TECHNIQUES DESIGNED TO MITIGATE THE IMPACT OF MASS MEDIA SEXUAL VIOLENCE ON ADOLESCENTS AND ADULTS Paper prepared for the Surgeon General's Workshop on Pornography and Public Health , June 22-24, 1986 in Washington D.C. Authors: Edward Jonnerstein Center for Communication Research University of Wisconsin-Madison Daniel Linz William S. Middleton Memorial Veteran's Hospital and University of Wisconsin-Madison Preparation of this paper as well as some of the research reported herein was supported by National Science Foundation grant YBNS-8216772 to Dr. Edward Donnerstein and Dr. Steven Penrod, University of Wisconsin-Madison. 1 I. Introduction The concern over the potential effects of med ia v iolence on both children and adults is not new. The research literature abounds with studies showing effects on behavior, attitudes, and perceptions. While there are disagreements regarding the relative contribution of mass media violence to actual aggressive behavior (e.g., Freedman, 1984), the research community is fairly unanimous in its conclusion that at least some individuals are affected and that exposure to many types of violent media can contribute in some part to the general incidence of violence in our society. The research findings linking exposure to mass media violence with anti-social behavior have been summarized in the 1982 NIMH report on media violence and will not be discussed here. Uhile the last two decades were devoted to research on the impact of mass media violence, the last few years have seen some emphasis on methods of countering or mitigating the impact of exposure to media violence. This research has nowap reached the point where we can say with certainty that a given message, educational program, or label on a particular media product is effective in reducing the impact of exposure to violence. We are, however, able to identify some promising procedures developed independently by several researchers that may counter the effects of exposure to media violence. The intent of this paper is to very briefly review these findings and the findings from empirical investigations on the impact of exposure to sexual violence, and then to review in greater detail attempts at mitigating the effects of exposure to violent pornography and mass released (R-rated) sexually. violent media presentations. Several of our suggestions for intervention programs for consumers of sexual violence are derived from the results of investigations of programs to counter the effects of exposure to milder forms of violence contained in network television portrayals among gradeschool children. II. The Effects of Exposure to Sexual Violence in the Media L By the latter part of the 197,0's many persons were becoming concerned about what appeared to be an increase in violence in pornographic materials. Feminists warned that: ". . . pornography is in fact escalating its misogyny, promulgating raw, mutilation, and even murder as average sexual acts, depicting the 'normal' man as a sadist and the 'healthy' woman as a willing victim" (Morgan, 1978, p. 55). Although it is not clear exactly to what extent pornographic depictions in American society have become more violent, there exists preliminary evidence from systematic content analysis of both "hard core" materials (Smith, 1976a, 1976b) and "soft core" materials such as Playboy and Penthouse magazines (Malamuth h Spinner, 1980) that violence in pornographic portrayals had increased at least throughout the 1970'9. The extent of the increase in the 1980's is at this point largely unknown. There is some suggestion, however, that materials more readily available to adolescents, because of cable and VCR access, are perhaps more sexually violent than traditional X-rate: materials (Palys, 1986). Regardless of whether there has been an increase in violent images in po~ography , the combination of portrayals of sex and violence in the material that is available may be especially potent for several reasons (Malamuth, 1984; Malamuth 6 Donnerstein, 1982). First, the pairing of sex and aggression in violent pornographic depictions may classically condition viewers to become sexually aroused to violence. It is assumed by many psychologists and psychiatrists (Abel, Blanchard, h Becker, 1978) that such conditioning processes are responsible for the behaviors of sexual offenders. Second, much aggressive pornography has as its predominant theme the idea that victims secretly desire assault, often deriving sexual pleasure from it. This may cause the viewer to believe that women want to be sexually assaulted, and even if initially resistant to sexual advances will eventually submit and even "enjoy" the aggression. Continued exposure to depictions which portray this message or "rape myth" may affect both attitudes toward sexual violence and behavior. It is precisely the 3 portrayal of this myth in violent pornography that may account for many of the increases in male sexual arousal to rape depictions, male perceptions and attitudes towards rape victims, and aggressive behavior against female victims following exposure to the material. Several studies have shown that if the victim is portrayed as becoming involuntarily sexually aroused by the assault, subjects show levels of sexual arousal as great and sometimes greater than those stimulated by mutually consenting sex (Malamuth & Check, 1980a, 1980b, 1983; Malamuth, Helm & Feshbach, 1980). On the other hand, more realistic rape portrayals which depict the victim as abhorring the experience are significantly less sexually arousing than consenting portrayals (Malamuth 6 Check, 1980a, 1980b; Malamuth, Helm, & Feshbach, 1980). Several studies have also shown effects for exposure to aggressive pornography on aggressive behavior in the laboratory. Subjects exposed to illustrated stories from Penthouse magazine depicting the rape of a woman with "some suggestion" of a positive outcome who were also given a communication designed to reduce inhibitions against aggression delivered significantly more electric shocks to a female confederate. Donnerstein (1980a, 1980b) conducted similar studies using short pornographic film clips. The results showed that the combination of exposure to aggressive pornography, a high level of pre-exposure anger, and pairing with a female victim resulted in the highest level of aggressive behavior (compared to neutral or nonaggressive pornographic films). However , even non-angered male subjects exposed to violent pornography showed significantly higher levels of aggression when paired with a female victim. In another experiment Donnerstein and Berkowitz (1981) showed that exposure to aggressive-pornographic films in which the movie victims appear to enjoy being roughed up or are sexually aroused while being raped tend to increase subjects' later aggression against a female even when the subjects are not angered. 4 Exposure to aggressive-pornographic depictions with a realistic outcome in which the victim is shown to abhor the experience, however, does not appear to result in greater levels of aggressive behavior against women by non-angered subjects, although the greater aggression is still found for angered subjects. There is also evidence that exposure to mass media that portray violence against women in sexually nonexplicit contexts may affect males' attitudes and levels of aggressive behavior as measured in the laboratory. For example, in a field experiment Halamuth and Check (1981a) tried to determine whether the depiction of sexual violence contained in mass media nonpornographic depictions such as The Getaway and in a mass-released film with similar content influenced the viewers' perceptions of women and their attitudes towards women. In their investigation male and female students participated in a study which they were led to believe focused on movie ratings. One group of subjects watched, on two different evenings, The Getaway and Swept Away (which also shows women as victims of aggression within erotic contexts). A group of control subjects watched neutral, feature-length movies. These movies were viewed in campus theaters and as part of the "Campus Film Program." The dependent measures were scales assessing the acceptance of interpersonal violence (AIV) against women, rape myth acceptance (WA), and beliefs abour adversarial sexual relations (Burt, 1980) which were embedded in a larger scale containing many items as part of a "Sexual Attitudes Survey." Subjects reported that they saw no connection between the survey and the movies. The results showed that viewing the sexually aggressive films significantly increased male but not female acceptance of interpersonal violence and tended to increase rape myth acceptance. In two more recent studies, Donnerstein (1983a, 1983b) and Donnerstein and Berkowitz (1983) tried to assess the impact of nonpornographic depictions of sexual violence against women on physiological arousal, attitudes toward women, and aggressive behavior in the laboratory. In the study by Donnerstein (1983) 5 subjects were first either angered by a female or angered by a male confederate and saw one of four types of short films. One group saw a sexually explicit aggressive-pornographic film in which a woman is attacked by a man at gunpoint, tied up and raped (violent pornography). The second group of subjects saw the aggressive-pornographic film with the sexually explicit parts edited out (a clip that might have been suitable for cable television). In this film the woman is still attacked by a man at gunpoint, tied up, slapped around and generally aggressed against; but, there was no nudity or even simulated sexual activity. The third group of subjects saw non-violent pornography. The fourth film was a non-sexually explicit, nonviolent neutral presentation. The results of this study showed that for subjects angered by a female confederate, both the violent-pornographic film and violence-only film conditions produced heightened levels of aggression. The Donnerstein and Berkowitz (1983) study showed a similar pattern of results. Angered subjects exposed to aggressive but nonpornographic materials behaved more aggressively towards a female confederate than control subjects exposed to nonviolent pornography. Attitudes about rape and subjects' willingness to say they might commit a rape were also measured. The most callous attitudes and the largest percentage of subjects indicating some likelihood of raping were found in the aggressive nonpornographic condition. Most recently, Linz, Donnerstein, and Penrod (1984), and Linz (1985) have studied the impact of prolonged exposure to nonpornographic materials that portray violence against women in an extremely graphic fashion (R-rated slasher films). In one study, (Linz et al., 1984), men who viewed five movies depicting violence against women came to have fewer negative emotional reactions to the films, to perceive them as significantly less violent, and to consider them significantly less degrading to women. There was also a tendency for this "desensitization" to the filmed violence to "spill over" into subjects' judgments of a female victim in another context. Men who were exposed to the large doses of filmed violence 6 against women judged the victim of a violent assault and rape to be significantly less injured and evaluate her as generally less worthy than a control group of subjects who saw no films. An additional study (Linz, 1985) examined the relationship between individual differences in psychoticism (as measured by the SCL-90, Derogatis, 1977) and responses to R-rated, mass released movies containing sexual violence and rape on later decision making about the victim of an acquaintance rape. The results showed several consistent interaction effects with subjects with relatively high psychoticism who were exposed to the high sexual violence (rape) films more likely to endorse the use of force in sexual relations and evaluate a victim portrayed in a reenacted rape case as less credible, less worthy and less attractive. III. Mitigating The Effects of Sexual Violence At the end of all the experiments described above, subjects were debriefed. These debriefings generally concentrated-on the unreality of the media depiction. Subjects are cautioned that pornographic and nonpornographic portrayals of women desiring, enjoying, or sexually aroused to forced sexual relations are fictitious. Material is also presented which tries to dispel common rape myths (Burt, 1980) especially any myths that have been portrayed in the stimulus materials used in the experiment. As Sherif (1980) has pointed out, it is extremely important for experimenters in this area to evaluate the effectiveness of these debriefings. Over the last few years several followup studies of the effectiveness of these debriefings have been undertaken. Donnerstein and Berkowitz (1981) compared the responses of debriefed subjects who had been exposed to violent and non-violent pornography in their study with control subjects exposed to a neutral film who received no debriefing on seven items taken from the Rape Myth Acceptance scale and the Acceptance of Interpersonal Violence scale (Burt, 1980). Subjects in this study were followed up anywhere from two weeks to four months after participation and asked to 7 complete the seven item questionnaire. The results of this follow-up showed that those subjects who viewed aggressive and non-aggressive pornography who were debriefed showed less acceptance of rape myths than nondebriefed subjects (who had been exposed to non-aggressive pornography or neutral films). More recently, Malamuth and Check (1984) conducted a study in which male and female subjects were exposed to sexually explicit stories depicting either rape or mutually consenting intercourse. Afterwards, subjects exposed to the rape version were given statements emphasizing that the depictions of rape in the stories they read were complete fantasy, and that in reality rape is a serious crime punishable by law and that victims of rape usually suffer severe psychological and physical damage after the assault. Subjects were also given specific examples of rape myths and assurance that these commonly held beliefs are indeed completely fictitious. Ten days later subjects received, in their classes, a "Public Survey." As part of the survey subjects-were asked to read newspaper articles and give their opinions, one of which was an article about rape. In addition, subjects were asked questions regarding the police's decision to bring charges, the victim's responsibility for her own rape, the recommended sentence for the man if convicted, and whether they thought they themselves might rape a woman if they could be assured of not being caught and punished. Subjects were also asked to what extent victim, rapist, and societal factors contribute to rape. There were no effects for subjects' evaluation of the newspaper article on rape. Subjects who were exposed to the'rape stories and then debriefed were, however, less inclined to see women as wanting to be raped, and victim behavior generally as a cause of rape compared to subjects who read the consenting story and received no debriefing. Debriefed subjects were also less likely to see rape as the result of a normal sexual tendency among males. In another experiment, Check and Malamuth (1984), using basically the same procedures, had subjects first read rape depictions in which the victim and 8 assailant were either acquainted or unacquainted with one another. As in the previous study, subjects were later (two to three days) presented several newspaper stories with a story about rape embedded among them. Subjects, when exposed to the rape debriefing, gave the rapist in the newspaper report a higher sentence and saw the rape victim as less responsible for her own assault. However, these effects only happened if subjects had been exposed to an example of a rape depiction which was relevant to both the rape myths discussed in the rape debriefing and the newspaper report of the rape. Specifically, only when subjects were exposed to an acquaintance rape scenario and given a debriefing which emphasized rape myths pertinent to that acquaintance situation (e.g., a woman who goes to a man's apartment deserves to be raped) was the debriefing effective. Further, there were no effects for receipt of the debriefing if it was preceded by a non-violent mutually consenting scenario instead of a rape scenario. The effectiveness of debriefings u&d in studies of long term exposure to movies that are nonpornographic but portray violence against women in an extremely graphic manner (R-rated slasher films) have also been assessed. One aspect of these studies that is particularly noteworthy is the long term nature of these followups. Subjects who are surveyed as long as seven to eight months after participation in debriefings show significant increases in their sensitivity towards rape victims. In a study designed to assess effectiveness of the debriefing used in the Linz et al. (1984) study on long term exposure to R-rated slasher films (Donnerstein, Penrod, &i Linz, 1984) subjects were assessed several weeks before participation in the study, one to three days after participation in the study and receipt of the debriefing, and again six weeks later. At each point in time subjects completed the entire Rape Myth Acceptance scale. The debriefing used in this study was a videotaped message in which one of the principle investigators cautioned subjects that "constant exposure to violence can desensitize or harden and make people callous to violence." The part of the 9 message particularly relevant to the rape myths portrayed in the films used in the study emphasized that "one problem with some of the scenes in the films that you saw this veek is that they tend to reinforce certain myths about rape such as the notion that women who wear provocative clothing are asking for or deserve sexual assault, or that women who put themselves in a risky situation are setting themselves up to be raped and hence are responsible for what happens. The filmL Spit on Your Grave, in which the woman was camping and canoeing by herself, illustrates this point." The subjects were told that research has shown that films of this type tend to reinforce and maintain these common but fictitious beliefs about rape. The results of the followup shoved that immediately after participation in the study and debriefing, subjects' average scores on the rape myth acceptance scale declined relative to the pre-study participation score by a few points (although the decrement was not statistically significant). The score obtained six weeks later were nearly identical to the immediate post-participation level scores (and still a few points lower than those obtained before the study). Perhaps more compelling is a followup study done by Linz (1985) which tried to assess the effectiveness of debriefings given to subjects who received either high (5 movies) or low (2 movies) doses of graphic filmed violence against women. As in the previous study, subjects' scores on the Rape Myth Acceptance scale were obtained before exposure to the violent films (in this case nearly two months) and again after participation in the study and debriefing. The films used in this study did not portray rape behavior per se although the films could be interpreted as propagating several myths about blaming the victim for her assault if she places herself in certain situations. Consequently, subjects were told: Although there were no rape scenes in these films, these films tend to reinforce certain myths about rape and other forms of sexual assault. Some examples of the types of rape myths that these films may have reinforced are that "only certain types of women get raped," or that "if 10 a women really vanted to fend off an attacker, she could," or that "women who dress provocatively are asking to be sexually assaulted." After receiving this message (conveyed by videotape), subjects were told they would be contacted later and were dismissed from the laboratory. As in the previous study the effectiveness of the debriefing was assessed but instead of surveying subjects six weeks after participation, the subjects were requested to complete the Rape Myth Acceptance scale six to seven months after participation in the study and debriefing. The results of this followup showed that subjects experienced a statistically significant decline in Rape Myth Acceptance relative to their pre-study levels. Further, the decline in this index was equal in size for both the large dosage and small dosage film exposure groups. Similar decreases in Rape Myth Acceptance after a proper debriefing can also be found among female participants in studies involving exposure to sexually violent media. Previous research on fen&e subjects' sexual responsiveness to rape depictions indicates a tendency for women to become sexually aroused to rape depictions if the depiction emphasizes the women victims of rape becoming sexually aroused by the experience (Stock, 1983). More recently Krafka (1985) has found that females exposed to R-rated non-pornographic sexual violence (slasher films) become emotionally desensitized (less anxious and depressed) to these depictions vith repeated exposure and later, when asked to evaluate a videotaped reenactment of a physical assault-rape trial were more likely to evaluate the victim of sexual assult harshly, than females exposed to other types of film. As with studies involving males, females exposed to sexual violence were thoroughly debriefed after participation in these studies. They were warned about the possibility of becoming desensitized to the violence in the film stimuli used in the study and are also debriefed about the portrayal of various rape myths in sexually violent movies. In addition to pointing out these concerns, females in the study by Krafka (1985) also received an additional message concerning the more 11 general belief prevalent in our society that women are sexually aroused by force: The sum total of rape-myth says that women enjoy rape, will be sexually aroused by force, and that they ask to be raped in subtle, if not direct, ways. One of the most common sexual fantasies for women, in fact, involves rape -a point that is not lost on the writers of best-selling romance novels. A common theme of the romance trade is that a handsome devil-hero ravishes the novel's heroine in a fit of uncontrolled passion; she later falls in love with her ravisher. The popularity of the romance novel is due in part, I think, to their ability to play on our fantasies. But important distinctions must be made between the fantasy and the fact of rape. First of all, fantasy does not typically involve violence or pain. The typical rape fantasy involves being overwhelmed gently by a man who considers one so desirable that he quote "simply cannot control himself." Second, fantasy is safe. Lots of people, male and female alike, experience things through fantasy that they would never want to experience in real life. Fantasy is okay -and yet we know that the typical real-world rape bears little resemblance to either rape as it is depicted most often in film or to the kind of pseudo-rape which might play out in a woman's fantasy. Rape is a crime, and it is a crime of violence that has little to do with the satisfying of sexual urges. It involves coercion.or threat in some form, and in a large percentage of cases, requires physical force to subdue the victim. This may res_ult in pain or serious injury to a woman, and most women who have been assaulted respond with extreme emotional trauma. Film-depicted rape tends to gloss over unpleasantries, so consequently, it doesn't reflect reality. Followvp evaluations of the effectiveness of a debriefing containing this type of message have been undertaken immediately, six weeks, and six'months after participation in the study and debriefing (Chapin, 1985; Krafka, 1985). The results of these evaluations indicate that exposure to sexually violent materials coupled with debriefings will produce significant reductions in rape myth acceptance (compared to a baseline measure taken several weeks before participation in the study) not only immediately but these effects will remain six months later. IV. What Factors Mediate the Effects of Debriefings? As we have seen, subjects participating in studies using violent pornographic depictions, as well as studies using nonviolent pornographic depictions of violence against women who have received a proper debriefing, emerge from this experience more sensitive to cultural stereotypes about violence against women. 12 Further, these effects have been found in studies in which subjects have been exposed to relatively large or relatively small doses of sexual violence during the experimental phase of the study, for male subjects as well as female subjects, immediately after participation in the debriefings and six to eight months after participation. What is it about the debriefings and/or participation in the studies themselves that lead to these changes? The results of two of the studies that we have reviewed (Check h Malamuth, 1984; Halamuth h Check, 1984) suggest factors about participation in sexual violence experiments and debriefings that may be of great importance. First, as Check and Malamuth (1984) note, it is probably necessary to specifically tailor the debriefings to the types of myths portrayed in the material used during the experimental phase of the study. Debriefings which focused on rape myths not specifically portrayed in the experimental phase of the Check and Malamuth (1984) study, for example, were not effective in reducing rape myth acceptance. This finding seems congruent with the research by Donnerstein et al. (1984), Linz et al. (1985), Krafka (1985), and Chapin (1985) where debriefings were presented in a videotaped format which were interdispersed with specific examples from the material presented to subjects during the preceding phase of the experiment. Second, messages dispelling rape myths might be most effective for subjects who had first been exposed both to the rape scenarios in the experimental portion of the study and the debriefing. Debriefings coupled with pre-exposure to consenting sex scenarios in the Malamuth and Check (1984) study were relatively ineffective in changing attitudes about rape. These findings imply that it may be necessary for subjects to first become aware of increased levels of sexual arousal in response to the rape passages, or experience desensitization in the face of violent portrayals and then to receive a debriefing which addresses these processes. As Halamuth, Heim, and Feshbach (1980) point out, the debriefing might provide the subject vith a certain kind of insight 13 comparable to that experienced by students who might become aware of racist feelings during a study on social prejudice. This may also be the case for subjects who have participated in studies using nonpornographic materials that portray rape myths. We might add to this Nisbett and Wilson's (1977) observation that many psychological phenomena such as the inhibition of helping due to the presence of bystanders (Latane & Darley, 1970) and many experiments within the insufficient justification paradigm (Aronson 6 Mills, 1959) probably would not occur in the first place if people were aware of the impact of certain critical stimuli. If subjects were aware of the effects of the presence of others on their tendency to help they vould undoubtedly try to counteract that influence. Similarly, as Kelley (1967) has also noted, results of insufficient justification experiments could probably never have been obtained if subjects were aware of the critical role of social pressure from the experimenter. If subjects realized that their behavior was governed by this pressure they would not have been motivated to move their attitudes into line with their behavior because they would realize that it was the result of external pressure, not their own attitudes. The debriefings used in sexual aggression experiments might provide subjects with similar sorts of insights about psychological processes. This insight might then be useful for subjects in short-circuiting the negative effects of exposure to sexual violence in the future. To say that this process might serve to "inoculate" subjects against uncritical acceptance of rape myths (in the classic persuasion and attitude change paradigm, e.g., McGuire h Papageorgis, 1961) as have Check and Malamuth (1984), may be somewhat inaccurate, however. According to Check and Malamuth "presenting a pro-rape communication (e.g., a rape depiction likely to result in subjects perceiving the victim as a willing and perhaps blameworthy participant in the assault) and then presenting counterarguments designed to dispel such rape myths 14 (e-g-, a rape debriefing) may serve to immunize subjects against uncritically accepting rape myths in the future" (p. 17). It is difficult to see exactly how decreased acceptance of rape myths would be predicted by an inoculation to persuasion approach as long as we adhere closely to McGuire's original formulation. It is possible, for example, that rape myths may be beliefs that are similar to the "cultural truisms" that, according to McGuire and Papageorgis, people are unpracticed at defending. As Petty and Cacioppo (1981) note, because people have no counterarguments with which to resist a persuasive message against these truisms, they remain highly vulnerable to influence. McGuire (1964) suggested that an inoculation treatment would consist of exposing people to a few pieces of counterattitudinal propaganda prior to exposure to the threatening message and showing them how to refute these arguments. According to the theory, presentation of weak counterarguments produces resistance to future attacks against the cultural truism because the inoculation poses a threat that motivates people to develop arguments to bolster their initial beliefs. Thus, relatively weak attacks (such as debriefings) might actually produce more resistance in people to future attempts to dispel rape myths, not greater critical evaluation of rape myths. Finally, a word of caution. As the research currently stands it is difficult to tell from any of the studies reviewed if 3 aspect of the debriefings actually accounts for changes in subjects' rape myth acceptance or if given enough time subjects would naturally experience a change in these beliefs. Those debriefings which have been effective in changing subjects' beliefs have always been preceded by exposure to the rape depiction phase of the experiment. Consequently, it is impossible to rule out the possibility that exposure to the rape materials alone and participation in some other activity besides the debriefing, or even sufficient time for subjects to reflect or rest, might result in lowered rape myth acceptance or at least a return of rape myth acceptance to 15 pre-study participation levels. From a study by Malamuth and Ceniti (1986), for example, we might suggest the possibility that a rest period might result in greater sensitivity towards rape victims. These authors have found that subjects exposed to violent pornography who are asked one week later to participate in an ostensibly unrelated experiment involving the administration of aversive noise shoved no increases in aggression. The findings of this study stand in apparent contradiction to the results of previous studies (Donnerstein, 1980a, 1980b, 1984; Donnerstein h Berkowitz, 1981; Malamuth, 1978). As Malamuth and Ceniti (1986) point out the important difference between these studies and the Malamuth and Ceniti study is that earlier investigations examined aggressive behavior immediately after subjects were exposed to violent pornography rather than testing for relatively long-term effects. These findings suggest that with time subjects exposed to sexual violence might return to baseline levels of hostility and aggression toward women. Rape myth acceptance among subjects exposed to violent pornography might well decline naturally with the passage of time also. However , examining the possibility of this decrement with time vould involve exposing subjects to rape depictions and not debriefing them -a procedure that would be unethical. V. Designing Interventions to Counter Sexual Violence in the Media One way to test the idea that providing subjects vith knowledge about psychological processes they may be experiencing while viewing sexual violence might counter the effects of these depictions, would be to design an experiment directed at reducing acceptance of rape myths and aggressive behavior towards women in the laboratory through the use of E-briefings. Messages could be constructed vhich inform subjects about the effects of exposure to violent pornography and other forms of aggression against women administered before participation in experiments. Recently, Bross (1985) has examined the effectiveness of a pre-film message 16 informing male viewers who were later exposed to large doses of R-rated slasher films about the psychological processes which might be operating. These pre-film messages explained to subjects what psychological effects might result from viewing sexually violent media. The message was similar to videotaped debriefing presented after participation in experiments involving sexually violent materials (e.g., Linz, 1985; Linz, Donnerstein, h Penrod, 1984). Clips of scenes from slasher films were interdispersed throughout the filmed message to assist subjects in understanding the effects that can be caused from viewing slasher films. Of special interest were the subjects' awareness of desensitization to violence and the possibility that viewers maight come to viev violent scenes less critically when they are juxtaposed with sexual ones. Thirty-three male introductory psychology students took part in the experiment. Eleven subjects were exposed to the manipulation, while the other twenty-two were examined for comparison purposes. Subjects were informed that they had been chosen to take part in a film evaluation study and were to view six full length feature motion pictures (all R-rated slasher films) over the period of two weeks. Subjects in the experimental condition saw the pre-film message before viewing the first film. The control group simply saw the first motion picture after signing the consent form. All subjects viewed four more slasher films over the next two weeks. After each film subjects indicated how much violence they perceived in the films, how degrading the films were to women, the realistic nature of the violence and self-reported physiological arousal. On the final day of the experiment, all subjects were contacted by phone and told that the last film they were originally scheduled to see had not arrived and were asked if they would take part in a study being run by the University's Law School. All agreed and reported the next day to the law school courtroom vhere they viewed a videotaped reenactment of a mock rape trial. Following the viewing, subjects evaluated a number of aspects of the trial (to determine if there were any "spill 17 over" effects into decisionmaking in a more realistic context). The pre-film message is a videotaped presentation with clips from slasher films edited into the tape. The message stated: During the next two weeks we will show you six full length feature films and then will ask for your reaction to these films. The six films you will see during the next two weeks have one thing in common. They are all R-rated which means they all contain scenes of explicit aggressive behavior. In addition, the films may contain scenes of a sexual and/or erotic nature. I would not like to discuss with you the effect that this type of material can have on viewers. One major problem with these films is that constant exposure to violence can desensitize or harden and callous people to violence in general. As an example, we will show you some scenes from these films. [Film Clip] While you may initially find some of the violence in these films disturbing, it will perhaps become a little easier to tolerate the violence after continued exposure. This is a normal reaction. You should, however, be aware that this type of film can make people less sensitive to violent acts. Another concern we have in reference to desensitization to violence is that many of the films that you will see are particularly violent in regard to women. For example, the Tollowing scenes are among those you vi11 see. We are particularly concerned about the issue of violence against women. Just as you may experience desensitization to violence after viewing these types of films you may experience the same effect with regard to violence against women. Again, you should be aware of this issue and sensitive to this problem. A second major problem is that in many of the films an erotic or sexual scene is combined or juxtaposed with a violent scene. As an example of this type of combination, here is a scene from one of the films. [Film Clip1 This type of scene map quicken the process of desensitization that makes one more tolerant of violence. Also, such a combination places the violence in a very positive context because it occurs while the viewer is aroused or in a very positive state. Many of the violent scenes you will see are preceded by sexual, sensual, or erotic content. One potential problem with this combination is that some individuals who become sexually aroused to the sexual portion of the scene also become sexually aroused to the violent portion of the scene. Again, this is a normal reaction for some individuals but you should be aware that this type of conditioning can occur. Further, you should be aware that some individuals become less sensitive to this type of violence, especially violence against women when the violence is placed in an erotic context. You should be aware that films like these can condition people to become sexually aroused during violent scenes. 18 Thank you very much for your cooperation in this film evaluation study. It should be noted that this message tries primarily to increase awareness among subjects of the psychological processes vhich might be the result of continued exposure to slasher films, instead of a direct attempt to persuade subjects that viewing sexual violence is wrong or harmful. The results indicated that, overall, subjects exposed to the pre-film message were less susceptible to the effects of the slasher films on the first day of the study than were those who were not exposed to the message (see Table 1). Although the differences were not statistically significant, subjects in the message condition reported seeing more violent scenes (number of scenes) and more violent scenes directed at women than did subjects in the no-message condition. Subjects exposed to the pre-film message also reported seeing more combinations of erotic and violent scenes than the non-exposure subjects on day one of the experiment. Subjects in the message condition found The day one film to be significantly more degrading to women than no message subjects. There was no discernible pattern of differences between prebriefed and control subjects' responses to the mock rape trial questionnaire. Another finding that should be noted was the high dropout rate of subjects who had been exposed to the pre-film message. Eleven subjects were originally shown the message on day one of the study. One subject decided to not continue his participation in the study immediately after viewing the message. Over the two week experimental period, four more subjects dropped out of the study (over 45% of those originally shown the message). The no message group had only a 32% drop out rate. Comparisons between subjects who discontinued participation in the experiment and those who decided to stay throughout indicate some interesting trends. Those vho dropped out of the study reported seeing more scenes of violence, more combinations of erotic and violent scenes, found the film to be more degrading to both men and women on day one than those who stayed for the 19 duration. Subjects who dropped out of the study also reported more physiological arousal to the first film than those who completed the entire study. The message might have convinced subjects to leave the study. In fact, one subject decided to discontinue his participation immediately after viewing the pre-film message. He stated that he did not want to be exposed to the slasher films because of the possible effects that were mentioned in the videotaped message. The findings from this small scale study are, of course, only suggestive. The effects for the most part are only trends and even these manifested themselves immediately after the first film viewing in a two week long study. More effective may be an intervention program that assists young adults in making critical evaluations of sexual violence in the media based on what has been learned from research on mitigating the effects of TV violence on school children, a subject which we will now adddress. Mitigating the Effects of Television Vioience Recently there has been interest among mass media violence researchers in teaching children skills to enable them to understand and evaluate what is being presented on television (Anderson, 1980; Corder-Bolz, 1982). The underlying premise has b.een, in general, that the media can teach behaviors, but the behaviors can also be unlearned (Eron, 1980). Of the studies undertaken, the majority have been directed at modifying children's beliefs about television-particularly beliefs about the unrealistic nature of much of what is presented (Dorr, Graves, 6 Phelps, 1980; Huesmann et al., 1983; Singer, Zuckerman, & Singer, 1980). Most successful of these attempts has been an intervention designed to change attitudes about aggression by Huesmann et al. (1983). The program developed by Huesmann et al. relied upon some of the major empirical developments in the area of aggression and television violence-including: the notion that children learn through the observation of aggressive models that violence can sometimes be revarding or an effective solution to problems; that 20 increased identification with aggressors may facilitate aggressive behavior in viewers; and that viewing violence that is portrayed as socially acceptable or permissible may increase behavior. a Huesmann et al. reasoned that even if children encoded the violent problem solving strategies they viewed on television they might be less likely to enact these strategies if they could be convinced that they are unrealistic, inappropriate, and unrepresentative of most people's behaviors. In their study Huesmann et al. chose a sample of 169 first and third grade boys and girls who had a history of high exposure to television violence who were randomly divided into control and experimental groups, both of which received three training sessions over a 6-8 week period. Before the intervention the children were pretested for the degree to which they considered the behavior of television characters as realistic. In the first study children in the experimental condition were taught, through a series of discussions and lectures, three principles: (a) the behaviors of the characters on shows such as Starsky and Hutch and Charlie's Angels do not represent the behavior of most people; (b) the camera techniques and special effects give the illusion that characters are actually performing highly aggressive and unrealistic feats; (c) the average person uses other methods to solve their problems. This technique resulted in little change in the children's aggressiveness, as measured through a peer nomination technique (Eron et al., 1971) or as measured by frequency of violence viewing, judgments of television realism, or television character identification. Consequently, Huesmann et al. made a more direct attempt to motivate the children not to encode and later enact aggressive behaviors based on counter-attitudinal advocacy research that has been found CO be effective in producing enduring behavioral changes in other domains (Cook h Flay, 1978). In the experimental group's training sessions children vere first credited with the attitudes that the experimenters wished them to adopt. The children were then asked to make 21 videotapes for other children who had been "fooled" by television and "got into trouble by imitating it" even though they themselves knew better. Finally, the children composed persuasive essays explaining how television is unlike real life and why it would be harmful for other children to watch too much television and imitate the violent characters. A videotape of each child reading his/her essay which was then played before the entire group. This second intervention was successful both in changing children's attitudes about television and in modifying aggressive behavior. Four months after the intervention there was a significant decline in peer nominated aggression and attitudes about the harmfulness of television violence for the experimental group. Further, analysis revealed that the strongest predictor of decreases in aggressive behavior and attitude changes about the harmfulness of television violence occurred for subjects who had a tendency to identify less with television characters. However, the intervention did not significantly reduce violence viewing or the judgments of the realism of television violence. As the reader will recall, Huesmann et al. (1983) reasoned there may be three factors contributing to the likelihood that a child would behave more aggressively as a result of violence viewing: (1) the child's perception of the violence as realistic, (2) the child's identification with the TV character; and (3) the child's beliefs about society's acceptance of aggression. Huesmann et al. (1983) have devised a successful intervention program based on these three factors. A very similar set of factors may account for the young male adult's acceptance of sexual violence against women and predisposition toward aggressive behavior following exposure to violent pornographic depictions. It is possible that an intervention designed to change males' perceptions that the portrayal of sexual violence or rape in the typical violent-pornographic film are realistic, reduce the likelihood of males identifying with the aggressor in the pornographic film and change males' beliefs about the acceptability of aggression against women, as well as providing subjects 22 with descriptions of the possible psychological effects of prolonged exposure to violence against women, may result in less acceptance of violence against women and lowered predispositions to aggress after exposure to violent pornographic films. Such an attitude change program as Huesmann et al. point out may be most effective if based on prior research on cognitive consistency and the persistence of attitude change (Cook 6 Flay, 1978). As Cook and Flay (1978) note, programs of attitude change which involve making salient inconsistencies between attitudes and behavior (emotional role play, behavioral rehearsal, modeling) or inconsistencies with cognitions about the self seem to produce persistent attitude change. For example, observing one's own behavior usually causes greater and more persistent change than observing others behave in some counternormal way (Bandura, Blanchard, d Ritter, 1969; Mann & Janis, 1968); an attribution manipulation (in which the subject can only justify his/her counter=attitudinal behavior by referring to an internal disposition or motivation rather than some external motivation) results in greater attitude persistence than the simple receipt of a persuasive message (Miller, Brickman, & Bolan, 1975); and writing one's ovn counter-attitudinal message will produce more persistent change than passively reading a message (e.g., Watts, 1967). Examples of some of the long term changes caused by consistency approaches include: teaching children self-control and honesty-with effects persisting for periods to 6 weeks (Freedman, 1965; Lepper, 1973); and, reducing smoking for up to 18 months (Mann h Janis, 1968). These approaches have also produced decreases in ethnocentrism (directed toward the physically handicapped) vhich have persisted over 4 months (Clore 6 Jeffrey, 1972); and reduced phobic behavior of many kinds for periods as long as 2 years (Cook h Flay, 1978). Operationalization of the consistency approach for changing attitudes about sexual violence might include: (1) crediting adolescent males for possession of 23 the attitudes we wish them to adopt; (2) inducing behaviors that would lead to the self-attribution of these attitudes; (3) inducing perceptions of personal responsibility for an outcome related to the attitudes; (4) inducing the perception of participation out of free choice, and (5) promoting the perception that the consequences of their behaviors are important. The subjects could be asked to help (in exchange for a small payment) prepare a videotape on sexual violence. This film, they would be told, will be used in area high schools to inform male adolescents who have been fooled by mass media depictions into thinking that women desire sexual violence. The subjects could then be informed by the experimenter that he/she assumes that the subjects do not really believe the message being advocated but a younger adolescent male might and therefore get himself into trouble for imitating such behavior. Subjects would then spend time composing essays on the "myths about sexual violence" which will be read and evaluated by the experimenter and then rewritten by subjects. These essays will then be read before a video camera. Subjects could be instructed to focus on the unreality of sexual aggression as presented in the media and why it is harmful to adopt attitudes which trivialize rape and sexual violence. These instructions will be presented only as rough guides as the experimenter will emphasize the need for subjects to contribute their own ideas. Each subject's videotaped reading will then be played before the entire group of subjects so that the group may evaluate the product. This intervention procedure could be tested against a direct persuasion approach designed to teach subjects: (a) that women do not enjoy, desire or become sexually aroused by violence and rape; (b) that repeated exposure to violence against women may desensitize subjects to violence not only in the films but perhaps co the plight of other actual victims of violence. This intervention would essentially be comprised of debriefing tapes and scripts we have used in previous studies. After subjects have participated in the interventions, several 24 measures of intervention effectiveness could be taken, including: (1) self-reported sexual violence viewing (in a follow-up assessment); (2) ratings of realism in portrayals of sexual violence (i.e. , portrayals of the myth that women desire sexual violence.); (3) general attitudes towards rape and rape victims; and (4) identification with sexual aggressors. VI. Mass Audience Interventions The interventions suggested above are only practical with relatively small groups of persons in a classroom or other controlled setting. Probably the most well known efforts at educating the general public about sexual violence in the media has involved large scale projects. As Malamuth (1984) notes there have been several large scale educational efforts directed at large audiences concerning subjects such as rape and rape myths (e.g. Cry Rape, Why Men Rape, A Scream of Silence) and pornography (e.g., Not a Love Story). These documentaries were expressly created to make the general public more aware of these issues. If the effectiveness of other mass media campaigns are taken as an indication, there can be little doubt that documentaries about rape or sexual violence have great potential for informing the public about these issues if they are seen by enough people. Research evaluating the effectiveness of anti-smoking televison information spots, for example, has demonstrated that these programs have been successful in increasing public awareness about the negative health consequences of smoking (Flay, 1986). But, this success has probably been the result of the large number of anti-smoking messages delivered during prime viewing times. Most public service announcement campaigns do not produce significant effects because they consist of a small number of spots, often of questionable quality delivered at unpopular viewing hours (Flay and Sobel, 1983). In order for informational programs on rape and sexual violence to be effective they must be viewed by a large proportion of the population. Even then, the most effective program for altering viewing patterns or changing attitudes about sexual violence 25 would include both a mass media information campaign conducted in conjunction with small more focused workshop interventions such as the program suggested above in schools or the home. Programs which have combined mass media campaigns with individualized skills training sessions have proven to be quite successful in helping the onset of adolescent smoking behavior, alcohol and drug abuse (Flay h Sobel, 1983). A preliminary investigation into the effectiveness of the film Not a Love Story to change attitudes about pornography and sexual violence by Bart, Freeman and Kimball (1984) suggest that viewers may benifit from exposure to this film. Bart et al, surveyed a group of 332 males and 318 females after they had viewed this film in an art film house in the Chicago area. The findings indicated that exposure to the film resulted in changes in beliefs and attitudes about pornography (e.g., "I didn't know pornography was that violent", "The film made me angrier about pornography"). Unfortunat16, lPWlC97. Malamuth, N., Heim, H., h Feshbach, S. (1980). Sexual responsiveness of college student% to :rape 'dep%criozis: ?I&I%Itury-tid&sinMbiti"?y &Y&?&. .Soutnalc-of Personality and Sociallfisychblogy,--r 38(:3); 3B9-40Bz Mann,irL.P; -& 3anis, I.L. W.96&)..;.A follow-up.s~y,'on.~tne-L;l;otn~ser;mr~~~~f -.- ---- emotional role playing. Journal of Personality and Social Pmyz&lXnry +A& 339-342. 34 McGu+re , '&,,;Jt (1964),, +ducing+resistance to persuasion: Some contemporary approaches. In L. Berkowitz (ed;), Advances in Experimental Social Psychology (Vol. 1). New York: Academic Press. i'fcCiiire, :idtJ., $ rapageorgls, D. ,(1961). The relative efficacy of various types _ _._. .~ gf &or-tel~"ef$efense in produ?ing immunity against persuasion. Journal of Abnormal and. Sqcial. Psycholo=: 62, 327-337. Miller,I,R,FL.;, BriSkman, P,,rb-Bolaa,,D. (1975). Attribution versus persuasion a~-fg~:~~edffying.behavioy~ __~_ Journal of Persbnality and Social Psychology, 31, wF%l. _ _ .-I-,--=; Morgan, R. -GlP@,~November)., Rowzto-run the pornographers out of town (and preserve the. First,.Amepdmgnt>. -,Ma,_-&$; 38330, National Institute of Mental Health. (1982). Television and behawor? Ten years of sciencecvrogresg,-and -implfcatigns for the eighties (Vol. I,-,Summary report). Rpekville.r~c,.#etipnal Inshitute,~ofrMental-Health. Neely, J. J., Reckel, R. V., C Leichtman, H.-M; m97-r). The effect of race-of,, model and response consequences to the model on imitation in children. Journal of Social Pspshdo~5gl~;;89i,225~Z3.1, Nicholas, R. B., UcCarter, K* E., & Heckel, R. V:, (1?!1>. Im$&at!on of adult and peer television models by white and N_egro children. JoUrrML br Social P~sYFh~10~~9~~ii7~3.~a. Nisbett, R. E., 6 Wilson, T. D. (1977);'TerIing more-tnan we can know: Verbal rewtm oa =MsL~ pboaws- P8ychological~Re~w,,..81i.,:23l-259~ Palys, T. S. (1986). Testing the common vlsuom: cne social content.of video. pornography. Canadian Psychology, 27, 22-35. Petty, R. E., & Cacioppo, J. T. (1961). Attitudes and persuasion: Classic and contemporary approaches. Dubuque, IA: Wm. C. Brown. Sears, D. O., Freedman, J. L., h Peplau, L. A. (1985). Social Psychology. (5th Ed.). Englewood Cliffs, NJ: Prentice-Hall. Sherif, C.U. (1980). Comment on ethical issues in Malamut+, Heim, and Feshbach's "Sexual responsiveness of college students to rape depictions: Inhibitory and disinhibitory effects." Journal of Personality and Social Psychologyc 38, 409-412. Singer, J., Zuckerman, H., & Singer, D. (1980). Teaching elementary school children critical television viewing skills: An evaluation. Journal Gf Communication, 30, 84-93. Smith, D. D. (1976). Sexual aggression in american pornography: The stereotype of rape. Paper presented at the annual meeting of the American Sociological Association, New York City. Smith, D. D. (1976). The social content of pornography. Journal of Communication, 26, 16-33. 36 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. Figure 1. Rape Myth Acceptance .Scale Items A woman who goes to the home or apartment of a man on their first date implies that she is.tiElbn.g to have sex. Any female can get raped. One reason that-women falsely rep0uf.a rape is that they frequently have a need to call attention to themselves. Any healthy w&an can srtecessfully'.re$ist a rapist&if she really wants to. When women go around braless or wearing short skirts and tight tops, they are just asking for trouble. In the majorit@ of rapes;. tne victd&.&s promiscu0us;or has a.-bad. reputation. If a girl engages in neekihg or petting and she Iets things @Lout of hand, it is her- own fault if her partner forces sex on her. ...-..-mII____ _I - _. Women ~o~~t-r~~~~~~hite~fk~g~.t_ahat,~hey,de~rrprr, -- --.-_ A woman who is stuck-up and_'nh$aks.'&e ;4%~&00;g~dlS~,C_~~;EZr;lguys on the street deserves to be taught a lesson. Many women have an unconscious wish JO be raped, and may thea unconsciously set up a situation in which they are fikelp to be attacked. If a woman gets drunk at a party and has intercourse with a man she's just met there, she should be considered "fair game" to other males at the party who want to have sex tith her too, whether she wants to or not. What percentage of women who report a rape would you say are lying because they arc angry and want to get back at the man they accuse? What percentage of reported rapes would you guess were merely invented by women who discovered they were pregnant and wanted to protect their own reputation? A person comes to you and claims they were raped. Row likely would you be to believe her/his statement if the person were: . . . your best friend? . . . an Indian woman? . . . a neighborhood woman? . . . a young boy? . . . a black WOIIUI? ? ? o a white woman? APPENDIX I This project is a multimedia educatioaal program des.&ed;.as an intervention strategy to help young people understand the nature and influence of sexually violent media and the ncitur&`o~+e%uiLIy Gvioienti behaviqc +and its impact `on individuals and society. Tatgc t Group me audience yt have determined to be mast at risk from sexually violent media are seventh through tvelfth grade, fyt166 e~&&tt .phrs .df .age 233 seventeen. This age group has the skills and developmental maturity to understand the concepts and yet are still in the f&&fft+ei $~&ig&s: air&i-bi-kely:tb assimilate new-att&&s. both. ~~ic:-4y. and.e,yyicq5aive- .* The men in this age group aIso shape G r a,nd-.$acb, their &*,ua?$,m$(urit)J' -$d'~l~~~ty* Ii :,q+i,.ide'&~Jy-yi$h $ers df `Ijoth'a.d~~&k'k' `&+ wor&a-~~8aabLiih thiir role e:' A .: ; T-r':;! ;?" .-1,y--. . . . . -. - 35 ..fi : , ; I : 3-,r:: ..( 7. i : 1.. . . >:,I,,.:_ ..z `-. :::. It is our intent t0 create a program'?t&t --can-- ach~ev6:`:aCEe~ta~ce.: in the school, church and home vith this,ra.r;get gr_oup. I, : 2: -y : c. t;r;. r,ri 5 , . . . I F 5. x >T < .-. c, L ; _ . i _ . . . 4. To change vieving habits; s-; :c; .<; ,:,+I 5 A presentation of the messages in myths reinforced by sexual violence in the media. This module portrays the nature ~~~~~ d~~~-~~ i n g a= misconceptions. Students discuss t eir ideqs about myths and reality and ;c-c, monitor their ovn attitude changes. 6 6s; -,rj $; LZ;; p':`i*`l se6947i. C.I' Interviews of victims and offenders of sex&' %~l%&&ipt%U%%J~t~~~sis for discussion. The impact of sexual violence is taken out of the realm of the media and into the real world for c~d~d"s'c~~~~'andL-di~~Sdofi~~'Sj -Y .i 0 4 EIDEETE OF. l'RE EDUCATIONAL CRWICDLDl4 Our project vi11 use both exposition, and interaction in its curriculum approach. `This ;u+kes the program a dynatiic and individualized process depending hiariry 00. strident input. It is respoa8ive to current programming and ,cu+&t `fir-4' &i&&o-.offeringt, aL any given. tiq- and. allows for the exptessioa "of ' i&iv'idu%l `%ntatests, insights and needs of each partic-ipating group of young people. The teach$r'r guide offeSs direction in presenting the expository. sections and of fe+s quest%% and: : Toll&through `.-act ivit~iia~." It.` iIso -.provides various dyne$c istf4@$si;;~ . i _ such `& siti'latiotis; iiii~rovijht~iohg ad oFher kinds of open-ended :$dttvitiee Fo facilithtie awar&nis'a.+ .>Theae arcivit%,es also provide the oppo:$tfkiCp- `foi r&debts L tb' `6oAitor