning foundations of behaviour therapy. Wiley, New York
Kaplan HI, Kaplan HS (1956) A historical survey of psychosomatic medicine. J Nerv Ment Dis 124:
546-568
References
147
Katz A, Webb L, Stotland E (1971) Cognitive influences on the rate of GSR extinction. J Exp Res
Pers 5: 208-215
Katz M, Cole JO, Lowery HA (1969) Studies of the diagnostic process: the influence of symbolic
perception, past experience, and ethnic background on diagnostic decisions. Am J Psychiatry
125: 937-947
Kelley HH (1948) First impression in interpersonal relations. Cited in: Krech D, Crutchfield R,
Ballachey E (eds) Individual in society. McGraw-Hili, New York
Kelley HH (1950) The warm-cold variable in first impressions of persons. J Pers 18: 431-439
Kelly DHW (1966) Measurement of anxiety by forearm blood flow. Br J Psychiatry 112: 789-798
Kelly DHW, Walter CIS (1968) The relationship between clinical diagnosis and anxiety, assessed by
forearm blood flow and other measurements. Br J Psychiatry 114: 611-626
Kelly G (1955) The psychology of personal constructs. Norton, New York
Kendell RE (1975) The role of diagnosis in psychiatry. Blackwell, Oxford
Kiloh LG, Andrews G, Neilson M, Bianchi G (1972) The relationship of the syndromes called
endogenous and neurotic depression. Br J Psychiatry 121: 183-196
Kiloh LG, Garside RF (1963) The independence of neurotic depression and endogenous depression. Br J Psychiat 109: 451-63
King LS (1958) The medical world of the eighteenth century. University of Chicago Press, Chicago
.
Klein M (1960) On mental health. Br J Med Psychol 33: 237-241
Kline P (1968) Obsessional traits, obsessional symptoms and anal eroticism. Br J Med Psychol41:
299-305
Knight RP (1941) Evaluation ofthe results of psychoanalytic therapy. Am J Psychiatry 98: 434-446
Knoff WF (1970) A history of the concept of neurosis, with a memoir of William Cullen. Am J
Psychiatry 127: 80--84
Kraeplin E (1896) Psychiatrie. Barth, Leipzig
Kraeplin E (1906) Lectures on clinical psychiatry. Bailliere Tindall, London
Kreitman N (1961) The reliability of psychiatric diagnosis. J Ment Sci 107: 876--886
Kreitman N, Sainsbury P, Morrissey J, Towers J, Scrivener J (1961) The reliability of psychiatric
assessment: an analysis. J Ment Sci 107: 887-908
Kuhn TS (1970) Logic of discovery or psychology of research. In: Lakatos I, Musgrove A (eds)
Criticism and the growth of knowledge. Cambridge University Press, Cambridge
Lader M (1967) Palmar skin conductance measures in anxiety and phobic states. J Psychosom Res
11: 271-281
Lader MH (1969) Psychophysiological aspects of anxiety. In: Lader MH (ed) Studies of anxiety.
Headley, Kent
Lader MH, Gelder MG, Marks 1M (1967) Palmar skin conductance measures as predictors of
response to desensitisation. J Psychosom Res 11: 283-290
Lader MH, Noble P (1975) The affective disorders. In: Venables PH, Christie MJ (eds) Research in
psychophysiology. Wiley, London
Lader MH, Wing L (1966) Psychophysiological measures, sedative drugs and morbid anxiety.
Oxford University Press, London
Lakatos I (1970) Falsification and the methodology of scientific research programmes. In: Lakatos
I, Musgrove A (eds) Criticism and the growth of knowledge. Cambridge University Press, Cambridge
Laplanche J, Pontalis JB (1973) The language of psychoanalysis. Hogarth, London
Laughlin HP (1967) The neuroses. Butterworths, Washington, D.C.
Lazare A (1973) Hidden conceptual models in clinical psychiatry. N Engl J Med 288: 345-351
Lazarus AA, Rachman S (1960) The use of systematic desensitisation psychotherapy. In: Eysenck
HJ (ed) Behaviour therapy and the neuroses. Pergamon, Oxford
Lemert E (1967) Human deviance, social problems and social control. Prentice Hall, New Jersey
Lesse S (1970) Anxiety: its components, development and treatment. Grune & Stratton, New York,
London
Levine M (1942) Psychotherapy in medical practice. Macmillan, New York
Levy R, Meyer V (1971) Ritual prevention in obsessional patients. Proc R Soc Med 64: 1115-1118
Lewin K (1935) A dynamic theory of personality. McGraw-Hili, New York
Lewinsohn PM, Mischel W, Chaplin W, Barton R (1980) Social competence and depression: the
role of illusory self-perceptions. J Abnorm Psychol 89: 203-212
148
References
Lewis AJ (1934) Melancholia: a clinical survey of depressive states. J Ment Sci 80: 277-378
Lewis AJ (1951) Henry Maudsley: his work and influence. J Ment Sci 97: 255-277
Lewis AJ (1958) Between guesswork and certainty in psychiatry. Lancet 1: 170-175, 227-230
Liberman RP, Raskin DE (1971) Depression: a behavioural formulation. Arch Gen Psychiat 24:
515-523
Libet JM, Lewinsohn PM (1973) Concept of social skill with special reference to the behaviour of
depressed persons. J Consult ain Psychol40: 304-312
Loevinger J (1955) Diagnosis and measurement: a reply to Eysenck. Psychol Rep 1: 277-278
Lorr M, O'Connor JP (1957) The relation between neurosis and psychosis: are-analysis. J Ment Sci
103: 375-380
Lucas CJ, Kelvin RP, Ojha AB (1965) The psychological health of the preclinical medical student.
Br J Psychiatry 111: 473-478
Ludwig AM (1975) The psychiatrist as physician. JAmMed Assoc 234: 603-604
Ludwig AM, Othner EO (1977) The medical basis of psychiatry. Am J Psychiatry 134: 1087-1092
Luria AR (1961) The role of speech in the regulation of normal and abnormal behaviour. Pergamon, New York
McGuire RJ, Mowbray RM, Vallance RC (1963) The Maudsley personality inventory used with
psychiatric inpatients. Br J Psychol 54: 157-166
Mackay D (1975) Clinical psychology: theory and therapy. Methuen, London
Macmillan MB (1976) Beard's concept of neurasthenia and Freud's concept of the actual neuroses.
J Hist Behav Sci 12: 376-390
Maher BA (1966) Principles of psychopathology. McGraw-Hill, New York
Mahoney M (1974) Cognition and behaviour modification. Ballinger, Cambridge
Mahoney MJ (1977) Reflections on the cognitive-learning trend in psychotherapy. Am Psychol32:
5-13
Mahoney MJ, Kazdin AE, Lesswing NJ (1974) Behaviour modification: delusion or deliverance?
In: Franks CM, Wilson GT (eds) Annual review of behaviour therapy and practice, vol 2. Brunner-Mazel, New York
Malmo RB (1957) Anxiety and behavioural arousal. Psych Rev 64: 276-87
Malmo RB, Shagass C (1952) Studies of blood pressure in psychiatric patients under stress.
Psychosom Med 14: 82-93
Mapother E (1926) Manic-depressive psychosis. Br Med J 2: 872-879
Marks 1M (1969) Fears and phobias. Heinemann, London
Marks 1M (1970) The classification of phobic disorders. Br J Psychiatry 116: 377-386
Marks 1M (1973a) Research in neurosis: a selective review. 1. Causes and courses. Psychol Med 3:
436-454
Marks 1M (1973b) New approaches to the treatment of obsessive-compulsive disorders. J Nerv
Ment Dis 156: 420--426
Marks 1M (1974) Research in neurosis: a selective review. 2. Treatment. Psychol Med 4: 89-109
Marks 1M, Gelder MG (1967) Transvestism and fetishism: clinical and psychological changes during
faradic aversion. Br J Psychiatry 113: 711-729
Marks 1M, Hodgson R, Rachman S (1975) Treatment of chronic obsessive-compulsive neurosis by
in vivo exposure. Br J Psychiatry 127: 349-364
Marx MH, Hillix WA (1963) Systems and theories in psychology. McGraw-Hill, New York
Marzillier J (1978) Outcome studies of skills training: a review. In: Trower P, Bryant B, Argyle M
(eds) Social skills and mental health. Methuen, London
Maslow AH, Mittelmann (1951) Principles of abnormal psychology. Harper, New York
Masserman JG, Carmichael HT (1938) Diagnosis and prognosis in psychiatry: with a follow-up
study of the results of short-term general hospital therapy of psychiatric cases. J Ment Sci 84:
893-946
Masters WH, Johnson VE (1970) Human sexual inadequacy. Little, Brown, Boston
Mather MD (1970) The treatment of an obsessional-compulsive patient by discrimination learning
and reinforcement of decision making. Behav Res Ther 8: 315-318
Maturana HR, Lettvin JY, McCulloch WS, Pitts WB (1960) Anatomy and physiology of vision in
the frog. J Gen Physiol 43: 129-175
Maudsley H (1899) The pathology of mind. Appleton, New York
Meehl PE (1946) Profile analysis of the MMPIin differential diagnosis. J Appl Psychol 30: 517-524
References
149
Meehl PE (1973) Psychodiagnosis: selected papers. University of Minnesota Press, Minnesota

