In J. Bryant and M.B. Oliver (Eds.), (2009). Media effects: Advances in theory and research. (3rd ed.). Erlbaum/Psychology Press
Effects of Sex in the Media
Richard Jackson Harris, Kansas State University, and Christopher P. Barlett, Iowa State University
Where do men and women, boys and girls, learn about sex? What is the impact of those influences? Throughout childhood, adolescence, and into adulthood people learn about sex from many sources, including parents, schools, friends, siblings, and media outlets such as movies, television, magazines, song lyrics, videos, and the Internet. For example, we may learn about French kissing from an older brother’s stories, orgasms from a pornographic movie, oral sex from an erotic web site, and rape from a television movie.
Sexual themes in entertainment have been around as long as fiction itself. Many classics were often highly sexual in content, such as Aristophanes’ Lysistrata, Chaucer’s Canterbury Tales or Shakespeare’s The Taming of the Shrew, all of which are filled with overtly sexuality and covert double entendres, some of which may be missed today due to the archaic language and the “classic” aura of these works. More broadly, sex has long been a part of popular culture. Roman gladiatorial contests sometime featured scantily clad women as combatants, and sex scandals, sexual entertainment, and young adults pushing the limits on acceptable dress and behavior have long diverted, and on occasion troubled, society.
According to a Time/CNN poll (Stodghill, 1998), 29% of U.S. teens identified television as their most important source of information about sex, up from 11% in 1986. Although the most-mentioned source (45%) was “friends,” only 7% cited parents and 3% cited sex education. In one study, 90% of Toronto adolescent boys and 60% of the girls (mean age =14) reported having seen at least one pornographic movie (Check & Maxwell, 1992, in Russell, 1998). Also, 29% of boys rated pornography as their most significant source of sex education, higher than schools, parents, books, peers or magazines (Check, 1995). Surveys of college men have shown that 35-55% report having consumed violent pornography in some form (Demare, Briere, & Lips, 1988; Garcia, 1986).
Throughout adolescence and early adulthood we continually learn about sex, and media are a major source of that information (Chia, 2006; Dorr & Kunkel, 1990; Sutton, Brown, Wilson, & Klein. 2002). Moreover, relative to other sources, media are becoming increasingly important (Check, 1995; Greenberg et al., 1993), especially women’s magazines and television (Kallipolitis, et al., 2004). The effects of this heavy consumption of sexually oriented media are the topic of this chapter. We begin by examining the nature of sex in the media, focusing on content analysis studies. The rest of the chapter presents a review of the research on how consuming sexually explicit media impacts sexual arousal, attitudes, and behavior.
THE NATURE OF SEX IN THE MEDIA
Types of Sexual Content
Sexually oriented media may encompass a wide variety of sources. Some materials in magazines, videos, films, and Internet web sites have labels like “erotic,” “pornographic,” “X-rated,” or “sexually explicit.” Pornography is big business, generating $13 billion just in the U.S. in 2006 (IT Facts, 2007). Although sex magazines have greatly declined in circulation since the mid-1990s, that drop has been more than compensated for by video sales and rentals, cable and pay-per-view TV, and especially the explosive growth of Internet pornography, producing over 20% of the total revenue in 2006.
Most scholars distinguish between violent sexual material, which portrays rape, bondage, torture, sadomasochism, hitting, spanking, hair pulling, and genital mutilation, and nonviolent sexual material. Further classifying the nonviolent sexual material is more difficult. Some nonviolent sexual material is entirely mutually consenting and affectionate (sometimes called erotica), depicting vaginal or oral intercourse in a loving, or at least non-coercive, fashion. On the other hand, some nonviolent sexual material is sexually dehumanizing, depicting degradation, domination, subordination, or humiliation. This nonviolent, but dehumanizing, material typically presents the woman with few human attributes besides body parts and sexual appetite. Although often verbally abused and degraded, she appears hysterically receptive and responsive to men’s sexual demands. The man appears in the sexually dominant position, and the woman is far more likely than the man to be more exposed or nude.
