PSYCHOTHERAPISIS' CONFLICTS: DEFENSE PREFERENCES, AND VERBAL REACTIONS TO CERTAIN CLASSES OF CLIENT EXPRESSIONS Thesis for the Degree of Dh. D. MICHIGAN STATE. UNWERSITY Edward J. Barnes 1963 This is to certify that the thesis entitled PSYCHOTHEBAPISTS' CONFLICTS, DEFENSE PREFERENCES, AND VERBAL REACTIONS TO CERTAIN CLASSES OF CLIENT EXPRESSIONS presented by Edward J. Barnes has been accepted towards fulfillment of the requirements for Ph. ngegree mismhszlogy /?55Z;fi71;§;<7¢%32/{~xn_w__ Major professor Due September 20, 1963 0-169 * . I *, _.__._‘.'. _ __ . , - 7 , .-.¥W .g V;— LIBRARY Michigan State Untversity rmeémaew It." ‘ M o K. “meMMtg , i ”j LIBRARY Michigan State Unlversity __ t PSYCHOTHERAPISTS' CONFLICTS, DEFENSE PREFERENCES, AND VERBAL REACTIONS TO CERTAIN CLASSES OF CLIENT EXPRESSIONS By Edward J. Barnes The purpose of this study was to investigate the relationship between therapist conflict (with reSpect to hostility, dependency and sexuality) his primary modes of defense, and his approach- avoidance reactions to conflictual expressions initiated by clients. The study was based on theoretical assumptions derived from psycho- analytic theory, as expressed in writings on countertransference reactions, and on recent formulations concerning defense preferences. Subjects were 26 individuals from two levels of experience: 18 psychotherapy interns, and 8 psychotherapy practicum students. The subjects were ranked on the following variables: hostility, sex, and dependency expressions. Therapist conflict scores were derived from these measures. Defense measures were derived from rank scores provided by two Judges on each of the following defenses: repression, regression, isolation, reaction formation, and undoing. Therapist approach-avoidance reactions were coded from 5# tape-recordings by means of a system (slightly modified) deveIOped by Winder and Bandura and their associates. Having Operationally defined conflict as a discrepancy between a subject's self-rating and the mean rating of a group of evaluators, it was hypothesized that 1) there is a negative relationship be- tween therapist conflict (for hostility, dependency and sex expres- sions) and the level of approach to conflictual (hostility, depend- ency, and sex) expressions, and that 2) non-conflicted therapists, -l 2 Edward J. Barnes as compared to conflicted therapists, respond to client conflictual expressions with relatively greater approach than avoidance reac- tions; it was also hypothesized that 3) therapists with "minus" conflicts receive higher ratings on "repressor" defenses than thera- pists with "plus" conflicts, and that therapists with "plus" con- flicts receive higher ratings on "sensitizer" defenses than thera- pists with "minus" conflicts; finally, that #) non-conflicted thera— pists receive lower overall ratings than conflicted therapists on the use of defenses. The predictions regarding the relationship between therapist conflict and approach reactions were supported for the intern group. They were not supported for the combined intern-practicum group, nor for the practicum group alone. The intern groups approach level for dependency and hostility was significantly higher than that of the practicum group. These differential findings for the two groups were in terms of the influence of therapy experience on therapist approach-avoidance reactions. These findings are considered highly tentative because of the small number of subjects in the practicum group. The hypothesis concerning hostility was not supported in a consistent manner. This lack of consistency was discussed in terms of the "object" of the hostility expression. It may be the case that therapists tended to avoid client hostility expressions when they (therapists) were the objects of such expressions. The predictions concerning the difference in approach reac- tions between conflicted and non-conflicted therapists were fully supported. It was also found that conflicted therapists tended to approach conflictual material less frequently as therapy progressed 3 Edward J. Barnes (in late as compared to early interviews, with the exception of hostility), while the Opposite reaction was observed for non-con- flicted therapists. This was discussed in terms of the increasing intensity of the therapist-client relationship as therapy progresses. The postulates concerning the relationship between direction of conflict and defense modes were not supported. These findings were considered inconclusive because of the lack of independence between the "repressor" and "sensitizer" defense measures. The postulate regarding the relationship between therapist conflict status and degree of overall defensiveness was supported for hostility, but not for sex and dependency. A speculative ex- planation was offered for this lack of consistency. There was an indication that degree of defensiveness is related to the nature and/or magnitude of conflict rather than to its mere presence or absence. These findings were discussed and specific suggestions were made for further research in this area. Approved:_fi CommIttee Chairman Batmfl 42; /Z’/_J° PSYCHOTHERAPISTS' CONFLICTS, DEFENSE PREFERENCES, AND VERBAL REACTIONS TO CERTAIN CLASSES OF CLIENT EXPRESSIONS By {2) _g3 J t Edward J. Barnes A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1963 $529030 é/ad/t% Dedication To Norma whose unswerving belief in me makes this entire venture worthwhile. 11 ACKNOWLEDGMENTS The writer is deeply grateful to his chairman, Dr. Albert I. Rabin, not only for his support during his proficient direc- tion of this investigation, but also for his long continuing belief in and encouragement of the writer. Acknowledgment is gratefully given to Drs. Bill L. Kell, Josephine Morse, and Charles Henley for the constant help extended both during the investigation and preceding it. Gratitude is also extended to Drs. Kell and Morse for the sustained support and encourage- ment which provided the motive force for special aspects of the writer's growth and deve10pment. Special appreciation is extended to Drs. Harry Grater, Bill L. Kell and Josephine Morse who acted as raters of the therapists studied; and to all the therapists who cOOperated so generously in the data collection process. Appreciation is also extended to Dr. William E. Mueller and Phillip Caracena for making available many of the tape- recordings used in this study; and to Hannah Lerman who col- laborated with the writer in coding the tape-recordings. *********§** 111 TABLE OF CONTENTS CHAPTER Page I. INTRODUCTION A. Initial Statement of Problem and Theoretical con81derat10ns oocoocooocooooooocooooooooocoocoooc 1 Bo Concept Of confliCt cooccooooooooocccoooooocoooooc 8 Co Relevant ResearCh oooooocoococoocoocoooooccocooooo 12 D. Defense Preferences and Direction of Conflict .... 16 E0 Final Statement Of PrOblem ooooooooooooooooooooooo 20 II. HYPOTHESES OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 21 III. THE METHODS OF INVESTIGATION Ac SUbJeOtS oooooocooooocoooooooocooocooocoocoooococc 23 B. Therapist Measures lo Ratings occoooococoooooooocooooooooocoocoooooo 26 20 Rating Procedure oooooooooooooooooooooooooocoo 26 3. Reliability Of Ratings ooocoooooooooocoooo 27 3. Evaluation Procedure for Therapist Conflict .. 28 . Measures of Client-Therapist Interaction ..... 31 a. COding ApprOaCh and AVOidance 000000000000 33 be InterJUdge Reliability 0.0000000000000000. 3“ 5. Measures of Therapist Defense Preference 3. Reliability Of Ratings ccooooooocooocooocc 35 IV. RESULTS A. Intercorrelation of Measures ..................... 39 B. Relationship between Conflict Measures and Approach-Avoidance RGECtions 0.0000000000000000... #1 C. Defense Preference and Conflict Status ........... 50 V. DISCUSSION OOOOOOOOOOOOOOOOOOO0.0000000000000000000000 60 VI. SUMMARY AND CONCLUSION OOOOOOOOOOOOOOOOOOOO0.00.0.0... 79 WMCES 0.0.0.000...0.0.CO...OOOOOOOOOOOOOOOOOOOOOO 82 APPENDICES OOOOOOOOOOOOOOOOOOOCOOOOOOOOOOOOOOOOOOOOOO. 86 iv TABLE 1. 2. 3. 6. 7. 8. 9. 10., 11.: 12. 13. LIST OF TABLES Summary of Personal Information on Two Experience Levels 000......OOOOCOOOOOOOOOOOOOOO0.0...0.0.0.0.... Subject's Self Rating, Mean and Standard Error of the Page 25 Mean of Evaluators' Ratings and t ratios for Dependency, HOStility, and Sex Expr3831on oooooocooccoooooooooooc Subject's Self Ratings, Mean Rating of Evaluators and Discrepancy Score for Dependency, Hostility and Sex Expr6331on cooocoooooocoooocooooooooooooooooooooo Defense Rankings: Interjudge Reliability (N = 18) Intern Group 0..0.....0.00000COOOOOOOOOCOOOCOO0...... Defense Rankings Interjudge Reliability (N = 8) Praeticum Group oooocoooooooooooooooooocoooooocoooooo Matrix of Correlations between Defenses (N = 18) Intem Group OIO'OOOCOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Matrix of Correlations between Defenses (N = 8) Praeticum GrOUP ooooooooococccoooccooococoooocoooooco Intercorrelations of Mean Ratings for Dependency, HOStility, and Sex Expr65510n8 ooooooooocoooooooooooo Intercorrelations of Discrepancy Scores for Dependency, Hostility and Sex Expressions ........... Intercorrelations of Self Rating and Mean Evaluator's Ratings for Dependency, Hostility, and Sex Expressions Rank Order Correlations between Observed Discrepancy Scores (Conflict) for Dependency, Hostility, Sex and Approach Reactions (for Total, Early, and Late InterV1eW3) (Intern and Praeticum) cocooccocccoooooco Rank Order Correlations between Observed Discrepancy Scores (Conflict) for Dependency, Hostility, Sex and Approach Reactions (for Total, Early, and Late InterV1eWS) Interns ococo.cocooccooooooooooooococo... Comparison of Approach Reactions of Conflicted and Non-Conflicted Therapists for Total, Early, and Late InterV1eW8 (InternS) oooooocoooooooocoocooocoooocoooo 31 37 37 38 38 40 no 41 “5 1+7 TABLE 14. 15. 16. 17. 18. 19. 20. 21. Differences between Total Approach Ratios for Conflicted and Non-Conflicted Therapists ("Minus" and ”Plus” Directions Considered separately) Interns coco.oooooocoooocoocoooocoooooocooocooooooooo Rank Order Cerrelation of Discrepancy Scores (for Dependency, Hostility and Sex) and Defense Measures (Repressor & SenSItizer) oooooooocoooooooooo Rank Order Correlations of Discrepancy Scores (for Dependency, Hostility and Sex) and Defense Measures (Repressor and Sensitizer) Practicum (N = 8) Rank Order Correlations of Defense Scores (for Dependency, Hostility and Sex) and Defense Measures (Repressor and Sensitizer) Intern (N = 18) Comparison of Conflict Groups (”Plus" and "Minus") on Repressor and Sensitizer Defense Measures Interns (N = 18) oooooooooooooooooooooooocoooooooooo Comparison of Conflict and Non-Conflict Groups on Overall Defense Measure (Interns N = 18) .......... Qualitative Summary of Hypotheses and Principal Findings ooooooooooooooooooooooooooooooooooooooooooo Comparison of Intern and Practicum Groups on Total Approach Ratios (for Hostility, Dependency, and Semlity OOOOOOOOOOOOOOO...OOOOOOOOOOOOOOOOOOOOOOO vi Page 49 52 52 53 55 55 58 59 LIST OF APPENDICES APPENDIX A. List of Personal Questions ........................ B. Instructions and Material for Banking Task ........ C. Sample of Tapes used in Investigation ............. D. E. Scoring Manual for Content Analy818 o o o o o o o o o o o o o c 0 List of Defenses, Definitions, and Instructions for Banking TaSk ooooooooooooooooocoooooooooooooooooooo vii Page 86 87 91 93 105 I. INTRODUCTION A. Initial Statement of Problem and Theoretical Considerations The purpose of this investigation was to examine the relation- ship between psychotherapists' conflicts with respect to certain needs or impulses, defense preferences, and the nature of their ver- bal reactions to client initiated expressions which impinge upon the conflictual area(s). In recent years there has been a trend in psychotherapy theory and research toward viewing the psychotherapeutic process as one in which both therapist and patient engage in a collaborative effort toward a common goal. Increasingly, attention is focused on the necessity of the therapist's personal involvement in the therapeutic process, and the nature of his involvement is seen as influencing his choice of interventions. In tracing the evolution of the doctorbpatient relationship, in terms of its historical context, Szasz, Koff, and Hollander (1958) call attention to the revolutionary notion of mutual participation and partnership inherent in the re- I lationship between the modern psychotherapist and his patient. A survey of the recent literature indicates that the therapy process is commonly studied from the point of view which emphasizes its impact upon the patient. The significance of the therapist‘s role in this process is readily acknowledged, but only recently has psychotherapy research concerned itself with personal attributes of the therapist. Alluding to this situation Strupp (1962) states: 2 ”One of the important failings of studies concerned with the results of psychotherapy has been the lack of specific infor- mation about the method of treatment and the person of the therapist. To describe a therapeutic method by a shorthand label like “psychoanalysis" often conceals more than it re- veals; moreover it is clearly unsatisfactory, for scientific purposes, to treat a complex variable in summary fashion. And a paragragh later: '... regardless of the theory or the technical desiderata, the therapist is a person with his own feelings, attitudes, and life history, and it is manifestly impossible to associate with another human being over a prolonged period of time without being affected by that person in some manner ..." (pp.580-581). Freud (1912) early recognized the significance of the analyst's personality in the therapeutic process. He states: “Amongst the factors which influence the prospects of an analysis and add to its difficulties in the same manner as the resistances, we must reckon not only the structure of the patient's ego, but the personal characteristics of the analyst .... The analyst because of the peculiar conditions of his ' work is impeded by his own defects in his task of discerning his patient's situation correctly and reacting in a manner conducive to cure ...” (p.351). Freud formulated the concept of countertransference to take account of difficulties arising in the therapeutic process which emanate from personal qualities inhering in the therapist. The source of such difficulties was assumed to be more or less eli- minated by the training program the prospective analyst had to undergo before he was considered a fully trained and qualified ana- lyst. Penichel (l9#5) stipulates certain therapist qualities he feels are necessary for the successful carrying out of analytic activities, but he, too, relates them to the nature of the thera- pist's own analysis (p.580). Thus, as Strupp (1962) indicates, despite Freud's concern with countertransference, actually he gave the problem scant attention (p.581). 3 Little (1951) also notes that much attention has been given to the study of transference attitudes of the patient, but until recent years very little attention has been paid to the study of countertransference, which had been assumed to be ”absent except in situations where the analyst was incompletely analyzed.” She feels that such a state of complete analysis does not exist. This assumption has gradually given way to the recognition that countertransference attitudes are present in all analytic situations- perhaps roughly prOportionate to the degree of success of the therapist's analysis - but nevertheless present all (Little, 1951; winnloott, 191m Fromm-Reichmann, 1950; Balint, 1950; Heiman, 1950; Ferenczi, 1950; Sullivan, 1949; Alexander, 1955; Reich, 1951; Berman, 1949). In analytic circles the trend in thinking about the effects of countertransference seems to be away from viewing it solely as a negative element in therapy. Awareness of one's counter- transference reactions and the use of them in a self conscious manner is now emphasized (Alexander, 1955). Strupp (1960) credits Alexander's concept of "corrective emotional experience” as playing a major role in the revival of interest in the therapist's personality as a significant ingre- dient in the therapeutic process. Alexander (1950) states: "No doubt the most important therapeutic factor in psycho- analysis is the objective and yet helpful attitude of the therapist, something which does not exist in any other relationship ...” (p.987). In a more recent paper Alexander (1958) restates his emphasis in these terms: I... “The theory of corrective emotional experience leads to still another technical conclusion. This concerns the most opaque (in my Opinion) area in psychoanalysis, the question of the therapist‘s influence on the treatment process by virtue of .being what he is: an individual personality, distinct from all other therapists” (p.311). In a similar vein, Heiman (1950) states that the ”analyst's emotional responses to his patient ... represent one of the most important tools for this work" (p.81). The assumption is that if the therapist is to be able to use his emotional reactions to the patient in a therapeutic manner, the awareness of such feelings must not be a source of anxiety for the therapist, i.e., the therapist must be relatively free of conflict over the feel- ings. prconflict does exist, the result will be the presence of anxiety in the therapist, whether at the level of awareness or defended against and kept from awareness. From this we would expect that when patient-therapist interactions, which touch up- on therapist conflict areas, are initiated anxiety is aroused in the therapist, and interferes with therapist-patient communi- cation. Hhatever influence the therapist exerts upon the patient in psychotherapy this influence is transmitted through his communi- cations. The relative effects of verbal aspects of the therapist's communications as opposed to non-verbal elements may not be measurable at this time, but there can be little argument that the verbal aspects are exceedingly important (Wolberg, 1954; Strupp, 1962). Thus, it makes sense to focus on the verbal as- pects of the therapist's messages, keeping in mind that tone of voice, gestures, and other non-verbal elements may be important also, even if more difficult to measure. 5 Rogers (1957) formulated certain conditions which he asserts must prevail in the therapeutic interaction if basic personality change is to be affected in disturbed individuals. Two of them are as follows: a) that the client perceives the unconditional positive regard and empathic understanding of the therapist, and b) that the therapist is congruent (to be explained below) in the relationship. The first condition listed places emphasis upon therapist-patient communication, or more specifically upon the therapist's communication to the patient. The implicit assump- tion is that if this condition does not obtain, the other condi- tions necessary for therapeutic change are not communicated to the patient. Thus, the verbal aspects of the therapist's communi- cations to the patient assume great import since it is by this mode, at least in part, that conditions for change are trans- mitted. The second condition gives recognition to the signifi- cance of the therapist being relatively free of internal conflict within the relationship. By congruence Rogers means a lack of discrepancy between one's awareness of his experience and one's actual inner (organismic) experience. When such discrepancy ob- taines, the therapist is vulnerable to anxiety and threat (Rogers, 1961). Discrepancy between onePs symbolization of his experience and actual inner experience is most likely to occur when integra- 'tion of the latter into the former constitutes a threat to the organism. When this state of affairs exists conflict is present. 0n the basis of the foregoing, we would expect that when there is impingement upon an area characterized by a discrepancy between the therapist's symbolized and unsymbolized experience, anxiety will be triggered off in the therapist, with the effect that 6 patient-therapist communication is interfered with by some alter- ation in the therapist's behavior. Thus, it seems that two dif- ferent theories of psychotherapy, psychoanalytic and Rogerian, converge in their formulations regarding the effects of therapist conflict upon therapistmpatient communication. Although studies indicate that expert therapists of diverse theoretical orientations do not agree as to the efficacy of a particular technique at a given moment in the therapeutic inter- action, they do concur in attributing primary significance to the therapist's communication to the patient and in feeling that such communication is influenced by personality factors of the psychotherapist (Strupp, l958a,b,c; Parloff, 1956; Betz and Whitehorn, 1956; Fiedler and Senior, 1952; Fiedler, l950a,b, 1953). An important therapeutic goal is bringing about effective therapist-patient communication. However the affective involve— ment required of the therapist, under certain conditions, may militate against this goal. As Bandura (1960) indicates, a mini- mal condition for resolution ofttpatient's conflicts would seem to be that the patient's conflictual feelings are permitted to occur within the therapy situation. Bandura believes that if certain classes of expressions are anxiety provoking for the therapist, it is expected that he would be less likely to permit or encourage the patient to express himself in these ways, and that when the patient does express tendencies that are threaten- ing to the therapist, it is expected that the anxiety elicited in the therapist would often motivate him to avoid continuation of the anxiety producing interaction. 7 Bordin and Cutler (1958) contend that the verbal behavior of the therapist in the therapy interaction can be classified into one of two general categories: “task oriented" or "ego oriented“ behavior. Behavior which belongs in the second category is said to be countertransference reactions, and not oriented toward help- ing the patient undergo therapeutic change. In "ego oriented” responses, the therapist himself becomes the focus, and his re- sponses are directed toward reducing threat, anxiety, maintaining self esteem, etc. This defensive behavior is apt to occur when the therapist encounters patient productions which are conflictual for him. The implicit assumption is that anxiety and/or guilt serve as the motive for the defensive behavior. In a review of recent writings on countertransference, Cohen (1955) noted that all treatments of the topic have included in their definitions of the concept attitudes which are both con- scious and unconscious; attitudes which are reSponses to both real and to fantasied attributes of the patient; attitudes which are stimulated by unconscious needs of the therapist; attitudes which are stimulated by sudden outbursts of affect on the part of the patient; attitudes which arise from reSponding to the patient as though he were some previously important person in the therapist's life, and attitudes which do not use the patient as a real object but rather as a tool for the gratification of some unconscious need. What is the common criterion for cats- gorizing the foregoing reSponses as countertransference? Cohen postulates that the common factor is the presence of anxiety in the therapist, whether conscious or unconscious. B. .A Concept of Conflict The determination of therapist conflict is a significant operation because of the focus of this investigation. Conflict is generally defined as the presence within the individual of two incompatible needs or impulses. Fenichel (1945) states that ”the neurotic conflict, by definition, is one between a tendency striving for discharge and another tendency that tries to pre- vent this discharge” (p.129). Psychoanalytic theory postulates that when forbidden impulses - forbidden because of internalized standards - press for expression, super ego anxiety and/or guilt are mobilized, and serve as the motive for calling forth defen- sive operations; these Operations serve to bind or ward Off further anxiety which would be experienced as beyond the ego's capacity to master. Repression is one such defense against over- whelming anxiety. Repression, commonly used in conjunction with other defensive maneuvers, refers to the withholding or expulsion of an idea or affect from conscious awareness (Freud, A., 1946). It is believed to be an outgrowth of the primitive defense of denial. Fenichel (1945) states that ”persons with specific countercathexes frequently avoid certain situations, objects, or activities ... sometimes without being aware of the avoidance, sometimes with full awareness of it” (p.169). Dollard and Miller (1950) state that no human being Operates so effectively that all his tendencies are congruent and well integrated. In their conceptualization of conflict, as in psychoanalytic theory, two tendencies are posited in the individual; an approach tencency and an avoidance tendency with reSpect to a goal. 9 In this investigation conflict is operationally defined as a discrepancy between an individual's self rating and the mean rating of a group of judges with respect to some specified be- havior system. Our conceptualization closely approximates the psychoanalytic formulation. Our operational definition sug-' gests a modified formulation to read "the presence within the individual of a need or impulse which is incompatible with the manner in which he perceives himself.” Thus if an individual's self evaluation with respect to a class of behaviors, for example hostility, is discrepant in this respect from that of others who are well acquainted with him, that person is said to be conflicted with reference to hostility; .His self percept which, in part, is an accumulation of internalized moral precepts, is incompat- ible with the recognition of the meaning of behavior said to be‘ hostile. Given this conceptualization, conflict viewed in terms of an absolute discrepancy between a subject's self evaluation and judge's evaluation of him, and/or in terms Of a discrepancy which reaches a specified statistical level. In either case it is possible for the subject to be categorized in one of three ways depending upon his self-rating as compared to the ratings of his colleagues. He could be categorized as having a ”plus” conflict, a ”minus" conflict, or as being non-conflicted. A "minus” conflict is defined as one in which a given therapist- subject's self rating is lower than the average of the ratings assigned to him by his colleagues. A ”plus" conflict is defined as one in which a given subject's self rating is higher than 10 the average of the ratings assigned to him by his colleagues, and non-conflict is defined as the absence of a discrepancy be- tween the two sets of ratings. But the psychological meaning of this concept of conflict has to be determined. Previous investigators (Cutler, 1953; Rigler, 1957; Munson, 1960) believe that a minus conflict is in- dicative of the presence Of a need or impulse which is incompat- ible with the self percept. It seems reasonable to assume that, in a self-evaluation, if the subject has to underemphasize or deny the meaning of certain aspects of his behavior, it is because they are not in accord with the way he must see himself. Aware- ness of the meaning of these behaviors is forbidden because they are the overt expressions of needs or impulses which are threat- ening. Thus, he underrates himself with respect to these be- haviors. The contradiction is between his self report regarding his behavior and the meaning assigned to it by others. This kind of maneuver sounds like the defense Of repression and/or denial e In the case of the ”plus” conflict, where the subject as- signs an exaggerated rating to certain classes of his behavior, it seems that he is highly sensitized to, or vigilant regarding, the expression of these behaviors. For example, the subject with a "plus” conflict for hostility assigns a meaning to his behavior which departs markedly from.that assigned by his assoc- iates. Such a person would seem to be highly sensitized to any expression hostility on his part. Apparentlyfhe would have fewer expressions of this impulse available to him. He is forced to ll restrict himself from the original impulse. The fact of slight hostility being perceived by the subject as strong hostility seems to be predicated upon anxiety about hostility. By this maneuver the subject seems to be protecting himself from manifesting be- haviors that may be objectively judged as strongly hostile. This kind of maneuver sounds like reaction formation. So it seems that "plus” conflicts are also "real" conflicts and not artifacts stem- ming from the manner in which conflict is Operationally defined. At any rate, previous research (Cutler, 1953, 1958; Rigler, 1957; Munson, 1960) suggests that they are genuine conflicts, since their effects upon therapist behavior were the same as that of ”minus” conflicts. This suggests that absolute discrepancy (direction of discrepancy not considered) is more crucial than direction Of dis- crepancy in its effect upon therapist behavior. "Non-conflict" is defined as the absence of an Observed dis- crepancy (or a discrepancy which fails to reach the out-off—point) between the subject's self-rating and the average of the ratings assigned to him by his associates. It is probably the case that an individual could see himself as others see him and yet not be satisfied with what he sees. Would this be indicative of conflict? The answer to this question would depend upon one's definition of conflict. According to our definition such an individual would not be conflicted. The significant element here is that the indi- vidual does not have to distort or deny his behavior. Bandura (1956) apparently had a similar idea in mind when he suggested that the indi- vidual who is able to evaluate his behavior objectively (objecti- vity defined as interobserver agreement) seemingly would possess greater insight into those characteristics in question than the 12 individual who denies or distorts the meaning of his behavior. For the purposes of this investigation it is of no consequence whether a therapist is considered non-conflicted or as having insight into the personality factor under study. In either case it is expected that he will respond with relatively greater approach than avoidance reactions to client dependency, hostility, and sexual expressions than will the therapist with a "plus" or ”minus” conflict with respect to these behaviors. C. Relevant Research Cutler (1953, 1958) studied the relationship between Speci- fic therapist personality factors and moment to moment inter- actions Of patient and therapist. He first identified areas of conflict, which were Operationally defined by discrepancies be- tween the therapist's self ratings and others' ratings of him. He was able to demonstrate that the therapist's recollection of his own and of the patient's behavior during therapy was selec- tively distorted in accordance with the therapist's conflict areas, when this recollection was compared with the transcript of a tape-recording of the session. The therapist's report of his own as well as of the patient's behavior overemphasized, dis- torted, or omitted those behaviors which impinged upon the thera- pist's area of conflict. He demonstrated also that the therapists as a group responded less adequately to patient expressions that impinged upon their conflict areas than they did to patient expressions which were relatively non-conflictual for them. The responses judged to be "ego oriented" were considered less 13 adequate, while those judged to be “task oriented" were con- sidered more adequate. Bandura (1956) investigated the relationship between the psychotherapist's competence, his anxiety level, and his self- insight into his anxieties. He attempted to establish that: a) competent psychotherapists are less anxious than those judged to be less competent, and b) competent therapists possess a greater degree of insight into the nature of their anxieties than do less competent therapists. His method was to use rating scales in each of three areas which he defined as being cone flictual: dependency, hostility, and sexuality. His group of subjects rated each other; the average rating of a given subject by the group constituted a measure of his level of anxiety for a given area. Insight into anxiety was defined in terms of the relative discrepancy between the therapist's self rating and the mean group rating. Anxiety and insight measures were compared with ratings of therapeutic competence made by therapy super- visors. A significant negative relationship between competence and anxiety was found, but no significant relationship between competence and.insight. An important limitation Of which Bandura is aware is that he based his measure of what happens in therapy upon supervisors' ratings rather than upon direct study of the therapeutic process itself. Another limitation of which Bandura does not seem to be aware is his use of a global measure of anxiety which is not specific to the events occurring between patient and therapist. For example he did.not attempt to relate therapist anxiety to specific patient productions which were known to be conflictual for the therapist. l4 Rigler (1957) investigated some determinants of therapist behavior, and among other things, postulated that certain thera- pists would exhibit greater anxiety during periods when dealing 'with conflict areas than when dealing with non-conflict areas. Conflict was determined by noting the disparity or concordance between ratings on a group of traits, made by the subject hime self and by a group of raters acquainted with him. The hypo- thesized relationship between arousal of therapist anxiety, which was assessed by GSR recordings, and the appearance of therapist conflict areas was not sustained when approach to conflict area was initiated by the therapist; however,when it was the patient who initiated the conflictual material, the relationship was decisively supported. The discrepancy was explained by postu- lating a differential need or ability to defend against anxiety arousing conflict under the two conditions.‘ Munson (1960) investigated the relationship between patterns of client resistiveness and counselor responses. The independent variable was therapist conflict with regard to nurturance and acquisitive needs. The dependent variable was therapist re- sponse to clients who are Oppositicnal and to those labeled as gratifying to the therapist. Among other things Munson hypo- thesized that conflicted counselors would be unable to respond to Oppositicnal clients in a manner which would facilitate thera- pist - client communication, whereas non-conflicted counselors would be able to respond in a facilitating manner. Conflict was determined by the method used by Cutler and Rigler, noted above. The hypothesis with reapect to nurturance was supported, but 15 no significant differential trend in approach to clients was found for conflicted and non-conflicted counselors with regard to acquisitiveness. Bandura and others (1960) investigated therapist's approach- avoidance reactions to patients' expressions of hostility. They predicted that the degree of therapist anxiety about hostility would differentiate therapists in the extent to which they dis- play approach and avoidance reactions to hostile patient-expres- sions. It was postulated that therapists with high hostility anxiety would tend to reapond to patient hostility expressions with avoidance reactions, whereas therapists with low hostility anxiety would tend to respond with approach reactions. Anxiety level of subjects was determined by colleagues' ratings. “Approach” reactions were defined as "reSponses designed to eli- cit further expressions of hostile feelings, attitudes, and behavior." ”Avoidance” reactions were defined as ”reSponses designed to inhibit, discourage, or divert the patient's hos- tile feelings, attitudes, and behavior.” Patient hostility was defined as ”expressions of dislike, resentment, anger, anta- gonism, opposition, or critical attitudes.“ Results indicated that therapists who typically expressed their own hostility in < direct forms were likely to permit and encourage patient hosti- lity expressions to a greater extentethan therapists who ex- pressed little direct hostility. Essential to all psychotherapy (approaches) is the necessity of the therapist's helping and understanding, and the theore- tical material as well as the research reviewed indicate that 16 the medium through which this takes place is communication for which the therapist has the major responsibility; and further, the nature of the therapist's communication, or the lack of it, seems to be contingent upon certain forces, feelings, or needs within himself. IL Defense Preferences and Direction of Conflict Some investigators believe that direction of conflict is related to particular modes of defense (Cutler, 1953, 1958; Rigler, 1957; Munson, 1960). Fenichel (1945) addresses him- self to the question of individual differences in defensive reactions. He points out that neurotic characters react to their experiences more or less rigidly with the same reaction patterns. ”They are ... fixated ... to certain mechanisms of defense. Even normal persons who are flexible and able to react adequately show habits in this reapect ... The problem of fixation to certain defense mechanisms is but a special case of the more comprehensive problem of the relative constancy of character traits in general” (p.523). Anna Freud (1946) makes a tentative statement regarding the connection between special forms of defense and particular neurotic reactions, as for example, between repression and hysteria, and obsessional neu- rosis and the defenses of reaction formation, isolation, unp doing and regression. Research on defense mechanisms with non-clinical subjects is informative. Generally the results seem to support the no- tion that defensive modes for a given individual tend to be l7 . stable, at least within the experimental situation. In the per- ceptual area, for example, some subjects may react with a low- ered threshold for tension producing material, and other sub- jects respond with:a heightened threshold (Carpenter, et.a1., 1956; Eriksen, 1952a). In the area of memory some subjects tend to recall tension producing material better than neutral material, while for others the reverse seems to be the case (Carlson, 1954; Eriksen, 1952b). Janis and Feshbach (1955) con- ducting an investigation into the effectiveness of fear arous- ing prepaganda obtained results somewhat incongruent with those cited above. It was found that a message with strong fear ap- peal was less effective in producing behavior (verbal recall) in a given direction than was a minimal fear appeal. The sub- jects as a group presumably adOpted similar protective mechanisms. They explain the failure of strong fear appeal by recourse to the concept of defensive reaction, and suggest that the arousal of an excessive amount of emotional tension may stimulate in a subject a defensive reaction that is incompatible with the re- call at a later date of the recommendations contained in the strong fear message. This sounds like repression. Goldstein (1959) hypothesized that certain defensive reactions facilitate the acceptance of propaganda, while others interfere with its acceptance. He selected subjects on the basis of their charac- teristic modes of responding to tension-arousing stimuli. The two modes of responding were labeled ”cOping” and "avoiding.” The results support the hypothesis that a strong fear appeal receives greater acceptance among ”cOpers” as compared to "avoiders", while a minimal fear appeal receives greater acceptance 18 among 'avciders' than among “cOpers”. But the obtained pat- tern of acceptance was largely due to the marked differential effectiveness of the two appeals on the ”avoider“ group, and.not as originally predicted, to any differential acceptance by "capers". Eriksen (1951) also differentiates two modes Of re- sponding to threat. Re labels them ”repression“ and ”sensitization". This study indicated that ”repressors" as compared to "sensitizers" had a higher threshold for material related to unacceptable needs. Eriksen (1954) also found that "repressors' tended to recall more completed than incompleted tasks. Gordon (1957, 1959) investigated interpersonal predictions, which were regarded as largely determined by response sets. The problem was to deter- mine how the assumed similarity response set changes as a func- tion of changes in the predicted object, characteristics of the predictors, set to predict, and presence of the predicted object. The sample was composed of an equal number of repressors, sen- sitizers, and neutrals. It was found that sensitizers assume similarity between self and predicted Object less frequently than did repressors. Altrocchi and others (1960), viewing the common elements in the results of the foregoing studies, derived the following de- finitions or descriptions Of ”repressors" (avoiders) and 'sensitizers' (cOpers), with respect to defensive Operations utilized in face of threat. They state that 'repressors' are those who tend to use avoidance (withdrawal), denial, and re- pressicn potential threat and conflict as a primary mode of adapting; and that “sensitizers” are those who tend to be alerted 19 to potential threat and conflict and tend to respond more readily with manifest anxiety, and to use intellectual and obsessive defenses: reaction formation, regression, undoing, and isolation. Altrocchi (1961) selected sensitizers and repressors accord- ing to the above descriptions and compared their'perceptions of themselves with the way in which they assume others perceive them. He found that sensitizers assumed more dissimilarity be- tween description Of self and the way others perceive them than did repressors. Repressors were also found to have a more posi- tive self-concept than sensitizers (described themselves as higher on Dominance and Love), and they are a more homogeneous group in self-description. Altrocchi states that many sensi- tizers see themselves as different from and as less desirable than other peOple. Altrocchi's description of repressors and sensitizers in terms Of defensive modes seems to bear some relationship to what we have termed direction of conflict. The ”plus” conflict seems to be related to the 'sensitizer' mode of responding to threat. It will be recalled that a “plus" conflict means that a subject overrates certain aspects of his behavior as compared to average ratings of him.in this respect by his colleagues. The ”minus“ conflict seems to be related to the repressor mode Of responding to threat. The hmtnus' conflict derives from the subject's underrating of himself with respect to certain aspects of his behavior as compared to the average of the ratings assigned to him by his colleagues. 20 E. Final Statement of Problem We are now in a pagition to give a complete statement of the problem. The purpose of this investigation was to examine the relationship between therapists' conflicts with reference to hostility, sexuality, dependency, primary modes of defending against threat, and the nature of their verbal reactions (approach- avoidance) to client initiated expressions which imping upon the conflictual area(s). II. HYPOTHESES .A. The general hypothesis with reSpect to therapist conflict and his approach-avoidance reactions was as follows: There is a negative relationship between therapist conflict and the level of therapist approach reactions to client initiated expressions which impinge upon therapist conflict area(s). From this general hypothesis the following specific hypo- theses were derived: 1. 2. 3. 4. 5. There is a negative relationship between therapist hostility conflict and the level of approach to hostility expressions. There is a negative relationship between therapist dependency conflict and the level Of approach to dependency expressions. There is a negative relationship between therapist sex conflict and the level of approach to sex expressions. Therapists who are non-conflicted with reference to hostility as compared to conflicted therapists re- spond to client hostility expressions with relatively greater approach than avoidance reactions. Therapistswho are non-conflicted with reference to dependency as compared to conflicted therapists reSpond to client dependency expressions with relatively greater approach than avoidance reactions. Therapists who are non-conflicted with reference to sexuality as compared to conflicted therapists respond 21 22 to client sex expressions with relatively greater approach than avoidance reSponses. 'B. The general hypothesis regarding the relationship between therapist conflict and defense preferences is as follows: Thera- pists who overrate themselves (in comparison with a group of judges' ratings of them) with respect to the expression of hostility, dependency, and sexuality (”plus” conflict) utilize a particular configuration of defenses to cOpe with threat; whereas therapists who underrate themselves with respect to the expression of these behaviors ("minus” conflict) cOpe with threat by utilizing a different configuration of defenses. l. Therapists determined to have ”minus” conflicts receive higher ratings on the use of defenses, said to characterize repressors, than therapists who are determined to have ”plus” conflicts. 2. Therapists determined to have ”plus" conflicts receive higher ratings on the use of defenses, said to charac- terize sensitizers, than therapists who are deter- mined to have minus conflicts. 3. Therapists who are non-conflicted receive lower over- all ratings on the use of defenses than therapists who are conflicted. III. THE METHODS OF INVESTIGATION The independent variable in this design is therapist con- flicts The dependent variables are therapist approach-avoidance reactions to client expressions which impinge upon therapist conflict area(s), and therapist primary modes of defending against threat e A. Subjects Subjects were 26 individuals from two levels of experience: 17 interns, 1 post doctoral trainee, and 8 practicum students. Criteria defining each level were: 1. 2. Interns - These were advanced graduate students accepted for training in psychotherapy at the MSU Counseling Center, and working there a minimum of twenty hours each week. Each intern had completed all the required courses in psychotherapy offered by either the Depart- ment of Psychology or the College of Education. For purposes of this investigation the post doctoral trainee was considered an intern since he did not dif- fer from the other interns on variables thought to be relevant; i.e., with reference to number of hours of personal therapy, number of clients seen in which he was supervised, and total number of clients seen in psycho- therapy. ngcticum - These were graduate students who had com- pleted first courses in psychotherapy; they saw a small number of clients and were supervised intensively. A11 23 24 ~had completed or were currently taking courses in psychOpathology and personality theory. All were in the Department of Psychology. All available interns who engaged in psychotherapy during the Spring and Fall terms of 1962 and the Winter and Spring terms of 1963 were used as subjects. Only those practicum students engaged in psychotherapy during the 1962-63 school year were used as subjects. Even though the two groups differed with reSpect to level of therapy experience, it was expected that variability within groups would be relatively small. Since subjects within a given group were of the same general level of training and ex- perience, differences in therapist reactions to emotionally significant material that may be due to differing amounts of experience and training seemingly were controlled for all es- sential purposes. Information was obtained on variables which could have some influence on the nature of the therapist's verbal reactions, such as experience level, personal psycho- therapy, etc. See List of Personal Questions, Appendix A. Table 1 summarizes the personal information obtained from the subjects. The average number of hours of personal therapy, number of clients seen under supervision, and total number of clients seen in psychotherapy; each differentiates the intern from the practicum group. The sample of therapists included 3 females. 25 Table 1.. Summary of Personal Information on Two Experience Levels = 26 Intern Practicum Personal Therapy 0 hours 3 1 l - 20 hours 3 3 21 - 100 hours 7 3 over 100 hours 5 1 Mean of those with therapy 124.2 (N=15) 67.5 (N=7) Standard Deviation 69 22.4 Median. 80 30 Number of clients seen Mean Median Mean Median under supervision 25.1 20 10.2 11.5 Standard Deviation 14.9 5.7 Total number of clients seen in psychotherapy 27.3 20 10.4 14.0 Standard Deviation 16.9 5.15 Curriculum Number Number Clinical 7 8 Counseling ll 0 Sex of Therapists Number Number Male 15 8 Female 3 0 26 B. Therapist Measures 1. Ratings Ranked measures were obtained for each of the subjects on each of the three variables: hostility, sexuality, and depend- ency. Each of the variables was defined in terms of descrip- tion of and inference from observed behavior. The definitions of the variables were as follows: Hostility Tendency to attack or belittle contrary points of view, to become angry when Opposed, to be critical of others, to ask questions or make statements that puts the other person "on the Spot”; to direct sarcastic remarks to others; may seem angry or irritated frequently. In general, tendency to be oppositional, antagonistic, argumentative, dero- gatory, critical, complaining, or resentful. Dependency Tendency frequently to ask for and to accept help, advice, suggestion, and direction from others; eager for approval by others; trusting and eager to please; concern to find out what others think; readiness to follow instructions and to do what is expected; lets others make decision; strong tendency to accept leadership of others; ready agreement with authority figures; tendency to form strong attachments, and to make as many friends as possible. Sexuality Frequent participation or readiness to participate in dis- cussions about sex; enjoys telling and listening to jokes involving sex; talks about one's own sexual interests and attitudes given the slightest opportunity; tendency to be seductive and flirtatious either openly or in subtle ways. 2. Rating Procedure Each subject in a group (intern or practicum) ranked all subjects including himself, with respect to expression of be- havior relevant to each of the three variables. A rank of l was 27 to be assigned to the subject judged to express the greatest amount of relevant behavior (hostility, dependency, or sex re- pectively) of his group, and the bottom rank was assigned to the individual judged to express the least amount of relevant be- havior. Each subject was asked to provide three sets of rank- ings, one for each of the three variables: hostility, dependency, and sexuality. Each ranking task was separated from the preceding one by an interval of a week. In those instances where limited inter- action rendered ranking of an individual or individuals unfeas- ible, the ranker was asked to omit such individual(s) from.the group being ranked. Instructions and materials for the ranking task have been reproduced in Appendix B. (a) Reliability of ratings. In accordance with the procedure presented by Guilford (1954), the assigned ranks of the intern group were converted into standard scores, with a mean of 5 and a standard deviation Of 2. The standard scores were treated as numerical ratings for purposes of determining reliability of ratings for this group e Since the raters were free to omit ranking those subjects with whom limited interaction rendered such ranking unfeasible, as expected, the number of ratings a given subject received varied from one to another, and the raters were not the same for all subjects in the sample. However no subject was ranked by fewer than 11 other persons and approximately 72% (or 13 subjects) were ranked by 14 or more raters. Horst (1949) 28 develOped a generalized reliability for such a set of measures. fPhe reliability coefficients for the intern group were .93 for lucstility, .94 for sex, and .88 for dependency, with the average reliability for the group being .92. The ranked measures of the practicum group were not con- ‘verted to standard scores since the number of subjects was less than 10. However, since the set of ranks was complete, the Kendall coefficient of concordance was applied to provide an estimate of reliability. Reliability coefficients for the prac- ticum group were .70 for dependency, .67 for hostility, and .82 for sex. 3. Evaluation procedure for therapist conflict. The evaluation method used for this purpose is modeled after that of Bandura (1956), and the rationale for determination of therapist conflict is analogous to that develOped by Cutler (1953. 1958). For the combined group (intern and practicum) the conflict measure was defined in terms of absolute discrepancy between a subject's self rating and the mean rating of the group for that subject (self minus group mean). The subject whose self rating was equal to the mean group rating was considered to be non-conflicted; a self rating which was lower than the group mean was considered "minus" conflict, and a self rating higher than the group mean was considered ”plus” conflict. Since the ranked scores for the intern group could be trans- formed rnto standard scores, it was possible to subject them to 29 statistical procedures for the purpose of defining conflict as a discrepancy of a specified magnitude. A priori it was decided that those areas would be designated as conflictual where the subject's self rating differed from the mean group rating by at least one standard deviation of the mean. When a discrepancy failed to meet this criterion for a given area for a given subject, that area was designated as non-conflictual. Table 2 presents the results of this analysis for the 18 interns, for the three variables: sex, hostility, and dependency. As noted above, since the number of subjects in the prac- ticum group was less than 10 their! ranks were not transformed into standard scores. Thus, the foregoing analysis could not be performed for this group. This presents some difficulty but does not prevent our using the practicum group, since ab- solute discrepancy also can be utilized as the conflict measure. Table 2 presents the observed discrepancies between subjects' self rating and the mean of the group of evaluators for the practicum sample. Since one set of rankings only was secured for each of the variables investigated, the reliability measures referred to earlier are indicative of interjudge consistency rather than to stability of the rankings over time. For the same reason we have no measure of the stability of the conflict scores e 30 Table 2. Subjects Self Rating, Mean Rating of Evaluators and Discrepancy Score for Dependency, Hostility and Sex Expressions Interns Dependency Hostility Sex Expression Expression Ex ression Subjects Self Group Discre- Self Group Discre- Self Group Discre- (N-18) Rating Mean pancy Rating mean pancy Rating Mean pancy Score Score Score .A n - 14 7.00 6.50 .50 5.00‘ 2.50' 2.50* 3.00 2.87 .13 n E 17 5.00 5.53 -.53 3.00 6.53 ~3.53* 4.00 1.82 2.18* n E 15 6.00 4.40 1.60* 4.00 3.62 .38 6.00 4.47 1.53* n 2 16 4.00 5.50 -l.50* 6.00 6.06 -.06 7.00 5.37 1.63* n E 14 6.00 4.50 1.50* 4.00 4.40 -.40 7.00 5.53 1.50* n: 16 2.00 3.00 -1.00 6.00 6.69 -.69 6.00 6.44 -.44 n S 15 5.00 6.27 -1.27* 6.00 5.00 1.00* 6.00 4.27 1.73* n E 15 5.00 4.40 .60 5.00 4.40 .60 6.00 4.73 1.27* n i 17 6.00 4.88 1.12 5.00 6.67 -1.67* 4.00 5.94 -1.94* n i 13 8.00 7.46 .54 4.00 1.69 2.31* 5.00 2.15 2.85* n15 17 6.00 5.59 .41 6.00 5.76 .24 7.00 6.47 .53 n E 16 2.00 5.31 -3.31* 5.00 4.44 .56 9.00 7.76 1.24* n E 15 6.00 6.40 -.40 9.00 5.00 4.00* 7.00 4.53 2.47* n E 12 5.00 4.75 .25 5.00 4.31 .69 7.00 4.27 2.73* n 3 14 5.00 4.93 .07 6.00 3.87 2.13* 6.00 5.85 .15 n E 11 3.00 2.36 .64 7.00 7.58 -.58 5.00 3.82 1.18* n S 12 5.00 2.50 2.50* 5.00 5.83 -.83 7.00 5.34 1.66* n E 11 6.00 5.36 .64 6.00 5.36 .64) 8.00 6.14 1.86* * Indicates conflict state. Subject's self rating differs from mean of evaluators' ratings by one standard deviation or more. Sign of discrepancJ score indicates direction of discrepancy. 31 Table 3. Subjects Self Rating, Mean Rating of Evaluators and Discrepancy Score for Dependency, Hostility and Sex Expressions Practicum N - 8 Dependency Hostility Sex Egpression Egpression Expression Subjects Self Group Discre- Self Group Discre- Self Group Discre- (Nh8) Rating Mean pancy Rating Mean pancy Rating Mean pancy §ppge 4§core Score S 7.00 6.43 .57 6.00 2.86 3.14 4.00 4.86 -.86 in 6.00 3.71 2.29 5.00 5.14 -.14 6.00 4.43 1.57 U 6.00 4.43 1.57 4.00 1.71 2.29 3.00 1.28 1.72 V' 6.00 2.28 3.72 3.00 2.00 1.00 4.00 2.57 1.43 W 5.00 6.28 -1.28 6.00 3.57 2.43 7.00 5.86 1.14 X 3.00 2.57 .43 7.00 5.14 -3.14 5.00 2.43 2.57 Y 3.00 3.00 0 7.00 4.14 2.86 4.00 4.28 -.28 z 5.00 3.43 1.57 6.00 5.86 .14 8.00 6.00 2.00 4. Measures of Client-Therapist Interaction Focus was upon sessions early in the psychotherapeutic process, i.e., on the interval encompassed by the first through sixth therapy sessions. Where possible two tape recordings were Obtained early in this interval (lst through 3rd sessions) and one at the end of the interval (4th through 6th session). However, this was not possible in all cases, and in two in— stances only one recording could be Obtained from each of two interns. One practicum student was unable to provide tapes in time for the coding process and only two recordings were ob- tained from.each of the remainder of the practicum students, one at the beginning of the aforementioned interval and one at the end of the interval. A total of 44 tapes were coded for the intern group and 14 for the practicum group. A list (with iden- tifying information, and therapist approach-avoidance ratios) 32 can be found in Appendix C. It was necessary to devise criteria for judging therapist reactions to client productions which were defined as conflict- ual or non-conflictual for the therapist. Bandura and others L (1960) develOped a system for coding therapists approach and avoidance reactions to certain patient productions. In their dependency study Winder et.a1., (1962) used a coding scheme analogous to that of Bandura. It was decided that Winder's analogue of Bandura's system would be utilized in this investi- gation. Some minor modifications were made in the scoring system. With reference to "patient's response categories,“ Winder's definitions Of sex and sex anxiety have been replaced by Murray's (1956) definitions. The sub-categories of ”non: ”factual information," and ”acknowledgement” were placed in the miscellaneous subcategory. Only the main categories were scored in this investigation. The coding system is reproduced in full in Appendix D. The primary elements of the system are presented in the following: 1. Thp_Scoripgpypipt A unit was an interaction sequence con- sisting of the beginning patient statement, the therapist's immediate response, and the immediately following patient response e 2. Patippt Behavior Categpries: Categories scored were de- pendency, hostility, and sex. Dependency was defined as any expression of approval seeking, information seeking, demand fcr'initiation of activity by the therapist, help seeking, com- pany seeking, and ready agreement with others. Hostility was 33 defined as any statement that includes description of or ex- pression of unfavorable, critical, sarcastic, depreciatory .remarks, Opposition antagonism, argument, expression of dis- like, resistance, irritation,annoyance, anger; expression of aggression and punitive behavior. Sexuality was defined as statements that include descriptions of or expressions Of a sexual nature, seductiveness and flirtation. 3. Therapist thavior Categories: Categories scored were therapist approach and avoidance reactions. Approach reactions were defined as responses of the therapist intended to elicit from the patient further verbalizations of the tOpic under discussion. Avoidance responses were defined as reactions by ' the therapist intended to inhibit, discourage, or divert the patient from further verbalizations about the tOpic under discussion. : (a) Coding therapist approach and avoidance reactions. Two coders were involved in the coding task; both completed a training period utilizing tapes not included in this investi- gation. Subsequently from a pool of 58 tapes a sample of 39, used for reliability purposes, was selected by means of a table of random numbers. A random order for coding the tapes was determined for each coder. Each individual coded the sample independently and according to his coding order. Approach ratios are defined as the sum of frequencies of approach reactions over the sum of frequencies of approach plus the sum of the frequencies of avoidance reactions. Avoid- ance ratios are defined as the sum of frequencies of avoidance 34 reactiOns over the sum of frequencies of avoidance plus the sum of frequencies of approach reactions. Since the focus was on therapist approach and avoidance reactions to client expressions of dependency, hostility, and sexuality, assessment of agreement on the following ratios was deemed crucial: approach and avoid- ance ratios for each of the three client categories of depend- ency, hostility, and sex. These ratios were determined for each interview of each subject in the reliability sample by both coders. The two sets of ratios were converted to prOpor- tions, and the Pearson Product Moment Correlational Method was applied to the porportions to determine interjudge agreement. (b) Interjudge agreement coefficients for approach ratios were as follows: .95 for dependency (N=39), .80 for hostility (N=39), and .90 for sex (N=27). When tapes in which it was agreed there was an absence of client sex expressions are in- cluded, the agreement coefficient for sex rose to .96. Reliance upon the coding of one individual (the writer) for the remaining 19 tapes in the writer's sample was consid- ered acceptable in view of the level of agreement in the relia- bility sample. In later analyses the combined ratios of the two coders were utilized. 5. Measures of Therapist Defense Preferences Three judges were asked to rank independently the subjects J on each of the defense mechanisms said to be preferred by "repressors" and ”sensitizers". Repressor defenses were avoidance, denial, and repression. Sensitizer defenses were isolation, reaction for- mation, undoing, and regression (Altrocchi, 1960, 1961). The 35 judges were Dr. B. L. Kell, Assistant Director of the MSU Coun- seling Center for Training, Dr. H. Grater, Assistant Director of the Center for Counseling Services, and Dr. J. Morse, CO-leader of the Psychotherapy Practicum for psychology students. Drs. Kell and Grater rated the intern group, and Drs. Kell and Morse rated the practicum group. The judges were provided with a list of the subjects to be ranked, a set of 3 by 5 cards - each card containing the name of a subject - and with ranking sheets containing name and definition of defenses on which subjects were to be ranked. They'were asked to assign a rank of l to the subject judged to use a given defense the most in his group, and to assign the last rank to the subject judged to use that defense the least. This procedure was followed for each of the defenses. The judges were asked to rank the subjects first on "sensitizer' defenses. (a) Reliability of ratings Two judges were unable to distinguish,on a behavioral basis, repression on one hand versus avoidance and denial on the other. Thus, for the 'repressor' defenses the subjects were ranked only on the defense of repression. Interjudge agree- ment was determined by applying the rank order technique (rho). The rank order correlations for the two judges ranking the . intern group are presented in Table 4, and those for the two judges ranking the practicum group are to be found in Table 5. -___-IT' A list of defenses with their definitions, instructions and materials for the ranking task can be found in Appendix E. 36 These measures seem to have adequate reliability in so far as interjudge consistency is concerned, but we have no information regarding the stability of the measures over time. The rankings made by each judge were combined and a new ranking for specific defenses was derived. The combined rank- ings of each of the five defenses are used as the measures of defense in subsequent analyses. The combined ranks of each defense were intercorrelated. The results for the intern group are presented in Table 6, and those for the practicum group are shown in Table 7. With re- ference to the intern group, the percentage of the variance that undoing, isolation, and repression share with each other would seem to indicate that these defense measures are not independent. A similar but weaker trend is observed between the intercorrelated measures of the practicum group. The mean intercorrelation of the defense measures for the intern and practicum samples is .60 ( p<.01) and .52 (p >.05) reSpec- tively. This lack of independence between defense measures will be discussed later. 37 Table 4. Defense Rankings: Interjudge Reliability (N = 18) Intern Group Defgnse rho p Spearman-Brown Correction Reaction Formation . 71 < .01 ' . 23 Undoing .73 < .01 . IBOlation e68 < e01 681 Regression .77 < . 01 . 87 Repression . 78 < . 01 . 88 Table 5. Defense Rankings: Interjudge Reliability (N = 8) Practicum Group Dgfgnse rho L Spearman-Brown Correction Reaction Formation .81 (.05 > .01 .89 Undoing .90 4.01 .95 Isolation .68 .05 > .01 .81 4 Regression .86 < .01 .92 Repression . 88 < .01 .94 38 Table 6. Matrix of Correlations between Defenses (N = 18) Integp Group ‘ Reaction Un- R ressi n Formation Isolation Re ression Doi Repression .89** .67** ~09 .54' Reaction Formation .82** .35 .73** Isolation .48* .80** Regression ,59ee ’ p e01 ** p< e01 Table 7. Matrix of Correlations between Defenses (N = 8) Practicum Group :Reaction Unp Rgpzession Pappgtion Isolat;on Rpgpessicn Doing Repression .80* .82* .10 .30 Reaction Formation .58 .41 .51 Isolation .35 .54 Regression .83* '. p (e05 > e01 IV. RESULTS A. Intercorrelations of Measures The initial step in the data analysis was to intercorrelate the mean of the ratings for each of the three variables: depcn-' dency, hostility, and sex expression. This analysis was to deter- mine whether these measures were independent. A separate analysis was performed for the intern and practicum groups reapectively since these measures for the two groups were not comparable. These intercorrelations are presented in Table 8. They were found to be small and statistically insignificant. The mean of the inter- correlations for the intern and practicum groups were - .03 and .18 respectively. Conflict (discrepancy) scores for the three areas were also intercorrelated to determine whether conflict is generalized or primarily specific to a given area. The results of this analysis for both groups are presented in Table 9. None of the correlations for the practicum group proved significant (p;>.05); a significant negative relationship (rs= -.43, p<:.05) between discrepancy scores for dependency and hostility was found for the intern group. The final set of intercorrelations was between the subjects' self rating and the mean of the evaluators' ratings for each of [the areas of hostility, dependency, and sex expression. For the intern group the intercorrelations were found to be of moderate size positive and statistically significant. For the practicum group a positive relationship was obtained between self and judge's ratings for sex expression only (r8= .76, p(.05). These inter- correlations are presented in Table 10. 39 40 Table 8. Intercorrelations of Mean Ratings for Dependency, Hostility and Sex Expression ___, Intern Group N=18 Practicum Gpgup N=8 Dependency Hostility Sex Dependency Hostility Sex Dependency -.34 -.18 .09 -.05 Hostility .26 .49 Table 9. Intercorrelations of Discrepancy Scores for Dependency, Hostility and Sex Expressions Intern Group N=l8 Practicum Group N=8 Discrepancy Scores Discrepancy Scores Dependenpy Hostility Sexg_5ependency Hostility Sex Dependency -.43* -.04 -.49 .28 Hostility .11 -.27 *Significant at the .05 level (one tail) 41 Table 10. Intercorrelations of Self Rating and Mean of Evaluators Ratings for Dependency, Hostility and Sex Expression Intern Group_N=_18 ,Ppacticum_group N= 8 Self Rating Self Rating Evaluator's Mean Rating Dependency Hostility Sex Dependency Hostility Sex Dependency .52* ~59 Hostility .42* .21 Sex .61* .76* *Significant at the .05 level or less. NO correlations were computed between the self rating and the conflict measures, since these measures are not Operationally inde- pendent. The rank order intercorrelations between the defense mea- sures have already been presented in the preceding chapter. B. Relationship Between Conflict Measures and Therapists Approach Avoidance Reactions Rank order correlations between conflict measures and approach ratios to client's dependency, hostility, and sex expressions were computed to test the hypotheses concerning the relationship between therapist conflict and approach-avoidance reaction. For each interview the number of times the therapist approached or avoided a client's expression, coded for one of the foregoing categories, was determined. The prOportion of approach reactions was calculated for each therapist, for each tape recording, with reference to each of the categories of dependency, hostility, and sex. The approach prOportions were used to test the hypotheses. 42 This was considered acceptable since the approach and avoidance prOportions are not independent. (If p equals the prOportion of approach responses for a given therapist, for a given reaponse cate- gory, then the prOportion of avoidance reSponses, in this regard, would equal 1 minus p.) Since predictions were made concerning the direction of the expected relationships, all tests of signifi- cance were one-tailed. The initial step in testing hypothesis A-l, involved calcu- lating the rank correlation between the absolute discrepancy scores (direction of discrepancy not considered) for hostility expression and proportions for total approach hostility; i.e., the approach hostility proportions for all tapes. The resulting correlation co- efficient was non-significant (r= -.05). The same analysis was performed for dependency and sex discrepancy scores and total ap- proach dependency and sex ratios. The results of this analysis are presented in Table 11. The resulting correlations were -.24 for dependency and -.45 for sex. The latter correlation is significant at the .05 level; the former is in the expected direction but does not reach the criterion of the .05 significance level. Further tests of hypotheses A-l, A-2, and A-3 bear on the re- lationship between discrepancy scores and approach-avoidance reac- tions at given temporal points in the therapy session. This analy- sis classified interviews as early Or late. Interviews 1 through 3 were categorized as early and interviews 4 through 6 as late. Then, rank correlations were calculated separately for early and late interviews. The resulting correlations are presented in Table 11. The only correlation reaching the .05 level of signifi- cance is that for sex conflict and approach sex in late interviews.v However, with the exception of hostility conflict and approach may. 11 p: a: 43 hostility in late interviews the remaining correlations are in the expected direction. The foregoing findings were for the combined practicum—intern group. Since the practicum and intern samples represent two ex- perience levels in terms Of therapy experience, a separate analysis was done for the intern and practicum groups. Analyses identical to those above were performed between discrepancy scores for hos- tility, sex, and dependency and total approach ratios for each of these are presented in Table 12 for the intern group. Only the correlation for hostility discrepancy scores and total approach hostility failed to reach the .05 level of significance, however, it was in the expected direction. None of the results for the practicum group were significant. The interviews were again categorized as early and late and approach ratios correlated with discrepancy scores for the three ‘behavior categories. These correlations are also presented in Table 12. Only the correlation between hostility discrepancy scores and approach hostility for late interviews (4-6) proves to be non-significant. It is also noted that with the exception of hostility, the relationship between conflict scores and approach ratios is stronger in late than in early interviews. For the practicum group only the correlation for sex discrepancy scores and total approach sex was significant at the .05 level (r= 4.74). The foregoing analyses are based upon conflict defined as an absolute discrepancy between the subject's self rating and the mean rating of a group of evaluators for some system of behavior. The analyses up to now have implicitly assumed a direct relationship between the magnitude of a discrepancy and the magnitude of app- roach ratios. This is probably not the case; at least the data Of 44 .ch>Huooamop msoa>aousa cued one mauoo pom ma one om ma 2 opens wow wow uncoxo .m3ca>uous« coma eon mfiamo you NN .maoa>uousw Houou pom munwz Ho>en no. no noooaaaowam t som.a mm.» smo.: wow NH. om.o mo.n auaaaunom mm.: ¢N.o eN.: hosoocoaon mung wanna Hooch mung mfiumm annoy coma maumm Hooch wow uaaaumom . zosoosoaoa newness: uowstoo asowuomom rumouaa< sneauooam one suousH Amaoa>nousH cued one .haumm .Houoa acmV occauooom sooouaa< one xom .hufiawunom .aocopsoama pom Auoaamsoov mouoom mosoaouomaa cunHOnn< consume macauoaonuoo acouo xcom .aa canoe 45 .%Hc>auucamca m3oa>nousa coma you NH .aasmc now ma .m3oa>ucuoa deuce new nan.z muons wow wow uaooxo m3~a>ucusa coma new ma .heuoo new ad .m3caShousH Hosea How wdnuz Ho>oa no. no screamsewam w smm.u woe.u emo.o wow #0. wmm.o mm.n zuaaaunom .em.- .ee.- .mm.- aooeeoeooo mung mflumm annoy mung manna. Hence coon macaw HouOH wow muwaaumom mosoooomon nonzero: uoaawcoo msofiuooom nooouaam aaousH Amaca>ncusH mung one .maamm .Houoa HomV msoauooom nomonaa< one wow .muwawumom .moooeocaon you Auowamsoov nouoom huooacuoman ouoHOmn< soo3uon mooauoaoauoo porno xsmm .NH canoe 46 this investigation suggest that such a relationship does not hold. We need to observe the relationship between approach-avoidance re- actions and discrepancies that are equal to or exceed one standard deviation of the mean of the evaluators' rating. The following analysis does this. The practicum group is not included in this analysis, since their ranked scores could not be transformed into standard scores because of the small number of subjects. Such a transformation was necessary in order to treat the ranks as number- ical ratings. Given this transformation we could apply statistical procedures to the ratings so as to determine whether or not a given discrepancy was indicative of conflict. The results of this analy- sis were presented earlier in Chapter 3 (Table 2). With reference to hypotheses 4-A, 5-A, and 6-A, the prediction was that non-conflicted therapists as compared to conflicted thera- pists respond to client expressions, which fall into the categories of sex, hostility, and dependency, with relatively greater approach than avoidance reactions. The conflicted and non-conflicted groups were compared with reSpect to approach ratios for sex, dependency, and hostility. The difference between the approach ratios for the two groups was tested by the Mann-Whitney U test (Siegal, 1956). The difference between the two groups for total approach dependency resulted in a U Of 18 (p.05). The relationship between sex discrepancy scores and overall defense measures was not significant (r-.13). For the final test of hypothesis B-3, only those discrepancy schbres (plus and minus) which were statistically significant were utilized as conflict measures. This analysis was done for the in- tern group only, since our decision rule defining conflict could be applied to their discrepancy scores. The prediction was that the overall defense scores of the conflict group would tend to be larger than those of the non-conflict group. For each area, dependency, hostility, and sex, there was a conflict and a non-conflict group. For each area the two groups were compared on overall defense mea- sures. The significance of the differences between groups were tested by Mann-Whitney U tests. The results of the analysis are presented in Table 19. Only the differences between groups conflicted and non-conflicted with reference to hostility reached the criter- ion level of significance. Again it is noted that conflict defined fink. firhuhwm . ’0 55 Table 18. Comparison of Conflict Group (”Plus" and ”Minus") on Repressor and Sensitizer Defense Measures Intern Group (N = 18) Dependency Hostility Sex Defense Type Plus Minus Plus Minus Plus Minus n1 31 n2 R2 n2 R2 n1 R1 n2 R2 n1 31 Sensitizer 3 8.5 3 12.5 5 20 2 8 13 98 l 7 U = 3 U = 5 U = 6 Repressor 3 10 3 11 5 20.5 2 7.5 13 100.5 1 4.5 U-‘-'-2 U=5.5 U21} Table 19. Comparison of Conflict and Non-Conflict Groups on Overall Defense Measure Intern Group (N = 18) Dependency Hostility Sex Conflict Non-Conflict Conflict Non-Conflict Conflict Non-Conflict n1 Hi n2 R2 n1 R1 n2 R2 112 Hz n1 R1 6 50.5 12 120.5 7 87.5 11 83.5 I“ 133 4 38 * Significant at the .05 level (one tail) n = number of cases in smaller of two independent groups = number of cases in larger of two independent groups = sum of ranks assigned to groups whose sample size is n l 2 ' I5 NwaN H = sum of ranks assigned to group whose sample size is n 56 as a discrepancy which deviates from the group mean by one standard deviation apparently is a more sensitive indicator of conflict than absolute discrepancy. The nature of both the data and the sample of subjects added considerably to the complexity of the analysis. The result is that the findings are often difficult to remember and to keep clearly differentiated. In an attempt to overcome this problem, a summary statement of the hypotheses and the principle findings is presented in Table 20. In testing each hypothesis the analysis of the data involved several components. Separate analyses were performed for total approach reactions and approach reactions in early and late interviews. Separate analyses were also done for the combined group of subjects, and for each group (intern and practicum) separ- ately. It will be recalled that the personal data indicated that the intern and practicum groups represent two therapy experience levels, the findings presented in Table 20 indicate that the hypotheses re- garding the relationship between conflict and approach to client expressions that impinge on the conflict area essentially were not supported by the data from the practicum group; nor by the data from the combined groups. However, when the same analysis was performed for the intern group alone the hypotheses were supported in a con- sistent manner. No hypothesis was advanced concerning the rela- tionship between experience level and therapist approach-avoidance reactions. But in view of the foregoing findings, the relationship between experience level and approach reactions was analyzed. To accomplish this the approach ratios of the intern and practicum groups were compared. Comparisons were made for approach ratios 57 for hostility, dependency and sex. The significance of the dif- ferences between groups was tested by the Mann-Whitney U test. The results of this analysis are presented in Table 21. The groups differ significantly with reSpect to total approach ratios for de- pendency and hostility. In both instances the intern group's ap- proach level is greater than that of the practicum group. The groups do not differ significantly with respect to approach ratios for sexuality. oophoaasm mamospoahs n .m .N oopaOQQSm no: mamoSpoah: u .m .2 .H mmosopamsomoo Hamnoeo a soda“ mmosobdmsohoo HHmaobo Inoo hosopsoaoo a poaawsoo How a .Nmm how .m .z .N mosoosoaoo you .m .z .N mmoSoadmsmmoc Ham mmosobdmsohop Iaobo @ podausoo Hamuobo a poaamsoo apdaapmos poo .m .H auaaasmos woe .m .H .m .2 .H mum mdmospoaam .m .2 .H .m .2 .H .m .2 .H mum mamonpoaam .m .2 .H .m .2 .H .m .2 .H Hum mammnpoaam As aaaae .m .H 6-4 mdmospoaam a) madam .m .H m-< mamuspoaam mamapwopns mama .m .z .m macabnopsa same. a sauce .m .H e-« mamospoaam Acosmoamasmam Sascha «cosmofimaswam mzoa>aoQCa mama and mpasmoav mama» nomoaaam mpasmonv a magma .m .z .N uneven Hmpop .m .z .m mamasnopna hays. .m .z .N mzodb mamabhopsa msoHSHOpsa mama names“ Hugo» .m .H opwa a mass. .m .H a Hmpop woe .m .H mud mamonnoaam .m .2 .H mass“ .m .H .m .2 .H Nu< msmospogam m30a>kopsa mama one Hopes woe .m .z .m maoapuopna .m .2 .H sans. you .m .H .m .2 .H a-< mammspoaam m a msouo Baoapomam enema shopsH macho snopsH asoapomam dosanaoo mamonpoamm mwsapsflh uwsacndm HmmHosanm use mmmmSponhm molwwmsssm abdpduaHdSG .om canes 59 Table 21. Comparison of Intern and Practicum Groups on Approach Ratios (For Hostility, Dependency, and Sexuality) for Total Interviews. Approach Ratios Dependency Hostility, Segpglity_ n1 7(p) 7(p) 6(p) n2 18 18 17 31 60.5 58 94 32 264.5 267 182 U 32.5* 30** 39 * p) 005 (.10 (two tails) ** p = .05 (two tails) (p) Indicates the practicum group V. DISCUSSION This study was designed to test two general hypotheses: A) There is a negative relationship between therapist conflict and the level of therapist approach reactions to client initiated res- ponses which impinge upon therapist conflict area(s), and B) Thera- pists who overrate themselves (in comparison with a group of judges‘ ratings of them) with respect to the expression of hostility, de- pendency, and sexuality ("plus” conflict) utilize a particular con- figuration of defense to cope with threat; whereas therapists who underrate themselves with respect to the expression of these be- haviors ("minus" conflict) cOpe with threat by utilizing a differ- ent configuration of defenses. Subsumed under each general statement are several specific hypotheses; six under A and three under B. In the following dis- cussion hypothesis A in its several components will be discussed first, followed by discussion those of B. Hypothesis A Because of the characteristics of both the sample of subjects and data, a thorough test of each hypothesis necessitated several analyses; one for the combined sample (practicum and intern) and total approach ratios} an analysis for the combined groups for early and late interviews, and one for the practicum and intern groups separately, since these groups represented two therapy experience levels. wITotal approach ratio refers to total approach reactions for a given category, for the total number of interviews, for a given subject. 60 61 Hypothesis A-l states that: There is a negative relationship between therapist hostility conflict and approach reactions to client hostility expressions. This hypothesis was not supported by the data for the combined" groups. However, when separate analyses were performed for the two“ groups, the hypothesis for the intern group was supported for early but not late interviews. The relationship between hostility con- flict and approach hostility in early interviews was in the pre- dicted direction, and significant at the .05 level of significance (Table 12). The finding for total approach ratios did not reach/ the criterion level or significance, though it was in the expected direction. The hypothesis was not supported by the data of the. practicum group. (Table 20) The findings suggest a differential interaction between thera- pist hostility conflict and approach reactions at different time- points in the therapy interval investigated; the significant rela- tionship, between conflict scores and approach reactions in early interviews, disappears in late interviews. InSpection of the data suggests that therapists with larger discrepancy scores are pri- marily responsible for this differential interaction. The ranks of those with smaller discrepancy scores remain fairly stable over interviews while the ranks of those with larger discrepancy scores change over interviews, with a resulting trend toward smaller diff- erence scores (larger approach ratios for late interviews). On first thought, one would expect the opposite, since the therapist- patiet relationship would provide a context more conducive to eli- citing therapist reactions indicative of conflict. Cohen's (1955) review of writings on counter-transference indicates that such re- actions are more likely to occur in later than in beginning stages 62 of therapy. Before any conclusion could be forthcoming from these findings, information regarding the object of client hostility ex- pressions would be necessary. The findings by Bandura and others (1960) suggest that therapists with hostility anxiety react differ- ently to hostility expressions depending upon its object, therapist or other. It does not seem likely, but possibly the object of hos- tility expressions in interviews, defined as late in this investi- gation, tended to be other~than-therapist, while those in early interviews tended to be therapist. Hypothesis A-2 states that: There is a negative relationship between therapist dependency conflict and approach reactions to client dependency eypressions. This hypothesis was not supported by the data for the com- bined intern-practicum group for total approach ratios, nor did the findings for early and late interviews reach the criterion level of‘ siginificance, but the relationships were in the expected direction= (Table 11). When separate analyses were performed for the intern and practicum groups, the hypothesis was consistently supported by~/ the data for the intern group (Table 12). The hypothesis was not ' supported by the data for the practicum group (Table 20). Again the differential interaction between therapist conflict scores and approach reactions for early and late interviews was ob- served. In this instance the differential reaction is present in both early and late interviews. The relationship between the order of conflict scores and approach reactions is stronger in late than in early interviews. Inapection of the rank scores reveal again that the ranks differences of therapists with small discrepancy scores tend to remain fairly stable from early to late interviews, while those of therapists with large discrepancy scores tend to 63 increase from early to late interviews. An increase in rank diff- erence is associated with a decrease in approach ratios. This trend is more in keeping with our expectations as outlined above in the discussion of the findings for hostility. Hypothesis A-3 states that: There is a negative relationship betwepn therapist sex conflict and approach reactions to client sex expressions. This hypothesis was supported by the data for the combined v intern-practicum.group for total approach ratios (Table 11). It was also supported by the data for the combined group for late in- terviews, but not for early interviews, although the latter rela-/ tionship approached the criterion level of significance (t = 1.68, p ) .05< .10)? When separate analyses were done for the two groups, the hypothesis was supported for the practicum group for total ap4/ proach but not for early and late interviews. The results for the intern group for total approach did not reach the criterion level of significance but was in the expected direction. The hypothesis was supported for early and late interviews, and again the relation- ship between discrepancy (conflict) scores and approach reactions was stronger in late than early interviews (Table 12). The discussion of the results thus far has been based upon conflict considered as an absolute discrepancy. The following discussion of the findings for the last three hypotheses under A are based upon conflict considered as a discrepancy which deviates from the mean of the evaluators' ratings by one standard deviation or more. As indicated previously only the intern group was in- volved in this analysis. N-2 —ZSignifIcance of r tested by t = r significant at .05 level (one tailed tesg). 64 Hypothesis A-4 states that: Therapists who are non-conflicted with reference to hostility, as compared to conflicted therapists respond to client hostility egpressions wlph relatively gpeater approach than avoidance reactions. This hypothesis was supported by the data for both total ap- proach and approach in early interviews, but the hypothesis was not supported by the findings for approach reactions in late interviews (Table 13). These findings are analogous to those discussed with ‘ reference to hypothesis A—l. Hypothesis A-5 states that: Therapists who are non-conflicted with reference to dependency, as compared to conflicted therapists, respond to client dependency expressions with rplatively greater approach than avoidance reactions. This hypothesis was supported by the data consistently. The/ differences between conflict and non-conflict groups were in the predicted direction, and significant at or beyond the criterion level of significance for total approach and approach reactions in early and late interviews (Table 13). InSpection of the mean of the ranks for the conflict group indicates that it tends to decrease from early to late interviews. With reSpect to approach ratios in early interviews the mean of the ranks is 6.28, and in late inter- views the mean is 5.18. With respect to the non-conflict group, these values are 10.21 and 11.22 for early and late interviews respectively. This finding seems to suggest that not only are the approach proclivities for the two groups different in initial inter- views, but also that differences in approach tendencies tends to become greater as therapy progresses over the therapy interval in- vestigated. Table 13 indicates that the difference between the 65 groups in early interviews is significant at the .05 level, and the difference for late interviews is significant at the .01 level. A corollary implication is that the two groups tended to behave differently, with respect to approach behavior, in late as compared to earlier interviews. This difference in.approach behavior for early and late interviews, with reference to the non-conflict group, does not reach the criterion level of significance but it approaches significance (U=l6, p=.10, two tails). This difference for the conflict group proves to be statistically significant (U=8.5, p=.02, two tails). The data indicate that as therapy progressed the con- flicted therapists tended to approach the conflictual material less frequently, whereas the non-conflicted therapists tended to approach such material with greater frequency. It should be remembered that these findings pertain only to the therapy interval investigated in this study. Also only two temporal points in the interval were compared. If finer time intervals were determined, the differences noted above might break down. Hypothesis A-6 states that: Therapists who are non-conflicted with reference to sexuality. as compared to conflicted therapists, reapond to client sex egpressions with relatively greater approach than avoidance responses. This hypothesis was also supported in a consistent fashion by the data. Differences between groups were in the predicted di- rection, and significant at or beyond the criterion level for total approach and approach for early and late interviews (Table 13). Again the trend toward differential approach reactions between groups appeared. For the conflict group the mean of ranks de- creased from early to late interviews, while the mean for the non- conflict group tended to increase. However, in this instance / I k ) difference in approach reactions over interviews did not reach the criterion level of significance for either group, though the direction of the difference for the two groups were congruent with those observed in discussion of the fifth hypothesis. The impli- cations of the findings in this instance should be analogous to those discussed in conjunction with hypothesis five. Before proceeding with the discussion of those hypotheses un- der B, an attempt will be made to extend the preceding discussion, with the goal of providing a general overview of the central trends. In general the data did not support the hypotheses (A-l, A-2, and A-3) when the intern and practicum groups were combined, nor were they supported when separate analyses were performed for the practi- cum group alone. The single exception was with reference to the relationship between sex conflict scores and total approach sex ratios. For the intern group, the hypothesis concerning the rela- tionship of hostility conflict and approach hostility received the least support, while the hypothesis concerning the relationship of dependency conflict and approach dependency was fully supported. The hypothesis regarding sex conflict and approach reactions was fully supported except in the instance of total approach ratios (Table 20). Thus the relationship between therapist conflict scores and approach reactions was less differentiated for hostility than for sex and dependency. One might inquire as to possible reasons for this. One possible explanation might be that therapist approach reactions to hostility expressions are, in part, a function of the object of the expression. As indicated earlier, Bandura and others (1960) found that when the therapist is the object of the expressed hostility, there is less of a tendency to approach than when the object is other-than-therapist. 67 It is noted that the magnitudes of the significant correla- tions are only moderate. This is not surprising since, undoubtedly, factors other than therapist conflict state influence his approach~ avoidance reactions. Bandura and others (1960) found that thera- pists who diSplayed a high need for approval tended to discourage or avoid patient hostility expression. Winder et.al. (1962) found patient characteristics to be a variable in therapist approach- avoidance reactions to patient dependency expressions. The indi- cation was that certain patients make dependency bids which are deemed unapproachable from the standpoint of usual psychothera- peutic concepts. Higler (1957) found that if the therapist ini- tiates discussion in a conflictual area, he is less likely to avoid subsequent client expressions in that area. The fact that conflict measures and approach reactions for the group with less therapy experience (practicum) consistently fail to demonstrate the ex- pected relationship, suggests that experience level may be a factor in therapist approach behavior. To check this out, the approach ratios of the intern and practicum groups were compared (Table 21). The two groups were found to differ significantly with reSpect to total approach to dependency and hostility expression. In both instances the intern group's approach level is greater than that of the practicum group. The difference between groups with res- pect to total approach ratios for sexuality was not statistically significant. These findings plus the fact that the approach ratios and conflict measures for the practicum group consistently fail to demonstrate the expected relationship seem to suggest that exper- ience is a factor in approach reactions for the categories of ex- pressions investigated. If these findings are reliable, an im- plication is that with increasing therapy experience, one learns 68 to select and to respond effectively, assuming that approach reac- tions are more effective than avoidance reactions. Fiedler (1950b) found that less experienced therapists reSpond in ways more discre- pant from the ideal than more experienced therapists. As Bandura (1960) indicated, a minimal condition for the resolution of a pa- tient's conflicts would seem to be that the patient's conflictive feelings are permitted to occur within the therapy situation. Thus it appears that approach reactions, on the part of the therapist, are more effective reSponses. In this reapect Winder pp_§l. (1962) found that when expressions of dependeney and aggression are approached by the therapist, the relative frequency of such expressions is sustained or increase and further, that when such expressions are approached the patient tends to remain in treatment. If, on the other hand, there is little positive or even negative reinforcement, then the patient tends to avnid the treatment situation. A comparison of the findings of the first three hypotheses, A-l, through Aa3, with the sesond three, Am4 threugh Am6, gives rise to an interesting observation. In the former instance the re» lationship between conflict scores and total approach ratios for hostility was not signifieant, but in the latter instance all dif- ferences were significant at or beyond the criterion level of sign- ificance (Table 13). With reference to the first three hypotheses, an observed discrepancy was considered indicative of conflict, where- as in the instance of the last three, a discrepancy was required to deviate from the mean group rating by one standard deviation or more before it was considered as indicative of conflict. This seems to indicate that the latter procedure of determining conflict pro- vides the more sensitive measure for testing the hypotheses in this investigation. 69 An analysis was performed to determine the relative effects of direction of conflict upon therapist approachnavoidance behavior, even though previous research and our interpretation of theory led the writer to assume that "plus" and "minus“ conflicts have the same influence or bear the same relationship to therapist approachm avoidance reactions. The findings suggest that though "plus" and J. "minus" conflicts generally tend to bear a similar relationship to approach-avoidance reactions, there was a tendency for "plus" con- flict to be more strongly associated with a tendency to avoid conflictual material (Table 14). Apparently those subjects who were overly sensitive to the expression of certain behaviors, on their part, (exaggerated perception of those behaior expressions in them- selves) tended to avoid client expressions in those areas with great- er frequency than did those subjects who minimized these behavioral expressions in their self evaluation. This seems to bear some rela- tionship to "perceptual vigilance". The exaggerated self evalua- tion seemingly would be indicative of undue sensitivity about cer- tain behavioral expressions, and theoretically would be suggestive of anxiety about its expression. This apparent trend is at variance ‘ with the findings of previous investigators (Cutler 1953, 1958; Rigler, 1957; Munson, 1960) who found that plus and minus conflicts had the same effect on therapist reactions. This finding has some implication for Bandura's (1956) study of anxiety, insight, and therapeutic competence. The findings revealed no significant re- lationships between therapist's degree of insight into the nature of their anxieties and ratings of psychotherapeutic competence. The significant factor here is that Bandura defined insight as a "plus" discrepancy and lack of insight as a "minus" discrepancy. 70 Our findings suggest that "plus" conflict is also indicative of a lack of insight. This may be possible reason he did not find a significant relationhip between therapist's insight and ratings of competence. However, because of the small number of cases in each group, this finding must be treated as tentative. Neverthe- less it does raise the question of whether direction of conflict is a variable in therapist approach-avoidance reactions. With the exception of hostility, it was found consistently that conflicted therapists tended to approach conflictual material less frequently as therapy progressed, while the opposite reaction was observed for non-conflicted therapists. With reference to hostility, inspection of ranks do not indicate a clearly defined pattern, but there was a tendency for conflicted therapists to have larger approach ratios in late than early interviews. One is tempted to say that these findings suggest that as therapy pro- gresses non-conflicted therapists assume a more active role, while conflicted therapists assume a less active role over time in the conflictual areas. It should be remembered that when reference is made to "conflicted therapist”, it is meant to refer to conflict for a given area. No subject in this investigation was found to be free of conflict for all areas investigated. But given this quali- fication, the above generalization is not justified. Only sessions relatively early in the therapy process were included in this study. We have no way of knowing whether or not these differential trends are maintained beyond the fifth or sixth interviews. Within the framework of this design it is not possible to know how much of this finding may be a function of the relative inexperience of the therapists, all of whom were still in training. 71 The general trends of the findings, as far as they relate to the hypotheses are positive for the intern sample, and thus offer some support for the theoretical notions advanced in writings on countertransference (Cohen, 1955; Reich, 1951; Little, 1951; Holt and Luborsky, 1958; Sullivan, 1949; Bordin and Cutler, 1958). They also provide additional support for the findings of previous studies, which take therapist approach-avoidance reactions as the point of focus (Cutler, 1953, 1958; Rigler, 1957; Numson, 1960; Bandura, _p§_§l, 1960). The findings in this investigation must be held as tentative because of the small number of cases and the small sample of behavior or each therapist. The nature of the sample, the manner in which subjects were selected, and the small phase of the therapy process investigated severely limits generalization of the findings. Within the confines of these limitations, we can conclude that the findings indicate that if certain classes of expressions are con- flictual (anxiety provoking) for the therapist, he is less likely to permit or encourage the patient to express himself in these ways; or when the patient does express tendencies that are conflictual for the therapist, the anxiety elicited tends to motivate the there- pist to avoid a continuation of the conflictual expressions, or the therapist tends to respond to the conflictual material in an ego- oriented rather than a task-oriented manner. Hypothesis B The hypotheses subsumed under the general formulation B will be discussed at this point. Hypothesis B-l predicts a negative re- lationship between direction of conflict and the level of ratings on repressor defense measures. The data for the combined intern- practicum group does not support the hypothesis. None of the 72 relationships between repressor defense measures and discrepancy scores reached the criterion level of significance; however the relationship between sex discrepancy scores and repressor defense scores is in the predicted direction, and approaches the criterion level (r8= .30, p (.10) .05). The results of this analysis are presented in Table 15. The hypothesis was not supported when a separate analysis was done for the intern and practicum groups (Table 16 and 17). Hypothesis B-2 predicts a positive relationship between direc- tion of conflict and the level of ratings on sensitizer defense measures. Again the data for the combined inter-practicum group does not support the hypothesis (Table 15). A significant relation- ship was found between dependency discrepancy scores and sensitizer defense measures, but the relationship was in the direction cp- posite to that predicted. The hypothesis was not supported when separate analyses were done for the intern and practicum groups (Tables 16 and 17). In fact, for the practicum group a sizable and significant relationship in the cpposite direction was found between dependency discrepancy scores and sensitizer defense mea- sures (Table 16). This suggests that the data of the practicum group was primarily responsible for the significant relationship in the cpposite direction observed in Table 15. This negative finding presents a problem, which will be discussed later. The final test of hypotheses B-1 and B-2 involved comparing the ratings of ”plus" conflicted and "minus" conflicted therapists on repressor and sensitizer defenses. Only the intern group was used in this analysis, since only their'discrepancy scores could be subjected to our decision-rule regarding conflict and non-conflict. 73 None of the differences between groups for either sensitizer or repressor scores proved to be statistically significant; in fact, none approach the criterion level of significance. Hypothesis B-3 predicts a positive relationship between dis- crepancy (conflict) scores and level of ratings on the use of overall defenses (combined repressor and sensitizer defense ratings). The data for the combined practicum-intern group did not sup- port the hypothesis. When an analysis was performed for the groups separately, the hypothesis was supported by the data of the practi- cum group for hostility discrepancy scores and overall defense mea- sures, but it was not supported with respect to dependency and sex discrepancy scores and overall defense measures. None of the corr- elations for the intern group were statistically significant. But when conflicted and non-conflicted subjects were compared the hy- pothesis was supported with reSpect to hostility, but not with respect to sex and dependency (Table 19). We can state that the first two hypotheses under B were not supported by the data. On first blush it appears that on the basis of the evidence we can conclude that there is an absence of a sta- tistically significant relationship between direction of conflict and preference for either a "sensitizer" or “repressor" defense mode. In fact the data suggest that there is a tendency for a given sub- ject to be rather consistent in his use of all five of the defenses used in this study. That is, if he is rated high on the use of one defense he tends to be rated in a similar manner with respect to the others. This finding seems to offer evidence against Altrocchi's (1960, 1961) thesis that given individuals tend to prefer the use of certain defensive modes to defend against threat. But inspection 74 of Tables 6 and 7 reveal that the "repressor" defense measure (repression) shares a large amount of variance with the ”sensitizer" defense measures (reaction formation, isolation and undoing). This lack of independence of the defense measures poses a problem. We have no way of knowing whether the high intercorrelation of the defenses indicate that the subjects tend to be consistent in the extent to which they use these defense or whether the consistency means merely that the raters adOpted a common frame of reference in the rating task. In other words are the ratings valid, or do they merely represent interjudge agreement about a common frame of reference? To answer this question an external standard against which the ratings could be compared would be necessary. If the ratings proved to be indicative of the true state of affairs, then the obtained results would seem to provide some evidence against the notion of preferences for given defenses in cOping with threat, considering of course the limitations placed upon the generaliza- bility of the findings because of the nature of the sample, and the small number of cases upon which the findings are based. If the correlations merely represented agreement between judges con- cerning a common frame of reference, the results would offer no evidence against the notion of defensive preferences, and we would simply have a phenomenon akin to the "halo effect". For this rea- son the meaning of the finding in the Opposite direction for the practicum group (Table 16) cannot be determined. In view of the foregoing, we can conclude that the question of whether or not given individuals tend to adOpt certain defensive modes when experiencing conflict (anxiety), must be left to future research. This points up the fact that other means of determining therapist defensive tendencies should have been used. However to the writer's knowledge 75 there is no more objective method of assessing defenses which could have been employed with the sample studied. Hypothesis 3 received partial support from the data. While therapists who were conflicted with reference to hostility tended to have higher overall defense ratings, no significant differences were found between defense measures of the subjects conflicted and non-conflicted with reference to sex and dependency. The inconsis- tency of the results presents an interesting problem. The factor pointed out in the foregoing discussion regarding the validity of the ratings has some relevance here. However, the lack of inde- pendence of measures would seem to be less crucial in a test of this hypothesis since we are concerned with overall defensive be- havior, rather than differential defensive behavior. If the re- sults cannot be attributed to an unreliability factor, then they suggest that degree of defensive behavior bears some relationship to the nature of the conflict which theoretically, gives rise to the defensive activity. InSpection of the data in Table 2 suggests that hostility conflict measures tend to be of greater magnitude than dependency and sex conflict measures. Inspection also indi- cates that the mean of the ranks of defense scores for those con- flicted with reference to hostility tends to be larger than the mean of the ranks for the groups with dependency and sex conflict; 8-4 for the dependency conflict group, 12.5 and 9.5 for the subjects conflicted with reference to hostility and sex respectively (Table 19). Thus it seems that the degree of defensiveness is associated with the magnitude or intensity of the conflict. This accords with what one would expect on the basis of psychoanalytic theory. But why should the conflicts of greater magnitude occur in the area of hostility rather than in the areas of sex and/or dependency? 76 The writer is aware of no theoretical reason of why this should be. However, one might conjecture that it bears some relationship to the beginning therapist's conception of what he, as a therapist, should be like. Perhaps the hostile impulse is the most difficult to integrate into image of the therapist as a warm, accepting, giving individual. Before becoming overly Speculative it should be pointed out that the above observations are only trends, for which significance tests could not be done. Suggestions for Future Research The inconclusiveness of the findings regarding the hypothesized relationship between direction of therapists conflicts and defense preferences has been discussed. An adequate test of this hypothe- sis demands that assessment of therapists' defense proclilities be determined by instruments whose validity can be or has been deter- mined. Future studies which attempt to investigate this relation- ship should be preceded by, or should include, the development of objective and quantifiable measures of defense suitable for use with psychologically SOphisticated subjects. But, had the "sensit- izer” and “repressor" defense measures been independent, negative findings for hypotheses B-1 and B-2 would not have been unequivocal evidence against such a relationship. Previous studies with posi- tive findings (Altrocchi, p3,gl., 1960; Altrocchi, 1961) utilized as subjects students from the general college pOpulation. Our subjects, no doubt, represent a highly homogeneous group, and it could be that their defensive behavior is not characterized by a clearly defined preference for certain defense modes. If this were the case then their defense measures probably would be correl- ated regardless of the nature of the measuring instrument used. 77 Also, it may be that therapists do have defense preferences, but that they form configurations different from those which have been labeled "sensitizer“ and "repressor" modes. Future studies in this area utilizing therapists as subjects should explore the relation between other configurations of defense and direction of therapist conflict. The data suggests that therapists(as a group) with hostility conflicts are more defensive than therapists with dependency and sex conflicts. An implication seems to be that therapists may have more difficulty handling their hostile impulses, i.e., these impulses are recognized in themselves less readily than dependency and sexual impulses. The smallness of the sample size on which this finding is based makes them highly tentative, but the finding does suggest a lead for further exploration. 5 The differential findings with reference to the relationship between hostility conflict and approach hostility on the one hand, and dependency and sex conflict and approach dependency and sex expression on the other, raises the question of whether approach reactions in the latter two cases are partly a function of the ob- ject of the expressions. Bandura (1960) found that when the thera- pist is the object of hostility expressions there is less of a tendency to approach than when the object is other-than-therapist. Future studies should pick to determine whether this relationship holds for the areas of sex and dependency. The intern group's approach ratios for hostility and depend- ency were significantly larger than those of the practicum group. Also the approach-avoidance reactions of subjects in the practicum apparently were independent of their conflict scores. This con- tinuation of findings suggests that experience is a factor in 78 therapists' approach-avoidance responses. However, the small number of subjects in the practicum group renders tentative these differ- ences between experience levels. Future research with a larger sample of subjects at the lower experience level should offer clari- fying evidence. For the intern group the data indicates consistently that conflicted therapists tend to approach conflictual material less frequently as therapy progresses, whereas the reverse trend was noted for non-conflicted therapists. Does this differential trend extend beyond the therapy interval investigated? An answer to this question would necessitate samples of therapists' approach-avoid- ance reactions over a broad Spectrum of the total therapeutic in- teraction. Finally, there was an indication that ”plus” conflicts tended to be more strongly associated with avoidance reactions than "minus“ conflicts. However, this trend is a weak one. This considered ‘with the fact that the number of subjects in each of the categories of‘conflict was extremely small, requires that these findings be considered only as suggesting a question for further exploration. VI. SUMMARY.AND CONCLUSION The purpose of this study was to investigate the relation- ship between therapist conflict (with respect to hostility, depend- ency and sexuality) his primary modes of defense, and his approach- avoidance reactions to conflictual expressions initiated by clients. The study was based on theoretical assumptions derived from psycho- analytic theory, as expressed in writings on countertransference reactions, and on recent formulations concerning defense pre- ferences. Subjects were 26 individuals from two levels of experience: lB-psychotherapy interns, and 8 psychotherapy practicum students. The subjects were ranked on the following variables: hosti- lity, sex, and dependency expressions. Therapist conflict scores were derived from these measures. Defense measures were derived from rank scores provided by two judges on each of the following defenses: repression, regression, isolation, reaction formation, and.nndoing. Therapist approach-avoidance reactions were coded from 54 tape-recordings by means of a system (slightly modified) develOped by Winder and Bandura and their associates. Having operationally defined conflict as a discrepancy be. tween a subject's self-rating and the mean rating of a group of evaluators, it was hypothesized that 1) there is a negative re- lationship between therapist conflict conflict (for hostility, dependency and sex expressions) and the level of approach to conflictual (hostility, dependency, and sex) expressions, and that 2) non-conflicted therapists, as compared to conflicted therapists, respond to client conflictual expressions with relatively greater 79 80 (approach than avoidance reactions; it was also hypothesized that 3) therapists with "minus" conflicts receive higher ratings on ”repressor” defenses than therapists with “plus" conflicts, and that therapists with ”plus” conflicts receive higher ratings on ”sensitizer" defenses than therapists with "minus" conflicts; finally, that 4) non-conflicted therapists receive lower overall ratings than conflicted therapists on the use of defenses. The predictions regarding the relationship between therapist conflict and approach reactions were supported for the intern group. They were not supported for the combined intern-practicum group, nor for the practicum group alone. The intern groups ap- proach level for dependency and hostility was significantly higher than that of the practicum group. These differential findings for the two groups were in terms of the influence of therapy experience on therapist approach-avoidance reactions. These findings are considered highly tentative because of the small number of subjects in the practicum group. The hypothesis concerning hostility was not supported in a consistent manner. This lack of consistency was discussed in terms of the ”object" of the hostility expression. It may be the case that therapists tended to avoid client hostility ex- pressions when they (therapists) were the objects of such expressions. The predictions concerning the difference in approach reac- tions between conflicted and non-conflicted therapists were fully supported. It was also found that conflicted therapists tended to approach conflictual material less frequently as therapy progressed (in late as compared to early'interviews, with the exception of hostility), while the cpposite reaction was observed for 81 non-conflicted therapists. This was discussed in terms of the increasing intensity of the therapist-client relationship as therapy progresses. The postulates concerning the relationship between direction of conflict and defense modes were not supported. These findings were considered inconclusive because of the lack of independ- ence between the "repressor” and "sensitizer" defense measures. The postulate regarding the relationship between therapist conflict status and degree of overall defensiveness was supported for hostility, but not for sex and dependency. ‘A speculative explanation was offered for this lack of consistency. There was an indication that degree of defensiveness is related to the nature and/or magnitude of conflict rather than to its mere pre- sence or absence. The various trends noted in this investigation raises the following questions which point the way for further research in this area. 1. Are therapist approach-avoidance reactions to conflictual material partly a function of experience, i.e., does the be- ginning therapist learn what to focus on and how to reSpond, so as to approach the client production? 2. Does the differential trend in the approach-avoidance reac- tions of conflicted and non-conflicted therapists extend beyond the phase of therapy investigated? 3. Is direction of conflict a variable in therapist approach- avoidanoe reactions? #. Is “object" of clients' dependency and sex expressions a variable in therapist approach-avoidance reactions, as previous studies have shown, with reference to client hostility expressions? .BEFEBENCES Altrocchi; J., Parsons, o. A., and Dickoff, Hilda. Changes in self-ideal discrepancy-in repressors and sensitizers. Jo adnome BOOe P8 OhOle, 1960, .6_]:, 67-720 Altrocchi, J. Interpersonal perceptions of repressors and sensi- tizers and component analysis of assumed dissimilarity scores. J. adnorm. soc. Ps chol., 1961, fig, 528-53“. Alexander, F. Analysis of the therapeutic factors in pszcho- analytic treatment. Psychoanal. Quart., 1950, 12, 82-500. Alexander, F. Unexplored areas in psychoanalytic theory and; treatment. Behav. Sci., 1958, 3, 293-316. Balint, H. Changing therapeutic aims and techniques in psycho- analysis. Int. J. Psycho-énglg, 1950, 3;, 117-12#. Bandura, A. Psychotherapists anxiety level, self-insight, and psycho-therapeutic competence. J. adn rm. soc. Ps hol , 1956,.5g, 333-337- ~Bandura, A., Lipsher, D., Miller, P. 3., Psychotherapist approach- avoidance reactions to patient expressions of hostility. J. consult Ps chol., 1960, g5, 1-8. Berman, L. Countertransference and attitudes of the analyst in the therapeutic process. Ps chiat , 1949, 1g, 159-166. Betz, B. J., Hhitehorn, J. C. The relationship of the therapist to the outcome of therapy in schizonphrenia. Psychiatric res Re , 1956, No. 5. Bordin, E. 8. and Cutler, R. L. Theory research project: pro- Ject title: sis 0 th ra eutic teractions: Sept. 1, 1952-Oct. 15, 19%g. fiate of report: got. 13, 1958, Univer- sity of Michigan. _ Carlson, V. 3. Individual differences in recall of word asso- ciation test word. J. Pers,, 195“, Q}, 77-87. Carpenter, C. w., Ueiner, H., and Carpenter, Janeth. Predictae bility of perceptual defense behavior. g&_3bgg§n&_ggg& P8!°h0;., 1956, 2, 380-3830 Cohen, H. B. Countertransference and anxiety. In C. Thompson, M. Hazer, and E. Uitenberg (es) An ggtling of Psychoanalzgis. New York: Random.House, 1955, pp. 9- . vCutler, R. L. Relationship between therapists' personality and certain aspects of psychotherapy. Unpublished Ph D dissertation, University of Hichigan, 1953. Dollard, J. and Miller, N. Pirsggalitz and Pszghotherapz. New York: HoGrawbHill, . 82 83 Eriksen, C. N. Perceptual defense as a function of unacceptable needs. J. adnorm. soc. Ps chol., 1951, 46, 557-564. Eriksen, C. N. Defense against ego threat in memory and per- ception. J abnorm. soc Ps chol., 1952a, 42, 230-235. Eriksen, C. w. Individual differences in defensive forgetting. J. e . Ps chol., 1952b, 44, 442-446. Eriksen, C. w. Psychological defenses and ego strength in the recall of com leted incompleted tasks. J. abnorm. soc. P8 OhOIe, 19 ’ £2, 45-500 Fenichel, 0. The Psychoanalytic Theogy of Neurosis. New York: Norton, 19 . Ferenczi, 8. Further Contributions to the Theo and Techni u of Psycho-Analysis. London: Hogarth Press, £555. Fiedler, R. E. The concept of an ideal therapeutic relation- ship. J xconsult Ps chol., 1950a, 14, 239-245. Fiedler, P. E. A comparison of therapeutic relationships in psychoanalytic, non directive, and Adlerian therapy. ,1; ggnsglt, ngghol., 1950b, 14, 436-445. Fiedler, F. E. and Senior, K. An exploratory study of unconscious feeling reactions in 15 patient-therapist pairs. J2 abnorm, sgc. Psychol., 1952, 4 , 446-453. Freud, Anna. The E o and the M chanisms f Def nse. New York: Internat. Universities Press, 1946. Freud, 8. The dynamics of transference. (1912) In.Collectgd Papers, vol. II, New Ibrk:, Basic Books, 1959, pp. 31 -322. Fromm-*bichmann, F. Principles of Intensive Ps ohothera . Chicago University Press, I950. Goldstein, H. J. The relationship between coping and avoiding behavior and response to fear-arousing propaganda. g,, abnorn..soo Ps chol., 1959, 58, 247-252. Gordon, J. B. Interpersonal predictors of repressors and sensi- tizers. J, Pers., 1957, 25, 686-698. Gordon, J. E. The stability of the assumed similarity response set in repressors and sensitizers. J. Pers., 1959, 21, 362-373. Guilfggd J. P. Psychometric Methods. New York: McCraw-Hill, Heiman, P. On countertransference. Int. J, Psycho-An§;., 1950, 1, 81-4. 84 Holt, 3. R. and Luborsky, L. Personalit Patterns of P3 chia- trists, New York: Basic Books, I958, pp. 272-293. Horst, P. A generalized expression for the reliability of measures. Psychometrika, 1949, 14, 21-31. Janis, I. L. and Feshbach, S. Effects of fear arousing communi- cation. J. abnorm. soc. Ps chol., 1955, 51, 663- 68. Little, Margaret. Countertransferenoe and the patient's response to it. Int. J. Psycho. Anal., 1951, fig, 32-40. Nunson, Joan E. Patterns of Client resistiveness and counselor response. Unpublished Ph D. disseration, University of Michigan, 1960. Murray, E. J. A content analysis method for studying Rsycho- therapy. Ps chol. Mono ., 1956, 29, (whole No. 20) 1-32. Parloff, M. B. Some factors affecting the quality of thera- peutic relationships. J. abnorm. soc. Ps chol., 1956, 5;, 5-10. Reich, Annie. 0n countertransference. Int. J. Ps choana1., 1951, 3;. 25-31. ~Big1er, D. Some determinants of therapist behavior. Unpublished Ph D. dissertation, University of Michigan, 1957. Rogers, C. R. The necessary and sufficient conditions of thera- peutic personality change. J. consult. Ps chol., 1957, g; 95-103. Siege1,52. Nonpgrametric Statistigs. New Yerk: McGraw-Hill, 19 . Strupp, H. H. The psychotherapist's contribution to the treat- ment process. Behav. Sci., 1958a, 3, 34-67. Strupp, H. H. The performance of psychoanalytic and client cen- tered therapists in an initial interview. J. consult, Psygoln 1958b, gg, 265-274. Strupp, H. H. The performance of psychiatrists and psycholo- gists in a therapeutic interview. J. lin. Ps hol., 19586, E, 219‘22 6 O Strupp, H. H. Patient-doctor relationships: psychotherapist in the therapeutic process. In A. J. Bachrach (ed) e 1- mental Foundations of Clinical Ps cholo . New Yor : sic Books, I962, pp. 576:615. Strupp, H. B. The nature of the psychotherapist's contr bution :to the treatment process. Arch. Gen. Psychiat., 19 0, 3, 219-231. Sullivan, H. S. The theory of anxiety and the nature of psy- chotherapy. Ps chiatr , 1949, 12, 3-12. 85 Szasz, T. S., Knoff, W. F., and Hollender, M. H. The doctor- patient relationship and its historical context. Amer. J. P8 chiat., 1958, 115, 522-528. Winder, C. L., Ahmad, Farrukh Z., Bandura, A., and Rau, Lucy, Dependency of patients, psychotherapists's responses, and aspects of psychotherapy. J. consult. Ps chol., 1962, 26, 129-134. Winnicott, D. W. Hate in the countertransference. Int. J. Psycho- Ann... 1949. 30. 69-74. APPENDIX,A LIST OF PERSONAL QUESTIONS Name of Respondent: Date: 1. Nature of training: Counseling Clinical Vocational Counseling 2. Theoretical Orientation: Rogerian Eclectic Analytic Vocational Counseling Other 3. Personal Psychotherapy: Yes No (If Yes) Orien- tation of your psychotherapist 4. Number of hours of personal psychotherapy 5. Number of clients seen in which you were supervised 6. Total number of clients seen in psychotherapy 86 APPENDIX B INSTRUCTIONS FOR RANKING You have been provided a list of individuals with whom you have had some interaction or have observed interacting with others in varying situations. You have also been provided a sheet con- taining a term referring to a personality disposition. This dis- position has been defined in terms of objective behavior, however, inferences from behavior are also included. Ybu are being asked to rank these individuals (including yourself) with respect to the degree of expression of certain kinds of behavior. The group itself is to serve as the frame of reference for ranking the individual member. If limited interaction or acquaintanceship with a particular individual or individuals make it impossible to rank him (them) with respect to the variable in question, he (they) should be omitted, and ranking confined to those individuals with whom you are adequately acquainted. In considering the individual for ranking, attempt to recall situations in which you have observed him interacting with others e.g., the psychotherapy practicum course, personal adjustment con- ference, informal social groupings, etc. NOTE: Ybur rankings will be treated confidentially. 87. 88 RANKING SHEET Name of Ranker Date Hostility Expression This variable is defined in terms of the diSposition to ex- press certain behaviors. It refers to the tendency or readiness of the individual to attack or belittle contrary points of view, to become angry when Opposed, to be critical of others, to ask questions or make statements that puts the other person "on the spot”, to direct sarcastic remarks to others; may seem angry or irritated frequently. In general, tendency to be oppositional, antagonistic, argumentative, derogatory, critical, complaining or resentful. These descriptive behaviors are merely to serve as a frame of reference for ranking the group on hostility, and are not meant to be exhaustive. You may be able to think of other hostility ex- pressions which can be utilized in making your judgments. PROCEDURE: Ybu have been provided with a list of individuals (including yourself) to be ranked on this variable. A rank of 1 should be assigned to the individual, in the group you judge to express the greatest amount of hostility, and the bottom rank should be assigned to the individual you judge to express the least amount of hostility. Indicate your ranking in the space below: Name Name lllllllllE lllllllllE 89 BANKING SHEET Name of Ranker Date Dependency Expression This variable is defined in terms of the disposition to ex- press certain kinds of behavior. It refers to the tendency fre- quently to ask for and to accept help, advice, suggestion and di- rection from others; eager for the approval of others; trusting and eager to please; concern about disapproval; concern to find out what others think; readiness to follow instructions and to do what is expected; lets others make decisions; strong tendency to accept leadership of others; ready agreement with authority figures; tendency to form strong attachments, and to make as many friends as possible. These descriptive behaviors are merely to serve as a frame of reference for ranking the group on dependency and are not meant to be exhaustive. You may be able to think of other dependency expressions which can be utilized in making your judgments. PROCEDURE: You have been provided with a list of individuals (including yourself) to be ranked on this variable. A rank of 1 should be assigned to the individual, in the group, you judge to express the greatest amount of dependency, and the bottom rank should be assigned to the individual you judge to express the least amount of dependency. Indicate your ranking in the Space below: Name Rank Name lllllllllE 9O RANKING SHEET Name of Ranker Date Sexuality Expression This variable is defined in terms of the disposition to give expression to certain kinds of behavior. The definition, in terms of descriptive behavior, refers participation or readiness to parti- cipate in discussions about sex; pleasure in telling and/or listen- ing to jokes involving sex; talks about one's own sexual interests and attitudes given the slightest Opportunity; tendency to be se- ductive and flirtatious either openly or in subtle ways. These descriptive behaviors are merely to serve as a frame of reference for ranking the group on sexual expression and are not meant to be exhaustive. You may be able to think of other sexual expressions which can be utilized in making your judgments. PROCEDURE: You have been provided with a list of individuals (including yourself) to be ranked on this variable. A rank of 1 should be assigned to the individual, in the group, you judge to express the greatest amount of sexuality, and the bottom rank should be assigned to the individual you judge to express the least amount of sexuality. Indicate your ranking in the space below: Name Name IIIIIIIIIE lllllllllE APPENDIX C Sample of tapes used in investigation DEPENDENCY HOSTILITY SEXUALITY The - _ _ .— pist & Inter- x of 2 Coded by X of 2 Coded by X of 2 Coded by Tap; No, view Coders 1 Coder Coders 1 Coder Coders l Coder A}. AVe Age Av. 1A_B° v. 20 AV, 20 lo Be AVo A-68 B-18 41 15 0-97 180 3222 93% E388 389 FTSS 221 PC G32 2.2 320 3232 261 1163 T-37 68 JT36 102 94 Kg75 9 45 L372 379 318 M277-1 317 277-2 N-93 285 “as P322 373 66 1 UHMOHJ NMNNU NU‘UWU‘HHU 0%»an NU‘UtU WU‘F‘NQ-F‘O‘HHU'INWNl-‘m .87 .13 e765e235 .84 .16 .685.315 .70 .30 .82 .18 .41 .59 .44 .56 ~505-395 .775.225 .715.285 .555.445 .705.295 .875.125 .84 . .73 .27 .63 .37 .87 .13 .715.285 .70 .30 :23 :22 .595.405 .46 .54 .88 .12 .875-125 .975.025 .77 .23 .875.125 .765.235 .69 .31 .62 .38 .77 .23 .86 .45 .63 .81 .55 .74 .67 ~53 .90 .14 .55 ~37 .19 .45 .26 .33 .47 .10 .75 .25 .50 .50 .63 .37 .81 .19 .70 .875.125 .665.335 .695.305 .755.245 .685.315 .905.095 .395.605 .5? .43 .675.325 .55 .35 «725-275 .75 .25 .88 .12 .785-215 .875.125 .64 .36 .695.305 .545.455 .79 .21 .825.175 .89 .11 .57 .43 .50 .50 .78 .75 .87 .89 .56 .85 .64 .57 .83 .22 .25 .13 .11 .44 .15 .36 .43 .17 :48 .52 .71 .29 .41 .59 .665 ~775-225 .705.295 ~47 .53 .845.155 .775.225 0 1.00 .865.135 .55 .33 .925.o75 .365.635 .75 «25 .765.235 .535.465 0 1.00 ~735-265 .515.155 0 1.00 .35 :-1 055 025 037 063 .89 .84 .56 .85 .37 .79 .33 .11 .16 .uu .15 .63 .21 .67 O 1.00 .74 .26 92 Thera- pist & Inter- Xof 2 Coded by X of 2 Coded by X of 2 Coded by Tap; No. view Coders l Coder Coders 1 Coder Coders l Coder 3.35 2 .81 .19 .75 .25 .37 .63 $263 5 .50 .50 .77 .23 ..... HPl 3 .63 .37 .Z6 .24 -- -- T-Sl47 3 .76 .24 . 9 .51 .43 .57 $282 5 .70 .30 .47 .53 1.00 .00 U-JSZ 5 .70 .30 .64 .36 .69 .31 T4? 3 .75..25 .62 ~38 .69.31 V-Lh 3 .68 .32 .70 .30 .75 .25 613a 6 .335.665 .74 .26 .765.235 w-250/1 4 .77 .23 .51 .49 .46 .54 250/2 5 .78 .22 .54 .46 .67 .33 x.3352 6 .76 .24 .70 .30 .85 .15 $175 3 .585.415 .73 .27 .85 .15 2.227 5 .21 .79 .51 .49 .50 .50 306 3 057 043 .62 038 039 e61 APPENDIX D SCORING MANUEL FOR CONTENT ANALYSIS 1. Scoring Unit. A. B. C. Definition. The scoring unit is composed of three parts: tKe client's statement, the therapist's total subsequent re- sponse, and the client's immediately following verbaliza- tions. (This last element not only completes the first in- teraction unit, but also serves as the first element of the next interaction unit.) 1. The term "client statement" means any communication which can be noted from a tape recording, such as a statement, a question. 2. The term "therapist response” means any communication which can be noted from a tape recording, such as a statement, a question, or a silence or seconds or more. 3. The term ”client response” means any communication which can be noted from a tape recording, such as a statement, a question, or a silence of 9 seconds or more. (Client silences are scorable only if it occurs immediately following a therapist response.) gzagplgg: (C= client; T= therapist) . elt humiliated. T. The way you feel now when an authority figure questions youe C. I don't know why I should still feel that way. C. Y0u haven't told me what to do about Paul. T. (Silence of 7 seconds) g. I guess I should tell you about my mother-in-law coming 0 see us 0 Interruptions. Statements of one person, either client or therapist, interrupted by the other will be one response if the continuity of the verbalization is not altered by the interruption. Wrasse C. I asked him to help me and... T. Why did you ask for help? C. He refused to do anything about it. If the interruption destroys the continuity of the verbali- zation, then the interrupted verbalization and the interrupt- ing verbalization become elements of a unit as defined above. Sep%rate client rgsponse . don t understand why he does not help me and... T. Did you ever ask him to help you? C. Yes I did, but it did.not work on him. In this instance, since the therapist's interruption broke the continuity of the ”client's statement", the interruption becomes the "therapist's response”, and the second of the ”client's verbalizations" becomes the "client response". This is a unit by the definition adopted above. 93 94 II. .Client response categories. A. Hostility. Hostility statements are those that include escription of or expression of unfavorable, critical, sarcastic, depreciatory remarks, anger, opposition, anta- gonism, argument, expression of dislike, irritation, re- sistance, annoyance, resentment, any expression of aggres- sion and punitive behavior. S boats rise 1. 2. 3. 4. 5. 6. Ange . (The client indicates that he feels angry) Co I am JuSt plain made C. I couldn't think...I was so mad. Dislike. (The client expresses dislike or describes action which would usually indicate dislike) C. I just don't get interested in them and I would rather be somewhere else. C. I've never felt I liked them and I don't expect I ever will. Rgsentment. (The client expresses or describes a per- sistent negative attitude which does or might change into anger on specific occasions) C. They are so smug; I go cold whenever I think about having to listen to their “our dog” and "our son”, boy! C. They don't do a thing for me so why should I ever ask them over. Antagonism. (The client expresses or describes anti- pathy or enmity) , C. It's nothing very definite, but we always seem at odds somehow. ‘ C. There's always this feeling of being enemies. Qpp%sition. (The client expresses or describes op- pos iona feelings or behavior) C. If he wants to do one thing I want to do another... and it's lousy. C. My roommate is always against things. She is even against things she wants. Critig%% attitudes. (The client expresses negative eva ua on, ascribes action which usually imply neg- ative evaluation, or expresses negative feelings in a veiled manner) B. C. 95 C. If I don't think the actors are doing well, I just get up and walk out. C. He is tied to his mother's apron strings. C. The lecturer just stands there and talks a mile a minute and nobody can take down what he is saying. C. Yen are a psychologist, you should know that. 7. Aggpessive action. (The client acts so as to hurt anot er person either physically or psychologically) C. He deserves to suffer and I'm making it that way every way I can. C. I hit him because he is always bragging about how tough he is. Hostility anxiety. Any expression which includes expres- sion of fear, anxiety or guilt about hostility, or state- ment reflecting difficulty in expressing hostility. C. I just felt so sad about our argument. C. I was afraid to hit her. C. After I hit her I felt lousy. C. My father was a punitve man; both my brother and I were afraid of him. Dppendgpcy. Dependency statements are those that include description or expression of help-seeking behavior or statements seeking approval or demanding initiation, de- manding or seeking company, information seeking, descrip- tion of support, and concern about disapproval as well as the agreement with the therapist's reflection or in- terpretation. Subcategpries 1. H l s ki . (The client asks for help, reports asEing for fielp, or describes help seeking be- havior) C. I asked him to help me out in this situation. C. I talked to her about it because she always has good ideas. 2. Approval spekipg. (The client requests approval or acceptance, asks if something has the approval of the therapist, reports having done so with others, or reports approval seeking behavior) 3. 4. 5. 6. 96 C. I hape you will tell me if that is what you want. C. I asked her if she thought I was doing it right. Company seekipg. (The client expresses a wish to be with others, describes making arrangements to be with others, describes efforts to be with others) C. I get so lonesome here; if I were at home I would have my friends. 0. I go to Kewpee's in the evening...I know other kids will be there. Information seekin . (These are statements in Wfiicfi Ehe cIient asks for some kind of information from others, or arranges to be the recipient of information primarily because of the relationship involved) . C. I asked her (roommate) to tell me why she never called up later. C. I came here to find out what my I.Q. is. Seeking for Initiation. (The client asks of the therapist that e start the discussion, select the topic, or take responsibility upon himself) C. What should I talk about today? C. I don't know how to start; could you give me a hint? C. I want you to tell me what to do in these situations. Concepp about disapproval. (The client expresses ear 0 disapprova , expresses unusual sensitivity about disapproval, or describes unusual distress about instances of disapproval) C. She (roommate) doesn't say anything I can put my finger on, but her tone of voice and the way she looks at me makes me wonder what I'm doing mug e A C. I get the feeling they don't understand me. C. I showed her one of my poems and when she read it and didn't say anything I couldn't write any- more 0 D. E. F. 97 7. Agpeement with others. (These are expressions of agreement or readiness to agree with or to accept a statement by the therapist or another because of the relationship between the client and that other person; includes ready agreement with tenta- tive interpretation by therapist) C. That is absolutely correct! I never thought of it that way before. (In reSponse to a tenta- tive interpretation by therapist) C. She (roommate) always hits the nail on the head and I have to agree with her. Dependency Anxiety. (A statement of dependency anxiety includes expressions of fear, anxiety, or guilt about de- _ pendency; or statements reflecting difficulty in ex- pressing dependency) C. I just felt so bad about asking for help. C. I feel terrible about not being able to take care of myself. C. I just can't ask you to do this...don't even want to bring it up. Sexuality. Sex statements are statements that include descriptions of or expressions of a sexual nature, se- ductiveness and flirtation, include descriptions of sexual attraction and arousal, sexual activity not mixed with fear or guilt; planning for sexual satisfaction, courtship and dating among unmarried pe0ple where the erotic element is present but institutionalized, descrip- tion of homosexual feelings and other perversions, des- criptions of masturbation, and discussion of normal sex education. C. I talked over my affiar with my father. C. He wanted to have sexual relations with me. C. I don't see any harm if a girl kisses a boy. C. I don't know what a woman looks like..her sexual organs 0 C. My mother would never talk to me about sex. C. We would play around, masturbating in condoms. Se; Angiety. These are statements that include expres- sions of fear, anxiety, and guilt about sex; denial of sex drive, negative attitude toward sex, feelings of sexual inadequacy or impotence, and statements reflect- ing difficulty in expressing sexual interests or in ex- pressingysex. G. 98 C. I just felt guilty after going out with her. C. I was nervous and disjusted when my brother would wrestle around with me. C. When I am with my wife I always imagine a strange girl who is behaving aggressively toward me...a1ways teases but never gives in. C. I was afraid someone would discover us. C. I got into bed with him then froze up when he wanted to go further. C. I want to strive for pure intellect and forget about the animal side. Other Client ReSponses. 1. 2. 3. Silence. Client fails to reSpond to therapist response before 9 or more seconds. Acknowledgement. Client response indicating accept- ance of therapist request of a generalized nature; includes reSponse to therapist statement which ends with a questioning tone of voice, also when client just says "yes" in reSponse to direct question from therapist. T. Could you tell me more about that. C. Yes, of course I will. an-acceptance. These are statements which that the client is either disagreeing with the therapist or is refusing to accept his reflection or interpretation. T. You seem to be thinking in the same way as your mother did when you were a child. C. No, not like her at all. Miscellaneous. These are statements which do not be- ong c any of the above mentioned categories. C. My brother is 17. C. I never knew my father very well. H. Object Categories. Every client expression falling into one o t e major categories (Hostility, Dependency or Sex) is scored for object, i.e., the person toward whom the client is oriented. 1. Therapist. (as object)- client has to specify therapist as object. 99 C. I don't know how to start; could you give me a hint? C. I want you to tell me what to do in these situa- tions. 2. Other (Objects). Objects in this category refer to all other persons, including the client himself, e.g., friends, teachers, siblings, parents, etc. III. Therapist Reaponse Categories. A. Therapist Approach ReSponses. The approach responses are those verbalizations by the therapist which are designed to elicit from the client further expressions and elabora- tion of feelings, attitudes, and behavior with reSpect to the t0pic under discussion, when that topic falls into one or more of the following categories: dependency, hostility, or sex. The approach responses are defined by the following subcategories: approval, exploration, reflection, labeling, interpretation, and instigation of discussion in response to clientes protracted silence. 1. Approval. (The therapist expresses explicit approval of or agreement with the client's attitudes, feelings, or behavior) C. Can we juasbe quiet for a moment? T. Certainly. T. C. I like to be with people. T. I would not blame you for that. C. I just blew up and hit her. T. I don't see how you could have done differently. 2. Egploration (probing). (Therapist remarks and ques- tions that encourage the client to describe or ex- press his feelings, attitudes, or behavior further; asks for further clarification, elaboration, descrip- tive information; calls for details or examples) C. How do I feel? I feel (pause) idiotic. T. What do you mean...you feel idiotic? C. I cannot understand my behavior mostly. T. What is it about your behavior that you cannot understand? 3. 5. 7. 100 Reflection. (The therapist repeats or restates the client's verbally expressed feelings, attitudes, or actions) C. I don't like that at all. I don't. I don't. T. YOur reaction is definite dislike. C. I wanted her help but could not bring myself to ask. T. It was impossible to ask her at the time. Labeling. (Therapist labels attitudes, feelings, and behavior) C. When she said that, I had a cold feeling go all through me. T. Y0u had a strong reaction to that...maybe fear, or anger, or some other feeling. C. When I had intercourse with her I nearly threw up. T. I guess you must have felt pretty disgusted. Suppopp. (The therapist expresses sympathy or under- standing of the client's feelings) C. I hate to ask favors from people. T. I can understand how difficult it would be for you. C. I find this very difficult to talk about. T. I know it is not easy for you to talk about it. Generalization. (Therapist points out to the client that his feelings, attitudes and behavior are common and natural) C. I don't know why I have to react that way. T. Under the circumstances, most peOple would act the way you did. C. When I first came here I was very lonely and a little afraid. T. Most students feel this way until they have made new friends. Instiggtion. (Therapist suggests discussion of de- pendency, hostility, or sex, or he shifts the B. 8. 9. 101 discussion to one of these areas, which may or may not have been expressed by the client earlier in the interview or in some previous interview; the therapist may sug est such topic after a protracted silence by client? C. I was pretty fatigued when I got home. I went to bed right off the bat but I just tossed and turned. When I woke up in the morning I had this pain in the leg. I went to the doctor and he gave me a shot. T. Yes, let's get back to that evening, that irrita- ted feeling you had. Intgppretations. (Therapist points out patterns or re tionships in the client's feelings, attitudes, and behavior; explains why the client does something; suggests casual and antecedent relationships; points out similarity in client's dependency, sexual, or hostile feelings, attitudes, or behavior in different situations or experiences or suggests relationships between present feelings or behavior and past exper- iences C. I am always asking her for things. T. You depend on her a good deal; you usually have a great deal of concern about not making her angry, and don't want to take the chance of doing anything until you know it meets with her approval, but this is the thing that makes her irritated at you. Seems that you are bringingabout the very thing you fear. Is that correct? Factual Information. (Therapist responds to client's epricit or implicit request for information) C. Who can I see about it? T. You might start at the housing office. Therapist Avoidance Rgspopses. These are expressions by he t erapist ic are designed to inhibit, discourage, or divert the client's dependency, hostility, or sexual expressions. The avoidance responses are defined by the following subcategories: disapproval, ignoring, topical transition, silence, and mislabeling. Spppgtpgories. 1. Disa roval. (These responses include unfavorable, crIticaI, sarcastic, antagonistic reactions toward the client for having expressed feelings or behaved in dependent, hostile, or sexual manner in the past or present; disapproval of the client's decisions to express dependent, hostile, or sexual feelings or behavior in the future; rejection of the client) 2. 3. 4. 102 C. 80 I blew my top and.hit her. T. Just for that you hit her. C. Why don't you make statements? Make a statement. Don't ask questions. T. I am the therapist here; you are coming to me for help. Igporipg. (An ignoring reSponse is one in which the therapist reSponds to the content of the client's response but ignores the hostile, dependent, or sex- ual affect) C. I lose my temper over his tardiness. T. What are the results of his being tardy? C. I am wasting your time but I am interested in what is important to you. .Aren't you proud? T. It pleases you. .To ical Transition. (The therapist changes the dis- cussion from a topic of hostility, dependency, or sexuality to one which is non-hostile, non-dependent, or non-sexual) C. My mother always annoys me. T. How old is your mother? C. I was nervous and disgusted when my brother would wrestle around with me. T. Your brother was older or younger? Silppcp. (Therapist makes no response for 6 seconds or more after the client's description or expression of dependency, hostility or sexuality) C. I just dislike it at home so much at times. r. (Silence) C. So I just don't know what to do. C. She is a big tease. T. (Silence) C. Builds up a big reputation by leading guys on but never comes across. C. 103 5. Muslabe;ipg. (The therapist mislables as non- osti e, non-dependent, non-sexual, feelings that are clearly hostile, dependent, or sexual) C. When are you going to give me the results of those tests? T. You seem to be almost afraid to find out. C. I want to strive for pure intellect and forget about the animal side. T. Intellectual achievement is very important for you. You fear not being able to reach your intell- ectual goal. Other Therapist Responses. 1. Retraction. (When the therapist takes back his in- terpretation after the client's negative or positive response to it) T. Maybe I was wrong in putting it that way. There can be various other explanations. ‘ 2. Miscellaneous. (These are the therapist's inter- ventions which do not fall into any of the above categories and also the ones that cannot be de- ciphered from the tape) IV. Additional Rules for Coding. 1. 2. 3. a. Therapist approach responses which are general, are scorable as approach to all major categories scored for the preceding client statement. For example; (T. Tell me about your difficulties.) would be an approach response to dependency, hostility, and sexuality if these cate- gories were scored in the client's preceding verbalization. If the therapist's response focuses on a specific aspect in the client's preceding expression, the response is scored.approach for that category but avoidance for any other scored category which is a major client category. For example, if the preceding client expression has been scored for dependency, hostility, and sex and the thera- pist's response focuses on dependency, the response is scored approach for dependency, but as avoidance for hostility and sex. If T. begins his response with either approach or avoid- ance but modifies his reaction before concluding the response, the scoring is to be on the basis of the modi- fication which concludes his response. Request by T. for repetition of C. verbalizations is scorable as approach, in the event C's original verbal- ization falls into any major category (sex, hostility, or dependency). 5. 6. 7. 8. 9. 10. 104 The object of client's expression does not in itself determine whether T's response to that expression is approach or avoidance. The determining factor is whether T's reSponse falls into one of the approach or one of the avoidance subcategories. Hostility, with T. as object, will be scored when voice quality of C. is indicative or irritation or anger, and T. is the object of the content of the expression. Such a C. statement may be scored for more than one category, e.g., for Dependency-Therapist and Hostility-Therapist. Client expressions having an undertone of dependency (”I don't know what to do”) are not scorable as depen- dency unless the dependency is made more specific and the object of the dependency can be determined as Thera- pist or other. Sometimes the client will finish a sentence or statement for the Therapist. In this instance the whole expression is scored as a client statement. T. You feel...... T. Understanding it now.... C. Angry. C. Will make a difference. Therapist verbalizations of the nature ”mmm-hmm', O.K., ”I see", etc., which are immediately followed by client change of t0pic are not to be considered as interruptions. If such a response is given in the instance where it is clear that the particular client expression is concluded it is to be scored. “Mmm-hmm”, ”hu-uh” of either client or therapist are not scorable as anything in themselves. For example, if a T. reSponded with "mmm-hmm" and did not elaborate, this verbalization would be scored as silence if T. did not respond further before 6 seconds elapse, or if the client did not continue before 6 seconds had elapsed. Such verbalizations may be scored when their meaning can be determined. as when in response to a question they take on the meaning of ”yes” or ”not, APPENDIX E List of Defenses and Definitions 1. Repression: The exclusion from awareness of ideas, thoughts, and feelings that arouse anxiety and threaten damage to the self image. The user of this defense seems to lack provision for af- fect discharge, in that certain ideas, feelings and attitudes that would be expected as adequate reactions to reality are missing. In general he broadcasts a persuasive message of ”emotional frigidity'. 2. Avoidan e Withdrawal : Refers to the tendency to avoid or maintain distance between oneself and any aspect of external or in- ternal reality; also implies any escape reactions which will even- tuate in the individual's failure to perceive aSpects of the exter- nal or internal environment which might threaten damage to the self image. An example would be, a subject who consistently attempts to avoid or withdraw from situations in which hostility is likely to arise. 3. Denial: The tendency to deny unpleasant parts of reality and/or painful sensations. The subject who uses this defense might seem to indicate that a particular image does not exist in reality, or if its existence is perceived, the subject reacts as if he does not apprehend its meaning. Will on certain occasions and to some extent seem to attach the focus of a certain personal observation away from the area for which it was intended, to another area. 4. Regyession: Refers to the adoption of modes of behavior that seem inappropriate agewise, as if it belongs more properly to a much younger age. Yet this behavior seems peculiar satisfying to the subject and brings him a feeling of security in its performance. 105 106 Especially during times of stress the subject seems "childlike" and helpless, and adOpts behavior patterns less mature than the be- havior patterns they replace. 5. Reactign-formatipp: A deve10pment in which conscious, social- ized attitudes serve as their primary function the hindering of ex- pression of unconscious, contrary, and opposite attitudes. Expres- sion of the unconscious attitudes would threaten the integrity of the self-image. An example would be an overt attitude of extreme docility which the judge feels masks unconscious attitudes of host- ility. The conscious attitude might give the impression of having a forced or compelling quality; in that the attitude is held in face of circumstances in which one would expect expression of contrary attitudes, e.g., docility rather than hostility in anger provoking situations. 6. undoing: Refers to a maneuver wherein activities are performed which are cpposite to activities performed at some earlier time. Undoing is a kind of negative magic which by means of a motor symbol- ism would ”blow away", as it were, not the consequences of an event (an impression, an experience) but the event itself. Examples: When the subject, after having been overtly aggressive toward an- other, behaves on the next occasion with excessive courtesy and kind- ness, without bothering to apologize for his previous behavior; or the subject who has a high need to succeed or perform well, but who consistently seems to Spoil his performance before a final stage. 7. Isolation: Involves a separation of the emotional from the cognitive aSpects of impulses. It is the emotional significance of the ideational content that is stripped away and stands apart. The user of this defense may remember that he was angry and aggres- sive in his behavior toward another without perceiving the quality 107 . of the anger itself. Such an individual is able to produce verbal expressions with reference to impulses, but such expressions are not accompanied by the expected feeling tones; he communicates a quality during such verbalizations which might be termed an "empti- ness of affect”. 108 INSTRUCTIONS FOR RANKING You have been provided with a list of individuals to be ranked on the use of several defense mechanisms. To facilitate the rank- ing task, a set of 3 by 5 cards (each containing the name of one of the persons to be ranked) as well as a set of ranking sheets are pro- vided. Each sheet contains the name and definition of a defense on which the group is to be ranked. If limited interaction or acquaintanceship with a particular person or persons makes it impossible to rank him (them) with reSpect to a particular defense, he (they) should be omitted and ranking cone fined to the remainder of the group. PROCEDURE: A rank of 1 should be assigned to the individual you judge to use a particular defense (on which you are ranking) the most in his group, and the bottom rank should be assigned to the individual you judge to use that particular defense the least. This procedure should be followed for each of the defenses.