Meichenbaum D (1976) Toward a cognitive theory of self-control. In: Schwartz GE, Shapiro D
(eds) Consciousness and self-regulation. Advances in research, vol I. Plenum, New York
Meichenbaum D (1977) Cognitive-behaviour modification. Plenum, New York
Meichenbaum D, Cameron R (1974) The clinical potential of modifying what clients say to themselves. Psychother Theor Res Pract 11: 103-117
Melges Fr, Bowlby J (1969) Types of helplessness in psychopathological process. Arch Gen
Mendels J (1965) Electroconvulsive therapy and depression. 2. Significance of endogenous and
reactive syndromes. Br J Psychiatry 111: 682-686
Mendelson M (1974) Psychoanalytic concepts of depression. Spectrum, New York
Menninger K (1948) Changing concepts of disease. Ann Intern Med 29: 318-325
Menninger K, Mayman M, Pruyser P (1963) The vital balance. Viking, New York
Meyer V, Chesser ES (1970) Behaviour therapy in clinical psychiatry. Penguin, Harmondsworth,
Middx
Meyer V, Crisp AH (1970) Phobias. In: Costello CG (ed) Symptoms of psychopathology. Wiley,
New York
Meyer V, Levy R (1970) Behaviour treatment of a homosexual with compulsive rituals. Br J Med
Psychol43: 63~7
Miles TR (1966) Eliminating the unconscious. Pergamon, Oxford
Miller NE (1948) Studies of fear as an acquirable drive. I. Fear as motivation and fear-reduction as
reinforcement in the learning of new responses. J Exp Psychol 38: 89-101
Miller NE (1959) Liberalisation of basic S-R concepts: extensions to conflict behaviour, motivation,
and social learning. In: Koch S (ed) Psychology: a study of a science, vol 2. McGraw-Hili, New
York
Miller NE (1969) Learning of visceral and glandular responses. Science 163: 434-445
Mischel W (1968) Personality and assessment. Wiley, New York
Mischel W (1969) Continuity and change in personality. Am Psychol 24: 1012-1018
Mischel W (1973) Toward a cognitive social learning reconceptualisation of personality. Psychol
Rev 80: 252-283
Mischel W (1977) On the future of personality measurement. Am Psychol32: 246-254
Mischel W (1979) On the interface of cognition and personality: beyond the person-situation
debate. Am Psychol 34: 740-754
Money-Kyrle RE (1957) Psychoanalysis and ethics. In: Klein M, Heimann P, Money-Kyrle RE
(eds) New directions in psychoanalysis. Basic Books, New York
Monroe LJ (1967) Psychological and physiological differences between good and poor sleepers. J
Abnorm Soc Psychol 72: 255-264
Mora G (1967) History of psychiatry. In: Freedman AM, Kaplan HI (eds) Comprehensive textbook
of psychiatry. Williams & Wilkins, Baltimore
Moss P, McEvedy C (1966) An epidemic of overbreathing among schoolgirls. Br Med J 2:
1295-1300
Mowrer OH (1938) Enuresis: a method for its study and treatment. Am J Orthopsychiat 8:
436-459
Mowrer OH (1939) A stimulus-response analysis of anxiety and its role as a reinforcing agent.
Psychol Rev 46: 553-565
Mowrer OH (1950) Learning theory and personality dynamics. Arnold, New York
Mowrer OH (1969) Psychoneurotic defences (including deception) as punishment-avoidance
strategies. In: Campbell BA, Church RM (eds) Punishment and aversive behaviour. AppletonCentury-Crofts, New York
Mowrer OH (1976) Enuresis: the beginning work. Paper presented to Annual Convention of Am
Psychol Assoc 1976, Washington, D.C.
Mowrer OH, Ullman AD (1945) Time as a determinant in integrative learning. Psychol Rev 52:
61-90
Muezinger KF (1934) Motivation in learning: I. electric shock for correct response in the visual
discrimination habit. J Comp Physiol Psychol 17: 267-277
Munroe RL (1957) Schools of psychoanalytic thought. Hutchinson, London
Myerson A (1936) Neuroses and neuropsychoses. Am J Psychiatry 93: 263-301
150
References
Nagera H (1969) Basic psychoanalytic concepts on the libido theory. Allen & Unwin, London
Nebylitsyn VD (1964) An investigation of the connection between sensitivity and strength of the
nervous system. In: Gray JA (ed) Pavlov's typology. Pergamon, London
Nichols KA (1974) Severe social anxiety. Br J Med Psychol47: 301-306
Nisbett RE, Schachter S (1966) Cognitive manipulation of pain. J Exp Soc Psychol 2: 227-236
Nisbett RE, Valins S (1971) Perceiving the causes of one's own behaviour. In: Jones EE et al. (eds)
Attribution: perceiving the causes of behaviour. General Learning Press, New Jersey
Norris V (1959) Mental illness in London. Chapman & Hall, London
O'Connor JP (1953) A statistical test of psychoneurotic syndromes. J Abnorm Soc Psychol 48:
581-584
Offer D, Sabshin M (1966) Normality. In: Freedman AM, Kaplan HI (eds) Comprehensive textbook of psychiatry. Williams & Wilkins, Baltimore
Offer D, Sabshin M (1974) The concept of normality. In: Arieti S (ed) The foundations of
Psychiatry. Vol I. Basic Books, New York
Osgood CE (1953) Method and theory in experimental psychology. Oxford University Press, New
York
Overmier JB, Seligman MEP (1967) Effects of inescapable shock on subsequent escape and avoidance learning. J Comp Physiol Psychol63: 28-33
Parsons T (1952) The social system. Tavistock, London
Parsons T (1965) Social structure and personality. Collier Macmillan, London
Paul GL (1966) Insight versus desensitisation in psychotherapy: an experiment in anxiety reduction.
Stanford University Press, Stanford, Ca.
Pavlov IP (1927) Conditioned reflexes. Oxford University Press, London
Pavlov IP (1928) Lectures on conditioned reflexes, voll. International, New York
Pavlov IP (1941) Lectures on conditioned reflexes, vol 2. Lawrence Wishart, London
Pavlov IP (1955) Selected works. Foreign Languages Publishing House, Moscow
Phillips L, Zigler E (1961) Social competence: the action-thought parameter and vicariousness in
normal and pathological behaviours. J Abnorm Soc Psychol 63: 137-146
Pilowsky I, Levine S, Boulton DM (1969) The classification of depression by numerical taxonomy.
Br J Psychiatry 115: 937-945
Pinel P (1801) A treatise on insanity. Facsimile of the London 1806 edition. Hafner, New York
1962
Plaut A (1960) A concept of mental health. Br J Med Psychol 33: 275-278
Pollin W (1976) Genetic and environmental determinants of neurosis. In: Kaplan AR (ed) Human
behaviour genetics. Thomas, Springfield, Ill.
Popper KR (1963) Conjectures and refutations. Routledge & Kegan Paul, London
Rabkin R (1964) Conversion hysteria as social maladaptation. Psychiatry 27: 349-363
Rachman S (1965) Aversion therapy: chemical or electrical? Behav Res & Ther 2: 289-299
Rachman S (1966) Studies in desensitisation: II. flooding. Behav Res Ther 4: 1-6
Rachman S (1972) Clinical applications of observational learning, imitation and modeling. Behav
Ther 3: 379-397
Rachman S (1974) The meanings of fear. Penguin, Middlesex
Rachman S (1976) Obsessional-compulsive checking. Behav Res Ther 14: 269-277
