THE BEHAVIOR THERAPY SYSTEMS OP JOSEPH WOLPE

AND HANS EYSENCK

by
Shelby Rice

Submitted as an Honors Paper

in the
Department of Psychology

The University of North Carolina

at Greensboro
1967
Approved by

Director

Examining Co...,.,ittce
 TABLE OP CONTENTS

Page
INTRODUCTION 1

BEHAVIOR THERAPY ROOTS IN HULLIAN THEORY 13

JOSEPH WOLPE'S SYSTEM 16

HANS EYSENCK'S SYSTEM 30

SUMMARY 41

BIBLIOGRAPHY 43
 INTRODUCTION

In 1917 John B. VJatson commented upon an experimentally-

conditioned fear response in a manner which is intriguing, prophetic,

and relevant for the discussion to follow.

The Freudians, twenty years from now, unless their

hypotheses change, when they come to analyze Albert's
fear of a seal-skin coat—assuming that he comes to
analysis at that age—will probably tease from him the
recital of a dream which will show that Albert at three
years of age attempted to play with the pubic hair of
his mother and was soolded violently for it. (We are
by no means denying that this might, in some other case
condition it.) If the analyst has sufficiently
prepared Albert to accept such a dream when found as
an explanation of his avoiding tendencies, and if the
analyst has the authority and personality to put it
over, Albert may be fully convinced that the dream was
a true revealer of the factors which brought about the
fear (Watson and Raynor, 1920, p. 14).

The attitude of spurious disregard for the psychoanalytic

approach to psychotherapy which is implicit throughout the

preceeding paragraph seems typical, not only of Watson, but of

psychologists in general. The science of psychology holds as its

major concern the study of the general behavior of animals and of

the human species—behavior which it believes to be as casually

determined as is any other natural phenomenon (Morgan and King,

1966). Given this framework, it is not surprising that

psychologists regard askance a theory derived primarily from

natural (clinical) observation which espouses that the behavioral

manifestations of "mental illness" are surface "symptoms" which

are the visible results of "unconscious causes". This is not to say

that the psychoanalytic approach does not postulate cause and effect

relationship by which behavior is determined, but that it has done

so in ill-defined terms—terms which do not lend themselves to the

experimental manipulation of specified variables which make such

statements of causal relationships acceptable.

Somewhat more surprising, however, and certainly more distressing

is the fact that despite the criticisms of this method, few alterna-

tive methods have been presented which prove to be more effective

(Eysenck, 1952). Goldstein and Dean (1966) reported that millions

of words have been written about psychotherapy in the past fifty

years, and many conflicting explanations and justifications have

been advanced concerning the process. The actual practice of

psychotherapy, however, has remained remarkably constant with few

exceptions. Most researchers have tended to restrict their

attention to the classical two-person verbal interaction. Notwith-

standing the early work of Pavlov and Thorndike and the more recent

studies of experimental neurosis, of fear and avoidance conditioning,

of the conditioning of autonomic reactions, of verbal conditioning,

and of individual differences in conditionability (Renya, 1966),

the actual incorporation of such findings into clinical practice has

been reluctant at best.

Such reluctance seems to be based primarily upon an erroneous

interpretation of the relationship between pure and applied science.

This interpretation, which prevents the expansion and utilization

of a science to its fullest potential, states that application, if

it is to have any verifiable basis, must wait until there is a pure

science ready to be applied. The corollary of this interpretation

is that once the pure science is in order, applications of its

:
principles will be both simple and immediate. Hilgard (1956)

refuted this misconception on grounds which defended both pure and

applied research. He argued that laboratory derived principles are

of value primarily because principles discovered in a well-controlled

situation can be validated in less well-controlled ones. Research

on practical problems, however, not only adjusts theory to practice,

but also contributes to the substantiation, refutation, or

extention of theoretical knowledge.

It is in relation to the problems stated above that this paper

will be concerned. The major thesis of the work will be that

within the discipline of psychology itself lies the responsibility

for providing the necessary alternative framework within which the

problems of disordered behavior might be considered. Attention

will focus upon the group known collectively as behavior therapists.

Their work will be regarded as a successful demonstration of the

manner in which new solutions to old problems may be sought. Not

only have the behavior therapists departed from the traditional

methods of psychotherapy; they also exhibit a willingness to seek

results from within an incomplete theoretical structure which they

have thus helped to substantiate and to extend. In order to

elaborate upon this contention regarding the behavior therapists,

an attempt will be made to present both the general rationale of this

approach and the more specific details of two related systems:

Joseph Wolpe's and Hans J. Eysenck's.

Despite theoretical and methodological differences among the

behavior therapists, all agree that general psychological principles,

particularly those derived from learning theorists and from the

experimental laboratories, can be efficaciously applied in the

clinical setting (Rachman, 1963). Their concern is with the

treatment of overt behavior rather than postulated underlying

psychodynamic processes (Matarazzo, 1965). This deoision concerning

what is to be treated follows logioally from the theoretical frame-

work which considers all patterns of behavior, whether adaptive or

maladaptive, as products of the learning process. The manner in

which these concepts of learning theory can be used to develop a

program for behavior change is expressed well in general terms by

Ullman and Krasner (1965). They have presented three questions

which any behavior therapist is likely to ask himself when confronted

with a problem. First, what behavior is maladaptivej that is, what

behavior should be increased or decreased. Second, what aspects of

the patient's environment currently support the patient's behavior,

either to maintain the undesirable behavior or to reduce the

likelihood of his performing a more adaptive response. Third, what

changes in the environment, usually through a consideration of the

reinforcing stimuli, can be manipulated to alter the subject's

behavior. Suoh a procedure indicates an a priori rather than a

post hoc usage of learning concepts. It is this point which

distinguishes the behavior therapists' approach from other therapies

employing learning concepts (Murray, 1963).

 The theoretical assumptions which underlie such an approach are

delineated by Eysenck (1963) in his summary of the main tenants of

behavior therapy. These tenants will be listed here in abbreviated

form to permit more accurate discussion of the most distinct

features:
(1) Neurotic behavior consists of maladaptive conditioned

responses of the autonomic system and of the skeletal responses made

to reduce the conditioned (sympathetic) reactions.

(2) While the term "symptom" may be retained to describe

neurotic behavior, there is no implication that such behavior is

symptomatic of anything.
(3) It follows that there is no underlying complex or other

"dynamic" cause which is responsible for the maladaptive behavior;

all that is dealt with in treatment ic the conditioned maladaptive

behavior.
(4) Treatment consists of deconditioning the maladaptive

behavior by such techniques as reciprocal inhibition, extinction,

conditioned inhibition, and of the conditioning of adaptive behavior

along orthodox lines.

(5) Treatment is a-historical and does not involve "uncovering"

of past events.
(6) Conditioning and deconditioning will usually proceed through

behavioral channels, but there is no reason why verbal methods

should not also be used. Words are also conditioned stimuli which

have an ascertainable position on the stimulus and response

generalization gradient of the patient.

 Despite Eysenck's contention that these are tenants which are

basic to all behavior therapy, careful reading would seem to indicate

that these postulates arose from a particular theoretical formulation.

Their inclusion at this point, however, will perhaps serve not only

to illustrate the manner in which the behavioral approach differs

from the traditional approach to psychotherapy, but also to point

out some of the issues of controversy within the field of behavior

therapy. To attain this end, each tenant will be considered in turn.