Sex in media is not limited to explicit portrayals of intercourse or nudity, however, but may include any representation that portrays or implies sexual behavior, interest, or motivation. Sex also occurs in many other places besides explicitly sexual materials. Many news stories, including reports of sex crimes, sex scandals, celebrity starlet social gossip, or tragic excesses like the Abu Ghraib prison abuses, involve sexual content. Sex is rampant in advertising, particularly for products like perfume, cologne, and after-shave, but also for tires, automobiles, and kitchen sinks. For example, one automobile ad on network television featured two women discussing whether a man’s choice of a car was related to the size of his penis (“I wonder what he’s got under the hood”). See Reichert and Lambiase (2003) for a set of papers on sex in advertising.
Since the advent of broadcast media in the 1920s, standards have usually been more conservative for radio and television than for print media, because it is easier to shield children from sexually oriented print media than from X-rated TV. With the advent of widespread cable and video technology, a sort of double standard has arisen, with greater acceptance of more sexual materials in video and premium cable channels than on network television. The logic appears to be that premium cable and rented movies are “invited” into the home, whereas network programming is there uninvited and accessible wherever a TV set is present. A greater problem is the easy availability of sexual materials on the Internet, which has virtually no effective restrictions (Ferguson & Perse, 2000). Although there is much interest in legally restricting children’s access to sexually explicit sites, there is considerable disagreement about what kinds of restrictions or blocking software would be both legal and effective, without blocking useful non-sexual sites like breast cancer information or art sites. According to a 2002 survey, it was reported that single males between the ages of 18-45 years old visited pornographic websites more frequently than any other demographic group did (Buzzell, 2005).
Turning to television, the most-studied medium, content analyses have shown that, although the sex on network television is not usually explicit, sexual talk and innuendoes are rampant, most often occurring in a humorous context. One extensive content analysis study found that 68% of TV shows on network and cable in 1999-2000 contained sexual content, with 65% containing talk about sex and 27% presenting physical sexual behaviors (Kunkel, Biely, Eyal, Cope-Farrar, Donnerstein, & Fandrich, 2003). References to premarital and extra-marital sexual encounters outnumbered references to sex between spouses by at least 6:1 (Greenberg & Hofschire, 2000), and as high as 24:1 for unmarried versus married partners in soap operas or 32:1 in R-rated movies with teen characters (Greenberg et al., 1993)! The latter study also found that nudity occurred in all R-rated films in its sample, with female exceeding male nudity by a 4:1 margin. Sex in media is largely without consequences. One study showed only 14% of the discussions about sex on prime-time TV contained any mention of risks or responsibilities of sex and only 3% of the portrayals of sexual behavior did (Cope-Farrar & Kunkel, 2002). For shows with “intercourse-related content,” the percentage of shows mentioning any risk or responsibility of sex rose from 14 to 26%, but that is still low (Kunkel, Eyal, Donnerstein, Farrar, Biely, & Rideout, 2007).
In 2007, a longitudinal meta-analysis of 25 content analyses (from 1975-2004) on sexual content appearing on U.S. prime-time network programming (NBC, ABC, CBS, and Fox) found a decrease in the frequency per hour for passionate kissing, touching and petting, and intercourse from the early 1990s to 2004. Interestingly, however, the amount of sex talk steadily increased from 1999 to 2004. This meta-analysis also found a significant positive correlation between the year and the amount of explicit sexual intercourse, although this type of content did not appear that often (0.025 occurrences per hour). Finally, results showed a recent increase in the frequency of unmarried intercourse and prostitution from 2000 to 2004 (Hetsroni, 2007).
Although content analyses of soap operas showed considerable sexual content in 1985, there was a 35% increase by 1994 (Greenberg & Busselle, 1996). Also in 1994, compared to 1985, there were more themes of (a) negative consequences of sex, (b) rejection of sexual advances, and (c) portrayals of rape. None of these three themes had been very common in the soaps of the 1970s and 1980s. Not surprisingly, R-rated movies and sex magazines had more explicit sex than appeared on television (Greenberg et al., 1993; Greenberg & Hofschire, 2000).
The major focus of this chapter is on sexually explicit materials, including, though not limited to, what is generally called “pornography” or “erotica,” both violent and nonviolent. The term pornography is highly value laden, however, and as such is rather scientifically imprecise. Thus, we will most often refer to such materials as “sexually explicit” rather than “pornographic,” although that term is so widely used it cannot be completed avoided. When we consider effects of sex in the media, however, we need to look more widely than only at what is typically considered “pornography.”
EFFECTS OF CONSUMING SEXUAL MEDIA
Although many might wish it otherwise, sex, even very explicit sex, does sell. Sexually oriented print, video, broadcast, and Internet materials are highly profitable commercially, a condition which in itself ensures their continued presence. Aside from these economic effects, three major classes of effects of exposure have been identified, namely arousal, attitudinal changes, and behavioral effects. See Gunter (2002), Huston, Wartella, and Donnerstein (1998), Linz and Malamuth (1993), Malamuth (1993), Malamuth and Impett (2001), Mundorf, D’Alessio, Allen, and Emmers-Sommer (2007), Oddone-Paolucci, Genuis, and Violato (2000), and Pollard (1995) for more detailed reviews of various types of effects and media.
Research on effects of sex in the media has been guided by a variety of theoretical perspectives. Although these theories are not the focus of this chapter, the reader is referred to other chapters in this volume for thorough explications and reviews of these different perspectives: Morgan (Cultivation Theory), Bandura (Social Cognitive Theory), Petty, Priester, and Briñol (Elaboration Likelihood Model), Roskos-Ewoldson (Priming), and Rubin (Uses and Gratifications). Each of these perspectives has informed and guided certain areas of research on the effects of sexual media. These theoretical influences are alluded to below, although the focus of the rest of the chapter is on empirical findings on the effects of sexual media.
One straightforward effect of consuming sexual media is sexual arousal, the heightened physiological state that energizes sexual behavior. Arousal is measured in either of two ways. The most common measures are self-ratings (e.g., “How aroused are you?” on a 7-point scale). It may also be measured more directly, albeit more obtrusively, through various physiological measures such as electronic sensors measuring penile tumescence, vaginal lubrication, or temperature (thermography).
By most measures, men are typically more aroused by sexual media than are women, especially in response to sexually violent or dehumanizing materials (Malamuth, 1996; Murnen & Stockton, 1997). Sexual violence may be particularly arousing to sex offenders and other violence-prone men and even to “normal” men if the victim is portrayed as being aroused by the assault. Sexually coercive men are more physiologically aroused by slides or verbal descriptions of coercive sex than are “normal” men, who may have developed the ability to inhibit a sexual response in the presence of coercive cues (Lohr, Adams, & Davis, 1997).
Sexual arousal in response to stimuli that would not normally be arousing may be learned through classical conditioning. For example, Rachman and Hodgson (1968) classically conditioned heterosexual men to be sexually aroused by women’s boots by pairing the boots with nude female photos, thus providing a model of how sexual “turn-ons” can be learned. This process could account for the vast individual differences in the specific stimuli that arouse people sexually. Through different experiences, people have all been conditioned to respond to different stimuli through associations with those we love. Myers (2007) reports a friend who was turned on by the smell of onion breath, because his first girl friend loved to eat onions and he has associated that odor with kissing her, through the process of classical conditioning. Because of its connection with a particular person, someone may be aroused by a certain perfume or cologne, type of clothing, or specific behavior. Media provide many of the images and associations for such conditioning.
The degree of arousal need not be highly correlated with the degree of explicitness. Sometimes people are actually more aroused by a less sexually explicit story than a more explicit one. A scene which cuts suddenly from a bedroom one night to the next morning may sometimes be more arousing than a more explicit version with the intervening night uncut! Censoring a sex scene may make a film more arousing because viewers can fill in their own scripts. When people are allowed to use their own imaginations to construct the ending of a romantic scene, they are more likely to construct a reality that is more arousing to them personally than if they view someone else’s idea of what is arousing. The individuality of sexual arousal is the concern that sex therapists have with certain sexual media from the Internet or adult video stores. It has been argued by Carnes (2001) that since the Internet has an unlimited number of websites that feature any sexual desire that the user wants, this leads to sexual arousal because the stimuli is “new.” For instance, because of Internet sex websites, a viewer can see images of any desired fantasy, many of which typically do not occur in most people’s sexual lives. These images are then “burned” into the brain and are fantasized about during sexual intercourse (Carnes, 2001).