Rachman S (1977) The conditioning theory of fear-acquisition: a critical examination. Behav Res
Ther 15: 375-387
Rachman S, Hodgson RJ (1968) Experimentally-induced 'sexual fetishism': replication and
development. Psychol Rec 18: 25-27
Rachman S, Hodgson R (1974) I. Synchrony and desynchrony in fear and avoidance. Behav Res
Ther 12: 311-318
Rachman S, Hodgson R, Marks I (1971) Treatment of chronic obsessive-compulsive neuroses.
Behav Res Ther 9: 237-247
Rachman S, Teasdale J (1969) Aversion therapy and behaviour disorders: an analysis. Routledge &
Kegan Paul, London
Rachman S, Seligman MEP (1976) Unprepared phobias: "be prepared". Behav Res & Ther 14:
333-338
Ramsay RW (1975) Research on anxiety and phobic reactions. In: Spielberger CD, Sarason IG
(eds) Stress and anxiety, voll. Hemisphere, Wiley, New York
References
151
Rescorla RA (1969) Establishment of a positive reinforcer through contrast with shock. J Comp
Physiol Psychol67: 260-263
Rogers CR (1965) Some thoughts regarding the current philosophy of the behavioural sciences. J
Humanistic Psychol 5: 182-194
Rosen A (1962) Development of the MMPI scales based on a reference group of psychiatric
patients. Psychol Monogr 76: 1-25
Rosenhan D (1973) On being sane in insane places. Science 179: 250-258
Roth M (1963) Neurosis, psychosis and the concept of disease in psychiatry. Acta Psychiatr Scand
39: 128-145
Rush AJ, Beck AT, Kovacs M, Hollon S (1977) Comparative efficacy of cognitive therapy and
pharmacotherapy in the treatment of depressed outpatients. Cog Ther Res 1: 17-37
Rust J (1974) Genetic factors in psychophysiology. Unpublished PhD thesis, University of London,
London
Rutter M, Korn S, Birch HG (1963) Genetic and environmental factors in the development of
'primary reaction patterns'. Br J Soc Clin Psychol 2: 161-173
Ryle A (1967) Neurosis in the ordinary family. Lippincott, Philadelphia
Sandifer MG (1977) The education of the psychiatrist as a physician. Am J Psychiatry 134: 50-53
Sandifer MG, Hordern A, Timbury GC, Green LM (1969) Similarities and differences in patient
evaluation by US and UK psychiatrists. Am J Psychiatry 126: 206-212
Sanford N (1965) Will psychologists study human problems? Am Psychol 20: 192-207
Sarbin TR (1964) Anxiety: reification of a metaphor. Arch Gen Psychiatry 10: 630-638
Sartory G, Rachman S, Grey S (1977) An investigation of the relation between reported fear and
heart rate. Behav Res Ther 15: 435-438
Sartory G, Lader M (1980) Psychophysiology and drugs in anxiety and phobias. In: Christie M,
Mellett P (eds) Psychosomatic approaches in Medicine, voll. Behavioural approaches. Wiley,
London
Saslow G, Peters AD (1956) A follow-up study of 'untreated' patients with various behaviour
disorders. Psychiatr Quart 30: 283-302
Schachter S, Singer JE (1962) Cognitive, social and physiological determinants of emotional state.
Psychol Rev 69: 379-399
Schachter S, Wheeler L (1962) Epinephrine, chlorpromazine and amusement. J Abnorm Soc
Psychol 65: 121-128
Scheff TJ (1966) Being mentally ill. Weidenfeld & Nicolson, London
Scheier IH (1972) Anxiety at a distance. In: Dreger RM (ed) Multivariate personality research.
Claitors, Baton Rouge
Schmidt HO, Fonda CP (1956) The reliability of psychiatric diagnosis: a new look. J Abnorm Soc
Psychol 52: 262-267
Schoeneman (1977) The role of mental illness in the witch hunts of the seventeenth centuries: an
assessment. J Hist Behav Sci 13: 337-351
Schofield M (1965) Sociological aspects of homosexuality. Longman, London
Sechenov 1M (1863) Reflexes of the brain. Reprinted in Selected Works. Bookmiga, Moscow 1935
Sederer L (1977) Moral therapy and the problem of morale. Am J Psychiatry 134: 267-272
Seligman MEP (1970) On the generality of the laws of learning. Psychol Rev 77: 406-418
Seligman MEP (1971) Phobias and preparedness. Behav Ther 2: 307-320
Seligman MEP (1975) Helplessness: on depression, development and death. Freeman, San Francisco
Seligman MEP, Johnston JC (1973) A cognitive theory of avoidance learning. In: Guigan F, Lumsden D (eds) Contemporary approaches to conditioning and learning. Winston, Washington
Seligman MEP, Klein DC, Miller WR (1976) Depression. In: Leitenberg H (ed) Handbook of
behaviour modification and behaviour therapy. Prentice Hall, New Jersey
Seligman MEP, Maier SF (1967) Failure to escape traumatic shock. J Exp Psychol 74: 1-9
Seligman MEP, Meyer B (1970) Chronic fear and ulcers in rats as a function of the unpredictability
of safety. J Comp Physiol Psychol 73: 202-207
Selye H (1950) The physiology and pathology of exposure to stress. Acta Inc, Montreal
Selye H (1957) The stress of life. Longman, London
Shagass C, Jones AL (1958) A neurophysiological test for psychiatric diagnosis. Am J Psychiatry
114: 1002-1009
152
References
Shakow D (1965) The role of classification in the development of the science of psychopathology
with particular reference to research. In: Katz MM, Cole JO, Barton WE (eds) The role and
methodology of classification in psychiatry and psychopathology. US Government Printing
Office, Washington, DC
Shaw BF (1977) Comparison of cognitive therapy and behaviour therapy in the treatment of
depression. J Consult Clin Psychol 45: 543-551
Sheflen AE (1958) Analysis of thought model which persists in psychiatry. Psychosom Med 20:
235-241
Shepherd M (1973) The prevalence and distribution of psychiatric illness in general practice. J R
Coli Gen Pract 23: 16-19
Shepherd M, Cooper B, Brown AC, Kalton G (1966) Psychiatric illness in general practice. Oxford
University Press, London
Sherif M (1936) The psychology of social norms. Harper, New York
Shey HH (1971) Iatrogenic anxiety. Psychiatr Quart 45: 343-356
Shields J (1962) Monzygotic twins. Oxford University Press, London
Sidman M (1964) Anxiety. Proc Am Phil Soc 108: 478-481
Siegelman M (1978) Psychological adjustment of homosexual and heterosexual men: a crossnational replication. Arch Sex Behav 7: 1-11
Silver RJ, Sines LK (1961) MMPI characteristics of a state hospital population. J Clin Psychol 17:
142-146
Simon B (1972) Models of mind and mental illness in Ancient Greece: II. the Platonic model. J Hist
Behav Sci 8: 389-404
Simon B (1973) Models of mind and mental illness in Ancient Greece: II. the Platonic model. J Hist
Behav Sci 9: 3-17
Simon B, Weiner H (1966) Models of mind and mental illness in Ancient Greece: I. the Homeric
model of mind. J Hist Behav Sci 2: 303-314
Slater E (1965) Diagnosis of 'hysteria'. Br Med J 1: 1395-1399
Slater E, Roth M (1969) Clinical psychiatry. Bailliere & Tindall Cassell, London
Slater E, Shields J (1969) Genetical aspects of anxiety. In: Lader MH (ed) Studies of anxiety.
Headley, London
Solomon RL, Wynne LC (1953) Traumatic avoidance learning: acquisition in normal dogs. Psychol
Monogr 67/354
