a; ”9." “1.41.1“... . e nun}, 4 i ”Esfiun13;...ohw.r .l .. 41.21.; .mnrufi ,. ‘ . ( .dwm :1 v1. :54. 1.. 56...; c... . fight“. .«navg an" 19:3. {03. an. Jlfik-J.‘ . l unwary .. . .152I..:uu.>le.¢9.u...x . «43$ : ,, 7b.... . 1 4: . ‘ . ‘ rdutfilhhfl .. 3.2.... i! .35.... . xiii?) $.13. . 1: )x: . .5 2.3.3.. 4 2 £35.“. a {.32}; .3: .3 fluxhwfidhuuz. 11% q . ... viii... t.) . £3,314”. , . . . 5a.” fl... ‘mfitLirh “a ., 31“,”: 1'H3~.&a [TY LIBRARIES lllllllllll "T” ll \\\\\\\\lllllllllll “IL 1293 01405 7578 l This is to certify that the dissertation entitled The Perception and Effect of Degradation and Violence in Commercially Released Videos presented by Jack A. Glascock, Jr. has been accepted towards fulfillment of the requirements for Ph .D degree in Mass Media Major profess r Date 3/ [1/ ‘75 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY MiChigan State University PLACE II RETURN BOX to remove this chock” from your record. TO AVOID FINES Mum on or baton date duo. DAIEKDJJE DATE DUE DATE DUE L__J l | | MSU In An Nfimaflvo ActioNEqml Opportunity Intuition mm: THE PERCEPTION AND EFFECT OF DEGRADATION AND VIOLENCE IN COMMERCIALLY RELEASED VIDEOS BY Jack A. Glascock, Jr. A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Telecommunication 1996 vi a9 ir we ex fa wa be ma Su as Vi Co: hit Sul Cor the mak ABSTRACT THE PERCEPTION AND EFFECT OF DEGRADATION AND VIOLENCE IN COMMERCIALLY RELEASED VIDEOS BY Jack A. Glascock, Jr. This study examines the contributions of degradation and violence in videos on male viewers’ subsequent verbal aggression. Stimuli were selected based on pre-test ratings in which video segments were rated on overall dimensions as well as male and female character attributes. Subjects were exposed to one of four film conditions in a fully crossed factorial design (degradation x violence). Verbal aggression was subsequently measured by assessing subjects’ retaliatory behavior toward a female confederate. Subjects exposed to materials rated high in degradation were more likely to support a female confederate’s reappointment as a research assistant while subjects exposed to portrayals rated high in violence tended to be more critical of the female confederate’s performance. While subjects' responses in the high violent condition were in the predicted direction, subjects in the high degrading conditions seemed to respond contrary to expectations. A possible explanation is that these subjects may have felt the depicted degrading behavior was morally wrong or improper and therefore guarded against making an overly aggressive response. ACKNOWLEDGMENTS First, I would like to thank the members of my dissertation committee, Dr. Lucinda Davenport, Dr. Steve Lacy and Dr. Todd Simon for their unwavering support; Dr. Bradley Greenberg for his invaluable advice and Dr. Robert LaRose for his thoughtful critiques of previous drafts. In addition I would like to thank members of the staff and faculty at the University of Windsor, in particular Dr. Charlene Senn and the Humanities Research Group, for their assistance and support during my stay in Canada and without whose help this project would not have been possible. Finally, my deepest gratitude to Cynthia and Campbell for their love and support during these past years. I am forever indebted. Li In Li Met R831 DiSC TABLE OF CONTENTS List of Tables ........... ........................ .. ..... vi Introduction .........e,. ................................ 1 Literature Review ....................................... 6 Usage, Content ..................................... 6 Policy Considerations .............................. 12 Behavioral Changes ................................. l6 Aggression ................................... 16 Aggression Procedures ... ....... . ...... .. ..... 34 Attitude Changes ................................... 39 Degrading Materials .. ......... ............... 39 Sexually Violent Materials ................... 50 Summary and Hypotheses ............. . ............. .. 54 Method ..... ............................................. 60 Pre-Test of Stimuli Materials ...................... 60 Pre-Test of Provocation ............................ 74 Main Experiment ......................... . .......... 82 Results 0 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 O O O O O O ....... O O O O O 85 Discussion .... .......................................... 95 Conceptual Analysis ................................ 95 Measurement/Procedural Limitations ................. 106 Policy Implications ............................ ....108 iv Suggestions for Future Research .................... 111 Appendix A - Description of Film Segments ............... 114 Appendix B - Pre-Test and Main Experiment Questionnaires ........... .......117 List of References ...... . ......... .. ................ ... 132 Ta Ta Ta Ta Ta Ta Ta Ta Ta Ta Ta Ta. Ta] Ta} Table Table Table Table Table Table Table Table Table Table Table Table Table Table 10 11 12 13 14 LIST OF TABLES Aggressiveness and Exposure to Erotica ...... 18 Pornography and Attitudes Toward Women ...... 40 Film Clips Evaluated ......OOOOOOOOOO ....... O 62 Factor Matrixes for the Film Evaluation Questionnaire ...... .... ..................... 64 Mean Scores for Film Evaluations - All Conditions ............ ................ 67 Analysis of Variance for Film Ratings Using Only Film Segments Selected for Main Experiment.......... ..... ............ ........ 72 Pearson Product-moment Correlations for Factors and Items on Film Evaluation Questionnaire .........OOIOOOOOOOOOOOOOO ..... 73 Factor Matrixes for the Provocation Scales .. 78 Results for Provocation Pre-Test .... ........ 80 Pearson Product-moment Correlations for Factors and Items for Provocation Pre-Test ......OOOOOOOOOOOOOOO......OOOOOOOO 81 Factor Matrixes for Experimenter/Experiment Form and Feelings Questionnaire ............ 87 Pearson Product-moment Correlations for Factors and Items on Experimental Questionnaire and Feelings Scale ........... 88 Results from 2 x 2 (Violence x Degradation) Analysis of Variance ....................... 89 Post Hoc Analysis Comparing Degrading, Violent and Erotic Conditions ...... ........ 93 vi effe "erc exp] (deg incr Ther migh mate exam porn nonv Spec larg. INTRODUCTION In the 1970s researchers became interested in the effect erotica might have on aggression. Some forms of "erotica," which at that time were defined solely by sexual explicitness with little regard for actual themes depicted (degrading, mutually consenting, etc.), were found to increase aggression by provoked males against other males. There was some indication that aggression against females might be even greater after exposure to certain nonviolent materials, however this line of research evolved into examining violent materials as opposed to nonviolent pornography. While attitudinal and perceptual effects of nonviolent materials have been studied, behavioral effects, specifically aggression of males against females, have largely been ignored. Researchers have found that subjects exposed to nonviolent, but degrading pornographic materials have reported more sexually calloused and aggressive attitudes toward women (Check and Guloien, 1989; Zillmann and Bryant, 1982). These materials have been typically described as portraying women "as eager to accommodate any and every imaginable sexual urge of any man in the vicinity" (Zillmann & Bryant, 1982). More recently, feminist authors have identified various themes present in degrading materials, several of which they argue may have more detrimental impact on viewers' attitudes and behaviors than the "availability" theme 0 It clarify the fem unequal (Steine delinea materia rated v availab. or comm; wishes) these tl availab: partici; another "Object: female a treatmei human cl femalets male's 5 Cox theme, 5 Donnerst degradin 2 theme characterized above (Cowan and Dunn, 1994). It may be appropriate at this point to attempt to clarify what is meant by degrading pornography. Central to the feminist criticism of pornography has been the unequalness portrayed between the male and female characters (Steinem, 1980). In a recent study, Cowan and Dunn (1994) delineated eight degrading themes found in pornographic materials. These themes, based on a content analysis of X- rated videos, ranged from "availability" (the female is available to anyone) to "domination" (the female is insulted or commanded to perform sexual acts without regard to her wishes). Cowan and Dunn (1994) argue that common to all of these themes, with the possible exception of the availability theme, is the notion that the female participant is less of a person than the male. For example, another theme described by Cowan and Dunn (1994), "objectification," is characterized by the male treating the female as an object or plaything in sexual relations. Such treatment is seen as depriving the female participant of human character or identity (Check, 1985). As a result, the female’s status as a person is lowered relative to the male’s status. Cowan and Dunn (1994) argue that the availability theme, which has been used by other researchers (i.e., .Donnerstein et al., 1987; Zillmann, 1989) to describe (degrading pornography, does not fully convey the inequality and subordination concepts that are integral to the other 3 degrading themes. The criticism is that women’s display of "unbridled sexuality" (p. 12) is what is described as degrading as opposed to any perceived inequalities or subordination. However, such characterizations have been theorized to result in a loss of respect for women. Weaver (1987) argues that traditionally women perceived as sexually permissive have been accorded a lower level of respect than other women. In other words their status is lowered relative to other women. Whether or not their status is lowered compared to male participants similarly portrayed (available and promiscuous) has not been addressed in the research. Commonsensically, one might expect male promiscuity not to be perceived as degrading or resulting in a lowering of respect for males. Males perceived as promiscuous are often described as "studs," which, if anything, connotes an enhanced status. Such terminology is in stark contrast to the general perception of promiscuous females as whores. In addition, since pornography is primarily a form of male entertainment, one would have to assume, that, in general, the portrayal of the male participants, with whom males would identify, is not perceived as degrading. In either case, one can reasonably make the argument that the status of the female is lowered, either relative to the male participant or relative to other women. This lowering of status is also characteristic of the degradation process which has been described as the lowering of the position, rank or standing of a person (Jarvie, 1991). Degradii lowers 1 availab: Wh: be cons: tacit er find no: feminist degradir to the n 1980). A false be 0r beate (Hill, 1 to lead 1980). Sex aggressi However, and some Ceniti, 1994). A Colleagu that has behavior. (Donners1 4 Degrading pornography, then, can be defined as that which lowers the status of the female participant and may include availability as well as other degrading themes. While depictions such as the ones described above can be considered degrading in and of themselves, it is their tacit endorsement in pornographic materials that feminists find morally objectionable. This endorsement, from the feminist perspective, is carried out by representing degrading behaviors as appropriate or as providing pleasure to the male and, even worse, female participants (Longino, 1980). As such, these depictions are seen as perpetuating false beliefs about women, i.e. they enjoy being humiliated or beaten; or they expect to be treated as sexual slaves (Hill, 1987; Longino, 1980). Such attitudes are hypothesized to lead to acts of sexual violence against women (Burt, 1980). Sexually violent materials have been found to increase aggression against females (Donnerstein & Berkowitz, 1981). However, only a few studies have been published in this area and some of them have produced inconsistent (Malamuth & Ceniti, 1986) or alternative findings (Fisher & Grenier, 1994). An additional issue is raised by Donnerstein and his colleagues who argue that it is the violence, not the sex, that has been shown to cause antisocial attitudes and behaviors after exposure to sexually violent materials (Donnerstein, Linz & Penrod, 1987). However, no study in a refereed journal has attempted to compare sexually violent and vi 1 disent violer initiz certai seems study conpor on the nateri sexual (1) th (2) su Radtke which 6 Compare reSulte male no leve to be C and violent materials.1 The intent of this study, then, is to attempt to disentangle the relative contributions of degradation and violence on aggression toward women. Given recent policy initiatives in the U.S. and, in Canada, the censoring of certain pornographic materials, research into this area seems particularly relevant. One procedure included in this study that hopefully will help sort out the various components is that the stimulus materials will be classified on the basis of a pretest in which subjects will rate the materials on the appropriate dimensions (violent, degrading, sexually arousing). Hence there will be verification that (1) the stimuli belong in their respective categories and (2) subjects can distinguish between the categories (Senn & Radtke, 1990). 1 Donnerstein (1984) reports a couple of studies in which aggressive and sexually aggressive representations are compared. In each study the sexually aggressive materials resulted in a higher increase in aggressiveness on the part of males subjects than the aggressive films. However since no levels of significance are reported these findings have to be considered somewhat tentative. to exe 90% add rep wer and hav Cane adul ente disc; Vide( 6 LITERATURE REVIEW Usage. Content It appears that a majority of adults have been exposed to pornographic materials at least once in their lives. For example, Bryant (Bryant & Brown, 1989) reported that over 90% of adults and high school students had looked at or read a sexually oriented soft-core magazine such as Playboy. In addition, 69% of adults, males and females, in the survey reported having seen an X-rated film. Similar percentages were reported in a survey for Time magazine with 84% of men and women reporting having seen nudity in a magazine and 62% having seen an X-rated movie ("Pornography," 1986). In terms of frequency of pornography usage, a 1985 Newsweek-Gallup survey, cited by the Meese Commission, found 37% of adults sometimes bought or read magazines like Playboy. Similar percentages have been found in other surveys. In the Time poll 43% of adults reported having seen nudity in a magazine in the past year. In a survey of Canadian attitudes toward pornography about a third (32%) of adults surveyed reported reading or leafing through adult entertainment magazines (Peat, Marwick & Partners, 1984). Percentages of yearly usage of X-rated videos have ranged from 7% to about 38%. One reason for these discrepancies may have to do with the rapid diffusion of the videocassette recorder (VCR) during the early and mid-19805. The Hefleyeek survey reported 7% of adults had gone to an X- rate vide 1986 rate surv male the the oppo surv ("PO porn that maga in t Mosh read Surv. that than the r9901 men 1 know] Nichc 7 rated movie and 9% had bought or rented an X-rated movie or videocassette in the past year (Press et al., 1985). In a 1986 survey, almost 25% of adults reported having seen an X- rated movie in the past year (Smith, 1987). In a more recent survey of college students Stock (1991) found that 27% of males and 11.9% of females had seen an X-rated videotape in the past year. Stock's figures may be somewhat lower than the norm since her respondents were college-age students as opposed to adults. The Time survey found 38% of adults surveyed had seen an X-rated video in the past year ("Pornography," 1987). It also appears men are more frequent consumers of pornographic materials than women. A recent survey reported that more men than women, in all age groups, sometimes had bought or read soft- and hard-core sexually explicit magazines and had bought or rented an X-rated videocassette in the previous year (U.S. Department of Justice, 1986). Mosher and MacIan (1994) also found that males reported reading more pornographic magazines than females. In a survey of Canadians, Check, Heapy and Iwanyshyn (1985) found that males reported more frequent viewing of adult videos than females. Stock (1991) found males used pornography on the average of once a month while most females (54.2%) reported never having used pornography. Not surprisingly, men reported gaining a greater amount of their sexual knowledge from pornography than females (Duncan and Nicholson, 1991). eX] mai am of fer get mec in pox The do the has con tha not st i For 901 The: Port Corn] as are 8 In addition, 25% of males reported trying or experimenting with some of the behaviors depicted in the materials they had seen in pornographic materials (Bryant and Brown, 1989). Such experimenting may be at the expense of the female counterpart. In Stock’s (1991) sample, 27% of females reported having been upset by someone who tried to get them to do something they had seen in a pornographic medium. Finally, most males use pornography by themselves or in groups (Stock, 1991). Females, when they do use pornography, tend to use it in conjunction with a partner. These findings lend support to the notion that when females do expose themselves to pornography it may often be due to the urging of a male counterpart. The less frequent exposure of females to pornography has been attributed to sociological factors as well as content characteristics. For example, Burt (1976) has argued that females are socialized to behave like "ladies" and to not show an active interest in such diverse forms of sexual stimulation as pornography. In addition women's access to pornography may be limited by social taboos against females going into adult bookstores. Examining adult magazines, Thomas (1986) suggests that one reason females may not find pornography oriented toward females exciting is that, compared to erotica provided for men, the models are neither as attractive nor as sexually or sensually suggestive. (They are rarely shown in states of partial undress or fondling their bodies.) 9 Perhaps even more to the point is that since most pornography is oriented toward male consumers, women do not buy or otherwise expose themselves to it because they do not like the messages presented. This is suggested by research by Senn (1993) who found most women to be critical of pornography and that the overall impact on their lives was negative in terms of harms perpetuated against them either physically or emotionally. The author also found evidence supporting the notion that women do not enjoy viewing pornography geared to males, but occasionally do so because materials that portray a more positive sexual imagery are generally not available to them. When exposed to X—rated materials intended for women, in contrast to materials intended for men, females have reported more sexual arousal and enjoyment, more identification and less negative affect (Mosher & MacIan, 1994). In sum, it appears that a significant portion of the population have viewed pornographic materials and about a quarter to a third of the adult population is exposed to pornography on a somewhat frequent basis. It is also apparent that more men than women are users of pornography, primarily because most pornographic materials are intended for men as Opposed to women. In the past, most content analyses of pornographic materials focused on sexual violence. In an analysis of cartoons and pictorials in Pleybey and Regeneeee magazines, Malamuth and Spinner (1980) found sexual violence (rape, 10 sadomasochism or exploitive/coercive sexual relations) in 10% of the cartoons and 5% of the pictorials in both magazines in 1977. In addition the authors found that sexually violent pictorials, but not cartoons, had increased over a five year period, 1973-1977. However Scott and Cuvelier (1987) found the number of sexually violent cartoons and pictorials in Playboy to be extremely small, on the average of 2 per year for cartoons and 1 per year for pictorials. These authors also found that the number of sexually violent cartoons and pictorials had actually decreased during the five-year period following Malamuth and Spinner’s study. In an analysis of pornographic novels Smith (1976) reported a third of all episodes in adult paperbacks contained force or coercion by a male to encourage the female into unwanted sex. In a study of dial-a-porn services, Glascock and LaRose (1993) found 13% of the recordings contained sexually violent acts, most of these involved sadomasochistic behaviors such as spanking or slapping. However, more violent acts such as rape or bondage were not found in their sample. Most recent studies have focused on X-rated Videocassettes, perhaps because of their increased visibility and accessibility due to the VCR. Palys (1986) and Yang and Linz (1990) found sexual violence in about 5% of all scenes in X-rated videos. Cowan, Lee, Levy and Snyder (1988) found 23% of all sexually explicit scenes in X-rated 11 videos contained acts of physical aggression. The difference in these studies is due to the base unit of measure used, sexually explicit scenes versus all scenes. Yang and Linz (1990) found no significant trends, increases or decreases, in sexual violence in videos produced from 1971 to 1986. In all of these studies males were found to be primarily the perpetrators and females the recipients of aggressive behavior. A more recent concern, expressed by feminist writers, has been the degrading portrayal of women in pornographic materials. Such depictions are objectionable to feminists in that they are seen to support an ideology of sexual inequity by presenting women as subservient to dominant males (Cowan et al., 1988). Such portrayals may be as common as violence in pornographic materials. In a survey of X-rated Videocassettes Cowan et al. (1988) found that 26% of all sexually explicit scenes were characterized by exploitation, in which one participant was "clearly using [another] without consideration of the used person," while 28% were characterized by dominance, in which one person controlled the sex act using verbal or physical force. Most of the dominating and exploiting was found to have been perpetuated by males. In dial-a-porn services, Glascock and LaRose (1993) found subservience to be one of the more frequent themes (38% of all sampled recordings). Subservience was defined as "when one party was actively encouraging or soliciting 12 sexual behaviors primarily for the satisfaction of others" (p. 317). Duncan (1991) also found more scenes (18.1% of all scenes) in a sample of adult videos contained acts of degradation (as defined by the Meese Commissionz) than acts of violence (13.6% of all scenes). In X-rated videos, it appears that degrading behaviors toward women, without violence, are as common as violent behaviors. It also appears that a significant portion of the adult population is exposed to these types of depictions, many on a somewhat frequent basis. The potential effect these portrayals may have on viewers, then, can be considered somewhat consequential in light of the substantial and somewhat frequent viewership of X-rated materials. Policy consideraeions In the United States and Canada, research into the effects of pornographic materials on attitudes and behavior has played a prominent role in policy initiatives as well as legal analysis. In the United States the Meese Commission was established during the administration of President Ronald Reagan and given the mandate to "explore possible ’ In its Final Repogg (Department of Justice, 1986) the commission defined nonviolent, degrading pornography as that which "depicts people, usually women, as existing solely for the sexual satisfaction of others, usually men, or that depicts people, usually women, in decidedly subordinate roles in their sexual relations to others, or that depicts people engaged in sexual practices that would to most people be considered degrading" (p. 331). l3 roles and initiatives the Department of Justice and agencies of local, state and federal government could pursue in controlling, consistent with constitutional guarantees, the production and distribution of pornography" (Department of Justice, 1986, p. 216). Relying heavily