Consideration of the first tenant indicates several important

characteristics of behavior therapy. First, these tenants were

specifically formulated to concern neurotic behavior. It is

significant that it is this group of disorders which has received

the most attention from the behavior therapists (Wolpe, 1958;

Eysenck, 1960c; Rachman, 1963). Indeed, Grossberg (1964) in a

rather critical review of the procedures of this approach, has

conceded that they are most successful when applied to neurotic

disorders with specific behavioral manifestations. Although work

with psychotic disorders has been attempted, such work has

involved primarily the application of the operant conditioning

techniques of the Skinnerian school (Eysenck, 1964a).

A second important aspect of the first tenant is that the

postulation of a two-factor theory of learning is very similar to

that of Skinner's (Hilgard, 1956). This position is implicit in

Bysenok's statement distinguishing between autonomic and skeletal

responses. On a more general level, however, this indicates

one of the ma.ior issues of controversy among the behavior therapists.

The dispute as to whether there are one, two, or many ways in which

learning occurs is an issue which has been adopted by virtue of the

adoption of the framework of the learning theorists (liilgard, 1956).

Murray (1963) has elaborated upon this point in an attempt to

distinguish between two approaches by which learning theory may

be applied to psychotherapy. The first of these is the biotropic

!
approach, whose exponents include Eysenck, Salter, Wolpe, and Shaw.

Hurry characterized the biotropic as •mphasizing classical

conditioning, primary motives, manipulative techniques, and an

impersonal attitude. Dollard and Miller, Kowrer, Shoben, and Murray,

however, are said to represent the sociotropic approach. This

approach is said to rely upon operant conditioning, social learning,

permissive techniques, the therapeutic relationship, and psycho-

dynamics. No attempt will be made to pursue Murray's argument

further. It should be noted that the sociotropic approach which

Hurray espouses would not be considered an example of behavior

therapy.
The relevance of his article lies primarily in the fact that

the distinction drawn between the two approaches illustrates that

divergence in methodology and in interpretation of fact can arise

from minor differences in the utilization of learning theory. The

sociotropic approach employs learning concepts to explain its

hypotheses} the biotropic approach uses learning concepts to

formulate its hypotheses. In another sense, the article is

relevant in that, although Murray states that the two approaches

represent different applications of learning theory, he never

explicitly states which learning theory he is using as a base.

The distinction he has drawn between conditioning methods employed

does not serve to specify the theoretical ap-roach. It seems pro-

bable that both these factors are sources cf ambiguit-- and contro-

versy (V.'olpe, 1964a; Mowrer, 1964). Eysenck (1960a) stated that the

theoretical points about which most of the arguments rage are of

academic rather than practical importance. It would seem, however,

that more explicit statement of the points in dispute is needed.

The second through the fifth tenants which Eysenck has

proposed are perhaps even more crucial to the behavior therapists'

position than is the first. They imply a more radical deviation

from the model of traditional psychotherapy. Ullman and Krasner

(1965) have explained the theoretical differences concerning the

formation and treatment of "symptoms" as being that which results

from an interpretation of behavior from within a psychological

model as opposed to a medical model. Most traditional psycho-

theranies consider behavior from within a medical model. Behavior

considered within this framework is regarded as being peculiar,

abnormal, or diseased because of some underlying cause. Symptoms,

therefore, are considered to be evidence of suoh causes, as evidence

of repression, and are not to be regarded as the disorder. It

follows from this conception that cures can be achieved only by

removing these underlying causes through an interpretation of the

symptoms, dreams, and acts as the products of the various

defense mechanisms. Any treatment which does not accomplish this

task may seem effective for a short time, but eventually the
removed symptom will be replaced by another.
The psychological approach, however, as is clearly shown by
Eysenok's postulates, explicitly rejects this dualistic, central-
conflict, peripheral-symptom model (Grossberg, 1964). One of the
best defenses of the behavioral position regarding symptoms is that
presented by Yates (195$)1 a defense which also serves to specify
more clearly this position. Yates objects to the distinction drawn
between a "fundamental" underlying anxiety and a "surface" symptom.
His first argument is that the term "symptom" has no specific
meaning in psychological terminology. To define a symptom as a
particular set of learned responses wouid be theoretically sound.
In addition, such a concept would be more amenable to experimental
validation than would the dynamic concept.
His second objection to the dynamic approach was that no
distinction is drawn between neurosis and neuroticism—the innate
predisposition to develop a neurosis under certain specifiable
conditions. Yates has made no attempt to prove that this concept
of neuroticism is anything more than a plausible alternative to the
dynamic concept. It would seem, however, that such a concept could
be operationally defined. For example, Eysenck (l96l) has
demonstrated individual differences in conditionability and autonomic
lability. One of the oausal factors of such differences has been
shown to be the genetic inheritance of the organism. In like
manner, Wolpe (1958) included the phrase "learning in a physio-
logically normal organism" in his definition of neurotic behavior.
The acceptance of a concept such as neuroticism would define a
 I
10

neurosis as a symptom and as nothing else. Psychological treatment

could be, therefore, nothing more than symptomatic, since treatment

of the predisposition must ultimately be by genetic or by chemical

means.
Yates's final point is one that has received much support

throughout psychological literature (Wolpe, 1958» Ullman and Krasner,

1965). He points out that the belief in symptom substitution is

based primarily on clinical observation. Such a phenomenon is seldom

found in instances in which a single subject has been subjected

to intensive investigation and symptomatic treatment, nor in those

cases where a large-scale follow-up has been conducted on groups of

subjects. A general statement by Ullman and Krasner (1965) appears

to be an appropriate summary of the position with regard to this

problem: all the evidence would seem to indicate that symptom

substitution is rarely observed} when and if such phenomena occur,

there are a number of more parsimonious explanations.

Eysenck's sixth tenant is primarily a statement of the basic

principles which underly the methodology and the techniques of the

behavior therapists. Ullman and Krasner (1965) have stated that

although there are many different techniques, there are few concepts

or principles which have been widely accepted and incorporated into

clinical practice. This tenant also illustrates the manner in which

the traditional method of verbal interaction may be considered within

a different framework.
The effects of such a theoretical position upon the methodology

employed in the clinical situation have been alluded to throughout

 11
this section. The distinction between the behavior therapist's

approach and the traditional therapist's approach is most apparent

in regard to the specific variables which each considers. The

focus of treatment for the traditional therapist is a particular

disease or some historical circumstance. The focus of treatment

for the behavior therapist is behavior. The area of assessment

within the traditional framework concerns the description and

measurement of internal dynamics. Within the behavior therapist's

framework, the total social situation is considered. The tradi-

tional therapist seeks to change the patient's behavior by altering

the internal "causes" of such behavior. Treatment for the behavior

therapist consists of the systematic manipulation of the environ-

mental contingencies which are judged to support the behavior

in order to alter the responses which have been judged to be mal-

adaptive. The goal of treatment, the change ii behavior requested ■

by the patient or b;- the society in which he lives, as well as the

procedures to be used to achieve this goal, are decided upon before

treatment begins. Behavioristic criteria, such as frequency of res-

ponse , are utilized to measure progress toward that goal. The

behavior therapist can thus evaluate his success and, if necessary,

change his procedure (Ullmann and Kr-.sner, 1965).

These are, perhaps, the only statements which can be made

concerning beh ivior therapy in general. Ullmann and Krasner (1965)

have stated that at present only the broadest, most thoroughly

established concepts are consistently utilized within a clinioal

setting. In order to examine the manner in which the postulations

1
 12
made by learning theorists and laboratory experimenters are

demonstrated to hive snecific application, it is necessary to

consider the particular systems which have developed.