Solomon RL, Wynne LC (1954) Traumatic avoidance learning: anxiety conservation and partial
irreversibility. Psychol Rev 61: 353-385
Solomon RL, Turner LH, Lessac MS (1968) Some effects of delay of punishment on resistance to
punishment in dogs. J Pers Soc Psychol 8: 233-238
Spence KW (1964) Anxiety (drive) level and performance in eyelid conditioning. Psychol Bull 61:
129-139
Spielberger CD, Gorsuch RL, Lushener RE (1968) The state-trait anxiety inventory. Florida State
University, Florida
Stein M, Ottenberg P (1958) Role of odours in asthma. Psychosom Med 20: 60-65
Storms MD, McCaul KD (1976) Attributional processes and emotional exacerbation of dysfunctional behaviour. In: Harver JH, Ickes WJ, Kidd RF (eds) New directions in attribution research.
Wiley, New York
Storms MD, Nisbett RE (1970) Insomnia and the attribution process. J Pers Soc Psychol 16:
319-328
Strauss JS (1975) A comprehensive approach to psychiatric diagnosis. Am J Psychiatry 132:
1193-1196
Stuart RB (1970) Trick or treatment: how and when psychotherapy fails. Research Press, Champaign, III.
Szasz TS (1960) The myth of mental illness. Am Psychol 15: 113-118
Szasz TS (1961) The myth of mental illness. Hoeber & Harper, New York
Taylor F Kraupl (1966) Psychopathology. Its causes and symptoms. Buttelworths, London
Taylor JA (1953) A personality scale of manifest anxiety. J Abnorm Soc Psychol 48: 285-290
Teasdale J (1974) Learning models of obsessional-compulsive disorder. In: Beech HR (ed) Obsessional states. Methuen, London
References
153
Thomas A, Chess S, Birch HG (1968) Temperament and behaviour disorders in children. New York
University Press, New York
Thompson C (1950) Psychoanalysis: evolution and development. Nelson, New York
Thomson R (1968) The pelican history of psychology. Penguin, Middlesex
Thorndike EL (1903) Educational psychology. Lemcke & Buechner, New York
Thorndike EL (1911) Animal intelligence. Macmillian, New York
Thorndike EL (1932) The fundamentals of learning. Teachers College, New York
Tolman EC (1948) Cognitive maps in rats and men. Psychol Rev 55: 189-208
Tredgold RF (1941) Depressive states in the soldier: their symptoms, causation and prognosis. Br
Med J 2: 109-112
Trouton DS, Maxwell AE (1956) The relationship between neurosis and psychosis. J Ment Sci 102:
1-21
Trower PE, Yardley KM, Bryant BM, Shaw PH (1978) The treatment of social failure: a comparison of anxiety reduction and skills acquisition procedures on two social problems. Behav Mod 2,
41--60
Truax CB (1966) Reinforcement and non-reinforcement in Rogerian psychotherapy. J Abnorm
Psychol 71: 1-9
Tuke S (1813) Description of The Retreat. Reprinted 1964. Dawson, London
Ullman LP, Krasner L (1967) The psychological model. In: Millon T (ed) Theories of psychopathology. Saunders, Philadelphia
Ullman LP, Krasner L (1969) A psychological approach to abnormal behaviour. Prentice-Hall, New
Jersey
Valins S (1966) Cognitive effects of false heart-rate feedback. J Pers Soc Psychol 4: 400-408
Valins S, Nisbett RE, Jones EE, Kanouse DE, Kelley HE, Weiner B (1971) Attribution processes in
the development and treatment of emotional disorders. In: Jones et al. (eds) Attribution. General
Learning Press, New Jersey
Vandenberg SG, Oark PJ, Samuels I (1965) Psychophysiological reactions of twins: heritability
factors in galvanic skin resistance, heartbeat, and breathing rates. Eugen Quart 12: 7-10
Wallace HER, Whyte MBH (1959) Natural history of the psychoneuroses. Br Med J 1: 144148
Ward CH, Beck AT, Mendelson M, Mock JE, Erbaugh JK (1962) The psychiatric nomenclature:
reasons for diagnostic disagreement. Arch Gen Psychiatry 7: 198-205
Warner EC (1952) Savill's system of clinical medicine. Arnold, London
Watson JB (1930) Behaviourism. Kegan Paul & Trench Trubner, London
Watson JB, Rayner R (1920) Conditioned emotional reactions. J Exp Psychol 3: 1-14
Weiss E (1942) Anxiety and the heart. Clinics 1: 916-931
Weiss 1M (1968) Effects of coping responses on stress. J Comp Physiol Psychol 65: 251-260
Weiss 1M, Glazer HI, Pohorecky LA (1976) Coping behaviour and neurochemical changes. In:
Serban G, Kling A (eds) Animal models in human psychobiology. Plenum, New York
Wexler M (1971) Schizophrenia: conflict and deficiency. Psychoanal Q 40: 83-99
Wilde GJS (1964) Inheritance of personality traits. Acta Psychol 22: 37-51
Williams HV, Lipman RS, Rickels K, Covi L, Uhlenhuth EH, Mattsson NB (1968) Replication of
symptom distress factors in anxious neurotic outpatients. Multivar Behav Res 3: 199-211
Willis MH, Blaney PH (1978) Three tests of the learned helplessness model of depression. J
Abnorm Psychol 87: 131-136
Wing JK (1973) Social and familial factors in the causation and treatment of schizophrenia. In:
Iverson LL, Rose SPR (eds) Biochemistry and mental illness. Biochemical Society, London
Wing JK, Cooper JE, Sartorius N (1974) The measurement and classification of psychiatric symptoms. Cambridge University Press, London
Wing L (1964) Physiological effects of performing a difficult task in patients with anxiety states. J
Psychosom Res 7: 283-294
Wolpe J (1952) Experimental neurosis as learned behaviour. Br J Psychol43: 243-261
Wolpe J (1958) Psychotherapy by reciprocal inhibition. Stanford University Press, Stanford, Ca.
Wolpe J (1960) Reciprocal inhibition as the main basis of psychotherapeutic effects. In: Eysenck HJ
(ed) Behaviour therapy and the neuroses. Pergamon, London
Wolpe J (1967a) Etiology of human neuroses. In: Millon T (ed) Theories of psychopathology.
Saunders, London
154
References
Wolpe J (1967b) Parallels between animal and human neuroses. In: Zubin J, Hunt HF (eds)
Comparative psychopathology. Grune & Stratton, New York
Wolpe J (1968) Learning theories. In: Howells JG (ed) Modern perspectives in world psychiatry.
Oliver Boyd, London
Wolpe J (1970) The experimental foundations of some new psychotherapeutic methods. In: Gantt
WH, Pickerhain L, Zwingmann C (eds) Pavlovian approach to psychopathology. Pergamon,
London
Wolpe J (1978) Cognition and causation in human behaviour and its therapy. Am Psychol 33:
437-446
Wolpe J, Brady JP, Serber M, Agras WS, Liberman RP (1973) The current status of systematic
desensitisation. Am J Psychiatry 130: 961-965
Wolpe J, Rachman S (1960) Psychoanalytic 'evidence': a critique based on Freud's case of Little
Hans. J Nerv Ment Dis 130: 135-148
Wortman CB, Dintzer L (1978) Is an attributional analysis of the learned helplessness phenomenon
viable? J Abnorm Psychol 87: 75-90
Wortman CB, Panciera L, Shusterman L, Hibscher J (1976) Attributions of causality and reactions
to uncontrollable outcomes. J Exp Soc Psychol 12: 301-316
Wyss D (1966) Depth psychology: a critical history. Allen & Unwin, London
Yarrow M, Schwartz C, Murphy H, Deasy L (1955) The psychological meaning of mental illness in