Individual Differences in Viewers. Some early studies examined convicted rapists and found them to be aroused by both rape and consenting sex, whereas normal men were aroused only by the consenting sex (Abel, Barlow, Blanchard, & Guild, 1977; Quinsey, Chaplin, & Upfold, 1984), although subsequent studies did not find this consistent arousal effect in sex offenders (Baxter, Barbaree, & Marshall, 1986; Hall, 1989). Under some conditions, however, even “normal” college men were aroused by scenes of sexual violence. For example, men, though not women, were at least as aroused by a rape scene as by a consenting sex scene but only if the victim was portrayed as enjoying the rape and coming to orgasm (Malamuth & Check, 1983; Ohbuchi, Ikeda, & Takeuchi, 1994). The men were not aroused if the victim was shown to be terrorized. Using a similar design, Bushman, Bonacci, van Dijk, and Baumeister (2003) found that men scoring high in narcissism found a rape scene preceded by affection between the parties as more entertaining and more sexual arousing than low narcissists did. Yates, Barbaree, and Marshall (1984) showed that normal men were equally aroused by depictions of rape and consenting sex but only after they had been angered by a female confederate. Otherwise the consenting sex scene was more arousing. Finally, Bogaert (2001) found that the personality traits of dominance, Machiavellianism, psychoticism, and hypermasculinity were correlated with the likelihood of viewing erotica containing violence, child pornography, or women with insatiable sexual appetites but not with the likelihood of viewing erotica lacking in these themes.
The Gender Skew. Explicit sexual materials have traditionally been designed by men and for men. As such, they have a distinctly macho, hypermasculine orientation. Although magazines and videos show all varieties of heterosexual intercourse, they place little emphasis on associated foreplay, afterplay, cuddling, or general tenderness. Women are shown eagerly desiring and participating in intercourse, often with insatiable euphoria. There is little concern with the consequences of sex or the relational matrix within which most people experience it. Men are much more active seekers and users of sexual material than are women, with an estimated 71% of sex videos viewed by men by themselves (Gettleman, 1999). However, this cannot necessarily be assumed to be due to greater intrinsic male interest in sex; it may merely reflect the pornography industry’s extreme slant to the hypermasculine fantasy. Indeed, a few studies have shown women to have more positive reactions to sexual videos written and directed by women and for women (Mosher & Maclan, 1994; Quackenbush, Strassberg & Turner, 1994; Senn & Desmarais, 2004), although men appear to be more likely to seek out sexual media and be aroused by it, even after controlling for content (Malamuth, 1996).
An evolutionary psychology explanation for sex differences in sexual behavior (Buss, 1995; Malamuth, 1996, 1999) would argue that men seek a greater number of sexual partners, while women are more interested in a longer-term commitment from a mate to help raise the offspring. These ideas are consistent with observed findings that men seek out and use sexual media more than women and are generally more aroused than women by them, especially media that visually represent many different potential partners. Women, however, are less aroused than men by typical pornography, preferring more contextually based sexual expressions like romance novels.
The Catharsis Legend. One often hears the argument that consuming sexually explicit material facilitates the expression of sexual urges and thus decreases arousal. This invokes the construct of catharsis, the emotional release that follows the expression of an impulse. This popular idea comes most directly from psychodynamic models of personality, notably Freud. Applied to sex, the catharsis argument predicts that consuming sexual media relieves sexual urges, with the magazine or video, perhaps in conjunction with masturbation, becoming a sort of imperfect substitute for the real behavior. Although a catharsis argument has been used to support loosening restrictions on pornography (Kutchinsky, 1973) and has been reported by sex offenders as a strategy for reducing impulses for committing an offense (Carter, Prentky, Knight, Vanderveer, & Boucher, 1987; Langevin, Lang, Wright, Handy, Frenzel, & Black, 1988), the research support for catharsis is weak to nonexistent (Bushman, Baumeister, & Stack, 1999). Viewing sexual material increases, not decreases, sexual arousal, and, after viewing, one is thus more, not less, motivated to engage in sexual behavior. Thus consuming pornography in order to reduce sexual arousal is likely to have the opposite effect. Nor will it reduce the propensity to rape, which is driven by a power motive, not a lack of sexual fulfillment (Prentky & Knight, 1991). See Scheele and DuBois (2006) for a recent conceptual examination of the history and current status of catharsis theory.
Sex and Values. Many concerns about sexually explicit media involve the attitudes and values they convey. Repeated exposure to media with a more-or-less consistent set of messages may cultivate a worldview that increasingly reflects the perspective of the media (Morgan, this volume). For example, watching numerous sitcoms and movies where characters are routinely sexually active early in a relationship with little concern of consequences may cultivate acceptance of such a position in the viewer and thus weaken family-taught values against casual premarital sex. Increasing numbers of ads and movies with themes of coercion and sexual violence may desensitize readers to violence toward women. Such effects are especially likely to happen if the characters holding those values are respected characters with whom viewers identify. Sexual promiscuity by a prostitute is less likely to influence the values of a viewer than is similar behavior by a respected suburban mother.
One of the major social criticisms of pornography is that it is ideologically anti-women (Buchwald, Fletcher, & Roth, 1993; Russell, 1998), a concern especially leveled at violent and nonviolent dehumanizing pornography. It is usually women, not men, who are the playthings or victims of violence by the opposite sex. For example, one sex magazine showed a jackhammer in a woman’s vagina as the opening photo to a story “How to Cure Frigidity,” and another showed a photo spread of a gang rape turning into an orgy with the women appearing to be aroused by the assault. A sex video showed a woman’s breast tied and squeezed for the entertainment of men who were watching.
Sexual attitudes. A large body of research has shown effects on a variety of sexual attitudes and values after exposure to nonviolent sexually explicit materials. After seeing slides and movies of beautiful female nudes engaged in sexual activity, men rated their own partners as being less physically endowed, although they reported undiminished sexual satisfaction (Weaver, Masland, & Zillmann, 1984). Men even reported that they loved their own partners less after seeing sexually explicit videos of highly attractive models (Kenrick, Gutierres, & Goldberg, 1989). Men who saw a pornographic video responded more sexually to a subsequent female interviewer than did men seeing a control video, although this result was only found for men holding traditional gender-role attitudes (McKenzie-Mohr & Zanna, 1990). It is as if the voluptuous model has become the norm or “anchor” (Tversky & Kahneman, 1974) to which real people are compared.
Such effects are not limited to men. Relative to control groups, both men and women who watched weekly pornographic films later reported less satisfaction with the affection, physical appearance, sexual curiosity, and sexual performance of their real-life partners (Zillmann & Bryant, 1988a, 1988b). They also saw sex without emotional involvement as being relatively more important than did the control group, and they showed greater acceptance of premarital and extramarital sex and placed lesser value on marriage and monogamy. They also reported less desire to have children and greater acceptance of male dominance and female submission. Using the same methodology, Zillmann and Bryant (1982) found that participants who had watched sexually explicit films 1-3 weeks earlier consistently overestimated the frequency of oral sex, anal intercourse, sadomasochism, and bestiality in the general population, relative to perceptions of a control group seeing nonsexual films. Teens watching a heavy diet of daytime talk television with frank discussion of sexual topics later overestimated the frequency of such behaviors, relative to a low-viewing group (Greenberg & Smith, 2002). Heavy viewing of soap operas, prime-time dramas and overall television predict a lower sexual self-concept in young adult women (Aubrey, 2007). Such results reflect the cognitive heuristic of availability, whereby we judge the frequency of occurrence of various activities by the ease with which we can generate examples (Glassner, 1999; Tversky & Kahneman, 1973, 1974). Recent exposure to vivid media instances thus raises the estimation of the frequency of such occurrences in the real world.
The sexual material need not even be explicit or graphic to help shape attitudes. Bryant and Rockwell (1994) found that, compared to controls, adolescents who watched a heavy diet of highly sexual prime-time programs were more lenient in their judgment about sexual impropriety and how much a victim had been wronged, although these effects were greatly attenuated by open family communication and active critical viewing. One may not even need the pictures. In one study all-verbal print descriptions of sex (e.g., the Penthouse Advisor column) were actually more conducive than photos to fantasizing about one’s own partner (Dermer & Pyszczynski, 1978).
In a meta-analysis of studies examining the relationship of the exposure to sexual media to the acceptance of rape myths, Allen, Emmers, Gebhardt, and Giery (1995) concluded that experimental studies show a consistent positive effect between pornography exposure and rape myth acceptance, while correlational and field studies show only a very small positive or nonexistent effect. The relationship was consistently stronger when the pornography was violent than when it was nonviolent, although some experimental studies obtained effects with both types.
Alcohol consumption may enhance existing tendencies to either harshly judge or empathize with a female victim, although it generally decreased sensitivity to victim distress, especially so in “hypermasculine” men (Norris, George, Davis, Martell, & Leonesio, 1999). Alcohol can even affect women’s judgments. Women reading an eroticized rape description while intoxicated were less likely than a sober control group to label coercive sex events as rape (Davis, Norris, George, Martell, & Heiman, 2006).
Pornography, especially videos, may be consumed for one or more of four different purposes (Gunter, 2002). Sexual enhancement creates the mood for sex or gives ideas about specific behaviors. Diversion offers an escape from boredom. Sexual release stimulates sexual fantasies. Substitution replaces a sexual partner. Men are more likely than women to report using pornography for sexual release and substitution. Those who used it for substitution were more likely to show acceptance of rape myths, although those who used it for sexual release were actually less likely to accept the rape myths (Gunter, 2002; Perse, 1994).
Slasher Films: Sex + Violence in Mainstream Movies. Although the studies discussed above used sexually explicit materials, attitudinal effects are by no means confined to clearly pornographic materials. Consider the highly successful horror-film series such as I Know What You Did Last Summer, Halloween, Child’s Play, Texas Chainsaw Massacre, Friday the Thirteenth, Scream, The Ring, Hostel, and Nightmare on Elm Street, as well as many lesser known films of the last forty years. Many are extremely violent with strong sexual overtones. Clover (1992, p. 21) defines a slasher film as a “generative story of a psychokiller who slashes to death a string of mostly female victims, one by one until he is subdued or killed, usually by the one girl who has survived.” Although some, such as the Scream series and Scary Movie, are framed as “satires” of the slasher genre, it is not clear that the youthful audiences receive them very differently than the non-satirical films.
Although many slasher films have R ratings in the United States, others are released unrated or direct to video to avoid the “accompanied by parent” requirement of R-rated movies. Given that so many viewings of movies are in DVD format among youth (sometimes in “uncut” versions), the ratings are of limited use, however. Oliver (1993) found that punitive attitudes toward sexuality and traditional attitudes toward women’s sexuality were associated with high school students’ greater enjoyment of previews of slasher films. Although there has been some trend toward stronger, less victimized, female characters in recent films such as Urban Legend, I Know What You Did Last Summer, and Bride of Chucky, the effects of such portrayals remains untested. Gender clearly remains a very salient aspect of these films and men and women respond somewhat differently to them. When young adults wrote descriptions of the most memorable slasher film they had seen and described their emotional reactions to the films, Nolan and Ryan (2000) found that men more often wrote of themes of fear of strangers, while women more often wrote of fear of horrors in the home and in intimate relationships.
Linz, Donnerstein, and Penrod (1984; see also Linz, Donnerstein, & Adams, 1989) examined the attitudinal effects of slasher films. After male college students were initially screened to exclude those with prior hostile tendencies or psychological problems, the remaining men in the experimental group watched one slasher film per day over one week. All of the films were very violent and showed multiple instances of women being killed in slow, lingering, painful deaths in situations associated with much erotic content (e.g., a woman masturbating in her bath is suddenly assaulted and killed by an intruder with a nail gun). Each day participants completed some personality measures and questionnaires evaluating that film.
Over the week, the men became generally less depressed, less annoyed, and less anxious in response to the films. The films themselves were rated as increasingly enjoyable, humorous, and socially meaningful, and progressively less violent, offensive, and degrading to women. Over the week’s time, the violent episodes in general and rape episodes in particular were recalled as less frequent. Although these data provide clear evidence of desensitization in men, there is still the question of generalization to other situations.
To answer this question, Linz, et al. arranged to have their participants later observe a rape trial at the law school and evaluate it in several ways. Compared to a control group, men who had seen the slasher films rated the rape victim as having been less physically and emotionally injured. These results are consistent with those of Zillmann and Bryant (1984), who found that massive exposure to sexually explicit media by jurors resulted in shorter recommended prison sentences for a rapist. Using Linz et al’s methodology of rating movies followed by evaluation an “unrelated” rape trial, Weisz and Earls (1995) showed men and women one of four films: man raping a man–Deliverance, man raping a woman–Straw Dogs, nonsexual male aggression toward both men and women–Die Hard 2, and nonaggressive action film–Days of Thunder. They found strong desensitizing effects of the two sexually violent films in men but not in women. Interestingly enough, it did not matter whether a man (Deliverance) or a woman (Straw Dogs) was the victim; both films desensitized men to the female rape trial victim, though neither effect appeared in women. Such findings show that attitudes inculcated by seeing slasher films do indeed transfer to new situations.
There have been some methodological (Weaver, 1991) and conceptual (Sapolsky & Molitor, 1996) criticisms of this research, and some effects have not been fully replicated in later work (Linz & Donnerstein, 1988). Some have questioned Donnerstein and Linz’ conclusion of sharply different effects of viewing violent versus nonviolent sexual materials (Mundorf, et al., 2007; Weaver, 1991).
Nor are violent sexual themes confined to horror, or even R-rated, movies. For example, the 1995 PG-rated James Bond movie Goldeneye featured a villain who seduces men to have sex with her and then crushes them to death. It also contains scenes of seduction with very violent mutual battering as a sort of foreplay. The major concern with such films is the juxtaposition of sex and violence. In countries like India and Japan, rape and other acts of violence against women are even more standard entertainment fare in action-adventure films.
Recently, Internet pornography has begun to be of interest to researchers (see Griffin-Shelley, 2003 for a review). Two early experimental studies (Barak & Fisher, 1997; Barak, Fisher, Belfry, & Lashambe, 1999) failed to find a consistent effect of amount of Internet sex exposure on any of several measures of misogynistic attitudes. Later work, however, has produced clearer effects. In an extensive correlational study investigating the effect of Internet pornography on adolescent male attitudes, a significant correlation was found between the amount of Internet pornography viewed and recreational sexual attitudes (which included items, such as “It is OK to have sexual relationships with more than one partner”), but this effect was mediated by the realism of the pornography (Peter & Valkenburg, 2006). Thus, when the sexual actions on the Internet website looked as though they could happen in real life, teenage boys were more likely to have more liberal attitudes toward sex. Similarly, Lo & Wei (2005) found that exposure to Internet pornography was significantly correlated with attitudes toward extramarital sex and sexually permissive behavior (holding hands to sexual intercourse), such that, as the viewing amount increased, so did more accepting attitudes of infidelity, as well as the amount of self-reported sexual behavior by participants. A survey of college students found that viewing sexually explicit material on the Internet was significantly correlated with masturbating online, sending and receiving pornography online, and seeking new people online (Boies, 2002). The same study also sought the uses and gratifications for viewing pornography on the Internet, and responses indicated that men (more than women) use this form of media because they find it sexually arousing, to satisfy sexual needs, to fulfill sexual fantasies, and to satisfy curiosity about new sexual techniques (Boies, 2002).
Thus there is considerable evidence that exposure to erotica, in whatever medium, affects attitudes and values about sex and various sexual issues. We now turn to examining effects on behavior.