 BEHAVIOR THERAPY ROOTS IN HULLIAN THEORY

The psychotherapist who would write his theory of neurosis and

of therapy in learning theory terms must first choose his learning

theory from the many which are available (Dittman, 1966). The

theory which serves as a framework for the systems of both Wolpe

and Eysenck is that of Clark L. Hull. Hull's theory has attempted

to present a bold and comprehensive theory of behavior and at the

same time to lend itself to empirical quantification (Hilgard, 1956)

His contribution to both the pure and applied areas of psychology

seems to derive primarily from this fact. Pew other theories have

been worked out in sufficient enough detail to generate the precise

predictions which can be derived from Hull's principles; nor has

any other system given rise to so many experimental attempts to

test its predictions and to modify its hypotheses. No attempt will

be made to assess the strengths and weaknesses of Hull's system

per se. Only to the extent that these principles have supplied a

valuable tool for the systems of psychotherapy to follow will they

be considered. The relevant concepts will be discussed briefly at

this point and more extensively throughout the paper.

Hull's central concept is that of habit—the modification of

the central nervous system through experience. This process

mediates learning and is not directly measurable but simply

inferred from the performance of the organism (Hilgard, 1956).

Eysenck (1960b) has stated that this dissociation of performance

 14
from the concept of habit is one of the most important contributions

of the theory. Performance—the behavior of an organism—is the

product of habit strength and of drive which results from

physiological needs. The formation of complex patterns of behavior,

therefore, whether adaptive or maladaptive, can be explained in

terms of the formation of habits. Within this framework, the

"symptoms" manifested by one considered to be mentally ill can be

conceived of as the products of the learning process obeying the

laws which Hull has formulated. It follows, that the point of

primary interest for the behavior therapist would be the manner in

which habits are formed and eliminated. Such knowledge could permit

more efficient and effective manipulation of behavior within the

therapeutic setting. Hull has postulated that one principle, rein-

forcement, is sufficient to explain learning or the formation of

habits. This reinforcement Hull has espoused to be either the

result of drive-reduction, as in the satisfaction of a physio-

logical need or of drive-stimulus reduction, as in the satisfaction

of a craving rather than a need (Hilgard, 1956). No distinction

is drawn, in this system, between the two major methods by which

a functional connection between an environmental stimulus and a

subject response is acquired. These methods are Pavlovian (respondent)

conditioning in which the stimulus elicits the response, and operant

conditioning in which the subject must emit the response to the sit-

uation prior to the environmental event that becomes associated with

and alters the frequency of occurrence of the response in the

future. Both are interpreted by Hull as illustrations of learning

 15
under the control of reinforcement.

Thus, the basic condition necessary for the formation of a

habit is contiguous association under the control of reinforcement.

The strength of the habit, however, is determined primarily by the

number of reinforced trials (Hilgard, 1956). Ullman and Krasner

(1965) feel that this theory with its concept of drive and of

mediation through habit serves particularly well as an explanatory

tool in the sense that it is capable of considering both the

internal and the external stimulus environments.

Further concepts of the system which are applicable to the

majority of experimental studies are those of reactive inhibition

and conditioned inhibition (Eysenck, 1960b). According to Hull's

theory of learning, all behavior produces some degree of inhibition

or fatigue in the mediating structures. This fatigue, it is

postulated, will act as a negative drive which dissipates with rest.

Such dissipation will, in turn, reinforce the state of rest, and

thus a negative habit, the habit of not responding, is formed

(Hilgard, 1956). These constructs present a rationale for the

removal of symptomatic behavior.

The manner in which these principles may be specifically

applied will be considered first in relation to the psycotherapeutic

system of Joseph Wolpe.

 JOSEPH WOLPE'S SYSTEM

Opinion concerning Wolpe's theory of psychotherapy by reciprocal

inhibition is diverse. Pranks (1961) has designated Wolpe's system

as being the "least unsatisfactory" theory based upon Pavlovian

conditioning techniques that is currently available. Eysenck (1964a),

in contrast, has stated that the system presents a method of treatment

which appears to be extremely successful in its application to many

diverse types of neurotic disorders. In addition to this, Wolpe

presents a theoretical account, closely linked with modern learning

theory which supplies this method with a rationale. Wolpe, himself,

has characterized his system as being:

...a serious alternative to the repression theory,

one that is based on the growing body of knowledge of the
processes by which change is wrought in the behavior of
organisms—modern learning theory—the fruit of the
efforts of Pavlov, Throndike, Watson, Tolman, Hull,
Skinner, and their followers (1958, p. ix).

Wolpe's alternative to the repression theory is made explicit

in his definition of neurotic behavior as any "persistent habit

of unadaptive behavior acquired by learning in a physiologically

normal organism"(l958). Anxiety is usually the central constituent

of neurotic behavior and is invariably present in the causal

situations. Within this definition, Wolpe includes the various types

of behavior usually labeled anxiety state, phobia, depression,

hysteria, neurasthenia, and obsessional state. Psychoses such as

 17
schizophrenia, however, are excluded on the grounds that they stem

from an abnormal organic state.

VIolpe's theoretical formulations concern primarily the manner

in which neurotic behavior is learned and can thus be unlearned

with the aid of therapeutic intervention. The need of such therapeutic

intervention to affect the removal of an unadaptive response calls

attention to one of the crucial aspects of Wolpe's theory. He

postulates that in any given situation, the behavior of an organism

may be judged to be either adaptive or unadaptive. Although he never

explicitly states his learning theory base (Ford and Urban, 1965)1

Wolpe's use of the drive-reduction model of Hull seems quite apparent

in his definition of these behaviors. Adaptive behavior is conceived

,0 be that which takes the form of progress toward the satisfaction

of a need or that which permits the avoidance of possible damage or

deprivation to the organism. The consequence of unadaptive behavior

would be the needless expenditure of energy or the occurrence of

damage or deprivation. A consistently unadaptive response, if only

in the sense that it produces fatigue, will tend to be progressively

weakened in habit strength and eventually extinguished. It is the

p_ersistence of an unadaptive habit which Woipe feels to be a

feature of neurosis. This extraordinary persistence Wolpe feels can

be accounted for as a matter of learning under special conditions.

Wolpe's account of the etiology of neurosis is, briefly, as

follows: General emotional reactivity differs from individual to

individual either as a result of maturationally established

physiological differences or as the result of an increase in

 18

sensitivity through the learning processes. These differences

can facilitate the acquisition of neurotic behavior in a subject,

if this subject is exposed either directly to stimuli evoking anxiety

or to ambivalent stimulation such that anxiety of high intensity is

evoked. The response of anxiety is defined as the autonomic response

pattern, or patterns which are characteristically a part of the

organism's response to noxious stimulation. When intense anxiety is

evoked in an individual, any event occurring at the same time may

become associated with the anxiety. These stimuli may be specific

or of a more pervasive situational nature. The variety of responses

which may be acquired as concommitants of the anxiety constitute

part of the symptom complex as the learned neurotic behavior. Wolpe

has postulated that the effectiveness of the learning of such

neurotic behavior is influenced by a number of factors. Included

among these are the intensity of the anxiety during each experience,

the number of learning experiences, the degree of similarity among

the experiences, and the responses which are open to the individual

at the time of exposure.