the family. J Soc Iss 11: 12-24
Yates AJ (1958) Symptoms and symptom substitution. Psychol Rev 65: 371-374
Yates AJ (1975) Theory and practice in behaviour therapy. Wiley, New York
Zajonj RB (1965) Social facilitation. Science 149: 269-274
Zilboorg G, Henry GW (1941) A History of medical psychology. Norton, New York
Zubin J (1967) Classification of the behaviour disorders. Annu Rev Psychol 18: 373-406
11. Author Index
Abraham K 96
Abramson L Y 84, 89-90
Adcock CJ 116
Adelman HM 60
Adelman MR 57
Adler A 2,94,101-102,104,106
Azjen I 120
AlbeeOH 19
AmselA 60
ArgyleM 86
AshP 34
AzrinNH 59
Bandura A 19,71,72,73,74-75,89
Barton WE 39
Beard OM 8, 15,98
BeckAT 34,83,87,136
BedfordA 33
Beech HR 58,60,73
BellakL 47
BemDJ 119
Benjamin S 69
Berger SM 74,89
Berkley HJ 12
Berlyne DE 71
Blanchard EB 108
BlockJD 129
BlumOS 15
Blumberg HH 34
Bockoven JS 19,20
BollesRC 64
Borkovec TD 76
Boulougouris JC 69
Brady JV 84
Braginsky DD 19
BregmanE 56
BreuerJ 17,92,93,96
Brewer WF 72, 73
BrillAA 18
Brill H 28
Briquet 9, 135
Britton RS 75
Broadhurst PL 129
BrownFW 130
Brown OW 77-78,79,80,134
Brown RT 60
Buchwald AM 85
Burt C 49
Carney MWP 49, 136
Caspari EW 131
CattellRB 28,41,42,50-51,109,110,
111-113,116,118-119,125,129,130,
132,133
CavenyEL 26
Charcot JM 8-9, 16, 135
Chesser ES 60, 68
ChurchRM 74
Clare A 14,26,37
ClaridgeOS 117,126-127,135
Coates DB 77
Cole S 44,79
Connell PH 34
CookSW 72
Costello CO 37
Cottrell NB 86
Craighead WE 15
CrispAH 107
Cullen W 3,8,11,18
Culpin M 11,16
Curran D 34
D'AmatoMR 67
DavisJF 124
Denker R 63,107
Derogatis L 35, 36, 50
Deutsch H 103
DietzPE 14
Dollard J 58, 60, 62, 72
Ducey CIS, 16
Eiser JR 81,120
Ellenberger HF 9,16,17
EllisA 88
Endler NS 118, 119
EngelOL 14
Erikson K 43-44
Essen-Moller E 32, 37, 44
Everitt BS 35, 36,48
156
Eysenck HJ 7,10,14,15-18,27,41,49-51,
55-58,61-67,74,82,86,92,106,109,
110,114-118,125-129,133,134,135
Eysenck SBG 49-50, 116
Fahrenberg J 121, 125-126
FarrellBA 105,106
Feldman MP 70
FenischelO 48
Fenton N 62
Fishbein M 120
Flew A 58
Forrest AR 118, 127
Foulds GA 14,33,37
Frank G 26,32
Freud S 1-2,8,9,40,46--49,50,59,91-108,
113,129,133,134-135
Frith CD 125
FrolovYP 22
Fromm E 2,101,103,104-105
Galen 7,117
GarciaJ 67
Garmany G 48--49
Gelder MG 69-70
GiorgiA 19
GlassDC 85
Gleitman H 60
GoffmanE 19
GoldbergDP 35,41,42
Goldstein IB 123
Gormezano I 20, 22
Gossop MR 34, 67, 88, 118, 120, 127
Gottesman II 130
GoughHG 51
Gove WR 44, 80
Grant DA 21,64
Gray J 59--60,62,63,64,117,127-128
Gray S 76
Greenson RR 93
Griesinger W 11, 13, 14
Grinker RR 33, 134
Guilford JP 110--111, 112
Guilford RB 110
Guthrie ER 24
Guze SB 135
Hagnell 0 44, 132
Hallam RS 69-70, 87
HanusaBH 85
Harris RE 72
Harris T 77-78,79,80, 134
Hartmann H 40
Hartshorne H 118, 119
HarzemP 106
Hearst E 59,67
HebbDO 59
Author Index
Heilbrunn G 107
Helmholtz H von 17,94
Hempel CG 27,36,132
Henderson D 39
Hendrick I 95, 102
Henry GW 7,13
Hersen M 44, 108
HineFR 36
Hippocrates 7,11,22
Hodgson R 58,63,69,76,121
Hoghughi MS 118,127
Hollingshead AB 76-77,80
HolzWC 59
Homer 3,15,16
Horney K 2,92,101,102-104,105
HoweES 123
Howells JG 27
Hull CL 24,59,61,126
Humphreys LG 72
HuntHC 19
HuntJMcV 118,119
Hunt WA 26, 34
Hunter R 8,14
InghamJG
117
Jacobsen CF 58
Jahoda M 40,42
JanetP 9,16,17,135
JaspersK 9-11,32-33
Jellinek EM 26
Johnson VE 82
Jones E 40
Jones MC 68,74
Jung CG 9, 10, 116
Kaelbling R 34
KaganJ 19
KamiyaJ 29
KanferFH 71
Kaplan HI 7
KatzA 72
KatzMM 37
KazdinAE 72
Kelley HH 37,87
KellyDHW 123,124-125,137
Kelly G 119
Kendell RE 30,32,48, 133
Kiloh LG 48, 49
King LS 18
Klein M 2,40,95
KlineP 106,113,118-119
Knight RP 107
KnoffWF 8
Kraeplin E 12,13-14,26,27
Krafft -Ebing R von 13
Krasner L 7,41,74,79
157
Author Index
Kreitman N 34, 35
Kristjansson I 118, 127
Kuhn TS 71,106
Lader~ 121,122,123-125,129,137
Lakatos I 28
Laplanche ] 92
Laughlin HP 15
Lazare A 14
Lazarus AA 68
LejeuneR 44
Lemert E 43, 79
LesseS 33
Levine~ 40
Levy R 69,108
LewinK 73
Lewinsohn P~ 44, 119
Lewis A] 11,39,40,48,49
Libet]~ 44
Locke] 20
Loevinger ] 50
LucasC] 45
Lucretius 11
LudwigA~
14
LuriaAR 71
Macalpine I 8
~cGuire RJ 127
~ackayD 29
~acmillan ~B 8
~aher BA 27,60
~ahoney ~ 72, 86
~aier SF 83, 123
~almo RB
123,124
~apother E 48
~arksI~ 11,69-70,74,75,79,85,86,108
~arx~H 16
~arzillier ]
87
~aslowAH 40
~asserrnan]H
33
~asters WH 82
~ather~D 60
~aturana HR 67
~audsleyH 11-12,129
~eehl PE 36, 51
~eichenbaum D
87-89
~elgesFT 83
~endels ]
136
~endelson ~ 48
~enninger K 26, 40
~esmer 8,15
~eyerB 84
~eyer V 68, 69, lO7, 108
~ilesTR
106
~iIIer NE 29,58,59--63,72
~ischel W 118,120
~oney-Kyrle RE 42
82
7,19
~oss P
135
~owrer OH 24, 56-57, 61--64, 74, 108, 134
~uezinger KF 24
~unroeRL 46
~yersonA 48
~onroeU
~oraG
NageraH 91
Nebylitsyn VD 23
Nichols KA 86
Nietzsche F 16, 17
Nisbett RE 43,81-82,133
NobleP 125
Norris V 34
O'Connor ]P 50
OfferD 39,40,41
OsgoodCE 71
OthnerEO 14
Overrnier JB 83, 122, 137
Parsons T 43, 46
Paul GL 69,107
Pavlov IP 3,7,20-24,58,64,71,114,126
Phillips L 44,71
Pilowsky I 35
Pinel P 18
Plato 15,16
PlautA 39
Pollin W 131
Pontalis JB 92
Popper KR lO6
RabkinR 136
Rachman S 55,57,58,63,66,67-70,74,76,
86,89,100,118,121,125
Ramsay RW 88,93
Rayner R 21,22,56,68
Redlich FC 76-77,80
Rescorla RA 64
Rogers CR 26, 105
RosenA 51
Rosenhan D 20,37,73,80
Roth ~ 12,26,32,39,46
Rush A] 90
Rust] 130
Rutter~
129
RyleA 45
Sabshin ~ 39,40,41
Sandifer ~G 14,36
SanfordN 19
SarbinTR 28
SartoryG 76,121,124
SaslowG 107
Schachter S 42-43,81,83, 133
158
Scheff TJ 44, 79
Scheier IH 28,41,50-51,112-113,116,
118-119,130,132
Schmidt HO 34
Schoeneman TJ 7
Schofield M 57
Schopenhauer A 16,17
Sechenov 1M 20
Sederer L 18, 19
Seligman MEP 66,67--68,83-84,122-123,
128,136--137
SelyeH 121-122
ShagassC 123,124,125,126
ShakovD 27
ShawBF 90
Sheflin AE 14
Shepherd M 45,132
SherifM 37
SheyHH 45
Shields J 130-131
SidmanM 55
Siegelman M 57
Silver RJ 51
Simon B 15,16
SingerJE 42-43,81,83,85,87,133
Skinner BF 3, 19,24,61
SlaterE 12,26,32,39,130-131,135
Solomon RL 57,63
Spence KW 24, 117
Spielberger CD 119
SteinM 107
StormsMD 81-82,133
Strauss JS 28
Stuart RB 15,107
SydenhamT 8
Szasz TS 7,9, 14, 135
Taylor F Kraupl 46
Taylor JA 117
Teasdale J 59, 60, 70
Author Index
ThomasA 129
Thompson C 104
Thomson R 12
Thorndike EL 3,23-24,61,72
Tolman EC 24--25
Tredgold RF 48
Trouton DS 48
TrowerPE 87
Truax CB 105
Tuke S 18-19
Ullman LP 7,41,61,74,79
Valins S 43,82,133
Vandenberg SG 129
Wallace HER 107
Ward CH 35
Warner EC 26,32, 133
Watson JB 3,21,22-24,56,60,68,74
WeissE 45
Weiss JM 122, 137
WexlerM 47
Wilde GJS 130
Williams HV 35
WillisMH 85
Wing JK 35,37
WingL 123-124,129
WolpeJ 55,58-59,60,64,66--67,68-69
71-73,75,87,100
Wortman CB 85
WyssD 103
YarrowM 80
Yates AJ 69
ZajoncRB 86
Zilboorg G 7, 13
Zubin J 32,34
12. Subject Index
-, instrumental 20, 23-24, 61-64

Adrenalin 42,81,121
-, theories (see also extinction, reinforcement,
Agoraphobia 66,68,69,79,85,124
stimulus (etc.) 55-70
Alcoholism 31,32,43,69,79
Conflict (see also intrapsychic conflict) 58-61,
Amnesia 9,29,97,132
62
Anxiety, basic 102-104
Consistency of behaviour 3, 109, 118-120
-,conceptof 1,28,55,75-76
Conversion and dissociative disorders 1,29,
-, factor 112-113
99, 113, 132, 136
-, free-floating 1,29,33,66--67,69, 134
Cortical inhibition 23,116
-, neurosis 1,29,30,32
Crime and delinquency 12, 56, 58, 118
-, state 9,10,34,55,58,113,119,123,125,
129-130
Defence 94,99,134
Arousal 43,72,75,81,123,126--128
Depersonalisation 31, 124
Asthma 8,31,33,107
Depression, and anxiety 33,75,134
Attribution 42-43,76,79,81-82,88,119,
-, endogenous or psychotic 30,48-49,125,
133
131,136
Autonomic nervous system 42,56,58,63,68,
-, neurotic or reactive 1,10,30,34,35,
75--76 121-126,129,133
48-49,68,76,79,83-85,113,116,124,131,
Aversion therapy 69-70
134,136
Avoidance 59-61,84,85,89,121
Determinism 91-92
Behaviour therapy 3, 19,40,56,68,79,88,
Diagnosis (see classification)
Drive 24,62,64-66,94,134
92
Drug dependence 31,32,34,43,69,79,118
Blood flow (forearm) 123, 124
Dysthymic neurosis 63,66,117,127
Blood pressure 55,111,118,121-125
Bodily humours 7
Effect, law of 23-24, 61, 106
Catharsis 17
Ego 40,46-47,94-97,100,102
Ego-strength 113
Character neurosis 31, 108
Electroconvulsive therapy, ECT 136
Classification and description 26-27,32,
Electrodermal response 72, 111, 118-125,
132-133
Classification and diagnosis 9-15,18,26-37,
129
Enuresis 31, 56, 108, 117
73,8~ 9~ 116,133,136,137
Classification, reliability and validity
Epilepsy 8, 12
of 34-36,51
Excitation-inhibition balance 22-23,126-127
Cognition 24-25,45,61,67,71-90,119,133, Expectancies 3,61,72,76,84,120,136--137
136
Experimental neurosis 22, 58
- and depression 83-85,136
Extinction 20,62-66,69,89,134
Cognitive triad 83, 136
Factor analysis 49,110-116
Compulsive thinking 1, 30, 93
-, primary factors 111-116,130
Conditioned fear response 22,74
- response 21,22,56--58,59,61-66,74,89, -, second-order and higher-order fac134
tors 112-118
- stimulus 20,21,23,56--58,62-70
Fear 22,56,58,60,61,64,68,84,86,121,
Conditioning, backward 22
128
-, classical 20-23,24,55-58,64-70,72
Fetishism 57-58,59,66,79
160
Subject Index
Flooding 66,69,89,108
Frustration 60-61
Frustrative non-reward 60-61,64,128
Nocive responses 65-66

