Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Non-Specific Treatment Factors And Deconditioning In Fear Reduction
(USC Thesis Other)
Non-Specific Treatment Factors And Deconditioning In Fear Reduction
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
This dissertation has been microfilmed exactly as received 69-5054 JAFFE, Lawrence William, 1931- NON-SPECIFIC TREATMENT FACTORS AND DECONDITIONING IN FEAR REDUCTION. University of Southern California, Ph. D., 1968 Psychology, clinical University Microfilms, Inc., Ann Arbor, Michigan NON-SPECIFIC TREATMENT FACTORS AND DECONDITIONING IN FEAR REDUCTION by Lawrence W. Jaffe A Dissertation Presented to the FACULTY OF THE.GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Psychology) August 1968 UNIVERSITY O F SO U TH ER N CALIFORNIA T H E G R A D U A TE S C H O O i. U N IV E R SIT Y PA RK L O S A N G E L E S, C A L IF O R N IA 9 0 0 0 7 This dissertation, written by ............ under the direction of hXs... Dissertation Com mittee, and approved by all its members, has been presented to and accepted by The Gradu ate School, in partial fulfillment of require ments for the degree of D O C T O R OF P H IL O S O P H Y t Dean n„,, Aug u s t 1 968 DISSERTATION COMMITTEE YT r 9 a To Malcolm Boyd Dana, a man knowledgeable In the ways of the human spirit. With respect and love. A CKNOWLEDGEMENTS I wish to express my appreciation to the members | of my dissertation committee., Dr. Norman Cliff, Dr. ; i i Milton Wolpin and Dr. L. Douglas DeNike for their assist- | ance in this project. I am grateful■to Dr. Cliff and j j Dr. DeNike who generously came to my assistance when it j j appeared that the beginning of the study would be delayed j at least a year because of the unavailability of a ! guidance committee chairman. An especially warm word of thanks is due Dr. DeNike who supervised the dissertation. His generosity, unaffected interest and devotion are appreciated and gratefully acknowledged. I also wish to thank Dr. J. W. McKelligott, Chief Psychologist at Wadsworth V. A. Hospital for allowing me to rearrange my work schedule to conform to ! I the exigencies of the dissertation and for other acts of I kindness and understanding too numerous to mention. Finally I wish to thank my fiancee Karen Betty j Lind for her loving acceptance of the withdrawal of time j and attention necessitated by the demands of completing this work expeditiously. Also her thoroughly competent and speedy performance of a number of complex clerical tasks hastened completion of the study. 111 ________________ TABLE OP CONTENTS Page ACKNOWLEDGEMENTS................... Hi ! LIST OP TABLES ..................... vii Chapter I. INTRODUCTION................. 1 The Placebo in Medicine The (Non-) Trait of Placebo Re activity Placebo Groups in Psychotherapy Patient Prognostic Expectancies Therapist Prognostic Expectancies Deconditioning of Pears Social-Psychological Effects Statement of the Problem Hypotheses II. METHOD . .............................. 43 Design Procedure Summary of Dependent Variables III, RESULTS.................... . 58 Pre-Treatment Equality of Groups Experimental Hypotheses Performance of Practice Effects Group (c) Treatment Comparisons Comparisons Between Treatment Means Treatment Effects Effects of the Expectancy Manipulation The Liking for the Experimenter Measure Paid vs. Unpaid Subjects Male vs. Pemale Subjects iv \ Chapter Page IV. DISCUSSION • 76 Practice Effects Comparisons Between Treatment Groups The Wolpin Deconditioning Treatment The Placebo Therapy Efficacy of the Expectancy Manipulation Relationship between Expectancy and Improvement The Manipulation of Expectancy Expectancy and Social-Psychological Effects Liking for Experimenter Miscellaneous Questionnaire and Interview Material V. SUMMARY................................ 102 APPENDIX A: Pear Survey Schedule............ . . 106 APPENDIX B: Description of the Study Read to Prospective Student Volunteers . 108 APPENDIX C: Pretest Manual ....................... 110 APPENDIX D: Generalized Subjective Pear Scale . 114 APPENDIX E: Behavioral Avoidance Test .......... 116 APPENDIX P: Subjective Pear S c a l e........ .. . 118 APPENDIX G: Instructions to D- Group....... 120 APPENDIX H: Instructions to D and P Groups . . . 123 APPENDIX I: Questionnaire B ................ 125 APPENDIX J: Neutral Scene ..................... 128' APPENDIX,K: Pear Scene .......................... 130 APPENDIX L: Work Sheet...................... 132 APPENDIX M: Rationale and Additional Instruction to P G r o u p .................. 134 APPENDIX N: Posttesfc Manual ................... 137 APPENDIX 0: • Questionnaire C ................... 140 APPENDIX P: De-Briefing Format--D Group .... 144 APPENDIX Q: De-Briefing Format--D- Group . . . 146 APPENDIX R: De-Briefing Format--P Group .... 148 APPENDIX S:- Raw D a t a ................. 151 REFERENCES............... 157 vi LIST OF TABLES Table Page 1. Summary of Analyses of Variance of Pre- Scores on the BAT, GSF and SF Scales for All Groups......................... 59 2. Summary of Performance of Groups on. the Criterial Measures of Fear Reduction . . 60 3- Summary of Single Mean _t Tests of Group Means on the Criterial Measures of j Fear Reduction................. 62 4. Summary of Duncan's New Multiple Range Tests of Differences Between Pairs of Treatment Means on the Criterial Measures of Fear Reduction............ 64 5- Summary of Analyses of Variance on the Performance Groups on the Criterial i Measures of Fear Reduction............ 65 | 6. Partial Product-Moment Correlation Co- I efficients for Expected vs. Actual ] Improvement on the Criterial Measures--. i with the Influence of Pretest Scores j Eliminated....................... 69 j 7. Summary of Analysis of Variance of "Liking I for Experimenter" Scores for all Groups. 70 j I 8. Summary of t Tests of Differences Between ! Pairs of”Treatment Means on the Liking I for Experimenter Scale 71 j 9- Summary of _t Tests of Differences Between Paid and Unpaid Group Means on the j Criterial Measures of Fear Reduction . . 74 10. Summary of _t Tests of Differences Between Male and Female Group Means on the Criterial Measures of Fear Reduction . . 75 vii CHAPTER I I INTRODUCTION ! I Non-specific treatment effects are as old as the j healing arts themselves. All pre-scientific techniques of healing may he considered largely non-specific in their effects since early healers had no very reliable way of determining precisely what in their procedures was efficacious, or why. An embarrassingly similar state of j affairs characterizes present-day psychotherapy. In the j shaman1s complicated ritual, what had curative powers and what was irrelevant was presumably not a question he con- i cerned himself with (Prank, 1961) . From a scientific j point of view any healing effect possessed by the ritual of the shaman may be termed a non-specific, expectancy, or placebo effect. The same may be said of the so-called j faith-cures--the sometimes indisputable and spectacular I ' - j I j cures'achieved at Lourdes, for instance (Prank, 1959)- It is the purpose of the present study to measure the influence of non-specific treatment effects such as expectation of relief, demand characteristics of the experiment, and placebo effects on the outcome of psycho- i therapy. Interest will be focused on the effects of such j i factors on a deconditioning treatment for fears of small animals. The Placebo in Medicine The history of medicine up to the present era has been the history of the placebo effect1 (Shapiro,, 1964a). Even after the advent of modern scientific medi cine., the placebo or psychological effect of medical treatment has accounted in substantial degree for what ever effect the treatment had in alleviating distress, curing ills, etc. (Shapiro, 1964a). Until recently, how ever the phenomenon has been little investigated (Pepper, 1945)• It seems to have aroused curiosity least of all in those professions which might be expected to take great interest in it, viz., psychologists and psychiatrist (Shapiro, i960). Please to these professions to take a greater interest in the phenomenon are frequent in the literature (Rosenthal & Prank, 1956j Shapiro, i960; Whitehorn, 195&)• Cupping, leeching, bleeding, sweating and purging j were each particularly favored by the medical profession j I at one time or another and the therapeutic effectiveness I of these procedures went unchallenged for long periods of | I ■^The term "placebo1 ’ will be used here in its j broad sense, similar to the usage proposed by Shapiro j (1964b) to refer to any procedure which has an effect on j the condition of a patient without its having specific activity for that condition. Placebos may include medi cations, (inert or active), and mechanical, surgical and psychotherapeutic techniques. Placebo effects may be negative but are most often positive (Honigfeld, 1964). 3 time. Now we know that any effect that these and many other treatments such as blistering, and freezing may have had were psychological. This may be regarded as histori cal evidence for the power of the placebo or psychological i effect of treatment since in spite of the fact that these, were placebos, Shapiro (,1964a) notes "...physicians main tained their positions of honor and respect throughout history in the face of thousands of years of prescribing useless and often dangerous medications ^p.73- ]*'' Examples of medications cited by Honigfeld (1964) are lizards' i blood, crocodile dung, the teeth of swine, the hoof of an ass., putrid meat and fly specs. With the event of increasingly sophisticated scientific methodology in the evaluation of medical i treatment, particularly drugs, the power of the placebo j 7 i has come to be increasingly recognized. For instance, thej placebo effect has been found capable of reversing the j action of the potent * emetic ipecac (Wolf, 1959)j being responsible for profound physiological changes in an j I l incurable terminal malignancy (Klopfer, 1959)., of being \ as effective as a cold vaccine in diminishing the i i frequency of colds (Diehl, 1940), in the form of a sham operation for angina pectoris of being more effective than the real operation (Dimond, Kittle & Crockett, i960). Placebo "drugs1 1 have been found as effective in suppress ing the cough reflex as a drug (codeine) widely used for 4 that purpose (Hillis, 1952), and of mimicking both the specific physiological effects (lowered white blood count, reduced fever) as well as the side effects of the anti biotic streptomycin (Kurland, 1957)* ! ; I j The potency of the placebo effect under the properj ' I I circumstances is therefore no longer in question. Consid- ; erable negative halo, however, still attaches to the word. I Interestingly it is psychotherapists whose treatment most | resembles the placebo who pay least attention to it : i (Honigfeld, 19641. Physicians are also quite defensive about the use of placebos, 75$ of them judging that other physicians use placebos more than they do (Shapiro, 1964a). Some of the negative connotations which seem to attach to the word "placebo" may be associated with its etymol ogy. The word means literally (in Latin) "I shall please" and historical definitions of the word focus on I its being given "to please rather than to benefit" a patient. The oldest known medical dictionary definition j | of the placebo is found in Motherby's Medical Dictionary-- j I i I the 1785 edition, and reads, "A commonplace method or | medicine" (Shapiro, 1964a). Despite its negative connota- j 1 tions the placebo is gaining in interest to both practi- J tloners and researchers (Honigfeld, 1964). One important reflection of the increased recognition of the placebo is the increased use of "double-blind" studies in the evalua tion of drugs. In these studies placebos or the presum 5 ably active drug under investigation are given to differ ent patients with neither the patient nor the physician i I aware of which is which. Thus nearly every well-designed j ! drug study will contain a report of a placebo effect. I | The (Non-) Trait of Placebo Reactivity } Looking at placebos., for a moment, from the point of view of the psychology of individual differences, the j comment of Shapiro (l964d) succinctly sums up our current I i state of knowledge: "...investigators have attempted to ! relate placebo effects... to the personality of the patient.! i ] But the personality traits reported in any one study tend j to differ from those reported in other studies £p. I80J." Some difference of opinion still remains, however. j Most investigators are of the opinion of Wolf (l957) and I his group who feel that there are as great intra-individual as inter-individual differences in reaction to placebo. To put it another way, knowing who was a placebo reactor ■ in one situation doesn't help to predict who will be one in another situation. At least one other research group (Lasagna, 195^-) continues the quest for the trait of placebo reactivity. I It perhaps should not need to be mentioned that placebo reactivity does not correlate with the trait of j suggestibility (Shapiro, Wilensky & Struening, 1968; I Stukat, 1958). This should not have to be mentioned because since placebo reactivity cannot be established as 6 a reliable phenomenon (trait) at all it cannot__correlate significantly with any other trait. Placebo Groups in Psychotherapy i i The inclusion of placebo control groups in psycho- j i therapy outcome studies was perhaps first suggested by ! Rosenthal and Prank (1965)* These authors pointed out that the improvement rate for neurotics in psychotherapy, about two-thirds (Eysenck, 1952; Eysenck, 1965)j was approximately the same as the improvement rate due to I placebo effects in illnesses where emotional factors are i ! ' ! i judged to play a significant role, e.g., colds (Diehl, j I 1933) and headaches (Jellinek, 19^6). These authors made j very clear a point that had been raised by others before j and since, that improvement in a psychotherapy patient | after treatment cannot be attributed to the particular j psychotherapeutic technique used without comparing this j i rate of improvement with that which would arise in response to another therapy in which the patients have | "equal faith and expectation of relief" (Rosenthal & j Prank, 1956). ! i Their suggestion seems to have been followed in a j few scattered investigations but with only one exception | i the researchers have neglected to equate their placebo- i I group with the treatment group on "faith in the therapist" I and "expectation of relief". It would seem to be a sine qua non of an adequately designed study of the placebo _ 7 effect to be sure the subjects in the placebo group believe they are receiving a bona-fide treatment. In reviewing many of the studies containing placebo control groups, it is hard to believe, from a common sense stand point alone, that the subjects in the placebo group could think they were actually being treated for their problem by means of a bona-fide technique and by a therapist who believed he could help them, e.g., Lang and Lazovik (1963) Lang, Lazovik and Reynolds (1965), Snyder and Sechrest (1959)• Perhaps because of'a behavioristic Zeitgeist j 1 which discourages inquiry into subjective processes, these investigators naively assume that if they have labelled a i particular placebo treatment "therapy" the subjects will be deceived thereby. On the contrary, there is a growing and already substantial literature (e.g., Criswell, 1958) which shows that subjects are likely to be responding to an experimental situation quite different from the one the experimenter thinks he has set up, that subjects, at times, have purposes and interests quite different from those of the.experimenter, and that they are very sensitive to his i i hidden purposes. . j It is commonly part of the experimental plan that the experimenter’s purposes remain at least partially hidden from the subject. The subject in turn attempts to divine the true nature of the situation so that he can 8 better present himself in a favorable light, prove or disprove the hypothesis, and so on. These and other like lihoods are set forth in an exceptionally clear and able manner by Riecken (1962). In support of such processes there is a growing body of experimental evidence such as the work of Martin Orne (1959a)and Robert Rosenthal (1966) . i One of the few outcome studies in psychotherapy j j j j which utilized a placebo control group was that of j j Snyder and Sechrest (1959)- These authors were interested ! ! I I in assessing the efficacy of directive group therapy with j | • i institutionalized mentally defective juvenile delinquents, j i They used objective, non-reactive indices of improvement | I such as number of demerits. The treatment group improved ! significantly and the placebo group also showed improve- | ment over the no-treatment group, though not significantly) I so. The placebo group in question, however, was inade quately constituted for a test of the placebo effect; the authors commented that it "...might not constitute a true placebo effect £p. 121- ] ." Among the factors which kept the placebo treatment j j ! from being one in which the patients could have "equal faith and expectation of relief" was the absence of i supportive statements such as were made to the treatment group, e.g., statements to the effect that the purpose of the group meetings was "adjustment" and that they (the treatment group) had "superior adjustment potential". 9. Instead, the situation went unstructured, the authors commenting, "...[placebo} group members... seriously felt their goal-lessness £p. 121]." Conversation was steered j ! f • I away from discussion of focal problems, a sense of group j j cohesiveness was lacking.and "overt resistance was ob- t servable at every session of the placebo group [p. 122]." j To judge from the comments of the authors themselves, it [ is probably a fair assumption that the placebo therapist | | i | was far from enthusiastic and confident about the value of j j i j his treatment. The variable of therapist expectancy, in j j this study, appears to have gone uncontrolled. It is probable that a stronger therapeutic effect of the placebo treatment would have been evident if the factors mentioned above had been controlled for. i i j Similar though milder objections must be made to i I Paul's (1966) attempt to construct a placebo group. Sub- | j jects In this group were given the task of picking out j j noises of a particular nature from a tape recording of i | mixed noises. They were told this was a stress task used j ! f | in training astronauts— (the task was actually "rather j relaxing"). They were also told that this procedure would j help them overcome their fears of public' speaking and to j 1 help in this they were given a "fast-acting tranquillizer" j S (actually an inert substance), at the start of each of their five treatment sessions. This treatment would not appear to have the 10 "face-validity" of either the traditional "insight” therapy approach, or systematic desensitization and there fore it must he considered doubtful that the placebo treatment aroused in the subjects the degree of expecta tion of relief of the other procedures with which it was compared. I ' i The design of this study included ratings by the j i therapists of the degree to which they expected subjects j i I in the various groups to improve. Subjects in the placebo j i group were expected to improve-least. Evidence will be ■ presented below to the effect that this factor alone could j undermine the therapeutic effectiveness of the placebo treatment. j Notwithstanding these weaknesses in this otherwise carefully designed study, the placebo treatment proved to be as effective as the short-term "insight" approach in relieving stage-fright and half as effective as desensi- i j i tization. Improvement was "unexpectedly stable" in the j I i i I placebo group as reported in a follow-up study two years I | I j later (Paul, 1967)- There was no more tendency for the j i placebo subjects to lose their gains than subjects in the regular treatment groups. A study by Paul and Shannon (1966), essentially a replication employing groups of Paul's (1966) study, obtained essentially the same results as the original investigation. The comments made above with regard to 11 Paul's (1965) study., therefore., are applicable to the study of Paul and Shannon (l966 ). A recent study of Zeisset's (1968) used a placebo therapy similar to that of Paul (1966) except that the pill placebo ("fast-acting tranquillizer") was omitted. j i Comparisons were made between systematic desensitization, I relaxation, placebo therapy and no-treatment in modifying j i the "interview anxiety" of schizophrenics. The desensi- ] tization and relaxation groups improved an equal amount j while the placebo and no-treatment groups failed to show improvement. The comments made above in reference to Paul's (1966)study are, of course, equally applicable to j ! Zeisset's (1968) use of essentially the same placebo j treatment. Why the placebo therapy failed to reach the level of effectiveness found by Paul (1966) and Paul and Shannon (1966) can only be guessed at but may be due to j population differences (college students vs. chronic ' | j schizophrenics) or perhaps the pill placebo is simply an j | ; essential part of the placebo treatment. ; In the recent experiment of Efran and Marcia (1967, ) ! I the potential power of the psychotherapeutic placebo 1 ■ i reveals itself more clearly. In this study.phobic subjects were divided into an experimental group, a "treatment control group" and a non-treatment .control group. None of the groups received treatment for which there was any 12 theoretical rationale. The experimental group, and the treatment control group were led to "believe that they were being treated for their phobias. They were, in fact, treated to an elaborate pseudotherapy (placebo) consisting of tachistoscopic presentations of blank cards and shocks.. Subjects were given a rationale for this treat- i j ment and were presumably under the impression that they ! i were receiving subliminal tachistoscopic presentations j of the feared stimulus. Marked therapeutic gains were j I made by the experimental group which believed it was being j treated with a new and effective technique. The other I experimental group (the "treatment-control" group), which believed that it was being treated with a new technique but minus an important ingredient, made lesser though significant gains. The no-treatment control group did not improve. Results were very similar to those obtained through the application of desensitization and similar I > ! procedures. j i j The following methodological drawbacks in the j i ' ■ i ' i ! above study should be noted however: The number of sub- | i . jects was relatively small (27 total for the three goups); there was an unequal initial assignment of subjects to groups; and the subjects were not matched for degree of j fear in assignment to groups--with a disproportionately | i i large number of severe phobics being -assigned to the j treatment group. Moreover, since while looking at the 13 slides which they thought contained scenes depicting the feared stimulus subjects were sometimes shocked* there was an aversive conditioning paradigm present which like the over-assignment of severe phobics to the pseudother apy group could have resulted in an underestimate of the I i pseudotherapy effect. In spite of the difficulties in j i i interpretation caused by the above the expectancy effects j f i [ ! seem to have been strong enough to warrant the conclusion | 1 ! j that expectancy or placebo effects alone can result in the J i I alleviation of selected fears. ! i This study