Once this primary learning of the neurotic response has been

made, Wolpe has hypothesized that two courses of action are open to

the individual. One course involves the passive acceptance of

suffering. Other aspects of performance, however, such as motor

coordination, concentration span, and sexual performance may be

severely impaired. The other manner in which relief from anxiety

may be sought is through recourse in activity which helps to

diminish anxiety. Thus, activity such as the physical avoidance of
 19
the stimuli conditioned to anxiety, displacement of attention, drug

taking, or anxiety relieving obsessions may be observed.

Wolpe has not specified precisely the characteristics of these

neurotic behaviors once they have been learned. Instead, he has

considered his cases in terms of the major symptomatology present.

The one observable characteristic common to all, the resistence to

extinction in ordinary environmental situations, is reiterated at

this point to permit a more precise explanation in terms of Hull's

learning theory. The first point to be made is that anxiety responses

apparently generate little reactive inhibition and thus there is

little opportunity for a conditioned inhibition of the response to

develop. Hull's principles indicate that the habit of responding to

a situation with anxiety can be effectively eliminated only by the

opposite habit of not responding. The second point is that because

many symptoms are learned as a means of avoiding the circumstances

which produced the anxiety, the individual tends to avoid the very

situations which would enable him to unlearn the symptom (Ford and

Urban, 1965). Ford and Urban (1965) succinctly stated VJolpe's

conception of neurotic behavior as "a collection of behavior learned

concommitantly with unreasonable fears as a means of reducing

unreasonable fears."
The most significant aspect of Wolpe's theoretical formulation,

however, lies in its role as a framework for the innovation of

effective therapeutic techniques. Basic to all these techniques is

the concept of reciprocal inhibition which is at once the most

distinctive and the most controversial aspect of VJolpe's approach.

 20
Colby (1964) has reported that the use of this physiological concept

as the principle by which maladaptive responses are unlearned,

clearly distinguishes Wolpe from other behavior theorists who explain

this process by use of the principle of extinction or the principle

of operant conditioning. Wolpe, however, considers the principle to

be of even wider applicability than does Colby. Wolpe (i960)

contends that it is the principle of reciprocal inhibition which

is fundamental to all successful treatment of neurotic disorders with-

out regard for the theoretical orientation of the therapist. The

specific rationale of such a statement and the theoretical arguments

which it has provoked will be considered later; the principle itself,

as formulated by Wolpe, is as follows:

If a response antagonistic to anxiety can be made

to occur in the presence of anxiety-evokin.'-; stimuli so
that it is accompanied by the complete or partial
suppression of the anxiety responses, the bond between
these stimuli and the anxiety responses will be
•..•eakened (1958, p. 71 )•
Wolpe regards experimental extinction, the other known process by

which habits are removed, as being an ineffective principle for the

removal of anxiety responses. Wolpe's methodology and therapeutic

techniques, therefore, involve primarily the selection of appropriate

responses which can effectively oppose the anxiety .-esponses of the

neurotic subject. Thus, responses were sought which largely implicated

the parasyzpathetic division of the autonomic nervous system, since

these would seem most likely to be incompatible with the predominately

sympathetic responses of anxiety (Wolpe, 1958).

Following is a list of those responses chosen and the manner

in which they are presumed to achieve therapeutic results. The

 21

choice of the appropriate technique requires that the behavior

therapist examine the patient's problem in detail to determine to

what stimuli the patient reacts with anxiety. Once the source of

anxiety has been determined, the effective response to inhibit the

anxiety can be selected.

(l) Assertive responses: This technique is used primarily to

overcome unadaptive anxiety aroused in the patient by other people.

Its methodology requires that the therapist assume a directive role.

He must point out the patient's fears, emphasizing how his fearful

patterns of behavior have incapacitated him and placed him at the

mercy of others. Occasionally, Wolpe feels, it is necessary to provide

even more extensive coaching for the patient and a role-playing

situation similar to psychodrama is arranged in the consulting room

(Wolpe, I960). Walton (1961) reported that a severe, chronic, and

apparently complex psychiatric disability proved amenable to only one

session conducted in this manner. The behavioral symptom of the

patient was the outbreak of violent agression while the patient was

apparently in deep sleep. Although the disorder had persisted for

over two years, a suggestion to the patient that he be less inhibited

in his expression of hostile feelings during periods of wakefulness

was apparently successful. A follow-up conducted two years later

revealed that the violent behavior of sleep had not returned, nor was

there evidence of symptom substitution.

(2) Sexual Responses; When anxiety responses have been

conditioned to-various aspects of the sexual situation, this technique

is employed. It is of particular value when the sexual inhibition

 22

is partial and varies according to definable properties of the rele-

vant situation. The patient is simply instructed that he must never

perform sexually unless he has an unmistakably positive desire to do

so. His full participation in sexual activity is to be accomplished

through a series of gradual graded steps. The degree of participation

at any particular time is determined by the wishes of the patient.

Wolpe and Stevenson (i960) presented three case histories of

relatively severe inhibition in which both the assertive technique

and the technique of graded sexual response proved effective.

Confidence was expressed that these methods were effective refutation

of the psychoanalytic approach which considers the treatment of

sexual deviations to be difficult and untenable.

(3) Relaxation response: Wolpe considers this technique to be

of extreme importance. Prom this approach is derived the technique

which has proved to be of extremely wide applicability; that of

systematic desensitization based upon relaxation. This technique

follows directly from the work of Jacobson who found that intense

muscle relaxation was antagonistic to anxiety. Treatment consists

of training the patient to relax according to Jacobson's technique

of "progressive relaxation". During this training, the therapist

constructs a list of stimuli to which the patient reacts with

unadaptive anxiety. The patient then ranks these stimuli according

to the amount of anxiety they arouse. The most disturbing items are

placed at the top of the list and the least disturbing are placed at

the bottom. This list becomes the "anxiety hierarchy" to be used

during treatment. After the patient has been hypnotized and

 23
instructed to relax as deeply as possible, he is told to imagine

the weakest item in the hierarchy. If relaxation is unimpaired,

the stimulus of the next greater intensity is presented. This

procedure is continued from session to session until the reciprocal

inhibition of the stimulus of maximum intensity is effected by

the relaxation response. Wolpe feels that this reciprocal inhibition

of the anxiety response will transfer easily from the clinical

setting. In addition, he has voiced confidence that this technique

is applicable to almost every source of neurotic anxiety which does

not involve anxiety arising from inadequacies in the handling of

interpersonal relationships (1^63)• The technique of desensitization

can be considered within the Hullian framework as the systematic

deconditioning of anxiety responses along a stimulus dimension of

generalization (Wolpe, 1962). Since each stimulus in the continuum

shares features with "adjacent" stimuli, the elimination of anxiety

responses to a stimulus remote from the central stimulus involves

the elimination of whatever fraction of the anxiety evoked by the

related stimuli is attributable to the shared features.

The wide applicability of the technique is evident from even

a superficial survey of the literature. It has been reported

successful in the treatment of exhibitionism (Bond and Hutchison,

1965), of bronchial asthma (Cooper, 1964)f of sexual disorders such

as impotence, voyeurism, and transvestism (Rachman, 1961), and in the

treatment of phobias (Lang and Lazovik, 1967). Wolpe (1964b)

presented a summary of the results of studies utilizing this

technique. These studies concerned sixty-eight phobias and allied

 24
neurotic anxiety habits. He reported that in a mean of 11.2 sessions,

45 of the neurotic habits were eliminated and 17 more were very

markedly improved. Follow-up studies of 20 of the 45 successfully

treated cases at intervals ranging from six months to four years

revealed neither relapse nor the emergence of new symptoms. Colby

(1964) has designated such statistics as being crude and the follow-up

studies questionable. However, they do illustrate the necessity of

the assessment of therapeutic procedures.

Wolpe (1958) has designated the above techniques as being his

"bread and butter responses" by virtue of their wide applicability

and generally favorable results. In addition, he has reported

research on responses that are not themselves physiologically

antagonistic to anxiety. Among these he lists motor responses,

feeding responses, respiratory responses, and interview induced

emotional responses. Of these, only the last will be considered more

fully. In conjunction with the technique of interview induced

emotional response, Wolpe postulated that reciprocal inhibition was

the basis of all success is therapy. His rationale for such a

statement derives from a consideration of the fact that the rate of

recovery is much the same regardless of the type of therapy employed.

The only feature common to all therapies, however, seems to be the

private interview in which the patient talks about his difficulties

-0 a person he believes to have the knowledge, skill, and desire to

help him. Wolpe has hypothesized that if the emotional response

evoked in such a situation is antagonistic to anxiety and of

sufficient strength, it will reciprocally inhibit the anxiety which

 25
occurs as a result of the content of the patient's discourse.

The centrality of the principle of reciprocal inhibition for

the treatment of neurotic disorders within Wolpe's system is more

than obvious. It seems almost superfluous to add that he explains

spontaneous improvement of neuroses with this principle. For this

reason, it is not surprising to discover criticism of the principle,

nor to find recommendations that it be submitted to more rigorous

experimental tests (Eysenck, 1960a).

One of the most thorough examinations of the principle of

reciprocal inhibition is that of Lomont (1965). His criticisms of

the concept will be consedered in some detail as a means of illus-

trating the manner in which clinical techniques may be subjected to

examination in the experimental laboratory. Lomont did not question

the efficacy of the technique. He challenged, hovjever, Wolpe's

statement that the technique of reciprocal inhibition can be reliably

differentiated from the classical extinction of a response. He has

suggested, instead, that Wolpe's technique differs from the classical

extinction of a response only in that it proposes to inhibit anxiety

by the presentation of a stimulus which is presumed to be anta-

gonistic to anxiety.
Wolpe's difinition of neurotic anxiety incorporates this

distinction between the two processes when he characterizes neurotic

behavior patterns as those which are resistant to extinction under

normal circumstances. Data from animal research provide the

experimental basis for his contention that conditioned avoidance

responses are resistant to exti iction, particularly if the

 ■

26

subject is free to make avoidance responses to the conditioned stimulus.

But, as Lomont has pointed out, in the process of reciprocal inhibition,

the patient cannot escape from the conditioned stimulus at will. For

example, in the process of desensitization, the patient is specifically

instructed to imagine scenes from the anxiety hierarchy which are

known to be related to anxiety.

To support his position that such a procedure may not differ

significantly from ordinary extinction, Lomont presented a summary

of the relevant literature which support the following contentions:

(l) evidence exists that the extinction of an avoidance response is

hastened if the subject is forced to remain in conduction with the

conditioned stimulus which prompted the fear response; (2) the

lengthening of each exposure to a fear conditioned stimulus will

reduce the number of trails required to extinguish the avoidance and

fear responses. Lomont further stated -hat even those studies which

ostensibly compare the two techniques are negligent in that they

do not control for excessive movement which could permit more

exposure to the conditioned stimulus in the group being treated

by the method of reciprocal inhibition. On a priori grounds, the

only technique Lomont found which might be exclusively a property

of the reciprocal inhibition method is identified as the "progressive

principle". This method ie basically that of the graded presentation

of the anxiety hierarohy. Lomont has reported that experimentation

with rats yields no evidence that such a process is operative in

the process of extinction. If further experimentation were to

indicate that this principle is an important component of the

 27
reciprocal inhibition technique, it would be a unique feature of

this approach and would reliably differentiate it from the procedure of

extinction. Lomont has further recommended that in order to substantiate

the theoretical explanation of the principle of reciprocal inhibition,

experimentation should be conducted which specifically compares

reciprocal inhibition with extinction. The former should differ

from the latter only in its attempt to reinforce a new response that

is incompatible with anxiety.

Such a comparison was attempted by Gale, Strumfels, and Gale

(19*06) • A group of rats were conditioned to fear a tone through the

pairing of a 1500 cycles per second tone with an intense shock.

These animals were subsequently divided into three matched groups

which were subjected to different deconditioning methods. Group I

was treated by the method of classical extinction. Group II differed

from Group I only in that food was presented to the animals in the

deconditioning trials. Great care was taken to control for excess

movement which might permit more exposure to the conditioned fear

cues of the apparatus. Group III comprised a control group. The

animals under this condition were kept in their home cages to control

for the natural loss of the fear response. The method of extinction

for Groups I and II was the presentation of an "anxiety hierarchy"

of tones which ranged from 300 cycles per second up to the original

150C cycles per second through increments of 100 cycles per second.

The results of thiB study provided support for Wolpe's hypothesis.

Group II, whose anxiety was hypothesized to be reciprocally

inhibited by the introduction of the feeding response, demonstrated

 28

loss of the conditioned emotional response much more quickly than

Groups I and III. In addition, if the process of extinction is
regarded as the learning of new responses to old stimuli, the
reciprocal inhibition technique is perhaps superior in yet another
respect. When this method is employed, the patient does not have to
search for an effective response. Instead, direction is given by
the therapist.
Application of the techniques based upon the principle by
reciprocal inhibition affords direction to the patient in a variety
of ways; among these are hypnosis, suggestion, role playing and
psychodrama. Such procedures are, no doubt, the basis for the
charges by critics (Hurray, 1963) that behavior therapy is unduly
directive and manipulative. Such criticism, however, seems to be
of a philosophical rather than of a methodological nature. Lomont
(1964) has effectively defended the position of the behavior
therapist through a consideration of the implications of determinism.
At a more practical level, Grossberg (1964) has pointed out that
manipulation occurs in all therapies. Therefore, criticism must be
directed against the kind of manipulation, rather than against
manipulation p_er se. Support for the kind of "manipulation" found
in the general approach to behavior therapy is given by Goldstien,
Keller, and Sechrest (1966). Empirical evidence supports the
position that psychotherapy will be more effective if emphasis in
therapy is placed upon the emitting of responses considered desirable
in other circumstances. The practice and differentiation of
appropriate responses should, in terms of learning theory, increase
 29

the tendency of the individual to give that response in the presence

of the appropriate stimuli. In general, Wolpe regards the acquisition

of desired behavior as a process which requires constant coaching

and feedback of results; a process which by-passes rational thinking

and deals directly with nonverbal processes (Metzner, 1963). Ford

and Urban (1965) have stated that this conceptualization suggests

that the first step in the recovery of neurotic disorders is not a

changed way of thinking, but a changed way of acting.

 HANS EYSENCK'S SYSTEM

The behavior therapy system of Hans Eysenck is perhaps unique.

Although much of his work has been done within the Hullian framework,

his emphasis upon the necessity of an adequate theoretical account

of neurotic disorders has led him to attempt to expand and to modify

this theory by the postulation of a two-factor theory of learning

(Eysenck, 1963; Matarazzo, 1965). His approach to the problem of

behavior disorders will first be considered from within a Hullian

framework.

Eysenck (1960b) distinguishes two ways in which Hull's theory

can be applied to the complex phenomena of social learning and

behavior modification. The first of the two approaches conceives of

the symptoms of behavior disorders as product:; of the general laws

of learning which Hull has presented in detail. Therapy, therefore,

would entail merely the removal of these symptoms (habits) through

a method of treatment based upon these same laws of learning. It is

this approach which characterizes VJolpe's work.

It is with the second mode of application, however, that Eysenck

is most concerned. This approach is primarily a typographical,

descriptive approach which considers individual differences in

behavior rather than general laws of behavior. The possibility of

an approach which considers individual differences in relating

diagnosis to therapy, was first suggested by Ivan Pavlov (Eysenck, 1960b).

 31
Pavlov hypothesized, after years of systematic observation of

patients, that hysteria is due to an exaggeratedly strong inhibition

of -he cerebral cortex; psychasthenia, in contrast, is considered

due to an exaggeratedly strong excitation.

The implications of an individual approach to the problem of

disordered behavior creates a system that differs greatly from

that of a general behavior theory. The basic premise of the

individual approach is that the factors which underlie and account

for the broad complexity of behavior, can be isolated and specified.

These factors are to be derived from large-scale studies of behav-

ior involving numerous subjects and multiple measures which permit 1

quantitative scoring. Given these surface indices, investigators

then apply the techniques of factor analysis to determine which

underlying factors determine or control variation in the surface

variables (Hall and Lindsey, 1957).

Eysenck (l96l) summarized the empirical studies which have

been made according to this method and offered the following as

major conclusions which have been reached:

(1) There are two main, independent factors in the psychia-

tric field, associated with the psychotic and neurotio disorders

respectively? psychotism and neuroticism.

(2) Both factors define continua which range all the way

from extreme disorder to normality; there are no breaks or qual-

itative differences which would enable the classification of

people into separate groups.

(3) Introversion-extraversion emerges as a third independent
 32

factor which interacts with neuroticism and possibly with

psychotism.

(4) Distribution of individual scores on these factors

reveals that no clusters corresponding to psychiatric disease

concepts such as hysteria or schizophrenia exi3t; all distribu-

tions are continuous and without the clustering predicted in

terms of such theory.

(5) Descriptively, factor scores give a much more detailed

and much more accurate picture of individual patients than does

psychiatric diagnosis.
(6) Individual patients tend to have scores on all factors,

not just one; therefore, the customary practice of allocating a

patient to just one diagnostic group is erroneous and misleading.

The major advantage of such a multi-dimensional approach,

is that the old "disease entity" concept of classification of

behavior can be discarded in favor of one which is operationally

defined by objective performance for each individual. Not only

does this classification deny that a specific "disease" is respon-

sible for a disorder, but also that no specific cure can be

applied. Rather, each diagnosis is regarded as a relative thing,

and may shift along one dimension or another in the course of

time, or as the consequence of specific experimental or therapeutic

manipulations. This scheme implies testable relationships between

diagnostic groups. Eysenck (1961) has stated that the method is

of particular value because errors may be detected and corrected

since the whole chain of argument is public and open to inspection.

 33
Eysenck (1961) is careful to point out that the correlations

derived from such a statistical treatment in no way implies

causality and can in no way be thus interpreted without additional

experimental proof. He states that "to know a persons position

on a given dimension is one thing; to know the reasons for his

being there and the methodology for changing his position is quite

another." This type of information must be derived from theories

of general behavior.
One variable which has been investigated extensively from the

standpoint of individual differences is the concept of condition-

ability. This concept is incorporated as one of the basic postu-

lates of the theory and techniques of all the systems of behavior

therapy. Despite its centrality, however, Pranks (1961) has

reported that it is not known whether it is meaningful to use the

term "conditionability". The existence of a general factor of

conditionability has yet to be demonstrated and the manner in which it

can be related to an individual organism is unspecified. To support

this statement, Pranks has pointed out that there is little agree-

ment concerning the influence and importance of even the most

fundamental parameters of the conditioning process. For example,

the relationship of the intensity of the unconditioned stimulus

to the conditioning process is not completely understood. Neither

the neurophysiological mechanisms by which a conditioned response

is established, nor its site or sites of action is known. Similarly,

there is little agreement concerning the term conditioned response,

or the aspects of behavior which should be included in this category.

 Despite these difficulties, however, the methodology related to

the conditioning of responses seems capable of contributing much

valuable information to the understanding of human behavior and

affords a promising tool in the attempt to establish behavior theory.

Pranks (l96l) proposed a theory of conditioning which was

based upon experimentally derived knowledge and which utilized the

concepts of Pavlov, Hull, and Bysenck. This conditioning theory is

to be related to the etiology, diagnosis, and treatment of certain

behavior disorders. The basic hypothesis which resulted from

Pranks approach is as follows:

Conditionability is related not to the degree of
neuroticism present, (Eysenck's dimension) but centrally
to the postulated excitation-inhibition balance (Pavlov)
and behaviorally to the introversion-extraversion
balance (EysenckN Oi the individual concerned (Pranks, 1961, p. 462).

It follows, that an introverted subject (cortical excitation),

whether neurotic or normal, should form conditioned responses readily,

and these responses once formed, should oe difficult to extinguish.

An extroverted subject (cortical inhibition) whether neurotic or

normal, should form conditioned responses poorly; these responses

once formed, should extinguish readily. Pranks listed in detail

the weaknesses of the theory but felt that its specific formu-

lation enables the predictions made from the theory to be explicitly

tested in a variety of situations and with a variety of conditioning

techniques.
Eysenck (1961) has demonstrated the manner in which such an

individual behavior theory can be utilized to predict the different

levels of personality organization which are usually explained with

I
 35
reference to the principles of general behavior theory. By causally

linking the personality dimension of introversion-extraversion with

its underlying central nervous system substratum, the excitation-

inhibition balance, specific predictions are made possible. Eysenck

postulated that persons whose central nervous systems are innately

prone to excitation will tend to develop introverted behavior

traits. In cases of abnormal functioning, such individuals should

demonstrate dysthymic neurotic symptoms such as anxiety. Conversely,

persons in whom inhibitory potentials predominate will tend toward

extraverted behavior patterns and hysteric-psychopathic symptoms.

The major advantages of such a system of classification are

threefold. First, the specific predictions made are capable of

disproof and self-correction. Second, the system demonstrates the

relationship between an individual theory of behavior and a general

theory of behavior. Third, in conjunction with behavior therapy,

such a system could serve as an important diagnostic tool. It is

hoped that eventually the suitability of any one patient for

behavior therapy, his likelihood of responding quickly, and the

circumstances under which such therapy is likely to be successful

may be predicted by making the appropriate laboratory tests before-

hand and by observing the individual's relative position on the

the introversion-extroversion continuum (Pranks, 1961).

Eysenck, however, has by no means confined his work to this

aspect of Hull's theory nor to Hull's theory in general. His

concern as a theoretician is primarily directed toward the establishing

of a general theory of neurotic behavior. He has stated (1963) that

 36
the acceptability of such a theory must ultimately depend upon the

ability of such a theory to present a "nomological network" within

which events of disordered behavior can be explained and understood.

Since it seems probable that it was the attempt to create this

"nomological network" which led Eysenck to reject the Hullian

framework as an inadequate theoretical basis, specific phenomena

and their relation to the theory will be examined.

The first of these, the phenomenon of transference, presented

no problem of interpretation within the Hullian framework (Eysenck,

1963). Eysenck has separated the fact of transference from the psycho-

analytic interpretation. He postulated that the therapist merely repre-

sents the conditioned stimulus to which the attitudes and emotions

appropriate to tho unknown cause of the cure is transferred.

VJhen Eysenck discusses symptom, extinction, relapse, and sponta-

neous remission in neurosis, however, his divergence from Wolpe and

from the framework of Hull becomes more apparent. His definition of

neurotic "symptoms" as unadaptive conditioned autonomic responses,

or the skeletal and muscular activities instrumental in moderating

these responses, implies this distinction. Eysenck distinguishes,

on theoretical grounds, between the formation of t-pes of neurotic

disorder and discusses the methods by which these two types may

be extinguished. Application of Wolpe's methods of recripocal

inhibition may be unilized in either of two ways, according to the

nature of the symptom.

(1) When the symptom is of a dysthymic character (anxieties,

phobias, depression, obsessive-compulsive reacito^.s, etc.) it is

 37
assumed that the disorder consists of conditioned sympathetic

reactions, and the treatment consists of reconditioning the stimulus

or stimuli to produce parasympathetic reactions. These reactions,

being antagonistic to the sympathetic ones, will weaken and

finally extinguish them.

(2) V/hen the symptom is of a socially disapproved type in

which the conditioned stimulus evokes parasympatheiio responses

(alcoholism, fetishism, homosexuality), or where there is an entire

absence of an appropriately conditioned response (enuresis, psycho-

pathic behavior), treatment (aversion therapy), consists of the

pairing of the stimulus in question with strong aversive stimuli

producing sympathetic reactions.

The major point of differentiation is that Eysenck distinguishes

between the procedures of classical and instrumental conditioning as

crucial elements in the genesis and treatment of both kinds of

neurotic disorders, whereas Wolpe, following Hull, does not.

Spontaneous remission was hypothesized by Eysenck (1963) to

be more probable in the case of disorders of the first type, in

which autonomic responses are classically conditioned to originally

neutral stimuli. It follows that on subsequent occasions the

presentation of the conditioned stimulus would not be followed by

reinforcement and in due course, extinction should take place.

To explain disorders of the first kind in which spontaneous remission

does not occur over time, Eysenck postulated that the phase of

classical conditioning is followed by a stage of instrumental

conditioning in whioh the patient withdraws from the conditioned

 38
stimulus upon encountering it. Such a withdrawal will lower

sympathetic arousal which acts as a reinforcement for the act of

avoidance. As the habit of avoiding the conditioned stimulus

builds up, the likelihood that the response will be naturally

extinguished lessens. Therapeutic methods, however, can be

utilized to effect the removal of such responses. Once the

symptom has been removed, whether by natural or by therapeutic

means, relapse should not occur in disorders. It is possible,

however, that new traumatic events may occur which produce a new

symptom and a new neurotic disorder.

Disorders of the second type present a much more complex

problem. According to Eysenck's classification, the basic charac-

teristic of these disorders is the formation of a strong bond

between a previously neutral stimulus and a strong positive

reinforcement. Extinction of such a disorder is unlikely in

ordinary life situations. Punishment may temporally decrease the

performance of the behavior, but it will not remove the habit.

Therapeutic treatment by aversion therapy also has inherent

difficulties. Aversion techniques require split-second timing

such that the aversive stimulus eliminates or at the least precedes

the positive reinforcement. In addition, aversive conditioning

may be extinguished in the same manner as another conditioned

response is extinguished. Thus, relapse rate would be predicted

to be much higher than that of disorders of the first type.

Another principle which Eysenck postulates to account for

the higher relapse rate illustrates a further deviation from

 39

the Hullian framework. Eysenck's reasoning is as follows (1964b):

Since disorders of the second kind may be quite pleasant and

agreeable to the patient, it is often society w:.ich provides the

motivation for treatment. Unlike Hull, who does not consider the

degree of drive in relation to the formation of a habit (Eysenck, 1960b),

Eysenck postulates that performance is very much determined by the

strength of the drive under which the individual learns the habit,

liince the patient for whom society demanded treatment would have

less desire to change his habits, he would be predicted to learn

less effectively in a therapeutic situation (Eysenck, 1964b). To

alleviate this high relapse rate, Eysenck (1964b) sug ests that prin-

ciples deduced from learning theory such as partial reinforcement,

overlearning, spaoed trials, and supportive conditioning be tcste*

in clinical situations.
In addition to these specific extentions of the theoretical

framework, Eysenck (1964a) also included in his latest book a section

on Skinner's operant conditioning as well as a consederation of

various other techniques. Eysenck's position, at present, is stated

as follows:
We should, in approaching the problem of treatment
in the neuroses, try to take as unbiased a view of modern
psychology as possible. Hullians and Skinnerians may have
their internecine quarrels within the academic stomping
ground, but when it comes to practical work, they should
leave their tomahawks and use whatever useful methods
may come to hand in relation to any particular problem
(1964a, p. 6).
Whether a theoretical formulation will eventually be advanced

which is capable of solving such "academic" problens remains to be

seen. Although Eysenck advocates practical application, it should

 40

be reiterated that his theoretical formulations have placed such

application within a more well-substantiated framework.

Metarazzo (196^) has stated that Eysenck's proposal of a

two-factor theory might eventually allow the incorporation of more

complex neurotic conditions related to the symbolic processes within

the behavioral framework. Ford and Urban (1965) have stated that

such an analysis of symbolic responses is necessary for an adequate

therapeutic approach. They postulated that symbolic processes

mediate both generalization of behavior and discrimination. Since

„he3c processes influence an individual's behavior, they may be

assumed to be involved in most kinds of disorders.

Thus the potential of behavior therapy would seem unlimited.

The development of this potential will depend ultimately upon the

elaboration, refinement and broadening of its bases: psychological

theory and research (Ullman and Krasner, 1965)*

 SUMMARY

The consideration of disorde.ed behavior within the framework

of general psychology presents a feasible alternative to the

traditional methods of psychotherapy.

The approach of the behavior therapists has been presented as

being that which utilizes most effectively the principles of behavior

derived from theories of learning and from the experimental

laboratories as a theoretical rationale for clinical methodology.

The basic principle underlying such methodology is that all behavior,

whether adaptive or maladaptive, is learned. Th=; techniques for the

romoval of behavior are many. The basic procedure for such removal,

however, involves the systematic manipulation of the environmental

continge.acies which control this behavior. In addition, the therapist

attempts to replace the removed responses with patterns of behavior

which are more socially acceptable.

Tie behavior therapist's approach is demonstrated to differ from

the traditional approach to psychotherapy with respect to the variables

which he considers important. His focus of treatment is overt beh-

avior, rather than postulated, underlying "causes". At present only

the concepts acceptable to all learning theorists are consistently

utilized within a clinical setting.

The learning theory of Clark Hull was presented as the conceptual

model for the construction of the behavior therapy systems of Joseph

 42
'•olpe and Hans Eysenck.

The psychotherapeutic system of Joseph V.'olpe is based primarily

upon the principle of reciprocal inhibition. This principle he has

postulated to be the cause of success in any therapeutic situation.

Wolpe has presented a learning theory of neurosis and his system is

primarily concerned with providing techniques which permit neurotic

behavior to be unlearned. Wolpe has hypothesized that if responses

antagonistic to anxiety can be made to occur in the presence of

anxiety, reciprocal inhibition of thin anxiety will occur. He has

presented several responses which he feels are very effective in

this respect. Wolpe's theoretical account of the formation and

removal of neurotic behavior has been a topic of controversy. The

techniques which he has proposed, however, have proved to be

remarkably effective.
The work of Hans Eysenck seems to make its most important
contribution at a theoretical rather than a methodological level.

He has distinguished between two types of behavior theory which are

possible within the Hullian framework. The first of these, a general

behavior theory, is typical of the work of Wolpe. The second, an

individual behavior theory, is that utilized by Eysenck. Eysenck's

later work has led him to reject Hull's theoretical model on the

■Trouds that it is not completely adequate as an explanatory tool.

Hie proposal of a two-factor theory of learning seems to broaden

significantly the theoretical base within which behavior theory may

develop.
 BIBLIOGRAPHY

Bond, I. K., and Hutchison, H. C. Application of reciprocal

inhibition therapy to exhibitionism. In Case Studies in
Behavior Modification. Ullmann L., and Krasner, L. Eds.
New York: Holt, Rinehart and Winston, Ino., 1965, 246-251.

Colby, Kenneth Mark. Psychotherapeutic processes. Annual Review

of Psychology,11964. 15* 347-371.

Cooper, A. J. A case of bronchial asthma treated by behavior

therapy. Behavior Research and Therapy, 1964, 1, 351-356.

Dittman, Allen F., Psychotherapeutic processes. Annual Review

of Psychology, 1966, 17, 51-79.

Eysenck, H. J. The effects of Psychotherapy. Journal of

Consulting Psychology, 1952,' 16, 319-324.

Eysenck, H. J. Learning theory end behavior therapy. In Behavior

Therapy and the Neuroses. Oxford: Pergamon Press Ltd., i960,
4-22. (a)

Eysenck, H. J. Modern learning theory. In Behavior Therapy and

the Neuroses. Oxford: Pergamon Press Ltd., i960, 79-83. (b)

Eysenck, H. J. Behavior Therapy and the Neuroses. Oxford:

Pergamon Press Ltd., 1966. (c)

Eysenck, H. J. Classification and the problem of diagnosis. In

Handbook of Abnormal Psychology. New York: Basic Books, Inc.,
1961, 1-32.
Eysenck, H. J. Behavior therapy, spontaneous remission and
transference in neurotics. American Journ .1 of Psychiatry,
1963, 119, 867-871.
Eysenck, H. J. Experiments in Behavior Therapy. New York: The
Macmillan Co., 1964* (a)
Eysenck, H. J. Behavior therapy, extinction, and relapse in
neurosis. In Experiments in Behavior Therapy. New York:
The Macmillan Co., 1964. (b)
 44
Pranks, Cyril M. Conditioning: and abnormal behavior. In
Handbook of Abnormal Psychology. Eysenck, H. J., Ed.
New York: Basic Books, Inc., I96I, 457-488.

Ford, Donald H. and Urban, Hugh B. The reciprocal inhibition

psychotherapy of Joseph VIolpe. In Systems of Psychotherapy.
New York: John Wiley and Sons, Inc., 1965, 211-273.

Gale, Diane, Strumfels, Gloria, and Gale, Elliot. A comparison

of reciprocal inhibition and experimental extinction in the
psychotherapeutic process. Behavior Research and Therapy,
1966, 4, 149-155.

Goldstein, Arnold P., and Dean, Sanford J. The Investigation of

Psychotherapy* New York: John Wiley and Sons, Inc., 1966.

Goldstein, Arnold P., Heller, Kenneth, and Sechrest, Lee B.

Psychotherapy and the Psychology of Behavior Change.
New York: John Wiley and Sons, Inc., 1966.

Grossberg, J. M. Behavior therapy: a review. Psychological

Bulletin, 1964, 62, 73-88.

Hall, Calvin, and Lindsey, G. Theories of Personality. New York:

John Wiley and Sons, Inc., 1957.

Hilgard, Ernest R. Theories of Learning. New York: Appleton-cent

Century-Crofts, Inc., 1956.

Lang, Peter J., and Lazovik, A. David. Experimental desensitization

of a phobia. In The Shaping of Personality. Babladelis,
Georgia, and Adams, Suzanne, Eds. Englewood Cliffs: Prentice-
Hall, Inc., 1967, 455-466.

Lomont, Jf P. The ethics of behavior therapy. Psychological

Reports. 1964, Hi 519-531.
Lomont, J. P. Reciprocal inhibition or extinction. Behavior
Research and Therapy, 1965i 3, 209-219.

Matarazsto, Joseph D. Psychotherapeutic processes. Annual Review

of Psychology, 1965, 16, 181-225.

Betznerj Ralph. Re-evaluation of Wolpe and Dollard-Miller.

Behavior Research and Therapy, 1963, 1, 213-215.

Morgan, Clifford L., and King, Richard A. Introduction to

Psychology. New York: McGraw-Hill Book Co., 1966.
 45
Mowrer, 0. H. Freudianism, behavior therapy and "self-disclosure".
Behavior Research and Therapy, 1964, 1. 321-339.

Murray, E. J. Learning theory and psychotherapy* biotropic versus

sociotropic approaches. Journal of Counseling Psychology,
1963, 10, 250-255.

Rachman, S. Sexual disorders and behavior therapy. American

Journal of Psychiatry, 1961, 118, 235-240.

Rachman, S. Introduction to behavior therapy. Behavior Research

and Therapy, 1963, 1, 3-15«

Renya, L. J. Conditioning therapies, learning theory, and research.

In The Investigation of Psychotherapy. Goldstein, Arnold, and
Dean, Sanford, Eds. New York: John Wiley and Sons, Inc.,
1966, 20-26.

Walton, Donald. An application of learning theory to the treatment

of somnambulism. Journal of Clinical Psychology, 1961, 1,
96-99. ~"
Watson, John 3., and Raynor, Rosalie. Conditioned emotional
reactions. Journal of Experimental Psychology, 1920, 3, 1-14*

Wolpe, Joseph. Psychotherapy by Reciprocal Inhibition. Stanford:

Stanford University Press, 195®«

Wolpe, Joseph. Reciprocal inhibition as the main basis of psycho-

therapeutic effects. In Behavior Therapy and the Neuroses.
Eysenck, H. J. Ed. Oxford: Pergamon Precs Ltd.," i960, 55-144.

Wolpe, Joseph. The experimintal foundations of some new psycho-

thera methods. In Experimental Foundations of Clinical
Psychology. Bachrach, Arther J., Ed. New York: Basic Books,
Inc., 1962, 5^4-576.

Wolpe, Joseph. Quantitative relationships in the systematic

desensitisation of phobias. American Journal_of Psychiatry,
1963, 119, 1062-68.
Wolpe, Joseph. Reply to Mowrer1s comments on reciprocal inhibition
therapy. Behavior Research and Therapy, 1964, 1, 339-343- (a)

Wolpe, Joseph. The systematic desensitization treatment of neuroses.

In Behavior Therapy and the Neuroses. Eysenck, H. J., Ed.
Hew York: The Macmillan Co., 1964, 21-40. (b)

Wolpe, Joseph, and Stevenson, Ian. Recovery from sexual deviations

through overcoming non-sexual neurotic responses. American
Journal of Psychiatry, I960, 116, 737-742.
 46

Ullmann, Leonard P., and Krasner, Leonard. Case Studies in

Behavior Modification. New York: Holt, Rinehart ana Winston,
Inc., 1965. 1-65.

Yates, Aubrey J. Symptom and symptom substitution. Psychological

Review, 1958, 65, 371-374.