Noradrenalin 122-123,137
Normality 39-46
General adaptation syndrome 121-122

Geneticfactors 3,12,22,51,109,128-131
Obsessional and compulsive disorders 1,9,

10,30,35,55,58,59,69,73,93,95,99,106,
108,111,113,130,132
Oedipus complex 15,26,95-97,100-101
Headache 8,31,33
Heart rate 42,55,70,81,118,121-125,130
Helplessness 82,83-85,89,102-103,
122-123,137
Heredity (see genetic factors)
Homosexuality 57,69
Hypnotism 8,15,16
Hypochondriasis 8, 31, 97, 112, 136
Hypothalamus 121,122
Hysteria (see also conversion and dissociative
disorders) 8-9,10,14,17,29,31,63,79,
92-94,97,108,117,126,130,135
Impotence, male sexual 69,82, 122
Incubation 64-66, 134
Inferiority feelings 101-102, 111
Inhibition (see excitation)
Insight 46
Insomnia 31,81-82,122,133
Intrapsychic conflict 3, 16,29,30,92,95,
97-98, 100, 133
Introversion-extraversion 41, 57, 110,
114-118,126-128,130
Labelling 43-44,78-81
Learned helplessness (see helplessness)
Libido 95,96,98,101,102,106,122,129,
135
Malingering 29
Masochism 59
Medical model of illness 3,8-9, 11-15, 18,
20,28,39,41,79,83
Modelling 69,74-76,89
Moral treatment 18-20
Muscle tension 55
Nervous disease 8, 11, 12, 18
Neurasthenia 1,8,9,10,15,22,30-31,32,
97-98
Neurosis (see also specific disorders, anxiety,
experimental neurosis, war neurosis etc.)
-,conceptof 8,11,28,31-32,50,78-79,
113,114,124,132,134
-, incidence of 1,44-45,75,76,132,135
-, mixed 31,50,98
-, moral connotations of 11, 18-20, 135
-, physical causes of 7-15
Neuroticparadox 61,64,133
Neuroticism 41,45,50,57,112,114-118,
125-126,129,133
Pain 8,56,58,61,64,70
Panic 30,82, 133
Paradigm 71
Paralysis 9,10,13,29
Person perception 37. 105
Personal construct theory 119
Personality disorder 10,30,33,34,124,126,
131
Phobic disorders 1,9,22,30,35,55,58,61,
66,68,75,8299,100,108,112,124-125,
132,137
Physiology and neurosis 10,22-23,59,70,
76,97-98,121-128,133,137
- and personality 3,109,125-128
Placebo 81,125
Pleasure principle 23,95,106
Possession by demons 7
Post-Freudian theories 2,101-105
Preparedness 67-68
Pseudo-patients 37,73
Psychasthenia 9, 10, 117
Psychoanalysis 1,2,3,9,15-18,23,39-40,
50,62,91-108,109,134,137
Psychopathy 116,117,126,128
Psychosis 8,10,14,34,44,46-51,76,99
-, affective 30,35,37,73,99
Psychosomatic disorders 8,33,59,84,97,
112,113,116,122
Psychoticism 41, 114
Pulse rate (see heart rate)
Punishment 58-62,127-128
Questionnaire data
110-116,130
Rational emotive therapy 88