on findings from experimental studies, the commission concluded that exposure to violent and, somewhat more tentatively, nonviolent pornographic materials might lead to antisocial beliefs and/or increased sexual aggression against women. As a means of "addressing the expansive scope... of the pornography problem" (p. 465), the commission published a list of 92 recommendations. One directive, using the Racketeer Influenced Corrupt Organization Act (RICO) to prosecute obscenity violators, was proposed because, it "could, in appropriate cases, virtually eliminate a large- scale pornography operation" (Department of Justice, 1986, p. 519). Under the RICO statutes, if a defendant is convicted of an obscenity violation, the forfeiture of any properties (bank accounts, inventories, buildings, yachts, etc.) involved in or derived from the business becomes mandatory (U.S. Code, 1988). Two cases illustrate the rather severe consequences of a RICO prosecution. In one case, after finding magazines and videotapes worth $105.30 to be obscene, a federal court ordered the forfeiture of a business (three adult bookstores and eight videotape clubs) with total annual sales of $2 million (U.S. v. ngbe, 1990). In another case a court ordered the forfeiture of a business 14 that included 13 adult bookstores and video stores as well as almost $9 million in personal assets (Alexander v. Ihermburgh, 1990). Such forfeiture judgements may have, as one defense lawyer asserted, "profoundly intimidat[ed] anyone who would test the waters of sexual expression of any sort" (Hansen, 1992, p. 28). In R. v. Butler the Canadian Supreme Court became the first court worldwide to explicitly accept the argument that certain types of pornography harm women (Lewin, 1992). Under Canadian law, in order for a work to be found obscene it must clearly depict the "undue exploitation of sex" (Canadian Criminal Code, 1985). Central to the determination of what is undue exploitation is the "community standard of tolerance" test which, until recently, was primarily concerned with what Canadians would find not tolerable for other Canadians to be exposed to (R. v. Towne Cinema Theatre LEQL, 1985). In Butler Sopinka J., writing the majority decision, incorporated the idea of harm into the community standard test: "The courts must determine as best they can what the community would tolerate others being exposed to on the basis of the degree of harm that may flow from such exposure" (R. v2 Beeler, 1992, p. 484). Sopinka J. divided pornography into three categories: (1) explicit sex with violence, (2) nonviolent, explicit sex that is degrading or dehumanizing and (3) explicit sex that is neither violent nor degrading. The court noted that "among other things, degrading or dehumanizing materials 15 place women (and sometimes men) in positions of subordination, servile submission or humiliation" (R_ye Butler, 1992, p. 479). Sopinka J. then concluded that the first category would almost always constitute the undue exploitation of sex, the second category would usually do so while the third category, explicit sex, would not. Citing the Meese Commission among others, Sopinka J. noted that there was a "substantial body of opinion that holds that the portrayal of persons being subjected to degrading or dehumanizing sexual treatment results in harm, particularly to women and therefore to society as a whole" (R. v. Butler, 1992, p. 479). This linkage of pornographic materials, violent and nonviolent, with harmful effects has set a somewhat ominous precedent in Canada (Kramer, 1992). For example, in a case decided less than six months after Butler and relying on the court’s rationale in Butler, a lower court upheld the government’s confiscation of gay erotica as obscene. The court found these materials degrading and dehumanizing, describing them as encounters "without real meaningful human relationship" and not "compatible with the proper functioning of society" and as such, inferring harm (glee Dey Bookshep, 1992). As noted, research into the potential effects of violent and degrading pornographic materials has had significant consequences for public policy in both the United States and Canada. In the United States this research 16 has been used to justify increased prosecution of the distributors of pornographic materials. In addition, the Draconian penalties exacted under the RICO statutes have reportedly served to "chill" sexually explicit expression. In Canada, these social science studies have been used to support an outright ban on sexually explicit expression that is deemed violent or degrading by the government. Behavioral Changes A s 'on. Research on the effects of sexually explicit materials has evolved from an arousal model to a more cognitive one. Much of the early research was based on a model of excitation transfer proposed by Zillmann (1971). This model VH“. describes a situation in which two arousing conditions occur in sequence. Arousal to the first does not dissipate immediately and can be transferred to the second condition, thereby adding to the arousal caused by the latter. So, for example, if a male is provoked, then exposed to an arousing communication and then given a chance to retaliate against his provoker, the strength of his retaliation will vary in accordance with the arousal elicited by the communication. In this model it is the arousal as opposed to the content that predicts the resulting aggression. What is meant by aggression in this paper is "any form of behavior that is intended to injure someone physically or psychologically" (Berkowitz, 1993, p. 3). 17 The excitation-transfer model was supported by a number of subsequent studies using either films (Meyer, 1972; Zillmann, Hoyt & Day, 1974) or textual passages (Fisher & Harris, 1976) as stimuli. (For a summary of these studies and others discussed in this section please see Table 1.) Other experimenters, notably Baron (1974a, 1974b), reported contrary results--that subjects, first angered and then exposed to erotica, subsequently exhibited leee aggressiveness toward a confederate. One discrepancy in these studies was the difference in stimuli used (Baron & Bell, 1977; Donnerstein, Donnerstein & Evans, 1975). All the studies in which aggression increased used films or text as stimuli while studies in which aggression decreased photos were used. Donnerstein et a1. (1975) reasoned that exposure to photographic stimuli may have resulted in a lower level of sexual arousal than that of the films and text. These authors theorized that exposure to erotic materials could result in an "attentional shift" in which a previously provoked person’s attention could be diverted, resulting in less aggressiveness. However if the level of arousal was great enough, this attentional shift could be overcome, resulting in increased subsequent aggression. Several subsequent experiments were generally successful in finding reduced aggression to be the result of exposure to mildly erotic materials (Donnerstein et al., 1975; Baron & Bell, 1977). In a study involving females Baron (1979) found subjects exposed to more explicit sexual 218 mGOHqucoo mufimoaxd on Ho «chasm: Ga duos» Gaza mxoozm whoa Umuo>waop on: Gofluwpdoo owuoum ca omega can» mxoonu wuofi pmum>flamp cofiuapcou unwaofl> ca m_m Hanudwd Gdnu mwmcoammu xoonm omcwucw duos cw pmuanmou nuan3 m>fimmmuomc Gaga mmmGOQmmu xoonm mucwucfl wuoE :fi pmuHSmmu cofiuapcoo Owuoum .uudmudn ououopoucoo umcwoOd mxoonm on: 0» madman Gw>wu Gong .mEHflu ou pomonxw .wuduopwucoo Eouu mxoonm >n Dmumocd m.m mumnwpmucoo on» nmwada ou mxoonm mm: o» mocdno o co>fiu can» .msaww ou pancake .muoumpmucoo Eouu mxuonm ocfi>woomu >n pmuoucd m.m flufldflummfl cofluwvcou muzmonxd on aw wum3 Ho udwfiumm Eafiu .cwEnv Hmuuamc no ..ma>oE .odum.. oeuouw .Auwudmfiaumnxm Sh umflmnumsn mdB nufin3 .u£u«u wuflcxv mucmpsum made we unmaofl> memfl> m.m Houuamc away .m>wmmwummm >2 U030aa0u adamaoud umoa Edam ufiuouw on» pw3onm Handoud ham umou lung 0 “mucmfiumm Sawu .cwa hum AHoGOHudoapm. Houusmc uo A>uflpsc deEdw sue: u0w>mnwn Hduaoowuav Ufluond ..unuflw amend udda0w>v muchSDm made mm o>wmmmu000 p03®fl> u.m Ambmflv Hw>wz Aabmav GGdEHHMN H magma dedtwdww IUOIfiQ5W hunch 219 muumnndm coupons icon unosd pcdou 0H0: mmocoumwwfip on “mopwam Houuswc 0» 60969500 muomwndm vmummad macaw mwwuwmcmucw xoonu uo m~0>ma um3oa Ga pmuadmmu odd onwmsoud duos mu woody mumB mopwam owuouw one UGdOu oum3 m.m pouadmcw icon no pouadmcw cmw3umn mwocmuouuflp uo mmucwumuwwp xom oz .Gowuwpcoo aduudmc aw muomnndm can» Am20wuozao>0 co ouoom acuou. w>flmmwuouo whoa we: macho owuouw one mcowuapcoo uwnuo Ham vaowoxm Goduwpcou uwuoum u0u Handout .ucowuwvcou nonuo Ga uuumnndm dds» omwoc ouduucfi whoa vmuw>waoo dawuwpcoo owuoum Cw u.m mUHSIOl ouuuwpoucoo ou uxuonm Ho>waop ou wocdno d cw>wo con» pad wasfifium any ad xooH Cu poxmd .mxuonm pad mcowudnao>m o>wudomc ha .uoa no .pwuwmcd mu03 m.m Monoummmwu mo Haw3 mm uddfiauwaxw pubesam>o con» .mwfiuoum wounu uo wco pwmu .uwwzm uo>oo moosooadfl an no ummnm um>oo onwuaducfi Gd >8 Do: uo poxo>oua muw3 m.m .GOMDQflHMDmH pad mMSmOQXo .GOwuduo>ouQ Hound pad cofluooo>oua 0u0wmn cwxMD oumu uuown pad whammmua pooHn no Mohandas “wuduwpmwcoo ou omaoc umumwcwavd ou mocdno d c0>wa can» .mpddfiumm Edam ou vmmoaxm .mmwoc m30wxoc Gaw>flmomu >3 Umxo>oun ouo3 m.m .flmmfiflmmmw .uum .>Mmcmom .muauwdHSN mo mquDUMQ Houuaod Mo Ndmfiddm Bouu cdxdu dwEOB muss mo umuauown mucmesum mama ow 0H Masada um owxooa u.m Amowuw>wuuo Hauwoo odd aduaoowumv ofiuoum Ho m>fimmmuumd .Hduusoc a new» m.m mquUdum made Hm pad wadfiww mm anu HUCOMDoospm dd Eouu Qaao xn meoHHow oumB mGOMDMUGoo Had “umudooumucw pad >daauuow adaxmuououon uo pmuuwmcou Esau owuoum “mumwsawMu Edda .awaum uwwouu can ucdaow> .o>wummua00 mucopduu mama om .Hduusoc ou Dononxv u.m ummaauwmu .flmmmwmu AMmeHV nonmm Amsmac magnum w “mamas Avboav .mmo a axon .cddfiaawn Nmmmm lil! l utn! I Use OXGUOGOGLU :4 u.m.. ho .76.~¢D:d 0.483 l.m GDCQUJDE 0:2: on Ca ufl Dmx00~ Him \Qth: 20.1.5 lfidflm In duflduudufl u ado “.415! u «slam w! N630! 2C) m.m >uucdncoc mcofid UGSOu mum3 mmocmumuufip 0c “mouona muoo Cu whamonxm Hound poocdncm pan mouona mxdowmmn ou whamonxw umuwm twosvmu Gmemwuumm m.m >umcm umxduuwmn Ho Houudoc Gena ocwumSUmflp whoa muod vouch omad m.m “mononu wuss Mo wxmowwwn .Hmuudwc can» have mcficflfidxw houud Dowdouo duos magma pmuuomou m.m afloum Heuuswc Cdnu mazoum have use mxdowmwwno aw m>fimmouOOd mmwa m.m >Hocd “asouo Houudwc ou pmumneoo azoum muod GM w>wmmmuomm mood m.m >Hucdlcoc “GOwuflpcou Hauuzm: any can» mcfimaouu duos Donna muw3 acoMufiUcou Had Aom.AaV o>wuoomwodw UGdOu mm3 Gowudasaflcda Human “anuo Hduusw: coca 0>Mmumuuod mama unawuwucoo mound pad wxduwmoo£o cw u.m .umwmuflm wumuovmucoo ou mxoonm Owuuomaw umumwawfipm ou wucdno cm>wm Gwnu Dad wazfiaum ou wmeQXm .pouwmcd yo: yo .mGOwudew>m >umeComqu w>fiuumwn n.0udhmvmwcoo maofiow >3 Uwuwocd m.w mxuonm unwuw>wamp >n mumuwpowcoo any umcwmmo mmmummc ou vm3oaad can» pad fladfiwum may cw>flm .uoc uo .mcoflumsam>m w>fiummmc cm>fio mum3 m.m mudumpmwcou o» mxoonm uo>wamp ou mucdSU d d0>wo nonv pad fiazewum onu ud xooH ou pmxud .mGOMDdst>w 0>wudooc >3 .uoc Ho .pmHOUGd ouw3 m.m Hag mmfiuw>fiuum Hmnxwmouwumn cw penance meQsoo Ho AHHdedN Eouu Gmxou mound wads .lmouoco .mxmoumwh.c madam UCMSDMD :« G08 ..uuo .uud .>chmuu no mummpSDm mHmEMu mv nobona ca um Umxooa u.m .>uw>fluod Hmsxwu wo mcoauuwuouwp vma>u 0H Damn Ho .uuod awaken uwowaqxo pocwducoo mouonn no nobona 003d wadsww ..mxmommwmzo. .amuudwc mucmpSDm made mm 0H um pmxooa m.m quxudm Eouw auxou mouonn mean Ho mucoswmfiuum>pd mdfiuoOQE Eouu cdxdu umuduowa .oxdoouoono. ..uum .>umdoom wo nuuduowa Huuuaoc mudwpaum odds on on an UmxooH a.m Hanna 4% AhbmHv Ambmav Conmm Hawm a coumm Andhmav QOHdm unammu 21 muowwuo on poocmpn>0 u.m vwnmocdacoc “GOnunUcoo adnuaoc dds» unnudoaa pad ucnunoxm mnofi mu Um>nwonwa Oman 0nm3 unenunpdoo 03» when» “unmannomxw Cu uucu>ondd mmma pmunoawn muoo pad mouse uo mouosn ou pmmoqxw m.m pmnwucd cenunpcoo Ahmed: wadewuuoadev .ocnmsond >Hnmn3. can an pmmdmnodn canmuonomo .GOnunpcou unamonxo 03D nuuwo Umuaamcw dn03 muooh3dm dm33 “donumonOOG puoavfln monona ops: .umnnu wouasudw onm3 muoon3du 6033 mHHJMHflfl mopnau o3» muwSHm>o Cu Umxmd 0nw3 m.m cwnu “mudnmpmucoo m3» wudsad>o Cu Umxmd con» pad nadfinum on meoaxw .Umnwmcd uoc no mGOnuomDUUd mmHMu >3 condone onm3 m.m mudnopowcou on mxoofiu nw>namp ou mucdnu d dd>nu Gmnu .vmuaamcw uoc no mouonn ud unnxooH nouud no on0u03 nonunu poDHSmcn on03 uuoow3dm .flmmflmmmmmm mucousum wade mm 35638 33 .3 mmmgflummu monun>nuod Hosxuu mSOwnd> an command nuansoo no dose? moss .uaonnmudfi Hanusmd no uouona on U03wn> m.m Hmnusmc A >unpzc oHdEwu A >unpdc oamfiudefiww nun; dnmuuda penumnxo pmonHOw Hduaond condonudw unmouna d amODOSQ >uw>nu0d addxom pmnHQEn nun? $05G odds nondsww no muse GHQEUu .3onuauc ud owxooH u.m ”Hilda” Abbmav >xwaoaom a GddEHHnN Amhmav mcd>m a :nmumnoddoo .dnoumnodcoa Nummmu 222 m.m pmnmocuncOc now DGSON wnw3 muomwwm 0c “unnmdonm pad acnnnsumne names“ snnu >Hco mm3 Eanw manuuoa w3u “wumnmpwwcoo Unc3ou n0n>unm3 >noudnHMDwn an ouuwnocn poocmpn>o Sana manuuwa ou pmmonxm m.m Dmnmocd m.m Dwnmucdnco: macaw endow onw3 neoconmuwnp on “noomwd 0>wuoumc 303 odd Homdond awaken 30w: unonnm on endow nafifinum on unamoaxw nouud 0>nmmonmmd mama u.m Umnmu:< mmmmmflfl pw3wn> Eanu 033 mo monunamdv w>nuomwud 03u pmuo33m>m muumw3dm swan “wudnopmucoo mudsau>m on woadnu Gd>nu can» .uoc no «Hafinuu ou pmuOQXm .Umnwocd uoc no mCOMDMuDUUd mmadw >3 UmnmmCd oan m.m wumnmpwwdoo xuosm on mucdnu u Gw>no Conn .Qsonu Honucoo ou Docmnmmc no .mQOnmcwfinp w>nuomuwo anode Dmuon onm3 30n33 MHSEnum o» Demomxm .pwnwOGd Doc no .ucowudsad>0 >unndcomn00 o>nu0moc >3 pwnwmcd ono3 m.m mummmmmmmm .mcns v uso3o mm3 Eanu 300m mo cowuansp 03» «G303m unaduncmu nuns Kan Mono no maunoo no G303m dwadundwu nun: maunoo .ucnuuwa .>unpsc .Uwuononcoc .mnamonxd o: "mGOnunpdou w>nw mo muown3om dame own mao an poodau onw3 m.m GOnunpcoo mndmomxo on 6 Oman mm3 anon» ammHuEww can mmadfi pmnuoHo >33dnunda pad @03u030 "denuded Avv .msocnandddu .Go«uo3ndumdfi dame "uncommon 0>wucumc Amy .msmcnanndsU pad enumHHOM "uncommon ucmam>n3fid Amy .OMDdHHou .mmnsoonoucn .uanGGOM Houncwm Hd9u55 ll III“ rimommdawuwlw>mmeon AHV mwflme OH NO muwm mucmpzum 0366 mm nSOu m0 060 vm3ow> m.m ”magnummm, .dalnu Anson. aaaennnn a sxmnoomm Amsmnc mung: \>// \ \ Mammmui\\ 3 2 0000003 030 uwzo 0nsmm0nm 00033 030 p0um3uc3n0>o m.m p0n0mCd “£030 0» 30050n0 0nOE 000G003>0 can 0030003Q030 0n06 00 3338300 U0unom0n E033000060000 pad >03303um03 on U0000X0 m.m 00300 no 030>03 n03033 000:00na .03u0n0ucoc no 03uon0 n030033 .no0uw0 0>3uou0c pad 30090nd 3033 00c33600 Donn 3325300 on 0nsmoaxm 0.m 00xo>0nacd n0u UGdOu 0n03 muu0wu0 on “00308 G030 om 0nOE Dad 0:30003Q030 .oc33n50030 >330500 00 0833M 0030003 pad usu300 030 00>300n0a 003ds0u “0c03030000 30nucoo no mc3uu0a c3 0.m c030 c3030 033du3c0o 3033 03u300 ou onsthX0 n0uwd 6030033du0n n0ud0no p00c003>0 0.m 00xo>0nm 53% uwSU 0.000n000wc00 0003mc3n0>0 0» 000030 G0>3m 003» 000 3335300 00 00moax0 you no 00mOQX0 .uudu 0n5000na 00033 mc3u03w03n0>0 >3 D0n0OCd 0n03 m.m 063nm 03a 000330>0 00 00x00 0n03 m.m >33003u .0u0n000uco0 03p 00 0030: 0:03xoc n0>330p 03300 03 30333 03 0600 0>3u3u0QE00 0 unnndp D0u3nmc3 03000 003 000w3dm 030 G030 .c03unpcoo 0ndmoax0u0c 0 G3 00003Q no 3355300 03D Cu p0momx0 .0u35003 303n0> pad C0uunn3 >3 U0n0ocd 0n03 0.m 0003 0n03 0000n000uc00 03080u uQ00X0 .0>03d 00 0500 «mmuummmm 00000500 0306 ov 00000300 0306 en mu00n3du 03dfi0u cm 3 acnumnap 03 0003036 w 0n03 muc0Eo0m E33u 03» “8033000060000 oc3u03a0p muc0em0m E33u no >03303um03 003>30>G3 muc0Em0u S33w .G0503 0030 no 000039 m3 00303> 0.m 0300n0ncoc no 03uon0v mm0Gu3o33Qx0 305X0m 000 33033 no 303v 300:0n0 .A0>3uoo0c no 0>303monv uu0uu0 >3 D03n0> 0n03 muc0Eu0m E33u .Gnficm “30nucoo 0 n0 0003030co0 E33w a no 000 on 00003000 0n03 0.m 0>030 00 0800 dale-dam A3mm3v nu0>nmu a 05.30 .5333 3?: $032 a 3358 2 .uc0>nm .GGME333N \ Ap.ucoov .3mm3. addfi333u a >xn3onom g 211 05330 30n0303 00 00000x0 0000n330 00030 3030 0>3000n000 0n05 0n03 05330 0300n0 00 00003X0 00003330 00n0030 “0030500 3033 03300 0.35 :05 030030 30 030>03 n03033 00n000333500 0305 3033 00n30Q 0000n330 00n0030 “0305 3033 00n30Q 00030 3030 5330 0000 00 0n30OQX0 n00u0 30030n0 mo 030>03 n03033 003 000n00 030503 3033 00n30n 0000w330 00n0034 0.0 00xo>0naa3o3 03050 03300 0n03 00030n0um30 03 “0303030300 030 n0300 33 0.0 3030 00303 30 030>03 n03033 00n0>3300 533w 0300n0 030 3000 003 033 0.0 0030>0n3 u0>3000n000 A 30n0303 A 0300n0 mo 3n0000Q 030 00303300 05333 00 30030n< a 000n000w300 0033000 000n000 00 003030 0 30>30 3030 030 3335300 030 00 00003x0 .00n0030 003 no .000n0000300 030500 no 0305 0 50nu 030030 030 030300330>0 30003n3 >3 00n0030 0n03 0.0 000n000w300 00 00303 n000333500 00 003030 0 30>30 3030 030 .00xo>0na 003 no .000n0003300 Eonu 00303 0303 0 033>3000n >3 0030>0nm .3335300 00303> 0.0 $0000 A>n003053000 03330333. 5330 30n0303 0 no >03303X000503 030500 00 3303 00 00n300n0033 3030 .30no 033003000 5330 0000 0 n03030 no 00300300 0305 m0 03330 .335uv 00303> 0.0 3330 5330 003335 hum A00n300n0033 33 0000030 033300. .0300n0 no .0303w0u3nav .0>3000n000 00300300 030503 on .30n0303 0 00303> 0.0 a 3.3% 300030 000nn0m 0 33000n03303 83: 30003330 .3. w §HHMN . HOUfiU.\ \ .L ..l. _ . 030000 030 0>3300303< dwwfiIOK >3 00n0030 0303 0.0 dHfiUOUGHH 00:00:00 030E 00 IHUIfime li«.l E.»\ Clgfi .335 h U0303> nubflmflzm .001me 30003» E0330: 0 23000n0220Q 3.0me 255 05330 30n0303 030 0300n0 o0 00n03500 000n00 030500 0033000 303000n000 00000n033 3330 030303> >3303X00 o0 0n30o3X0 un0.0 00n0030a303 n00 0005 003 0330330 n033530 0 “00030>3000n000 00000n033 030303> >3303X00 >330 .000n00 3950 533 03300 0.0 n00 “00030300n 0>3000n000 00000n033 5330 030303> >3303x00 030 0300n0 .000n00 0305 0 3033 00n303 0.0 n00 “05330 30n0303 00 0030 3030 n0000n0 003 05330 030303> >3303X00 030 0300n0 030 00 3003on3 0000n00 0305 00 00n03500 030500 0033000 030030 00 030>03 n03033 00030on3 5330 0300n0 030 00 0n3003x0 .>03330no330 303000n000 030000 030 33 300303330n0330 303000n000 3003 n0000 030030 00 030>03 n03033 000300n3 05330 0300n0 030 0>3000n00< Mlmlmlua .3 61200 no 03050 000n0000300 0 00 030030 n0>3300 00 003030 0 30>30 3030 030 3335300 030 00 00003x0 .00n0030 003 no .030030 030 030300330>0 30003n3 >3 00n0030 0n03 0.0 n0003 .335 03 030 033303> 5330 n0000 >3000300553 3003 000n0000300 30030 00 003030 0 30>30 3030 .33on0 30n0300 00 00303000 no 3335300 00 0000330 .000n0000300 030500 no 0305 >3 030030 030 030300330>0 30003n3 >3 00n0030 0n03 0.0 04000300 30300n30 33 .0335 v 03030 003 5330 3000 “5330 030303> >3303x00 0 no 5330 30n0303 .5330 00300300 0305 om3 0300n0 30 00303> 0.0 .30330 0333 039. 50n0 03330 5330 0>3000n000 no >03303X000503 030500 030 300 3030 .30n0 033n30000 03330 5330 0000 00300300 0305 cm .335 b 00303> 0000n330 a «3 30003. 33000n03300 30003. 50330: a 33000n03303 Hflmwm .. .. II 11:41 lit.-. LCCC I 12¢ 03.00026 .5330 030.070 ...:330 06330 0300n0 00n£0 33¢ >3 30030030 0n03 I.0 00300300 I35: 00 30n0303 0. 300303) 0.0. 003303503 .0 20000n303hwn0350 - . . \ 30.330 03 033000000 10100 03.30 003.: 0.0 \ a a u.‘ .\ .\ 265 0000000 03003033030 02 300000 030 033>0w30 00 00003300 03 030500 030 30333 33 5330 030303> >3303x00 030 00 0n30o3X0 n0000 >330 00030>3000n000 0n05 0003003>0 00003330 00n0030n3o3 “05330 030303> >3303X00 0o 003>0 3003 00 0n3003X0 n0000 00030>3000n000 0n05 0003003>0 0.0 00n0030 “5330 30n0303 030 3030 30030n0 0n05 00030330 05330 030303> >3303x00 030 0300n0 03B 05330 030303> >3303X00 0o 003>0 3003 00 0n3003x0 n0000 00030>3000n000 00000n033 0003003>0 0000n00 030500 3033 00n303 0.0 “033030n0 >330300 00 0000n 0n03 05330 0300n0 00n30 334 Hudmflflfl X0033 030m 00 0003330333 “000n0000300 030 0033000 000n000 00 003030 0 30>30 3030 030 000n0000300 030500 0 50n0 030300330>0 0>300003 >3 00n0030 0n03 0.0 0n3003X0 0003 n0000 3003 030 .3335300 00 0000330 0n03 0.0 030500 0 0>0330 003 000n00 030 030 00033033 003 303030300 00n0030u303 0 0030 030030 0>o30 00 0500 000n0000300 030 0n030n no 30030 0 n0>3300 00 003030 0 30>30 3030 .3325300 030 00 vomoaxm .000n0000300 030500 no 0305 0 >3 30300330>0 n003 0 030 030030 >3 00n0030 0n03 0.0 IMHflflmmHmw 003n03 3003 00n30 0 n0>o 05330 x30 00303> 0303030300 0n3003x0 33 0.0 00300300 “303030300 0n3003X0no3 000300n0n0033 no 030303>303 >3303X00 0n03 03o .030303> >3303X00 033 330 .00305 No 00 00303000 0.0 00300300 0305 om 0>030 00 0500 300303 33 0003335 m 003 5330 3000 “03300333533 300000 030 00330 30503 030 30333 33 2330 0503o3> >330nxmm 0 no 300000 030 0>0w30 30503 030 30333 33 5330 030303> >3303x00 .5330 0300n0 .5330 00300300 0305 cm 30n0303 0 00303> 0.0 03000300 dwamduu 30033. 303000 0 nus203mz 30.0couv NUM30¥me aw CfimumeflnCOQ 33mm3v 003303n00 0 33000n03303 Hulda 00 @3030 aucerflfln m ”“4111‘ 0 >2 $03 Hln‘l‘ll ‘Fll“ 00:0 0n03 0.0 WHflUUUfiWN IZT TIL-\C3E37w 63.13 m .UUCQCZamw EdCE NN UWWWfime v£>3w fl . :Cw y wt—hrfv 6:0 00 30533000 a.m WWII..." 0W «V0030 hfifltthuv Q hmflCfiflk NW1d” 127 30030 03n00030 0300 00 00030 0000m330 N 030 000n0000300 030 3033 30030 00 00030 n .30300330>0 030 3330 00 00030 0000n330 0 .0330000 n030000030 30300330>0 030 3330 no 30300330>0 303n0> 0 n0>3300 .030030 03n00030 n0>3300 "0303030 00n30 30>30 030 3323300 030 cu vwmoQXm .30300330>0 0>300003 0 030 030030 03n00030 >3 000n0000300 030500 0 >3 00n0030 0n03 0.0 .H OOHQ 03 3 00 033500 30330 0 33 03303300n .303033030 00 030 0000335330 0n03 0 .00300300 0305 mm mmmflfimmu 033030 0>303003 0 3033 >330n0o3no3 030303> 00 0305000 003335 u0>30 0 .303030300 030 00 00303000 0.0 mwmmmmm Avmm30 n0330n0 0 n03033 H0300 28 activity did evidence an increase in aggression. Subjects in this condition also reported more disgust after viewing the stimuli leading Baron to suggest that a person's affective reactions to erotica might mediate subsequent aggressive tendenciesziépecifically, if arousal is labelled in a negative manner (disgust) later aggression will increase since a person's anger or annoyance should be enhanced; if arousal is labelled in a positive fashion, then subsequent aggression should be reduced: However, studies designed to support the arousal-affect model have yielded inconsistent results. Zillmann and Sapolsky (1977) found erotic photos, which were perceived as more exciting and pleasing than neutral stimuli, reduced reported annoyance with the experiment, but had no effect on retaliation against the experimenter. White (1979) found that sexually explicit slides rated low on negative affect decreased aggression, but erotic slides rated high on negative affect failed to increase aggression. Several subsequent studies have found increased aggression after exposure to arousing, disturbing erotica, but no effect on aggression for arousing, pleasant erotica (Sapolsky §:EEEEmEiEl’1981; Zillmann, Bryant & Carveth, 1981; Zillmann, Bryant, Comisky and Medoff, 1981). In some of these studies disturbing erotica was described as containing sexually violent—(sadomasochism) and/or degrading be 'ors (bestiality) ZillmannfiiBryant & Carveth, 1981: '“Zillmann> Bryant, Comisky and hedoff, 1981), lending support to the notion that the message (violence or degradation) of 29 these types of materials may lead to increased aggression. An alternative explanation for the findings discussed above is offered by Feshbach and his associates (Feshbach & Malamuth, 1978; Malamuth, Feshbach & Jaffe, 1977) who argue that subjects could have been disinhibited by exposure to materials commonly tabooed and hence this disinhibition carried over to aggressive behavior which is also tabooed. In effect sexual arousal does not lead to aggression; it is the reduction in sexual inhibition that does so. Support for Feshbach’s arguments comes from a series of studies in which subjects, not provoked, exhibited an increase in aggressive behavior after exposure to erotic materials (Jaffe & Berger, 1977; Jaffe, Malamuth, Feingold & Feshbach, 1974). However, other researchers have not been able to duplicate these findings. Studies that have included a condition in which subjects were not provoked have reported null findings (Baron, 1974a; Baron, 1979; Cantor et al., 1978; Donnerstein & Barrett, 1978; Donnerstein et al., 1975; Sapolsky & Zillmann, 1981; White, 1979; Zillmann & Sapolsky, 19770) or, occasionally, less aggression after exposure to sexually explicit materials (Baron, 1974b, Baron & Bell, 1977). To summarize the findings up to this point, it appears that exposure to erotica can increase aggression if subjects are provoked and experience negative affect. However, when provoked subjects experience positive affect during exposure to erotica, aggression is inhibited, typically resulting in null findings. In general, exposure to nonviolent erotic 30 materials, without prior provocation, has failed to produce aggressive urges. It should be noted here that erotica simply meant any form of sexual explicitness, which could range from nudity to intercourse to sadomasochistic acts. Films rated as disturbing, and shown to increase aggression, included those containing violence as well as degrading acts such as bestiality. In some cases, what was described as negative affect, then, was also violent and degrading. Also up to this time, researchers had only examined same—sex (usually male-male) aggression. However, a more policy-oriented approach has proven to be whether exposure to pornography could result in increased male aggression toward females. Beginning in the late 1970s Donnerstein and his colleagues undertook a number of studies to determine the effect of pornography on aggression against women. In an initial study Donnerstein and Barrett (1978) found that while angered subjects paired with a female confederate reported more arousal after exposure to a stag film, they aggressed less than subjects paired with a male confederate. The authors theorized that perhaps social disapproval of aggressing against a female might have inhibited responses toward the female confederate. In a follow-up study, when male subjects (both angered and non-angered) were exposed to a stag film and given a second chance to aggress, aggressive responses toward the female target increased significantly in comparison to a male target (Donnerstein & Hallam, 1978). Even though the erotic films used in these two experiments contE oral (197s passj that Othex simi] above 1980] mater 31 contained only non-violent content such as anal intercourse, oral sex and female homosexuality, Donnerstein and Hallam (1978) reasoned that the depiction of women in submissive, passive roles might have contained "subtle aggressive cues" that in turn elicited increased aggression against women. Other researchers have suggested that sexual activities similar to those described in the non-violent condition above be characterized as degrading to women (Russell, 1980). Hence one could infer that nonviolent, degrading materials might lead to increased aggressive behavior. Subsequent experiments have shown that even non-angered subjects, when exposed to sexually violent pornography, exhibit an increase in aggression toward a female target compared to a neutral or erotic film (Donnerstein, 1980; Donnerstein & Berkowitz, 1981). The rationale for these findings can be found in the work of Berkowitz (1974) who has shown that individuals can take on an "aggressive cue" value if they have been associated with observed acts of aggression. So, for example, showing a female as a victim of aggressive acts, should increase the aggression-eliciting properties of females in general. As a result, the viewing of such portrayals may increase subsequent aggression toward females. This seems to be especially true when the aggression portrayed is shown as having positive consequences. In Donnerstein and Berkowitz’s (1981) study, when the rape victim was depicted as enjoying the sexual assault, even non-angry subjects later exhibited more aggressi1 to either Apparent the sexu: inhibitic the posi‘ aggressi( the nega‘ someone z with this aggressi‘ In a found no subjects week per: 0f time ; aggressic had asses Stimuli t DonnerstE I Berkowitz prodUCes COgnitiOn tendencig certain I 32 aggressiveness towards a female target than subjects exposed to either neutral, erotic or negative ending films. Apparently the positive outcome ending (the female enjoys the sexual aggression) justified and reduced subjects’ inhibitions about aggression. Among provoked subjects, both the positive ending and negative ending films increased aggression towards a female. Provoked subjects exposed to the negative ending video may have been predisposed to hurt someone and the association of the assault victim’s pain with this disposition may have evoked a heightened aggressive response (Berkowitz & Donnerstein, 1981). In a more recent study, Malamuth and Ceniti (1986) found no effects on aggression toward women after exposing subjects to violent and nonviolent pornography during a 4- week period. One reason for this might have been the length of time (1 week) between subjects' last exposure and the aggression evaluation. Previous experiments finding effects had assessed aggression immediately after exposure to stimuli (Donnerstein, 1980; Donnerstein & Berkowitz, 1981; Donnerstein & Hallam, 1978). Malamuth and Ceniti, citing Berkowitz (1974), note that exposure to mass media usually produces only short-term effects because its influence comes from the temporary activation in viewer’s minds of cognitions that may intensify individual behavior tendencies. Before behavior can be again intensified in some future point in time, the message must be reactivated, i.e. certain "retrieval cues" must occur in the individual’s 33 environment. In other words, subjects would need to be placed in situations similar to those depicted. Fisher and Grenier (1994) designed an experiment to test the ecological validity of studies showing aggression against women after exposure to sexual violence. In this experiment, which was similar to Donnerstein and Berkowitz's study (1981), subjects were given three options at the end. They could evaluate the female confederate by delivering electric shocks (the method used by Donnerstein and Berkowitz); they could give a verbal evaluation; or they could skip the evaluation altogether. Given these alternatives, most subjects (86%) in the study chose the non-aggressive response. While the results of this study seem somewhat plausible, they do not lend themselves to easy interpretation. For example, subjects could have chosen to skip the evaluation simply to get out of the study early. While this may indicate a lack of aggression, it may also indicate these students had better things to do than to continue participating in an experiment once they had earned their credits. Based on a relatively limited number of studies, it appears the effects of exposure to violent, sexually explicit materials on aggression may be somewhat transitory. For non-angered subjects, aggression may be increased after exposure to sexual violence depicting positive consequences. For provoked subjects, aggression may be increased regardless of ending. In any event, it appears that the 34 violence is the key ingredient, with the positive ending and/or provocation serving as further disinhibiters. Aggression procedures. The provocation-retaliation sequence used in the previous experiments utilized several different techniques to induce anger as well as measure retaliation. Most commonly, shocks or some other physical measure of aggression (loud noise) were used for both the provocation and retaliation phases (Baron, 1974a, 1974b; Donnerstein & Barrett, 1978; Donnerstein & Hallam, 1978; Meyer, 1972; Zillmann, 1971; Zillmann et al., 1974). Other studies used some form of verbal aggression for both the provocation and retaliation procedures (Fisher & Harris, 1976; Sapolsky & Zillmann, 1981; Zillmann & Sapolsky, 1977). Still others used a combination of the two (typically a negative evaluation, then shocks or noise; Baron, 1974b, 1979; Baron & Bell, 1977; Donnerstein et al., 1975; White, 1979; Zillmann et al., 1981). There are advantages and disadvantages for each method. The advantage of using physical aggression is that it permits the most direct measure of physical assaults (Baron, 1977). However one problem with this technique, which usually requires the subject to give shocks to a confederate as part of a "learning" exercise, is the possibility that some subjects may select higher shocks, not out of a desire to inflict pain, but to help the confederate learn the subjects presumabl 1974). Ban aggressi verbal c serious is commc subject: in a fa] indirec4 as such or inhi) is that aggressi HOWeVer, but in a verbal me they havg OCCur to Anot Potential 35 materials rapidly in order to avoid future shocks (Baron, 1977). Another problem Baron points out is that of demand characteristics. Some subjects might think they are expected to deliver the same number of shocks as they received. In addition the use of physically aggressive measures to anger subjects has been reported to be a frequent objection, presumably of human subjects committees (Zillmann et al., 1974). Baron (1977) notes several advantages to using verbal aggression measures. Since they involve only ratings or verbal comments there is usually no possibility of any serious harm to participants. Also, since verbal aggression is common in everyday life, these procedures provide subjects with an opportunity to aggress against the victim in a familiar manner. Finally, verbal aggression involves indirect rather than direct attacks against the victim, and as such is less prone to the influence of strong restraints or inhibitions. Of course, the drawback of verbal techniques is that they provide a less direct measure of physical aggression than do shocks or other physical assessments. However, verbal aggression measures do measure aggression, but in a different form. One requirement for interpreting verbal measures of aggression as actual aggression is that they have to include the possibility of some type of harm to occur to the victim (Baron, 1977). Another procedure is the evaluation-shock method. Potential problems with this design were investigated by 36 McDaniel, O’Neal and Fox (1971) who found that mode of attack (shocks or ratings) made no significant difference in attacked subjects’ negative evaluation of a confederate. However the attacked subjects evaluated the confederate more negatively if the mode of evaluation was the same for both subject and confederate. Thus, the authors argue, subjects are more aggressive, and less inhibited, when the opportunity for retaliation is less obvious, i.e., the same as that used for the anger instigation. Verbal measures used in aggression studies have generally followed the procedure introduced by Berkowitz, Corwin and Heironimus (1963). These experimenters gave subjects a page from a multiple choice intelligence test and had them complete problems on the page as quickly as possible. In the provocation condition the confederate explained the test in an extremely condescending manner, in general insulting the intelligence of the students at the university. The confederate also badgered and insulted the subjects while they worked on the problems. Immediately after exposure to a film stimulus subjects were given a form with questions evaluating the experiment as well as the experimenters. Zillmann and his colleagues have used a similar procedure in studying aggressive behavior. In one study subjects first interacted with a polite experimenter in doing a relatively easy task (Zillmann et al., 1975). Next subjects were treated rudely by a second experimenter while financial aid the ‘ subjects (Bryant a The laboratox addressec rsported aggressic hiqher lg 1969; She 1971) . n PriSQnerE higher 16 9t al. ( high in BOUndn p‘ Shocks an 37 performing a more difficult task. Later subjects were asked to evaluate the experiment as well as rate the experimenters. Subjects were told the ratings of each experimenter were to be used in determining his future financial and academic support as a graduate student ("To aid the departments in determining stipends for research assistantships, we are including he following questions to be forwarded the experimenter’s major department."). Other aggression studies, not necessarily involving erotic materials, by Zillmann and his colleagues have employed the same general procedure but with slight variations on how subjects were verbally provoked by the rude experimenter (Bryant & Zillmann, 1979; Zillmann & Cantor, 1976). The question of whether aggression found in the laboratory is indicative of real-world aggression has been addressed in a number of studies. Several studies have reported significant correlations between observed aggression in naturalistic settings and the deliverance of higher levels of shock via an aggression machine (Hartmann, 1969; Shemberg, Levanthal & Allman, 1968; Wolfe & Baron, 1971). For example, Wolfe and Baron (1971) found that male prisoners with a history of highly aggressive acts delivered higher levels of shocks than male college students. Shemberg et al. (1968) found that male and female teenagers, rated high in aggressiveness by their counselors in an I'Upward Bound" program, chose to deliver significantly stronger shocks against a confederate than subjects rated low in CC ‘39 prc the Ver pro dis; 3991 aggr real aggh and e Kinne aggre. 38 aggressiveness. A more recent study by Malamuth (1983) found that subjects who held attitudes favorable toward aggression against women (high scores on Burt's (1980) Rape Myth Acceptance and Acceptance of Interpersonal Violence scales) exhibited higher levels of aggression toward a female confederate in a laboratory experiment. Construct validity of laboratory measures of aggression was demonstrated by Carlson, Marcus-Newhall and Miller (1990) who examined results of such measures reported in more than a hundred publications. Over all the studies examined, these authors found that variations in the experimental conditions consistently led to expected differences in the laboratory aggression measures used. Regardless of the procedure used, it appears the provocation-retaliation sequence works most effectively when the same method is employed for both parts. That is, a verbal retaliation measure works best when the provocation procedure is also verbal. While there are advantages and disadvantages for using either measure, verbal and physical aggression are two quite different concepts. Verbal aggression measures verbal aggression and can not realistically be considered a measure of physical aggression. However both modalities are a form of aggression and each can have detrimental effects on recipients (See Kinney, 1994, for a discussion of the effects of verbal aggression on receivers). 39 Attitude Changes Degrading materials. In the 19805 the emphasis of research in pornography turned to examining the effect pornographic materials might have on attitudes and perceptions toward women. (For a summary of studies discussed in this section please see Table 2.) A catalyst for these studies were feminist writings contending pornographic materials were degrading to women and female sexuality (Zillmann & Bryant, 1982). In an initial study involving nonviolent pornography Zillmann and Bryant (1982) described these materials as portraying females as "anonymous, panting playthings that men liberally exploit for sexual gratification" and as "eager to accommodate any and every imaginable sexual urge of any man in the vicinity" (p. 12). These authors found that subjects massively exposed to this type of pornography, compared to those intermediately or not exposed, estimated various adult sexual practices to be more pronounced, considered pornography less offensive and objectionable, recommended a shorter prison sentence for a rapist (trivialized rape) and reported less support for the women’s liberation movement. In addition, male subjects exposed to massive doses of pornography evidenced greater sexual callousness toward women than males in the other two conditions. Zillmann and Bryant’s study has been criticized on a number of methodological grounds, including experimental design (Gross, 1983), the use of unrealistic stimulus 4() hnaouoochoa ou Ummomxo m.m >3 mam>ma Hm3oH xaucdowuficoflm cw vouuoamu wuw3 umcuuna Hosxmm m.w:o nua3 mmmcflaoon ou unacflouumn MEMuH cwsoz vu63ou mmwcmsoHHoo Hmsxmm Houowuo vmuwnwnxm vmmoaxm >Hm>wmmua m.m ofinfi acauwvoo ca “umflQmu a uOu wocmucmm conflua unoflcwa «Hos pmpcmafioomu can quEm>OE Sofiuduwnwa m.cmEoz HOu uuooadm mmwa vmuuoawu .mmauw>fiuoo andxwm SOEEooGd unwowuooun muazuu wo womucwouwa Monown vmunaaumo >namu00duoa ou nmmoaxm >Hm>wmmufi m.m dudJHQfl “memo“ hav oamom woodwaacm accomumm uo >u0ucm>cH mason CoEoE nMMZOB mmwcmsoHacu Hmsxwm uquEw>oE Cowuuumnfia m.cw803 HOu quQQDm “umfiawu w uOu oocwucwm comwun a no CoflumpcmEEoumu “muofl>unmn Hudxmm COEEOUG: find GOEEoo ocfiowuooun muaaod wo mooucwouma mo Godunfiaumm dduflflflflfllddflUGHQifl cofiuwoo new mmusoouwucfl Hand .xwm Homo OGHHSuwwu mompfi> counn-x vm3mfl> m.m coauflocoo >£Qmumochoa on» a“ .mmwvaoo cofluosuwm >8 0030w> m.m coflufipcoo Houucou mquvSum on» cw “oOAqu xwo3 ucoc can muchSum w u uw>o madam «0 undo: oamewu can wads ooa m 0» vmmoaxo muo3 m.m oofluma xmwBuXMm m Hm>o Amsaww ofiuoumncoc wmv whomomxm on no Amuoon x pad my muwfivmfiuquM .Amuson m usonnv w>flmmME boondocfi musmoaxm mo mam>mq .mmudoonwucfl Hood can mmufiooumucw .mnocwHfiCGdo .oHunHHmw onwpdaocfi mmfiufi>fluod Hosxwmoumumn mo mfiaflu Um3ww> m.m mucmpdum oHdEwu om pad made om umuufianu «Janina Ammmflc uac>um a acceafifin Aummav unusum a causaafln Human CQEOB UHMSOB mm 3UHUH4 U36 3 mH OGHO m manna 41 nonuo zoom Eouu umuuwv uoc pap gown? Aamuusmc pad mono on .mmfio o>wmmwfiuwaucocv madouo moHSu Hwfiuo on» non» mxoonm uo mam>ma umnufis pouo>flamv cowuwpcou mono w>flmmw8hma any :« m.m .umwanu u MOM oocwucom comfiua uo3oH m vmvcwfifiouou muomhndm muc0H0w> vmuauwuouo pad poouwou:mamfi .vmuomfiumcatoadewm “mHmHQwaw oHSmOQanqu 0>ammfisndm >HHuSXmm pad mzosomMEOuQ >aannxwm oxwa ouofi one: >o£u uonu om mumon wadewu o>wmmdfihmaucoc mo anewuamouoa vauuwnu mcofiuwpcoo Hmnuo Ham Cw mamflumuus .maufluouufi Hmuunmc ou commafioo “acauwvcoo Houuaw: cw m.m Gdnu m>wmmwfiuoa ouofi an muowa madfiww 0>Hmnfifiuwa-noc no>aoouon macauwvcoo xua Husmcmmcoo and xwu voududumnw Iwadfiou cw u.m can: 04% wumumpmwcoo maofiwu m ou Umumumficaevw mxoonm wo >uwmcmucH ummeu HOw mocmucum comwua pad mmOMEdU >hdumcofi UwvcmEEooou can Aoaou 6cm uasmmmo Hwofimhnav madauu Hmowa 03» flown m.m mufimomxw hound Oman “mm>wuomnvn vm uo umfla a co mwvwam chamomxmuumoa No you can madmonxo unawwn mudHQwam mudu ou vwxma cmnu .munaafiwxm m>wunuudc NO you cm>flo m.m NI‘ mmso o: no w>wmmflEqu scoc .m>flmmweuwn moms wameww u now£3 ocflusv mwpfiam owuouw ho mmvfiam MUCNUSUW 0H9: 0v HMHUfiQC U030fl> m.m oocaaow> Hmzxwm .xmm vouuwoo uoana .xmmnkumuwumdw :0HMEMN .xmm andmcmmcou .Houu50G. mucmfiumm aaflw uHOSN 0>ww mo moo ou vomonxu m.m mudmpaum wamewu can cans om mg «daily Ammmav acaxwa w pumcowq AhmmH. Ho>603 a 42 owpa> owzacuuocuoo on» 0» vomoaxm m.m wdoc>OOHUGo ho omvw> Houuamc on» o» vomomxu muomnndm quay monouowu Havamhna m.uouc®8«uwnxw on» unond duos Umaanomu pad Awo.vmv Houcwfifiuonxo maceww u o» HomoHo mufimso ufimnu 0>OE ou UmUGmu .pouo>fiuofi >Haczxmm once on o» pmovzn oum3 ompfi> Uflsacumocuoa may ou meoaxm m.m Umm>uaxwm mafiasomcz ucmpflocw o£u usond Sunny gnu mcwaamu mu3 mam pony muoaamn uwonu vmmomuocw pad uadcmmo on» usonc sash» mnu onwaawu no: Ewuow> 000M 0:» unnu wwwfiawn .muuonndm mans Ummoouuwv wasfiwum uomnno xwu ou ouamoaxw .9 CK nw3wfi>uwucw oaosww a uo mmHSunww Hoowm>£Q wo Hauoou mo wEHu cofluumwu can Haoomu vac mocuumab accomhmaumucw .cowuo>fiuofi Hosxwm pwUsHocfl mwuzmmmfi ucwvcmawo u>houcm>cH waom xwm son on» Co Umummunmua mnt m.m wuwucd0wumonu Emuwum a Cu Gnommou on pwxmu amnu pad unopwocw wash 6 mo mcoflmum> mzoMHo> dw>fiu m.m me Houo van monsoouwucfi mafivsaocfl ompfl> ufloflaaxm >aa0fixwm no auuudwc mucmvsum made om o Hmnuflw ©w3mfi> m.m maflnmcofiuuamu accomuwn wmoao USU Gmemmuumo vaaaocw voaaua mquUGoo Monuo “muomnno Hoaxmm mo c0503 UGMuUMQmw mono mcfipaaocw moflwam mo mama Houm>wm «0 mac ou vwmonxw 0&03 m.m mucwwaum vaufiwu mm can 0a68 mm Hula—”fl «39m Aommav MEN 6 EOZIm H Ngvmnvz Ammmav GQEHOU w cwmsuncwpom .umhz an 43 60wmuw> unnamwv nucflucwmcoococ can» Humsouo:mcfiucwmcoococ ou oudmonxw Hound on Ou uad3 uoc pap >0£u audxwm 080m ma«op Mo vouch unwon >onco vases cwEo3 uo wooucwouma uwfiown unmsonu whomows wand uo poonflamxfla no saw: unwuoom m.m “ceauwpcoo umdbmap nonwucmmcooco: GM m.m can» huoum moon N wound cw whamcoaa Euuoa> wuoE 00>«wuuon mGOwuoflQwv Humsoudnocfludwmcoucoc chow: 0&3 m.m Ummoaxw uoc muuwnndm unawfiau Gaza Hoa>o£mn 0>wmmmumuu haacaxmm duos vmuuoamu can c0203 pun3ou mmmcmsoHHnu Husxwm mo mam>wa Honofln Umocmow>0 .mcflacu mo GOOSAwawH Hmuooum w vmuuoamu >£chmocuoa mafinwchdnmv ou 00mOQXm muwfidmcoo >£Qnuuochoa ucwnvmuw “moon xmm vaHOu pad much u0 poonflaoxwa uwuomuu vmuuoawu macauwocoo ocwuficdfidnmo odd unmaow> aw canon Suaveuono>ma no can: onwuoou n.m umfldmflflfl onu>fi ouch a“ uuwwaon Houocmu unwchmocoo wHfiQQGOMummnv 60H03mcd u.m was» ucw3oHHow “mocowumaxw u.:6803 no anodmmmuafid uwmnu ouoowpcfl 0» wamo pad mmuaoouuucw oawucwmcoo >Hacnuafi no ouch uo mowuoum whom: m.m waoom unmeasoHHou ansxwm .AmEUUfl NV mmcwuou muo< xwm monom\maom «0 boozeawxfia .OHnom oUGmH0w> accomumnuwucH wo wchuQmoo< Eouu uewuw w>au .mauom mumaamm xom Hoflunmu0>c< .maoom woaduawood nu»: 006m Idmmfiflmmfljammammmumm AumSUmwp .m> Homsouo m.:d603 x neon on .m> Ceca m.cdfio3 x ucmwcooncoc .m> ucmmcoo m.cd503v Gawmwp Hnfluouuou oommouo haaau 6 Ga newMOuu unmam wucw65um mace mva mo 0G0 ou vmcwumaa u.m Houucoo mquOQXm on bad cofiuoum .mcwuwchSSmU .ucmaow>coa .ucoH0w> >Hauaxum umcowuwvcoo Moon uo 0:0 0» vocuammn u.m mucwvaum Icon made man can mucwvaum odds can 54mg Hanan Ammmav xumnu w nudfidaoz Amman. nowadao a xoonu a 44 mamom >H4 co uwnmfin >Hucdowwfinmflm mfiaww quHofi> haawsxmm ou pmmomxm muomeSM mans mo mmuoom mndmmmfi >nuuQEw mama Howuu long so mmuoom umBoH vmuuoamu mfiaww ucwHofl> pwumuum ou vmmonxw m.m >£uu03 mama can kuswcfl mood Eau0w> ouch powwow maafiu ucmHOA> vmuuunm ou vomoaxo m.m ”and mwacom wwwflawm HMSXmm Hafiummuw>p< pad mucmuamood nu»: QOm .moGwH0w> chomumnuwucH mo mucounwood ucmvcmwmo pad EduuM> mo mucwfiflonsn whammofi Ou Umcufimmp wuflocdofiumwsv vmum3mcd pad ompw> Hmfluu mama anoumz m.m sway “muomnno dexmm mm cwEo3 3wfi> Ou >0Gmpcwu pad mewumawu Hmsxmm aw 00H0u mo quEomuovcm .mmaou xmm 0>fium>ummaoo :« wmfiamn .mUGMuQmoo< £u>z mo acamuw> wawfivofi .>£quEm mamm OGwDSHoGM mwamom mdowuu> UmuwumMCwfiom m.m musmooxw Mound ucdaflommw Una ucdpcwump uo macauawouoa ucwpumuou oufloccofiummsv 60u03mcd can downy many a wo uGoEuomcwmu ponuunB m.m EAwu umwa uwuun “6&503 ou mnz u“ unavouuwv 3oz pad mcofimcofifip oocwaow> odd Kan do anu on» nounu pad unwaxuwno @006 a ocuoaasoo n.m .Eaau £060 unasmwb “muw4 11.3% pmuuuux mm3 mfiawu onu wo mac: .mfiawu ucoaofi>coc 03u Ho Ndidludumw pad Nmmmwum umH .msafiu unwaow> haamaxwm o3» Umbw> m.m mucmvzum mama om Una mawfiwm mm xup Rondo hho>w woo .mEAau w>ww Ho 03» Hwnufim 603mfl> m.m “madmoaxw on no madam .xmm momcwwu. .mfiafiw unmaofi>coc pmumuux .mfiawu ucwaofi> pwumunm nmcofiuwpcoo H50u mo oco mucmooum wane wma ou bmcmwmmm 0H03 m.m nfiawu on 3am no «>66 0>ww How and a aawu uaoHo«> mucwpzum moans an pwuduum oao 003mw> I.m g «3mm Aammav xumnu a nufifidaoz Ammmflv Gounod a dwmumumccoo .Ncwq .mmmav Ncfiq .Avmma. panama a nwouuuoacoo .nde g 455 mucouum uncufiuficufiu oz umwmumm mwamom Ramma .xumnu w zuasdauz. Emflun>uwmcoo HMSXmm bad Aomma .uusmv oocmHOM> AmcomquhoucH no mocduawoo< .wocduamood nu>z aqua .mumflamm Hosxwm Howucmum>pd .o:wn>uomuoum maom xwm .Ambma .gomeEHo: a mquva :0803 uvum308 mmpduwuu< Eouu mewuw UGMUSHUCM oHMGGCOMumwaO glam? modufi an odd mmacfiwu no a xw03 wGo new xcv >u0>m Eaflu woo 003ma> m.m “mfiafiu owuououcoc no >£Qnumocuon uaoao«>coc umnuwm vozmw> m.m «a um .mmmav Huoz a "usam-awamfium .uuwuoom Hasluu 46 materials (Linz & Donnerstein, 1988) and confounding measures (Brannigan & Goldenberg, 1987; Christensen, 1986) as well as for theoretically paradoxical results (Brannigan, 1987). Worth noting are the measurement issues raised by Christensen (1986). Of primary concern is the sexual callousness scale, which the Christensen criticizes as confounding. The scale contains a number of "Anglo-Saxon four-letter words" and, as Christensen argues, exposure to pornographic films could have reduced the shock value of these words, thereby resulting in more positive responses. The validity of this criticism was conceded by Zillmann and Bryant (1986) who later retracted this part of their study. One further criticism of Zillmann and Bryant’s study of note is the theoretical paradox noted by Brannigan (1987). In an account of their experiment published elsewhere Zillmann & Bryant (1984) reported that subjects became less aggressive toward their targets after exposure to massive amounts of pornography. This decrease in aggression corresponded to a reduction in negative affect and increase in enjoyment to the pornographic stimuli, lending support to the arousal-affect model. However, if male subjects were becoming more sexually calloused toward women, why didn’t they behave more aggressively toward them? The answer to this may lie in the fact that the aggression examined in Zillmann and Bryant’s experiment was between subjects and targets of the same gender, not male toward female as Brannigan apparently assumes. Males, having developed 47 certain anti-social attitudes toward women as a result of exposure to pornography, might have aggressed more against females than other males, as did male subjects in Donnerstein and Hallam’s study (1978). A number of other studies have reported results lending support to Zillmann and Bryant's findings. The potential of exposure to pornography to lead to increased expectations of or dissatisfaction with one’s sexual partner, was supported more directly by a subsequent experiment (Zillmann & Bryant, 1988) in which subjects massively exposed to pornography reported more sexual dissatisfaction with their current sexual partner. This finding is significant because, as the authors theorize, such disappointment could lead to conflict, and ultimately violence against one's sexual partner. The rape trivialization finding was replicated in a follow-up study by Weaver (1987) in which subjects were exposed to materials characterized as "female-instigated sex" in which women were portrayed as "eagerly and indiscriminately seeking and participating in sexual endeavors" (p. 47). Exactly what cognitions might be primed by exposure to nonviolent materials have been examined in other studies. This priming process occurs when exposure to a communication makes certain cognitions more easily available in memory (Malamuth & Check, 1985). In turn, these recently activated cognitions will have a stronger influence than alternative cognitions on ensuing judgements. In a study designed to 48 prime the concept of "women as sex objects," Wyer, Bodenhausen and Gorman (1985) found slide portrayals of women depicted as sex objects decreased male perceptions of a rape victim’s credibility and increased their beliefs that she was responsible for the assault. In Weaver's (1987) study, exposure to mutually-consenting and female—instigated film depictions, compared to the neutral stimuli, prompted male subjects’ to perceive sexually discriminating female peers ("nice girls") as more permissive. Additionally, exposure to various types of R-rated films, including those in which the sexual activity was female-instigated, male- coerced or the violence was eroticized, shifted subjects' perceptions of a non-permissive female peer closer to that of a sexually promiscuous and a submissive prototype. The greater this shift in perception of women as permissive and submissive, the more lenient subjects were toward a rapist in terms of recommending a prison sentence. Of note here is that other research has shown that priming cognitions of female permissiveness may lead to increased inter-gender aggression (Leonard and Taylor, 1983). In a study somewhat analogous to the sex-object priming demonstrated by Wyer et a1. (1985), McKenzie-Mohr and Zanna (1990) hypothesized that exposure to pornographic stimuli would prime a "heterosexuality subschema" in male subjects. According to Bem (1981) such a subschema leads individuals to "code all cross-sex interactions in sexual terms and all members of the other sex in terms of sexual attractiveness" 49 (p. 361). In this latter study, subjects scoring above the median on the masculine scale of the Bem Sex Role Inventory (Bem, 1974) and below the median on the feminine scale were classified as masculine sex typed. The authors found that masculine sex-typed male subjects tended to be more sexist when interacting with a female after exposure to a sexually explicit video than when exposed to a neutral video. Similar results were reported by Check and Guloien (1989) who found that subjects who were "somewhat sexually aggressive" to begin with reported a greater likelihood of engaging in forced sexual activities including rape after exposure to violent and degrading pornography. In sum, it appears that exposure to nonviolent, degrading pornography can prime certain cognitions in males that may lead to increases in anti-woman attitudes. These cognitions might best be described as viewing women in sexist terms, primarily as sex objects. Theoretically, the effect such cognitions might have on aggression comes from the literature on dehumanization, a process which strongly parallels that of degradation. In fact, the terms appear to be used interchangeably in the literature (Check & Guloien, 1989). Common to both is the loss of personhood by an individual, i.e., being treated as less than a human being or having no personal identity (Hill, 1987; Kelman, 1973; Zimbardo, 1969). Treating a woman as a sex object would fall into this category. Her personal identity and human qualities become secondary to her physical attributes. This 50 process of dehumanization has been theorized to lead to increased aggression toward those dehumanized. Placed into a social learning perspective, dehumanizing others is seen as a means of reducing self-reproach (a form of disinhibition) for using aggressive or cruel actions against others (Bandura, Underwood & Fromson, 1975). Once a person is dehumanized, he or she becomes a more inviting target for aggressive actions. As noted by Bandura et a1. (1975) "inflicting harm upon individuals who are regarded as subhuman or debased is less apt to arouse self-reproof than if they are seen as human beings with dignifying qualities." Considering the above, exposure to degrading materials should make cognitions of females as having subhuman qualities more easily accessible in viewers’ memory. In turn, these recently activated cognitions should increase subsequent aggressive tendencies toward women by reducing self-reproach or certain inhibitions among viewers. Sexually violent materials. The attitudinal effects of portrayals of aggression against women in the mass media have been examined by Donnerstein and his colleagues. (See Table 2 also for a summary of studies discussed in this section). In a study repeatedly exposing subjects to R-rated violent films, Linz, Donnerstein and Penrod (1984) found that, compared to a no- exposure control group, these subjects judged a rape victim to be less injured and less worthy. In a subsequent account 51 of tflris study, Linz (1989) reported similar results for subjects exposed to the X—rated nonviolent and X-rated violent films. In a more recent study, Linz, Donnerstein and Penrrxi (1988) found subjects exposed repeatedly to R-rated violent films over a two-week period compared to subjects in all other conditions (X-rated nonviolent, R-rated nonviolent, no-exposure) reported significantly less empathy for victims of rape. However none of the exposure conditions was significant when compared to a no-exposure, control group. One possible explanation for the lack of significant findings in this study is that subjects rated each film immediately after viewing on various sex and violence dimensions as well as how degrading the films were to women. Such evaluations could have created an awareness of the experimental objectives and subjects may have guarded against being influenced by the stimuli (Zillmann & Weaver, 1989). Extending this direction of inquiry to a natural setting, Malamuth and Check (1981) found that subjects who had seen sexually violent, but not X-rated, films as part of a campus film program reported more acceptance of interpersonal violence against women. The R-rated films used by Donnerstein and his colleagues fit into a genre of movies known as "slasher filmsJ'The sex depicted in these movies is typically suggestiwe, not explicit. The violence is quite graphic and usually directed at women (Donnerstein, Linz & Penrod, 1987L.However, most often scenes containing sex are 52 juxtaposed with scenes of violence that is not sexual in nature. (In a content analysis of slasher films, Weaver (1991) reports a negligible number (0.5%) of scenes depicting sexual violence). One example of this genre, described by Linz et al. (1984), is that of a woman who is shown masturbating in the bath when she is suddenly and brutally attacked. Donnerstein and his colleagues have argued that it is the violence as opposed to the sex in media portrayals that contributes to people’s attitudes about women and rape (Donnerstein & Linz, 1986). These authors report two experiments, in which the sexual and violent components were systematically manipulated, to bolster their claim (Donnerstein et al., 1987). In the first study males exposed to a sexually violent film exhibited more aggression against a female confederate than subjects exposed to an aggression- only film, which, in turn, produced more aggressive behavior than a X-rated, sex-only film or a neutral film. In the second experiment, the greatest acceptance of rape myths and the largest percentage of subjects indicating a likelihood of rape or using force were found in the aggression—only condition. Subjects in the X-rated, sex-only condition scored lowest on these two measures while subjects exposed to sexual violence scored somewhere in between. In addition, subjects in the sexually violent condition evidenced higher levels of aggression than subjects in a sex-only condition. Subjects in the aggression-only condition also exhibited 53 higher levels of aggression than sex-only subjects but only under certain conditions (They were given a name association one or were previously angered). However since these studies do not report significance levels, judgements about their validity are not possible. In addition, the findings seem at odds with Donnerstein and Linz’s assertion since the sexually violent materials seemed to produce greater aggressive behavior than the violent materials. Theoretically, if the sex did not matter, one would expect them to produce equivalent amounts of aggression. At the very least, replication of these results seems warranted. An interesting point is raised by Fisher and Grenier (1994) who note a number of studies, including their own, in which sexually violent materials have failed to produce anti-woman thoughts and acts. These authors argue that certain methodological limitations such as subject awareness, selective attrition and chance findings may have contributed to the inconsistencies among studies in this area. Similar concerns regarding chance findings have been expressed by other authors (Agreed, 1988; Linz, 1989: Mould, 1988) who refer to a number of studies in which many dependent variables are used while a relatively small number of significant results are reported. This being the case the impact of sexually violent materials on anti-social attitudes and behaviors could have been exaggerated by previous researchers (Fisher & Grenier, 1994). 54 Summar an H otheses Understanding the potential effects of violent and/or degrading materials on viewers is important for two reasons. One, a large portion of adults have been exposed to these types of materials at least once, and many adults report repeated usage. Since most pornographic materials are geared primarily toward males, not surprisingly males are more frequent users than females. These materials, which frequently portray degrading and violent sexual activities, at the very least could foster vastly disparate expectations of sexual relationships between viewers and nonviewers. Secondly, research into the potential effects of violent and degrading materials has been used to shape public policy and restrict, or chill, expression in Canada and the United States. However many questions remain. For example, does viewing females as sexual objects lead to anti-social behavior? Is it sexual violence or violence alone that produces anti-social attitudes and behaviors. Only until these questions are addressed more directly in the research can we began to more fully understand the process whereby certain media depictions lead to anti-social attitudes and behaviors and thus make more informed policy decisions. While exposure to pleasing "erotica" has not been shown in: produce aggression, exposure to erotica that is «disturbing has been shown to increase aggression. There is some indication that materials that have increased aggression have contained violent and/or degrading images. 55 In fact, research has shown that subjects exposed to sexually violent materials (Donnerstein, 1980; Donnerstein & Berkowitz, 1981) exhibit more aggressiveness, especially toward a female target. What has not been demonstrated is whether degrading materials may have a similar effect. Research has also shown that nonviolent, degrading materials can have a detrimental effect on peoples’ attitudes and perceptions toward women (Check & Guloien, 1989; Linz, 1989; McKenzie—Mohr & Zanna, 1990; Wyer, Bodenhausen & Gorman, 1985; Zillmann & Bryant, 1982, 1988). If these types of materials can foster or, at least reinforce, sexually aggressive or calloused attitudes toward women shouldn’t they also encourage males to act more aggressively toward women? A link between aggression and degrading portrayals comes from the literature on dehumanization, in which individuals have been theorized to aggress more against persons who have been dehumanized (Bandura, Underwood & Fromson, 1975; Check & Malamuth, 1986; Kelman, 1973; Zimbardo, 1969). Dehumanizing others is seen as a means of reducing inhibitions against aggressive actions (Bandura et al., 1975). Depictions that dehumanize women may thus lead to greater subsequent aggressive behavior toward women in general. A similar type of disinhibition process may take place in violent media portrayals. According to Berkowitz (1993) , portrayals of violence can activate aggressive thoughts in viewers. If the violence is depicted as having positive 56 consequences, for example, then some people may become more aggressive because their existing inhibitions against aggression have been lowered (Berkowitz, 1993). Recently feminist researchers have argued that degrading materials used by Zillmann and Donnerstein and their colleagues have focused primarily on women’s display of sexuality as opposed to themes that have been the focus of feminist criticism of pornography such as "dominance" and “objectification" (Cowan & Dunn, 1994). Until the attitudinal and behavioral effects of these types of themes have been examined, these authors argue, a true test of the effects of degrading pornography has not been demonstrated (Cowan & Dunn, 1994). This study, then, will attempt to use materials whose theme falls into one of the more degrading categories described by Cowan and Dunn (1994). An additional issue that will be addressed in this study is that raised by Donnerstein et al. (1987) concerning the relative contributions of the sexual and violent components of sexually violent materials in affecting attitudes and behaviors. In the experiments reported by Donnerstein et al. (1987) sexually violent materials appeared to have greater affect on aggression than violent- only or sex-only materials. This suggests the possibility that materials combining both degrading sex and violence may have more powerful effects than materials with only one of these components. if the perpetrator is rewarded, 57 The intent of this study then is to test the relative contributions of degradation and violence on male aggression toward females. In order to do this stimuli were selected and rated on violent and degrading dimensions so that a fully crossed factorial design could be used in the main experiment. As a result there were four experimental conditions: High Violence, High Degrading (sexually violent materials); High Violence, Low Degrading (violent materials); Low Violent, High Degrading (degrading materials) and Low Violent, Low Degrading (erotic materials). Due to the ethical considerations of using physical aggression against subjects, verbal aggression procedures and measures were used in the study. This is consistent with previous research recommending consistency between the two and indicating no significant differences between modes (verbal or physical) in evaluation of a confederate (McDaniel, O’Neal and Fox; 1971). Based on the preceding discussion, the following hypotheses were tested. H1: Provoked male subjects will exhibit significantly (p < .05) greater verbal aggressiveness toward a female confederate after exposure to materials rated high in violence toward women than provoked subjects exposed to materials rated low in violence toward women. H2: Provoked male subjects will exhibit significantly (p < .05) greater verbal aggressiveness toward a female confederate after exposure to materials rated high in degradation toward women than provoked subjects exposed to materials rated low in degradation toward women. H3: Provoked male subjects will exhibit significantly (p < .05) greater verbal aggressiveness toward a 58 female confederate after exposure to materials rated high in degradation toward women and high in violence toward women than subjects exposed to materials rated high in degradation toward women and low in violence toward women or high in violence toward women and low in degradation toward women or low in both violence and degradation toward women. The purpose of the pre-test film evaluations was to achieve objective ratings of the segments. Since previous research has shown some sex differences in the perception of degrading behaviors with males rating some materials less degrading than females (Cowan & Dunn, 1994), it was decided to use male and female subjects for the film evaluations. One of the documented effects of exposure to pornographic materials has been that of desensitization to violent and degrading behaviors (Linz et al., 1988). Since males have reported more frequent exposure to pornographic materials than females, one might expect that even if males and females originally had similar perceptions, they might be expected to differ following their life experiences. Specifically, males may report less violence and degradation than females, who might be less desensitized to such portrayals. Using all male subjects to evaluate the films might increase the likelihood that low violence and/or low degrading materials could be selected that would be perceived much differently by females. As such the generalizability of the study might be perceived by some as somewhat limited. In an attempt to avoid this potential criticism, both genders were used to evaluate the films. 59 In addition, subjects’ affective reactions to the experimental stimuli were measured in order to assess the possibility that either arousal or negative affect may have contributed to subjects’ verbally aggressive responses. Since the above hypotheses rely on a cognitive model, a null relationship between arousal or negative affect and the experimenter or experiment evaluation variables was expected. 60 METHOD Pre-test of Stimuli Materials Evaluations of and reactions to the film stimuli in terms of violence, degradation, sexual arousal and several other variables were determined in a pre-test. Male and female students were recruited from introductory psychology courses at a publicly funded Canadian university. A total of 70 students participated in this phase of the study. Students received extra credit toward their final grade for their participation. All subjects consented to participating in an experiment that might involve exposure to sexually explicit and/or violent materials. Four experimental conditions were evaluated: sexually violent (HV-HD), violent (HV-LD), nonviolent degrading (LV- HD) and erotic (LV-LD). Male and female subjects, in separate groups of 6 to 9 participants, viewed and rated film segments in one of the conditions. Each condition consisted of 3 to 5 film segments, each about 5 to 10 minutes in length. Segments were selected for each of the film conditions based upon the following definitions. LV;LQ (erotica): mutually pleasurable, sexual activity that includes affection, warmth and/or sensuality between participants (Steinem, 1980). LV-HD (nonviolent; degrading): sexual activity in which the male is dominant and commands the female to do what he wants without regard for her desires (dominance) or the female is treated as a sexual 61 object or "plaything" to be used at the whim of the male participant (objectification) or the activity revolves around the worship of the penis (penis worship). These themes were found to be among the most degrading in a sampling of X-rated videos by Cowan and Dunn (1994). HV;HQ (sexually violent): portrayals in which violence is depicted as a part of the sexual activity. Violence is defined as an "overt expression of physical force against self or other, compelling action against one’s will on pain of being hurt or killed" (Gerbner & Gross, 1980). This definition includes the intention to cause harm and is limited to physical aggression. Because in Canada X-rated violence is banned, R- rated materials were used in this category. HV-LD (violent): activities in which physical force of a nonsexual nature is used against females by males. Titles and film lengths for all of the films pre-tested can be found in Table 3. Each stimulus was evaluated immediately after exposure. Subjects were administered a questionnaire in which the degradation (degraded, humiliated), dominance (dominant, submissive (reversed), aggressive), and willingness (willing, promiscuous) of the male and female characters were assessed on a 7-point Likert scale ("not at all" to "very much"). Overall film ratings of violence (violence, brutal) and arousal (sexually arousing, exciting, stimulating) were assessed in the same manner. Other items on the questionnaire such as affectionate, tranquil and disgusting were used as filler items. 62 Table 3 Film Clips Evaluated Category/ Title Length Number LV-LD (erotica) 1 "Femme-TV Idol" 8:06 2* "Femme-The Dressing Room" 6:54 3* "Urban Heat" (Disco) 7:15 4 "Urban Heat" (Morning) 6:32 LV-HD (degrading) 1 "Damiano's Hot Fuckers" (Love To Eat Cum) 5:28 2* "Amanda by Night" 6:25 3 "Damiano’s Hot Fuckers" (Wide World of Spurts) 6:40 4 "Nothing to Hide" 6:47 5 "Insatiable" 6:52 6* "Barbara Broadcast" 6:12 7 "Talk Dirty to Me" 10:40 8 "Debbie Does Dallas 8:38 HV-HD (sexual violence) 1* "Blue Velvet" 5:17 2 "Looking for Mr. Goodbar" 6:07 3 "Swept Away" 8:26 4* "Last Tango in Paris" 4:32 5 "I Spit on Your Grave" 5:15 HV-LD (violence against women) 1 "Body Double" 5:27 2 "Savage Attraction" 7:41 3 "Raging Bull" 5:40 4* "Dressed to Kill" 6:40 5* "Halloween" 3:40 note: Asterisks indicate those films selected for the main experiment. 63 After the film evaluations were completed subjects were debriefed about the experiment. (Female subjects were debriefed by a female experimenter.) The experimenter explained the purpose of the experiment and discussed relevant research findings from exposure to violent or pornographic materials. The experimenter also offered assistance to anyone who might have been upset by the sexually explicit or violent materials by referring them to the university psychology services center. In addition, participants were given a "Personal Safety Card" issued by the campus Sexual Harassment Office on which phone numbers for various local resources were listed. Each set of attributes (overall, male and female characters) was factor analyzed. Since all of the scales used in the study can be thought of as interval scales (the intervals on the scales are equal), permissible mathematical operations include analysis of variance and factor analysis (Nunnally, 1978). Factors were extracted using principal components analysis and varimax rotation. Cronbach’s alpha was used as an estimate of reliability. Factor loadings above .40 were considered significant while variables with overlapping factor loadings were discarded (Nunnally, 1978). The cut-off point for reliability estimates was .70 (Nunnally, 1978). Table 4 presents the varimax-rotated matrix resulting from the factor analysis. Two factors were extracted for the overall film ratings, Violence (Factor 1), Cronbach’s alpha = .93, and Arousal (Factor 2), Cronbach’s 1'5» Table 4 Factor Matrixes for 64 the Film Evaluation Questionnaire Overall ratings Item Factor 1 Factor 2 (Arousal) (Violence) Entertaining 422 -.15 Sexually arousing lg; -.28 Exciting #26 -.08 Stimulating L2; -.22 Violent -.11 424 Brutal -.13 i2; Eigenvalue 3.94 1.57 Percent of variance 65.8 26.2 Male Character Item Factor 1 Factor 2 Factor 3 (Degraded) (Dominant) (Willingness) Humiliated l2; -.17 .10 Willing -.49 .44 46; Submissive -.13 .17 -L84 Promiscuous .27 .24 _LZ§ Dominant -.13 i8; .00 Degraded 422 -.02 .19 Aggressive -.05 i§§ .08 Eigenvalue 1.98 1.82 1.51 Percent of variance 28.3 26.0 21.5 Female Character Item Factor 1 Factor 2 Factor 3 (Degraded) (Dominant) (Willingness) Humiliated igg .07 .01 Willing -.63 -.09 __51 Submissive -.36 .24 —i§§ Promiscuous .01 .32 i8; Dominant -.08 igg_ .16 Degraded All -.07 .33 Aggressive .11 i2; -.10 Eigenvalue 2.41 2.07 1.08 Percent of variance 34.5 29.7 15.4 ngt_: Underlined factor loadings indicate items used in calculating reliabilities for the factors. 3 70. 65 alpha = .96. Ratings for the female character(s) indicated factors for Female-degrading (Factor 1), Cronbach’s alpha = .88, and Female-dominant (Factor 2), Cronbach’s alpha = .68. Ratings for the male character(s) indicated factors for Male-degrading (Factor 1), Cronbach’s alpha = .69, and Male- dominance (Factor 2), Cronbach's alpha=.82. The reliabilities for the Female—dominant and Male—degrading factors were deemed close enough to the cut-off point to include in the analysis. Another factor, Willingness (Factor 3), was extracted for both the female and male character evaluations. However, because of the dispersed factor loadings of the willing item for both characters and the low reliabilities of the Willingness factor for both males (alpha = .47) and females (alpha = .65), it was decided to use the willing item individually as an indicator of female willingness. This particular item was of interest because previous research has shown that portrayals of the female as the "willing victim" of sexual violence may increase aggression (Donnerstein & Berkowitz, 1981). The willing item was considered the most direct measure of this attribute. One distinction between the LV-LD (erotic) and LV-HD (degrading) films is the unequal roles of the male and female. It was expected that males would be rated relatively high on dominance in all conditions, including the LV-HD (degrading) condition, compared to the LV-LD (erotic) segments. Females were expected to be rated relatively low on the dominance item in all conditions except the LV-LD 66 (erotic) condition. Analyses of variance were performed for each film segment to help narrow the choice of films to use in the main experiment, the objective being to select 2 or 3 clips that best represented each condition. Results from these analyses are presented in Table 5. The only significant gender difference found was that males were more aroused than females by portrayals that are degrading to women and to a lesser extent by depictions of sexual violence against women. Interestingly, two of the clips in the LV-HD (degrading) condition were rated relatively high on violence. These two segments, taken from "Nothing to Hide" (film No. 4) and "Insatiable" (film No. 5), were the same as those used by Cowan and Dunn (1994) to represent objectification and dominance, but had been edited to exclude overt physical acts of violence. Further testing revealed that what subjects were actually coding as violence in these segments was primarily the verbal abuse directed toward the female character by the male character, as opposed to overt acts of physical violence. Based on these evaluations the following procedures were followed: (1) the two LV-HD (degrading) films rated high on violence, "Nothing to Hide" and "Insatiable," were set aside and not used in this study since they differed from the rest of the films in this category, which were rated relatively low on violence; and (2) two clips were selected from each of the four categories to be used in the (a Table 5 Mean Scores for Film Evaluations - All Conditions 67 AROUS VIOL DEG-F DEG-M DOM-F DOM-M WIL-F WIL-M LV-LD (erotic; 3:13, females=6, males=7) "Femme-TV Idol" (f) 3.45 1.00 1.67 1.50 2.83 4.25 6.83 6.83 (m) 4.25 1.07 1.14 1.14 3.00 3.50 7.00 7.00 (t) 3.92 1.04 1.38 1.31 2.92 3.85 6.92 6.92 "Femme-The Dressing Room"* (f) 3.80 1.08 1.83 1.50 2.17 4.42 6.83 6.50 (m) 3.70 1.64 1.21 1.14 2.43 5.64 6.57 7.00 (t) 3.75 1.38 1.50 1.30 2.31 5.08 6.69 6.76 "Urban Heat" (Disco scene)* (f) 3.42 1.00 1.00 1.50 3.50 4.17 6.67 6.67 (m) 3.89 1.50 1.35 1.00 3.43 5.00 6.86 6.42 (t) 3.67 1.27 1.19 1.23 3.46 4.62 6.76 6.54 "Urban Heat" (Morning scene) (f) 2.86 1.00 1.50 2.00 2.50 3.59 6.50 6.83 (m) 4.07 1.00 1.50 1.00 2.57 4.71 6.43 7.00 (t) 3.51 1.00 1.50 1.46 2.54 4.19 6.46 6.92 LV-HD (degrading; 3:15, females=9, males=6) "Damiano’s Hot Fuckers" (Love to Eat Cum) (f) 1.75“ 1.56 3.22 2.55 5.38 2.11 6.78 6.44 (m) 5.13b 1.17 2.92 1.17 5.67 2.00 6.67 6.00 (t) 3.10 1.40 3.10 2.00 5.50 2.01 6.73 6.26 "Amanda by Night"* (f) 2.19“ 2.33 5.06 2.28 1.33 6.27 4.67 6.78 (m) 5.67b 2.00 5.08 2.08 1.67 6.00 4.33 6.50 (t) 3.58 2.20 5.07 2.20 1.47 6.17 4.53 6.67 "Damiano’s Hot Fuckers" (Wide World of Spurts) (f) 1.44“ 1.50 3.94 2.11 3.75 1.56 6.89 6.44 (m) 5.04b 1.42 3.25 2.00 5.00 1.17 6.16 6.33 (t) 2.88 1.47 3.67 2.07 4.29 1.40 6.60 6.40 "Nothing to Hide" (f) 1.67“ 5.11 5.44 2.39 1.78 6.72 6.44 5.33 (m) 4.33b 4.58 6.16 1.58 1.67 7.00 7.00 5.00 (t) 2.73 4.90 5.70 2.06 1.73 6.83 6.67 5.20 68 AROUS VIOL DEG-F DEG-M DOM-F DOM-M WIL-F WIL-M "Insatiable" (f) 1.67“ 4.22 5.28 2.27 2.11 6.78 6.22 5.89 (m) 6.08b 3.08 5.25 1.25 1.67 7.00 6.67 5.83 (t) 3.45 3.77 5.27 1.87 1.93 6.87 6.40 5.87 "Barbara Broadcast"* (f) 2.17“ 1.89 5.22 2.56 3.22 5.00“ 6.67 7.00“ (m) 4.21B 2.08 5.00 3.92 3.50 3.17b 6.00 5.83b (t) 2.98 1.97 5.13 3.10 3.33 4.30 6.40 6.53 "Talk Dirty to Me" (f) 2.61“ 2.33 4.00 1.55 1.56“ 6.33 5.56 7.00 (m) 5.178 2.50 5.25 2.75 2.67b 5.92 4.83 6.33 (t) 3.63 2.40 4.50 2.03 2.00 6.17 5.27 6.73 "Debbie Does Dallas" (f) 2.56“ 1.83 4.11 1.33 1.44 6.50 6.72 7.00 (m) 4.538 2.36 3.86 3.21 2.43 6.21 5.57 6.43 (t) 3.42 2.06 4.00 2.16 1.88 6.37 5.94 6.75 HV-LD (violent; n=l3, females=6, males=7) "Body Double" (f) 1.92 5.83 4.83 1.42 1.33 6.67 1.00 6.17 (m) 2.50 5.07 4.21 1.79 1.14 6.71 2.14 4.14 (t) 2.23 5.42 4.50 1.61 1.23 6.70 1.61 5.08 "Savage Attraction" (f) 1.21 4.08 6.00 1.16 1.50 7.00 1.67 5.67 (m) 2.07 3.42 6.14 1.14 1.14 6.78 1.86 4.28 (t) 1.67 3.73 6.07 1.15 1.31 6.88 1.77 4.92 "Raging Bull" (f) 2.29 6.00 4.16 3.50 4.33 6.75 3.17 4.67 (m) 3.25 5.07 4.14 4.50 2.57 6.43 2.86 4.14 (t) 2.80 5.50 4.15 4.03 3.38 6.58 3.00 4.38 "Dressed to Kill"* (f) 1.65 6.58 4.58 1.16 2.50 3.75 3.00 3.50 (m) 2.50 4.92 3.93 1.75 2.00 5.00 3.86 4.86 (t) 2.10 5.69 4.23 1.45 2.23 4.38 3.46 4.23 "Halloween"* (f) 1.04 5.00 2.92 1.42 1.50 5.25 5.17 5.67 (m) 2.82 4.07 2.57 1.64 1.71 6.86 4.14 4.71 (t) 2.00 4.50 2.73 1.54 1.65 5.58 4.62 5.15 69 AROUS VIOL DEG-F DEG-M DOM-F DOM-M WIL-F WIL-M HV-HD (sexually violent; 3:14, females=8, males=6) "Blue Velvet"* (f) 1.81“ 6.31 5.86 2.88 1.38 6.94 4.75 6.00 (m) 3.088 5.83 6.17 2.33 1.00 7.00 5.33 5.83 (t) 2.35 6.11 6.00 2.64 1.21 6.96 5.00 5.92 "Looking for Mr. Goodbar" (f) 1.78 6.87 4.19 3.31 3.13 6.87 4.25 5.00 (m) 3.12 6.58 4.42 3.75 2.50 6.33 3.50 5.50 (t) 2.36 6.75 4.29 3.50 2.86 6.64 3.92 5.21 "Swept Away" (f) 2.50 3.88 4.88 2.25 2.38 6.25 5.00 5.63 (m) 3.37 4.25 4.42 2.00 2.00 6.00 4.67 4.83 (t) 2.88 4.04 4.68 2.14 2.21 6.14 4.86 5.28 "Last Tango in Paris"* (f) 1.91 4.19 5.81 2.00 1.75 6.75 2.87 5.83 (m) 3.42 3.75 5.08 2.08 1.50 6.33 3.33 5.87 (t) 2.55 4.00 5.50 2.04 1.64 6.57 3.07 5.86 "I Spit on Your Grave" (f) 1.25A 6.25 6.56 1.50 1.13 6.75 1.00 6.33 (m) 2.508 5.58 6.91 1.58 1.50 6.94 1.00 7.00 (t) 1.78 5.96 6.71 1.54 1.29 6.86 1.00 6.71 AROUS=arousing; VIOL=violent; DEG-F=degrading-female character; DEG-M=degrading-male character; DOM-F=dominance- female character; DOM-M=dominance—male character; WIL— F=willing-female character; WIL-M=willing-male character; f=females; m=male; t=total. note: Asterisks indicate those films selected for the main experiment. Within each film segment and column, means with different subscripts differ at the .05 (uppercase) and .01 (lowercase) levels of significance. A second session was used to rate the films numbered 6 through 8 in the LV-HD (degrading) condition, n=15, females=9 and males=6. 70 main experiment. These clips were selected on the basis of how they were rated on the appropriate attributes (primarily violence and degradation). For example, the films in the HV- LD (violent) condition were selected because of their relatively low ratings on female degradation. In addition some consideration was given to the total time subjects would be exposed to the film segments, the objective being to equalize exposure times across conditions as much as possible. Descriptions of the selected film segments can be found in Appendix A. The two film segments selected for the LV-HD (degrading) category might best be characterized by themes of objectification ("Amanda by Night") and penis worship ("Barbara Broadcast"). As mentioned previously, these two themes were rated among the most degrading in a sample of X-rated materials (Cowan & Dunn, 1994). As such, these segments were selected to provide as strong a test as possible for the effect of nonviolent, degrading materials on aggression. Total film length for each category ranged from just under 10 minutes to about 14 minutes. A 2 X 4 (Sex X Film Type) analysis was used to analyze the data from the selected segments. Since the number of subjects was not equal among treatment groups, an analysis of unweighted means was used (Keppel, 1982). The primary reason for the unequal sample sizes is that two groups of male and female subjects were used to rate films in the LV— HD (degrading) condition. Since one film was selected from each group, each film, then, was rated by a different 71 subject. In the other conditions, both films were rated by the same subject. An interaction between sex and group was found for the Arousal scale. Compared to the one condition devoid of sexual activity (violent), males were more aroused by the degrading portrayals (HD-LV) while females were more aroused by the erotic depictions (LD-LV). In general, males (M = 3.87) reported more arousal to the films than females (M = 2.20), 3(1, 69) = 28.18, p < .001. There were no other significant differences between males and females in the ratings. Main effects for the other dependent variables are shown in Table 6. As expected films selected for their violent content were rated high on violence. The LV-HD (degrading) and HV-HD (sexually violent) films were rated high and the LV-LD (erotic) films low on female degradation. The HV-LD (violent) segments were rated moderately degrading, not significantly different from either the high or low degrading ratings for the other segments. It appears that violence is considered somewhat degrading regardless of the context. In fact, the correlation between violence and female-degradation over all conditions proved to be highly significant (See Table 7). As mentioned previously, the films selected for the violent condition were those that subjects had rated as the least degrading to the female character(s). As expected, male characters were rated relatively low on degradation in all of the conditions. 72 Table 6 Analysis of Variance for FilmiRatings Usinngnly Film Segments Selected for Main Experiment Attribute LV-LD LV-HD HV-LD HV-HD E (erotic) (degrading) (violent) (sexually violent) Q 13 301 13 14 Arousal (females) 3.62b 2.18‘“b 1.31a 1.86‘“b 3.43* (males) 3.79"b 4.94” 2.66“ 3.25%b 4.62** Violence 1.33“ 2.08“ 5.10b 5.05b 54.20** Female- 1.35“ 5.10b 3.48‘”b 5.66b 39.91** degrading Male- 1.35“ 2.65b 1.50“’b 2.34“’b 3.7l* degrading Female- 3.25b 2.93b 1.94“’b 1.50“ 5.32** dominant Male- 4.85“ 5.23“ 5.06“ 6.77b 5.11** dominant Female- 6.73“ 5.47b 4.048 4.04° 10.3o** willing Male- 6.65b 6.60b 4.69“ 5.89“’b 8.16** willing ** p < .01; * p < .05 Note: Within each row, means having no letter in their superscripts in common differ at p < .05 by Tukey’s test for comparisons between pairs of cell means. Scales were 7-point Likert scales with 1 = "not at all" and 7 = "very much." 1 The larger than average cell size for the degrading category is due to the circumstance that each film used in this category was rated by a different group of subjects. In the other categories each film selected was rated by the same group of subjects. 73 Table 7 Pearson Product-moment Correlations for Factors and Items on Film Eva luation Questionnaire AROU VIOL DOM-M DEG-M DOM-F DEG-F M-WIL (In) (f) AROU VIOL — - 3 9* -.39* DOM-M --44* -.15 .63** DEG-M --38 -.35 .19 .01 DOM-F -51“ -.07 -.36** -.33** .26* DEG-F --09 -.46* .64** .46** .41** -.19 M-WIL -33 .09 -.31** -.06 -.3o* .16 -.13 F-WIL -50* .31 -.80** —.44** -.17 .46** -.36** .47“: *p<.05, **Q<.Ol AROU = Arousing; VIOL = Violent; DOM-M = Male dominance; DEG-M = Degrading to the male; DOM-F = Female dominance; DEG-F = Degrading to the female; M-WILL = Male willing; F- WILL = Female willing; m = males; f = females. 112293 H = '70 except for Arousal in which Q = 38 for females and I1 = 3 2 for males. Females were not very dominant in any of the videos, hilt most dominant in the LV-LD (erotic) and LV-HD (degrading) conditions. Male characters, on the other hand, were rated relatively high on dominance in all the videos, especially in the HV-HD (sexually violent) videos. As prediCted , the most equal relationship appeared to be in the LV-LD (erotic) films. Female characters were perceived as most Willing in the LV-LD (erotic) and LV-HD (degrading) videos: however female willingness was rated as at least moderate OVer all conditions. The films in the HV-HD (sexually Violent) condition were selected, in part, based on their relatively high rating, compared to other films rated in this category, on the female-willingness item. This 74 parallels a common theme found in violent pornography of the willing victim, mentioned previously. A similar rating on the female-willingness item was evident in the HV-LD (violent) condition. One explanation for this might be that sexual relations were implied in both films before the violent acts were carried out. The fact the female character was apparently agreeable in this regard may have lead subjects to rate her somewhat higher on this item than anticipated. Pre-test of Provocation Male students were recruited from an introductory psychology course. Students received extra credit toward their final grade for their participation. All subjects consented to participating in an experiment that would involve a problem solving exercise in which they would have a certain amount of time to complete each problem set. Subjects were contacted by telephone and told to report to a room on campus where the study was to take place. The provocation treatment was based on a procedure used by Zillmann and his colleagues (Bryant & Zillmann, 1979; Sapolsky & Zillmann, 1981; Zillmann, Bryant, Cantor & Day, 1975; Zillmann, et al., 1981; Zillmann & Cantor, 1976; Zillmann & Sapolsky, 1977). However one difference was that students were not insulted or criticized in the present experiment. This was due to ethics committee concerns about students' self-esteem. It was felt that criticizing 75 students’ ability to follow instructions (Sapolsky & Zillmann, 1981; Zillmann & Cantor, 1976; Zillmann & Sapolsky, 1977; Zillmann et al., 1981) or insulting their intelligence (Bryant & Zillmann, 1979; Zillmann et al., 1975) would lower their self-esteem. Even though this condition was proposed as being only temporary, it was nonetheless deemed an unacceptable risk by the ethics committee. In the provocation condition, the female experimenter arrived 10 minutes after the scheduled time for the study to begin. Then the experimenter rummaged through her files and folders, announced that she had forgotten the consent forms and would be back in a couple of minutes. Upon returning, the experimenter passed around the consent forms and then administered the problem—solving exercise which consisted of a timed questionnaire made up of sample items taken from various aptitude tests such as the Graduate Record Exam (See Appendix B). The questionnaire consisted of three parts and was designed so that subjects would not have time to finish each part during the allotted time. During each part the experimenter urged the subjects to work faster, for example, "OK, you’re going to have to work faster on this part because we don’t have a lot of time left. We’ve only got a half an hour for this study, you know." Throughout the exercise the experimenter acted in an abrupt manner. In the non-provocation condition the experimenter arrived promptly, passed out the consent forms shortly after arriving and did 76 not make any comments during the study other than to instruct subjects as to when to start and stop working on the exercises. In addition the questionnaire was modified so that each part could be completed during the allotted time. After the last exercise was completed, the experimenter collected the tests and left the room to get Experimenter 2, the "research coordinator," who administered the experimental evaluation forms. The evaluation forms were similar to those used by Zillmann and his colleagues, but also included a mood assessment scale (angry, irritated, calm, etc.) based on a "Feelings" scale developed by Byrne and Sheffield (1965). (See Appendix B for a copy of the evaluation form.) The evaluation form contained three items assessing Experimenter 1. Item 1: "How well did the above named research assistant perform in her/his role as an experimenter?" Responses were based on a scale with 20- point increments ranging from -100 ("poorly") to 100 ("excellently"). The remaining two items were based on similar scales with -100 being the most negative response and 100 the most positive. Item 2: "How would you rate his/her manner of interacting with others?" Item 3: "In your opinion, should this student be reappointed as a research assistant?" In addition three items were used to rate the experiment in general. Item 4: "Were you in any way mistreated by the experimenter in the experiment?" The scale for this question ranged from 0 ("not mistreated at all") to 100 ("mistreated a great deal") with 10-point increments. 77 The remaining items were based on similar scales (0 the most positive response, 100 the most negative). Item 5: "Did you think the demands made upon you in this experiment were excessive?" Item 6: "Are you in any way dissatisfied by the way you were treated in any aspect of the experiment?" In general, Items 1-3 have been used as a measure of retaliation against the first experimenter, while items 4-6 have been used to assess subjects’ annoyance with the experiment (Sapolsky & Zillmann, 1981; Zillmann et al., 1975; Zillmann & Cantor, 1976; Zillmann & Sapolsky, 1977). However, no reliability estimates have been reported for either of the evaluation scales. Before passing out the forms, Experimenter 2 explained that it was the faculty advisor’s policy in experiments like this to assess subjects’ opinions of the experiment and the way they had been treated by the research assistant conducting the experiment. Experimenter 2 also explained that the experimenter evaluation form was to be used by the faculty advisor in evaluating research assistants in terms of offering financial support for the upcoming semester. Ratings of the experiment were said to be used as a check on the experiment procedures. After the evaluation forms were filled out and collected, Experimenter 2 debriefed the subjects about the true purpose of the experiment and explained why deception had been used in the study. Comparisons between the provocation and non-provocation means were performed for items on the 78 Table 8 Factor Matrixes for the Provocation Scales Evaluations Item Factor 1 Factor 2 (Retaliation) (Annoyance) Performance 489 -.22 Interaction igg -.08 Reappointment 492 -.16 Mistreated 115 .45 Excessive demands -.12 155 Dissatisfied 151 .44 Eigenvalue 3.72 1.26 Percent of variance 62.1 21.0 Mood Assessment Item Factor 1 Factor 2 Factor 3 (Relaxed) (Irritable) (Anger) Relaxed 492 -.12 -.12 Angry -.07 -.10 $82 Anxious -.02 486 -.34 Pleased 410 -.19 -.45 Irritated -.25 #16 .48 Calm 428 -.01 .06 Eigenvalue 2.75 1.29 1.06 Percent of variance 45.9 21.6 17.7 note: Underlined factor loadings indicate items used in calculating reliabilities for the individual factors. 3 = 13. 79 experiment/experimenter evaluation form and the mood assessment scale. Factors were extracted using principal components analysis and varimax rotation. Cronbach’s alpha was used as an estimate of reliability. Table 8 presents the varimax-rotated matrix resulting from the factor analyses. The items from the experimenter/experiment form were converted to z-scores in order to compare the two scales. In addition the items for the experimenter rating scale were reverse coded so that the higher the score on all items, the greater the indication of verbal aggression. Two factors were extracted for the experiment/experimenter ratings. Factor 1 (Retaliation) had a reliability estimate of .91. Factor 2 (Annoyance) included only one item, excessive demands. Three factors were extracted for the mood assessment scale. Factor 1 (Relaxed) had a Cronbach’s alpha = .87. Factor 2 (Irritable) had a Cronbach’s alpha = .55. Factor 3 (Anger) consisted of only the angry item. Because of the low reliability found for the Irritable factor, it was decided to use the item measuring irritation individually. Results for the t-test comparisons are presented in Table 9. Differences between the provoked and non-provoked groups were significant for Retaliation factor as well as the individual items making up the factor. In addition, subjects in the provoked condition indicated significantly more irritation than subjects in the non-provoked condition (9 < .01). A similar trend was found for the anger item 80 Table 9 T-tests for Provocation Pre-test Measure Provocation Non-Provocation t Mean SD Mean SD D. 8 5 Experimenter /Experiment Scale Retaliation .50 .71 -.80 .13 3.98*** Performance .54 .90 -.86 .22 3.37*** Interaction .67 .56 -1.08 .21 6.58*** Reappointment .52 .92 -.84 .22 3.18*** Mistreatment .35 1.16 -.56 .00 1.73 Dissatisfied .41 1.09 -.67 .00 2.20* Annoyance Excessive -.20 .90 .32 1.17 .92 Demands Mood.Assessment Relaxed 4.54 1.70 5.27 1.86 .72 Angry' 1.63 .74 1.00 .00 1.85* Irritated 3.63 2.07 1.00 .00 2.79** *p < .10, **p < .05, ***p_ < .01 EQ£_- Ifigher z-scores for items on the EXPerimenter/Experiment Evaluation scale indicate greater verbal :aggression. Ratings for the Mood Assessment scale were based on a 7-point Likert scale with 1 = "not at all" and 7 == "very much." 81 although the significance level was only marginal (p = .09). It should be noted that the tests for homogeneity of variance (Cochran’s C and Bartlett-Box F) were significant for the Retaliation factor as well as for Items 1-3 evaluating the first experimenter. As can be seen in Table 9, the variances for these items were much greater in the provocation condition than in the non-provocation group. However, as Keppel (1982) points out, even flagrant violations of the homogeneity assumption do not appear to seriously distort the F distribution. Correlations between the factors and evaluation items on the provocation questionnaire are presented in Table 10. As can be seen, significant correlations were found between the irritated item and the Retaliation factor as well as the Table 10 Pearson Product-moment Correlations for Factors and Items for Provocation Pre-test RETAL PERF INTER REAP MISTR EXCES DISS REL IRRI RETAL PERF .86** INTER .89** .73* REAP .90** .96** .79** MISTR .81** .51 .62 .51 EXCES -.04 -.18 -.18 -.12 .15 DISS .85** .56 .69* .61 .85** .17 REL -.11 -.21 .10 -.19 -.02 .12 .03 IRRI .78** .81** .67* .75* .64* .02 .49 -.41 ANGR .32 .48 .17 .32 .15 -.31 .27 -.21 .24 IRETAL = Retaliation; PERF = Performance; INTER = Interaction; REAP = Reappointment; MISTR = Mistreatment; EXCES = Relaxed; Excessive demands; IRRI = Irritated; ANGR = * E < 005’ p < .01. DISS ‘ Dissatisfied; REL = Angry . H = 13. 82 individual items assessing the experimenter. No significant correlations were found between the angry item and any of the items on the experimenter/experiment evaluation form. It appears that subjects, while not significantly angered by the provocation treatment, were at least irritated by it. It also appears their irritation was directed specifically at the first experimenter as opposed to the experimental conditions in general (i.e. excessive demands, dissatisfaction with the experiment) Main experiment Subjects were 73 male undergraduate students enrolled in an introductory psychology course who participated in the experiment in order to earn extra credit toward their final grade in the course. All subjects had been informed during recruitment that there would be two separate experiments available to participate in, one of which might involve exposure to sexually explicit and/or violent materials and the other would involve a problem-solving exercise. Subjects were recruited in class and telephoned to arrange a time to participate in the study. Subjects participated in small groups of about 6 persons. The first phase of the experiment was administered by Experimenter 1, who arrived and administered the problem-solving exercise following the procedure described previously in the provocation pre-test section. At the end of the last exercise, Experimenter 1 collected the tests and left the 83 room. Experimenter 2 entered the room, passed out the consent forms and briefly explained the purpose of the second experiment. After collecting the consent forms, Experimenter 2 turned on the VCR containing the film segments and then left the room. After the last film segment was shown, Experimenter 2 reentered the room and informed the subjects that the evaluation forms for the first experiment had not been passed out and that now would be a convenient time to do so, before going on to the next phase of the present experiment. Experimenter 2 then explained the advisor’s policy as described in the previous section. After subjects completed these forms, they were given the feelings questionnaire, similar to the one administered in the pre-test and consisting of 15 items assessing subjects’ mood (See Appendix B). Experimenter 2 then collected the questionnaire and evaluation forms, administered the suspicion check and debriefed the subjects. In the debriefing the experimenter explained the purpose of both phases of the experiment, including why deception had been used and the relevant research findings from studies exposing subjects to sexually degrading and/or violent materials. A male experimenter was used for the second part of the experiment to show the films because using a female experimenter might have primed cognitions of female permissiveness, which in turn could have led to increased inter-gender aggression. Such was the case in the experiment conducted by Leonard and Taylor (1983). Having a female 84 experimenter show male subjects sexually violent or sexually degrading depictions might have given subjects the idea that she, and perhaps females in general, approved of such behavior. Such approval, or permissiveness, might therefore lead to increased aggressiveness in these conditions. 85 RESULTS The data were collected during two different time periods. Subjects participated in the experiment in small groups of about 6 persons. During the first session (n=48), subjects were tested on four consecutive week days. This session included two groups for each experimental condition. The next session took place about two weeks later and one group for each condition was tested (3:25). During the second session it became evident to the experimenters that subjects appeared, in general, more knowledgeable about the study. During the first session, subjects had frequently knocked on an adjoining door in the experimental room about 7 or 8 minutes into the study, presumably wondering where the first experimenter was. This did not happen during the second session. Also, subjects appeared less concerned with their performance on the problem-solving test and made comments, beforehand, about certain procedures used in the study ("Don't we have to do a couple of questionnaires before the film clips?"). During the coding of the questionnaires, there was also a noticeable difference in the experimenter evaluation scores between the two sessions with ratings in the second session appearing much more negative than in the first session. This was supported by statistical analysis in which differences between subjects in the first and second session reached the p < .001 level for each of the items specifically rating the 86 experimenter (performance, interaction and reappointment), with the ratings being much more negative in the second session. Because of a possible confound brought about by the passage of time between the first and second sessions (Students had talked to each other.), the sub-sample (n=48) of subjects in the first session was used to analyze the data. A factor analysis was performed on the data, following the same methodology described above, for the experimental evaluation and film questionnaires. Factor matrixes for both forms are presented in Table 11. For the experimental evaluation forms two factors were extracted. Factor 1 was labeled Retaliation (performance, interaction, reappointment), Cronbach’s alpha = .91. Another factor was extracted (Annoyance) consisting of the mistreatment, experimental demands and dissatisfaction items; however since this factor had a relatively low reliability (alpha = .60), it was decided to treat these items individually. For the feelings questionnaire, three factors were extracted. Items calm, happy and inhibited were eliminated because they exhibited relatively low loadings spread over more than one factor. Factor 1 for the mood questionnaire was labeled Negative Affect (angry, relaxed, afraid, depressed, disgusted, nauseated), Cronbach’s alpha = .87. Factor 2 was labeled Arousal (entertained, bored (reversed), aroused, excited), Cronbach's alpha = .86. One item, anxious, was dropped from Table 11 87 Factor Matrixes for the Experimenter/Experiment Evaluation Form and Feelings Questionnaire Experimental Evaluations Item Factor 1 Factor 2 (Retaliation) (Annoyance) Performance .86 .35 Interaction 12; .09 Reappointment 121 -.01 Mistreated -.02 .80 Excessive demands .08 .68 Dissatisfied .45 155 Eigenvalue 3.04 1.32 Percent of variance 50.8 22.0 Feelings Scale Item Factor 1 Factor 2 Factor 3 (Negative Affect) (Arousal) (Curious) Entertained -.41 .76 -.09 Anxious .29 Lg; .20 Bored -.09 -.62 -.02 Aroused -.08 121 -.11 Angry 125 .04 .26 Curious .17 .06 121 Relaxed -418 -.37 .23 Afraid 419 .07 .25 Depressed 484 -.12 .10 Disgusted 414 -.35 .00 Excited -.08 121 .06 Calm -.55 -.51 .36 Nauseate 48; -.06 -.02 Happy -.59 .52 .02 Inhibit .46 .35 .04 Eigenvalue 4.84 3.93 1.18 Percent of variance 32.2 26.3 7.9 Note: Underlined factor loadings calculate reliabilities. n = 48. indicate items used to 88 this factor due to its lack of face validity. A third factor contained only one item, curious, which was used as a filler item and not of interest in this study. Correlations among the dependent variables (evaluation and mood assessment items) used in this study are presented in Table 12. The only significant correlation between the evaluation items and the affect variables was that found between the Arousal factor and the mistreatment item (; = .37, p < .05). An analysis of variance was performed on the data using violence and degradation as independent variables. These results are presented in Table 13. Subjects in the high violence conditions were more retaliatory than those in the low violence conditions. While this finding was in the hypothesized direction (p = .08), this seemed to be the Table 12 Pearson Product-moment Correlations for Factors and items on Experimental Questionnaire and Feelings Scale RETAL REAPP INTER PERF MISTR DEMAN DISS NEG RETAL REAPP .92** INTER .93** .81** PERF .89** .70** .73** MISTR .16 .34* .11 -.01 DEMAN .19 .23 .18 .12 .24 DISS .49** .58** .45** .32* .40** .38** NEG -.03 -.02 -.05 -.00 .03 -.21 -.10 AROU -.20 -.12 -.19 -.25 .37* .18 .15 -.16 * p < .05, ** p < .01 RETAL = Retaliation; REAPP = Reappointment; INTER = Interaction; PERF Performance; MISTR = Mistreatment; DEMAN = Demanding; DISS Dissatisfaction; NEG = Negative Affect; AROU = Arousal. p 48. Table 13 2 x 2 (Violence x Degradation) Analysis of Variance 89 Source p Mean Sum of DF F_ Sig. Eta Squares of F Squared Retaliation Violence 3.01 1 3.19 .081 .07 LOW 25 -.25 High 23 .27 Degrading 2.30 1 2.44 .125 .05 Low 22 .25 High 26 -.21 Interaction .02 1 .02 .885 .00 LV,LD 11 -.02 LV,HD 14 -.42 HV,LD 11 .52 HV,HD 12 .04 Within 41.46 44 .94 Reappointment Violence 1.69 l 1.82 .185 .04 Low 25 -.19 High 23 .21 Degrading 4.11 1 4.42 .041 .09 LOW 22 .33 High 26 -.28 Interaction .00 1 .00 .970 .00 LV,LD 11 .13 LV,HD 14 —.45 HV,LD 11 .52 HV,HD 12 -.08 Within 40.98 44 .93 Interaction Violence 2.05 1 2.07 .157 .04 LOW 25 -.20 High 23 .22 Degrading 1.37 1 1.38 .245 .03 LOW 22 .19 High 26 -.16 Interaction .05 1 .05 .822 .00 LV,LD 11 -.05 LV,HD 14 -.33 HV,LD 11 .43 HV,HD 12 .03 Within 43.40 44 .99 90 Source p Mean Sum of DF F_ Sig. Eta Squares of F Squared Performance Violence 4.20 1 4.43 .041 .09 Low 25 -.29 High 23 .31 Degrading .90 l .95 .335 .02 Low 22 .16 High 26 -.14 Interaction .02 1 .02 .902 .00 LV,LD 11 —.15 LV,HD 14 -.39 HV,LD 11 .48 HV,HD 12 .17 Within 41.73 44 .95 Mistreated Violence 1.66 1 1.78 .190 .04 Low 24 .14 High 23 -.24 Degrading .02 1 .02 .878 .00 LOW 22 -.07 High 25 -.02 Interaction .01 1 .01 .910 .00 LV,LD 11 .10 LV,HD 13 .17 HV,LD 11 -.25 HV,HD 12 -.23 Within 40.21 43 .94 Experimental Demands Violence 7.13 1 8.08 .007 .15 LOW 24 .37 High 23 -.40 Degrading .50 1 .57 .456 .01 Low 22 .10 High 25 -.09 Interaction 1.48 1 1.68 .202 .03 LV,LD 11 .30 LV,HD 13 .44 HV,LD 11 —.1o HV,HD 12 -.67 Within 37.92 43 .88 91 Source p Mean Sum of DF F_ Sig. Eta Squares of F Squared Dissatisfied Violence .01 1 .01 .935 .00 Low 24 -.02 High 23 .02 Degrading 1.65 l 1.65 .213 .04 LOW 22 .20 High 25 -.18 Interaction .00 1 .00 .968 .00 LV,LD 11 .18 LV,HD 13 -.18 HV,LD 11 .22 HV,HD 12 -.17 Within 44.34 43 1.03 Arousal Violence 44.18 1 30.18 .000 .40 Low 24 4.95 High 23 3.00 Degrading 1.87 1 1.27 .265 .02 LOW 22 3.76 High 25 4.20 Interaction 1.31 1 .89 .350 .02 LV,LD 11 4.91 LV,HD 13 4.98 HV,LD 11 2.61 HV,HD 12 3.35 Within 63.07 43 1.47 Negative Affect Violence 8.74 1 6.58 .014 .10 Low 24 2.04 High 23 2.88 Degrading 22.63 1 17.06 .000 .25 Low 22 1.72 High 25 3.09 Interaction .01 1 .01 .935 .00 LV,LD 11 1.30 LV,HD 13 2.67 HV,LD 11 2.14 HV,HD 12 3.56 Within 57.05 43 1.32 LV Low violence; HV = High violence; LD = Low degrading: HD High degrading. 92 Note to chart on previous page: The factors and items on the experimental evaluation form were converted to z-scores and the items for the experimenter evaluation form were recoded so that the higher the score on these items, the greater the verbal aggression. The scores for the Arousal and Negative Affect factors were based on a 7-point Likert scale ranging from "Not at all" to "Very much." trend for all of the items contributing to the Retaliation factor. In fact, subjects in the high violence conditions were significantly more critical of the experimenter’s performance than subjects in the low violence conditions. Interestingly, subjects in the low violence conditions reported the experimental demands of the first experiment significantly more excessive than subjects in the high violence conditions. In sum, the first hypothesis that materials rated high in violence would produce more verbal aggressiveness than materials rated low in violence was only partially‘supported. No significant effects were found for the degrading conditions on the Retaliation factor. However, a significant effect was found for one of the items contributing to the measure, the reappointment of the experimenter as a research assistant. Subjects in the high degrading conditions were significantly more likely to agree that the first experimenter should be reappointed than subjects in the low degrading conditions. It should be noted here that previous researchers have found this item to be strongest indicator of retaliatory behavior among the items measuring retaliatory behavior (Sapolsky & Zillmann, 1981). The second 93 hypothesis that the degrading materials would lead to greater verbal aggressiveness was not supported. In fact, just the opposite appeared to be the case--the high degrading materials appeared to produce less verbal aggressiveness than the low degrading materials. The third hypothesis that materials rated high in both violence and degradation toward women would induce greater verbal aggressiveness than materials with one or none of these attributes was also not supported. No significant findings were reported for any of the interactions between violence and degradation on any of the dependent variables. Table 14 Post Hoc Analysis Comparing Degrading, Violent and Erotic Cpnditions Measure Degrading Erotic Violent F (HD,LV) (LD,LV) (HV,LD) n 14 11 11 Retaliation -.42a -.02"'b .52b 3.70* Performance -.40 -.15"b .48b 3.64* Interaction -.33 -.05 .43 1.98 Reappointment -.44a .13"'b .52b 3.90* *p < .05 LV Low Violence; HV = High Violence; LD = Low Degrading; HD High Degrading. Note. Higher z-scores for factors and items on the Experimenter Evaluation scale indicate greater verbal aggression. Within each row, means having no letter in their superscripts in common differ at p < .05 by Tukey’s test for comparisons between pairs of cell means. 94 After analyzing the data it became apparent that the greatest discrepancy among the conditions was between the HV-LD (violent) and LV-HD (degrading) conditions. A ppst hpg comparison of these two conditions, using the LV-LD (erotic) condition as a control, and their effect on the variables most directly assessing the experimenter’s effectiveness was undertaken to further explore this relationship. The results of this analysis are presented in Table 14. On all but the interaction item, subjects in the HV-LD (violent) condition exhibited significantly more retaliatory behavior than subjects exposed to the LV-HD (degrading) condition. However neither condition was significantly different from the erotic, or "control" condition. 95 DISCUSSION Conceptual Analysis No significant effects were found for either independent variable, degradation or violence, on the retaliatory factor nor was any interaction found between degradation and violence. In fact, these two variables appeared to have distinct, somewhat opposite tendencies. For example, male subjects exposed to highly violent materials were significantly more likely to depreciate the female confederate’s performance--one of the items used to measure retaliatory behavior. Contrary to predictions, subjects exposed to highly degrading materials, compared to subjects in the low degrading conditions, were significantly more likely to recommend a female confederate’s reappointment-— perhaps the single most direct measure of retaliatory behavior. This, of course, was contrary to the hypothesized effect of depicted degradation on verbal aggression. How can such a finding be explained? One possibility is suggested by the arousal-affect model proposed by Zillmann and his colleagues and shown to decrease aggressiveness after exposure to arousing materials (Zillmann & Bryant, 1984; Zillmann & Sapolsky; 1977). In fact, Zillmann and Bryant (1984) relied on this model to explain their finding that the more subjects were exposed to nonviolent, degrading materials the less aggressive they became toward a same-gender confederate. The less arousing and repulsive the films were to subjects, the less 96 aggressive they became. However, in the present study the relationship between the evaluation measures and subjects’ affective responses was found to be minimal--the only significant correlation being between Arousal and the mistreatment item--an item for which no effects were found. In Zillmann and Bryant’s study (1984), the authors simply noted a correspondence between arousal/affect and aggression, but did not report any significant correlations. In the present study, no comparable pattern was evident between the affective and evaluation responses. In fact, subjects reported more negative affect after viewing the segments rated high in degradation than subjects exposed to the low-degrading segments. There were no differences in arousal between the high and low degrading groups. According to the arousal-affect model, this should have led, if anything, to greater aggressiveness on the part of those subjects exposed to the highly degrading portrayals, not less. Another explanation may lie in the portrayal, in some types of degrading pornography, of the male, who seems to be depicted as a somewhat passive recipient of female-initiated sexual activity. This would appear to be characteristic of the "availability" theme described by Zillmann and Bryant (1982) and others. In such instances female characters are typically portrayed as "raving, slobbering nymphomaniacs" (Brownmiller, 1980, p. 32) and "eager to accommodate any and every imaginable sexual urge of any man in the vicinity" 97 (Zillmann & Bryant, 1982). This being the case, the message imparted to male viewers may be that nonaggression, or passiveness, in relations with the opposite sex may be rewarding. As such, nonaggression could be their response, at least initially, when placed in a similar situation. In retrospect, one of the degrading stimuli, "Barbara Broadcast," may have contained a comparable message. In this segment an almost oblivious male waiter was the recipient of female-initiated oral sex. Subjects exposed to the degrading-only materials (LV-HD), which included "Barbara Broadcast," tended to exhibit the least aggressive responses. In fact, these subjects reported significantly less retaliatory behavior than subjects in the violence-only (HV-LD) condition. However such an explanation would not satisfactorily explain the possible contribution of the other three degrading segments, in which the male characters were rated relatively high on the dominance factor. The preceding does point out the need to more carefully _//gistfnguishmbetween various types of sexually explicit ‘ materials and, more specifically, degrading portrayals. As defined in the literature degrading materials seem to encompass a wide range of themes, from availability to objectification and dominance. While it is argued here that these themes have in common the lowering of the status of the female participant, the portrayal of the participants may vary. For example, in the availability theme the female is characterized by unbounded sexuality while the male 98 appears to be a somewhat passive recipient. In other themes, the male may be dominant while the female is, at least initially, somewhat passive or reluctant. In terms of an imitation effect, the depiction of the male, which has not received as much attention from researchers, as well as the portrayal of the female, may be an important determinant of subsequent behavior. Another possibility is that subjects may have thought that the depictions in the sexually violent segments were morally improper and thus guarded against an aggressive _fi.- -response. Such an occurrence has been theorized by Berkowitz (1993) who argues that if media aggression is perceived as improper or morally wrong then the depiction will not promote aggression-enhancing thoughts or actions. Disapproval of an aggressor’s behavior may actually activate inhibitions against aggression. As noted by Fisher and Grenier (1994) contemporary educational efforts to decrease anti-woman attitudes and behaviors could influence male subject’s responses in current research. In fact, much of the research reporting significant findings was conducted in the late 19705 and early 19805. An increased awareness of the negative consequences of such attitudes and behaviors on the part of subjects could result in null findings or at least "mask" the true effects of sexually violent and degrading materials. For example, Donnerstein et al., (1976) cite a number of studies reporting that subjects exposed to debriefings designed to dispel common myths about sexual 99 violence later (up to 7-8 months) have exhibited significantly less acceptance of similar types of myths. In addition, as noted by Malamuth (1984), recent documentaries and dramas concerning sexual violence such as "Cry Rape," "Why Men Rape," "A Scream of Silence," and pornography in general, "Not A Love Story," have been created to make the general public more cognizant of the myths surrounding these types of portrayals. In Canada, the heightened public awareness of the harmful effects of sexually violent and degrading pornography has prompted the Supreme Court to censor such materials (R v. Butler, 1992). As such, subjects’ disapproval of sexual violence and degrading behaviors may have inhibited their retaliatory responses. Put into a theoretical perspective, viewing degrading behavior towards females might depersonalize women in general and lead to subsequent aggression, except when the degrading behavior is perceived as morally improper, in which case subjects’ aggressive responses might be inhibited. There is ample evidence in the present experiment that male subjects were able to detect degrading behaviors. Their pre-test ratings did not differ from those of female subjects, and they exhibited significantly more negative affect towards the high-degrading films compared to the low- degrading segments. A5 a result, they may have guarded against making an aggressive response towards the female confederate. Future research could address this proposition by the administration of a pre-test assessing subjects’ 100 attitudes toward women, the assumption being that the more negative their attitudes, the less likely they would be able to perceive degrading behaviors and guard against making an azgjessive response. It may also be possible that degrading portrayals do not lead to increased aggressive behavior. In Zillmann and Bryant’s (1984) study of prolonged exposure and in this study (reappointment item), subjects exposed to degrading materials subsequently exhibited less aggressive behavior. This contrasts with the McKenzie-Mohr and Zanna (1990) study in which exposure to nonviolent pornography was found to increase subjects’ willingness to treat females as sex objects (measured by how close subjects sat next to a female confederate and how much they recalled about her physical features). One distinction between these studies is that the aggression measures in the first two studies were constructed so that subjects were aware that their responses might have harmful consequences for the female confederate (either by negative ratings or the infliction of physical pain). On the other hand, the type of behavior exhibited by subjects in the McKenzie-Mohr and Zanna study did not pose harmful consequences for the confederate. This being the case, exposure to degrading materials may lead males to relate to women as sexual objects, but without the intention of harm that might accompany more aggressive actions such as sexual violence or coercion. As mentioned previously, regarding women as sexual objects, while offensive to some, 101 can be viewed as an efficient means for the perpetuation of the species (Zillmann & Bryant, 1989). In addition, since the control group in McKenzie-Mohr and Zanna’s study was shown a video of a House of Commons (Canada) question period, the somewhat important question of whether or not sexually explicit materials per se (erotica) would have lead to similar results was not addressed. Theoretically, instead of the target being depersonalized and subjects disinhibited, depictions of degrading behavior may result in the behaviors being imitated by viewers. Such behaviors may not be perceived to have harmful consequences, in fact the typical portrayal in pornographic materials is just the opposite--women seem to enjoy such treatment. As such the potential impact of degrading materials on society may be much more subtle than that of violent portrayals. As McKenzie-Mohr and Zanna (1990) note, the social implications of having males relate to women as sex objects are relatively unexplored in the research, although intuitively such behaviors and attitudes would seem to foster antisocial activities such as sexual harassment and discrimination. Considering the above discussion, the theoretical perspectives leading to the hypotheses tested in this study need further elaboration. Berkowitz's (1974) aggressive cue theory was used to predict subjects’ responses to the violent depictions and subsequent aggression opportunity. This theory posits that individuals take on an aggressive cue value if they can be associated with observed acts of 102 violence. Showing the female as a victim of violence should increase the aggressive-eliciting properties of females in general. Therefore viewing such portrayals should increase subsequent aggression toward females. This theoretical approach was partially supported by the data in this study in that the trend for exposure to the violent portrayals was increased retaliatory behavior (p = .08). In addition, one of the items used to measure verbal aggressiveness, ratings of the female confederate’s performance, was found to be significantly more negative for subjects in the violent conditions (p < .05). Significant findings may have been limited by some of the procedural limitations, loss of subjects and provocation procedure for example, discussed in the following section. However, it appears that the theoretical approach used here to predict effects from exposure to violent portrayals is a reasonable one. Perhaps the most parsimonious theoretical explanation for findings for the degrading conditions would be that of social learning theory (Bandura, 1977)4 This theory describes a modeling process whereby individuals observe and imitate media portrayals based on the likelihood of being rewarded. For example, when degrading behavior is depicted as rewarding, viewers are apt to imitate the behavior in similar circumstances. If passive behavior is depicted as rewarding in some pornographic materials, for example the passive male being serviced by the wildly euphoric nymphomaniac, then the chances of that behavior being 103 imitated should be increased. If degrading behavior is depicted as having positive consequences, for example the female is treated solely as a sex object during sexual activity by the male, then that behavior is more likely to be imitated, especially if the viewer believes the behavior will bring about the same rewarding result. This might explain why after viewing degrading pornography viewers might exhibit sexist behavior toward a female but not necessarily aggressive behavior-~defined here as intending to inflict harm upon another. Obviously if such behavior is thought to be morally wrong, it would not be imitated. The dehumanization process, while still a legitimate approach and certainly not disproven here, does not appear to offer as thorough an explanation for the results in this study. If subjects thought the degrading behavior was morally wrong (and certainly this is not a given at this point), then the dehumanization process should theoretically have been able to overcome that. This is because the dehumanization process has been used to explain how behavior that would be considered morally offensive, for example the Nazi’s extermination of Jews during World War II, can be carried out. As Kelman (1973) argues, "to the extent that the victims are dehumanized, principles of morality no longer apply and moral restraints ... are more readily overcome" (p. 48). If the female characters in this study were truly dehumanized, then any moral restraints against aggressing against them should be overcome. 104 The only effect from the experiment evaluation items was that subjects in the low violence conditions (degrading and erotic films) rated the experimental demands as more excessive than subjects in the high violence conditions. Such a finding seems at odds with the theoretical concepts discussed previously in which depictions high on violence would be expected to produce greater annoyance. One explanation for this discrepancy, is that subjects may have simply compared the demands of the first part of the experiment (the problem-solving study) with that of the second part (film exposures). In the two conditions that led subjects to express the most displeasure with the first study were also the same two that subjects reported the most positive affect to. Subjects in these conditions reported being more aroused than subjects in the other conditions. These subjects may have compared the general feeling of pleasure experienced in the second experiment with the somewhat negative experience of the first experiment in having the experimenter arrive late, the test being difficult and not having enough time to finish, and rated these demands more as being more excessive than those subjects whose level of arousal was not significantly increased during exposure to the high violence films. Finally, subjects’ pre-test ratings of the film segments indicated a relationship between violence and degradation. In fact the correlation between the two proved to be highly significant (r = .64, p < .01). It appears that 105 while degrading materials can be something other than violent, violent materials may be perceived as at least moderately degrading. Apparently subjects believe that a violent attack is to some extent degrading for the victim. However there may be some limitations, not present in this study, in perceiving violent portrayals as degrading. Violence against an inanimate object, for example, may not be perceived as degrading at all since no person is victimized. In violence depictions in which two protagonists are portrayed in a somewhat even struggle against each other there also may not be a perception of degradation. However when one party is portrayed as relatively defenseless and is victimized by another, similar to the depictions used in this study for the violent conditions, then the perception seems to shift toward the victim being degrading. The greatest shift appears to come about for sexually violent portrayals, in which the female is most intimately violated, and apparently most degraded. In the present study it appeared that perceptions of violence and degradation, for reasons discussed above, had contrasting tendencies. Depictions rated high on violence tended to increase subject’s verbal aggressiveness while depictions rated high in degradation tended to decrease verbal aggression. However, in regard to degrading materials, given the paucity of previous research and the inconsistent findings reported here, further investigation is need to draw more definitive conclusions. 106 MeasurementZProcedurai Limitations Other factors, relating to measurement and procedural considerations, may also have influenced the results of this study. One area of the study that proved somewhat problematic was the provocation procedure. Because subjects could not be insulted due to the ethical considerations previously discussed, the provocation used in this study was relatively mild compared to other studies in which insults or physical aggression were used to provoke subjects. As a result, male subjects may not have been disinhibited enough to aggress. This possibility is supported to some extent by the pre-test provocation results in which subjects were significantly irritated (p < .05), but only marginally angered (p < .10). A related factor may have been the stimuli selected for the experiment. Because violent X-rated films are censored in Canada, only R-rated films could be used for the sexually violent (HV-HD) condition. As mentioned previously, a common theme in x-rated violence is that the female victim eventually becomes aroused and ends up enjoying the sexual activity. Such a portrayal may not be as prevalent in R- rated films. For many of the films previewed for this study, sexual violence was depicted as having negative consequences for the female. In the ones selected, the female’s reaction to the sexual aggression was ambivalent at best. She was certainly not the unwilling victim turned into "a raving, slobbering nymphomaniac" described by Brownmiller (1980, p. 107 32). As Donnerstein and Berkowitz (1981) have shown, if subjects are not angered (and they may not have been to the extent needed in this study), then only a positive ending will induce an aggressive response. This being the case, subjects in the sexually violent (HV-HD) condition may not have reacted aggressively, in part, because the female victim’s reaction was not overtly positive. Another procedural consideration was subject loss and the resulting loss of power in detecting treatment effects. For example, for the Retaliation factor in Table 14, an acceptable degree of power (.80), given the derived means and total error variance for the analysis, would have required 18 subjects per treatment condition (Keppel, 1982). The harmonic mean for the three conditions considered in Table 14 was less than 12, resulting in a relatively low degree of power (.60) for the analysis, indicating the possibility of a type II error (failing to reject the null hypothesis when the alternate was true). Previous studies’ subject pools have ranged widely, from 49 subjects (Linz, Donnerstein & Penrod, 1984)) to 160 subjects (Zillmann & Bryant, 1982) to over 400 subjects (Check & Guloien, 1989) with typically four experimental conditions having been used. The present study falls at the low end of that range (n = 48) making it relatively less powerful than most other studies in the field. If the present study had more power, it may have been possible to determine whether, compared to 108 the erotic control group, violent portrayals increased aggression and/or degrading nonviolent portrayals decreased aggression. The following procedural recommendations should be considered in future research endeavors. A provocation condition should be used so that subjects are clearly angered. It may not be sufficient for subjects to be irritated or simply predisposed to downgrade the experimenter. The question for ethics committees is whether the benefits of investigating the possible anti-social effects of sexual violent and degrading depictions on viewers outweigh the risks to subjects’ self-esteem. The ongoing concern over these types of portrayals as well as the adequacy of the proposed debriefing procedure would seem to offer a compelling argument in favor of the potential benefits gained. Another recommendation is that the data should be collected in as short a time span as possible. Participants in this study may have talked between sessions, resulting in substantially more negative evaluations. This suggestion would appear to be particularly appropriate for experiments involving sexually explicit materials, since the temptation to talk about the stimuli, novel to say the least, may prove irresistible for many. Po '0 m l'ca 'ons To a large extent, exposure to degrading materials has been shown to lead primarily to the reinforcement of sexist 109 attitudes and behavior (McKenzie-Mohr & Zanna, 1990; Wyer et al., 1985). As such they may not pose a threat to society (Zillmann & Weaver, 1989). In addition, many types of media offerings have been found to contain sexist images and perpetuate sexist attitudes. For example, Signorielli (1989) has found that women have been portrayed in a sexist fashion in prime-time television programs over the past two decades. Exposure to such programming has been positively correlated with more stereotypical views of women (See Gunter, 1986, for a summary). As discussed previously, the rape trivialization finding in Zillmann and Bryant’s (1982) study was replicated by Weaver (1987) who used excerpts from R- rated movies. In effect, degrading or sexist materials in many contexts have been found to have what some may consider antisocial effects. A similar argument can be made for sexually violent materials since an increase in aggressive behavior after exposure seems to depend on the female’s reaction to the sexual violence and not the sexual violence per se. Only when the female is aroused does depicted sexual violence lead to increased aggression. Such depictions are not limited to X-rated materials, as illustrated by Check and Malamuth's (1981) field study in which commonly available, R-rated movies portraying sexual violence with positive outcomes were found to increase male subjects’ acceptance of interpersonal violence against women. Indeed many studies of violent-only materials have found increases in subjects’ 110 aggressive behavior after exposure (See Berkowitz, 1993, for a review). Such findings are similar to that of Donnerstein and Berkowitz (1981). In addition, in the present study exposure to violent R-rated materials was found to increase verbal aggression relative to X-rated degrading materials. Paradoxically, in Canada, the R-rated violent materials, which were found to induce more harmful behavior, would not be banned while the X-rated degrading materials, which produced relatively less aggressive behavior, would be censored. Taken as a whole, these findings call into question the selective targeting of degrading and violent X-rated materials for increased prosecution and restriction. Since other, more mainstream, materials have been found to depict comparable behaviors and perpetuate similar views, the harmfulness approach to pornography, advocated by some feminists and embraced by the Canadian Supreme Court and the Meese Commission, seems misguided. If harmfulness is the concern, then it seems a broad range of materials should be considered since messages of violence and degradation have been found in x-rated as well as more mainstream media offerings. In addition, if subjects indeed found the degrading depictions to be morally wrong in the present study and thereby exhibited less aggressiveness, then perhaps the best procedure to follow would be to continue and/or increase educational efforts to counteract some of the common myths surrounding degrading and sexually violent portrayals (the recipient enjoys being 111 degraded or physically abused). Suggestions for Future Research One category that was revealed in the pre-testing, but not examined in this study, is characterized primarily by verbal abuse. While this type of depiction has been mentioned in previous research, its potential significance has been relatively unexplored. Content analyses of X-rated videos have focused primarily on aspects of physical violence, such as rape or bondage. However, verbal abuse may also be a significant component, as indicated by the analysis of X-rated films by Cowan et al. (1988) in which verbal aggression, characterized by "abusive, derogatory language, swearing at the person" was found in 20% of all sexually explicit scenes (G. Cowan, personal communication, March 18, 1991). While little attention has been paid to the imitation effects of verbal aggression (Greenberg, 1980), the potential importance of such effects can be seen by the impact of verbal abuse in interpersonal relationships. The effect of such abuse on an individual’s self-esteem and general physical and psychological well-being has been well documented (Kinney, 1994). As such, further research into this aspect of sexually explicit materials and media presentations in general seems warranted. Certainly, the perception of such presentations as violent lends support to the notion that they may have significant effects on anti- social attitudes and behaviors. In addition, based on 112 subjects’ perceptions in this study, it appears that Gerbner’s definition of violence may be too limiting and should be broadened to include verbal aggression as well as physical acts. As mentioned previously, suggestions for improving the procedural and measurement elements of the present study include using a stronger provocation technique, the use of X-rated, as opposed to R-rated, violent materials and compressing the time period for data collection. Another suggestion would be to measure pre-existing attitudes and behaviors before assigning subjects to experimental conditions. A similar procedure was followed by McKenzie- Mohr and Zanna (1990) who pre-tested male subjects on their sex role identification. An advantage of this method is that it permits more powerful statistical analyses to detect differences in means. For example, McKenzie-Mohr and Zanna found no overall effects for the analyses of variance used in their study. However, using planned comparisons these authors found that subjects characterized as masculine sex- typed were significantly more likely to treat women as sex objects after exposure to a pornographic video than similar- typed subjects in a control group. Another possible pre-test variable might be frequency of previous pornography consumption. Check and Guloien (1989) found this variable to be significantly related to various "anti-woman" attitudes such as rape myth acceptance, acceptance of violence against women, sexual callousness and reported likelihood of rape 113 and forced sex. In addition, exposure to violent and degrading materials was found to reinforce certain sexually aggressive attitudes (Check & Guloien, 1989). By pre-testing subjects on such variables, future researchers may be able to more consistently delineate the effect of degrading and sexually violent materials. 114 Description of Film Segments pr vioienge, Low degrading (Erotica) "Femme-The Dressing Room" A couple is shown kissing, caressing and touching backstage in'a theater dressing room. The sexual activity progresses from cunnilingus to intercourse. There is no dialogue during the scene, only music. "Urban Heat-Disco" A male and female are shown undressing each other. Sexual activities include kissing and touching with the male caressing the female’s breasts and genital area. The male character performs cunnilingus, then the female performs fellatio before the couple engages in intercourse with the male on top. Low violence, High degrading (Degrading) "Amanda by Night" Two vice squad detectives visit a massage parlor and decide the new girl needs to be "broken in." The female, who is nude, performs fellatio on both males at the same time. The sexual activity consists of double penetration (intercourse with one male and fellatio with the other). One male is shown chewing gum and partially dressed during the sexual activity while the other, after telling the female to turn over, comments, "You got to position them around, Fender, you know what I mean?" 115 "Barbara Broadcast" In one scene a female waitress performs fellatio on the male maitre d’ after she drops and breaks a tray full of dishes. In another scene a female customer requests to perform fellatio on a male waiter. After the waiter, who is expressionless during the sexual activity, ejaculates in her face she says, "Thy rod and thy staff, they comfort me." All the sexual activity takes place in the midst of a busy restaurant. High violenceI High degrading (Sexually violent) "Blue Velvet" The male character enters the room and disgustingly asks the female where his drink is, "You shithead...can’t you fucking remember anything?" She gets his drink and they sit opposite each other. The male then orders the female to spread her legs. The male character then tells the female not to look at him ("Don’t you fucking look at me!") and slaps her. He then throws her down on the floor, massages her genital area roughly and has intercourse with her. Next he hits her again, telling her not to look at him. The scenes ends with the male leaving and the female still lying on the floor. "Last Tango in Paris" The male character tells the female to get a stick of butter. While they are both sitting on an empty apartment floor, he opens the front of her pants, turns her over on 116 her stomach and puts a handful of butter on her bottom. They then have anal intercourse during which time the male demands she repeat an epitaph after him while the female is depicted as crying. High violence. Low degrading (Violent) "Dressed to Kill" The female character is shown leaving the apartment of someone she has spent the night with. As she is leaving she finds a medical report indicating the man she has just slept with has been diagnosed with a venereal disease. As she is riding the elevator down from the apartment, she remembers she has forgotten her ring. She rides the elevator back up and when the door opens again a person wearing a blond wig enters and with a razor-like knife proceeds to slash the female character with it. "Halloween" This entire scene is shot from the perspective of the perpetrator. A couple is shown sitting on a couch in the female’s house. They then go upstairs and a short while later the male friend comes down and leaves. The male perpetrator then goes to the kitchen, picks up a butcher knife and heads up the stairs to the female’s bedroom. Noting the messed up bed, the perpetrator spots the nude female who is sitting at her dresser and begins to stab her with the knife as she shouts, "Michael, no!" 117 Pre-test and Main Experiment Questionnaires Problem-solving Exercise INTRODUCTION This questionnaire contains questions about your background as well as various problem-solving exercises. Your answers on this questionnaire are completely anonymous. You will be given a certain amount of time by the experimenter to complete each section. The problem-solving items were designed to be somewhat difficult but please try to answer as many of them as you can. DEMOGRAPHIC INFORMATION Age Place of Birth: Country: Sex: Male Female Year in university lst 2nd 3rd 4th or higher What is your field of study? Are you: Black White Asian Indian Other What is your religious affiliation (if any)? 118 PART 1 Please answer the following items by writing in your answer in the space provided below each item. 1. A person spends one-eighth of his spare change for post cards and four times as much for a box of letter paper, and then has 90 cents left. How much money did he have at first? 2. A certain military division contains 3,000 artillery, 15,000 infantry and 1,000 cavalry. If each branch is expanded proportionately until there are in all 20,900 men, how many will be added to the artillery? 3. In a certain year, Manitoba produced 2/3 and Ontario produced 1/6 of all the iron ore produced in Canada. If all the other provinces combined produced 18 million tons that year, how many million tons did Manitoba produce that year? 119 PART 2 Below are 10 mixed-up sentences. Please write under each sentence what it would say if the words were straightened out. 1. nights seen the moon not be some can 2. cardinal cultivated the be virtues should 3. with that apple a and begin ant words acorn are 4. repeated call human for courtesies associations 5. friends in us disaster never appearances are 6. size now of guns use are great in 7. sadness lists great casualty cause 8. health necessary camp a is to clean 120 PART 3 Please write your answers to the following questions below each question. You may use the space provided below each question for your calculations. Please try to do as many as you can. 1. A grocer will offer a particular kind of fruit for sale only on the day it is delivered. The delivery schedule is the same each week. The grocer’s store is open for business Monday through Saturday only. Cherries are delivered only every other business day. Oranges are delivered every business day. Peaches are delivered on two non—consecutive business, but not on Monday, Friday or Saturday. Lemons are delivered on the same days as cherries but also on Saturdays. Grapefruit are delivered only on the two consecutive days on which lemons are delivered. What is the maximum number of days on which both lemons and peaches can be purchased? On which days are the fewest kinds of fruit mentioned above for sale? 2. A half tone is the smallest possible interval between notes. Note T is a half tone higher than note V. Note V is a whole tone higher than note W. Note W is a half tone lower than note X. Note X is a whole tone lower than note T. Note Y is a whole tone lower than note W. What is the relative order of the notes from the lowest to the highest? 121 3. A farmer plants only five different kinds of vegetables-- beans, corn, kale, peas and squash. Every year the farmer plants exactly three kinds of vegetables according to the following restrictions: If the farmer plants corn, the farmer also plants beans that year. If the farmer plants kale one year, the farmer does not plant it the next hear. In any year the farmer plants no more than one of the vegetables the farmer planted in the previous year. What would be a sequence of combinations for the farmer to plant in two successive years? If the farmer plants beans, corn and kale in the first year, what combinations must be planted in the third year? 122 Evaluatipns for ExperimenterZExperiment EXPERIMENTER EVALUATION Please indicate your answers to the questions below by checking the appropriate space. All answers are on a continuum from -100 (most negative) to +100 (most positive). Remember your responses are completely anonymous. Please answer these questions as honest as possible. Experiment Experimenter’s name 1. How well did the above named research assistant perform in her/his role as an experimenter? -100 +100 POORLY EXCELLENTLY 2. How would you rate his/her manner of interacting with others? +100 -100 Extremely Extremely UNPLEASANT PLEASANT and and DISCOURTEOUS COURTEOUS 3. In your opinion, should this student be reappointed as a research assistant? +100 -100 Definitely Definitely NOT YES 123 EXPERIMENT EVALUATION Please indicate your answers to the questions below by checking the appropriate space. All answers are on a continuum from 0 to +100. Remember your responses are completely anonymous. Please answer these questions as honest as possible. Experiment 1. Were you in any way mistreated by the experimenter in the experiment? 0 100 NOT at MISTREATED all to a GREAT MISTREATED EXTENT 2. Did you think the demands made upon you in this experiment were excessive? 0 100 NOT at EXTREMELY all EXCESSIVE EXCESSIVE 3. Are you in any way dissatisfied by the way you were treated in any aspect of the experiment? 0 100 NOT at EXTREMELY all DISSATISFIED DISSATISFIED 124 (used only for provocation pre-test) Please indicate by marking the following items in the appropriate space how you feel right now. 4. Relaxed Not at : all much Very 5. Angry Not at all much Very 6. Anxious Not at : all much 0. 0. Very 7. Pleased Not at all much Very 8. Irritated Not at all much Very 9. Calm Not at all much Very ADDITIONAL COMMENTS: 125 Pre-Test of Film Segments INTRODUCTION The purpose of this study is to measure your reactions to the video segments you will see. You will need to fill out part of the questionnaire after viewing each segment. Each film section on the questionnaire is the same and has three parts. The first part asks for your OVERALL impressions of the segment, the second part for your impressions of the FEMALE character(s) and the third part for your reaction to the MALE character(s). Remember there are no "right" or "wrong" answers--the only correct responses are those that are true for you. REMEMBER THE QUESTIONNAIRE IS BEING USED SOLELY FOR RESEARCH PURPOSES AND IS COMPLETELY ANONYMOUS. Work at a fairly high speed through this questionnaire. Do not worry or puzzle over individual items. It is your first impression we are interested in. At the same time, please do not be careless, because we want your true impressions. The ratings are based on a continuum from "not at all" to "very much." EXAMPLE: If you felt the video was very humorous you would mark the following item in this way: Humorous Not at : : : : : :__X Very all much Film i Your impressions of the tape OVERALL: Entertaining Not at : : : : : : Very all much Violent Not at : : : : : : Very all much 126 Sexually Arousing Not at : : : : : : Very all much Disgusting Not at : : : : : : Very all much Tranquil Not at : : : : : : Very all much Exciting Not at : . : : : : Very all much Brutal Not at : : : : : : Very all much Stimulating Not at : : : : : : Very all much Your impression about the way the FEMALE character(s) was portrayed in this video: Affectionate Not at : : : : : : Very all much Humiliated Not at : : : : : : Very all much Willing Not at : : : : : : Very all much 127 Submissive Not at : : : : : : Very all much Promiscuous Not at : : : : : : Very all much Dominant Not at : : : : : : Very all much Degraded Not at : : : : : : Very all much Aggressive Not at : : : : : : Very all much Your thoughts about the way the MALE character(s) was portrayed in the video: Affectionate Not at : : : : : : Very all much Humiliated Not at : : : : : : Very all much Willing Not at : : : : : : Very all much Submissive Not at : : : : : : Very all much 128 Promiscuous Not at : : : : : : Very all much Dominant Not at : : : : : : Very all much Degraded Not at : : : : : : Very all much Aggressive Not at : : : : : : Very all much 128 Promiscuous Not at : : : : : : Very all much Dominant Not at : : : ° : : Very all much Degraded Not at : : : : : : Very all much Aggressive Not at : : : : : : Very all much 129 Post—Exposure Mood Assessment Scale INTRODUCTION The purpose of this study is to measure your reactions to the video segments you have just seen. Your responses need to be based on how you feel at the present time. Work at a fairly high speed through this questionnaire. Do not worry or puzzle over individual items. It is your first impression we are interested in. At the same time, please do not be careless, because we want your true impressions. The ratings are based on a continuum from "not at all" to "very much." EXAMPLE: If you feel very confident right now you would mark the following item in this way: Confident X Not at : Very all much Remember there are no "right" or "wrong" answers--the only correct responses are those that are true for you. REMEMBER THE QUESTIONNAIRE IS BEING USED SOLELY FOR RESEARCH PURPOSES AND IS COMPLETELY ANONYMOUS. Now, go ahead and mark for the following items to indicate how you feel at the present time. 1. Entertained NOT AT : : : : : : VERY ALL MUCH 2. Anxious NOT AT : : : : : : VERY ALL MUCH 3. Bored NOT AT : : : : : : VERY ALL MUCH 4. Sexually Aroused NOT AT : : : : : : VERY ALL MUCH NOT ALL NOT ALL NOT ALL NOT ALL NOT ALL NOT ALL NOT ALL NOT ALL NOT ALL NOT ALL NOT ALL AT AT AT AT AT AT AT AT AT AT AT 130 5.Angry 6. Curious 7. Relaxed 8. Afraid 9. Depressed 10. Disgusted 11. Excited 12. Calm 10. Nauseated 11.Happy 12. Inhibited VERY MUCH VERY MUCH VERY MUCH VERY MUCH VERY MUCH VERY MUCH VERY MUCH VERY MUCH VERY MUCH VERY MUCH VERY MUCH 511591 2. Wh you t 4. Dj expel 131 Suspicion Check 1. What did you think these experiments were all about? 2. What did you think the hypotheses were (i.e., what did you think we were looking for, trying to study?). 3. During the experiment did you have any suspicions? 4. Did you think there any relationship between the two experiments? Alexanl Bandur N Bandur D IDUH Baron, Bem, ¢ 132 LIST OF REFERENCES Agreed, K.E. (1988). Donnerstein, Malamuth and Mould: The conduct of research and the nature of inquiry. The Journal of Sex Research, 24, 358-362. Alexander v. Thornburgh, 943 F.2nd 825, 829 (8th Cir. 1990). Bandura, A., (1977). Social Learning'flheory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A., Uhderwood, EL. & Michael, M.E. (1975). Disinhibition of aggression through diffusion of responsibility and dehumanization of victims. Journal of Research in Personality, 9, 253-269. Baron, R.A. (1974a). The aggression-inhibiting influence of heightened sexual arousal. Journal of Bersonaiity and Social Psychology, ;Q(3), 318-322. Baron, R.A. (1974b). Sexual arousal and physical aggression: The inhibiting influence: of ‘cheesecake’ and. nudes. Bulletin of the Bsychonomic Society, 3(5A) 337-339. /Baron, R.A. (1977). Human Aggression. New York: Plenum Press. / Aaron) R.A. (1979). Heightened sexual arousal and physical aggression: An extension to females. on a 0 ese in Personalit , ii, 91-102. £//’Baron, R.An & BeliJ, .A. (1977). Sexual arousal and aggression males: effects of type of erotic stimuli and prior provocation. Journai pf Eersgnaiity ang Sogiai Psychology, 15(2), 79-87. Bem, S.L. (1974). The measurement of psychological androgyny. Journal, of Consuipi g and Clinical Bsychoiogy, 42(2), 155-162. Bem, S.L. (1981). Gender schema theory: A_cognitive account of sex typing. Bsychologicai Review, 88(4), 354-364. Berkowitz, L. (1974). Some determinants of impulsive aggression: Role of mediated associations with reinforcements for aggression. psychelggical_gexiew..81. 165-176. Berkol Berk01 Brann Brann Brown Bryan Bryar Burt, Burt, Byrne Byrne Carl: Check 133 Berkowitz, L. (1993). Aggression: Its causesl conseggences and control. New York: McGraw-Hill, Inc. Berkowitz, L., Corwin, R. & Heironimus, M. (1963). Film violence and subsequent aggressive tendencies. Public Opinion Quarterly, 21, 217-22 Brannigan, A. (1987). Pornography and behavior: Alternative explanations. Journal of Communication, 21(3), 185-189. Brannigan, A. & Goldenberg, S. (1987). The study of aggressive pornography: The vicissitudes of relevance. Critical Studies in Mass Communication, 4, 262-283. Brownmiller, S. (1980). Excerpt on pornography from Against our will: Men. Women and Rape. In L. Lederer (Ed.), Take back the night: Women on pornography (pp. 30-34). New York: Morrow. Bryant, J. & Brown, D. (1989). Uses of pornography. In D. Zillmann & J. Bryant (Eds.), Pornography: Research advances and, gpolicy' considerations (pp. 25-55). Hillsdale, N.J. : Lawrence Erlbaum Associates, Publishers. Bryant, J. & Zillmann, D. (1979). Effect of intensification of annoyance through unrelated residual excitation on substantially delayed hostile behavior. Journal of Experimental Social Psychology, l5, 470-480. Burt, M.E.H. (1976). Use of pornography by women: A critical review of the literature. Case Westerp Reserve Jggrnal of Sociology, 2, 1-16. Burt, M.R. (1980). Cultural myths and supports for rape. Journal of Personalityjand.Socigl Psychology, 22(2), 217- 230. Byrne, D., Fisher, J.D., Lamberth, J. & Mitchell. (1974). Evaluations of erotica: Facts or feelings? Journal of Personality and Social Psychology, 29(1), 111-116. Byrne, D. & Sheffield, J. (1965). Response to sexually arousing stimuli as a function of repressing and sensitizing defenses. Journal pf Apnomal Psycholggy, E, 114-118. Carlson, M., Marcus-Newhall, A. & Miller, N. (1989). Evidence for a general construct of aggression. Personality and Sogial Psychglogy Bullegin, l5, 377-389. Check, J.V.P. & Guloien, T.H. (1989). Reported proclivity for coercive sex following repeated exposure to sexually violent pornography , nonviolent dehumanizing pornography, 134 and erotica. In D. Zillmann & J. Bryant (Eds.), Eornography; Research advances and poligy considerations (pp. 159-184). Hillsdale, N.J.: Lawrence Erlbaum Associates, Publishers. Check, J.V.P., Heapy, N.A. & Iwanyshyn, O. (1985). A survey of Canadians’ attitudes regarding sexual content in the media. The LaMarsh research programme reports on yialence and. conflict. resolution. Toronto, Ontario: York University. Check, J.V.P. & Malamuth, N.M. (1986). Pornography and sexual aggression: A social learning theory analysis. In M.L. McLaughlin (Ed.), Communication yearbook . Beverly Hills, CA: Sage. Christensen, F. (1986). Sexual callousness re-examined. Journaliof Communication, 35(1), 174-184. Cowan, G. & Dunn, K.F. (1994). What themes in pornography lead to perceptions of the degradation of women? The Journal of Sex Research, 2l(1), 11-21. Cowan, G., Lee, C. Levy, D. & Snyder, D. (1988). Dominance and inequality in x-rated Videocassettes. Esycholagy pf Woman Quarterly, l2, 299-311. Canadian Criminal Code. (1985). R.S.C. 1985, c. C-46, s. 163. Donnerstein, E. (1980). Aggressive erotica and violence against women. Journal of Personality and Social Esychology, 22(2), 269-277. Donnerstein, E. & Barrett, G. (1978). Effects of erotic stimuli on male aggression toward females. Journ of Personaliry and Social Psychology, 35(2), 180-188. Donnerstein, E. & Berkowitz, L. (1981). Victim reactions in aggressive erotic films as a factor in violence against women. our al 0 ers al't ' 710-724. Donnerstein, E., Donnerstein, M. & Evans, R. (1975). Erotic stimuli and aggression: facilitation or inhibition. Journal of Personality and.Social Esychology,.12(2), 237- 244. Donnerstein, E. & Hallam, J. (1978). Facilitating effects of erotica on aggression against women. gourpal at W. 35(11) . 1270-1277- Donnerstein, E. & Linz, D. (1986, December). The question of pornography. Psyahglggy Today, pp. 56-59. Donne] Dun081 Dunca: Feshb Fishe Fishe Fishe Garc: Gerb: Gree Grif Grif 135 Donnerstein, E., Linz, D. & Penrod, S. (1987). The question of pornography: Research findings and policy implications. New York: The Free Press. Duncan, D.F. (1991). Violence and degradation as themes in ‘adult videos.’ Psychological Reports, 22, 239-240. Duncan, D.F. & Nicholson, T. (1991). Pornography as a source of sex information for students at a southeastern state university. Psychological Reports, Q2, 802. Feshbach, S. & Malamuth, N. (1978). Sex and aggression: Proving the link. Psychology Today, pp. 111-117, 122. Fisher, J.L. & Harris, M.E. (1976). Modeling, arousal and aggression. The Journal of Social Psychology, 100, 219- 226. Fisher, W.A. & Byrne, D. (1978). Sex differences in response to erotica? Love versus lust. Journal of Personality ang Social Psychology, 22(2), 117-125. Fisher, W.An & Grenier, G. (1994). Violent pornography, antiwoman thoughts and antiwoman acts: In search of reliable effects. The‘Journal of Sex Research, 22(1), 23- 38. Garcia, L.T. & Milano, L. (1990). A content analysis of erotic videos. Journal pf Psychology & Human Sexuality, 2(2), 95-103. Gerbner, G. & Gross, L. (1976). Living with television: The violence profile. Journal of Communication, 22(2), 176- 199. Glad Day Bookshop Inc. v. Deputy Minister of National Revenue (Qustoms and Excise) (1992), 90 D.L.R. (4th) 527 (Ont. Gen. Div.). (Slascock, J. & LaRose, R. (1993). Dial-a-porn recordings: The role of the female participant in male sexual fantasies. Journal pf Broadgaating ang Elacrronic Meaia, 21(3) , 313- 324. Greenberg, B.S. (1980). Life on television. Norwood, NJ: Ablex. Griffitt, W. (1973) . Response to erotica and the projection of response to erotica in the opposite sex. Jonrnal pf Experimental Research in Personality, a, 330-338. <3;Jriffitt, W. & Kaiser, D.L. (1978). Affect, sex guilt and the rewarding-punishing effects of erotic stimuli. Jgurnal of Gros Gunt Hans Har1 Hat: Her Hil Jaf Jaf Jar Kel Kin Kra Leg 136 Personality and Social Psychology, 22(8), 850-858. Gross, L. (1983). Pornography and social science research: Serious questions. Journal of Communication, 22(4), 107- 111. Gunter, B. (1986). Television and sex role stereotyping. London: J. Libbey. Hansen, M. (1992, March). Porn forfeiture stymied. ABA Journal, 28. Hartmann, D.P. (1969). Influence: of symbolically' modeled instrumental aggression and pain cues on aggressive behavior. Journal of Personality and Social Psychology, 11, 280-288. Hatfield, E., Sprecher, S. & Traupmann, J. (1978). Men’s and women’s reactions to sexually explicit films: A serendipitous finding. Argnives of Sexual fienaviar, 1(6), 583-592. Herrell, J.M. (1975). Sex differences in emotional responses to "erotic literature." gournal of Consulting and Qlinical Psychology, 22(6), 921. Hill, J.M. (Summer 1987). Pornography and degradation. Hypatia, 2(2), 39—54. Jaffe, Y. & Berger, A. (1977). Psychologigal Raports, 21, 335- 336. Jaffe, Y., Malamuth, N., Feingold, J. & Feshbach, S. (1974). Sexual arousal and behavioral aggression. on na of Eeraanality and Social Esycnology, 22(6), 759-764. Jarvie, I. (1991). Pornography and/as degradation. Inrarnational Jonrnal 9f Lag ang Esychiarry, 12, 13-27. Kelman, H.C. (1973). Violence without moral restraint: Reflections on the dehumanization of victims and victimizers. Journal pf Social lssnes, 22(4), 25-61 Kinney, T.A. (1994) . An inductively derived typology of verbal aggression and its association to distress. Hanan Qommunication Researcn, 21(2), 183-222. Kramer, R. (1992) R, y. Burlar: A new approach to obscenity law or return to the morality play? Criminal Law Quarterlx. 35. 77-90. Leonard, K.E. & Taylor, S.P. (1983). Exposure to pornography, permissive and nonpermissive cues, and male aggression 137 toward females. Motivation and Emation, 1, 291-299. Lewin, T. (1992, February 2). Canada court says pornography harms women and can be barred. New York Times, pp. A1, B9. Linz. D. (1989). Exposure to sexually explicit materials and attitudes toward rape: A comparison of study results. Journal of Sex Research, 2a(1), 50-84. Linz, D. & Donnerstein, E. (1988). The methods and merits of pornography research. Journal of Communication, 22(4), 180-184. Linz, D., Donnerstein, E. & Penrod, S. (1984). The effects of multiple exposures to filmed violence against women. Journal of Communication, 22(3), 130-147. Linz D., Donnerstein, E. & Penrod, S. (1988). Effects of long- term exposure to violent and sexually degrading depictions of women. Journal of Personality and Social Esychology, 22(5), 758-768. Longino, H.E. (1980). Pornography, oppression and freedom: A closer look. In L. Lederer (Ed.), Take back the nignt: Women on pornography (pp. 40-54). New York: Morrow. Malamuth, N.M. (1983). Factors associated with rape as predictors of laboratory aggression against women. gournal.of Personality and Social Esychology,.22(2), 432- 442. Malamuth, N.M. & Ceniti, J. (1986). Repeated exposure to violent and nonviolent pornography: Likelihood of raping ratings and laboratory aggression against women. Aggressive Behavior, 12, 129-137. Malamuth, N.M. & Check, J.V.P. (1981). The effects of mass media exposure on acceptance of violence against women: A field experiment. Journal of Research in Personality, 12, 436-446. Malamuth, N.M. & Check, J.V.P. (1985). The effects of aggressive pornography on beliefs in rape myths: Individual differences. Journal of Research in Eersonality, 12, 299-320. Malamuth, N., Feshbach, S. & Jaffe, Y. (1977). Sexual arousal and aggression: Recent experiments and theoretical issues. Jaurnal pf Social Issues, 22, 110-133. Malamuth, N.M. & Spinner, B. (1980). A longitudinal content analysis of sexual violence in the best-selling erotic MCDE McKe Meyc M052 Mou Pal Pea Pre Sal; 138 magazines. The Journal of Sex Research, 12(3), 226-237. McDaniel, J.W., O’Neal, E. & Fox, B.S. (1971). Magnitude of retaliation as a function of the similarity of available responses to those employed by attacker. Psychonomic Science, 22(4), 215-217. McKenzie-Mohr, IL. & Zanna, M.P. (1990). Treating women as sexual objects: Look to the (gender schematic) male who has viewed pornography. Personality and Social Psychology Bulletin, 12(2), 296-308. Meyer, T.P. (1972). The effects of sexually arousing and violent films on aggressive behavior. The Journal of Sex Research, 2(4), 324-331. Mosher, D.L. & MacIan, P. (1994). College men and women respond to X-rated videos intended for male or female audiences: Gender and sexual scripts. The Journal of Sex Researcn, 21(2), 99-113. Mould, D.E. (1988). a critical analysis of recent research on violent erotica. The Journal of Sex Researcn, 22, 326- 340. Palys, T.S. (1986). Test the conventional wisdom: The social content of video pornography. Canadian Psychology, 21, 22-35. Peat, Marwick & Partners. (1984). A national population study of pornography and prostitution, Report #6. Canadians’ attitudes toward and perceptions of pornography and prostitution. Ottawa, Canada: Department of Justice. Press, A., Namuth, T., Agrest, S., Gander, M., Lubenow, G.C., Reese, M., Friendly, D.T. and McDaniel, A. (1985, March 18). The war against pornography. Newsweek, 58-67. R. v. Butler, [1992] 1 S.C.R. 452. R. v. Towne Cinema Theat e Ltd., [1985] 1 S.C.R. 494. R. v. Wagner (1985), 43 C.R. (3d) 318. Russell, D.E.H. (1980). Pornography and violence: What does the new research say? In L. Lederer (Ed.), Take back rne night; Women on pornography (pp. 218-238). New York: Morrow. Sapolsky, B.S. & Zillmann, D. (1981). The effect of soft-core and hard-core erotica on provoked and unprovoked hostile behavior. The rna of Se Research, 11(4), 319-343. 139 Schmidt, G. (1975). Male-female differences in sexual arousal and. behavior during' and after‘ exposure» to sexually explicit stimuli. Archives of Sexual Behavior, 2(4), 353- 362 Schmidtg G4 & Sigusch, V. (1970). Sex differences in responses to psychosexual stimulation by films and slides. The Journal of Sex Research, 2(4), 268-283. Schmidt, G., Sigusch, V. & Schafer, S. (1973). Responses to reading erotic stories: Male-female differences. Archives of SexuallBehavior, 2(3), 1973. Scott, J.E. & Cuvelier, S.J. (1987). Sexual violence in Playboy magazine: A longitudinal content analysis. Archives of Sexual Behavior, 12, 279-287. Senn, C.Y. (1993). The research on women and pornography: The many faces of harm. In D.E.H. Russell (Ed.), Making violence sexy:_Feminist views on pornography (pp. 179- 193). New York: Teachers College Press. Senn, C.Y. & Radtke, H.L. (1990). Women’s evaluations of and affective reactions to mainstream violent pornography, nonviolent pornography and erotica. Violence and Victims, 2(3), 143-155. Shemberg, K.M., Levanthal, D.B. & Allman, L. (1968). Aggression machine performance and rated aggression. gournal pf Experimental Research in Personality, 2, 117- 119. Signorielli, N. (1989). Television and conceptions about sex roles: Maintaining conventionality and the status quo. Sex Roles, 21, 341-360. Smith, D.D. (1976) . The social content of pornography. gonrnal pf Communication, 22(1), 16-24. Smith, T.W. (1987). The polls-a review: The use of public opinion data by the Attorney General’ 5 Commission on Pornography. Bublig Opinion Quarterly, 21, 249-257. Steinem, G. (1980). Erotica and pornography: A clear and present difference. In L. Lederer (Ed.), Take pack tha ' h ' 0 en or o a (Pp. 35-39). New York: Morrow. Stock, W. (1991). Qiffarenrial responses of women and men to pornography; a feninist interpretation. Paper presented at the annual meeting of the American Psychological Association, San Francisco. SS: 140 Thomas, S. (1986). Gender and social-class coding in popular photographic erotica. Communication Quarterly, 22(2), 103-114. U.Sl Code. (1988) Title 18, Section 1963. k-/U/.S. Department of Justice. (1986). Attorney general’s commission on pornography: Final report. Washington, DC: U.S. Government Printing Service. U.S. v. Pryba, 900 F.2nd. 748 (4th Cir.), cert. denied, 111 S.Ct. 305 (1990). Weaver, J.B. (1987). Effects of portrayals of female sexuality and violence against women on perceptions of women. Unpublished doctoral dissertation, Indiana University. Weaver, J.B. (1991). Are "slasher" horror films sexually violent? A content analysis. Journal of Broadcasting and Electronic Media, 22(3), 385-392. White, L.A. (1979). Erotica and aggression: The influence of sexual arousal, positive affect and negative affect on aggressive behavior. Journal of Personality and Social Psychology, 21(4), 591—601. Wolfe, B.M. & ZBaron, R.A. (1971). Laboratory' aggression related to aggression in naturalistic social situations: Effects of an aggressive model on the behavior of college student and prisoner observers. Psychonomic Science, 22, 193-194. Wood, W., Wong, F.Y. & Chachere, G. (1990). Effects of media violence on viewers’ aggression in unconstrained social interaction. Psychological 2ulletin, 109(3), 371-383. Wyer, R.S., Bodenhausen, G.V. &' Gorman, T.F. (1985). Cognitive mediators of reactions to rape. Journal of Eersonaliry and Social Psychology, 22(2), 324-338. Yang, N. & Linz, D. (1990). Movie ratings and the content of adult videos: The sex-violence ratio. Journal of Qommunication, 22(2), 28-42. Zillmann, D. (1971). Excitation transfer in communication- mediated aggressive behavior. gonrnal of Experimenral Social Psychology, 1, 419-434. Zillmann, D. (1984). Connections between sex and aggression. Hillsdale , N.J. : Lawrence Erlbaum Associates , Publishers . Zillmann, D. (1989). Effects of prolonged consumption of pornography. In D. Zillmann & J. Bryant (Eds.), 141 Pornography: Research advances and policy considerations (pp. 127-157). Hillsdale, N.J.: Lawrence Erlbaum Associates, Publishers. Zillmann, D. & Bryant, J. (1982). Pornography, sexual callousness, and the trivialization of rape. Journal of Communication, 22(4), 10-21. Zillmann, D. & Bryant, J. (1984). Effects of massive exposure to pornography. In N.M. Malamuth & E. Donnerstein (Eds. ) , Pornography and sexual aggression (pp. 115-138) . Orlando, Florida: Academic Press, Inc. Zillmann, D. & Bryant, J. (1986). A response. Journal of Communication, 22(1), 184-189. «Zillmann, IL. & Bryant, J. (1988). Pornography’s impact on sexual satisfaction. Journal of Applied Social Psychology, 12, 438-453. Zillmann, D., Bryant, J. & Carveth, R.A. (1981). The effect of erotica featuring sadomasochism and bestiality on motivated inter-male aggression. Personalitygand Social Psychology Bulletin, 1(1), 153-159. Zillmann, D., Bryant, J., Comisky, P.W. & Medoff, N.J. (1981). Excitation and hedonic valence in the effect of erotica on motivated inter-male aggression. European Journal of Social Psychology, 11, 233-252. Zillmann, D. & Cantor, J.R. (1976). Effect of timing of information about mitigating circumstances on emotional responses to provocation and retaliatory behavior. Journal of Experimental Social Esychology, 12, 38-55. Zillmann, D., Bryant, J., Cantor, J.R., & Day, K.D. (1985). Irrelevance of mitigating circumstances in retaliatory behavior at high levels of excitation. Journal of Research in Personality, 2, 282-293. Zillmann, D., Hoyt, J.L. & Day K.D. (1974). Strength and duration of the effect of aggressive, violent and erotic communications on subsequent aggressive behavior. Communication Research, 1(3), 286-306. Zillmann, D. &.Sapolsky, B.S. (1977). What.mediates the effect of mild erotica on annoyance and hostile behavior in males? Journal of Personality and Social Psychology, 22(8), 587-596. Zillmann, D. & Weaver, J.B. (1989). Pornography and men’s sexual callousness toward women. In D. Zillmann & J. Bryant (Eds.), Bornography; Research advances and policy 142 considerations (pp. 95-125). Hillsdale, N.J.: Lawrence Erlbaum Associates, Publishers. Zimbardo, P.G. (1969). The'human‘choice: Individuation, reason and order versus deindividuation, impulse and chaos. Nebraska symposium (n1 'motivation, Lincoln, Neb.: University of Nebraska. MICHIGAN STRTE UNIV. LIBRARIES 1NW”WWI“WWWINWHIWII‘IWHI 31293014057578