is the only one found in the literature which focused its attention on the placebo or expectancy effect in psychotherapy. It is also notable because it is the only study reviewed in which it appeared that the authors succeeded in arousing in their placebo sub- \ jects the degree of "faith and expectation of relief" in { j the placebo treatment necessary in order to demonstrate 1 i the phenomenon in its full strength. In no other study j was sufficient care taken to construct a placebo treatment\ t that subjects could believe was a bona fide treatment. j i In an attempt to ascertain the effective ingred- j i ients in systematic desensitization* Lang* Lazovik and Reynolds (19^5) used a "pseudotherapy" group which i received 5 sessions of relaxation training plus 11 | sessions of pseudotherapy. In the pseudotherapy subjects were permitted to talk about the anxiety hierarchy but not 14 about their fears. It is difficult to see how this could have been accomplished very smoothly but the authors do .not describe the pseudotherapy in any detail except to comment that "...the pseudotherapist gently steered the | conversation away from phobic or other sensitive material [P- 396J," and also that their object was "...to avert affect-laden statements [p. 40lJ." Not very surprisingly. the pseudotherapy subjects showed no improvement. It is,, j | ! of course, quite understandable that the experimenters j ; . I I ' ’ j j wished to keep their pseudotherapy patients from receiving real treatment in this case desensitization, but at the same time one must doubt the "face-validity" of such a placebo treatment. In another attempt to separate "the wheat from the chaff" as far as the ingredients of systematic desensitization are concerned Lang and Lazovik (1963) in an "own control" design, first gave their phobic sub- | jects (later to be desensitized,) training in relaxation j and hypnosis and tested them to determine whether the j "suggestive effect" of this training induced any therapeu tic gain. It did not. After administration of the desensitization component of the therapeutic procedure, however, there was a significant improvement on the therapeutic criterion. This study is subject to the same criticisms (lack of control of therapist and patient expectancies) 15 as the studies reviewed before. Two other studies will now be briefly reviewed which, though they did not make use of psychotherapy placebo groups, are relevant to this subject. In a frequently quoted but relatively poorly- j I ! designed and reported study Gliedman, Wash, Imber, Stone i and Frank (1958) found that pill placebos administered to i | outpatients who formerly had gotten brief psychotherapy i reduced the patients' discomfort satisfactorily and pill i placebos were as satisfactory as an "active" drug in ! ! reducing subjective discomfort. j Friedman (1963) gave psychiatric outpatients a symptom rating scale at their first appointment. They were 1 then given another copy of the symptom rating scale and j asked to fill it out as they expected to after their course of treatment. They were then given an interview j t which was designed to be non-therapeutic. The question naire was then re-administered (for the third time). j Those who expected to feel better after the course of i | treatment did so after the initial interview and tended to | feel relief of the symptoms which they expected to be relieved. The author interprets his results as follows: "...patient expectancy of help is activated at the first j j physician-patient contact and accounts for the reported j t decrease in discomfort after the initial interview [P. I 66]." 16 As mentioned above the last two studies summar ized did not actually make use of expectancy or placebo control groups in the evaluation of the effect of psycho therapy. Still further afield are the following three studies which are briefly cited here because their results may cast light upon the phenomenon of interest. All three of these were psychotherapy outcome studies and all employed patients on waiting lists as control group subjects. The often quoted study of Barron and Leary (1955) found that patients in individual and group psychotherapy j i 1 j . r did not improve on the MMPI significantly more than ! i i waiting list patients. Cartwright and Vogel (i960) used patients on a waiting list as controls for psychotherapy patients and got a number of results which contradicted their hypo theses. Patients on the waiting list seemed to improve j more than psychotherapy patients assigned to inexperienced j | therapists. Patients on the waiting list for longer j I periods of time improved significantly on the TAT. As for improvement on a self-report measure, which is likely to be more sensitive to placebo effects, therapy was superior. G-rummon (195^) took measures on wait-list patients to see if they improved in the two months before therapy--they didn't but thos'e who left therapy early - 17 did! It .would seem., based on the literature summarized above, that when a placebo therapy has adequate "face- validity'1 powerful and long-last In.grtherapeutic benefits may follow. On the other hand the placebo effect may be weak or absent if sufficient care is not exercised In setting up and administering the placebo therapy. Patient Prognostic Expectancies The entire literature on the placebo effect is at least partially relevant in this context since any effect the placebo may have must be imparted to it by some action of some level by the patient. Moreover the therapeutic procedure, in accordance with the etymology of the word "placebo" would have to "please" (conform to the expectations of) the patient in order tojbe of maxi- 2 mum benefit. Historically faith healing is probably the out standing example of the healing power of strong positive patient expectancies (Prank, 1959; Prank, 1961). Refer ences of an anecdotal or intuitive nature alluding to the importance of the patient--expectancy variable are fre quent in the psychological and medical literature (Goldstein, 1962; Honigfeld, 1964; Shapiro, 1964). Of ? Perhaps the difficulty In treating lower class patients by means of a verbal psychotherapy is because a belief in the efficacy of "just talking" requires more sophistication than these patients possess. 18 particular importance, perhaps, is the already-cited paper of Rosenthal and Frank (1956) which, in calling for i placebo control groups in psychotherapy, emphasized the ! j essentiality of equating the groups on the variables of j ; I "faith in the therapist" and "expectation of relief". j In a reply to Rosenthal and Frank (1956), Cart- j wright and Cartwright (1958) put themselves on record as I j being of the opinion (based upon their "clinical bets") I j ! j that patient expectancy bore no relationship to the out- | come of psychotherapy. ! Experimental studies of the patient-expectancy variable in psychotherapy have been both infrequent and contradictory though the preponderance of the evidence appears to be that there is a relationship between patient-expectancy and outcome of psychotherapy (Goldstein, 1962). Studies by Brady, Zeller and Reznikoff j (1959 ); Friedman (1963); Goldstein (1961) and Lipkin (1964) [ i have suggested that a positive relationship does exist ! while Brady, Reznikoff and Zeller (i960) and Goldstein ! (I96C) failed to find a relationship. i Studies such as that of Efran and Marcia (1967 ) | and Paul (1966) and others which attempted to manipulate subject expectancy in therapy like situations have been quoted in another context and studies of this nature have tended to be contradictory.. There as yet appears to be insufficient evidence to draw any more firm conclusions 19 then Goldstein did In his (1962) review of the litera ture : In summaryj It may be observed that growing evidence tends to support the contention...that patient prog nostic expectancies relate to obtained patient im provement in psychotherapy. The exact nature of this relationship remains somewhat equivocal,, with theo retical and experimental material increasingly point ing toward curvilinearity. (p. 24) The evidence from level of aspiration studies, though not j | unequivocal, points in the same direction (Goldstein, | 1962). | ! Therapist-Prognostic Expectancies j ■ j The voluminous literature on placebo effects is also relevant to the subject of therapist prognostic expectancies (Goldstein,- 1962). Some of the drug and i ’ ’medical" placebo research has already been cited-- ; excellent recent reviews are those of Honigfeld (1964) and Shapiro (1964). Two additional studies by way of i examples, will be cited here: I | Batterman and Grossman (1955) found that patients j j found no more pain relief from aspirin than from placebo | i unless the dispensing physician—knew which was which. | j Feldman (1956) found that the effectiveness of chlorpromazine seemed to a function of the physician1s initial orientation to the use of tranquillizers--those j i who were enthusiastic met with success--those who were j "against" met with therapeutic failure. The following relevant comments of researchers 20 in the area may be of interest: Honigfeld (1964) comments., "The zeal of the doctor toward the particular treatment modality may well be translated by the patient into treatment benefit £p. 239]•" Lesse (1964), states: It is well known that the greatest number of i positive placebo effects and the most enduring j are obtained, in general, by therapists who are optimistic and enthusiastic and who have strong convictions as to the response they expect to achieve from a specific drug or procedure. The statement most often attributed to Trousseau j (Honigfeld, 1964), "Use the new drugs while they still | have the power to heal", is also relevant in this context. In summary it may be said that there is perhaps ! no more eloquent testimonial to the importance of ther- j apist expectancy on the effectiveness of drugs than the j necessity for double-blind methodology in drug research. J With regard to the outcome of psychotherapy there j l ] appears to be general acceptance of the importance of the j i variable of therapist prognostic expectancy. The expert- j mental literature on the subject will now be briefly j i j summarized: I Goldstein and Shipman (1961) and Stone, Imber j and Prank (1966) have shown that attitude of the therapistJ 1 toward psychotherapy bears a positive relationship to the ; 1 success of his patients in psychotherapy. Prognostic expectancies of therapists in a study by Goldstein (i960) bore a positive relationship to actual outcome of 21 therapy. Negatively, Paul (1966) assessed the degree to which his five therapists expected therapeutic results from the three procedures used. Here there was no rela tionship between expectancy and outcome. The preponderance of evidence, as with the variably of client prognostic expectancy, is that therapist prog- i nostic expectancy influences the outcome of psychotherapy. Deconditioning of Pears Systematic desensitization is the most widely practiced, most thoroughly researched and most influential of the currently popular behavior therapies (Weitzman, 3 1967)• Using learning theory as a rationale and direct ing itself toward the alleviation of neurotic fears, systematic desensitization lends itself to experimental analysis and evaluation in part because of its clearly specified and isolable components. It contains three j essential components: (l) The patient learns deep muscle I relaxation. (2) The patient and therapist construct an j 1 ! "anxiety hierarchy" which is a list of feared situations j | on a particular theme (such as "death" or "criticism") j graded from most to least anxiety arousing. (3) Items on the anxiety hierarchy starting with the least threaten ing ones are presented to the imagination of the relaxed 3 For details on behavior therapy methods, theories and pertinent studies and case histories see Eysenck I960; Eysenck, 1965a,* Wolpe, 1958; Wolpe & Lazarus, 1966. 22 patient.^ Since an anxiety response presumably cannot occur at the same time as a relaxation response, the latter is assumed to reciprocally inhibit the former. This recipro cal inhibition presumably spreads by means of stimulus generalization to the more threatening items on the anxiety hierarchy, thereby diminishing their "anxiety loading" and making them more amenable to reciprocal in hibition by means of relaxation. In this way progress is made from the less threatening to the more threatening items on the hierarchy. Generalization to real life situations is presumably made spontaneously. It is assumed to be essential to the efficacy of the technique that the anxiety aroused by the imagined situation be outweighed by the reciprocally inhibiting response--in this case relaxation. If the anxiety pro duced by imagining a feared situation were to be strong enough to disrupt the relaxed state of the patient it ; would be predicted that the response of anxiety would thereby be strengthened resulting in an exacerbation of j i the patient’s symptomatology. j . ^This component of systematic desensitization will, in this paper, be denoted by the term "desensitiza- tion", • this term will not be used to denote Wolpe’s sys tematic desensitization by reciprocal inhibition proced ure for which the initials "SDRI" will be employed. The term systematic desensitization or classical desensitiza tion will refer to SDRI. 23 It is of more than a little interest that there is a behavior therapy extant which has as its main object the stirring up of as much anxiety as possible at a sit ting by means of the therapist describing and elaborating upon the feared stimulus. This is the implosive therapy | i of Thomas G. Stampfl (Stampfl, 1961; Stampfl, 1967; j j Stampfl & Levis, 1965; Stampfl & Levis, 1967; Stampfl & j Levis, 1968). Stampfl has used implosive therapy with j i 1 patients from a number .of diagnostic categories (including! psychotics and alcoholics) and has reported a nearly 100$ j success rate (London, 1964). Experimental evidence of the efficacy of implosive therapy may be found in Hogan (1966), Hogan and Kirchner (1967), Kirchner and Hogan (1966), and Levis and Carrera (1967)- The procedure has features in common with Wolpin's in that neither employs relaxation or hierarchies and both appear to regard the process of imagining the feared stimulus the sine qua non of therapeutic success. Stampfl, however, attempts to generate a maximum I of anxiety in the therapeutic situation while Wolpin I simply focuses on having the patient imagine himself in | i the feared situation without regard to the amount of anxiety generated thereby. Wolpin (1968)^ seeks neither 1 to generate a great deal of anxiety (as does Stampfl) nor to prevent anxiety from being elicited (as does Wolpe). _______ 5M. Wolpin. personal communication. July 1 1 , 1 9 6 8 . i 24 Looking at implosive therapy from a Wolpeian standpoint it may be said that Stampfl, without any relax ation training begins by presenting a stimulus situation to the imagination of the patient which is at or above the $ i top of the anxiety hierarchy and watches carefully for signs of anxiety or fright from the patient* not so he can quickly withdraw the stimulus situation (as Wolpe would) j but for the purpose of obtaining cues so as to intensify j j j the patient’s fear. In Stampfl’s own words, "The greater | i i I i ' i the degree of anxiety elicited the greater the reason for j i continuing the presentation of anxiety-eliciting stimuli" i (Stampfl & Levis, 1967, P- 499)* Wolpe, it would seem would find it difficult to j f account for the success of such a therapy from the stand point of his own theory of how anxiety is reduced. Both Wolpe1s and Stampfl1s techniques, however, are based upon similar theories of learning. The basis of implosive therapy is 0. H. Mowrer’s (i960) two-factor learning 1 theory. Like Wolpe, Stampfl assumes fears are learned by means of contiguity and maintained by a conditioned instrumental avoidance response which is the neurotic symptom. For Stampfl the cure is experimental extinction of the anxiety response. In order to hasten this process high levels and prolonged periods of anxiety are desirable. Thus it may be seen that though both implosive therapy and systematic desensitization have a -plausible basis in 25 learning theory and both have achieved rather amazing rates of success* the techniques themselves are in some respects contradictory. Some of the implications of the foregoing may he summarized as follows: (!) If two techniques contain operations which contradict each other in important re spects and both purport to be derived from the same body of theory* a question needs to be raised as to the rele vance of the theory to the technique. (2) If two techniques contain operations which contradict each other in important respects and both achieve a similar (high) rate of success a question needs to be raised as to the assumed relevance of these operations to the therapeutic success of the techniques. Such questions as these have been raised by a number of authors such as Breger and McGaugh (1965)* Chomsky (1959)^ Lomont (1965) . , London (1964)* and Weitz- man (1967) as regards the relevance of the theory for the technique as well as the substantiality of the theory and Lomont (1965), Lomont and Edwards (1967), Rachman \ (1965)* Rachman (1966)* Rachman and Hodgson (1967), Wolpin and Pearsal (1965) and Wolpin and Raines (1966) as regards the degree to which various components of the techniques* such as relaxation* are essential to its success. Clearly these two issues are related since if the theory upon which the technique is purportedly based 26 Is questionable or if its relevance to the technique is questionable (such that the technique cannot be deduced from the theory) then the technique has no justification but the empirical one that it "works" (London, 1964). j i That systematic desensitization by reciprocal j inhibition (SDRI) "works" seems well established both by j clinical report (Eysenck, I960; Eysenck, 1965,* Wolpe, I 1958) and in controlled studies (Cooke, 1966; Davison, j j I 1968; Lang & Lazovik, 1963; Lazarus, 196I; Paul, 1964; Paul & Shannon, 1966). What about systematic desensitiza tion is responsible for its efficacy; which components j are essential and which superfluous is and has been a matter for speculations and experimentation. Two kinds of studies which bear upon this problem will now be briefly reviewed. Studies of the first type addressed themselves to the question of what components j or combination of components of systematic desensitization j I if any were therapeutic and what components were irrele- j ! • I vant. In reviewing these studies an attempt will be made j ! to be exhaustive. Studies which addressed themselves to ! the question of whether relaxation per se is an effective therapeutic technique will be excluded. Studies of the second type, while not addressing ■ j themselves to this problem bear upon it. Here no attempt will be made to review the entire literature on such studies. 27 First, a general statement is in order (which may help to explain why a review of the latter type of study cannot be exhaustive) to the effect that it seems to be j i typical of experimenters for purposes of control or con- j i I j venience to make ad hoc modifications in the therapeutic j i : ! ! technique with which they are experimenting. Thus the I number of studies which are relevant to our question approaches the number of studies which have been performed with systematic desensitization. It must also be pointed ! j out that Wolpe’s desensitization procedure seems to have | i withstood this rough handling by experimenters rather j well, i.e., few of the studies which modified the proce dure found significant decrements in therapeutic efficacy. j i Modifications, some of which will be mentioned in greater detail later, have included the following: Group desensitization (Lazarus, 1961; Paul & Shannon, 1966), use of common hierarchies (Lazarus, 1961 Davison, 1968) ' use of a priori hierarchy (Myerhoff, 1967; Paul, | 1966 ; Wolpin & Pearsal, 1965; Wolpin & Raines, 1966), j i I omission of hierarchical order of presentation and relaxa tion (Myerhoff, 1967; Wolpin & Raines, 1966). In addition it is the impression of this writer that experimenters 1 ! have continued to get excellent therapeutic results with j i systematic desensitization in spite of the failure to report or include such clinical niceties as history 28 taking and rapport-"building. Pre-treatment interviews have tended to be-shortened to the point of non-exis tence; hynposis--once a favored method for achieving deep relaxation has been very little-used by experimenters with systematic desensitization witti no decrement in therapeutic efficacy; and the amount of time devoted to relaxation training has grown shorter and shorter (e.g., Paul; 1966) until in some studies (Myerhoff, 19^7; Wolpin | & Raines, 1966). Relaxation has been dispensed with | altogether. j | ! j An example of the gradual abandonment of Wolpe's j 1 I technique, is the study of Rachman (1965)- This author I found that relaxation and desensitization were each inef- | I ! fective while the two combined were effective in the i i ! reduction- of fear of spiders. He notes, however, that j anxiety in each subject was not extinguished before going on to the next item on the hierarchy (as SDRI requires) with apparently no decrement in the efficacy of the sys- | tematic desensitization procedure. He remarks, "This I i procedure differs from the Wolpeian method in an important I respect. In the normal therapeutic procedure, one does not present the next item on the hierarchy until the • ' I patient reports little or no disturbance from the item j i i under consideration. For purposes of the present experi ment however this precaution was disregarded [p. 246]." Rachman (1966) pursued the question further in a 29 follow-up study in which two groups of spider-phobic subjects got all or one-half of their subjective anxiety reduced before-proceeding to the next item on the hier archy. He found no significant difference in results as | between the two groups both of which improved significant ly over a no-treatment control. Kondas (1967) compared SDRI in groups with rela xation by autogenic training in reducing "stage fright" and found the effects of relaxation weak and "transient" j 1 while SDRI produced a significant and stable fear reduc- j tion. ! Cautela (1966) reports employing a modification j 1 of Wolpe's systematic desensitization techniques in his ! I clinical practice such that the patient is often exposed to "doses" of anxiety of up to 15 seconds at a time. Prom a purely Wolpeian standpoint this should be expected to result in exacerbations of the fears of patients, yet no decrement in success ratio is reported. Cooke (1966) compared "in vivo" therapy with systematic desensitization and a no-treatment control. There was no significant difference between the treatment groups both of which improved, significantly over the j i controls. Since the in vivo group was desensitized while j standing up while the desensitization group was treated while comfortably seated and since the in vivo group was actually in the presence of the anxiety-evoking stimulus., 3° the author points out that the (non-signifleant) super iority of the in vivo group over the desensitization group on all five fear measures is evidence that relax ation, per se, is not an effective therapeutic procedure.^ By the same reasoning it is equally evident that it is | not necessary for demonstrating the efficacy of SDRI that the patient be deeply relaxed when he confronts the feared stimulus (in vivo or in imagination).. While the studies are not strictly comparable, Folkins et al. (i960) addressed themselves to the same question as Rachman (1965) viz., what are the separate ! and combined effects of relaxation and desensitization (which Polkins calls "behavior rehearsal") on the reduc tion of fear. Folkins found that the separate components of SDRI were as effective apart as in combination. Davison (1968) found the contrary; that in reduc ing fears of snakes desensitization and relaxation (SDRI) were effective together but not apart. Similarly Johnson j and Sechrest (1968) compared systematic desensitization j and progressive relaxation in treating test anxiety and found SDRI effective while relaxation alone produced no significant improvement. Zeisset (1968) however found that relaxation and systematic desensitization were ^This appears to be a fairly well-established finding though Polkins, Lawson, Opton and Lazarus (1968), Graziano and Kean (1967), and Zeisset, (1968) obtained contradictory results. 31 separately effective and equally so in reducing the "Interview anxiety" of chronic schizophrenics. In a well-designed study Lomont and Edwards (1967) compared the efficacy of two procedures in reducing I subjective and objective snake fear. Both groups of 1 I ! subjects got relaxation training. One group then com- ! bined the relaxation with visual imagery while the other j group tensed its muscles while being presented with the j feared stimulus situations. The authors were less than j cautious In their interpretation of the results since they attribute a clear superiority to the reciprocal inhibition group yet the R. I. group was superior to j the tension group at the .05 level of significance (the level presumably chosen by the authors) on only one of five measures of fear--and that one a verbal report mea sure. Though on two other measures of snake fear the | superiority of the R. I. group approached statistical : | significance--the evidence from this study., in the opinion! j of this writer, should not be interpreted as anything more j I i I than suggestive. j A study by Myerhoff (1967) had certain similar- j ; Ities to that of the one just quoted as well as contra- j dictory results. One treatment group tensed their j i muscles during desensitization; the other treatment group j I i did not. Both groups reduced their fears of snakes over the no-treatment control group with the tension group 32 improving more than the no-tension group. Neither group received relaxation training. In an experiment by Lazarus and Abramovitz (1962) j on the treatment of childrens' fears, relaxation training { was omitted. Instead they induced their child patients to imagine situations the children were fond of and in geniously introduced into these scenes at first small, ! then larger "doses" of the feared stimulus. The treatment! ; j I was found successful and interpreted by the authors on j ! 1 the basis of the reciprocally inhibiting effect of the "positive imagery" on the fear brought about by the in troduction of the feared stimulus in imagination. Rachman (1966) compared a group that received SDRI I for its spider fears with a group that received only de sensitization (SDRI minus relaxation training and hier archies) . The latter group showed no improvement while the former group showed distinct improvement. i ! j In a similar study Wolpin and Raines (1966)' treat-j | ed six women patients for their snake phobias. None re- j J ceived relaxation training. Instead two of the six were j 1 ! instructed to tense their muscles during desensitization, j i All six lost their snake phobias in- a maximum of five " j sessions. In addition the two subjects who worked "at j the top of the hierarchy" (a la Stampfl) improved as much j as those that worked with a 20 step anxiety hierarchy. Similarly, Wolpin and Pearsall (1965) trained a snake-phobic patient in relaxation for seven sessions and on the eighth session worked through an entire 20 step j anxiety hierarchy resulting in a complete cure of the j i ' | patient’s phobia. j I The contradictory results as between studies which find relaxation essential to successful deconditioning of fear and those who find it superfluous has attracted a ; 7 certain amount of recent attention.' Lazarus and Davison (in press) state, "...if the level of anxiety is very high,! | j | a premature exposure to the feared situation will pro- j bably lead to increased sensitivity. When treating | avoidance behavior maintained by relatively low levels of anxiety, however, almost any non-reinforcement contingency j may be expected to extinguish the anxiety." -j A doctoral dissertation recently completed i (Schubot, 1968) offers confirmatory evidence of this thesis. This investigator compared Wolpe’s systematic desensitization technique with desensitization (minus rela-j xation) in modifying snake phobic behavior. Both treat- j | ments were effective with SDRI showing superiority with j the more fearful subjects (those who could not approach j within five feet of the caged snake on pre-test) while 1 with the less fearful subjects both treatments were 'Reference has already been made, above, to - writers who have treated in a more general fashion the discrepant theories, procedures and results in behavior therapy. equally effective. While this thesis might hear on discrepant I I results such as that between Lang, Lazovik and Reynolds j ? (1965) and Wolpin and Raines, (1966), the former study i j l | having employed patients who presumably had higher levels of anxiety, it does not explain such contradictory findings as between that of Wolpin and Raines (1967) and j | Rachman (1965). Wilson's (1967) comment bears on this j | j j issue. This author claims that in the Wolpin and Raines j j j (1967) study subjects rehearsed accomplishment of the i t j final criterion of fear reduction while Rachman had his j | i subjects rehearse "horrific"- scenes such that they were ! | rehearsing fear responses to these situations. j Whether these kinds of discrepant results are due j to population differences as Lazarus and Davison (in press)' claim, easily remediable methodological differencesj such as Wilson (1967) claims, or other non-specific treat-j J ment factors such as demand characteristics, is a ques- j I j tion which the present study is intended to illuminate. i j Social-Psychological Effects j While the discussion thus far has focused upon certain aspects of the treatment situation known, some what nebulously, as "non-specific treatment factors" | there are a related group of variables which are claiming increasing attention. They are most often grouped together under the heading of "social-psychological 35 factors (Friedmen, 1967; Honigfeld, 1964). Particular classes of related variables have been studied under the names of "demand characteristics of the experimental sit uation" (Orne, 1959) and "experimenter bias" (Rosenthal, j 1966) as well as "Interviewer effect", (Crespl, 1948), "psychology of the scientist" (Cordovo & Ison, 1963) "situational and interpersonal variables", (Masling, i960) and "evaluation anxiety" (Rosenberg, 1965)- Their similarity to non-specific treatment effects such as the j j placebo effect and therapist expectancy may perhaps 'be i ! most clearly elucidated by two research examples, one showing the parallel between the placebo effect and I demand characteristics and the second drawing a parallel between therapist prognostic expectancies and experiment- • er bias. Demand characteristics of the experimental situa tion are defined by Orne ( . 1962) as "...the totality of j | cues which convey the hypothesis to the subject (p. 779)-"! : i If the placebo effect can be thought of as a response to an inert treatment as if it were an "active" one, the • * 1 demand characteristics effect may be considered a 1 response to an "inert" experimental procedure as if it contained the necessary conditions to produce the pheno menon. For example, Orne and Scheibe (1964) convinced a j group of subjects that they were in a sensory deprivation experiment (they were isolated for four hours but not V- . 36 sensorily deprived),, while another group— exposed to the same physical conditions was under the impression that it was a control group. The first group, reported a constel- j lation of responses (e.g.j perceptual aberrations) typi- j j cal of sensory deprivation subjects while the second j j group did not. I In the case of the parallel between the effect of j therapist prognostic expectancies and experimenter bias i the matter may be stated thus: It has been shown in the I I treatment situation that the expectation of the therapist can influence the course of treatment. Similarly it has been shown in the experimental situation that the expect ancy of the experimenter can influence the results he obtains. Rosenthal (1966) has demonstrated the experi- j menter bias effect at least a dozen times over using j variations on a procedure he first used in his Ph.D. j ! dissertation on the mechanism of projection. I In this procedure subjects are given a series of J photographs to rate as "successful" or "failure" on the j ! basis of appearance. Ratings are on a 20 point scale j i from +10 to -10. The norms of the ratings of these photos! | are such that they all have an average rating of 0. Typically there are two groups of experimenters who are the real subjects in the study. One group is led to expect average ratings from their subjects of +5 and the other group -5. Though there is no "unprogrammed" or 37 unstandardized verbal communication between subject and experimenter there have, over and over, been significant differences in the results obtained by the two groups of experimenters in accordance with their expectations. There are a good many phenomena of interest rele vant to experimenter effects which will only be mentioned here since they are not essential to the discussion. Examples are as follows: i It has been impossible thus far to determine j exactly how information is transmitted from experimenter to subject in spite of careful analysis of sound motion pictures, verbatim transcripts and other relevant data (Friedman, 1967)• The experimenter effect has been shown ! i to be capable of determining the outcome of experiments with worms (Cordovo & Ison, 1963) and rats (Rosenthal, 1966) as well as with people. And finally bias has been demonstrated to be transmitted from the experimenter to j the subjects through the experimenter’s research assist ants even though the latter were not explicitly informed of the experimental hypothesis (Rosenthal, Persinger, . Vikan-Kline & Mulray, 1963)• Martin Orne has been one of the more frequent contributors to the literature on social psychological factors in the experimental situation both by way of theoretical papers (Orne, 1959b, Orne, 1962) and experi ments (Orne, 1959a, Orne & Scheibe, 1964; Orne & Evans, 38 1965)• Generally his work has had as its aim the demon stration that experimental results may be explicable on the basis of demand characteristics--much of this work having been done with hypnosis. Barber (1964) has also had an interest in this field. For example he gathered together the relevant literature, including his own prolific contributions, to show that psychologically induced physiological effects (e.g., welts, deafness) need not be interpreted as being due to an hypnotic "state” but may be ascribed to the influence of "direct suggestion". The profound methodological implications of the research on social-psychological factors has been comment ed upon by many of the authors cited above. Perhaps a \ remark by Shand.s (i960) quoted in Goldstein (1962) and Goldstein's comment thereon would be an apt summing up: Shands, (i960) states: The measuring instrument...has been found in many important situations to exert a major influence upon the object measured; what emerges is no longer a measurement but a changed pattern of relations involving two aspects which can be separated only by definition. This realization has made it impos sible to "study nature" in any simple sense: instead it has become inevitable that one study himself studying nature (p. 217)- This, then, seems to be psychology1s version of Heisenberg's principle of indeterminacy. Goldstein's (1962) comment follows: As applied to the "therapy vs. no-therapy" experi- m e . n t . a i . . . . d . - e . s . i . g n . , . . - . t h i . s .. principle... would... translate., tha.t.___ 39 the very act of attempting to measure personality or symptomatic change in the absence of formal psychotherapy is in itself a form of non-specific therapy which, when combined with favorable j i patient prognostic expectancies, may bring about significant degrees of remission of symptomatology (p. 92). Statement of the Problem Early in the history of psychotherapy research Fiedler (1950) adduced suggestive evidence that thera pists may not know what it is about their therapy that makes it effective— that it may not be the particular theoretical framework on which the technique is ostensib- | l ly based. His pioneer investigation suggested that the j effective ingredient or ingredients may be something ] j found in the techniques of therapists of various theoreti cal allegiances. Similarly, Frank, Gliedman, Imber, Nash and Stone (1957) have suggested that since the success rate is about the same for all forms of conventional j psychotherapy the effective ingredient may be something | common to all forms of therapy. If Frank's reasoning is j correct the higher success rate of desensitization and j related therapies (e.g., those of Stampfl and Wolpin) j suggests that these therapies may contain a "specific" for irrational fear reduction. However the results of Efran and Marcia's (1967) study, discussed earlier, suggest that an "inert" therapy is capable of obtaining similar results. Perhaps, then, therapies based on an imagining process that systematical 40 ly and focally confronts the feared stimulus, (e.g., those of Wolpe, Wolpin & Stampfl) are better placebos than traditional psychotherapies. At any rate it would seem ! important to attempt to separate out the specific from ) i the non-specific components of a desensitization-like j I ! | procedure as studies such as those of Lang, Lazovik ! t I I and Reynolds (1965) and Paul (1966) discussed earlier | i j • have. i ! ! 1 5 | The therapeutic "vehicle" to be employed here is | 1 the treatment model of Wolpin. The theoretical and pro cedural affinities and differences between this model 1 ; and those of Stampfl and Wolpe have been discussed above. Wolpin1s theory of. treatment, simply stated, is that Imagining situations in which the feared stimulus appears is, per se, capable of reducing fear of that stimulus. Hypotheses Wolpin1s and Stampfl1s (Stampfl, 1961; Stampfl & ; j Levis, 1967; Woipin & Pearsal, 1965; Wolpin & Raines, 1966) theories of the treatment of irrational fears, as discussed above, would predict: j i Subjects receiving the Wolpin treatment will show j 1 fear reduction on behavioral and subjective measures. j Stated formally: j 1. Subjects visualizing fear scenes will (a.) report less fear of the feared stimulus on post-treatment than on pre-treatment tests and (b) show less avoid ance behavior of the feared stimulus on post-treat ment than pre-treatment tests. 41 Subjects receiving the Wolpin treatment will show more fear reduction than subjects receiving the placebo | therapy. Stated formally: I i ! 2. Subjects visualizing fear scenes will (a) report j more decrease in fear between pre and post-Treatment j j measures than a group receiving placebo treatment and j | (b) show more decrease in avoidance behavior between ! pre and post-treatment measures than a group receiv- • ; ing placebo treatment. j i Theories and research (Friedman, 1967; Honigfeld., j 1964; Orne, 1964; Riecken, 1962) on non-specific treatment! effects in the areas of medical, drug and psychotherapeu tic placebo effects, prognostic expectancies and social- psychological factors such as demand-characteristics of the experimental situation would predict: Subjects receiving the placebo treatment will show fear reduction on behavioral and subjective measures. Stated formally: | 1 3. Subjects receiving the placebo treatment will i (a) report less fear on post-treatment than on pre-treatment tests and (b) show less avoidance behavior on post-treatment than on pre-treatment tests. Subjects receiving the Wolpin treatment and under 1 the impression that it is a bona fide treatment will show j more fear reduction than subjects receiving the same | i i treatment but under the impression that they are in a j control group and that the procedure is ineffective. j j Stated formally: 4. Subjects visualizing fear scenes under conditions of high expectancy will show greater decrease between pre and post measures of (a.) subjective fear and (b) avoidance behavior than subjects visualizing scenes under conditions of low expectancy. CHAPTER II METHOD i I Design j j In order to assess the degree to -which the success of a fear-reduction therapy might be influenced by patient expectancy, two experimental treatment groups were compared which received a deconditioning treatment j modelled on that of Wolpin and Raines (1966) and Myerhoff j (1967). Subjects in the D (deconditioning) group were given instructions designed to induce expectancies of symptom relief while subjects in the D- (deconditioning minus expectancy of relief) group were told that they were in a control group and receiving a dummy treatment, with j i the clear implication that they were not expected to im- ! prove. Following these instructions the groups received the same treatment. Differences between these groups on j the dependent variables, therefore, would be attributable j to their differences in expectancy. j A third group, 'the P (pseudotherapy) group, was | set up to serve as a placebo control group in line with the suggestions of Rosenthal and Frank (1956). This group was given instructions designed to induce 44 expectancies of symptom relief but was administered a pseudotherapy similar to that employed by Efran and Marcia (1967). Thus any difference between this group and the deconditioning group (D group) would be attributable to the specific effect of the deconditioning procedure. In addition a small no-treatment control group (C group) was included. This group was set up in order to obtain at least a crude measure of the practice effect and not as a full-fledged control group since previous fear reduction studies had unanimously concluded that no- I treatment control groups did not improve, (e.g.,, Cooke, j 1966j Lang & Lazovik, 1963; Paul, 1966). Prospective student volunteers were administered the Pear Survey Schedule (PSS, Myerhoff, 1967). Those who indicated MUCH or VERY MUCH fear of snakes, rats or mice and who volunteered for the study began by taking a paper and pencil Pear Scale, devised by the experimenter and i | designed to provide a subjective measure of their parti cular fear. They were then pretested individually on both behavioral and subjective-measures of fear while confronting the feared animal. This completed session 1. The subjects were then assigned to treatment groups on a random basis. Session 2 was the first treatment session. Treatments were administered in groups „of from two to seven subjects. Subjects were first given their instruc tions designed to manipulate expectancies. These instruc- ■ 45 tions were identical for the D and P groups and intended to foster a positive expectancy while for the D- group j the instructions were to the effect that D- subjects were in a control group and would not be receiving a bona fide treatment. Following these instructions Questionnaire B was administered to assess the effect of the expectancy i manipulation, followed by the appropriate treatment j (deconditioning or placebo therapy). Session 3 consisted | i of another administration of the treatment as did Session j 4 which was then followed by individual posttesting which I i . employed the same behavioral and subjective measures of i fear as during pretesting. Subjects then took Question naire C to provide data on the question of whether sub jects talked about the experiment while it was still in j progress, to help assess the effectiveness of the expect- j ancy manipulation and to measure the degree of acquaint ance with and liking for the experimenter by the subject. Subjects in the no-treatment control group were given only pre and post measures of their fear of the phobic object. Procedure _Subjec_ts--Subjects were 67 undergraduate college students at the University of Southern California. Twenty-six were students in introductory psychology and received class- credit for participation in the experiment. Forty-one were students in other social science courses Bes- siai Nature of Procedure Procedure 46 Notes Screening PSS Administered in class. Sub jects with fears of snakes, rats or mice who volunteer proceed to pretesting stage. Pretesting and further screening GSB BAT SB Subjects who lift the animal are disqualified. Others are assigned to groups D Group (Decondition ing) Treatment ex pectancy in structions Que st ionnaire B Deconditioning D- Group (Deconditioning — negative ex pectancy) Control ex pectancy in structions Questionnaire B Deconditioning Deconditioning Deconditioning Posttesting: BAT GSB SB Question naire C Post- experimental interview Deconditioning Deconditioning Posttesting: BAT GSB SB Question naire C Post experiment al interview P Group (Pseudother apy) Treatment ex pectancy in structions Questionnaire B Pseudotherapy Pseudotherapy Pseudotherapy Posttesting: BAT GSB SB Question naire C Post- experimental interview 0 Group Tno-- treatment) Post testing: BAT SB Big. 1— Experimental Design 47 who were paid five dollars for their participation. 1 Fifty-two subjects were female, 15 male. Qualifications of prospective subjects--Frospec- I tive subjects were first administered a version of the ; i ( Fear Survey Schedule (FSS, Appendix A), similar to that j used by Myerhoff (1967)* a short questionnaire on which j j subjects rated their degree of fear of a number of | 2 i j commonly-feared objects. They were then read a descrip- j ! ; ] tion of the study (Appendix B) which made mention of the ! ( j fact that some subjects would be assigned to a control i i I group which would not be receiving treatment during the course of the experiment. Then those subjects who indi cated MUCH or VERY MUCH fear of snakes, rats or mice were invited to participate in the study. Only after subjects j j qualified for the study in this manner were they informed j that they would be paid for their time, if indeed this j was the case. Physical Arrangements The room used for pretesting was 17 ft. x 12 ft. j with a door at one end and a window at the other. In i i front of the window was a wooden office desk upon which j . Because of the preponderance of female subjects the feminine gender ("she") will be used when referring to a subject. 2 Exceptions were two students who responded to an advertisement for subjects in the student newspaper. These subjects were not administered the FSS. 48 j the appropriate caged animal was placed.