Reality principle 17
Reciprocal determinism 73
Regression 46,48, 50, 96-97, 99
Research psychologists 109
Reinforcement 20,24,61,72,75
Repression 16,17,56,60,93-94,97,99-100,
102,106,133
Resistance to treatment 88,93
Reticular activating system 127-128
Ritual behaviour 1,30,59,79,87,128
Ritual prevention 69,108
161
Subject Index
Safety signals 63--64, 84
Schizophrenia 14,30,34-35,37,47,73,124,
131
Secondary gain 29,30,102,108,135
Sedation threshold 125,126
Self-actualisation 40
Sick role 9,43,79, 134
Skin conductance (see electrodermal response)
Social anxiety (social phobia) 85-87,124
- class 76-78,134
- context 36-37,40,42-44,73-74,76-79,
101-105,134
Social skills 44, 85-87
Societal reactance theory (see labelling)
Spiral after-effect 117, 126
Spontaneous recovery 20,63,107
Stimulus, concept of 66--67,71-72
Stomach ulcers 59,84, 122
Stress 45,77,82,109,116,121-122,133
Stuttering 31,82,102
Sweating (see also electrodermal response)
Symptom substitution 17, 108
Systematic desensitisation 68--69,87,89,108,
124
Syphilis 10, 13
Thanatos 15,95
Traits 3,109-120,133
Transference 19
Trepanation of the skull 7
Twin studies 129-131
Unconscious
16-17,92-94
Voyeurism 59,69
War neurosis 58,62,122
Witches 7
w'H.Gaddes
Learning Disabilities and

Brain Function
A Neuropsychological Approach
With a Foreword by W, M. Cruickshank
1980. 45 figures, 4 tables. XVI,403 pages
ISBN 3-540-90486-7
Contents: Neurology and Behavior: Some Professional
Problems. - Prevalence Estimates and Etiology of Learning
Disabilities. - The Nervous System and Learning. - The
Use of Neuropsychological Knowledge in Understanding
Learning. Disorders. - Perceptual Disorders. - Sensory and
Motor Pathways and Learning. - Cerebral Dominance,
Handedness, and Laterality. - Languange Development,
Aphasia, and Dyslexia. - The Neuropsychological Basis of
Problems in Writing, Spelling, and Arithmetic. - Remediation, Therapy, and the Learning-Disabled Child. - Postscript. - Appendices. - Glossary. - References. - Indexes.
Neural Mechanisms in Behavior
A Texas Symposium
Editor: D. McFadden
With contributions by H. B. Barlow, R M. Boynton,
E. V. Evarts, E. R Kandel, F. Ratliff, 1. E. Rose,
R F. Thompson
1980. 151 figures, 1 table. XI, 308 pages
ISBN 3-540-90468-9
Contents: Neural Correlates of Some Psychoacoustic
Experiences. - Design for an Eye. - Form and Function:
Linear and Nonlinear Analyses of Neural Networks in the
Visual System. - Cortical Function: A Tentative Theory
and Preliminary Tests. - The Search for the Engram, II. Brain Mechanisms in Voluntary Movement. - Cellular Insight into the Multivariant Nature of Arousal. - General
Discussion. - Index.
D. w'pfaff
Estrogens and Brain Function
Springer-Verlag
Berlin
Heidelberg
New York
Neural Analysis of a Hormone-Controlled Mammalian

Reproductive Behavior
1980. 109 figures, 1 color plate, 20 tables. X, 281 pages
ISBN 3-540-90487-5
Contents: Introduction. - Triggering the Behavior: Sensory
and Ascending Pathways: Stimulus. Primary Sensory
Neurons. Spinal Intemeurons. Ascending Neural
Pathways. - Facilitating the Behavior: Sex Hormones in
the Brain: Steroid Sex Hormone Binding by Cells in the
the Vertebrate Brain. Hypothalamic Mechanisms. Hypothalamic Outflow. Midbrain Module. - Executive Control
over the Behavior: Descending and Motor Pathways:
Brainstem to Spinal Cord. Motoneurons and Response
Execution. - Building on this Paradigm: Logical and
Heuristic Developments. - Summary. Epiloque. References.
1. C. Eccles
The Human Psyche
The GIFFORD Lectures

University of Edinburgh 1978-1979
1980. 76 figures, 2 tables. XV, 279 pages

ISBN 3-540-09954-9
H. 1. Eysenck
The Structure and Measurement of

Intelligence
With Contributions by D. W. Fulker
1979. 69 figures, 38 tables. V, 253 pages

ISBN 3-540-09028-2
W. W. Henton, I. H. Iversen
Classical Conditioning and Operant

Conditioning
A Response Pattern Analysis
1978. 106 figures, 16 tables. XVII, 355 pages

ISBN 3-540-09326-7
1. P. Henry, P. M. Stephens
Stress, Health,
and the Social Environment
A Sociobiologic Approach to Medicine
1977. 171 figures. XII, 282 pages

(Topics in Environmental Physiology and Medicine)
ISBN 3-540-90293-7
K R Popper, 1. C. Eccles
The Self and Its Brain
Springer-Verlag
Berlin
Heidelberg
New York
1977.66 figures. XVI, 597 pages

ISBN 3-540-08307-3
Distribution rights for Great Britain: Global Book
Resources of Henley-on-Thames
E. Szekely
Functional Laws of Psychodynamics

1979.63 figures, 10 tables. VIII, 353 pages
ISBN 3-540-90371-2

Theories of Neurosis

Uploaded by

Copyright:

Available Formats

Theories of Neurosis

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

Theories of Neurosis

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What theories of neurosis are discussed in the book?

What theories of neurosis are discussed in the book?

Michael Gossop

Michael Gossop, B. A., Ph. D.

Softcover reprint of the hardcover 1st edition 1981

In view of the practical importance of neurotic disorders (with

advantages and disadvantages of different theories. The book

Part I. The Concept of Neurosis

The Neurotic Disorders

Neurosis, Normality and Psychosis

Normality and Abnormality

Part n. Theories of Neurosis

Conditioning Theories of Neurosis

Social and Cognitive Theories . . . . . . .

Dissatisfaction with Conditioning Theories

Psycho-analytic Theories of Neurosis

Trait Theories of Neurosis

The Physiology and Genetics of Neurosis

Physiology and Neurosis

In preparing this book I was helped in different ways by many

of my friends and colleagues. I am particularly indebted to

The Organic Model

1.2 The Organic Model

The Organic Model

The Psycho-dynamic Model and the Concept of Intrapsychic Conflict

1.3 The Psycho-dynamic Model and the Concept of Intrapsychic Conflict

Identity is closely linked to family, friends and

Concept of sanity as ideal functioning. Therapy

Mental disorders are caused by external events

Emphasis upon internal life of the individual,

Mental disturbances can be relieved by attempts

Mental disturbances as a result of intrapsychic

The Psycho-dynamic Model and the Concept of Intrapsychic Conflict

1.4 Moral Treatment

1.5 Learning Theory and Behaviour Therapy

Learning Theory and Behaviour Therapy

Fig. 1. The classical conditioning paradigm

Learning Theory and Behaviour Therapy

Learning Theory and Behaviour Therapy

2. The Neurotic Disorders

2.1 Psychiatric Diagnosis

The Neurotic Disorders

In psychiatry there is considerable disagreement about the requirements which should

2.2 The Neuroses

The Neurotic Disorders

2.2.4 Phobic Neurosis

2.3 Problems of Diagnosis

The Neurotic Disorders

The Neurotic Disorders

2.4 Reliability and Consistency

The Neurotic Disorders

2) Obsessive-Compulsive-Phobic. The items that showed strong positive loadings for

2.6 Social Context

Eight 'norma!' volunteers presented at 12 different psychiatric hospitals, complaining

The Neurotic Disorders

3. Neurosis, Normality and Psychosis

3.1 Normality and Abnormality

Neurosis, Nonnality and Psychosis