Television Programs

Although media options have expanded in recent decades and now include social media and video games, television still anchors many media diets, with estimates that American youth view TV nearly four hours a day (Rideout, Foehr, & Roberts, 2010), and Americans, overall, view five hours a day (Nielsen, 2011). Themes, storylines, dialogue, and jokes related to sex and sexuality are a prominent feature of this content. The most recent large-scale efforts to assess the prevalence of sexual content were conducted at the start of the new millennium, and published from 2004–2007. Fisher et al. (2004) coded 1,276 television programs representing a composite three-week sample (recorded in 2001) from 11 major networks: ABC, CBS, Fox, NBC, UPN, WB, BET, Cinemax, HBO, MTV, and Showtime. In total, 82.1% of the episodes studied contained at least one instance of sexual talk or behavior. In their analysis of 1,154 programs from the 2004–2005 television season, Kunkel, Eyal, Finnerty, Biely, and Donnerstein, (2005) found sexual content in 77% of primetime programs. Talk about sex was present in 68% of programs and sexual behavior was present in 35% of programs. Findings across multiple studies indicate slight changes from year to year in these rates, with both increases and decreases being reported.

Although sexual content is highly prevalent, it is not uniform, and evidence indicates that some types of sexual content are more prevalent than other types. More specifically, findings indicate that sexual talk is more prevalent than depictions of sexual activity (Kunkel et al., 2005). Also, when sexual behavior is depicted, it tends to be mainly kissing and flirting. For example, of the eight sexual behaviors coded by Fisher et al. (2004), physical flirting was in 49.5% of programs, kissing or touching in 48.6%, implied intercourse in 10.5%, and depicted intercourse in 3.9%. Looking more specifically at programming preferred by gay and lesbian youth, Bond (2015) found that of the ten categories of sexual behaviors coded, the most common sexual behaviors were LGB romantic kissing (31%), LGB physical flirting (29%), and heterosexual physical flirting (14%).

One type of sexual content that is consistently minimal or absent is content about the risk and responsibilities of sex, such as discussions of safe sex practices, depictions of condom use, or discussions about disease prevention. Fisher et al. (2004) reported that, of programs noted to feature sexual content, 2.9% contained messages of sexual patience, and 5.2% mentioned sexual precautions. Analyses performed by Eyal, Kunkel, Biely, and Finnerty (2007) of programs preferred by teens found that only 4% of sexual scenes contained risk and responsibility messages in 2001–2002, and 5% in 2004–2005. Rates were somewhat higher across all programs in their larger study, whereby 14% of programs with sexual content were found to contain risk and responsibility messages (Kunkel et al., 2005).

Second, sexual content has been found to vary greatly based on the genre. Analyses indicate that levels of sexual content are especially high in comedies and TV movies, and are much lower in children’s programming. For example, in their analysis of 1,276 television programs across 11 networks, Fisher et al. (2004) found sexual behavior in 66.8% of episodes overall, but in 100% of the comedy-dramas and television movies, and in 92.7% of the feature films; it was least prevalent in the children’s cartoons (20.7%), talk shows (28.4%), and news magazines (29%). Sexual talk was found in 76.5% of episodes, overall, and was most prevalent in the comedy dramas (100%), situation comedies (93%), feature films (93.2%), and variety/comedies (93%). It was least prevalent in the children’s cartoons (14.9%) and news magazines (41.9%).

Acknowledging the diversity of sexual content, some studies have looked at the presence of a particular type of sexual theme or script. One script examined is the presence of the heterosexual script and its components. The heterosexual script describes the courtship strategies, commitment orientations, and sexual goals considered appropriate for women and men in heterosexual relationships in (Kim et al., 2007). This script expects men to actively pursue sexual relationships, to objectify women, and to prioritize sex over emotion; conversely, women are expected to be sexually passive, to use their looks and bodies to attract men, to serve as sexual limit setters, and to prioritize emotions over sex (Kim et al., 2007). References to the heterosexual script have been noted to occur 15.5 times per hour in primetime television preferred by teens (Kim et al., 2007), and to appear in 11.45% of the interactions between characters on “tween”-oriented programming (Kirsch & Murnen, 2015). Analyses of reality dating programs indicate that references to men as always looking for sex occur 3.6 times per hour, and references to women as sex objects occur 5.9 times per hour (Ferris, Smith, Greenberg, & Smith, 2007). Particularly prominent are messages linking masculinity to sexual prowess and interest (Kim et al., 2007; Kirsch & Murnen, 2015).

Motion Pictures

Sexual content is a regular feature of motion pictures, most often depicted among characters who are white, heterosexual, and newly involved with each other (Gunasekera, Chapman, & Campbell, 2005; Hefner & Wilson, 2013). One of the largest projects to explore this issue, led by Bleakley, Jamieson, and Romer (2012), examined 855 films, which included 15 of the 30 top-grossing movies for each year from 1950 to 2006. Each film was coded for the presence or absence of sexual content in five-minute segments, and sexual content was defined to include kissing on the lips, nudity, sexual behavior, or sexual intercourse, implicitly or explicitly shown. Any sexual content emerging was rated for explicitness on a five-point scale. Several analyses have emerged from this large dataset, finding differences in movie sexual content by gender, year, and movie rating. Of the 855 films, 84.6% contained sexual content in at least one 5-minute segment, with lower levels in G-rated films (68.2%) than in R-rated films (88.3%) (Nalkur, Jamieson, & Romer, 2010). Explicitness of the sexual content increased with the ratings but did not increase over time. In terms of character gender, analyses indicated that across these 855 films, 57% of female characters were involved in sexual content compared to 30% of male characters (Bleakley et al., 2012).

Other research teams have focused on films of a particular genre, such as romantic comedies, or on films directed at a particular audience, such as teens. In their analysis of 90 films directed at teens that were released from 1980–2007 (30 from each decade), Callister, Stern, Coyne, Robinson, and Bennion (2011) found that 28% of films contained adult sexual activity, and 80% contained teen sexual activity, with an average of 6.5 teen sexual acts per film. The most common sexual activity depicted was passionate kissing. No differences were found over time in the overall prevalence of sexual acts, nor in the presence of sexual dialogue. In their analysis of the 52 highest-grossing romantic comedy films from 1998–2008, Hefner and Wilson (2013) found an average of 7.21 romantic ideal expressions per film and an average of 14.21 instances per film of challenges to romantic ideals. The most prevalent romantic ideal was the notion of soul mates/the one and only true love. Similarly, in their analysis of 40 top-grossing romantic comedies, Johnson and Holmes (2009) observed 3,470 relationship-oriented incidents. The largest category observed was kissing and the second largest category was compliments, most of which were expressed by a male character. Relationships were not always rosy, however, and incidents of deception, arguing, and of breakups were also reported. However, as observed with television content, there was little depiction in movies of safe sex practices, risks, or consequences (Callister et al., 2011).

Music Videos

Sexual imagery is also prevalent in music videos, noted to appear in 58.5% of music videos in one study (Turner, 2011) and in 84% of videos in another (Ward et al., 2013). The sexual acts appearing most frequently are sexual and suggestive dance, sexual objectification, and self-touching. For example, King, Laake, and Bernard (2006) examined the presence of 19 sexual behaviors for women and 16 sexual behaviors for men in 411 music videos appearing on four networks. The most frequent sexual behaviors for women were sexual dance, flirting, and caressing/stroking of self. For men it was hugging/embracing, sexual dance, and groping of own genital area.

These patterns have been found to vary based on artist gender and music genre. One consistent theme emerging is that in music videos, women and female artists are more often shown in sexual ways than men and male artists. More specifically, across several studies, analyses indicate that female artists and performers reveal more body parts, are more provocatively dressed, are more often sexually objectified, engage in more sexually suggestive dance, and engage in more sexual behavior than do their male counterparts (Aubrey & Frisby, 2011; Frisby & Aubrey, 2012; King et al., 2006; Turner, 2011; Wallis, 2011; Ward et al., 2013). For example, Aubrey and Frisby (2011) found that in comparison to male artists, female artists were significantly more likely to be provocatively dressed (35.3% of female artists vs. 5.2% of male artists) and were more likely to engage in sexually suggestive dance (31.4% vs. 4.2%). Analyses indicate that music genre also matters, such that rhythm and blues (R&B), rap/hip-hop, and pop music videos have been found to contain more sexual content than rock or country music videos (Frisby & Aubrey, 2012; Turner, 2011).

Music Lyrics

A few studies have examined sexual content in music lyrics, alone, outside of the visual format of the music video. Indeed, of six teen-oriented media formats examined, Pardun, L’Engle, and Brown (2005) found that sexual content was most prevalent in popular music, with sexual content emerging in 40% of the units coded (each line of lyrics). This level compares to 11% of units for TV programs and 12% for movies. Within the music data, the most common sexual content was references to relationships (52%), sexual innuendo (19%), and references to sexual intercourse (15%). In their analysis of 279 top songs from 2005, Primack et al. (2008) found that 37% of songs contained references to sexual intercourse. Of these references, 65% were classified as references to degrading sex, in which one partner (nearly always women) was objectified, the other had a voracious sexual appetite, and a heavy emphasis was placed on physical characteristics. Furthermore, 88% of the songs with degrading references to sex were sung by men.

Video Games

Although fewer studies have analyzed sexual portrayals in video games, the findings emerging paint a consistent picture. Analyses indicate that not only are women underrepresented in video games, but that when present, they are much more likely than men to be shown with a sexualized appearance or in sexually revealing clothing. This pattern has been demonstrated among gaming magazine advertisements and articles (Dill & Thill, 2007; Miller & Summers, 2007; Summers & Miller, 2014), on video game covers (Burgess, Stermer, & Burgess, 2007), and within the game (Downs & Smith, 2010). For example, Downs and Smith (2010) played 20 top-selling games, for 20 minutes, and encountered 489 characters, only 14% of whom were women. Coding all primary and secondary characters for eight variables representing overt sexuality, they found that, in comparison to male characters, a greater proportion of female characters were portrayed in sexually revealing clothing, were depicted as partially or fully nude (43% of female characters vs. 4% of male characters), were shown with unrealistic body proportions, and were shown wearing clothing that would be inappropriate for completing the task at hand.

Theoretical Explanations for Media Effects

Sexual media research has sought to determine whether exposure to media content related to sexuality can influence attitudes, beliefs, and expectations about sexuality and also whether these effects extend to sexual behaviors. This research draws on two main theoretical approaches. One approach is cultivation theory (Gerbner, Gross, Jackson-Beeck, Jeffries-Fox, & Signorielli, 1978), which posits that heavy consumption of television’s particular version of reality predicts viewers’ adoption of this version. Therefore, if television regularly portrays casual sex with minimal risks, cultivation theorists argue that frequent television consumers might grow to cultivate or adopt a comparable view, including adopting a permissive stance towards casual sex, dismissing possible health risks and consequences, and being more likely to engage in casual sex themselves. Traditional cultivation research relies on two key assumptions; first, that cultivated views of reality are caused by the volume of television consumed, and second, that television presents a unified, homogenous view of reality that reflects the mainstream values of society (Kahlor & Morrison, 2007). As these assumptions have been questioned by many researchers, it is common to see variations on the theory that examine genre or content-specific exposure to television rather than total TV exposure as a unified whole (e.g., Gottfried, Valla, Bleakley, Hennessy, & Jordan, 2013; Marron & Collins, 2009; Wright & Qureshi, 2015). Additionally, although TV remains the main focus of cultivation theory research, this theory has been applied to other media beyond television (e.g., Wright & Qureshi, 2015).

In addition to cultivation theory, Bandura’s social learning theory (Bandura, Ross, & Ross, 1963) and its later revision to social cognitive theory (Bandura, 2002) have been applied to research on media and sexual content in order to better understand the process of media influence. Social learning theory is based on the idea that observing relevant, attractive media models can shape viewers’ understandings of what behaviors and norms are socially appropriate and positively reinforced. In sexuality research, social learning theory is used to explain how adolescents may observe sexual content in the media and then model their own behavior after that content. Social cognitive theory adds to this approach by incorporating the agency of viewers in the engagement and interpretation of media (Bandura, 2002). The expectation of social cognitive theory is that viewers’ scripts, schemas, and normative beliefs are shaped by their engagement with media content, and that these beliefs and values lead to behaviors. Social cognitive theory would suggest that, if women are regularly exposed to messages that men are sex-driven and women are more sexually passive, they might internalize those sexual scripts, and perhaps ultimately limit their own sexual agency.

In addition to these approaches, researchers examining the role of media in sexual socialization have used broader ecological frameworks to understand the contexts in which socialization occurs (Chia, 2006; L’Engle, Brown, & Kenneavy, 2006; Price & Hyde, 2009). This work explores the unique contributions of different socialization agents (e.g., family, peers, schools, media, religion) in predicting sexual behaviors (Chia, 2006; L’Engle et al., 2006), embedding media into the larger social framework surrounding adolescents.

One contemporary ecological approach to the media’s role in sexual socialization is Wright’s (2011) 3AM model, which builds on sexual scripts theory (Gagnon & Simon, 2005) to propose a specific, testable model involving the acquisition, activation, and application of sexual scripts from media. This model incorporates influences, such as motivations for viewing (e.g., Ward & Rivadeneyra, 1999) and affective responses to media (e.g., van Oosten, Peter, & Valkenburg, 2015), which have been shown to impact the likelihood of adopting sexual scripts. Several researchers have applied this model to examinations of the media’s impact on sexual socialization (e.g., Ezzell, 2014; Kratzer & Aubrey, 2015; Weed, Nicholson, & Farris, 2015). For example, Braithwaite, Coulson, Keddington, and Fincham (2015) have used this model to examine the ways in which the effects of pornography viewing on sexual behavior are mediated by sexual scripts. The 3AM model and other ecological approaches that take into account contextual factors surrounding the adoption of sexual attitudes and behaviors depicted in media will likely continue to play a large role in sexual socialization research.

Effects on Attitudes Toward Uncommitted Sexual Exploration

The most substantial body of literature linking media and sexual attitudes explores if media use contributes to attitudes toward uncommitted sexual exploration. Higher amounts of exposure to television (Chia, 2006; Ward, 2002), prime-time television (Ward & Friedman, 2006), reality dating shows (Ferris, Smith, Greenberg, & Smith, 2007; Zurbriggen & Morgan, 2006), music videos (ter Bogt, Engels, Bogers, & Kloosterman, 2010; Zhang, Miller, & Harrison, 2008), movies (Ward, Epstein, Caruthers, & Merriwether, 2011), and media in general (Chia & Lee, 2008) are all associated with or predictive of more positive attitudes toward uncommitted sexual exploration. However, some research finds only conditional support for these associations, reporting that higher levels of media use are associated with more positive attitudes toward uncommitted sexual exploration for only female (ter Bogt et al., 2010; Ward, 2002) or only male (Vandenbosch & Eggermont, 2012a; Ward, 2002) participants, depending on the medium. Further, research considering exposure to a range of media forms—including total amount of television (Zurbriggen & Morgan, 2006), prime-time television (Ward et al., 2011), and music videos (Ward et al., 2011; Ward & Friedman, 2006)—fails to find a link between media use and attitudes toward uncommitted sexual exploration. Still, it is noteworthy that our review of the literature did not uncover any research demonstrating that higher levels of media use were associated with more negative attitudes toward uncommitted sexual exploration; if it was indeed the case that there was not an association between media use and attitudes toward uncommitted sexual exploration, one would expect a roughly equal number of studies to uncover positive and negative effects. Considered together, then, survey research suggests that higher amounts of media exposure are associated with an increased tendency to hold positive attitudes toward uncommitted sexual exploration.

In addition to highlighting the influence of mere exposure, survey research demonstrates that viewer cognitions also shape viewers’ attitudes toward uncommitted sexual exploration. Social cognitive theory suggests that we are more likely to adopt the attitudes expressed in media content if we perceive that content as realistic (Bandura, 2002). Consistent with this line of reasoning, correlational research demonstrates that people who perceive media content as more realistic hold more positive attitudes toward uncommitted sexual exploration (Chock, 2011; Ferris et al., 2007). The reasons we use media are also consequential cognitions. Here, research finds that consuming media to learn (Ward & Rivadeneyra, 1999), for companionship (Ward & Friedman, 2006), and to be entertained (Zurbriggen & Morgan, 2006), each predicts a greater tendency to hold more positive attitudes toward uncommitted sexual exploration. One recent paper (Bond & Drogos, 2014) suggests that our perceptions of sexual media content may in fact explain the positive association between exposure to sexual media content and attitudes toward uncommitted sexual exploration. Bond and Drogos surveyed college-aged viewers of the popular sex-laden reality television program Jersey Shore and found that heavier viewers of the program reported stronger parasocial relationships (i.e., feeling like their favorite Jersey Shore character was a friend) and wishful identification (i.e., the desire to be like a favorite Jersey Shore character), each of which predicted more positive attitudes toward uncommitted sexual exploration. Further, when the mediating influences of parasocial relationships and wishful identification were statistically controlled, the relation between Jersey Shore exposure and attitudes toward uncommitted sexual exploration dropped to non-significance.

The evidence for a causal link between media use and attitudes toward uncommitted sexual exploration, however, is relatively sparse. Experimental exposure to sexual media content typically produces no main effect on attitudes toward uncommitted sexual exploration, and this null finding has been repeated across media forms, including hip-hop music videos (Kistler & Lee, 2010), music (Sprankle & End, 2009), and television (Ward & Friedman, 2006; though Ward, 2002 finds effects for women). It may be that experimental effects only emerge under certain conditions. For example, although Taylor (2005) found no main effect of exposure to sexual media content on attitudes toward uncommitted sexual exploration, he found that those participants who believed the content they saw was more realistic expressed more positive attitudes toward uncommitted sexual exploration. The nature of sexual media portrayals may also be consequential: in one study, watching a teen drama that depicted sex with positive consequences had no effect on attitudes toward uncommitted sexual exploration, but watching a teen drama that depicted sex with negative consequences led participants to express more negative attitudes toward uncommitted sexual exploration (Eyal & Kunkel, 2008). Impressively, this effect was still evident two weeks after experimental exposure. Because media are more likely to feature positive consequences for sexual activity than negative ones (Eyal & Kunkel, 2008; Kunkel et al., 2007), the results from Eyal and Kunkel’s (2008) study suggest a possible explanation for the null findings described above: in selecting “typical” media texts to bolster the external validity of their studies, experimenters are likely not choosing media that feature sex with negative consequences.

Women Versus Men: The Heterosexual Script and Related Constructs

A second line of research considers how media use contributes to endorsement of components of the heterosexual script. As indicated earlier, the heterosexual script describes the dominant cultural scripts (i.e., normative expectations) for heterosexual romantic/sexual relationships, identifying different but complementary roles for women and men in these contexts (Kim et al., 2007). According to the heterosexual script, men are expected to actively pursue women, avoid emotional commitment, and value women primarily for their appearance. Women, by contrast, are expected to express interest in men using passive strategies like self-objectification, set sexual limits, and prioritize emotional commitment over sexual fulfillment. Implicit in this conceptualization is an understanding of male-female relations as inherently adversarial, because the roles women and men are expected to uphold are often at odds. Also implicit in this conceptualization is a sexual double standard, whereby men are normatively expected to have (and are thus rewarded for) uncommitted sexual encounters, whereas women are normatively expected to not have uncommitted sexual encounters and thus could expect to face social consequences for enacting the same behavior that would reap rewards for men.

Higher levels of regular media exposure tend to be associated with a greater tendency to endorse components of the heterosexual script, including the beliefs that men are sex-driven and women are sex objects. These associations have been demonstrated in studies considering total amount of television exposure (Ward, 2002), exposure to dating-themed reality television programs (Ferris et al., 2007; Zurbriggen & Morgan, 2006), and music video exposure (Ward et al., 2011), although some studies report significant links for women only (ter Bogt et al., 2010) or men only (Vandenbosch & Eggermont, 2011). Experimental research generally supports these correlational results, finding that experimental exposure to media that depict the sexual stereotypes described by the heterosexual script leads to increased endorsement of these beliefs (Ward, 2002; Ward & Friedman, 2006; Ward, Hansbrough, & Walker, 2005).

Media exposure is also linked to constructs conceptually related to the heterosexual script. Watching dating-themed reality television, for example, predicts an increased tendency to endorse adversarial sexual beliefs and the sexual double standard (Zurbriggen & Morgan, 2006). Similarly, frequent exposure to music videos is associated with an increased tendency to endorse the sexual double standard (Zhang et al., 2008), and experimental exposure to music videos depicting sexual stereotypes leads to increased expression of adversarial sexual beliefs (Kalof, 1999). More recently, a panel survey reported that adolescent girls’ exposure to music videos featuring male artists predicted increased acceptance of token resistance one year later (van Oosten et al., 2015). Token resistance is the belief that women often say “no” to sexual activity when they really mean yes, and is perhaps a logical outcome of the heterosexual script, which teaches women that they should avoid openly expressing sexual availability. The results from this panel survey are particularly concerning because acceptance of token resistance is a risk factor for sexual victimization (Krahé, Scheinberger-Olwig, & Kolpin, 2000).

As in the literature on attitudes toward uncommitted sexual exploration, our perceptions of and motives for using media have implications for how media affect us. Perceiving media as more realistic (Ferris et al., 2007) and identifying more strongly with media characters (Ward & Friedman, 2006) is each associated with stronger endorsement of heterosexual script components. Motives for consuming media that reflect more active modes of engagement (i.e., to learn, to be entertained, for companionship) are also associated with a greater tendency to endorse components of the heterosexual script, though often only for women (Ward, 2002; Ward & Friedman, 2006; Zurbriggen & Morgan, 2006). Consuming media out of habit, however, which reflects a more passive mode of engagement, is unrelated to these beliefs (Ward & Friedman, 2006). Thus, how we consume media matters as much as—if not more than—the amount of media we consume.

Perceptions of Others’ Sexual Behavior

Survey evidence links increased estimates of peers’ sexual behavior to higher levels of both total television exposure (Eggermont, 2005; Ward, 2002) and exposure to specific television genres, such as soap operas and music videos (Ward, 2002). More active forms of engagement with television content are also consequential, at least for women: among female viewers, believing television portrayals are true-to-life, identifying with same-sex television characters, and watching television to learn or to be entertained are all associated with increased estimates of sexual activity among peers (Ward, 2002; Ward & Rivadeneyra, 1999). However, experimental findings paint a different picture, with laboratory exposure to sexual prime-time television (Taylor, 2005) and music videos (Sprankle & End, 2009) failing to affect estimates of peers’ sexual behavior, and exposure to relationship-themed reality TV leading to increased estimates of peers’ sexual experience among male viewers only (Vandenbosch & Eggermont, 2011).

Romantic Relationships: Hopes, Expectations, and Experiences

Media also help shape what we hope for, expect, and experience in romantic relationships. Consumption of romantic-themed television and movies—which generally portray romance idealistically—tends to be associated with holding more idealized beliefs about and expectations for romantic relationships (Hefner & Wilson, 2013; Lippman, Ward, & Seabrook, 2014; Segrin & Nabi, 2002), although exposure to more conflict-ridden portrayals of romance (i.e., sitcoms) is associated with a decreased tendency to hold idealistic romantic beliefs (Lippman et al., 2014). Here, too, more active forms of engagement with media, including perceived realism (Lippman et al., 2014) and viewing to learn (Hefner & Wilson, 2013), are significant predictors of holding more idealistic relationship beliefs, and in fact relate more strongly to these beliefs than mere exposure.

Media also affect the types of romantic partners who interest us, as well as our expectations about the nature of the sexual activity we will engage in with them. Both correlational (Eggermont, 2004) and experimental (Carpentier, Knobloch-Westerwick, & Blumhoff, 2007) research indicate that media consumption is predictive of assigning more weight to sexual, appearance-based characteristics in partner preferences. Further, among women, watching more sexual television content predicts expecting sexual activity to occur earlier in a relationship (Aubrey, Harrison, Kramer, & Yellin, 2003). Among men, watching sexual television content (Aubrey et al., 2003) is associated with expecting to engage in a greater variety of sexual activities in a relationship.

In addition to affecting preferences expressed with regard to hypothetical relationships, media are also associated with the functioning of established relationships. Within established relationships, heavier consumption of relationship-themed programming predicts increased relationship conflict, decreased relationship satisfaction, and decreased relationship commitment (Osborn, 2012; Reizer & Hetsroni, 2014). Again, perceived realism is an even stronger predictor of negative relationship outcomes than mere exposure, predicting increased expected and actual relationship costs (e.g., restrictions on freedom), increased perceived quality of alternative partners, and decreased relationship commitment (Osborn, 2012).

Cross-Sectional Survey Research

A number of studies examine the relation between media exposure and sexual behavior or intentions using one-time surveys. Looking at overall television exposure, Barr, Moore, Johnson, Merten, and Stewart (2014) found that high screen time (three or more hours per day) was associated with increased risky sexual behaviors, including sexual initiation prior to age 11 and having three or more sexual partners. Specifically examining the sexual content in the media (rather than overall media exposure), other studies have found that increased levels of exposure to sexual media content was associated with increased reports of intentions to have sex, light sexual behavior (kissing, holding hands), and heavy sexual behavior (oral sex, intercourse) (Fisher et al., 2009; L’Engle et al., 2006; Somers & Tynan, 2006). There are some studies whose data do not fit the pattern (e.g., Marron & