THE PERCEPTEQ‘E’ OF WGLEé‘éCE-I AS A FUNC?!GN 0F HESTCRE‘SfiL m CURRENT BEHAVEOR Thesis for the Degree of M. A MICHEGA-N STATE 83 NERSETY IH-OMAS F. REtF 196? 1Hh5‘.5 LIB RA K Y Michigan "rate University ABSTRACT THE PERCEPTION 0F VIOLENCE AS A FUNCTION OF HISTORICAL.AND CURRENT BEHAVIOR by Thomas I. Reif Recent studies in stereoscopic perception have indi- cated that both.familiarity with content, and the intent to become involved with the content, will increase a per- son's perceptual sensitivity to that content in the stereo- scope. Thus in different studies, both pe0ple who have hist- ories of involvement with violence, and people who event- ually become involved with violence, have readily perceived violent scenes in the stereoscope. This study was designed to investigate the relationp ship between these two perceptually sensitizing factors-hist- orical involvement with content, and current partaking of the same content. Subjects were institutionalized Juvenile de- linquents. The scenes which.were stereoscopically presented to them were scenes or violence, simultaneously paired with ”neutral" scenes. Six groups of nine delinquents each were divided into varying degrees of historical and current in- volvement with violence. Historically, delinquents were con- sidered either violent aggressive, non-violent aggressive, or non aggressive. Currently, these same delinquents were Thomas F. Re if considered either aggressive or non aggressive. Consequent- ly, each group of delinquents had both a historical and a current behavior status, e.g., historically violent aggres- sive and currently aggressive, historically nonpviolent ag- gressive and currently non aggressive, historically non aggressive and currently aggressive, etc. It was hypothesized that those groups who were either historically familiar with some form of aggression, or who were currently aggressive within the institution, would be perceptually sensitized to violent scenes in the stereo- scope. In addition, it was hypothesized that the more a de- linquent's historical behavior approached physical violence, the greater his perceptual sensitivity to violence would be. Results indicated that when both historical and current behavioral variables are included in the same study, an in- teraotion effect occurs. It was found that those delinquents who have historically been violent or nearly violent, and who still maintain their aggressive behavior within the in- stitution, perceive relatively little violence in the stereo- scope. However, those whose histories have been violent or nearly violent, and who currently are not aggressive within the institution, perceive relatively large numbers of vio- lent scenes in the stereoscope. In addition, those delinquents who have only recently begun to involve themselves in aggres- sive behavior perceive relatively many violent scenes in the Thomas F. Relf stereoscope. In general, the explanation for these results is offerred in terms of two factors which seem to contribute to a delin- quent's sensitivity to, or disinterest in, perceiving vio- lence in the stereoscope. The first factor is the phenomen- ological reality of the individual, i.e., his past patterns of behavior, his current consistencies or inconsistencies with those patterns, and the psychological experience which results. The second factor is intimately tied to the first. It is the degree to which perceiving violence is functionp ally important to the individual. In cases where it is func- tionally important, perceptual sensitivities increase; in cases where there is no need to look for violence, perceptual sensitivities decrease. K“. . _ Approvedq%y”7geeCuglcirtézg;1, o mmittee Chairman Thesis Committee: Dozier Thornton Ehns Toch Charles Eanley THE PERCEPTION OF VIOLENCE AS A FUNCTION OF HISTORICAL AND CURRENT BEHAVIOR BY Thomas F. Reif A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1967 TABLE OF CONTENTS ACKNOWLEDGEMENT.................................. INTRODUCTION..................................... STATEMENT OF THE PROBLEM......................... Hypotheses.................................... METHOD........................................... 1. SubjectS.............4..................... B. Criteria for Selection..................... 0. ApparatuS.................................. D. Administration Procedure................... E. Scoring Procedure.......................... RESULTS.......................................... DISCUSSION....................................... SUMMARY.......................................... BIBLIOGRAPHY..................................... APPENDIX......................................... 111 Page ii 1 6 9 10 10 10 15 16 l7 18 21 31 33 35 ACKNOWLEDGEMENT I would like to thank the members of my committee, chairman, Dozier Thornton, Hans Toch, and Charles Hauley for their helpful suggestions in developing and writing this thesis. The assistance of Bruce Vogel, clinical psychologist of Lansing Boys Training School was invaluable. Without it this study could not have been accomplished. I am also greatful to the boys who participated in this study, either as subjects or as helpers. Their willing- ness to cooperate and to assist was greatly appreciated. My wife Ellen's spirit and strength gave me my free- dom to work quickly and in solitude. 11 INTRODUCTION The present use of the stereoscope as a perceptual in- strument rests heavily on the assumption that the time of visual exposure to two different, yet structurally equal scenes (one presented to each eye) is short enough to elim- inate binocular rivalry. Thus, when an individual is visu- ally exposed to two scenes for a period of approximately one-half second, he is unable to see one scene and then the other. Ordinarily he perceives one of the two possible scenes. That which he does see may be determined by a var- iety of factors. Engel (1956) originally demonstrated that the scene which was more meaningful to the observer would be the one perceived. He simultaneously presented two pictures of a person (one upright, the other upside-down) in the stereo- scope. Individuals tended to perceive only that scene which was upright. In another study, Engel (1958) presented dif- ferent, yet structurally equal faces simultaneously in the stereoscope. He found that individuals tended to “fuse" their perceptions into more attractive faces. 1. 2. Since Engel’s studies there have been a number of ex- periments in stereoscopic perception which have supported the assumption that the scene which had more meaningful con; tent, in terms of the experience of the individual, was the one which would be more readily perceived. Bagby (1957) showed that more culturally relevant scenes are perceived when paired with scenes of another culture. Hastorf and Myro (1959) Presented upright and inverted postage stamps simultaneously and confirmed Engel's findings. Davis (1959) used words with different frequency usage and different emotional content and found that words with low frequency usage and words with strong emotional content tended not to be seen in the stereoscope. Lo Sciuto and Hartley (1963) demonstrated that words pertaining to one's own religion are perceived more often than words relevant to other re- ligions. In addition, there was a significant correlation between open mindedness and the tendency to see words re- lated to a religion other than one's own. Stereoscopic perception has also been studied in the area of criminology. Toch and Schulte (1961) prepared a series of "violent" vs. "neutral" scenes and presented them to men in their fourth year of police training, as well as to men in their first year of training, and to college soph- mores. They found that men in their fourth year of training perceived significantly more violent scenes than either of 3. the other two groups. These results suggest that percept- ual choice may be more than a function of the individual's experience with the content; it may, in fact, reflect a "readiness” to perceive violence in the environment. The tendency to perceive certain content may be related to an important functional readiness to look for the kinds of behavior with which one expects to become involved. Police- men are trained to identify and cope with violence in the environment. Their stereoscopic perceptions seem to function in accord with their expectations. It was then demonstrated by Shelley and Toch (1962) that the readiness to perceive violence in the stereoscope is positively correlated with a tendency to behave violent- ly in the future. In this study, all inmates of a work camp were presented with the series of violent vs. neutral scenes in the stereoscOpe. Those inmates who perceived the greatest number of violent scenes subsequently became disciplinary problems and were transfered to prison. Those inmates who saw the least number of violent scenes subsequently made a satisfactory adjustment to camp life. A further indication of the highly functional relation- ship between perceptual choice and overt behavioral tendencies is evident in a study done by Berg and Toch (1964). Their results indicate that those prisoners who are characterized as highly impulsive people are perceptually predisposed to perceive ”impulsive'scenes (blatant attempts at the grati- fication of different impulses) as opposed to more social- ' 4. ized depictions of need satisfaction. The impulsive prison- ers of this study were contrasted with prisoners who ex- hibited much evidence of intrapsychic conflict. This latter group saw significantly fewer impulsive scenes than the for- mer. The results are explained in terms of an increased per- ceptual sensitivity towards opportunities for gratification. That is, those people who habitually tend to gratify their impulses in an unsocialized manner have also deve10ped a facility for spotting opportunities for gratification. Their increased perceptual vigilance apparently functions in accord with their behavioral predispositions. In general, it seems that the way in which a person characteristically adapts in life influences the way in which he is predisposed to per- ceive scenes in the stereoscope. Intuitively, this relation- ship might be extended to environmental perceptual predis- positions. A study done in a detention home for juvenile delinquents (Collier,l963) extends the notion of the relationship be- tween perceptual prediSpositions and characteristic overt behavioral tendencies. In this study, the tendency to per- ceive violent scenes in the stereoscope is positively cor- related with assaultive themes on the Holtzman Inkblots as well as with subsequent recidivism rates. It is shown that one half of the delinquents who perceive the greatest number of violent scenes are detained again within three 5. months, whereas only one tenth of the low violent perceivers are subsequently detained. From the above descriptions of previous studies in stereoscOpic perception, it is evident that both content (experiential meaningfulness) and behavioral predispositions can play an important part in determining the perceptual choice. Stereoscopic perception might thus be studied as an index of historical relevance to the individual, and/or as an index of contemporaneous behavioral predispo- sitions. STATEMENT OF THE PROBLEM The present study is designed to investigate the relation- ship between stereoscopic perceptual choice and experiential and behavioral factors in Juvenile delinquents. Since Moore (1966) has shown that the readiness to perceive violence in the stereoscope is in part a function of age, sex, and socialization experience, it is questioned whether worthwhile and consistent differences in perceptual predispositions would be found among a group of mid-adolescent offenders. Since experience has been shown to be an important determin- ant of perceptual choice, it may be that the life styles of younger delinquents are sufficiently undeveloped so as to diminish the discriminating power of the stereoscope. It is assumed, however, that the stereoscOpe may differentiate high violence perceivers from low violence perceivers in a group of institutionalized mid-adolescent offenders. This assumption is based on the findings that anti-social, ag- gressive habits have developed in some youth (Bandura and Walters, 1959). Furthermore, it is assumed that these in- dividuals would be perceptually predisposed or sensitized to violent content. The rationale for this, and in no way in- consistent with previous assumptions in stereoscopic studies, 6. 7. is offered in terms of the "cue" properties of some stimulus events for Specific personality patterns (Dollard and Miller, 1950). Thus, the underlying assumption which we sought to con- firm was that those delinquents who consistently engage in aggressive, anti-social behavior would be more sensitive to the stimulus cues of aggressive behavior than character- istically non aggressive delinquents. Appropriate differences in perceptual sensitivity may also be reflected by the qual- ity of the stereoscopic perceptual choice. Since the experiential factor has only proved to be a significant determinant in studies with wide discrepancies in behavioral classifications, e.g., impulsive vs. neurotic, adjusting vs. non-adjusting, high frequency vs. low frequency, etc., the present study attempts to exPand its classifications. In terms of understanding the processes involved in percept- ual choice, and in terms of understanding the development of delinquent behavior, historical delinquent behavior might be advantageously divided into degrees of aggressiveness. Buss (1961) has defined aggression as ”behavior which delivers a noxious stimulus to another organism". Berkowitz (1962) implicitly defines aggression as behavior which intends to harm, and which has a damaging effect on another animate or inanimate object. In dividing aggression into different de- grees, this study attempts to consider both the psychological intent, and the effect which the behavior has on another per- son. Thus, historical delinquent behavior will be considered 8. either violent aggressive, non-violent aggressive, or non aggressive. The differentiation also suggests a develop- mental process, which has its origins in non interpersonal forms of transgression, and its ultimate, most direct form in painful physical attacks on other people. The first hy- pothesis of this study is that there will be an increasing tendency to perceive violence in the stereoscope, as a de- linquent's historical behavior pattern approaches physical violence. In addition to this experiential relationship, the present study also attempts to investigate the relationship between stereoscopic perceptual choice and contemporaneous behavioral predispositions. It is questioned whether delinquents who currently exhibit aggressive behavior will perceive more violence than delinquents who are currently non aggressive, regardless of previous historical behavior patterns. In other words, what will the effect be (on perceptual choice) when both historical, experiential variables, and current behavior- al variables are included in the same design? For example, what effect on perceptual choice occurs in a group of boys who historically were violent individuals, but who currently, within the institution, have not engaged in any aggressive behavior? Or, similarly, is the readiness to perceive violent content altered when a delinquent who has historically been non aggressive suddenly becomes aggressive within the instit- ution? Is the determination of the perceptual choice merely a function of the experience of the individual, or of his cur- 9. rent attitudes, or of both? This is the principal question which we attempt to answer. Hypotheses The hypotheses tested in this study are the following: 1. Delinquents who have historically engaged in aggressive, anti-social behavior will perceive significantly more vio- lent scenes in the stereoscope than delinquents who have not historically engaged in aggressive behavior. Specifically it is hypothesized that the perception of violence is in part a function of the degree of historical aggressive be- havior perpetrated, i.e., historically violent aggressive delinquents should perceive more violence than historically nonpviolent aggressive delinquents, who, in turn, should perceive more violence than historically non aggressive delinquents. 2. Delinquents who currently (within the institution) engage in aggressive behavior will perceive more violence than de- linquents who do not currently exhibit aggressive behavior. METHOD A. Subjects Fifty-four delinquents were selected from Lansing Boys Training School, Lansing, Michigan, a medium security in- stitution for delinquent boys. The ages of the boys in this institution range from 12-19. Boys actually selected ranged in age from 13-18. Since it was particularly dif- ficult to use subjects from the "closed program” (high truancy risks), all subjects were selected from the open program.“ This program consists of varying combinations of work-study details, all of which take place within the institution. B. Criteria for Selection Since the study was designed to measure differences between and within two variables, historical behavior and current behavior, the following criteria were used for se- lecting groups of boys: Historically, behavior was divided into three categories; violent aggressive, non-violent ag- gressive, and non aggressive. Currently, behavior was con- sidered either aggressive or non aggressive. The definitions *with the exception of one subject 10. 11. of these kinds of behavior will now be given. A historically violent aggressive delinquent is defined as one who has at least one mention of assault in his social history. Furthermore, the assaultive behavior described must not have been denied by the boy. In addition, the nature of the assault must have entailed actual physical contact, with obviously physically painful consequences, i.e., shoot- ing, stabbing, fist fighting, etc. Finally, if there is only one mention of assault on record, then it is also re- quired that there be some other mention of fighting be- havior in the record. In cases where a boy is charged with more than one assault, no other mention of fighting behav- ior is required. The objective in this group is to select boys who are habitually violent aggressive individuals, whose personalities might be characterized as aggressive (Buss, 1961, Berkowitz, 1962). A historically non-violent aggressive delinquent is defined as one who has involved himself in behavior which came very near to being physically assaultive, but which did not actually involve painful, physical contact. In addition, there must be no mention of fighting behavior in the social history; this is required to be more certain that the boy has not historically been violent aggressive. The objective in this group is to obtain a sample of boys who have apparent- l2. 1y begun to express their needs, hostilities, desires, etc., towards other people, but who have not as yet become habit- ually involved in painful physical attacks on other people. Examples of such historical behavior included in this group were: threatening with a weapon, homosexual molestation (for- cing another boy to commit sodomy), heterosexual molestation (touching parts of another person's body) armed robbery, and strong-arming. ‘ A historically non aggressive delinquent is defined as follows: one who has no history of assaultive or fighting behavior, and who has limited his offenses to non-interper- sonal forms of transgression. Specific types of historical behavior included in this group were unlawfully driving away an automobile, glue sniffing, indecent exposure, breaking and entering, theft, and truancy. A currently aggressive delinquent is defined as one who fights within the institution, either with the staff or with other boys. In order for a boy to be considered currently aggressive, he must have been in at least two fights within the last two months. Thus, a boy who has been in many fights within the institution, but not fought for two months, is not considered currently aggressive. In addition, no boy who has not been at the institution for at least two months is 13. eligible for selection. Currently non aggressive boys are simply defined as those boys who cannot be considered currently aggressive, and who have been in the institution for at least two months. Thus, these boys have not been in at least two fights over the last two months. ' The determination of an individual's current behavior status was achieved through brief interviews with all coun- selors. Each counselor has charge over approximately 50-60 boys, and is fairly well acquainted with the behavior of each of their boys. The interview consisted of asking the counselor whether or not certain boys met the criteria outlined. The determination of an individual's current behavior status within the institution is a particularly difficult task, since counselors are often uncertain about specific instances of fighting, their causes, conspirers, etc. In addition, since degrees of permissiveness vary from cottage to cottage, the amount of fighting also varies as a function of different residences. Because of the inexactness inherent in the procedure for determining current behavior status, an attempt was made to minimize error in the following ways: other staff were questioned (clinical psychologist, social worker); deten- tion lists and intra-institutional notes were checked; in a few cases boys were asked about the behavior of other boys. In cases where discrepant impressions of an individual were 14. given, the individual was automatically eliminated from the test groups. It was also noted in the process of determining current behavior status that fighting was itself a behavior in which most boys engaged. There seemed to be few boys in the institu- tion who had never fought with a peer in the cottage or else- where. In addition, there seemed to be few boys in the insti- tution (at least in the cpen program) who might be considered "violent”. Those boys who do become violent are sent to the closed program, or to another, more confining institution. Thus, there is some question as to whether the classification of aggressive vs. non aggressive (currently speaking) be- havior is representative of aggressive tendencies or exper- iences to the same degree that the historical variable achieves representation. Total number of §fs used was fifty-four. Based on the two variables discussed above, historical behavior and current behavior, six groups of nine subjects each were classified as follows: 1) historically violent aggressive and currently aggressive, 2) historically violent aggressive and currently non aggressive, 3) historically non-violent aggressive and currently aggressive, 4) historically non-violent aggres- sive and currently non aggressive, 5) historically non aggres- sive and currently aggressive, 6) historically non aggressive and currently non aggressive. Table 1 (Appendix) presents a 15. summary of the group differentiations. Each group contained six Negro and three white subjects. However, in four of the groups, a Mexican boy was used in place of one of the white subjects. This occurred because of the lack of available white subjects meeting the criteria outlined. Mean ages for each group are indicated in Table 2 (Appendix). C.Apparatus The stereoscope has been described elsewhere (Tooh.and Schulte, 1961). Six of the original nine violence slides were used as well as one slide from the series used by Berg and Toch (1964). The six from the original series, in the order in which they were presented, were the following: 1. man with.knife in back 2. man standing over body 3. man hanging 4. man.shooting himself 5. two men fighting 6. man stabbing another postman farmer man.with suitcase man with.microphone man showing pictures man with drill press The seventh pair of stereograms consisted of a man and wo- man dancing, paired with a man struggling to take off a wo- man's clothing. The time exposure uded in this study was consistent with.most of the previous studies (0.5 seconds). Light intensity was set at 80 volts for each field (compar- 2 able to 0.2 candles/ft. Each pair of slides was presented l6. twice, once to each eye. This procedure controlled for pos- sible eye dominance. D. Administration Procedure All subjects were tested within a period of five days. Subjects were brought to the preliminary waiting room (the office of the clinical psychologist) in groups of three. Each subject was called out randomly. An office boy was sent by the psychologist to bring back three boys. These boys, having arrived, were given the following instructions by the psychologist: There is a student from Michigan State upstairs who is doing some research on how well people see things. He wants each of you to help him out for a few minutes. He has nothing to do with the school. There's no reason to worry about anything because he won't even know who you are. I'll give you each a number, and I'll introduce you to him as that num- ber. Each boy was then presented to the experimenter as a subject number. As the experimenter escorted the subject upstairs, he gave the following instructions :- I'h a student at Michigan State, and I'm doing some research on visual acuity. Do you have good eyes? Do you wear glasses? 0.K., I’ll show you what I'd like you to do...(arriving at the testing room)...this is a stereoscope. It's a machine which flashes a picture on for a very short time. I'm interested in how much of the picture all boys can see, if I flash it for just half a second. I have a number of pictures that I'd like to show you if you'd be willing to cooperate. Would you? 0.K., now you should understand that these are pictures of people doing different things, and 17. that I' m interested in how much you can tell me about what they' re doing. The picture comes real fast so you 've got to look real quick. First we have to line up your eyes. If you look in the stereosc0pe you should see a dot and a circle.... Subjects were then asked to set the knob for their indivi- dual fusion points. E. Scoring Procedure The following code, created by the author, and in general, consistent with previous means of scoring perceptions in the stereoscope, was used: Q;la non-violent perception, in which either a perception of the neutral picture, or a neutralized perception of the "violent picture (6.8., "man shaving himself" instead of "man shooting himself in the head" ) occurs. F- a nondviolent fusion, in which both pictures are seen, but no evidence of violence is reported (e.g.,' 'man waving his arms' '). V- a violent perception, in which either the violent picture alone, or both pictures are seen (e.g.,' 'man with a knife in his back" or "man dancin with a lady and he's try- ing to take her clothes off" RESULTS Neither of the two hypotheses were confirmed. Specif- ically, results did not indicate a linear relationship be- tween the degree of historical aggressive behavior and the perception of violent scenes in the stereoscope. Also, re- sults did not suggest that currently aggressive individuals are more predisposed to perceiving violence. The appropriate analysis of variance (Table 3, Appendix) indicates a highly significant interaction between historical and current be- havioral variables. That is, it seems clear that both exper- iential and current motivational states are interacting in some way to influence the tendency to perceive violence in the stereoscope. An examination of paired comparisons of means of each group was made in an effort to determine the nature of the interactional process. Using Tukey's method (Winer, 1962) tiscores were obtained for selected paired comparisons (Table 4, Appendix). A summary of these comparisons is represented below. 1. When equated on historical kinds of aggressive behavior (either violent or non-violent) currently non aggressive delinquents perceive significantly (p<.01, p<.05) more vio- lent scenes than do currently aggressive delinquents (pairs 1. 2). 19. 2.'When equated on historically non aggressive behavior, de- linquents who are currently aggressive perceive significant- ly (p<;05) more violent scenes than.currently non aggressive delinquents (pair 3). 3. When equated on current non aggressive behavior, delin- quents who are historically non aggressive perceive signif- icantly (P<;Ol, p