^ Between the J desk and the door 18 in. segments of blue cloth tape were J o i t placed on the floor at intervals of 2, 4, 6, 8 and 12 feet from the cage and marked appropriately. One wall contained a one-way mirror which was draped and not in use during the experiment proper though it had been used for training the research assistant and in pilot work. The treatment room next door to the pretesting room,was of approximately the same dimensions., 17 ft. x j 11 ft. also with a window at one end and a door at the I i i other. The room was furnished with a movie screen at the window and six or seven student chairs with arms for writing in two rows facing the window. Behind the chairs., j \ at the door end of the room was a 26 in. x 42 in. table upon which stood a "Carousel" "800" slide projector and tachistoscope. Depending upon the treatment being admin- i istered the experimenter stood near the projector, opera- I | ting it, or sat at the front of the room with the tape | recorder. Fear Pretest Subjects who agreed to participate in the experi ment were given an individual appointment for the fear ^Two animals were used: a four ft. speckled king snake (for 6o subjects) and (for six subjects) a seven inch adult, black and white male hooded rat. For the snake a ten-gallon glass cage with wire mesh top and one inch of gravel on the bottom was used. For the rat a 15 in. x 10 in. wire cage with one inch of cedar shavings o . n . _ _ th.e_b_0J1t_Q.m_ M3,s_n§e d,_______________________________ 49 pretest. They were greeted in the waiting room of the U.S.C. Psychological Research and Service Center by the research assistant (RA) who was a 38-year old housewife 4 and psychology major at a local college. The research assistant had been trained to ad minister the pretest in accordance with detailed written instructions (Appendix C). She approached the subject seated in the waiting room and after verifying her name handed her the Generalized Pear Scale (GSF, Appendix D), j 1 ; a graphic rating scale designed to measure the subject's i fear of the phobic object. The RA then instructed the j 1 subject to knock on the door of the testing room after j i completing the form. The RA then returned to the testing room which was only a few steps away. She then lifted the snake and held it until the subject knocked at the door, which was usually a matter of less than a minute. The snake was handled in order to insure that it was moving since there is some indication (Myerhoff, 1967) I that a moving animal is a more fearful stimulus. She I then replaced the snake in the cage, opened the door for j the subject and went through the procedure for administer- ^Thanks are due to Mrs. Gloria Tishkoff for the j careful and intelligent way she performed her duties as j RA. The RA was unacquainted with the experimenter's ! hypotheses as well as with the groups to which the sub jects belonged on both pre and post testing. She was employed to help reduce experimenter bias in accordance with the suggestions of Breger and McGaugh (1965). 50 ing the Behavioral Avoidance Test (BAT, Appendix E) a behavioral measure of fear of the phobic animal similar I to that of Myerhoff (1967) modified from one used by : Rachman (1965). The test measures how closely a subject I will approach the feared object under a standardized set of circumstances. If the subject succeeded in picking up and holding the phobic object she was excused from par ticipation in the experiment; otherwise the subject re turned to the waiting room where she completed the Sub- i jective Pear Scale {SP, Appendix P) in which, on a graphic I rating scale, she rated her fear at the closest point to j the phobic object. She then completed a form showing the times she had available for participation in the experi- ' f I ment and was told that she would be contacted by tele phone. The subject was then assigned to a group by means i of a predetermined random schedule. Twenty-two subjects I I were originally assigned to each of the D, D- and P j J groups, and six subjects to the no-treatment control ! | | group (C). Dropouts during the experiment reduced these j i ! group N's to 20, 20, 21 and 6 respectively. i i Treatments were conducted in groups when possible, though in two or three instances they were conducted i i individually. The groups ranged in size from two to I seven persons but averaged three or four. The Expectancy Manipulation After the first session, devoted to the pretest, 51 subjects assigned to a particular group and having In common a particular time available were contacted by telephone and an appointment was made for them to return to the Psychological Research and Service Center (PRSC). Subsequent appointments were ordinarily made at the con clusion of each visit. Subjects assigned to a particular treatment were summoned from the waiting room by the experimenter (the author) and escorted to the treatment room adjacent to the testing room. They were assured that there were no animals of any kind in the room and told to take any seat.! I The expectancy manipulation was then accomplished by meansj of differential instructions. Group D- received control j group expectancy instructions (Appendix G) in which they were told that they had been assigned to a control group and would be receiving a "dummy treatment" instead of the regular treatment. Groups D and P both received identical! treatment group instructions (Appendix H) in which they were told that they would be receiving a treatment which,, though it was being refined by research, had already achieved "promising results". Subjects were then admin istered Questionnaire B (Appendix I) in order to assess the effect of the expectancy instructions by requiring the subjects to estimate how close they would be able to approach the phobic object and how afraid they would be of it. after treatment was completed. 52 The Deconditioning Treatment The D and D- groups were hoth treated in exactly the same way aside from the instructions read to them. They received a treatment along the lines of that of Wolpin and Raines (1966) and Myerhoff (1967). According i to the principles of this treatment., no relaxation train- | ! j ing is given. The subjects are simply instructed in the | ! | detailed visualization of a scene involving the feared | i object. They were first given practice imagining a short j neutral scene (Appendix J), and then were told, "Now you will be imagining a scene involving the animal you saw in his cage during your pretest in the adjoining room." The subjects then heard the Pear Scene (Appendix K) presented by means of a tape recording of the experimenter1s voice repeating the•scene ten times in each session. Since there were three sessions the scene was presented to the i imagination of the subject a total of 30 times. Each scene took about 90 seconds to present with appropriate i I I ! pauses to allow subjects to visualize each element in the j scene clearly. The to5al time for the treatment part of I each session was about 15 minutes. During presentation of the Pear Scene the experi menter sat at the front of the room beside the tape i recorder obliquely facing the subjects and avoiding eye contact with them. After the last scene in any treatment session was presented subjects were told, "That ends this 53 session, I'll see you again ______ at P.M." Appointments were ordinarily one week apart. Each sub ject was seen a total of four times--the first time for pretesting. After the third treatment session, the | fourth and last session all told, subjects were told to j wait in the waiting room where they were then contacted and individually posttested by the HA. i i The Placebo Therapy | | The placebo therapy, based on that of Efran and Marcia (1967), consisted of high-speed tachistoscopic projections of slides of travel scenes and slides of additional problems. There were 80 of each, mixed toge ther in no particular order. The arithmetic problems were barely perceptible and the subjects were instructed to write down the correct sums on a worksheet (Appendix L) given them for that purpose. The problems consisted 27 of pairs of two digit and one digit numbers, e.g., 4, 88 ~ The content of the travel scenes could not be identified. The P group subjects were given an elaborate rationale for the treatment (Appendix M) to the effect that they were seeing "subliminal" fear stimuli and that the purpose of working the arithmetic problems was to "reciprocally inhibit" the fear responses brought about by the fear stimuli which operated unconsciously. The sequence of operations for the subjects in 54 the placebo group was as follows: In their first treat ment session they received their expectancy instructions and completed Questionnaire B. They were then given the rationale for the treatment and handed a copy of the work- j sheet. After handing out the worksheet and pencils to i those who required them the experimenter went to the back of the room and set the slide projector in motion. One hundred and sixty slides were then projected on the j j screen, at the rate of one slide every five seconds. The j i total time for administration of this treatment averaged j 134- minutes. Slides were projected automatically at the i five second interval by means of a Kodak ''800" Carousel slide projector. Two separate trays which hold 80 slides each were used. Tachistoscopic exposures of 1/75 of a second with the diaphragm halfway closed (set at 34 or "dim") projected the slides such that the arithmetic slides could be just barely perceived while the travel ("fear") slides could not be identified although varying j I j patterns could be distinguished. The lights in the j I i treatment room were left on. j The completed worksheets were left by the subjects j with the experimenterj subjects were under the impression j i i that these were scored. As in the deconditioning treat- ; I mentj the placebo treatment was given in three sessions-- at the completion of the third treatment session the sub jects were asked to take a seat in the waiting room. 55 Posttesting Posttesting was conducted -toy the RA In much the i same manner as pretesting and In accordance with detailed written instructions., (Appendix N). She contacted each subject in the waiting room, had the subject follow her j i to the testing room where she administered the Behavioral j Avoidance Test (BAT) in the same manner as she did on S pretesting. Again the subject was asked to approach the phobic object as closely as she could and again she was j | asked to return to the waiting room where she filled out ; I | the Subjective Pear Scale (SF, Closest). This was a | rating from the subject of her fear of the phobic object j ! at the closest point she was able to approach it. In i ! addition, if there was a pre-post difference, she was j asked to return to the testing room and stand at the same place she stood at the closest she could approach the I phobic object on the occasion of her pretest. She then j ' r returned to the waiting room and filled out another SP scale (SP, Same). She was then given Questionnaire C (Appendix 0) to complete which asked the subject to rate her fear of the phobic object, a repeat of the Generalized Pear Scale (GSP) administered before treatment. It also asked questions relating to the efficacy, of the expectancy manipulation, the confidentiality of the content of the | experiment and the degree to which the subject was acquainted with the experimenter prior to the experiment. 56 In addition, on a separate sheet of paper the subject was asked to rate her degree of liking for the experi menter. For this, the subj-eet was assured that this rating was necessary for scientific purposes and would not be associated with her name. On its completion by the subject, Questionnaire C, with the Experimenter Rating Scale (page 3) detached, | was given to the experimenter by the RA. The experimenter! then held a post-experimental interview and debriefing i with each subject individually. These sessions were in- j formal in nature and the format (Appendices P, Q and R) differed somewhat depending upon the group to which the subject belonged. The purpose of this interview was to | inform subjects of the true nature of the experiment, to allow them an opportunity for questioning and/or ventila tion, and to help the experimenter evaluate the effective- j i f ness of the experimental manipulations and to help him | understand more fully how the experiment appeared to the I i subjects. | Summary of Dependent Variables The dependent variables employed in the study | were: j Behavioral Avoidance Test (BAT), (Appendix E), j a behavioral measure. I Generalized Subjective Fear Scale (GSF), (Appendix D), a measure of subjective fear of the phobic 57 object- in general. Subjective Pear Scale (SP) (Appendix P), a measure of subjective fear of the phobic object at the closest point the subject was able to approach the phobic i / i object, (SP, Closest). The same scale, when the subject j i i rates her fear on the posttest at the same point in space j as she rated it on the pretest is termed (SP, Same). j CHAPTER III RESULTS Pre-Treatment Equality of Groups In order to assess whether the random assignment of subjects to groups resulted in groups that were in fact equated on the criterial measures,, one-way analyses of variance of the BAT* GSF and SP scores were performed (see Table 1). These disclosed no significant pre-treat- ment differences between the groups on any of the var iables. Thus it was concluded that the treatment groups did not differ on relevant variables before treatment. Experimental Hypotheses Por a summary of experimental results see Table 2 In order to determine whether the groups that received the actual treatments (D and D-) showed fear reduction, t tests were performed on the appropriate behavioral and subjective measures. The obtained _t's for BAT, GSP, SP (Same), and SP (Closest), respectively, were 6.24, 7*32, 4.46, and 3*91 all significant at the ,001 level and with 39 hf in each case. Thus Hypothesis (a and b), which predicted that subjects receiving Wolpin's deconditioning treatment would show fear 58 59 TABLE 1 j I I SUMMARY OF ANALYSES OF VARIANCE OF PRE-SCORES ON THE BAT, GSF AND SF SCALES FOR ALL GROUPS Source df Analyses of MS Variance F P BAT Between groups 2 3-11 .58 NS Scale Within groups 55 5-37 GSF Between groups 2 12.50 00 0 NS Scale Within groups 58 447.10 ft CO Between groups 2 781.50 1.24 NS Scale Within groups 58 630.83 60 TABLE 2 SUMMARY OF PERFORMANCE OF GROUPS ON THE CRITERIAL MEASURES OF FEAR REDUCTION Measure Pretest Posttest j Difference j Gp. N Mean S.D. Mean S.D.' Mean S.D. BAT D 20 4.75 2.32 2.35 2.00 2.40 1.63 D- 20 5.00 2.39 4.10 2.81 .90 1.30 P 21 4.24 2.07 3.19 1.83 1.05 1.40 C 6 6.16 3.29 4.83 4.14 1.33 1.11 GSF D 20 75.85 24-. 00' 25.60 23.71 50.25 31.70 D- 21 75.70 20.58 53.30 30.82 22.40 22.97 P 21 ■ = f 1 - 1 c- 16.82 46.71 27.94 30.43 25.14 C 6 75-00 22.29 SF D 20 67.70 26.07 24.70 24.50 43.00 30.64 (Same) D- 20 55.60 26.46 52.00 29.19 3.60 20.48 P '21 64.33 20.70 35 • 95 31.25 28.38 35.22 C ' 4 65.00 36.86 31.50 32.51 43.50 31.73 SF D 20 67.70 26.07 32.20 30.94 35.50 33-35 (Clos est) D- 20 55.60 26.46 50.90 28.56 4.70 21.69 P 21 64.33 20.70 39-95 30.10 24.38 31.54 C 5 65.OO 36.86 48.80 31.15 11.20 38.61 *Due to an oversight the GSF posttest was not administered to the C group. 6l reduction on behavioral and subjective measures of fear of the phobic object, was confirmed. In order to compare the performance of the groups that received the actual treatment (D and D-) with the placebo therapy group (P), t_ tests were performed. The obtained _t's for BAT* GSF, SF (Same), and SF (Closest), respectively were 1.4l, .74, .56, and .49 all nonsignifl- i | cant and all with 60 _df. Thus there were no significant ! differences in improvement on behavioral or subjective 1 j measures of fear reduction between the actual treatment j and placebo groups. Thus Hypothesis 2 (a and b) which predicted that subjects receiving Wolpin’s deconditioning treatment would show greater fear reduction than the placebo group on behavioral and subjective measures of fear of the phobic object was not confirmed. In order to determine whether the group that received the placebo therapy (P) showed fear reduction on behavioral and subjective measures, _t tests were per formed (see Table 3)- The obtained values for _t were significant at the .01 level or beyond on all measures. Thus Hypotheses 3 (a and b) which predicted that subjects receiving the placebo therapy would show fear reduction on behavioral and subjective measures of fear of the 1 phobic object was confirmed. In order to compare the performance of the group that received the deconditioning treatment under condi- 62 TABLE 3 SUMMARY OP SINGLE MEAN t TESTS OP GROUP MEANS ON THE CRITERIAL MEASURES OF PEAR REDUCTION Meas. Group Mean SD t df p BAT D 2.40 1.63 20.34 19 .001 D- • 90 1.30 3.02 19 .01 P 1.05 1.40 3.37 20 .01 C 1.33 1.11 2.69 5 .05 GSP D 50.25 31.70 6.86 19 .001 D- 22.40 22.97 4.25 19 .001 P 30.43 25.14 5-41 20 .001 C n o t a dminis ter e d SP D 43.00 30,64 6.12 19 .001 (Same) D- 3.60 20.48 .77 19 N.S. P • 28.38 35-22 3.43 20 .01 C 43.50 31.73 2-37 3 N. S. SP D 35.50 33-35 4.64 19 ,001 (Closest) D- 4.70 21.69 .95 19 N.S. P 24.38 31.54 3-46 20 .01 C 11.20 38.61 • 58 4 N.S. 63 tions of high expectancy (D) with the group that received the same treatment under conditions of low expectancy (D-) Duncan's New Multiple Range Tests were performed (see i 1 ’ ! S Table 4). The obtained values were significant at the .01 j I level or beyond on all measures. Thus Hypothesis 4 (a | i and b) which predicted that subjects receiving Wolpin's ■ deconditioning treatment under the impression that it was i a bona-fide treatment would show greater fear reduction ! than subjects receiving the same treatment but under the I impression that they were in a control group and that the procedure was ineffective was confirmed. Performance of Practice Effects Group (c) In order to determine whether the group that i i i received pre and post testing only (C group) showed fear j reduction on behavioral and subjective measures t tests j were performed (see Table 3)* There was a significant ! decrease in avoidance behavior (j? > .05) but’subjective j X j fear measures showed no significant reduction. j Treatment Comparisons j In order to test for differences among groups ! following the application of the various treatments, one way analyses of variance were performed on the four - j critical measures (see Table 5)- All four measures ■'"Due to an oversight the GSF scale was not admin istered to the C group subjects at the time of the post test. Therefore difference scores on this measure are not available for the C group. 64 TABLE 4 SUMMARY OF DUNCAN'S NEW MULTIPLE RANGE TESTS OF DIFFERENCES BETWEEN PAIRS OF TREATMENT MEANS ON THE CRITERIAL MEASURES OF FEAR REDUCTION i Behavioral Avoidance Test (BAT) , Treatment Group Mean D D- P D 2.40 0.00 1.50** 1•35** D- .90 0.00 - .15 P 1.05 0.00 ■ Generalized Fear Scale (GSF) D 50.25 0.00 27.85** 19.82* D- 22.40 0.00 - 8.03 P 30.43 0.00 I Subjective Fear Scale (Same point as pretest), (SF, Same) j D 43.00 0.00 39-40***14.62 D- 3.60 0.00 -24.78* P 28.38 0.00 Subjective Fear Scale (Closest Point), (SF, Closest) D 35.50 0.00 30.80** 11.12 D- 4.70 . 0.00 -19.68* P 24.38 0.00 *p> .05 **p > .01 ***p > .001 65 TABLE 5 SUMMARY OP ANALYSES OP VARIANCE OP THE PERFORMANCE GROUPS ON THE CRITERIAL MEASURES OP PEAR REDUCTION Source df Analyses MS 1 of* Variance F p BAT Between Groups 2 13-60 6.18 .01 Within Groups 58 2.20 ! ! ' GSF Between Groups 2 4117.50 5-44 .01 Within Groups 58 757.. 46 SF Between Groups 2 7939.50 8.65 .01 (Same) Within Groups 58 917.52 SP Between Groups 2 4869.50 5.38 (Closest) Within Groups 58 905.79 66 distinguished among treatment groups at the .01 level of significance leading to the conclusion that the three treatments had' produced significantly different levels of fear reduction. i i j Comparisons Between Treatment Means j ! j In order to determine which pairs of group means \ i j differed significantly, Duncan1s New Multiple Range tests i were performed (see Table 4). On all criterial measures | improvement was greatest for the D-group followed by the I i P group and then the D- group. The D group improved significantly more than the D- group on each of the four ] measures while improving significantly more than the P group on the behavioral measure (BAT) and one of the sub jective measures. As far as comparisons between the P and D- group are concerned} improvement for the P group exceeded Improvement for the D- group only on the SP measures (same and closest point). Treatment Effects i In order to determine whether improvement on the criterial measures of fear reduction for each group was significantly greater than that which might have occurred by chance single mean _t tests were performed (see Table 3)* Pear reduction was greatest on all criterial measures for the D group whose improvement departed from chance expec- j tation at the .001 level on all four measures. The P group also showed gains on the criterial measures at the 67 .01 level or better on all four measures. For the D- group, Improvement was significant on the BAT and GSF ! scales but not on the two SF scales. j j Effects of the Expectancy Manipulation j i ! As a check on whether expectation scores for the j D and P groups could be treated as having come from the ! i same population, _t tests were performed. There were no j j significant differences between the two groups on any of j the measures. The obtained _t1s for BAT, GSF and SF, j respectively, were .30, (38df), NS; .20, (36 df), NS; and j 1 l •78, (38 df), NS. Therefore it was concluded that expec tation scores for the D and P groups could be considered as having come from the same population and combined. In order to determine whether the expectancy mani pulation resulted in different expectations in the treat ment groups, t tests were performed. The data for the D j i and P groups, both groups having received positive expect- ! ancy instructions, were combined and compared on expect ancy measures with the D- group which had received nega- I tive expectancy instructions. As was anticipated, the combined D and P groups expected more improvement than the D- group on all three criterial measures, though on only I one measure (GSF) was the difference statistically signifH | 1 cant. The obtained _tfs for BAT, GSF and SF respectively > were 1.49, (58 df), NS; 2.31, (56 df), > .05; 1.91, (58 df), NS. 68 In order to determine whether there was a differ ence in expectancy of improvement between the D and D- groups _ f c tests were performed. There were no significant differences between the two groups on any of the measures though all differences were in the hypothesized direction. The obtained _t' s for BAT., GSF and SF respectively were | 1.07, (37 df), NS; 1.14 (35 df), NS; and .72 (37 df), NS. ; | In order to determine the extent of the relation- | ship between expected and obtained scores on the criterialj measures, partial product moment correlation coefficients were computed for each of the criterial measures, eliminat ing the influence of the pretest score (see Table 6). The partial correlation coefficients for BAT, GSF and SF (Same) ! repectively were .60, .41 and .53, all significant at the . 0 1 level. This findings is taken to indicate that there was a strong relationship between the subject's expecta tion and her actual performance, j The Liking for the Experimenter Measure' i " “ “ r ■ ' ”'T " • 1 1 — 1 '• j 1 I In order to compare the treatment groups on the j ] degree to which they rated themselves as ”liking” the j ! experimenter, a one-way analysis of variance was performed] (see Table 7)• The obtained F = 5-14, was significant at I the .01 level suggesting that subjects in the different j groups liked the experimenter to differing degrees. ! In order to determine which,pairs of group means differed significantly, _t tests were performed (Table 8). 69 TABLE 6 PARTIAL PRODUCT-MOMENT CORRELATION COEFFICIENTS FOR EXPECTED VS. ACTUAL IMPROVEMENT ON THE CRITERIAL MEASURES--WITH THE INFLUENCE OF PRETEST SCORES ELIMINATED Measure Group i Partial Correlation i i BAT D .70** 1 D- .41 ! P .59** i All Groups .60** | GSF D ■ .27 D- .29 ' P .62** | All Groups .41** | SF D .44 j (Same) D- .62** ! P .64** | All Groups ■53** | ! *P > .05 **P > .01 ***p 7 .001 70 TABLE 7 SUMMARY OP ANALYSIS OF VARIANCE OP "LIKING FOR EXPERIMENTER" SCORES FOR ALL GROUPS Analysis of Variance ; Source df MS F P | Between Groups 2 797 5-14 .01 Within Groups 58 155 71 TABLE 8 SUMMARY OP t TESTS OP DIPPERENCES BETWEEN PAIRS OP TREATMENT*-MEANS ON THE LIKING POR EXPERIMENTER SCALE Treatment Group Mean SD D D- 1 P 1 “ i i D 55.05 12.39 0.00 2.30* 1 1.00 ; D- 46.90 10.34 0.00 3.21** j i P 59.19 13.44 0.00 1 f f f | I I j i j 72 While there was no significant difference between the D and P groups on this measure, both groups scored signifi cantly higher than the D- group, at the .05 and .01 levels respectively. Since the D and P groups received differing treatments but the same positive expectancy while the D- group received the same treatment as the D group but with a negative expectancy the results suggest that "liking for the experimenter" was a function of expectancy rather than treatment and offers further evid ence of the success of the expectancy manipulation. In order to determine whether improvement on the criterial measures of fear reduction was associated with scores on the Liking for the Experimenter scale, product- moment correlation coefficients were computed. For the BAT, GSF, SF (Same), and SF (Closest) measures the r's were, respectively, .07, -03, .01 and .15 j none of which were statistically significant. Therefore it was con cluded that there was no relationship between improvement on the criterial measures and liking for the experimenter. In order to determine whether expectation of improvement was associated with scores on the Liking for the Experimenter Scale, product-moment correlation co- j efficients were computed. For the BAT, GSF, and SF meas- l j ures the r's were, respectively, .11, .21, and .13j none j of which were statistically significant. Therefore it was concluded that there was no relationship between expecta 73 tion of improvement on the criterial measures and liking for the experimenter. Paid Vs. Unpaid Subjects In order to determine whether paid and unpaid j i j subjects performed differently on criterial measures of j I fear reduction* _t tests across all groups were performed ! I (see Table 9)- These disclosed no significant differences; i between paid and unpaid subjects on any of the criterial j i measures. Thus it was concluded that paid and unpaid j subjects could be treated as having come from the same | i population. Male Vs. Female Subjects In order to determine whether male and female subjects performed differently on criterial measures of fear reduction* _t tests across all groups were performed (see Table 10). These disclosed no significant differ ences between males and females on any of the criterial measures. Thus it was concluded that male and female subjects could be treated as having come from the same population. TABLE 9 SUMMARY OP t TESTS OP DIFFERENCES BETWEEN PAID AND UNPAID GfKOUP MEANS ON THE CRITERIAL MEASURES OP PEAR REDUCTION Measure Group Mean SD t df P BAT Paid Unpaid 1.51 1.31 1.71 1.26 • 52 66 NS GSP Paid Unpaid 33-46 35.58 30.80 23.37 • 27 60 NS SF (Same) Paid Unpaid 25.33 27.56 42.43 31.87 .26 64 NS SP (Closest) Paid Unpaid 20.24 21.68 35.18 29.90 .17 65 NS 75 TABLE 10 SUMMARY OF t TESTS OF DIFFERENCES BETWEEN MALE AND FEMALE GROUP MEANS ON THE CRITERIAL MEASURES OF FEAR REDUCTION Measure Group Mean SD t df P BAT Male Female 1.80 1.33 1.51 I.56 1.02 66 NS GSF Male Female 33-69 34.46 33.70 27.91 .08 60 NS SF (Same) Male Female 33.29 24.24 34.33 33.43 .87 64 NS SF (Closest) Male Female 19-13 21.27 38.29 30.73 .22 65 NS i I CHAPTER IV DISCUSSION The results of this experiment, In the main, confirmed the hypotheses as stated. First, as hypothesiz ed, subjects who received the deconditioning treatment improved on both subjective and behavioral measures of j fear of the phobic animal. Second, as hypothesized, sub jects who received the placebo treatment likewise im proved on these measures. Contrary to hypothesis, the difference that had been anticipated between the decondi tioning and placebo groups on these measures did not materialize. Finally, as hypothesized, subjects who received the deconditioning treatment and were told it was a bona-fide treatment improved more on behavioral and subjective measures than those subjects who received the same treatment but were told they were in a control group. The foregoing results do not take into account the strong practice effects which on the basis of pre vious studies were not anticipated when the hypotheses were drawn up. Findings relating to the formal experi mental hypotheses will be elaborated upon after the discussion of practice effects which follows. 76 77 Practice Effects Previous research in the area of fear reduction by means of desensitization and similar procedures, | (e.g., Cooke, 1966a, Davison, 1968; Lang & Lazovik, 1963; \ I | I Lang, Lazovik & Reynolds, 1965; Paul, 1966) has been | j unanimous in finding no significant fear reduction in no- ; treatment control groups. For this reason, and because j a placebo group was included for purposes of comparison, j a no-treatment group was not originally a part of the j experimental design. During pretesting however it was | I found that a much larger proportion of allegedly fearful subjects than had been expected were able to lift and hold the feared animal. These results could be interpre ted as suggesting that there was something in the testing situation which was causing subjects to behave differently] than found in previous experiments (Myerhoff, 1967; Wolpin 1968).^ Therefore a small (n = 6) no-treatment group was added to serve as a rough gauge of practice effects. Con trary to previous findings this group improved on the j behavioral measures of fear reduction, although consistent j i with previous studies, it showed no significant improve- I ment on the subjective measures. The unanimity with which similar studies have found no significant practice effect, together with considerations to be discussed in"the next section, lead j _______ ' S / I . Wolpin, personal communication, 2-IQ-68._______ 78 the writer to the judgment that the significant practice | effect found in the present study may safely he regarded | as■artifactual in nature. s Explanation of Strength of Practice Effect | I I | I II I ! - > - _■■ I .■ I n I I I mnm, > 1 1 ■ ■ I I I . > 1 I I a | I ■ | . , 1 ■ I I ■ ■ ! | Three causes which could have operated separately | | i ! or together might plausibly have accounted for the j j • i unusually strong practice effects: ! i . . ! | (1) Two subjects, both males, were recruited by i ! ! j means of an advertisement in the student newspaper which j offered treatment to student volunteers with fears of snakes or rats. By chance both subjects were assigned to i the no-treatment control group. Thus two of the six sub jects in the no-treatment control group came from a differ ent population from the rest of the subjects. These sub jects may have been atypical in that they were the only two respondents to an advertisement that presumably thousands of students saw, and they were the only two sub jects who volunteered for the study before they knew there would be a reimbursement of some kind (monetary or course credit.) Therefore it may be assumed that they were more ! interested than the main body of subjects in being "cured" of their fear of the phobic animal. t These two subjects, furthermore, represent half [ i (two of four) of the control group subjects who improved since two of the six did not .improve at all. Even more striking, these two "atypical" subjects improved 5 BAT 79 points between them while the total improvement for the no-treatment group was 8 points. Their average improve ment was 2.50 points while the average for the other four subjects was .75, about the same as the .7^ averaged by | Myerhoff's (1967) no-treatment control group. i ! | It is the author's opinion, based upon the fore- ' | going that the presence of this atypical pair of subjects • in the small no-treatment control group is the most likely j j explanation for the atypically high improvement score of i this group. Why these two subjects should have improved j | more than the others can only be a matter of speculation at this point. One possibility is that they were more highly motivated to approach the feared stimulus, and their greater improvement is due to higher motivation. Their pretest scores indicate that in spite of their desire to rid themselves of their phobias they were no more fearful than the average subject. What is clear and what is not speculation is that these subjects were different in some j ways from the other subjects, that both of them, by j i chance, were placed in the same group in which they made I up one-third of the subjects and that they improved much j more than the other subjects in that group. j I I (2) The present study differed from all studies but Davison's (1968) and Myerhoff's (1967) in that a research assistant was employed to administer pre and post testing. The RA was uninformed as to the experiment- 80 al hypotheses as well as to the group to which the subject belonged. It may be, therefore, that the lack of practice effect found in other studies is due to experimenter bias. Weakening this explanation somewhat is the fact that Davison (1968) and Myerhoff (1967) used RAs and got no significant practice effects. It should be noted how ever that Davison had only four subjects in his no-treat ment control group. Myerhoff reports that, due to an oversight in the carrying out of the design, his RA could have known the groups to which the subjects belonged. However evidence is presented that the RA did not, in fact, have such knowledge. (3) There is evidence, alluded to above, that in the present study subjects found it easier to approach the feared stimulus during pretesting than has been found in previous studies. Myerhoff (1967), drawing from the. same population of subjects as in the present study, found i 6 subjects out of 84 or 7$ were willing to handle the j presumably fearful stimulus during pretesting while In the ! present study 24 out of 103 or 23$ of the subjects were ! willing to do so. The difference in percentages is significant at the .001 level and suggests that the pre testing situation in the present experiment was less j frightening, that the subjects felt more impelled to over come their fear, and/or that the situation was more therapeutic. When the two testing situations, the present 81 one and Myerhoff’s are compared, the following three differences emerge: (a) Most importantly, in the author’s view, is that in the present study the RA was a warm, friendly, attractive and motherly Jewish woman (the mother of five) while in Myerhoff's study the RA was a male i I i (nothing more was said about him). (b) Tending to make j the present study's testing situation more fearful was j that the snake used in the present study was a much larger and more active snake. (c) Tending to make Myerhoff's (1967) testing situation more fearful was that i one of the animals used by him was a California tarantula ! i I t which is probably least likely of the animals used to be | i i picked up. j What is being suggested here, In short, is that the practice effects found in the present study may be due | to what might be called a "Jewish Mother Effect"--an j atmosphere of testing more comfortable, more therapeutic, I ! less fearful and more full of implicit demands for s improvement than that present in most other testing situa- j j tions. Any such effect would have ample opportunity to operate since administration of the BAT included a standard degree of urging and assuring the subjects. Many of the students felt pressure upon them to approach the phobic object and some even seemed to feel that the testing sit uation was a part of the treatment that had been promised them. 82 Practice Effects as a Reliable Phenomenon Research with in vivo desensitization (e.g., Cooke, 1966b) indicates that under the proper circumstances confrontation with the feared stimulus may result in fear reduction. The practice effects found in this study can probably be understood as resulting from high motivation j on the part of the subjects combined with contact with j the feared stimulus in a therapeutic (relaxing) environ ment . Comparisons Between Treatment Groups ! " ‘ \ Although a difference between the D and D- groups i was predicted, the magnitude of that difference, which must be attributed t.o the expectancy variable, was unex pectedly large. This result lends credence to the central thesis of this study, viz., a large proportion of the j improvement brought about by a psychotherapy method can j j be accounted for by the expectancy of the patient. If | this expectancy is lacking, the treatment will lose much I of its effectiveness. j The D group is also significantly superior to the I P group on the behavioral measure as well as on one of the j three self-report measures of fear reduction. On the i remaining two self-report measures the D group is likewise superior, though not significantly so. Nevertheless the evidence appears to indicate that the Wolpin decondition ing treatment with positive expectancy is more effective 83 than positive expectancy alone both behaviorally and, less conclusively, subjectively. The behavioral-subjective difference may indicate that expectancy has a greater effect upon subjective than behavioral measures. This is consistent with prior research in the area of non-specific treatment effects. There is considerable evidence that placebos affect sub- J jective or experiential variables more than physical j or physiological ones (Beecher, 1956; Beecher, 1962; Honigfeld, 1964). The superiority of the P group over the i D- group on two out of three subjective measures of fear reduction, though not on the behavioral measure, may similarly be understood as due to the greater impact of ■expectancy upon subjective than behavioral measures. To j rephrase the matter: receiving the deconditioning treat ment under no matter what expectancy tends to affect j behavior proportionately more than subjective feelings, j while the opposite is true of the placebo therapy. 1 If the claims of behavior therapists are correct, . i that behavior therapy affects behavior more than tradition-; al psychotherapy does while if the latter is effective at all, it is mostly with subjective feelings, ("I have the same problems, but I feel better about them,") then it would follow, although admittedly the chain of logic is not without weaknesses, that traditional psychotherapy contains more "placebo" ingredients than behavior therapy, 84 or to put it the other way around, behavior therapy contains more "specific" ingredients than traditional psychotherapy. If there is any truth to this notion it would suggest that augmenting the placebo and other non specific treatment qualities of behavior therapy would j have a great impact upon its subjective effectiveness as j well as a lesser impact upon its influence on behavior. Motivation for Therapy The positive relationship between motivation and success in therapy is a generally accepted finding (Goldstein, 1962). In the present study the variable of motivation was uncontrolled. Therefore differences in improvement between the groups may be due to pre-treatment | differences in motivation for therapy. This possibility is deemed unlikely because of the random assignment of subjects to groups but the ubiquitous nature of the j motivation variable and its relationship to both expect ancy of improvement and actual improvement make it | desirable that it be measured in future studies of this j nature. ; I The Wolpin Deconditioning Treatment In contrast to the manner in which the experimental hypotheses were framed, for the purpose of discussing treatment effectiveness only the D group will be considered as having received the Wolpin deconditioning j i treatment. It was on the basis of both the D and D- j 85 groups receiving instruction in visualizing fear scenes that when the experimental hypotheses were formulated it was decided to combine the scores of subjects in both groups to determine the effect of the deconditioning treatment. However the great difference In performance between the D and D- groups suggests that combining them was inappropriate. In the ordinary course of events patients receiving the Wolpin treatment would have the impression that it is a bona-fide treatment. Furthermore tests of hypotheses of treatment effectiveness have never, to the writer's knowledge, been conducted with subjects under the impression that they are in a control group and receiving a dummy treatment. given under conditions which, on a priori grounds, might be expected to attenuate its effectiveness. The treatment i was administered in groups and by means of a tape record- ! I ing rather than by the therapist directly. Moreover, the j fear scene was phrased generally enough to include any j small animal. In spite of these possibly limiting condi- j tions the degree of improvement was comparable to that found by Myerhoff (1967) who read his subjects fear scenes fitted to their particular fear and at a rate of be done without argues for both the effectiveness and the trdiness of the technique. It should be noted that the Wolpin treatment was speed suited to their progress. That these niceties could It should also be noted that Wolpe's desensitiza- tlon technique has also been successfully administered in groups (e.g., Kondas, 1967; Lazarus, 1961; Paul & Shannon, 1966; Ritter, 1968) and by means of recordings (Kahn & Baker, 1968; Migler & Wolpe, 1967). Stampfl's technique, too, (Hogan & Kirchner, 1967) has been employed in groups and administered by means of tape recordings with impres sive results. The readiness with which these techniques appear to lend themselves to simplification, automation and self-administration seems a point very much in their favor. It is the author1s impression that a treatment effect as large as the one obtained in the present study could have been obtained with two instead of three treat- I ment sessions. Even one session would probably have resulted in a significant treatment effect. This con jecture is based on the following, admittedly impressional,| evidence: In the first treatment session D group subjects seemed alert and interested. In the second and third j sessions, however, particularly in the later stages of j those sessions, subjects showed signs of boredom, sleepi- j i ness and lack of attentiveness. On post-experimental interview and questionnaire report, furthermore, a number of subjects complained of the boredom and monotony of the sessions. 87 The Placebo Therapy The placebo group, it will be remembered, received a pseudotherapy modelled upon that of Efran and Marcia (1967). The effectiveness of the pseudotreatment was comparable to that obtained by those authors. The improvement of the P group on all criterial measures provides additional evidence of Efran and Marcia's (1967) contention that, "subjects will respond to treatment involving expectancy alone [p. 240J." However this writer would take exception to those authors' position that, "subjects' expectations during and following treat ment may constitute the crucial factor in achieving successful results (p. 23p)." The evidence provided by the present study in which the D group was superior to the P group on the behavioral measure and one of the self- report measures while there was no difference between the P and D- groups on those measures points to two crucial ingredients in a successful treatment. These may be conceptualized as a specific ingredient, i.e., an effect ive therapy and non-specific ingredients such-as expect ancies and demand characteristics of the experimental situation. In further support of this view reference is made to drug placebo research (e.g., Dinnerstein, Lowenthal & Blitz, 1966) in which drug effects and placebo effects appear to interact such that meprobamate, for example, is 88 no more effective than placebo except when it is adminis tered by physicians who "believe" in it. Then it is superior to placebo, even when the physicians don1t know which is which’ . (Fisher, Cole, Rickels & Uhlenhuth, 1964.) In other words factors such as expectancy of improvement■ would appear to be necessary in order to "bring out" the specific factors in a treatment. It has already been mentioned that the fear reduction achieved by the P group in the present study was comparable to that found by Efran and Marcia (1967). On subjective measures of fear reduction, however, Efran and Marcia found no effect. This result is surprising since the present study as well as the drug placebo literature discussed previously suggest that expectancy affects subjective measures more than behavioral measures. The discrepancy in results as between Efran and j Marcia (1967) and the present study is difficult to re- j solve. Efran and Marcia themselves ascribe their findings j to the lack of independence and unreliability of their j subjective fear measure. A second explanation offered by j these authors is less plausible. They state that there j may not be a positive relationship between increased j ability to approach a feared object and a reduction in subjective fear since approaching a feared object more j closely tends to bring more fear and conversely standing further away will make for less fear. This notion. 89 however, is not borne out by the results of the present study which found a correlation of .93 between SF (Same) and SF (Closest) improvement measures. Efficacy of the Expectancy Measures Expectancies were manipulated by means of differ ential instructions given to subjects at the beginning of their first treatment session. Evidence as to whether the manipulations were successful comes from three sources: 1. Questionnaire B, administered immediately after the expectancy manipulation asked the subjects to rate what they expected their fear to be at the end of treatment on three criterial measures, BAT, GSF, and SF (Same). On only one of these measures (GSF) did the subjects who received positive expectancy instructions (D and P groups) expect significantly more improvement (P >-05) than the negative expectancy (D- group) subjects although on the other two measures the difference was in the predicted direction, at the .10 and .20 levels for BAT and SF (Same) respectively. Comparing the D and P groups, who received the same expectancy instructions but different treatments, disclosed that on none of the three measures did their difference approach significance. These results are interpreted as indicating that while the expectancy manipulation did not succeed completely, there was a trend toward significant differences between the positive and negative expectancy groups on the variable 90 of expectancy. 2. The D and P groups, who received the same expectancy instructions but different treatments, both rated themselves as liking the experimenter more (p > .05 and p > .01 respectively) than the D- group who received the same treatment as the D group but was told that they were in a control group. This result is interpreted as indicating that the subjects in the D- group were con vinced that they were control subjects, receiving a dummy treatment and therefore they had a more negative attitude toward the experimenter. 3- Comments from the subjects on the question naires and during the post-experimental interview indicat ed that the expectancy manipulation had influenced at least some of them. Pears of having "ruined the experi ment" were voiced both by D- group subjects who improved 2 and I) and P group subjects who failed to improve. This 2 An alternative explanation for the difference in rate of improvement between the D and D- groups--that it j is due to conscious cooperation with the experimenter on j the part of the D- group subjects such that they inten- j tionally do more poorly on the posttest cannot be com- j pletely ruled out. However the fact that some subjects j were willing to risk the displeasure of the experimenter j by performing in a manner opposite to the one in which they thought they were expected to perform argues against such j an explanation. Also'" in similar work Orne (1959b, 1964) j has claimed that conscious cooperation with the purposes of the experimenter is not a factor in the demand charact eristics effect and Riecken (1962) agrees that while the subject in an experiment certainly wishes to present him self in a favorable light the dishonesty implicit in con scious complicity with the experimenter to confirm his 91 is taken as evidence that many subjects were aware of what was expected of them and that demand characteristics (Orne, 1962) were operating to effect the results of the experiment. At the same time it must be reported that a few D- group subjects would praise or criticize the de conditioning treatment as if they had never been told that they were in a control group and receiving a "dummy" treat ment. With these subjects the experimenter regularly asked if they remembered being told that they she was in a control group and knew what this meant. But it seemed as j if they had temporarily "repressed" this memory which seemed nevertheless to be influencing the subjects’ behav- i ior since these same subjects, as hypothesized, often had not improved. In summary, evidence from questionnaires and post-experimental interview behavior is suggestive j i that the expectancy manipulation influenced at least some j i subjects. | 1 It might be argued that the fact that there was no i j significant differences in expectancy of improvement between the D group and D- group suggests that the differ- j ence in improvement between those groups cannot be attributed to differences in expectancy. Though plausible enough this contention is open to dispute; and even if it were entirely acceptable it does not weaken the his hypothesis makes such conscious complicity unlikely. 92 conclusions of the study appreciably. It is argued here that the reason that no signifi cant differences in expectancy between the D and D- groups emerged may be due to the crudity of the measuring instrument. The relevant literature (e.g., Cronbaeh, i960) makes it abundantly clear that intensive refinement and much revision is usually necessary before a test emerges as a dependable measuring instrument. A rating scale requiring subjects to predict their performance on three posttests, a rather complex task, was employed in [ the present study. The measure had been constructed by the author and had not been employed in previous research. It is considered likely that factors such as improper wording and format contributed sufficiently to error variance to obscure differences in expectancies that may I have been present. That such differences were present j f is suggested by the evidence presented earlier In this I section and by the fact that all differences were in the ! predicted direction. : The D and D- groups were different only in the j i instructions they received. There were significant | differences In the extent of their fear reduction. There-I ! fore it may be concluded that the differences in .fear reduction are due to differences in instructions. i | Whether differences in instructions caused differences in expectations, demand characteristics or some other factor 93 such as motivation is not question this experiment was designed to answer. It is quite likely., in fact, that telling one group that they are in a control group and receiving a dummy treatment while the other group is told they are receiving a bona-fide treatment sets up varying demand characteristics and the differences in performance between the two groups may well be due to a combination of non-specific or social-psychological treatment factors including expectancies and demand characteristics. Whether factors of this nature can produce significant differences in treatment effectiveness is the crucial question and the one this experiment was designed to deal with. In the present study verbal communication between experimenter and subject was minimal; as well as program med. However, the experimenter necessarily knew the groups to which the subjects had been assigned. It seems ! i well-established that in such a situation the unintended, j non-verbal influence of the experimenter may operate to I bias ,the subject in the direction of the experimenter's hypothesis (Friedman, 19^7; Goldstein, 1962; Rosenthal, I i 1966). In the present study a RA ignorant of the groups | to which the subjects were assigned was assigned to administer the criterial measures. However, the possibil ity of experimenter bias operating to influence the results in the present study cannot be ruled out. \ j 94 Relationship Between Expectancy and Improvement While a relationship between subjects' expecta tion and actual performance was anticipated, the strength of that relationship was impressive. The finding supports Goldstein's (1962) contention that client prognostic ex pectancy is related to the outcome of psychotherapy. It is in line with level of aspiration studies as well as with the results obtained in correlational studies by Brady, Zeller and Reznikoff (1959)? Friedman (1963)? Goldstein (1961) and Lipkin (1964) all of whom found a positive relationship between expectancy and improvement. The Manipulation of Expectancy Granted that positive expectancy is an important if not crucial constituent of an effective therapy, the question, "How may a positive expectancy be engendered?" I naturally arises. It is the impression of this experi- j menter that college student experimental subjects bring : with them a positive expectancy about the treatment. The problem with these subjects is rather, for purposes of I experimental control, in engendering a low or negative ; expectancy. The experimenter found in pilot work that in j i order to make an adequate impression it was necessary to j inform the low expectancy subjects repeatedly and in different ways that the, treatment they would be receiving O was ineffective. Similarly, Efran (1968) found that j<I. Ef. ran, jparso-nal.. c ommun Icai-Lorr,. . . . 1 - 5= 1 6 8 - . . 95 subjects that had been told that they were not receiving the complete treatment and that there was more to come "forgot" those instructions and he recommended that some way be found to reinforce and repeat negative instructions. There seems to be a dissonance-reduction effect such that when subjects have a negative expectancy and the treatment is one with face validity, they tend to forget or disregard the negative expectancy instructions. On the I other hand, as in the case of the placebo group subjects j in this experiment, when subjects were given a positive j expectancy and the treatment received did not possess great face validity, the subjects often evidenced a puzzled and sometimes doubtful attitude towards the treat ment . | Constructing a truly inert placebo treatment j requires that care be exercised that no ingredient of any effective therapy be present in the placebo treatment. _ j Because of the plethora of treatments and theories of I therapy extant it would appear difficult if not impossible | to avoid including in a placebo group an element of some j treatment or other. For example, in constructing a placebo therapy the problem for which the patient is being treated ' should not be so much as mentioned, because imagining the feared stimulus is an instance of the Wolpin decondition- j ! I ing paradigm. There seems to be little doubt but that setting up 96 an adequate placebo therapy requires extreme care. Care is all the more urgent since experimenters who set up placebo controls usually hypothesize that the actual treat ment being evaluated will show superiority to the placebo treatment. To judge from the literature (e.g., Lang, Lazovik & Reynolds, 1965; Davison, 1968) and the research on experimenter bias (Rosenthal, 1966) they would tend to be satisfied with less than a treatment of equal face validity to the actual treatment. In that case, the placebo therapy will be deprived of much of its effective ness. | Expectancy and Social-Psychological Effects No previous study of psychotherapy which had as its object the induction of a negative expectancy has come I to the attention of the author. The dramatic effect of i the D- group manipulation suggests that a careful analysis j of the negative expectancy manipulation might be in order. ; It will be recalled that all subjects who volun- ; teered for the experiment were first told that there was a | I possibility that they would be placed in a control group. Subsequently D- group subjects were notified that, by j chance, they had indeed been assigned to the control group and would be receiving a dummy treatment instead of the real treatment. Essentially this comprised the manipula tion. What Is of interest here Is the following: In order to reinforce the negative expectancy and also to 97 prevent dropouts subjects were told that if they so desired they could take the regular treatment after the control group sessions were over. They were asked to indicate before they received their first treatment if they thought they might want the regular treatment after | the control group sessions were over. Interestingly less than one quarter of the control group subjects thought there might be a chance they would desire the "real" treatment. At the post-experimental interview., of course, the control group subjects were apprised of the fact that they had, in fact, received the "real" treatment so it is S i not surprising that none of them requested to receive it. For the placebo group subjects, however, who were told at the post-experimental interview that they had not in fact received a bona-fide treatment it is noteworthy that | none of these subjects inquired about receiving the real treatment. What this suggests is that there was not a very i j strong interest on the part of the subjects in receiving treatment for their fears. This is of importance because I j the literature, on medical and drug placebos (e.g., Beecher,j 1955; Honigfeld, 1964) suggests that a heightened desire for treatment enhances the placebo effect. Thus the placebo effect in the present study may have been atten- | uated by the lack of interest on the part of subjects in obtaining treatment. 98 This is of some interest if an attempt is made to separate placebo from demand characteristics effects. It is conceivable that at least as important as the placebo or expectancy effect in the present experiment is the social-psychological or demand characteristics effect. In future research it may be worthwhile to attempt to sepa rate out demand characteristics from expectancy effects by manipulating them in opposite directions, e.g., arranging J the situation to convince the subject that the treatment he is receiving is of known effectiveness with the ex perimenter holding the belief that it is, in fact, inef- j | fective. This could be done the other way around, of course, e.g., that the experimenter is trying to validate a treatment of known ineffectiveness. I Pre-Treatment Level of Fear It is likely that the subjects’ pretreatment level of fear was somewhat lower in the present study than in Myerhoff's (1967) experiment. This is because Myerhoff gave his subjects two successive pre-tests and disqualified 4 those who improved more than two score points. It may be j i argued therefore that the strength of the non-specific | I treatment effect in the present study may be exaggerated by the relatively low levels of fear being dealt with. However the literature on non-specific treatment effects ^This procedure may have resulted in a lower prac tice effect in that author’s study. 99 suggests that high levels of distress are associated with heightened reactivity to placebos and other non-specific treatment effects (Beecher., 1955; Beecher, 1962; Goldstein, 1962). It is perhaps surprising from an intuitive stand point but it is nonetheless a relatively unambiguous find ing that the greater the extent of a wound, for example, the greater the placebo effect. In laboratory situations, contrariwise, where relatively low levels of pain are induced experimentally, placebos are much less effective (Beecher, 1962). On this basis the results of the present study are if anything an underestimate of the influence of expectancy and placebo factors on treatment effective- | I ness in actual clinical practice. j Liking for Experimenter A few comments are in order in regard to the liking for experimenter measure. This measure did not correlate either with improvement or expected improvement. The j ' ' ' ■ i reason for this may be the markedly restricted range of j ratings that were obtained. The standard deviation was j I 12.07--on a 100 point scale. An intuitive grasp of the lack of variability of this measure may be obtained by considering the fact that 32 of 6l subjects, more than ! i half, checked the precise middle of the graphic rating | scale (50); only 3 scores were less than 40 and only 10 were greater than 60. 100 It Is not surprising, considering this restriction in range, that the measure did not correlate with other relevant variables. It would appear that subjects, even when they are assured that the rated person will not know of their rating, are loathe to make personal judgments of this nature. It would seem that some alternative method should be devised to measure the variable of liking for the experimenter if there is an interest in relating it to other variables in future,research. Some non-reactive index of liking might be appropriate. Some suggestions are number of smiles, number of words spoken or comments made, number of glances at the experimenter or amount of time spent lingering in the laboratory after time is called. Miscellaneous Questionnaire and Interview Material Comments from a good many subjects support Myerhoff's (1967) claim that a moving snake is a much more fearful stimulus than a stationary one. In future experi ments, as in the present one, efforts should be made to see to it that the snake is moving or stationary for all subjects. There was no evidence of symptom substitution except from one female placebo group subject who claimed, "Since the start of the experiment I have become less afraid of rats but much more afraid of other animals and insects." This subject was also one of the seven or eight 101 (all female) subjects who mentioned dreaming about the phobic animal. Some of these subjects proceeded to re count their dreams which appeared to have made a strong impression on them since the subject of the dreams was entirely outside the area covered by the post-experimental interview. Finally evidence was obtained which indicates that subjects do indeed preserve the confidentiality of the experimental procedure inasmuch as no subject reported saying or hearing anything which might have prejudiced the result of the experiment. Moreover it was the impression of this experimenter., based upon the conscientiousness with which even the most minor transgressions of secrecy were reported and explained,, that the subjects' reports on this issue were trustworthy. CHAPTER V SUMMARY In the field of psychotherapy there has recently been a good deal of interest in the experimental evaluation of various behavior therapies. A number of behavior therapy techniques containing differing and even contra dictory operations have nevertheless achieved similar high rates of success. In psychology* medicine, and pharmacology evidence has been accumulating that non-spec ific treatment factors such as patient expectancies, experimenter effects and demand characteristics are capable of influencing both the success of treatment and the out come of experiments. This study was designed to produce evidence bearing on the question of whether the similar success rates found in evaluations of different behavior therapies are due to non-specific treatment factors common to all therapies. More precisely, the purpose of the study was to obtain an estimate of the extent to which non-specific treatment factors were determinative of the success of a desensitization-like procedure in reducing fear of small animals. 102 103 Subjects were 67 college students claiming fears of snakes or rats. Two groups of 20 subjects each received the deconditioning treatment. The only difference between these two groups was that subjects in the D group had the ordinary expectation that they were receiving a bona-fide treatment while subjects in the D- group were told that they had been assigned to a control group and would be receiving a ’ ’dummy treatment”. The third group, the P group, which was set up to serve as a placebo group was made up of 21 subjects who were given instructions designed to induce expectancies of symptom relief. These subjects actually received a pseudotherapy. Finally a small (n = 6) no-treatment group was included in order to obtain at least a crude measure of the practice effect. All subjects received pre and post testing which provided self-report measures of fear and a measure of avoidance behavior in the presence of the phobic animal. I i I Testing was conducted by a research assistant in order to | ■ j preclude experimenter bias. She was unacquainted with the j j experimenter1s hypotheses as well as with the groups to j which the subjects belonged. Subjects assigned to the treatment groups received three sessions of either decon ditioning or pseudotherapy. Treatments were administered in groups. The deconditioning procedure consisted of tape-recorded instructions in visualizing scenes of contact with the phobic animal in great detail. The pseudotherapy 104 consisted of high-speed tachistoscopic projections of travel scenes and simple addition problems. The content of the travel scenes could not be identified while the addition problems, which the subj'ects were required to solve, were barely , perceptible. All subj'ects were given a rationale for the efficacy of the treatment they were receiving. The results of the experiment were in general confirmatory of the experimental hypotheses. All three treatment groups showed fear reduction suggesting that both specific and non-specific treatment factors were operating. The D group improved significantly more than the D- group on all measures and was superior to the ? group as well on the behavioral measure and one of the self-report measures of fear reduction. The P group was superior to the D- group on two of the three self-report measures, though not on the behavioral measure. Expect ancy appeared to affect self-report measures of fear reduction more than behavioral measures. The evidence, though not unequivocal, suggested that the expectancy manipulation was effective and there were significant correlations between measures of improvement and measures of expected improvement. The no-treatment group also showed improvement on the behavioral measure however this improvement was interpreted as artifactual in nature. APPENDICES 105 APPENDIX A PEAR SURVEY SCHEDULE 106 107 APPENDIX A PEAR SURVEY SCHEDULE NAME: AGE: MALE: FEMALE: (please print) This questionnaire contains a list of some of the many things and situations that can make people uncomfort- able, fearful, or anxious. Please check,, in the appropri ate column to the right of each statement, the amount of fear or anxiety that each produces in you. If you are not sure, make a guess. Please answer every item. Many thanks for your cooperation! DEGREE OP PEAR OR ANXIETY A A PAIR VERY NONE LITTLE AMOUNT MUCH MUCH 1. Crawling insects 2. Pire 3- Mice 4. Going up plane in an air- 5. Lizards 6. Heights 7. Enclosed places 8. Rats 9- Blood 10. Moths 11. The dark 12. Snake s 13- Deep water 14. Progs 15- Surgical operations 16. Spiders APPENDIX B DESCRIPTION OP THE STUDY READ TO PROSPECTIVE STUDENT VOLUNTEERS IN THEIR CLASSROOMS 108 109 APPENDIX B DESCRIPTION OF THE STUDY READ TO PROSPECTIVE STUDENT VOLUNTEERS IN THEIR CLASSROOMS The general purpose of this study Is to understand better what makes psychotherapy effective. It Is diffi cult to study psychotherapy under laboratory conditions for a number of reasons among which are the complicated constellation of problems that a patient usually has, the difficulty in evaluating the degree of improvement in those problems and the generally long-term nature of the process. To simplify studying the problem we are working with an experimental procedure in reducing fears of small harmless animals. Such fears are common in otherwise j normal, healthy persons. At the same time it is felt that j if a psychotherapeutic procedure is effective with such fears then it is also likely to be effective in helping neurotic people overcome some of their fears which make them miserable. . The procedure to be used, while still in its ex perimental stage, is absolutely safe and has been found effective in similar situations. Subjects in the experi ment will not be obligated to do anything they don't want to do, nor will they be subjected to embarrassing or humiliating experiences. (Subjects will be paid $5*00 on completion of treatment.)* | | Those whose questionnaires have been marked "Elig- I ible" may volunteer for the study. Four sessions aver- j aging 20 minutes are required. The location is the I Psychological Research and Service Center at 73^ West j Adams Boulevard, 2nd Floor. (Directions given.) j One last thing: Not all of those who volunteer ! for this study will receive immediate treatment for their j fear during the course of the experiment. This is because j on a random basis some subjects will be assigned to a | control group. Treatment given to this group will lack ! the essential component which is presumed to cause the reduction in fear. However, subjects who are assigned to this group will, if they wish, receive the regular treat ment for their fears right after the completion of the experiment. * This sentence was omitted if inappropriate. MANUAL: APPENDIX C FOR RESEARCH ASSISTANT PRE-TESTING 110 Ill APPENDIX C MANUAL: FOR RESEARCH ASSISTANT PRE-TESTING You will have in advance the name of the _S who is assigned to any particular 15 minute block of time. You will also know which animal the _S claims to be afraid of. These instructions will refer to snakes but will be applic able to the other animals as well. At the appointed time go to the waiting room and ask if the person presumed to be the S is, in fact, "John Doe", for instance. Hand the S the "Tear scale" form (FS) on the clipboard and tell him To knock on the door to Rm. 209 when he has completed the form. Then go to room 209 and get the snake moving. The door should be kept locked so that the j3 cannot enter it until you open the door for him. When he knocks say "just a minute," replace the snake in the cage leaving the top open and go to the door. Take the completed FS and clipboard from the S at the doorway and check to see that the FS was completed correctly, insuring that the _S does not see the snake while you do so. If the form was completed incorrectly give the S the appropriate instructions and ask him to return when he has completed it correctly. Take the FS form and clipboard from the jS and retain them. Administer the BAT in accordance with the manual. As soon as you have his score ask him to go back to the waiting room and ask him not to discuss the procedure with anyone (since there may be other Ss in the waiting room). In the case that the _S lifts the snake up dis continue the test and tell the _S, "We are looking for £>s for this experiment who are afraid to approach snakes. Since you don't show much difficulty in this regard we con-i gratulate you but must excuse you from the study." Cautionj the _S not to discuss the experiment with anyone. j If the S was unable to lift the snake write down on the FS his score and a description of his behavior, e.g. "Touched the outside of the cage," or "Stood within 4 feet of the cage." Then take the £ > the SF scale and Subject Availability Schedule and clipboard and ask him to fill them out and hold on to them and you will pick them up. This is so that you can go right on to testing the next jS. The next time you are in the waiting room and the S appears to be finished, take the completed forms from him XinsPeGt- ing them to see that they were done correctly) tell him 112 that he will he contacted about his next appointment and caution him again not to talk to anyone about the pro cedure explaining that if a S knows what to expect when he enters the room, this may influence his behavior. When questions are asked answer them either by repeating the appropriate directions, or saying that you are not permitted to answer any questions about the ex periment and that Mr. Jaffe will answer all questions at the completion of the experiment. If Ss attempt to lift snake approach the cage in order to (rj be in a position to see to what degree they succeed in lifting and holding the snake and (2) to be in a position to avoid wear and tear on Lemuel. If they get him off the floor of the cage and/or if they raise the top part of his body out of the cage ask them to hand you the snake or to replace him gently and discontinue the test and excuse the _S. Say, at the doorway: (1) "You indicated on the questionnaire that you filled out in class that you are afraid of snakes. We have in the cage there (point) a non-poisonous snake. I would like you to approach it as closely as you can." As _S approaches the cage, close the door behind him and take your position by the door. (2) When S stopd moving closer wait 10" and say, j "I assure you that the snake is non-poisonous, may I urge j you to approach as closely as you can." I j j a. If the.S does not move closer within 10" after! urging This completes the test. j b. If £ 5 does move closer repeat the procedure ! described just above (2). Begin timing after each "standard urging request. (3) If S approaches within two feet of the cage, j say, "Would you "be willing to step right up to the cage and look down at the snake?" a. If the £ > does not comply, adopt the usual pro cedure, timing 10", urging, timing another 10" and scoring "6" if he has not complied through out this interval, ("Stands within two feet of cage.") b. If the S complies by stepping up to the cage and loolcing down, adopt the usual procedure _________ asking him to touch the outside of the cage, 113 place his hand on the floor of the cage and so on as long as he complies, c. If the S steps back beyond two feet, close the cage ancF adopt the usual procedure for £>s who retreat (See "4" below). (4) In the event £ > retreats from a closer to a more distant point, his score is the closer point provided he remained at that closer point for at least 10". If this is not the case wait for S to stop retreating and adopt the procedure described Tn (2) above. Notes: If _S shows signs of fear which he does not appear able to cope with, terminate the test immediately and give him an appointment with the author. If _S asks any questions about the procedure or the snake (e.g., "How do I know he can’t jump out?") do not answer--just go on timing. After administration of the SF scale you may refer any remaining questions to the author. Do not engage in any conversation with a £ > except that outlined in these instructions. Make each statement as nearly verbatim as possible to insure maximum uniform ity of your contact with each S. The point is to get each _S to move as closely as possible using the same amount of urging. _Ss are given 20 sec. from point of being asked to lift and hold snake to do so _if they make an attempt to do so within the allotted 10 sec. APPENDIX D GENERALIZED SUBJECTIVE PEAR SCALE (GSF) 114 APPENDIX D GENERALIZED SUBJECTIVE BEAR SCALE (GSF) Room 209 Clinical Research Lab. A Name PEAR SCALE Put a mark through the line below at a point which corresponds to your present degree of fear of snakes. i___________________________________________________________________________________ I___________________________________________________: ___________________________________1 0 50 ; 100 complete absence maximum fear (or of fear (or anxiety) anxiety) ever felt or Imagined APPENDIX E BEHAVIORAL AVOIDANCE TEST (BAT) 116 117 APPENDIX E BEHAVIORAL AVOIDANCE TEST (BAT) Score Behavior 1...................... Lifts and holds the snake. 2...................... Touches the snake. 3...................... Places hand on floor of cage. 4......................Touches outside of cage. 5...................... Stands over cage looking down at snake. 6...................... Stands within two feet of cage. 7...................... Stands within four feet of cage. 8...................... Stands within six feet of cage. 9...................... Stands within eight feet of cage. 1 0..................... Stands within 12 feet of cage. 1 1..................... Stands In doorway or refuses to enter room. Note: Snake Is moving, cage is open, research assistant stands near doorway during testing. APPENDIX P SUBJECTIVE PEAR SCALE 118 APPENDIX F SUBJECTIVE FEAR SCALE Name Please rate your fear at the closest point you were able to approach the snake by making a mark at the appropriate place on the line. 0 50 100 COMPLETE ABSENCE OF FEAR OR ANXIETY MAXIMUM FEAR (OR ANXIETY) EVER FELT OR IMAGINED IN THIS SITUATION APPENDIX G INSTRUCTIONS: D- GROUP 120 121 APPENDIX O r INSTRUCTIONS: D- GROUP This Is the first of three sessions. You will not be asked to look at any of the animals that you are afraid of until after the third and last session. You have been assigned to the control group and therefore will not be receiving treatment for your fears at this time. Instead during this session and the next two sessions you will be receiving what might be called a j "dummy treatment". The reason for this is that in order to constitute a proper control you must spend the same amount of time in the experimental situation doing the same kinds of things as the regular treatment group. This is the only way we have of isolating the factor of the treatment itself from all other factors involved in the situation. The only thing different between your group and the treat ment group is that you will not be receiving the actual treatment we are evaluating. Now of course assignment to groups was made on a purely random or chance basis and as I said in class., those of you who wish to will begin receiving the real treatment at the completion of these three control group sessions. Now I would like some idea of how many might want to receive the regular treatment after these control group sessions are over. You don't have to commit yourself. I just want an estimate for scheduling purposes. Finally I would like to urge you to participate in the experiment as wholeheartedly as you can since the control group is just as important as the treatment group in any' experiment since without a control group there would be no way of evaluating the treatment. Now I would like you to complete a questionnaire that I will hand out at this time. (Hand out Questionnaire B). Please take as long as you need to fill it out-1 -when you're finished turn it over on your desk so I'll know when you're through. (Does everyone have something to write with?) By the way, when the questionnaire refers to "treatment" it means the dummy treatments or control group treatments you will be receiving today and the next two times you come here not the regular treatment that you may 122 want to receive afterwards. APPENDIX H o INSTRUCTIONS: D AND P GROUPS 123 124 APPENDIX H INSTRUCTIONS: D AND P GROUPS This is the first of three treatment sessions. You will not be asked to look at any animals that you are afraid of until after the third and last session. You have been assigned to the actual treatment group. You will be receiving a treatment for your fears which, while still being refined by research, is based on established principles of learning theory and has been employed in other laboratory investigations of fear reduc tion with quite promising results. Before we go any further I would like you to com plete a questionnaire that I will hand out at this time. (Hand out Questionnaire B). Please take as long as you need to fill it out. Does everyone have something to write with? t i APPENDIX I QUESTIONNAIRE B j | i j I j 125 126 APPENDIX I QUESTIONNAIRE - Br NAME PEAR SCALE (l) Marked below In red is the degree of fear of snakes you estimated you had the first time you used this scale. This time we would like you to predict your degree i :of fear of snakes after treatment is completed. Put a mark through the line below at a point which j corresponds to how afraid of snakes you think you'll be j after completion of treatment. ! | ■'---------------------- — ------------1________________________________________ i : 0 5 0 ruo j complete absence maximum fear (or I iof fear'(or anxiety) anxiety) ever j felt or imagined (2) Marked below in red is the extent to which you were able to approach the feared object the first time you took the test. We would like your estimate as to how close you think you will-he able to approach it after completion of :treatment. Just put a checkmark by the approriate line. Stand in doorway or refuse to enter room. ! Go within 12 feet of it. Go within 8 feet of it. Go within 6 feet of it. Go within 4 feet of it. Go within 2 feet of it. Stand over open cage and look down at it. 127 ______________ Touch outside of cage with top open.! Place hand on floor of cage. I - - - - - - - - - - - - - - - - - - - i i ______________ Touch the animal. ______________ Lift and hold the animal. (3) Marked below is your rating of your degree of fear of ! the feared object the first time you took the test. How j afraid do you think you'll be of it at that same degree of ; closeness when you repeat the test aTf the end of treat- ’ ment? j Your first rating: complete absence maximum fear (orj :of fear (or anxiety) anxiety) ever j | felt or imagined! i in this situa- I > tion j Your estimate of how afraid you'll be at the end of treatment— at the same degree of closeness as you came to it the first time. maximum fear (or anxiety) ever, felt or imagined in this situa tion (4) In the space below we would appreciate any thoughts iyou may have about the experiment so far. Use the back of this sheet if you want to. If you have no thoughts or ^comments please write "None". !complete absence |of fear (or anxiety) I APPENDIX J NEUTRAL SCENE: D AND D- GROUPS I I t I j I i ! S 128 129 APPENDIX J NEUTRAL SCENE: D AND D- GROUPS You will be asked to visualize a scene Involving the animal you saw in its cage in the next room. In order to acquaint you with the procedure, I’m going to ask you to practice imagining a scene you may have experienced on your way up to keep this appointment. Close your eyes. At the entrance to this building is a big, heavy wooden door with a big brass knob. Can you imagine yourself approaching it? Try to really see it. jReally put yourself into the scene. Now experience your- I self pushing it open. It is a rather heavy door. As you ;open it you see a large paneled room. Experience it in your imagination very clearly and vividly. (Pause.) You may open your eyes. | Now you will be imagining a scene involving the |animal you saw in the cage during your pre-test in the !adjoining room. Follow the instructions on the tape. I(Play tape.) APPENDIX K PEAR SCENE 130 APPENDIX K 131 PEAR SCENE Now you will be imagining a scene involving the animal you saw in his cage during your pretest in the adjoining room. 1. Close your eyes. Picture yourself going over to the cage with the top open. You can clearly see him moving around inside the cage. 2. You reach inside the cage., put your hands around him., hold him, lift him out. He doesn’t bite. 3. As you hold him you can feel his full weight. 4. You can see his head move and feel him wriggle around in your hands. 5. As he crawls over your hands and arms you can feel his body quiver. 6. Peel yourself stroke him across his back as he crawls over your hands and arms, very vividly, very clearly. 7. Picture yourself very vividly as you hold him in your hands and play with him. You may open your eyes, (pause) and repeat (l) APPENDIX L WORK SHEET 132 133 APPENDIX L WORK SHEET Print. Name 1. ___________ 22. 43. 64. 2. _______ 23. 44. 65. 3- 24. 45. 66. 4. ___________ 25. 46. 67. 5. ___________ 26. 47. 68. 6. 27. 48. 69. 7. 28. 49. 70. 8. ________ 29. 50. 71. 9- 30. 51. 72. I. 0. ___________ 31. 52. 73- II. ___________ 32. 53. ' - 74. 12. ___________ 33- 54. 75- 13. ___________ 34. 55- 76. 14. ___________ 35- 56. 77. 15. 36. 57. 78. 16. 37. 58. 79- 17. 38. 59. 80. 18 . 39- 60. 19. ___________ 40. 61. 20. ___________ 4l. ________ 62. 21. ___________ 42. 63. APPENDIX M RATIONALE AND ADDITIONAL INSTRUCTIONS: P GROUP 134 135 APPENDIX M RATIONALE AND ADDITIONAL INSTRUCTIONS:' P GROUP The treatment you will he receiving is based upon two well-established principles. One is the state of reciprocal inhibition which exists between central and autonomic nervous system impulses. An example of a cen tral nervous system impulse is the kind of abstract think-1 ing that goes into solving an arithmetic problem. An example of an autonomic nervous system impulse is fear. Now these impulses inhibit each other so that for example ; if you are concentrating your attention on a problem in a careful orderly manner you are not likely to be afraid. ; On the other hand, if you are really fearful about some- j thing it is hard to concentrate on a task like solving ! arithmetic problems. The other principle upon which this treatment is j based is the unconscious nature of the fear response. An ! example of this principle is that you cannot will to be j afraid. In line with this principle the fearful stimuli ! to which you will be exposed will be perceptible only un- j consciously, in other words they will be subliminal stim- : uli--below the threshold of conscious awareness. Though each slide has been judged as above the median in fear arousing potential, the exposure times will be too short for them to become consciously perceptible so that though j you will not be perceiving the slides consciously, they j will be making an impact unconsciously. I By the way, these fear stimuli will not include j any of the animals you are afraid of. j As part of your treatment you will be doing addi tion problems. These problems will appear on the screen just long enough to be consciously perceptible. Doing these problems activates central nervous system impulses and counteracts the unconscious fear responses. I will now pass out a sheet of paper on which you are to write your answers. You may not see some of the addition problems at first and you may not be sure you | are putting down an answer in the right place but you will j soon get used to the idea and as long as your answers are ! in the right order they will be scorable. Don’t skip any | spaces unless you think you missed a problem. 136 (PROM BACK AT ROOM, AT PROJECTION) Remember you will not be able to identify the fear stimuli which operate unconsciously but you should try to identify the addition problems which will be intermingled with them. APPENDIX N MANUAL: POST-TESTING (FOR RA) 137 138 APPENDIX N MANUAL: POST-TESTING (FOR RA) You will have the names of _Ss to be tested but you won't know their pre-test scores until after the first part of the post-test is completed. The author will pro vide you with a folded sheet of 8-§- x 11 paper. On the outside will be _S's name and number. After the first I part write the AS score on the outside of the paper and open the paper to get £>' s AS score which you will need to ! administer the second part of the test. If there is no I change between pre and post testing do not administer the second part of the test. I Get the snake moving and go to the waiting room j and ask the _S to come with you. Enter the room ahead of i the _S and say: "in the cage there (point) is the same | non-poisonous snake you saw in your pretest. I would ! like you to approach it as closely as you can." I Then administer and score the AS in accordance with the manual and ask the S to wait for you in the waiting room as soon as you "Have his score. Then go to j the waiting room and give the _S the SF scale and clip- | board and ask him to come back to the testing room and | knock on the door when he is through. ; When he knocks on the door, get the snake moving I as you did in the pretest and go to the door and take the j SF scale and clipboard from him and say, "The first time j you took the test you..." and complete the sentence with i whichever one of the following sentences is appropriate, j using the marker where it is relevant: "...stood in the doorway" or "...refused to enter the room" "...came within 12 feet of the cage" "...came within 8 feet of the cage" "...came within 6 feet of the cage" "...came within 4 feet of the cage" "...came within 2 feet of the cage" "...stepped right up to the cage and looked down at the snake" "...touched the outside of the cage" "...put your hand on the floor of the cage" "...touched the snake" 139 Where you are using the marker say., "Would you stand there again please/' or ask the jS to perform the appropriate action again. In the former case time 10" and then ask the S to go to the waiting room. In the latter case ask hTm to go to the waiting room after they have performed the action. Then take the SF scale, Questionnaire C and the clipboard to them and ask them to complete and return the forms. When they have done so tell them to go back to the waiting room and the experi menter will see them in a moment. When you collect Questionnaire C separate the third page from the first two and ask jS to hold on to the first' 2 pages--E will take it from him. APPENDIX 0 QUESTIONNAIRE C 140 APPENDIX 0 QUESTIONNAIRE C NAME (l) Were you tempted to talk to anyone connected or un connected with the school about the study? Yes No___ (2) To whom did you actually talk about the study, if anyone? (This won't affect your experimental credit at all.) (3) What did you say about the experiment? (4) What did you hear about the experiment before or during participation from any source other than the ex perimenter and his research assistant? (5) Howwell did you know the experimenter (the person who gave the treatment sessions) before the experiment? (Circle the most nearly correct statement.) fa) Never saw him before. (b) Saw him around'' but never spoke to him. fc) Was acquainted with him slightly. id) Knew him moderately well. — (e) Knew him very well. (6) If you experienced a reduction in fear, what factors (whether in or out of the experimental situation) do you think were responsible? (7) Put a mark through the line below at a point which corresponds to your present degree of fear of the feared 142 object. o ‘ 50 " ; nio Complete absence Maximum fear (or of fear (or anxiety) anxiety) ever felt or Imagined (8) Any comments that you would care to make about the experiment would be welcome here. Use the space below and the back If you need It. If you have no comments, please write "None". Do not forget the last page of this questionnaire. 143 Note: This sheet will he detached from the rest of this questionnaire and will not he associated with your name. The Information Is requested for scientific purposes. Please be as honest as you can. How well did you like the experimenter?--(the person who gave the treatment sessions). Put a mark through the line below at the appropriate point. _______________ I 100 : more than any one I've ever met t_______________________________ I __ o 50 less than anyone I've ever met APPENDIX P DE-BRIEFING FORMAT: D GROUP 144 145 ! APPENDIX P DE-BRIEFING FORMAT: D GROUP The purpose of this session is to give you a birds-: eye view of the experiment and to answer any questions you might have, but first I would like to ask you a couple of questions: What did you think of the treatment for your fears , that you received? \. Did you think, at any point, that you were not actually receiving a bona-fide treatment? Did you think the pre-test was part of the treat ment? Your part of the experiment was concerned with an j answer to this question: To what extent is a treatment ! which is known to be effective in reducing fears effect- ! ive because the patients "believe in it". In order to I obtain an answer to this question, two groups were set up j which received the same treatment, which is a bona-fide j treatment. One group, your group, was informed, correct- I ly, that it was a bona-fide treatment. The other group j was told that they were a control group and were led to believe that the treatment they were receiving was not the; real treatment. Thus the two groups differed only in the j extent to which they believed that they were getting a | real treatment. Now since both groups received the same ; treatment and differed only in their expectation--any difference in improvement between the groups should be due to expectation. Please be sure not to discuss the experiment till after the semester is over--thank you for your coopera tion. i APPENDIX Q DE-BRIEFING FORMAT: D- GROUP 146 147 APPENDIX Q, DE-BRIEFING FORMAT: D- Group The purpose of this session is to give you a birds-; eye view of the experiment and answer any questions you might have, but first I would like to ask you a couple of questions: What did you think of the treatment for your fears ; that you received? Do you remember being told you were in a control group? What is your understanding of the term "control group"? Did you, at any point think that you were not j actually in a control group? If yes, question | further to learn circumstances, etc. i ; Did you think the pre-test was part of the treat- ! ment? j I Your part of the experiment was concerned with an ' j answer to this question: To what extent is a treatment which is known to be effective in reducing fears effect ive because the patients "believe in it". In order to . obtain an answer to this question, two groups were set up j which received the same treatment, which is a bonafide treatment. One group was told they were getting a bona- I ! fide treatment and the other group, your group was told j that they were a control group and were led to believe j : that the treatment they were receiving was not a real j s treatment. Thus the two groups differed only in the ex- j I tent to which they believed that they were getting a real j : treatment. Now since both groups received the same treat ment and differed only in their expectations--any differ ence in improvement between the groups should be due to : expectation. ; Now you may have some feelings about what I've | said--perhaps about being misled as to the treatment you i received (Allow subjects to ventilate). ; Please be sure not to discuss the experiment till I after the semester is over--thank you for your cooperation APPENDIX R DE-BRIEFING FORMAT: P GROUP 148 149 APPENDIX R DE-BRIEFING FORMAT: P GROUP The purpose of this session is to give you a birds- eye view of the experiment and answer any questions you might have, but first I would like to ask you a couple of questions: What did you think of the treatment for your fears that you received? What was the purpose of being shown tachistoscopic slides of fear stimuli? What was the purpose of working arithmetic pro blems? Could you identify any of the fear slides? (If so, what were they?) What kind of slides did you think you were seeing? (if appropriate) Did you think at any time that you were not actually receiving a bonafide treatment? Did you think the pre-test was part of the treat ment? You may have heard about placebos. Placebos are inert substances which are sometimes given to a patient with the understanding that they are actually beneficial. Actually placebos have been found to be quite effective in medical treatment, in many cases as effective as the "real" medicine. Now what you received was a placebo treatment. There were no "fear stimuli" the scenes flashed on the screen at high speeds were travel scenes. So if your fear was reduced it was probably a result of your expecta tion that your fear would be reduced--which can be a very powerful treatment in itself. This was the factor we were interested in measuring, expectation. Now you may have some feelings about what I've said--perhaps about being misled as to the treatment you received. (Allow subjects to ventilate.) 150 Please be sure not to discuss the experiment till after the semester is over— thank you for your coopera tion. | APPENDIX S RAW DATA 151 152 APPENDIX S RAW DATA Code 1 • • . Subject number 2 • • . Sex 3 • • . Phobic animal, Snake (s) or Rat (r) 4 • • . BAT pretest score 5 • . BAT posttest score 6 • • . GSP pretest score 7 • • . GSP postest score 8 • • . SP pretest score 9 • • . SP (Same) posttest score 10 • • . SP (Closest) posttest score 11 • • . BAT posttest score expected 12 • • . GSP posttest score expected 13 • • . SP (Same posttest score expected 14 # , . Rating of liking for experimenter 153 N = 20 DECONDITIONING GROUP (D) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 3 F R 2 1 30 0 40 0 0 1 2 4 50 6 P S 6 3 70 70 80 8 0 80 5 50 4o 50 10 M S 4 1 96 16 20 16 16 2 36 4 80 14 M S 3 2 96 90 90 50 100 1 10 0 50 17 P s 4 2 100 10 90 0 10 3 50 0 50 18 M s 4 2 90 4o 100 ' 6o 6o 4 82 92 41 22 P s 6 6 90 50 6o 50 50 6 90 50 40 24 P s 4 1 50 10 70 10 10 3 4o 40 80 27 P s 4 1 90 4 28 4 6 1 20 4 60 31 P s 2 1 70 10 70 10 10 1 10 10 80 34 P s 8 4 100 35 100 34 34 7 98 96 5 0 38 M s 4 l 90 4 90 4 4 2 50 10 50 43 M s 4 2 66 10 80 12 90 3 45 20 5 0 44 P s 2 2 4o •60 30 30 30 2 80 90 50 49 P s 7 4 20 10 20 10 10 4 10 10 50 50 F s 4 2 100 49 80 50 6o 4 70 4o 70 53 P s 6 1 99 4 100 4 4 - __ __ 50 61 P s 2 1 80 20 70 20 20 2 50 20 60 63 P s 11 9 60 20 60 50 50 8 - - 4o 40 71 P R 8 1 80 0 76 0 0 4 — 40 50 154 DECONDITIONING GROUP--LOW EXPECTANCY (D- ) N = 20 1 2 3 4 5 6 7 8 9 10 ll 12 13 14 2 F S 2 2 70 56 40 56 56 2 50 30 46 5 F S 4 3 100 50 80 50 50 4 80 20 50 12 P S 8 9 90 90 90 90 87 8 90 92 50 15 P S 11 ll 90 90 90 90 90 11 90 90 50 20 F S 2 l 70 20 20 20 20 1 70 10 20 23 M S 6 2 50 8 20 8 10 5 4o 10 30 25 P S 4 1 80 40 60 10 10 1 80 0 50 28 P S 4 4 60 6o 4o 60 60 3 52 34 50 32 P S 2 1 10 0 4 0 0 1 10 0 60 35 M S 4 4 100 80 80 88 88 4 96 86 30 37 P S 4 4 70 6o 60 60 60 2 30 30 50 42 P S 6 4 100 50 30 20 20 4 30 10 50 45 M R 3 2 6o 86 80 90 84 2 40 20 40 47 P S 7 ~ 6 70 60 50 60 60 7 6o 4o 50 51 P R 6 4 70 32 80 35 35 4 50 4o 50 54 P S 7 4 90 42 76 67 46 7 70 60 68 58 P S 4 4 8o 22 23 22 22 2 50 10 41 64 F S 3 2 90 60 80 60 60 3 80 80 50 67 P S 4 4 74 80 74 80 80 4 75 75 50 72 P S 9 10 90 80 35 74 8 0 8 80 50 53 PLACEBO GROUP (P) 155 N = 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 4 F S 3 2 100 50 70 50 50 2 50 50 80 7 F S 4 2 80 50 56 0 10 1 10 0 4o 11 F s 4 3 00 6o 80 60 60 3 50 50 50 13 F s 4 3 90 50 90 0 70 2 30 20 60 16 F s 4 2: 80 50 55 50 50 2 50 10 50 19 F s 4 4 70 10 70 10 10 4 70 40 50 21 M s 8 8 94 92 70 88 88 7 94 60 50 26 F s 4 4 100 50 10 30 30 3 10 0 69 29 M s 4 4 50 70 60 20 20 3 4o 4o 85 30 M s 2 l 76 26 90 • 0 5 - 10 10 50 36 F s 4 l 30 0 30 0 0 3 20 20 50 39 F s 4 4 90 70 60 50 50 3 20 10 50 46 M s 7 2 60 8 70 8 8 4 30 30 59 48 M s 4 4 70 30 70 20 20 3 30 20 80 52 F s 4 3 70 70 40 84 84 4 6o 20 60 55 F s 2 2 70 20 70 20 20 2 50 0 60 56 F s 11 7 70 00 IS - 70 69 68 8 70 4o 90 59 F R 4 4 80 30 90 30 30 2 30 4 60 65 F s 2 2 94 00 00 80 88 88 2 80 80 50 69 F s 4 4 72 78 80 78 78 4 81 30 50 70 F s 2 1 76 1 34 0 0 1 10 0 50 156 NO-TREATMENT CONTROL GROUP (C) N = 6 123 4 5 6 7 8 9 10 11 12 13 14 1 F S 2 1 84 - 80 0 8 8 F S 11 11 100 - 90 - - 4o M s 6 4 90 - 30 20 20 41 F s 4 2 70 - 90 20 80 62 F s 10 10 76 - 100 86 86 66 M R 4 1 30 0 — — 50 j REFERENCES 158 REFERENCES Bandura, A., Grusec, J. E., and Menlove, F. L. Vicar ious extinction of avoidance behavior. Journal of Personality and Social Psychology, 1967, 5, T 5- - 2 3 : ---------------------------------- -- Barber, T. X. and Calverley, D. S. Toward a theory of hypnotic behavior: Effects on suggestibility of defining the situation as hypnosis and defining response to questions as easy. Journal of Ab- normal & Social Psychology, 1964, bd, 585-593* Barron, F. and Leary, T. F. Changes in psychoneurotic patients with and without psychotherapy. Journal of Consulting Psychology, 1955* 19/ 239-245. Batterman, R. C. and Grossman, A. J. Effectiveness of salicylamide as analgesic and antirheumatic agent. Journal of the American Medical Association, 1955, 159* 1619-1522. ; Beecher, H. K. The powerful placebo. Journal of the American Medical Association, 1956* 159, 1602-1606 Beecher, H. K. The subjective response and reaction to sensation. American Journal of Medicine, 1956, . 20, 107-113. Beecher, H. K. Non-specific forces surrounding disease and the treatment of disease. Journal of the American Medical Association, 1962', 179, 437-440. Brady, J. P., Reznikoff, M. and Zeller, W. W. The rela tionship of expectation of improvement to actual improvement of hospitalized psychiatric patients. Journal of Nervous and Mental Disease, i960, 130, 41-44.----------------------- --------- Breger, L. and McGaugh, J. Critique and reformulation of learning theory approaches to psychotherapy and neurosis. Psychological Bulletin, 1965, 63, 338- 358. : -------------- Cartwright, D. S. and Cartwright, R. D. Faith and im provement in psychotherapy. Journal of Counsel- ing Psychology, 1958, 5* 174-TTT Cartwright, R. D. and Vogel, J. L. A comparison of 159 changes In psychoneurotic patients during matched periods of therapy' and no-therapy. Journal of Consulting Psychology, 1960, 24, 121-127. Cautela, J. R. A behavior therapy approach to pervasive anxiety. Behavior Research and Therapy, 1966, 4, 99-109.“ Chomsky, N. Review of B. P. Skinner’s ’ ’Verbal Behavior". Language, 1959* 35* 26-58. Cooke, G. Identification of the efficacious components of reciprocal inhibition therapy.. Unpublished doctoral dissertation, University of Iowa, 1966a. Cooke, G. The efficacy of two desensitization procedures: an analogue study. Behavior Research and Therapy, 1966, 4, 17-24. ' i Cordovo, L. and Ison, J. R. Psychology of the scientist: X. Observer bias in classical conditioning of the planarian. Psychological Reports, 1963, 13* 787- 789. : Crespi, L. P. The interview effect on polling. Public Opinion Quarterly, 1948, 12, 99-111- Cronbach, L. S. Essentials of Psychological Testing, New York: Harper & 'Row, i960. “ Criswell, J. H. The psychologist as perceiver. In R. Taguiri and L. Petrullo (Eds.) Person perception and interpersonal behavior. Stanford University Tress,' 1"95«, 95-109- Davison, G. C. Relative contributions of differential relaxation and graded exposure to the in vivo de sensitization of a neurotic fear In, Proceedings of the 73rd Annual Convention of the APA, 1965. Davison, G. V. Systematic desensitization as a counter conditioning process. Journal of Abnormal Psycho- JLogy, 1968, 73, 91-99. Diehl, H. S., Baker, A. B. and Cowan, D. ¥. Cold vaccine: a further evaluation. Journal of the American Medical Association, 194'0, T15, 593-594. Dimond, E. G., Kittle, C. P. and Crockett, J. E. Com parison of internal mammary artery ligation and sham operation for angina pectoris. American 160 Journal of Cardiology, i960, 5, 483-486. Dinnerstein, A. J., Lowenthal, M. and Blitz, B. The in teraction of drugs with placebos in the control of pain and anxiety. Perspectives in Biology and Medicine, 1966, 10, I03-II7. Efran, J. and Marcia, J. E. Treatment of fears by ex pectancy manipulation: an exploratory investiga tion. Proceedings, 75th Annual Convention of the APA, 1967, 239-240. Eysenck, H.'J. The effects of psychotherapy: an eval uation. Journal of Consulting Psychology, 1952, 16, 319-32^ ; : Eysenck, H. J.- Behavior Therapy and the Neuroses. Oxford, Pergamon Press, i960. Eysenck, H. J. Experiments in behavior therapy. Oxford, Pergamon Press, 1965 (a). Feldman, P. E. The personal element in psychiatric re search. American Journal of Psychiatry, 1956, 113, 52-545 ; Fiedler, F. E. A comparison of therapeutic relationships in psychoanalytic, non-directive and Adlerian therapy. Journal of Consulting Psychology, 1950, 14, 436-4455 : Fisher, S., Cole, J. 0., Rickels, K. and Uhlenhuth, E. H. Drug-set interaction: The effect of expectations on drug response in out-patients. In: Bradley, P. B. Neuropsychopharmacology, v. 3, N.Y. Elsevier^ 1964, 149-156. Folkins, C. H., Lawson, K. D., Opton, E. M. and Lazarus, R. S. Desensitization and the experimental re duction of threat. Journal of Abnormal Psychol ogy, 1968, 73, 100-113: ; Frank, J. D. The dynamics of the psychotherapeutic re lationship: Determinants and effects of the therapist’s influence. Psychiatry, 1959, 22, 17-40. Frank, J. D. Persuasion and healing. Johns Hopkins Press, 1961. Frank, J. D., Gliedman, L. H., Imber, S. B., Nash, E. H., 161 and Stone, A. R. Why patients leave psychother- apy. American Medical Association Archives of Neurology and Psychiatry, 1^57, 77, 2fc>3- Friedman, H. S. Patient-expectancy and symptom reduction Archives of General Psychiatry, 1963, 8, 61-6 7. Friedman., N. The social nature of psychological research. New York: Basic Books, 1967- Gliedman, L. H., Nash, E. H., Imber, S. D., Stone, A. R. and Frank, J. D. Reduction of symptoms by pharmacologically inert substances and by short term psychotherapy. American Medical Association Archives of Neurology and Psychiatry, 1958, 79, ■ 345-351"------------------------------------ Goldstein, A. P. Therapist and client expectation of personality change in psychotherapy. Journal of Counseling Psychology, i9 6 0, 7, iSo-lST! Goldstein, A. P. Therapist-patient expectancies in psychotherapy, I\few York: Macmillan, 1 9 6 2. Goldstein, A. P. and Shipman, W. G. Patients' expectan cies, symptom reduction and aspects of the init ial psychotherapeutic interview, Journal of Clinical Psychology, 1961, 17, 12f9rT33. Graziano, F. M. and Kean, J. E. Programmed relaxation anc reciprocal inhibition with psychotic children, Proceedings, 75th Annual Convention, APA, 1967- Grummon, D. L. Personality change as a function of time • • in persons motivated for therapy. In: C. R. Rogers & R. F. Dymond (Eds.) Psychotherapy and Personality Change, Chicago: University of Chicago Press, l'964. Hillis, B. R. The assessment of cough-suppressing drugs. Lancet, 1952, 1, 1230. Hogan, R. A. Implosive therapy in the short term treat ment of psychotics. Psychotherapy: Theory, Research, Practice, 1'$'66, 3^ 25-32. Hogan, R. A. and Kirchner, J. H. Preliminary report of the extinction of learned fears via short-term implosive therapy. Journal of Abnormal Psychol ogy, 1967, 72, 106-iw: 162 Honigfeld, G. Non-specific factors in treatment. Dis eases of the Nervous System., 1964, 25, 145-155, 225-234. ------------------------- Jellinek, E. M. Clinical tests on comparative effect iveness of analgesic drugs. Biometric Bulletin, 1946, 2, 87. ' Johnson, S. M. and Sechrest, L. Comparison of desensi tization and progressive relaxation in treating test anxiety. Journal of Consulting and Clinical Psychology, 196STJ 323 280-255. Kahn, M. and Baker, B. Desensitization with minimal therapist contact. Journal of Abnormal Psycho logy, 1968, 73, 198-^88: Kirchner, J. H. and Hogan, R. A. The therapist variable in the implosion of phobias. Psychotherapy: Theory, Research, Practice, 1966', 3, 97-101. Klopfer, B. Psychological variables in human cancer. Pharmacological Review, 11,' 1959V " 898. Kondas, 0. Reduction of examination anxiety and "stage fright" by group desensitization and relaxation. Behavior Research and Therapy, 1967, 5* 275-281. Kurland, A. A. The drug placebo: its psychodynamic and conditional reflex action. Behavioral Science, 1957, 2, 101-110. Lang, P. J. and Lazovik, A. D., Experimental desensitiza tion of a phobia. Journal of Abnormal and Social Psychology, 1963, 687"5T9-~5'25• Lang, P. J., Lazovik, A. J. & Reynolds, D. Desensitiza tion, suggestibility and pseudotherapy. Journal of Abnormal Psychology, 1965, 70, 395-402. Lasagna, L., Mosteller, F., Felsinger, J. M. and Beecher, H. K. A Study of the placebo response. American Journal of Medicine, 1954, 16, 770-779- . Lazarus, A. A. New methods in psychotherapy: a case study. South African Medical Journal, 1958, 33, 660. Lazarus, A. A. Group therapy of phobic disorders by systematic desensitization. Journal of Abnormal and Social Psychology, 1961, 63", 504-510. 163 Lazarus, A. A. and Abramovitz, A. The use of "emotive imagery1 1 in the treatment of children's phobias. Journal of Mental Science, 1962, 108, 19I-I95. Lazarus, A. A. and Davison, G. C. Reciprocal inhibition and desensitization therapy. In: R. M. Jurgevich (Ed.) Handbook of direct and behavior psychotherapies, in press. Lesse, S. Placebo reactions and spontaneous rhythms. American Journal of Psychotherapy, 1964, 18, 99- T W - ------------------------------------- Levis, D. J. and Carrera, R. Effects of 10 hours of implosive therapy in the treatment of outpatients: A preliminary report. Journal of Abnormal Psychology, 1967. , 72, 504-508. Lipkin, S. Client's feelings and attitudes in relation ship to the outcome of client-centered therapy. Psychological Monographs, 1954, 68, 1-30. Lomont, J. P. Reciprocal inhibition or extinction. Behavior Research and Therapy, 1965, 3, 209-219. Lomont, J. P. and Edwards, J. E. Role of relaxation in systematic desensitization. Behavior Research and Therapy, 1967^ 5* 11-25. London, P. Modes and morals of psychotherapy. Holt, Rinehart & Winston, 1964. Masling, J. The influence of situational and interperson al variables in projective testing. Psychologi- cal Bulletin, i960, 57, 65-85. Migler, B. and Wolpe, J. Automated self-deBensitization: a case report. Behavior Research and Therapy, 1967, 5s 133-135^ Mowrer, 0. H. Learning Theory and Behavior. John Wiley & Sons, 'I960'. Myerhoff, L. Tension and anxiety in deconditioning. . Unpublished doctoral dissertation, University of Southern California, 1967. Orne, M. T. The nature of hypnosis: artifact and essence Journal of Abnormal and Social Psychology, 1959, 58, 277-299 (a). 164 Orne, M Orne , M Orne, M Orne, M Park, L Paul, G Paul, G Paul, G Pepper, Rachman Rachman Rachman T. Demand characteristics of an experimental design and their implications. Paper read at APA, Cincinnati, Sept. 1959 (a). T. On the social psychology of the psychologi cal experiment: with particular reference to demand characteristics and their implications. American Psychologist, 1962, 17, 776-783. T. and Scheibe, K. E., The contribution of non deprivation factors in the production of sensory deprivation effects: The psychology of the "panic button". Journal of Abnormal and Social Psychology, 1964, 68, 3-12. T. and Evans, P. J. Social control in the psychological experiment: Anti-social behavior and hypnosis. Journal of Personality and Social Psychology, 19655 169-200. C. and Covi, L. Non-blind placebo trial: An exploration of neurotic patients responses to placebo when its inert content is disclosed. Archives of General Psychiatry, 1965, 12, 336- L. Insight vs. Desensitization in Psychotherapy. Palo Alto: Stanford University Press, 1966. L. Insight vs. desensitization in psychotherapy two years after termination. Journal of Consult- ing Psychology, 1967, 31, 333-34?n I » . and'JShannon, D. T. Treatment of anxiety through systematic desensitization in therapy groups. Journal of Abnormal Psychology, 1966, 71, 124-1' w . 0. H. P. A note on placebo. American Journal of Pharmacy, 1945, 117, 409-412. S. Studies in desensitization: I. The separate effects of relaxation and desensitization. j Behavior Research and Therapy, 1965, 3, 245-251. S. Studies in desensitization: II. Blooding. Behavior Research and Therapy, 1966, 4, 1-6. S. and Hodgson, R. J. Studies indesensitization: IV. Optimum degree of anxiety reduction. Behavior Research and Therapy, 1967. 165 Riecken, H. W. A program for research on experiments in social psychology. In: N. F. Washburne (Ed.) Decisions, values, groups, v. II New York: Pergamon Press, 1962. Ritter, B. The group desensitization of children's snake phobias using vicarious and contact desensitiza tion procedures. Behavior Research and Therapy, 1968, 6, 1-6. ---------------------------- Rosenberg, M. J. When dissonance fails: An eliminating evaluation apprehension from attitude measure ment. Journal of Personality and Social Psychol ogy, 1965 , 1, 26-42. Rosenthal, D. and Prank, J. D. Psychotherapy and the placebo effect. Psychological Bulletin, 1956, 53, 294-302. Rosenthal, R. Experimenter effects in behavioral re- search, New York: Appleton-Century-Crofts, 1966. Rosenthal, R., Persinger, G. W., Vikan-Kline, L. L., and Mulray, R. C. The role of the research assistant in the mediation of experimenter bias. Journal of Personality, 1963, 31, 313-335- Schubot, E. D. The influence of hypnotic and muscular relaxation in systematic desensitization. Unpublished doctoral dissertation, Stanford University, 1967- Shands, H. C. Thinking-and psychotherapy. Cambridge: Harvard University Press, I960. j Shapiro, A. K. Factors contributing to the placebo j effect. Behavioral Science, 1964, 18, 73-88 (a). ! i Shapiro, A. K. Contribution to the history of the placebo effect. Behavioral Science, i960, 5, 109-135• Shapiro, A. K. Historic and heuristic definition of placebo. Psychiatry, 1964, 27, 1, 52-57 (b). Shapiro, A. K: . Etiological factors in placebo effect. Journal of the AMA, 1964, 187, 712-714 (c). Shapiro, A. K. Historic and heuristic definition of placebo: a rejoinder. Psychiatry, 1964, 27-2, 178-181, (d). 166 Shapiro, A. K., Wilensky, H. and Struening, E. L. Study of the placebo effect with a placebo test. Comprehensive Psychiatry, 1968, 9, 118-137- Snyder, R. and Sechrest, L. An experimental study of directive group therapy with defective delin quents. American Journal of Mental Deficiency, 1959, 64,""TT7~T2"3• ------------------------------ Stampfl, T. G. Implosive therapy: the theory. In: S. G. Armitage (Ed.) Behavior modification tech niques in the treatment of emotional disorders. Battle Creek, Michigan: Veteran's administration Publications, 1967- Stampfl, T. G. A learning theory derived psychodynamic therapeutic technique. Unpublished manuscript, Illinois State University, 1961. Stampfl, T. G. and Levis, D. J. Implosive therapy: The theory, the subhuman analogue, the strategy and the technique. Unpublished manuscript, University of lows,, undated. Stampfl, T. G. and Levis, D. J. Essentials of implosive therapy: A learning-theory based psychodynamic behavior therapy. Journal of Abnormal Psychology, 1967, 72, 496-503. Stampfl, T. G. and Levis, D. J. Implosive therapy-- a behavioral therapy. Behavior Research and Therapy, 1968, 6, 31-3 > 6 " . Stone, A. R., Imber, S. D. and Prank, J. D. The role of non-specific factors in short term psychotherapy. Australian Journal of Psychology, 1966, 18, 210- -ZTT.------------------------------------ Stukat, K. Suggestibility--a factorial and experimental analysis. Almgrist & Willsel: Stockholm, 1956. Volygesi, P. A. School for patients: Hypnosis, therapy and psychoprophylaxis. British Journal of Medical Hypnosis, 1959, ~ 5 ~ , 10-17. Weitzman, B. Behavior therapy and psychotherapy. Psych- ological Review, 1967, 7^, 3OO-317. Whitehorn, J. C. Psychiatric implications of the "placebo effect" American Journal of Psychiatry, 1958, 114, 662. 167 Wilson, Wolf, S Wolf, S Wolpe, , Wolpe, . Wolpin, Wolpin, Zeisset G-. D. Efficacy of "flooding" procedures in de sensitization of fear: a theoretical note. Behavior Research and Therapy, 1967* 5* 138. The pharmacology of placebos. Pharmacological Review, 1959* H* 689-704. , Doering, C. R., Clark, M. L., and Hagans, J. A. Chance distribution and the placebo reactor. Journal of Laboratory Clinical Medicine, 1957* 49* 837-841. Psychotherapy by reciprocal inhibition.- Palo Alto! Stanford University Press, 1958---- ' . and Lazarus, A. A. Behavior therapy techniques. New York: Pergamon Press, 1956. M. and Pearsall, L. Rapid deconditioning of a fear of snakes. Behavior Research and Therapy, 1 9 6 5* 3* 1 0 7-1 1 1. : M. and Raines, J. Visual imagery, expected roles and extinction as possible factors in reducing I fear and avoidance behavior. Behavior Research ! and Therapy, 1 9 6 5* 4, 2 5-3 7. i R. Mf “Desensitization and relaxation in the modification of psychiatric patients' interview behavior. Journal of Abnormal Psychology, 1 9 6 8, j 73* 18-24.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Tension And Anxiety In Deconditioning
PDF
Effects Of Group Behavior Therapy Imagery On Basketball Performance
PDF
Prognostic Expectancy Effects In The Desensitization Of Anxiety Over Invasion Of Body Buffer Zones
PDF
The Effects Of A Self Shock Procedure On Hallucinatory Activity In Hospitalized Schizophrenics
PDF
Mental Imagery As A Function Of Muscular Tension And Suggestion
PDF
Imagery And Response Styles In Desensitization
PDF
Stimulus and response generalization of classes of imitative and nonimitative behavior as a function of reinforcement, task, cues, and number of therapists
PDF
The Effect Of Anxiety And Frustration On Muscular Tension Related To The Temporomandibular-Joint Syndrome
PDF
Factors Of Adaptation And Rehabilitation In Home Hemodialysis
PDF
The Effects Of Nonreward Dissonance And Secondary Reward On Extinction And Attractiveness
PDF
Age, Sex, And Task Difficulty As Predictors Of Social Conformity: A Search For General Tendencies Of Conformity Behavior
PDF
The Enhancement Of Eeg - Alpha Production And Its Effects On Hypnotic Susceptibility
PDF
Transfer Of The Partial Reinforcement Extinction Effect Across Tasks In Normal And Retarded Boys
PDF
A Cognitive Dissonance Analysis Of Conformity Behavior, As Applied To The full Denture Patient
PDF
Effects Of Different Treatment Procedures On Reading Ability And Anxiety Level In Children With Learning Difficulties
PDF
An Application Of A Two-Stage 'Attention' Model To Concept Formation In The Mentally Retarded
PDF
Effects Of Video Tape Feedback Versus Discussion Session Feedback On Group Interaction, Self Awareness And Behavioral Change Among Group Psychotherapy Participants
PDF
Some Factors Affecting Ucr Diminution
PDF
The Effects Of Anxiety And Threat On Self-Disclosure
PDF
The Effects Of Sex, Assigned Therapist Or Peer Role, Topic Intimacy, And Expectations Of Partner Compatibility On Dyadic Communication Patterns
Asset Metadata
Creator
Jaffe, Lawrence William
(author)
Core Title
Non-Specific Treatment Factors And Deconditioning In Fear Reduction
Degree
Doctor of Philosophy
Degree Program
Psychology
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,Psychology, clinical
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Cliff, Norman (
committee chair
), De Nike, L. Douglas (
committee member
), Wolpin, Milton (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c18-651186
Unique identifier
UC11360934
Identifier
6905054.pdf (filename),usctheses-c18-651186 (legacy record id)
Legacy Identifier
6905054.pdf
Dmrecord
651186
Document Type
Dissertation
Rights
Jaffe, Lawrence William
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA