Behavior Therapy Versus Psychoanalysis Therapeutic and Social Implications JOSEPH WOLPE Temple University School of Medicine ABSTRACT: Although the specific efficacy of psycho- fundamental importance" (p. 159). Nevertheless, analytic therapy in the treatment of the neuroses has it is psychoanalytic theory that has continued to never been demonstrated, psychoanalytic theory and be the most pervasive influence in psychothera- practices continue to dominate the field of clinical psy- peutic practice. chology. That psychoanalytic theory has not been dis- placed by the behavioral theory of neurosis is remark- able in mew of the persuasive evidence that exists for the efficacy of behavior therapy. One reason for this The Reign of Psychoanalysis seems to be the persistence of widespread mispercep- tions of behavior therapy. It has been represented to According to psychoanalytic theory, mental activ- the public as an "inhuman" treatment that routinely resorts to electric shocks and other unpleasant agents ity is partly conscious and partly out of reach of and to the profession as a therapy incognizant of the consciousness in the "unconscious mind." Neurotic patient's feelings or thoughts and applicable only to symptoms are regarded as the manifestations of neuroses that are "simple," such as phobias—an image emotional forces that have been "repressed" in the regularly reinforced by "authorities" who are misin- unconscious. Freud (1922/1950) regarded these formed. This article attempts to correct these misper- symptoms as "compromise formations between the ceptions. It also draws attention to the suffering im- repressed sexual instincts and the repressive ego posed on many by years of psychoanalysis. The promise instincts" (p. 107). Psychoanalytic therapy aims to of widespread availability of behavior therapy as an overcome a neurosis by bringing the putative re- alternative will only be fulfilled when more high-quality pressed impulses into consciousness, through such training is funded. means as free association and dream analysis. All The present century has seen the birth (Jones, the derivatives of psychoanalysis (the theories of 1924) and the development (Ayllon & Azrin, 1968; Adler, Sullivan, and others) exert their main ther- Wolpe, 1958) of behavior therapy—methods of apeutic effort toward making the unconscious con- psychotherapeutic change founded on principles scious (Munroe, 1955). of learning established in the psychological labo- The clinical effectiveness of psychoanalytic ratory. Its results in the treatment of human neu- therapy has never been established. Eysenck (1966), roses have been quite impressive. Paul (1966), on in a review of 24 studies encompassing over 7,000 the basis of a survey of controlled studies on sys- cases, concluded that the data failed to show that tematic desensitization, has stated, "For the first psychoanalytic therapy facilitates the recovery of time in the history of psychological treatments, a neurotic patients. Erwin (1980) has convincingly specific therapeutic package reliably produced defended Eysenck's conclusion against challenges measurable benefits for clients across a broad range by Bergin (1971), Bergin and Suinn (1975), and of distressing problems in which anxiety was of Brown and Herrnstein (1975). A particularly note- worthy study is that of the Fact-Gathering Com- This article was the address of the recipient of the Distinguished mittee of the American Psychoanalytic Association Scientific Award for the Applications of Psychology at the meet- ing of the American Psychological Association, Montreal, Sep- (Note 1). Out of 595 patients, 306 were judged to tember 1980. have been "completely analyzed" (in a mean of Requests for reprints should be sent to Joseph Wolpe, De- about 600 sessions); 210 of these were followed up partment of Psychiatry, Temple University, c/o Eastern Penn- sylvania Psychiatric Institute, Henry Avenue, Philadelphia, afterwards, and 126 were stated to have been cured Pennsylvania 19129. or greatly improved. This is 60% of the completely Vol. 36, No. 2, 159-164 AMERICAN PSYCHOLOGIST • FEBRUARY 1981 • 159 Copyright 1981 by tlic American Psychological Association, Inc. 0003-066X/81 /3602-0159S00.75 analyzed group but only about 31% of the original imental neuroses" in animals—a long-lasting sus- total. ceptibility to the triggering of strong anxiety It is customary to turn a blind eye to such poor responses by particular stimulus conditions, a sus- results and to contend that the treatment is nev- ceptibility in many respects similar to the neuroses ertheless on the right track because the psycho- of human beings (Wolpe, 1967). Many experi- analytic theory of neurosis is true. In actuality, not menters in the United States subsequently con- a single one of the theory's main propositions has firmed Pavlov's observations, often using variations ever been supported by scientifically acceptable of his procedure (for a review, see Wolpe, 1952). evidence (see, e.g., Bailey, 1964; Salter, 1952; Val- Using a method described by Dimmick, Ludlow, entine, 1946). But this too is brushed aside. That and Whiteman (1939), I produced experimental this happens is a tribute to the expository brilliance neuroses in cats by administering painful but non- with which Freud presented his theories. His writ- damaging electrical stimuli (high voltage, low am- ing weaves a magic web from which few can ex- perage) of two seconds' duration to an animal in tricate themselves once enmeshed. To the con- a small cage (Wolpe, 1952, 1958). This stimulation vinced it is sacrilegious to suggest the need for elicited strong fear reactions: The animal's pupils anything so mundane as empirical testing. dilated, its hair stood on end, and it breathed rap- After a phase of outraged opposition to psycho- idly. Repeating the stimulation at irregular inter- analysis early in the 20th century, converts to it vals of minutes resulted in the animal's becoming were legion. In 1939, it was officially approved by very fearful of the cage and surrounding stimuli the American Medical Association. By then it had in between 5 and 20 repetitions. The autonomous become widely accepted by Western intellectuals power of these stimuli strongly to arouse fear as a philosophy of life. They saw it, in vibrant would have lasted the life of the animal if left contrast to the dry abstractions of academic psy- untreated (Gantt, 1944). The fear was undimin- chology, as a psychology of reality, dealing with ished by exposures, short or long, to the experi- things that mattered and revealing dark and mys- mental cage, nor was it alleviated by months of terious aspects of the mind. absence from the cage. In every animal, however, Not everybody was persuaded. There were the fear could be systematically weakened, even- many who saw the flaws in the theory and some tually to zero, by arranging for small amounts of who vigorously criticized it. One of the most note- it (evoked at first by generalized stimuli) to be worthy critics was Wohlgemuth, whose Critical inhibited by the competition of eating behavior. Examination of Psychoanalysis appeared in 1923. This suggested that a therapeutic principle resided But critiques like his had little effect—illustrating in response competition. Clinical trials showed that Conant's (1947) maxim that theories are not aban- the competition of feeding also overcame chil- doned on the basis of contradictory evidence. On dren's fears (Jones, 1924) but not those of adults the other hand, Conant attests that they are aban- (Wolpe & Wolpe, in press). Fortunately, a consid- doned when better theories arrive on the scene. erable number of other responses were found to There was no better theory in 1923. But now, in have the ability to inhibit and consequently to 1981, a seemingly better theory—the behavioral overcome adults' fears. Fear can be inhibited by theory of neuroses—has been with us for a quarter the calmness generated by deep muscle relaxation, of a century. Yet psychoanalysis sits firmly in the by the expression of legitimate anger in the context clinical saddle. The first step in attempting to ex- of certain inappropriate social fears, and by the plain why this is so is to examine what evidence use of sexual responses in cases of sexual fear, as there is that behavioral theory and the therapy that well as by a number of more esoteric methods, emerged from it are really better. including flooding (Wolpe, 1973). The important question, however, is whether these experimentally derived methods actually Conceptual Origins and Therapeutic achieve an unusual percentage of favorable results Efficacy of Behavior Therapy of and are significantly more economical of time and Neuroses effort. The answer lies in a comparison with the well-documented fact that the practitioners of I shall now review the foundations of the behav- practically any system of psychotherapy obtain ioral theory of neurosis and the evidence of the recoveries or marked improvements in 40%-50% clinical efficacy of behavior therapy. At the be- of the cases they treat (e.g., Eysenck, 1966). If the ginning of this century, Pavlov produced "exper- followers of different systems—Freudian and Jung- 160 • FEBRUARY 1981 • AMERICAN PSYCHOLOGIST ian analysts, nondirective therapists, encounter psychotherapy and waiting-list groups, were con- groupers, and primal screamers—all achieve this sidered either improved or recovered, a difference percentage, there must be a common process work- significant at the .05 level. Patients treated by be- ing for all of them that has nothing to do with havior therapy also showed significant improve- their respective techniques.. The distinctive pro- ment in work and social adjustment, while cedures of a therapeutic system cannot be said to psychotherapy patients showed only marginal be helpful unless a recovery rate significantly improvement in work and none in social adjust- above the common baseline can be shown. ment. At one year, only those who had been treated The question then becomes, Does behavior ther- by behavior therapy showed greater improvement apy improve on the common run of results? The in target symptoms than waiting-list subjects.