AGGRESSIVE AND SEXUAL FANTASIES IN VIOLENT AND NON-VIOLENT PRISON INIIIATES Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY BENJAMIN BEIT-‘HALLAHMI 1970 AGGRESSIVE AND SEXUAL FANTASIES IN VIOLENT AND NONvVIOLENT PRISON INMATES Thesis for the Degree of Ph, D. MICHIGAN STATE UNIVERSITY BENJAMIN BEIT-HALLAHMI 1970 ''''''''''''''''' new IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII mum .NdhflhfiyanIStauc University This is to certify that the thesis entitled Aggressive and Sexual Fantasies 1n Violent and Non-violent Prison Inmates presented by Binymnin Bait-Hal lahtni has been accepted towards fulfillment of the requirements for Eh . D . degree in mm yXML Major professor Date-W74 0-169 ~. IIIJAB & sIIIs I ennxemum ‘mc. ILI ARV BINDERS R an '1 B APIiuelIseg Vie; I: .‘i" EM; ABSTRACT AGGRESSIVE AND SEXUAL FANTASIES IN VIOLENT AND NON-VIOLENT PRISON INMATES BY Benjamin Beit-Hallahmi This study investigated the aggressive and sexual fantasies of prison inmates in relationship to their actual behavior. Sixty-five inmates of the State Prison of Southern Michigan were divided into three groups: 1. Inmates with a history of violent crimes and a record of disciplinary problems inside the prison (Group 1). 2. Inmates with a history of violent crimes but no record of disciplinary problems inside the prison (Group 2). 3. Inmates with a history of non-violent crimes and no record of disciplinary problems inside the prison (Group 3). Two instruments were administered to the Ss in groups: 1. A Fantasy Questionnaire (FQ) containing 82 items, that was constructed by the investigator. Benjamin Beit-Hallahmi 2. An experimental TAT booklet (ETAT), containing five pictures, two of which were rated "sexual" and two of which were rated "aggressive." The fifth picture was considered neutral. There were four hypotheses regarding predicted dif- ferences between the three groups, as follows: 1. Ss in group 2 will produce more aggressive fan- tasies on both the F0 and the ETAT, compared to Ss in group 1. 2. Group 3 will be significantly lower, on both Sex and Aggression measures, compared to the other two groups combined. 3. Frequency of aggressive fantasies will be posi- tively correlated with frequency of sexual fan- tasies, on both the FQ and the ETAT. 4. Sex and Aggression measures on the ETAT will be positively correlated with the Sex and Aggression scales on the F0. Since it was shown that ETAT scores were correlated with verbal fluency, and the latter correlated with 1.0. and education, it was decided to correct the raw ETAT scores for verbal fluency by dividing every score for every story by the number of words in the same story. Out of four hypotheses, only hypothesis 3 was un- equivocally supported. Benjamin Beit—Hallahmi 2. An experimental TAT booklet (ETAT), containing five pictures, two of which were rated "sexual" and two of which were rated "aggressive." The fifth picture was considered neutral. There were four hypotheses regarding predicted dif- ferences between the three groups, as follows: 1. Ss in group 2 will produce more aggressive fan- tasies on both the F0 and the ETAT, compared to Ss in group 1. 2. Group 3 will be significantly lower, on both Sex and Aggression measures, compared to the other two groups combined. 3. Frequency of aggressive fantasies will be posi- tively correlated with frequency of sexual fan- tasies, on both the F0 and the ETAT. 4. Sex and Aggression measures on the ETAT will be positively correlated with the Sex and Aggression scales on the FQ. Since it was shown that ETAT scores were correlated with verbal fluency, and the latter correlated with 1.0. and education, it was decided to correct the raw ETAT scores for verbal fluency by dividing every score for every story by the number of words in the same story. Out of four hypotheses, only hypothesis 3 was un- equivocally supported. Benjamin Beit-Hallahmi The differences among the groups on the Aggression measures were in the direction predicted by Hypothesis 1, but they were not statistically significant. The F0 data were in the direction predicted by Hy- pothesis 2, but the ETAT data were in the opposite direction. The results were in line with the reasoning behind the original hypotheses, and the lack of significant differ- ences was discussed in relation to the nature of the instru- ments. Both instruments were shown to be applicable to this population. The support given to Hypothesis 3 is in agreement with previous studies. AGGRESSIVE AND SEXUAL FANTASIES IN VIOLENT AND NON-VIOLENT PRISON INMATES BY Benjamin Beit-Hallahmi A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1970 ACKNOWLEDGMENTS I would like to express my thanks to Dr. Albert I. Rabin, the chairman of my doctoral committee, who has guided me through the final stretch of my way toward the Ph.D. His patience and encouragement helped me to alleviate some of anxieties involved. I owe a special debt of gratitude to Dr. Robert A. Zucker, who has been a source of encouragement and help not only in research, but also in my clinical work. Dr. Bertram P. Karon and Dr. Andrew M. Barclay have helped me with this study through invaluable advice and suggestions. I want to mention my debt to Dr. Barclay for contributing one of the main hypotheses for this study. As this marks the end of my doctoral studies, I would like to thank the faculty and graduate students of the Psychology Department for making the last four years so exciting and fruitful for me. ii TABLE OF CONTENTS Chapter INTRODUCTION . . . . . . . . . . . Aims . . . . . . . . . . . . Theoretical Overview . . . . . . . Studies of Fantasy and Overt Aggression . Studies Correlating Aggressive and Sexual Fantasy . . . . . . . . . . . . Studies Using Fantasy Questionnaires . . AIMS OF PRESENT STUDY . . . . . HYPOTHESES . . . . . . . . . . . Hypothesis Hypothesis Hypothesis Hypothesis QWNP o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o 0 METHOD . . . . . . . . . . . . . Instruments . . . . . . . . . . Fantasy Questionnaire (FQ) . . . . . Experimental TAT (ETAT) . . . . . . Subjects . . . . . Procedure . . . . . Scoring the F0 . . . Scoring of the ETAT . . Testing the Effects of Relevant Variables The Effects of the Length of Time Spent in Prison. . . . . The Effects of the Length of Stories . The Revised Scoring . . . . . . . Operational Definitions of Hypotheses . . Hypothesis 1 . . . . . . . . . Hypothesis 2 . . . . . . . . . Page 30 35 35 35 35 Chapter Hypothesis 3 . . . . .. . . . . . Hypothesis 4 . . . . . . . . . . RESULTS 0 O O O O O O O O O O I 0 Testing the Hypotheses . . . . . . . . Hypothesis 1 . Hypothesis 2 . Hypothesis 3 . Hypothesis 4 . Additional Findings . . . . . . . . . Differences on Single ETAT Scores . Rank Pattern of Group Means on the ETAT Scores . . . . . . . . . . . . The Effects of ETAT Stimuli . . . . . Black and White Differences . . . . . Achievement Ratings for Picture T . . . DISCUSSION 0 O O O O O O O O O O O Level of Violence and Fantasy Content . . . The Relationship Between Sexual and Aggressive Fantasies . . . . . . . . The Relationship Between the Two Instruments The Effects of Background Factors The Nature of the Population . SUMMARY REFERENCES APPENDICES Appendix I . Appendix II Appendix III Appendix IV Appendix V . Appendix VI Appendix VII Appendix VIII Appendix Ix iv Page 36 36 37 37 37 39 40 41 41 41 42 43 43 44 46 47 50 50 52 52 54 56 60 60 72 73 84 85 89 90 91 95 LIST OF TABLES Table Page 1. I.Q. Rating System . . . . . . . . . . 26 2. Means and Standard Deviations of I.Q. Ratings . . . . . . . . . . . . . 26 3. Analysis of Variance on I.Q. Ratings . . . . 27 4. Correlations Between I.Q. Ratings and Raw ETAT Scores. N = 60 . . . . . . . . 27 5. Educational Achievement Ratings . . . . . 28 6. Means and Standard Deviations of Educational Level Ratings O O O O O O I O O O O 29 7. Analysis of Variance Table for Educational Level Ratings 0 O O O O O O O O O O 29 8. Correlations between Educational Level and Other Variables. N = 60 . . . . . . . 29 9. Means and Standard Deviations of Time Spent in Prison During the Present Term, in Months 0 O O O I O O O O O O O O 30 10. Analysis of Variance on the Means of Time Spent in Prison . . . . . . . . . . 30 11. Significant Correlations between Time in Prison (in Months) and FQ Scales. N = 65 O O O O O O O O O O O O O 31 12. Means and Standard Deviations of Time Spent in Prison, Including Previous Terms, in Months . . . . . . . . . . . . 31 13. Means and Standard Deviations of the Number of Words in the Five ETAT Stories . . . . . . . . . . . . . 33 Table . Page 14. Means and Standard Deviations for the Sum of Words for all Stories . . . . . . 34 15. Correlations between the Number of Words and Sex and Aggression Scores on ETAT Stories . . . . . . . . . . 34 16. Means and Standard Deviations of FQ Aggression Scores for Three Groups . . . . 37 17. One Way Analysis of Variance for F0 Aggression Scores . . . . . . . . . 38 18. Means and Standard Deviations of ETAT Aggression Scores for Three Groups . . . . 38 19. Means and Standard Deviations of F0 Sex and Aggression Scores . . . . . . . 39 20. Means and Standard Deviations of ETAT Sex and Aggression Scores for Groups W & VN and NV 0 O O O O O O O 0 O 4 0 21. Correlations Between Sex and Aggression Measures . . . . . . . . . . . . 41 22. Summary of Rank Order of Group Means on Eight ETAT Scores . . . . . . . . . 42 23. Significant Differences Between Black ‘and White SS 0 O O O O O O O O O O 44 24. Means and Standard Deviations of Achieve- ment Ratings for Picture T . . . . . . 44 25. Significant Correlations Between Achieve- ment Ratings and Other Scores . . . . . 45 26. Summary of Results Related to Hypotheses . . 46 27. Summary of Group Differences by Rank Ordering of Group Means . . . . . . . 48 28. Significant Correlations Between Sex and Aggression Scores (Corrected) on the ETAT I O O O O O O O O O O 9 0 29- Group Means for Sex and Aggression Scores on Four ETAT Stories . . . . . . . . 95 vi INTRODUCTION This study was designed to investigate the aggressive and sexual fantasies of prison inmates, and to answer three basic questions: 1. Is there a relationship between the level of fan- tasy aggression (and sex) and the inmates' crim- inal history and prison behavior? 2. What is the relationship between aggressive and sexual fantasies, in this group? 3. What is the relationship between the expressions of fantasy through two instruments, a fantasy questionnaire and a TAT-like task? Theoretical Overview What McClelland (1966) called "The Relation of Thought to Action" has long been a major concern of psychol- ogists with both theoretical and practical implications. Fenichel (1945, p. 50) represented a "purist," intra- psychic approach when he wrote: "As long as thinking is not followed by action it is called fantasy." Since Rorschach (1921) and Murray (1937) fantasy processes have been studied as representing stable personality characteristics or action tendencies (Lindzey, 1952). Psychologists, with the demands of clinical work and academic research, have become inter- ested in those thoughts that are followed by action. Traditional psychoanalytic theory (Freud, 1949) has emphasized the wish-fulfilling function of fantasy. Satis- faction, according to this approach, could be achieved in fantasy when it was absent in reality. Fantasy is always a satisfaction of some need, and masochistic daydreams satisfy masochistic needs. A behavioristic approach would see fan- tasies as drive reducing. In this case the behavioristic approach is similar to the psychoanalytic approach (Wagman, 1965). The early psychoanalytic position on fantasy proposed that all forms of imagination grew out of suppressed desires. Freud went so far as to state in his paper on the poet and daydreaming that "happy people never make fantasies, only un- satisfied ones do" (Freud, 1962). In terms of energy con- cepts Freud postulated that thought and fantasy discharged small quantities of energy and in this fashion permit delay in gratification. This drive reducing thought decreases the pressure on the child and opens the way for planning, organ- ization of behavior and the synthetic and defensive capaci- ties of the ego. Psychoanalytic ego psychology has emphasized the fact that ego functions such as thought, language and motor skills may derive energy from sources not specifically re— lated to libidinal conflicts (White, 1959; 1964). Hartman (1958) suggested that fantasy itself need not arise only out of frustration or defense, but may serve adaptive functions for the organism. Today the earlier psychoanalytic conception of fan- tasy as involving a partial drive reduction and serving as a defense mechanism remains the most widespread. Another point of view, associated with the appearance of the TAT technique and the New Look school in perception in the 40's was one that emphasized the direct relationship be- tween fantasy expressions of drives and actual behavior. This approach culminated in the need Achievement studies by the McClelland group (McClelland gt_al., 1953). Lazarus (1961, 1966) criticized the direct-expression View and presented a substitutive-defensive analysis of fan- tasy productions. According to Lazarus, the direct relation- ship may exist for ego-oriented fantasy, while the substitu- tive relationship exists for more "primitive" fantasies. The problem of the relation of aggressive fantasies to aggressive actions has served as a paradigm for the more basic question of the relationship between fantasy and be- havior. The literature dealing with this problem will be re- viewed below together with other studies which have contrib- uted to the design and content of the present one. Studies of Fantasy and Overt Aggression Studies correlating fantasy and aggressive behavior before 1960 were reviewed by Buss (1961), and such a compre- hensive review will not be attempted here. This review will concentrate on studies done on deviant populations, since they are the most relevant to the present study. Adolescents who had been semi-institutionalized for truancy, stealing, and disorderly behavior were studied by Mussen and Naylor (1954). TAT aggression was found to be significantly related to behavioral aggression, as rated by cottage supervisors. The most aggressive boys also had the least number of punishment themes. Stone (1956) divided army prisoners into three groups: 1. Non violent crimes and no history of violence. 2. Desertion with no history of violence. 3. Offenses of murder and attempted murder. The violent group (3) showed significantly more TAT aggression compared to the two non-violent groups. Army men were also used in the study by Purcell (1956). Fifty-seven patients in an Army psychiatric clinic were rated as to antisocial aggressiveness. Their TAT protocols were scored for: l. Aggression. 2. External punishment. 3. Internal punishment. The most antisocial group manifested the following significant differences: higher aggression scores; more direct and undisguised aggression; less punishment antici- pated; more themes of aggression as an excuse for aggres- sion; and fewer themes of guilt or shame. Jensen (1957) had teachers select three groups of high school boys: 1. Aggressive Bad-~disruptive, unruly, destructive. 2. Aggressive Good--active, energetic, competitive. 3. Passive--quiet, retiring, unassertive. The Aggressive Bad group was highest compared to the other two in the proportion of TAT aggressive stories having no punishment or defense themes, sex themes, tabooed sex themes, tabooed violence and tabooed lan- guage. The ordinary (non-taboo) measures of TAT aggression did not separate the aggressive from the passive boys. Weissman's (1964) study evaluated the usefulness of the TAT for predicting acting out behavior. The 83 were de- linquent and nondelinquent adolescents, who were adminis- tered the standard TAT. The three dimensions that revealed significant and consistent differences were: 1. The number of aggressive stories. 2. The aggressive stories to cards not ordinarily associated with aggression. 3. Reaction time. The differentiation between aggressive and non— aggressive groups was clearest when the differences in actual aggression were extreme. The best predictor of acting out was low reaction time, and not fantasy pro- ductions, but Weissman for some reason suggested a com- bined measure as a better predictor in the future. This study is the most similar to the present one found in the literature, and thus served as a source of ideas, but it was felt that the methodology, and especially the statistical treatment, left something to be desired. Weissman used multiple, individual t tests and median tests, while it seems that a single analysis of variance and range tests would have been more appropriate. Schaefer and Norman (1967) compared anti-social adolescents to a control group of normal youngsters. They found that the control group produced significantly more aggressive TAT stories than the anti-social group. Support to the assumption that stimulus factors affect the relation between fantasy and behavior was given in the study by Colemen (1967). In the case of the TAT only card IBBM ("high Aggressive") differentiated between aggressive and non-aggressive children. Megargee and Cook (1967) attempted to determine if the contradictions in the literature regarding relationships between projective test measures and actual aggression is due to the way different scales and criteria were used. They used four different TAT scales, but there was only cri- terion measure with which the four scales had more than a chance relation. It was School Conduct. For the remaining ten behaviOral criteria only one of the forty correlations was statistically significant. Their results were described as "discouraging." James and Mosher (1967) found that thematic aggres— sion on high stimulus relevance cards was significantly re- lated to physically aggressive behavior, but no such relation existed for the low stimulus cards. Silber and Courtless (1968) compared fantasy aggres- sion, as measured by the TAT of mentally retarded and nonre- tarded offenders. It was predicted that retarded offenders would produce more fantasy aggression and that the more serious offenders among the retarded would manifest more aggressive fantasies, compared to the less serious ones. Contrary to expectation, the serious offenders mani- fested the least fantasy aggression, and there were no dif- ferences between retarded and non—retarded offenders. Buss (1961), in his summary, writes: "The clinical studies yielded one clear-cut positive finding: TAT aggres- sion is directly related to assaultiveness. This generaliza- tion holds for prisoners of both sexes and several kinds of patient populations. TAT aggression is not related to assertiveness, uncooperativeness, and other behaviors peri- pherally associated with aggression" (p. 153). However, the criteria for classification as "aggres- sive" varied in the various studies. They can be divided into (a) past history of anti-social actions and (b) ratings of relatively more recent behavior. It still remains to be seen if there are any differ- ences related to the two kinds of criteria. Is "past his- tory" more important in determining the fantasy content of institutionalized offenders, or is institutional discipline a more important factor? Support for the former hypothesis will support notions of a violent "life style," while support for the latter will aid notions of situational violence. Studies Correlating Aggressive and Sexual Fantasy In addition to zoological and physiological evidence for the sex-aggression link (MacLean, 1965), the connection between sexual and aggressive motivations has been discussed by several authors. Freud (1955) mentioned this connection in relation to the hypothesized mating behavior in prehistoric humans. Ac- cording to this conception, which is quite popular, the female had to be "grabbed by the hair" and subdued before sexual intercourse took place. Thus the aggressive component became part of the male sexual arousal. From a different theoretical angle, this connection between two drives can be viewed in the light of the Hullian concept of D, the generalized drive level (Hull, 1943). According to Hull, increasing the strength of any drive will increase the energy available to all drives. Clark (1953) was the first one to show the correla- tion between sex and aggression in thematic fantasy produc- tions. His Ss were aroused sexually, while under the influ- ence of alcohol, and then wrote TAT stories that showed an increase in aggressive and sexual imagery, compared to a control group. Barclay and Haber (1965) showed that Ss could be aroused sexually following an increase in aggression. Their Ss, students who were treated in an irritating manner, showed an increase in sexual imagery on the TAT, as compared to a control group. Barclay (1967) showed that the correlation exists for both men and women, after the above mentioned studies es- tablished its existence for men only. All studies showing the sex-aggression correlation were conducted with college populations. All of them also used an arousal technique as part of the experimental proce- dure. It still remains to be seen if the same correlation will appear with a different population and without an arousal procedure. 10 Studies Using Fantasy Questionnaires Most of the research using fantasy questionnaires have been done by J. L. Singer and his associates, who used them in studies of daydreaming. The general daydreaming questionnaire, developed by Singer and his colleagues (Singer and Antrobus, 1966) calls for the subject to indicate the frequency, on a five point scale, with which he has experienced each of a series of ap- proximately one-hundred daydreams. Singer concentrated on a presumably normal, adult population of at least college education. The first normative study, by Singer and Craven (1961), reported on a sample of 250 presumably normal adults between the age of 19 and 50 of at least college education. Ninety-six per cent of the respondents reported that they en- gaged in some form of daydreaming daily. Visual imagery was found to be the predominant modality for fantasy and day- dreaming is reported to have occurred chiefly when the sub- jects were alone. Items dealing with future actions and particularly interpersonal contact were high in frequency. Most subjects reported that they enjoyed daydreaming and a general conclusion was that daydreaming is a widespread com- mon occurrence when people are alone or in restful states. Large differences in daydreaming frequency and con- tent were found among persons from different subcultural groups. In the Singer and McCraven study (1961) respondents were classified into the following subcultural groups: 11 Negro, Jewish, Anglo-Saxon, mixed parentage and other. The Jewish and Negro groups showed the highest daydream frequency and the Anglosaxons the lowest. In a subsequent study (Singer and McCraven, 1962) adults from several socio-cultural groups: Italian, Irish, Jewish, Negro, Anglo-Saxon and German subcultures, were com- pared on fantasy content. All of them were college educated Americans born of at least the second generation but with both parents from the same national origin background. The order of daydreaming frequency reported was (from high to low) Negro, Italian, Jewish, Irish, German and Anglo-Saxon. This finding reflects the relative upward mobility, insecurity and even the pattern of integration of the various groups in the United States. The last three groups represent subcultures which are in superior economic positions in the United States. The Negro respondents show a high degree of relative- ly concrete, realistic daydreams while the Jewish and Anglo- Saxon groups go to the opposite extreme, minimizing reliance on highly realistic fantasies. The Irish subjects showed by far the greatest resort to mystical and otherwise fantastical fantasies. No significant group differences emerged on day- dreams classified as involving direct outward aggresSive and destructive drives, but when controlled aggression is con- sidered significant group differences occurred, with the Irish showing by far the highest and the Anglo-Saxon the lowest scores in these categories. The Negro, JeWiSh: and 11 Negro, Jewish, Anglo-Saxon, mixed parentage and other. The Jewish and Negro groups showed the highest daydream frequency and the Anglosaxons the lowest. In a subsequent study (Singer and McCraven, 1962) adults from several socio-cultural groups: Italian, Irish, Jewish, Negro, Anglo-Saxon and German subcultures, were com- pared on fantasy content. All of them were college educated Americans born of at least the second generation but with both parents from the same national origin background. The order of daydreaming frequency reported was (from high to low) Negro, Italian, Jewish, Irish, German and Anglo-Saxon. This finding reflects the relative upward mobility, insecurity and even the pattern of integration of the various groups in the United States. The lastthree groups represent subcultures which are in superior economic positions in the United States. The Negro respondents show a high degree of relative— 1y concrete, realistic daydreams while the Jewish and Anglo- Saxon groups go to the opposite extreme, minimizing reliance on highly realistic fantasies. The Irish subjects showed by far the greatest resort to mystical and otherwise fantastical fantasies. No significant group differences emerged on day- dreams classified as involving direct outward aggresSive and destructive drives, but when controlled aggression is con- sidered significant group differences occurred, with the Irish showing by far the highest and the Anglo—Saxon the lowest scores in these categories. The Negro, Jewish, and 12 Italian groups were particularly high on erotic drive fantasies. The Negro subjects also showed considerable preoccupation with fairly concrete, sensual gratifications and material security. Cazavelan and Epstein (1966) gave a fantasy inven- tory, consisting of 202 daydreams and questions about day- dreaming to 20 paranoid schizophrenic women and 20 closely matched controls. The reported daydreams of the paranoid schizophrenics were more mystical, more manifestly patho- logical and dealt less often with problem solving that day- dreams of the controls. The authors conclude that defensiveness to hostility and absence of heterosexuality in fantasy are supportive evidence for the Freudian hypothesis of reaction formation to homosexual love which is denied and projected outward. This is an example of the use of a fantasy inventory for testing an hypothesis regarding psychopathology. The Wagman fantasy questionnaire (based on the Singer questionnaire) consisting of 120 specific daydream items and obtaining scores on 24 sub-scales has been used in several studies. Wagman (1965) tested the general hypothesis that the sheer amount of daydreaming is positively correlated with general personality characteristics of anxiety or neuroticism and negatively associated with repressive or denial type characteristics. He used A and R MMPI scales and also the 13 L and Si scales. The hypothesis was supported by the findings. For men there was a positive correlation between daydreaming frequency and anxiety measures and negative cor- relation with repression and defensiveness measures. For women the same correlations were found. Similar findings were obtained in the Singer and Schonbar (1961) study with a group of middle aged women. The relationships between university achievement and daydreaming behavior were studied in another study by Wagman (1968). The investigation hypothesized an inverse associa- tion between academic achievement level and frequency of daydreaming activity. This hypothesis was supported for women but not for men by the findings of statistically sig- nificant negative correlations between university GPA and each of the following types of daydreams: guilt or superego, outward aggression, death, passivity, pre-genitality, pre- genital narcissism, physical attractiveness, sexual (per- verse), achievement, self-aggrandizement, parent and family, money and possessions, improbably, pleasant, and total. This is the only study in the literature correlating scores on the daydreaming questionnaire with actual behavior. From the above it is clear that studies using fantasy questionnaires were mainly normative, and were done mainly with college populations. There have been few attempts to relate reported fan- tasies to any kind of actual behavior, or to use fantasy questionnaires as a projective or diagnostic tool. AIMS OF PRESENT STUDY There were three main aims for this study. 1. Comparing the relative effects of habitual vio- lence and temporary inhibition of aggression on fantasy content. Meyer (1967) showed that inmates who were in prison for violent offenses, and who did not have any discipline problems inside the prison, had a higher blood pressure, com- pared to those violent offenders with disciplinary problems. Meyer interpreted his results to mean that this lack of dis- ciplinary problems was actually an inhibition of habitual aggression. The present study tried to see if this inhibition will be reflected in aggressive fantasies. It had been predicted that those violent offenders who are "well behaved" inside the prison would report and produce more aggressive fantasies, compared to those violent offenders with disciplinary problems. 2. Comparing two instruments for the assessment of fantasy. Most studies dealing with fantasy aggression and overt behavior have used only one instrument, in most cases 14 15 a projective one. This study uses two separate instruments, the Fantasy Questionnaire (F0), and an experimental series of TAT pictures (ETAT). This is the first study using such a fantasy ques- tionnaire with a prison population. 3. Testing the correlation found with other popula- tions between sexual and aggressive fantasies with this population. In the other studies (Clark, 1953; Barclay and Haber, 1965; Barclay, 1967) some means of aggression or sexual arousal was used as an independent variable (e.g. insulting the $3). In the present study no manipulated arousal exist- ed, except for the FQ and ETAT themselves. HYPOTHESES Hypothesis 1 Ss who have a history of violent offenses, but no record of disciplinary problems inside the prison, will re- port and produce more aggressive fantasies, compared to Ss who have a history of both violent offenses and disciplinary problems inside the prison. Hypothesis 2 83 with a history of non-violent offenses and no prison disciplinary problems will be lower on both sex fan- tasy and aggressive fantasy measures, compared to all 85 with a history of violent offenses. Hypothesis 3 Frequency of aggressive fantasies will be positively correlated with frequency of sexual fantasies, for all 83 on both instruments. Hypothesis 4 Fantasy measures based on indirect reports will be positively correlated with direct fantasy productions. 16 METHOD Instruments Fantasy Questionnaire (FQL The F0 has been developed on the basis of similar ones, authored by Singer and Antrobus (1966) and Wagman (1968). With the permission of the above authors, about eighty items previously used by them were combined with ori- ginal items, written by the present writer, to comprise a questionnaire which included 136 items. This preliminary questionnaire was used in a pilot study with fourty-four State Prison of Southern Michigan (SPSM) inmates. The aim of the pilot study was to test the suitability of the FQ and the procedure to the prison popu- lation, and to find a basis for modifications. The 136 items were subjected to an item analysis and to a group analysis, using the group intercorrelation analy- sis technique developed by Dr. John Hunter and the computer program written by Mr. Stanley Cohen. As had been expected, the FQ was considerably modified as a result of the pilot study. Fifty-four items were eliminated, since they were found to be unrelated to other items and scales. Eighteen 17 18 content scales were devised, based on the statistical group analysis. Appendix I contains the FQ and Appendix II contains a breakdown of the items into scales. Experimental TAT (ETAT) The set includes five pictures, and the S is in- structed to tell a structured story in writing, in response to each picture, by answering four questions. (See Appendix III). An initial group of ten pictures were selected by this writer from a collection of pictures previously used in similar studies. They were presented to six judges, male graduate students in psychology, with the instructions to rank the pictures according to their sexual and aggressive arousal potential. The judges were told that the pictures would be used in a study with a male population. They were also asked to select the two most "neutral" pictures. On the basis of the judges' ratings, five pictures were selected. One was classified as neutral (picture T), one as High Sexual (picture B), one as Moderate Sexual (picture C), one as High Aggressive (picture W) and one as Moderate Aggressive (pic- ture A). Pictures C and W were used by Kalin, McClelland and Kahn (1965). The order in which the five pictures appeared in the ETAT booklet was randomized, to control position ef- fects. 19 The scenes shown in the pictures are described below: Picture T.--A young man sitting on a chair in what looks like a farm yard, playing a trumpet. Behind him a large tree, a fence and a house. Picture B.--A young man and a young woman on a beach, dressed in swimsuits only. The girl is lying on her back and the man is sitting behind her. Picture C.--A man and a woman, fully dressed and with overcoats, looking at each other, walking together on a side- walk. Picture W.--A White man seen from the back is looking towards a group of Negroes. He may be running. There is also a photographer (White), whose head is not seen. Picture A.--A group of young men in what looks like a dormitory room. One of them, who looks a little older, talking. At least one of those listening is a Negro. Subjects Subjects for the present study were selected on the basis of several criteria: 1. Age. Only inmates under fifty years of age were selected. 2. I.Q. Only inmates with I.Q. of ninety or above were selected. 3. At least two prison terms, including the present one. 4. At least two years spent at SPSM. 20 The rationale for these criteria was the following: we were interested in inmates with a history of habitual criminality, not in occasional offenders. We were also in- terested in inmates who are intelligent enough and intact enough to complete the experimental task. Additional information gathered was in regard to the specific offense and to the disciplinary record over the last two years. This information was used to classify the Ss in terms of violence. The actual selection process took place in the following manner: official listings of the Michigan Department of Correction were used to select names of inmates with at least two prison terms. The listing in- cluded the legal definition of the offense. The following offenses were used to classify inmates as violent: Murder First Degree, Murder Second Degree, Man- slaughter, Kidnapping, Assault with Intent to Commit Gross Bodily Harm Less Than Murder, Assault with Intent to Commit Murder, Felonious Assault, Assault with Intent to Rob and Steal, Armed. The following offenses were used to classify inmates as non-violent: Uttering and Publishing, Forgery, Larceny from Building, Simple Larceny, Possession of Burglary Tools, Entering without Breaking, Breaking and Entering. Following this initial classification, each inmate's internal file, kept by the Deputy Warden was checked. In this file the following information is included: transfers to and from other prisons, disciplinary reports, I.Q., and 21 information regarding any serious problems, such as psychi- atric hospitalization, homosexual activities, involvement in illegal activities inside the prison, etc. At this stage inmates with I.Q. of less than ninety, less than two years at SPSM, or a history of psychiatric hospitalization were elim- inated. Inmates reported to be in isolation because of active homosexuality were also eliminated. At this stage the existence and number of any disci- plinary reports over the previous two years were also noted. On the basis of the initial classification as violent or non-violent and the presence or absence of disciplinary reports, the subjects were classified into one of three groups: 1. Violent offenders with disciplinary reports (VV). 2. Violent offenders without disciplinary reports (VN) . 3. Non-violent offenders (without disciplinary re- ports) (NV). The number and race of subjects in each group was as follows: 1. Twenty-four Caucasian and twenty Negro Ss. 2. Eighteen Caucasian and fifteen Negro Ss. 3. Eighteen Caucasian and thirteen Negro 55. These number were considerably greater than the num- ber of Ss originally desired for groups, since only some of the 85 had been expected to cooperate. 22 A letter was sent to each one of the 108 Ss (Appen- dix IV), telling him about the study and asking for coopera- tion, with a reward of two packs of cigarettes, which for most inmates equal two days' pay. The 108 names were given to the Prison administra- tion, and the inmates were notified in time for the study. The names were divided into four testing groups, which in turn included about a fourth of each of the three experi- mental groups. Actually, eighty men showed up for the study, but fifteen of them decided not to take part and so only sixty- five completed both the FQ and the ETAT. Those sixty-five were divided according to groups, as follows: 1. Eleven Caucasians and thirteen Negro 85 (group W). 2. Ten Caucasians and eleven Negro 53 (group VN). 3. Eleven Caucasians and nine Negro 53 (group NV). The original design called for the following numbers: 1. Twelve Caucasians and twelve Negro Ss. 2. Twelve Caucasians and twelve Negro Ss. 3. Ten Caucasians and ten Negro 85. Procedure The instruments were administered to the sixty-five Ss in four groups. In two of the groups the F0 was adminis- tered first and in the other two the ETAT was administered first. 23 The instructions (Appendix V) were read by the inves- tigator, who was present in the room to answer additional questions and make sure that instructions were followed. The group sessions took place in one of the classrooms in the Academic School at SPSM. When the completed materials were returned to E, each inmate received two packs of cigarettes. Scoring the FQ The FQ answer sheets were mechanically scored by the MSU scoring office, and the data were transformed to punched cards. To test the internal consistency of the FQ, a sepae rate scoring was done for the even-numbered items (forty-one items) and for the odd-numbered items (forty-one items). The correlation between the two scores was .907 (a < .001). Items on the FQ were responded to on a five point scale. The range of total scores was 2 to 172. Scoring of the ETAT After reviewing the responses to picture T (Neutral), it was realized that the picture served as a strong stimulus for achievement imagery but otherwise was truly neutral. Sexual imagery was found only in two cases, and aggressive imagery only in one. It was then decided to drop the re- sponses to picture T from the analysis of results related to the hypotheses. 24 Responses to pictures A, W, B and C were rated on a five point scale for Sex and Aggression, according to the method developed by Barclay (1967b). The scoring manual is reproduced in Appendix VIII. The ratings were done by a fellow graduate student, who did not know about the research hypotheses or the groups studied. He was told that the 83 were prisoners, and was alerted to the possibility of faulty grammar or lower class slang expressions. Reliability was established in the following way: both the rater (LN) and the writer (BBH) rated the responses of thirteen 53 to pictures A, W, B and C. A total of fifty- two stories were rated. Sex and Aggression ratings for the two raters were then correlated. The correlation for Sex ratings was .925 (a < .001), and for Aggression ratings .975 (a < .001) . The range of total Sex scores was four to sixteen, and the range for total Aggression scores was four to fifteen. Testing the Effects of Relevant Variables Before testing the main hypotheses, it was decided to consider the possible effects of variables which may cause artifacts in this particular group of Ss. I.Q., educational background, productivity on the ETAT stories and length of time spent in prison were consid- ered such variables, and their values were obtained for all 85. 25 Group means for these variables for the three exper- imental groups were compared and the variables were also correlated with all FQ and ETAT scores. The results are reported below. The effects of I.Q. and educational background.-- During the selection process, described above in the Method section, individual I.Q. scores were checked, and only Ss with an I.Q. score above ninety were included. These scores were based on the Army General Classification Test (AGCT), administered to every incoming inmate at SPSM. For the purpose of comparing groups and checking the possible effects of I.Q. level, I.Q. ratings and Educational Level ratings were used. They were obtained from the files‘ of the Michigan Department of Corrections, and were available for sixty out of sixty-five 85. Table 1 presents the rating system. From the selection procedure it is clear that level five was the lowest rating permitted. Table 2 presents the means and standard deviations for I.Q. for sixty Ss. An analysis of variance was performed on the I.Q. rating data for the three groups, and is presented in Table 3. While the differences in I.Q. among the three groups were not significant, there were some significant correlations 26 Table l.--I.Q. rating system. I.Q. Score Rating 0-49 0 50-59 1 60-69 2 70-79 3 80-89 4 90-99 5 100-109 6 110-119 7 120-129 8 130- 9 Table 2.--Means and standard deviations of I.Q. ratings. Group N Mean S.D. Total 60 6.06 1.33 VV 21 5.80 1.72 VN 20 5.85 1.34 NV 19 6.47 1.02 between I.Q. and ETAT raw scores, for the total group of sixty 83. They are presented in Table 4. There were no significant correlations between I.Q. ratings and any of the F0 scales. 27 Table 3.--Analysis of variance on I.Q. ratings. Source SS DF MS F Significance B 5.408 2 2.704 1.396 .262 W 112.524 57 1.974 T 117.933 59 Table 4.--Correlations between I.Q. ratings and raw ETAT scores. N = 60. Score r Significance No. words, story A .360 .01 No. words, story W .254 .05 No. words, story C .275 .05 No. words, story B .361 .01 No. words, story T .301 .05 Sex, story A .369 .01 Sex, stories A and W .416 .01 Aggression, story A .259 .05 Aggression, story W .258 .05 Aggression, stories A and W .337 .05 Sex, all stories .280 .05 Aggression, all stories .350 .05 No. words total .338 .05 28 The use of the corrected ETAT scores (see page 35) was aimed at eliminating the effects of verbal fluency and I.Q. After the corrections were made, none of the Sex or Aggression scores (corrected) correlated with I.Q. The only significant correlation was with the corrected Achievement score -.286 (significant at the .05 level). Ratings for the level of educational achievement were also obtained for the same sixty 85. Table 5 presents the rating system. Table 5.--Educational achievement ratings. Level Rating Illiterate Gr. 1-2 Gr. 3-4 Gr. 5-6 Gr. 7-8 H.S. 1-2 H.S. 3-4 Coll. 1-2 Coll. 3-4 CDQONU'Iwal-‘O Means and standard deviations for the educational level ratings are presented below. (See Table 6.) An analysis of variance was performed on the educa- tional level ratings and is reported in Table 7. Significant correlations between educational level ratings and other scores are reported in Table 8. Table 6.--Means and standard deviations 29 of educational level ratings. Group N Mean S.D. Total 60 4.65 1.07 VV 21 4.47 .87 VN 20 4.30 1.26 NV 19 5.00 .81 Table 7.--Analysis of variance table for educational level ratings. Source SS DF MS F Significance B 5.145 2 2.572 2.467 .094 W 59.438 57 1.042 T 64.583 59 Table 8.--Significant correlations between educational level and'other variables. N 60. Score r Significance I.Q. .364 .05 No. words story C .266 .05 story B .255 .05 story T .295 .05 Time in prison, present term -.299 .05 Sex, picture W (corrected) -.397 .05 Sex, pictures A and W (corrected) -.382 .05 Sex, total score (corrected) -.350 .05 30 The Effects of the Length of Time Spent in Prison Information on the length of time spent in prison during the present term, and including previous terms, was obtained from the Ss through the Personal Questionnaire (Appendix VI). Means and standard deviations of the number of months spent in prison for the latest sentence are given in Table 9. Table 9.-—Means and standard deviations of time spent in prison during the present term, in months. Group N Mean S.D. Total 65 44.40 31.98 VV 24 54.54 36.12 VN 21 46.33 33.09 NV 20 30.20 19.14 An analysis of variance performed on the data for the three groups is presented below. Table lO.--Analysis of variance on the means of time spent in prison. Source SS DF MS F Significance B 6579.774 2 3289.887 3.4639 .037 W 58883.825 62 949.739 T 65463.600 64 31 All significant correlations between the length of time spent in prison during the present term and F0 scales are reported below. Table ll.--Significant correlations between time in prison (in months) and FQ scales. N = 65. Scale r Significance Self Destruction .382 .01 Sadism .320 .05 Escape .313 .05 As the table shows, there were only three signifi- cant correlations, out of eighteen correlations with all FQ scales. Means and standard deviations of the total time spent in prison, including previous terms, are reported in Table 12. Table 12.--Means and standard deviations of time spent in prison, including previous terms, in months. Group N Mean S.D. Total 65 107.69 64.29 VV 24 105.62 55.23 VN 21 108.19 66.04 NV 20 109.65 75.02 32 As Table 12 shows, differences among the three groups on the total time spent in prison, including previous terms, were slight. No significant correlations were found between this variable and any other measure. The Effects of the Length of Stories Previous studies (Kagan and Lesser, 1961; Murstein, 1963) have shown that the length of a TAT story may affect its content ratings considerably. For this reason, it was decided to check for the possible influence of this factor in the present study. The number of words in each story was counted. Means and standard deviations of the number of words for all five stories are reported in Table 13. Analyses of variance have been performed on the data for all five stories, but only on the case of story T did the F ratio reach significance below .10. (F = 2.6907, sig- nificance = .076.) It was then decided to sum up the number of words for all stories. The results are presented in Table 14. An analysis of variance showed that the differences between the groups were not significant. Table 15 presents the correlations between the number of words in each story and the Sex and Aggression scores for the 65 Ss. The table shows a clear connection between produc- tivity in a story in terms of length and the scores for the 3 3 om.om o~.mm mm.mm ~m.vo mm.sm so.om ms.mm mo.ao mv.m~ mm.mm Hence av.mH m~.mm nm.~m om.mo em.em oe.mm mm.- oo.mw om.a~ o¢.mm >2 mm.m~ mo.m¢ «m.- so.mm vv.- oo.em mm.o~ mo.ms om.m~ we.mm z> v~.o¢ mm.mm ms.nm mo.ms ~m.~m m~.mm oa.mv ea.on ms.am mo.mm >> .o.m new: .a.m new: .o.m use: .o.m new: .o.m new: scene a muoum m muoum o muoum 3 muoum a muoum .mowuoum 849m o>fim on» ca mcuo3 mo Hones: ecu mo mGOHDMH>mc cumpcmum cam mcmozul.ma manna 34 Table l4.--Means and standard deviations for the sum of words for all stories. Group N Mean S.D. Total 65 299.96 134.35 VV 24 337.79 174.19 VN 21 259.80 101.57 NV 20 296.75 98.01 Table 15.--Corre1ations between the number of words and Sex and Aggression scores on ETAT stories. Score Sex Aggression Story A .23 .12 W .02 .25* C .26* .32* B .26* .29* All stories .43** .36** * Significant at the .05 level. ** Significant at the .01 level. same story. This connection is more clearly indicated for the Sex scores on all stories, and for both Sex and Aggres— sion scores on the stories given to the Sex pictures, B and C. 35 As was shown above, productivity on stories (number of words) was significantly related to both I.Q. and educa- tional level, and the later were also significantly related to the raw ETAT scores themselves. The Revised Scoripg To control for the effects of story length on s scoring, and possibly also the effects of I.Q. and educa- tion, a corrected score was created by dividing the raw score for each story by the number of words in the story. These corrected scores were the ones used in testing the hypotheses. The range of the total Sex corrected scores was be- tween .01204 and .05699. The range of the total Aggression corrected scores was between .01290 and .09615. Operational Definitions of Hypotheses Hypothesis 1 85 who have a history of violent offenses, but no record of disciplinary problems inside the prison (Group 2 or "VN"), will report more aggressive fantasies on the FQ, and will produce more aggressive fantasies on the ETAT, compared to Ss in group 1 ("VV"), who have a history of both violent offenses and disciplinary problems. Hypothesis 2 Group 3 ("NV"), consisting of non-violent inmates with no disciplinary record, will be significantly lower, 36 on both Sex and Aggression measures, compared to the other two groups. Hypothesis 3 Frequency of aggressive fantasies will be positively correlated with frequency of sexual fantasies, on both the FQ and the ETAT. Hypothesis 4 Sex and Aggression scores on the ETAT will be posi- tively correlated with the Sex and Aggression scales of the F0. RESULTS Testing the Hypotheses Hypothesis 1 55 who have a history of violent offenses, but no record of disciplinary problems inside the prison (Group 2 or "VN"), will report more aggressive fantasies on the FQ, and will produce more aggressive fantasies on the ETAT, com- pared to Ss in group 1 ("VV"), who have a history of both violent offenses and disciplinary problems. Hypothesis 1 was tested by comparing the Aggression scores of all three groups (VV, VN and NV) on both the FQ and the ETAT. Table 16 shows the means and standard deviations of Aggression scores on the FQ, for the three groups. Table 16.-—Means and standard deviations of PO Aggression scores for three groups. Group VV VN NV Mean 12.79 13.61 9.55 S.D. 10.40 10.18 8.99 37 RESULTS Testing the Hypotheses Hypothesis 1 83 who have a history of violent offenses, but no record of disciplinary problems inside the prison (Group 2 or "VN"), will report more aggressive fantasies on the FQ, and will produce more aggressive fantasies on the ETAT, com- pared to Ss in group 1 ("VV"), who have a history of both violent offenses and disciplinary problems. Hypothesis 1 was tested by comparing the Aggression scores of all three groups (VV, VN and NV) on both the FQ and the ETAT. Table 16 shows the means and standard deviations of Aggression scores on the FQ, for the three groups. Table 16.--Means and standard deviations of F0 Aggression scores for three groups. Group VV VN NV Mean 12.79 13.61 9.55 S.D. 10.40 10.18 8.99 37 38 An analysis of variance was performed and is presen- ted in Table 17. Table l7.--One way analysis of variance for FQ Aggression scores. ======a====== m Source SS DF MS F Significance Between 189.893 2 94.946 .964 .387 Within 6103.860 62 98.449 Total 6293.753 64 Table 18 presents the means and standard deviations for the Aggression scores on the ETAT, for all three groups. Table 18.--Means and standard deviations of ETAT Aggression scores for three groups. Group VV . VN NV Mean .0313 .0349 .0348 S.D. .0181 .0137 .0112 An analysis of variance comparing the three groups produced an F ratio of .44476, which is not significant at any acceptable level. 39 We can say now that the group means came out in the order predicted by the hypothesis, but the differences were not statistically significant. The trend was similar on the two instruments. Hypothesis 2 Group 3 ("NV"), consisting of non-violent inmates with no disciplinary record, will be significantly lower, on both Sex and Aggression measures, compared to the other two groups. For testing hypothesis 2, the data for groups VV and VN were combined and they were treated as one group, that was then compared to group NV. Table 19 presents the comparison on FQ Sex and Ag- gression scores. Table 19.--Means and standard deviations of PO Sex and Ag- gression scores. Variable Group ‘VV & VN NV t Siggiii- N 45 20 Sex Mean 14.44 11.55 1.372 NS S.D. 9.35 7.06 Aggression Mean 13.17 9.55 1.101 NS S.D. 10.19 8.99 Table 20 presents the comparison for the ETAT data. 40 Table 20.--Means and standard deviations of ETAT Sex and Aggression scores for groups VV & VN and NV. Variable Group VV & VN NV t Siggigi- N 45 20 Sex Mean .0301 .0310 .291 NS S.D. .0102 .0120 Aggression Mean .0329 .0348 .543 NS S.D. .0165 .0112 The differences between groups VV&VN and NV on the FQ Sex and Aggression scales are in the predicted direction, though they fail to reach significance. The differences on the ETAT Sex and Aggression scores are in a direction contrary to that hypothesized. Hypothesis 3 Frequency of aggressive fantasies will be positively correlated with frequency of sexual fantasies, on both the FQ and the ETAT. To test hypothesis 3, product moment correlations (r) were computed between: the FQ Sex scale and the FQ Aggres- sion scale: the ETAT Total Sex score (corrected for length of story) and the ETAT Total Aggression score (also corrected). The results are presented in Table 21. From Table 21 we can conclude that the hypothesis was clearly supported. 41 Table 21.--Correlations between Sex and Aggression measures. N - 65. Instrument r Significance F0 .683 .001 ETAT .434 .001 There were also a number of significant correlations between Sex and Aggression scores on individual ETAT stories. They are reported in Appendix VII. Hypothesis 4 Sex and Aggression scores on the ETAT will be posi- tively correlated with the Sex and Aggression scales of the FQ. To test this hypothesis the correlations between the FQ Sex scale and the ETAT Total Sex score (corrected), and between the FQ Aggression scale and the ETAT Total Aggression score (corrected) were computed. The correlation between the Sex measures was -.033, and between the Aggression measures .171, both not signifi- cant. It is clear that this hypothesis was not supported. Additional Findings Differences on Single ETAT Scores The three groups were compared on the Sex and Ag- gression scores (corrected) for the four ETAT pictures. Out 42 of the eight scores, only in one case was the difference significant: on the Sex score for picture W, an analysis of variance produced an F ratio of 3.4038, significant at the .040 level. The rank order of group means (from highest to lowest) for this picture was VN, VV, NV. Rank Pattern of Group Means on the ETAT Scores When the group means on the eight single ETAT scores were rank ordered, a clear cut pattern emerged, with group VN first on most of the scores, group NV second, and group VV last. This pattern is presented in the table below. Table 22.--Summary of rank order of group means on eight ETAT scores. Group VV VN NV Rank 1 l 5 2 2 1 2 5 3 6 l 1 This pattern may be seen as lending support to the main hypothesis and the reasoning behind it. 43 Appendix IX includes the actual group means for the ETAT scores. The Effects of ETAT Stimuli As expected, ETAT pictures (B and C) rated ini- tially as sexual, aroused significantly more sexual im- agery, as reflected in the Sex corrected scores, than the pictures rated as aggressive. The opposite was also true, and aggressive pictures elicited higher ag- gressive ratings. Both differences were significant at the .001 level (t for Sex scores - 7.629; t for Aggres- sion scores - 8.791). There was also a significant difference on Ag- gression scores between the High Aggressive picture W, and the Moderate Aggressive picture A (t - 5.466, a < .001). This was also in the expected direction. The difference between the High Sexual picture B and the Moderate Sexual picture C, though in the expected direc- tion, was far from being significant. Black and White Differences Testing for possible race differences was done by dividing the whole group of 65 83 into a Black group (N=33), and a White group (N=32), and comparing them. Only two significant differences (out of eighteen) were found on the F0 scales, when comparing the Black- White groups, as presented in Table 23. 44 One significant difference (out of eight) was found on the ETAT. Table 23.-~Significant differences between Black and White Ss. Black White Scale Signifi- Mean S.D. Mean S.D. t cance Food 5.45 3.92 3.59 3.32 2.068 .05 Bizzare 2.15 2.10 .96 1.33 2.748 .01 Aggres- sion, story A (cor- rected) .0420 .0332 .0280 .0237 2.00 .05 Achievement Ratings for -: Picture T Picture T was rated for achievement imagery, accord- ing to the method presented by McClelland gp_g£. (1958). Means and standard deviations of corrected achieve- ment ratings are presented in Table 24. Table 24.--Means and standard deviations of achievement rat- ings for picture T. Group N Mean S.D. Total 65 .0542 .0505 VV 24 .0590 .0526 VN 21 .0513 .0510 NV 20 .0516 .0496 45 An analysis of variance performed on the Achievement data for the three experimental groups produced an F ratio of .16567, which is not significant. The only significant correlations between Achievement ratings and other scores are reported below. Table 25.--Significant correlations between Achievement rat- ings and other scores. Score r Significance No. words, story W .260 .05 No. words, story B .297 .05 No. words, story T .326 .05 I.Q. -.286 .05 DISCUSSION The table below presents a summary of the findings relating to the original hypotheses. Table 26.--Summary of results related to hypotheses. Hypothesis FQ ETAT 1 Predicted direction, NS Predicted direction, NS 2 Predicted direction, NS Opposite direction, NS 3 Supported .001 Supported .001 4 Not supported From the table it can be concluded that one out of four hypotheses was supported, and one was not. In the other two cases, no significant results were obtained, but the trend was as predicted. It seems that it will be more fruitful to discuss the results in terms of the three questions posed in the Intro- duction, rather than discuss each hypothesis. The three questions were: 1. Is there a relationship between the levels of fantasy aggression, sex fantasies and the in- mates' criminal history and prison behavior? 46 47 2. What is the relationship between aggressive and sexual fantasies in this group? 3. What is the relationship between the expressions of fantasy through the two instruments, a fantasy questionnaire and a TAT-like task? Level of Violence and Fantasy Content Our 53 were classified into three levels of aggres- sion, on the basis of their past criminal history and their more recent disciplinary record inside the prison. Hypothesis 1 tested the differences between VV and VN, assuming that the restraint showed by the VN Ss inside the prison will affect their responses to the FQ and the ETAT. The results on both instruments were in the predicted direction, though far from being statistically significant, but the rank order of the three group means was not exactly as predicted. The table below presents a summary of the findings by rank ordering of group means. The difference between group VV and group VN shows that the more recent behavior was important, i.e., the be- havior inside the prison. These two groups were similar in their level of initial aggression, as evidenced in their con- victions for violent crimes. What made the difference was the more recent restraint in regard to prison discipline, shown by the VN group but not the VV group. We can assume that the two groups were habitually violent, and this 48 Table 27.--Summary of group differences by rank ordering of group means. (Rank 1 is highest) Score 1 2 3 Significance FQ Aggression VN VV NV No ETAT Aggression VN NV VV No FQ Sex VV VN NV No ETAT Sex VN NV VV No Sex, story W (corrected) VN VV NV .040 Total ETAT words VV NV VN No Time in prison VV VN NV .037 I.Q. NV VN VV No Educational level NV VV VN .094 violence became a part of their criminal life style. When the VN Ss restrained themselves from actual breaches of prison rules, they "compensated" by expressing more aggres- sion in fantasy. The Ss in group NV expressed no criminal aggression and it did not become a part of their life style. For them there was not that additional restraint of not breaking the prison rules. Their aggressive fantasies may be a reflection of their long standing "deprivation" in the area of aggres- sion. In terms of the model presented by Lazarus (1961), we can say that the aggression expressed in the fantasies of 49 groups VN and NV was not "direct," since it was not accom- panied by actual aggression; it was rather "substitutive." The differences between the three groups on the ETAT Aggression scores seem to be in direct contradiction to the conclusion drawn by Buss (1961), in his review of the liter- ature. As quoted earlier, he claimed a direct relationship between TAT aggression and violent behavior. They are also in contradiction to more recent re- views, such as the one by Megargee and Cook (1967), who wrote (p. 48): "Empirical studies have generally found either no significant relationships or significant direct ones. The notion of significant inverse relations between projective test measures and overt aggression has received little em- pirical support. . . . “ It should be remembered that our own results were not significant, but the trend in them was contrary to what had been expected, on the basis of the literature. It may be that the explanation for this trend lies in the interaction between the two instruments used, the FQ and and ETAT, and the generalization offered by Lazarus (1961) regarding the difference between "secondary" and "Primitive" fantasies. These will be discussed in a later section. It is interesting to note that the largest difference between the groups was found on the Sex score to picture W, that was intended as an "aggressive" picture. 50 The Relationship Between Sexual and Aggressive Fantasies We found a positive correlation, and also a "cross- over" in terms of aggressive imagery in response to sexual pictures and vice versa. Moreover, this positive correla- tion did not exist for Sex and Achievement scores or Ag- gression and Achievement scores. Thus, our results cannot be explained away by the big-D (Hull, 1943) approach. If we assume a general expres- siveness rather than specific drives, we should expect a correlation with Achievement scores. It may be correct to assume a bond between sex and aggression, as more "primi- tive" drives, as opposed to the more "socialized" achieve- ment motivation. No arousal procedure has been used in this study except the ETAT pictures and the FQ items them- selves. The Relationship Between the Two Instruments On the FQ we presented the S with ready made themes, while on the ETAT he is asked to produce his own stories, for which he is responsible. There is a clear difference in the fantasy level: the ETAT is less structured, and, compared to the FQ, closer to the internal "primary" fantasy. Going back to the differentiation between primary and secondary process (Fenichel, 1945) and to the differences in their functions, we can hypothesize that the primary process fantasy functions as a substitute, while a secondary process 51 fantasy is a direct expression of actual behavior (Lazarus, 1961). This hypothesis can explain the differences between the ETAT and the FQ in this study. While the more violent groups, VV and VN were higher on Aggression on FQ, the "secondary" measure, the non-violent group NV, was higher on Aggression on the ETAT, compared to the other two groups combined. There were differences among the three groups in the length of time spent in prison. The mean time for group VV was the highest. We can assume that the length of time Spent in prison is a frustration factor. It was significantly cor- related with the Self-Destruction and Escape scales of the FQ. This frustration, if it exists, was shown only on the FQ, where VV was second highest on Aggression, but not on the ETAT. In this case the "secondary" measure was more affected by the frustration factors, while the VV Ss probably were more defensive on the ETAT. Going back to the Lazarus (1961) hypothesis, we can say that the FQ was more sensitive to long term tendencies in actual behavior, since the two more violent groups were higher on the Aggression measure thereof. The ETAT was more sensitive to long term tendencies in fantasy, since the non- violent group was higher on Aggression there, compared to the other two groups combined. 52 The Effects of Background Factors I.Q. ratings correlated significantly with a number of ETAT scores and with productivity on the ETAT (number of words), which in turn correlated with other ETAT scores. The NV group was the highest on I.Q. though the dif- ferences were not significant. Educational level (which correlated .364 with I.Q.) also correlated with various ETAT scores, ETAT productivity and two FQ scales. The NV group was highest in educational level and the differences were significant at the .093 level. The use of corrected ETAT scores eliminated the ef- fects of I.Q., but there were still some significant nega- tive correlations with educational level. There was a -.350 correlation between the expression of Sex on the ETAT and educational level. Since this cor- relation is with the corrected ETAT Sex scores, we can assume that it is "real," i.e. not due to other factors. This may mean that the better educated Ss learned, among other things, to control the expression of their sexual fantasies (and behavior?). The Nature of the Population A discussion of our results cannot be complete with- out a comment on the special characteristics of this popula- tion. 53 We dealt with a group that by any popular standards could be described as "hardened criminals"--convicted felons, from a lower class background. The majority of psychological studies on human behavior and fantasy have used college students as subjects, and our whole conception of the human mind is based on them. This study tried to move away from the world of the college sophomore and the white rat to the harsh reality of prison existence and misspent lives. This transition involved considerable doubts and dif- ficulties. Prisoners are supposedly less "psychologically minded" than better educated people, but the Ss in the study showed, in the way in which they responded to the instru- ments, that these can be used with meaningful results. For more successful studies in terms of clear group differences, it is clear that further refinement, and prob- ably lengthening, of the instruments is needed. SUMMARY Two instruments, a fantasy questionnaire (FQ) and an experimental TAT booklet (ETAT) were administered to sixty- five SPSM inmates, divided into three groups on the basis of their past history of violent crimes and their more recent history of discipline problems inside the prison. It was hypothesized that both past history of vio- lence and recent disciplinary record would be related to expressions of aggressive fantasies. It was also hypothe- sized that sexual and aggressive fantasies will be positively related. The main hypothesis was that restraint in prison be- havior in a group of habitually violent inmates will be re- flected in more aggressive fantasies in this group, compared to other violent inmates who did not show such restraint. The data regarding this hypothesis were in the pre- dicted direction but the differences were not statistically significant. The only hypothesis that was unequivocally supported was the one predicting a positive correlation between ag- gressive and sexual fantasies. The findings were discussed in light of the research literature and the nature of the two instruments. 54 55 The main points of the discussion were as follows: 1. Behavior inside the prison, in terms of disci- pline, is more related to fantasy content than habitual criminal violence. The differences in the nature of the task between the FQ and the ETAT were reflected in the find- ings. The FQ, which calls for admission to "ready made" fantasies, was directly related to habitual behavior. The ETAT, which calls for the production of original fantasies, was inversely related to recent behavior. The correlation between aggressive and sexual fantasies, found before in various studies, seems to exist for this particular group. Instruments for the assessment of fantasy at various levels can be used with this population, though further refinement is called for. REFERENCES REFERENCES Barclay, A.M. The Effect of Hostility on Physiological and Fantasy Responses. Unpublished Ph.D. dissertation. University of Minnesota, 1967. Barclay, A.M. Manual for Scoring Sexual and Aggressive Imagery, Defensiveness, and Distortion on the TAT. Second revision. East Lansing, Mich.: Michigan State University (mimeo), 1967b. Barclay, A.M., & Haber, R.N. The relation of aggressive to sexual motivation. J. Personality, 1965, 33, 462- 475. Buss, A. The psychology of aggression. New York: Wiley, 1961. Cazavelan, J., & Epstein, S. Daydreams of female paranoid schizophrenics. J. of Clinical Psychol., 1966, 27- 32. Coleman, J.C. Stimulus factors in the relation between fan- tasy and behavior. J. proj. Tech., 1967, 31, 68-73. Clark, R.A. The effects of sexual motivation on fantasy. i. Fenichel, O. The Psychoanalytic Theory of Neurosis. New York: Norton, 1945. Freud, S. Complete Psngological Works. London: Hogarth Press, Vol. 18, Ch. 3, 1955. Freud, S. A general introduction to psychoanalysis. New York: Morton, 1949. Freud, S. Creative writers and daydreaming. In J. Strachey (ed.), The Standard edition of the complete psycho- logical works of Sigmund Freud. London: Hogarth, 1962, Vol. IX. Hartmann, H. Ego psychology and‘the‘problem of adaptation. New York: Inter. Univ. Press, I958. 56 57 Hull, C.L. Pringiples of Behavior. New York: Appleton- Century-Crofts, 1943. James, P.B., & Mosher, D.L. Thematic aggression, hostility, guilt and aggressive behavior. J. proj. Tech., 1967, El, 61-67. Jensen, A.R. Aggression in fantasy and overt behavior. Psychol. Monog£., 1957, 2;, No. 445. Kagan, J., & Lesser, G.L. (ed.). Contemporapy Issues in THematic Apperceptive Methods. SpringfieId, IIl.: Thomas, 1961. Kalin, R., McClelland, D.C., & Kahn, M. The effects of male social drinking on fantasy. J. pers. soc. Psychol., 1965, I, 441-452. Lazarus, R.S. A substitutive-defensive conception of apper- ceptive fantasy. In J. Kagan & Lesser, G.L. (ed.) Contemporary Issuesiin Thematic Apperceptive Methods. Springfield, 111.: Thomas, 1961. Lazarus, Richard S. Story telling and the measurement of motivation: The direct versus substitutive contro- versy. J. Consult. Psychol., 1966, 32, 483-487. Lindzey, G. Thematic apperception test: Interpretive as- sumptions and related empirical evidence. Psychol. Bull., 1952, 32, 1-25. MacLean, P.D. New findings relevant to the evolution of psychosexual functions of the brain. In J. Money (ED.), Sex Research New Developments. New York: Holt, 1965. McClelland, David C. Longitudinal trends in the relation of thought to action. J. consult. Psychol., 1966, 29, 479-483. McClelland, D.C., Atkinson, J.W., Clark, R.A., & Lowell, E. L. The achievement motive. New York: Appleton- Century, 1953. McClelland, D.C. et al. A scoring manual for the achievement motive. In J.W. Atkinson (Ed.) Motives in Fantasy, Action, and Sociepy. New York: Van Nostrand, 1958. Megargee, E.I., & Cook, P.E. The relation of TAT and inkblot aggression content scales with each other and with criteria of overt aggressiveness in juvenile delin- quents. J1_proj, Tech., 1967, HI, 48-60. 58 Meyer, R.G. The Relationship of Blood Pressure Levels to pHe Chronic Inhibitibn ofiAggression. Unpublished Ph.D. dissertation. Michigan State University, 1967. Murray, H.A. Explorations in Personality. New York: Sci- ence Editions, 1967. Murstein, B.I. Theory and Research in Prgjective Techniques. New York: Wiley, 1963. Mussen, P.H., & Naylor, H.K. The relationship between overt and fantasy aggression. J. abnorm. & soc. Psychol., 1954, 22, 235-240. Purcell, K. The Thematic Apperception Test and antisocial behavior. J. consult. Psychol., 1956, 32, 449-456. Rorschach, H. Psychodiagnostik. Bern: Huber, 1921. Schaefer, J.B., & Norman, M. Punishment and aggression in fantasy responses of boys with antisocial character traits. J. pers. soc. Psychol., 1967, p, 237-240. Singer, J.L. Daydreaming: An Introduction to the Experi- mental Study of’Inner Experience. New York: Random House, 1966. Singer, J.L., & Antrobus, J. Tpe Imaginal Processes Inven- tory. New York: City College, 1966. Singer, J.L., & McCraven, V. Some characteristics of adult daydreaming. J. of Psychol., 1961, El, 151-164. Singer, J.L., & McCraven, V. Patterns of daydreaming in American subcultural groups. Int. J. Soc. Psychiat., 1962, H, 272-282. Singer, J.L., & Schonbar, R. Correlates of daydreaming: a dimension of self-awareness. J. Consult, Psychol., 1961, 3Q! 1-6. Silber, D.E., & Courtless, T.F. Measures of fantasy aggres- sion among mentally retarded offenders. Am. J. mental. Deficien., 1968, la, 918-923. Stone, H. The TAT aggressive content scale. J.Aproj: Tech., 1956, 32, 445-452. Wagman, M. Daydreaming frequency and some personality meas- ures. J. Consult. Psychol., 1965, £2, 395. 59 Wagman, M. University achievement and daydreaming behavior. J. consult. Psychol., 1968, $2, 196-198. Weissman, S.L. Some indicators of acting out behavior from the Thematic Apperception Test. J. proj. Tech., 1964, 3Q, 366-375. White, R.W. Motivation reconsidered: the concept of compe- tence. Psychol. Rev., 1959, pp, 297-333. White, R.W. Ego and reality in psychoanalytic theory. Psychological Issues, 1964, 2, Monograph 11. Winer, B.J. Statistigal Principles in Experimental Desigp. New York: McGraw Hill, 1962: APPENDICES Ml-—v APPENDICES APPENDIX I THE FANTASY QUESTIONNAIRE (FQ) PSYCHOLOGICAL STUDY QUESTIONNAIRE Name Number Part II. Each question has 5 possible answers. For each question, choose the answer which is most true or appropriate for you. Each answer corresponds to one of the numbers 1 through 5. Locate the number of each question on the answer sheet. Then fill in between the lines under the number that indicates your answer. PROCEED WITH PART II I daydream l. infrequently. 2. once a week.l 3. once a day. 4. a few times during the day. 5. many different times during the day. 2. Daydreams or fantasies make up 1. no part of my waking thoughts. 2. less then 10%.of my waking thoughts. 3. at least 10%,of my waking thoughts. 4. at least 25% of my waking thoughts. 5. at least 50% of my waking thoughts. 3. I have a night dream 1. rarely or never. 2. once a month. 3. several times a month. 4. several times a week. 5. once a night. 4. I can recall a dream 1. rarely or never. 2. once a month. 3. several times a month. 4. several times a week. 5. once a night. 5. daydream at work I 1. infrequently. 2. once a week. 3. once a day. 4. a few times during the day. 5. many different times during the day. 10. 11. 12. I lose myself in daydreaming l. infrequently.‘ 2. once a week. 3. once a day. 4. a few times during the day. 5. many different times during the day. Whenever I have time on my hands I daydream . never. . rarely. . sometimes. . frequently. . always. U'IowaH I recall my dreams vividly l. rarely or never. 2. once a month. 3. several times a month. 4. several times a week. 5. once a night. When I am at a meeting or show that is not very interesting, daydream rather than pay attention . never. rarely. sometimes. . frequently. . always. I l 2 3 4 5 HOW well adjusted to prison life do you consider yourself: 1) poorly, 2) less than average, 3) average, 4) better than average, 5) excellent. Are you satisfied with your present job? 1) Not at all. 2) A little. 3) O.K. 4) Really like it. 5) Love it. How is your supervisor satisfied with your performance on your present job? 1) Not at all. 2) A little. 3) O.K. 4) Satisfied. 5) Very satisfied. PART IV Each item £2 Part Iy_describes g daydream pp fantasy*which may be similar, more or less, to one which you have had. Answer each item on the basis of how often you have had the fantasy--or one similar Use the following key as an approximate guide to indicate how frequently you have had a given daydream. to it. Answer Answer Answer Answer .Answer "- [w E. on the answer sheet if you have never had the daydream. if you remember having the daydream only once or twice in your lifetime. if you remember having the fantasy several times in the past few years, but not in the past 6 months. if you have had the daydream at least once in the past 6 months, but not as often as once a week. on the answer sheet if you have the daydream frequently-- as often as once a week. When you have chosen your response to an item, locate on the IBM sheet the number which corresponds to the item. Then fill in between the lines under the appropriate number that indicates your response to the item, 6.9., I, 2, 1, g, or g, CONTINUE ON THE SAME ANSWER SHEET INSTRUCTIONS FOR PART IV WILL BE REPEATED BEFORE EVERY RAGE OF ITEMS. THIS WILL MAKE YOUR WORK EASIER. PART IV Each item ip Part I! describes §_daydream pp fantasy which may be similar, more or less, to one which you have had. Answer each item on the basis of how often you have had the fantasy-~or one similar Use the following key as an approximate guide to indicate how frequently you have had a given daydream. to it. Answer Answer Answer Answer Answer "- Iw .5. on the answer sheet if you have never had the daydream. if you remember having the daydream only once or twice in your lifetime. if you remember having the fantasy several times in the past few years, but not in the past 6 months. if you have had the daydream at least once in the past 6 months, but not as often as once a week. on the answer sheet if you have the daydream frequently-- as often as once a week. When you have chosen your response to an item, locate on the IBM sheet the number which corresponds to the item. Then fill in between the lines under the appropriate number that indicates your response to the item, e.g., I, g, s, g, orIg. CONTINUE ON THE SAME ANSWER SHEET INSTRUCTIONS FOR PART IV WILL BE REPEATED BEFORE EVERY RAGE OF ITEMS. THIS WILL MAKE YOUR WORK EASIER. 13. I imagine meeting a woman who will give me true happiness. 14. I imagine that people notice me as I enter a room and think I'm awfully good-looking. 15. I imagine having an exciting experience with an attractive woman who compliments me on my performance. 16. I anticipate the coming of another World War and picture the consequences of an atomic bombing of my hometown. 17. I picture myself eating foods in the finest restaurants. 18. I think of what Heaven or Life Eternal might actually be like. 19. I see myself getting revenge in a clever way on a teacher or supervisor who has criticized me. 20. I imagine being known as the best sexual performer in the world, admired by everyone for my potency. 21. I see myself in Heaven and see myself transformed. 22. I finagine the death of the judge that sentenced me to prison. 23. I see myself eating and drinking at a great banquet with Innusual delicacies. 24. I imagine the death of the prosecutor that sent me to prison. 25. I picture my own funeral. 26. I think about how my digestive system is working. 27. I see myself married to a woman who is seriously ill and who needs constant loving attention from me. 28. I visualize sexual intercourse using a great variety of positions and forms of satisfaction. PART IV Each item ip Part IX describes g daydream pp fantasy which may be similar, more or less, to one which you have had. Answer each item on the basis of how often you have had the fantasy-- or one similar to it. Use the following key as an approximate guide to indicate how frequently you have had a given daydream. Answer I. on the answer sheet if you have never had the daydream. Answer 3 if you remember having the daydream only once or twice in your lifetime. Answer §_ if you remember having the fantasy several times in the past few years, but not in the past 6 months. Answer 3_ if you have had the daydream at least once in the past 6 months, but not as often as once a week. Answer g. on the answer sheet if you have the daydream frequently-- as often as once a week. When you have chosen your response to an item, locate on the IBM sheet the number which corresponds to the item. Then fill in between the lines under the appro- priate number that indicates your response to the item, 6.9., ._1_: 2.! 2! it or _5_- CONTINUE ON THE SAME ANSWER SHEET 29. 30. 31. 32. 33. 34. 35. 36. 37. .38. 39. 40. 41. 42. I imagine myself having an inexhaustable supply of my favorite foods. I imagine humiliating the woman with whom I am having sex relations. I imagine being released due to a new law passed by Congress. I imagine a miraculous escape from prison. I imagine being pardoned by the governor. As a child, I pictured myself as crippled or disfigured in some way. I see myself as losing my mind and being placed in a mental hospital. I imagine what would happen if a plane crashed on the prison. I visualize the End of the World. I see myself at the side of a dying loved one soothing and comforting him or her. I imagine living in Ancient Rome and participating with complete abandon in a wild sex party. I imagine reading in black headlines that New YOrk has just been attacked by enemy missiles. I imagine vactioning in a resort like Hawaii where I have a very lavish suite at the Royal Hawaiian Hotel. I imagine that I have secretly had sexual relations with the Spouses or sweethearts of all of my friends. 29. 30. 31. 32. 33. 34. 35. 36. 37. .38. 39. 40. 41. 42. I imagine myself having an inexhaustable supply of my favorite foods. I imagine humiliating the woman with whom I am having sex relations. I imagine being released due to a new law passed by Congress. I imagine a miraculous escape from prison. I imagine being pardoned by the governor. As a child, I pictured myself as crippled or disfigured in some way. I see myself as losing my mind and being placed in a mental hospital. I imagine what would happen if a plane crashed on the prison. I visualize the End of the World. I see myself at the side of a dying loved one soothing and comforting him or her. I imagine living in Ancient Rome and participating with complete abandon in a wild sex party. I imagine reading in black headlines that New YOrk has just been attacked by enemy missiles. I imagine vactioning in a resort like Hawaii where I have a very lavish suite at the Royal Hawaiian Hotel. I imagine that I have secretly had sexual relations with the Spouses or sweethearts of all of my friends. PART IV Each item ip Part I! describes g daydream pp fantasy which may be similar, more or less, to one which you have had. Answer each item on the basis of how often you have had the fantasy-- or one similar to it. Use the following key as an approximate guide to indicate how frequently you have had a given daydream. Answer‘I Answer Answer Answer Answer 3 on the answer sheet if you have never had the daydream. if you remember having the daydream only once or twice in your lifetime. if you remember having the fantasy several times in the past few years, but not in the past 6 months. if you have had the daydream at least once in the past 6 months, but not as often as once a week. on the answer sheet if you have the daydream frequently-- as often as once a week. When you have chosen your response to an item, locate on the IBM sheet the number which corresponds to the item. Then fill in between the lines under the appro- priate number that indicates your response to the item, 6.9.. ,1_. Z: _3_. g. or _5_. CONTINUE ON THE SAME ANSWER SHEET 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. I imagine inventing fantastic forms of torture, to humiliate my sexual partner. I imagine owning a considerable amount of clothing some of which is made to order for me in London and Paris. I see myself as being seduced by an older woman of great wealth and influence who can protect me or further my career. I picture the reaction of my friends as I commit suicide. I imagine that I have enough money to insure security for myself and an inheritance for my children. I imagine the death of the lawyer that defended me in my trial. I see myself being judged for punishment in the after life. I picture what it would be like to inherit one million dollars from a long-lost relative. I imagine humiliating my sexual partner. I imagine that a screen or TV producer sees me and asks me if I would be interested in a screen test. I see myself committing the perfect crime: killing all my enemies on the outside without getting caught. I imagine myself being acquitted by the Supreme Court of the United States. I imagine different ways of making an initial advance to a woman. I see myself committing the perfect crime: breaking into a laank and becoming a millionaire, then living in wealth for the rest of my life. PART IV Each item ip Part I! describes g daydream p£_fanta y which may be similar, more or less, to one which you have had. Answer each item on the basis of how often you have had the fantasy-- or one similar to it. Use the following key as an approximate guide to indicate how frequently you have had a given daydream. Answer Answer Answer Answer Answer .1. IN [w lh lm on the answer sheet if you have never had the daydream. if you remember having the daydream only once or twice in your lifetime. if you remember having the fantasy several times in the past few years, but not in the past 6 months. if you have had the daydream at least once in the past 6 months, but not as often as once a week. on the answer sheet if you have the daydream frequently-- as often as once a week. When you have chosen your response to an item, locate on the IBM sheet the number which corresponds to the item. Then fill in between the lines under the appro— priate number that indicates your response to the item, 6.9.. _1_. 3. g. 4. or _5_. CONTINUE ON THE SAME ANSWER SHEET 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. I imagine myself being able to fly over the walls of the prison. I imagine torturing the woman with whom I am having sex rela- tions. I imagine a revolution and general release of all prisoners. I imagine how life would be if I had had different parents. I picture myself as a very different and more successful person than I am. As a child, I imagined myself as a great saint of the Church or Biblical hero. I sometimes imagine myself beating another person with my fists. I imagine myself involved in sexual relations that are too embarrassing to admit to even on a questionnaire. I imagine myself driving a high-powered race car. I imagine beating up the woman with whom I have sexual rela- tions. I see myself having a relationship with a married woman. I see myself "telling off" my parents. I sometimes picture myself as having all my needs satisfied. I like to imagine myself lying on a quiet beach in a warm afternoon sun. I sometimes imagine myself having a sexual relationship with a teenage girl. I daydream that someone I know dies. PART IV Each item ip Part Iy_describes §_daydream pp fantasy which may be similar, more or less, to one which you have had. Answer each item on the basis of how often you have had the fantasy-- or one similar to it. Use the following key as an approximate guide to indicate how frequently you have had a given daydream. Answer Answer Answer Answer Answer .1. IN lw lb lm on the answer sheet if you have never had the daydream. if you remember having the daydream only once or twice in your lifetime. if you remember having the fantasy several times in the past few years, but not in the past 6 months. if you have had the daydream at least once in the past 6 months, but not as often as once a week. on the answer sheet if you have the daydream frequently-- as often as once a week. When you have chosen your response to an item, locate on the IBM sheet the number which corresponds to the item. Then fill in between the lines under the appro- priate number that indicates your response to the item, 9.9-: ll 3! _3_I £1 or _5_0 CONTINUE ON THE SAME ANSWER SHEET 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. I imagine myself undressing a woman. I imagine a general prisoner rebellion in all prisons. I imagine the second coming of Jesus. I sometimes imagine myself lying on my back in a large field of grass and just gazing into the blue sky. I see someone I know being injured or killed in an accident. I picture myself taking my next important step in life. I see my parents upset or crying because of something displeas- ing I have done. I imagine myself living the life of a cow, gently eating grass in a quiet pasture. I see myself in a daydream failing miserably to satisfy my sex partner. I picture myself getting back at someone I dislike. APPENDIX II FQ SCALES FANTASY QUESTIONNAIRE (FQ) SCALES £0312 Daydreaming Nightdreaming Adjustment Sex Satisfaction Self Destruction Aggression Food Security Religion Unreal Bizzare Nurturance Sadism Escape Release Achievement Total ates: 1,2,5,6,7,9 3,4,8 10,11,12 15,20,28,39,42,45,55,64,67,71,73 13,14,4l,44,45,47,52,6l,65,69,70,76, 78 25,34,35,46,49,81 16,36,37,40,59,74,l9,22,24,48,53,63, 68,72,77,79,82 l7,23,26,29,80 13,18,41,45,47,69,76,80 18,21,37,62,75,49 50,54,56,57,60 26,35,80 27,38 30,43,51,58,66 32,57,59,74 31,33,54 44,47,50,52,53,56,6l l3-82 72 APPENDIX III THE EXPERIMENTAL TAT (ETAT) BOOKLET NAME (Please Print) NUMBER Instructions for Storytelling On the following pages you are to write out some brief stories that you make up on your own. There are no right or wrong answers. This is designed to give you an opportunity to be as fanciful as you wish, to imagine a situation quickly and write out a brief story about it. In order to help you get started, there are a series of pictures that you can look at and build your stories around. When you have finished reading these instructions, you should look at the first picture briefly, then turn the page again and write a story suggested by the picture. To help you cover all the elements of a story plot in the time allowed, you will find four questions spaced out over the page. They are: 1. What is happening: Who are the people? . 2. What has led up to this situation? That is, what has happened in the past? 3. What is being thought? What is wanted? By whom? 4. What will happen? What will be done? Ybur overall time for each story is about five minutes. There is no strict time limit, but don't write much over five minutes. I'll annOUnce when five,minutes are up. iv" 3141-- ' . ' 'g- E‘Fm’a shaman-amp». Kuhn»... an”... Just look at the picture briefly (10-15 seconds), turn the page and write out the story it suggests. Spend about 5 minutes on this story. 1. What is happening? Who are the peOple? 2. What has led up to this situation? That is, what has happened in the past? 3. What is being thought? What is wanted? By whom? 4. What will happen? What will be done? Just look at the picture briefly (lo-l5 seconds), turn the page and write out the story it suggests. Spend about 5 minutes on this story. \ 1. What is happening? Who are the people? 2. What has led up to this situation? That is, what has happened in the past? 3. What is being thought? What is wanted? By whom? 4. What will happen? What will be done? Just look at the picture briefly (lo-15 seconds), turn the page and write out the story it suggests. Spend about 5 minutes on this story. 1. What is happening? Who are the people? 2. What has led up to this situation? That is, what has happened in the past? 3. What is being thought? What is wanted? By whom? 4. What will happen? What will be done? Just look at the picture briefly (lo-15 seconds), turn the page and write out the story it suggests. Spend about 5 minutes on this story. 1. What is happening? Who are the people? 2. What has led up to this situation? That is, what has happened in the past? 3. What is being thought? What is wanted? By whom? 4. What will happen? What will be done? Just look at the picture briefly (lo-15 seconds), turn the page and write out the story it suggests. Spend about 5 minutes on this story. 1. What is happening? Who are the people? 2. What has led up to this situation? That is, what has happened in the past? 3. What is being thought? What is wanted? By whom? 4. What will happen? What will be done? APPENDIX IV LETTER TO ALL SS APPENDIX IV LETTER TO ALL 83 Dear Mr. YOu have been selected to participate in a psychological study of SPSM inmates conducted by me. Ybu will be put on call for the actual study, which will take place in one of the classrooms in the Academic School. It will take about an hour, and for your cooperation you will receive two packs of cigarettes. This study is done by me independently, and the information is confidential. No one inside the prison or the dept. of corrections will have access to the data. A similar study was done by me during the summer, and those who participated know that everything was kept confidential. Your cooperation is very important. I hope to see you in the study group. YOurs, Benny Beit-Hallahmi APPENDIX V INSTRUCTIONS TO SS Instructions to Subjects General we are going to ask you to take part in a research project. The purpose of this particular project is to find out something about the psychological world in which prisoners live. We hope that the information we gain from this study will be helpful in making treatment of prisoners more scientific and more humane. In this way we will decrease the number of prdblems that a man faces while he is in prison as well as increase the effectiveness of rehabilitation efforts. Another purpose of this study, which is certainly important, is helping the main investigator to get the Ph.D. degree in psychology. This is the benefit that I will get from it. The benefit that you will gain from it is: (1) Every participant in the study will receive two packs of cigarettes as a token of appreciation for his efforts. (2) We hope that you will enjoy taking part in this study for your own personal benefit, even without getting the cigarettes, but obviously we are sure that you will enjoy this payment. The reason We ask you to write your name is because we may be interested in asking some of you to take part in another stage of this project. we need your names so that we can get in touch with you. Instructions to Subjects General We are going to ask you to take part in a research project. The purpose of this particular project is to find out something about the psychological world in which prisoners live. We hope that the information we gain from this study will be helpful in making treatment of prisoners more scientific and more humane. In this way we will decrease the number of problems that a man faces while he is in prison as well as increase the effectiveness of rehabilitation efforts. Another purpose of this study, which is certainly important, is helping the main investigator to get the Ph.D. degree in psychology. This is the benefit that I will get from it. The benefit that you will gain from it is: (1) Every participant in the study will receive two packs of cigarettes as a token of appreciation for his efforts. (2) We hope that you will enjoy taking part in this study for your own personal benefit, even without getting the cigarettes, but obviously we are sure that you will enjoy this payment. The reason We ask you to write your name is because we may be interested in asking some of you to take part in another stage of this project. We need your names so that we can get in touch with you. HQ Instructions The questionnaire has two parts, and each part has to be answered in a different way. Ybu are going to answer part I by writing on the questionnaire itself, but the other part is going to be answered by marking with a pencil on an IBM answer sheet. Together with the questionnaire you will get an IBM.answer sheet and a pencil. You are not supposed to write your answers on the questionnaire but on the answer sheets and we are going to help you with making sure you put your reply in the right place, in the right manner. In case of any problems just raise your hand. In the second part of the questionnaire you are supposed to make a choice between five possibilities. In this part most of the questions deal with your psychological habits more or less; with your thinking processes or with your personal preferences. And we have for each question a scale from one to five that describes the frequency for each one of you. For example, let's take the question "I go to the dentist." The options are 1 never, 2 rarely, 3 sometimes, 4 often, 5 very often. Now the choice is clear. Ybu have to determine for yourself what is the right frequency for you. Now you have to do the same thing for each one of the questions in this part. We are going to ask you in these questionnaires questions that are sometimes very personal and we expect you to be as frank as possible. We are interested in frank, honest answers. This is the reason that we ask only volunteers to take part in this study. Now, your replies to these questionnaires in no way affect your future in the institution or in the outside. Your answers will be seen only by me. In all research projects information is completely confidential. Some of the questions on the questionnaire may look like repetitions of other questions, but they are not. Please follow the instructions carefully and answer all questions. Please raise your hand if you have any questions. ETAT Instructions. In this booklet you will find five pictures, and you will be asked to write out a short story, in response to each one of them. I will read the instructions as they appear on the first page of the booklet. Instructions for Stopytelling On the following pages you are to write out some brief stories that you make up on your own. There are no right or wrong answers. This is designed to give you an oppor- tunity to be as fanciful as you wish, to imagine a sit- uation quickly and write out a brief story about it. In order to help you get started, there are a series of pictures that you can look at and build your stories around. When you have finished reading these instructions, you should look at the first picture briefly, then turn the page again and write a story suggested by the picture. To help you cover all the elements of a story plot in the time allowed, you will find four questions spaced out over the page. They are: 1. What is happening: Who are the people? 2. What has led up to this situation? That is, what has happened in the past? 3. What is being thought? What is wanted? By whom? 4. What will happen? What will be done? Your overall time for each story is about five minutes. There is no strict time limit, but don't write much over five minutes. These instructions are quite clear and you shouldn't have any problems. If you have any questions, please raise your hand. ETAT Instructions. In this booklet you will find five pictures, and you will be asked to write out a short story, in response to each one of them. I will read the instructions as they appear on the first page of the booklet. Instructions for Storytelling On the following pages you are to write out some brief stories that you make up on your own. There are no right or wrong answers. This is designed to give you an oppor- tunity to be as fanciful as you wish, to imagine a sit- uation quickly and write out a brief story about it. In order to help you get started, there are a series of pictures that you can look at and build your stories around. When you have finished reading these instructions, you should look at the first picture briefly, then turn the page again and write a story suggested by the picture. To help you cover all the elements of a story plot in the time allowed, you will find four questions spaced out over the page. They are: 1. What is happening: Who are the people? 2. What has led up to this situation? That is, what has happened in the past? 3. What is being thought? What is wanted? By whom? 4. What will happen? What will be done? Your overall time for each story is about five minutes. There is no strict time limit, but don't write much over five minutes. These instructions are quite clear and you shouldn't have any problems. If you have any questions, please raise your hand. Instructions to Subjects General We are going to ask you to take part in a research project. The purpose of this particular project is to find out something about the psychological world in which prisoners live. We hope that the information we gain from this study will be helpful in making treatment of prisoners more scientific and more humane. In this way we will decrease the number of problems that a man faces while he is in prison as well as increase the effectiveness of rehabilitation efforts. Another purpose of this study, which is certainly important, is helping the main investigator to get the Ph.D. degree in psychology. This is the benefit that I will get from it. The benefit that you will gain from it is: (1) Every participant in the study will receive two packs of cigarettes as a token of appreciation for his efforts. (2) We hope that you will enjoy taking part in this study for your own personal benefit, even without getting the cigarettes, but obviously we are sure that you will enjoy this payment. The reason we ask you to write your name is because we may be interested in asking some of you to take part in another stage of this project. We need your names so that we can get in touch with you. . . 4 ‘l . , n l. I . . - u I. .. k .; ... . 0. . \ n . n. \.I ,. .. . Q. I .5. . . I. . §Q_Instructions The questionnaire has two parts, and each part has to be answered in a different way. YOu are going to answer part I by writing on the questionnaire itself, but the other part is going to be answered by marking with a pencil on an IBM answer sheet. Together with the questionnaire you will get an IBM answer sheet and a pencil. You are not supposed to write your answers on the questionnaire but on the answer sheets and we are going to help you with making sure you put your reply in the right place, in the right manner. In case of any prdblems just raise your hand. In the second part of the questionnaire you are supposed to make a choice between five possibilities. In this part most of the questions deal with your psychological habits more or less; with your thinking processes or with your personal preferences. And we have for each question a scale from one to five that describes the frequency for each one of you. For example, let's take the question "I go to the dentist." The options are 1 never, 2 rarely, 3 sometimes, 4 often, 5 very often. Now the choice is clear. Ybu have to determine for yourself what is the right frequency for you. Now you have to do the same thing for each one of the questions in this part. We are going to ask you in these questionnaires questions that are sometimes very personal and we expect you to be as frank as possible. We are interested in frank, honest answers. This is the reason that we ask only volunteers to take part in this study. Now, your replies to these questionnaires in no way affect your future in the institution or in the outside. Your answers will be seen only by me. In all research projects information is completely confidential. Some of the questions on the questionnaire may look like repetitions of other questions, but they are not. Please follow the instructions carefully and answer all questions. Please raise your hand if you have any questions. ETAT Instructions. In this booklet you will find five pictures, and you will be asked to write out a short story, in response to each one of them. I will read the instructions as they appear on the first page of the booklet. Instructions for Stopytelling On the following pages you are to write out some brief stories that you make up on your own. There are no right or wrong answers. This is designed to give you an oppor- tunity to be as fanciful as you wish, to imagine a sit- uation quickly and write out a brief story about it. In order to help you get started, there are a series of pictures that you can look at and build your stories around. When you have finished reading these instructions, you should look at the first picture briefly, then turn the page again and write a story suggested by the picture. To help you cover all the elements of a story plot in the time allowed, you will find four questions spaced out over the page. They are: 1. What is happening: Who are the people? 2. What has led up to this situation? That is, what has happened in the past? 3. What is being thought? What is wanted? By whom? 4. What will happen? What will be done? YOur overall time for each story is about five minutes. There is no strict time limit, but don't write much over five minutes. These instructions are quite clear and you shouldn't have any problems. If you have any questions, please raise your hand. APPENDIX VI PERSONAL QUESTIONNAIRE TO SS SPSM PSYCHOLOGICAL STUDY Questionnaire. Part I. Name Number How long have you been in prison this time? (in months) How long have you been in prison including previous times? Are you an only child? (Circle one) YES/NO Are you the oldest child in your family? (Circle one) YES/NO Gr APPENDIX VII SIGNIFICANT CORRELATIONS BETWEEN SEX AND AGGRESSION SCORES ON THE ETAT 90 .Hm>oa mo. one as unmowmwcmflm mum mcowumHouHoo Has Sm. 3m. H33. Hmm. m can 0 monsuoam omm. mov. 3 one a on: Us mum. m u .m ohm. «av. 5mm. 3 musuofim mmm. mvm. « ououoam mouoom Hmuoe 3 one 4 O onsuoflm 3 ousuoflm 4 ousuoflm xom monoom cowmmonmmd co Aoowoouuoov mouoom coflmmonmmo «948m may one xmm consumn mGOADMHoHuoo unmowwflcmwmtl.mm wands APPENDIX VIII ETAT SCORING MANUAL Reproduced from Manual for Scoring Sexual and Aggressive Imagery, Defensiveness, and Distortion on the TAT Developed and Compiled by Andrew M. Barclay Michigan State University Second Revision December 1967 SCORING MANUAL Sexual Imagery Give a score of: l to no sexual reference: boy and girl working as an assignment to the following references: gone out a few times studying together in the library good friends good time together "hawking" admired fond of going together keeping an eye on desires girl-friend boy-friend study date couple dependent run around with physical contact husband wife blind date going out "bird dogging" to the following references: win the love of necking petting simple love references serious marriage sit on a lap make a pass date fall in love something physical fiance 91 going parking lovers' lane looking at pictures of nudes going to be married invite up to a room adores reads sex into everything (they are in love) magnificant attraction much too friendly kiss 92 4 to the following references: heavy petting to make sexual demands alone in library or lab proposal of marriage "stepping out on" lustful direct desires such as sexual affairs advanced love "hot date" references "he wants her" 5 to the following references: sleeping together rape abortion Don Juan prostitution mistress seduce lovers debauch strong love references pregnancy nymphomaniac "balling" screw make love to Score each story once, giving the score that is most appro- priate. For example, if the story is loaded with sex, score it as a 5; if it is only moderately so, score it as a 3, etc. Score one category higher for multiple references from the same category (3 or more). When there are multiple references from two different categories, score the higher. 93 Aggressive Imagery Give a score of: 1 to no aggressive references: things have gone wrong 2 to the following references: dismayed sarcastic statements disgust tell to scram does not care for bored with someone heck, darn annoyed tease scorn someone is a "pain: challenge disagree criticize hurt force aloof "creep" 3 to the following references: argue never hear the end of prejudice alimony reprimand arrest dispute angry beat (game) cannot stand rebel obnoxious fire from a job resent detests furious scold laugh someone out indignant curse sick of jealous battle of wits break an engagement 4 to the following references: strong anger (violence) damn hell (in any contest) makes trouble for divorse hit or kick berates hostile despise beats (strikes) hate bitching 94 5 to the following references: murders wars (invasions, battles, revolutions (of any etc.) sort) suicide violence rape battle of strength Give each story one score as for manifest sex. Count wishes the same as the deed itself. For example, "He wishes she were dead" rates a 5, etc. Score one category higher for multiple references from a single category (3 or more). Multiple references from two or more categories are given the higher score. APPENDIX IX GROUP MEANS FOR CORRECTED SEX AND AGGRESSION SCORES ON FOUR ETAT STORIES APPENDIX IX GROUP MEANS FOR CORRECTED SEX AND AGGRESSION SCORES ON FOUR ETAT STORIES TABLE 29.--Group means for sex and aggression scores on four ETAT stories. Groups Scores VV VN NV Sex, Story A .0211 .0219 .0216 Sex, Story W .0200 .0286 .0186 Sex, Story C .0402 .0433 .0412 Sex, Story B .0387 .0396 .0532 Aggression, Story A .0328 .0396 .0333 Aggression, Story W .0608 .0718 .0678 Aggression, Story C .0238 .0199 .0222 Aggression, Story B .0167 .0216 .0259 95 11041an STATE UNIV. LIBRARIES 1|11111111111111111 “I111“111111111111 31293106411261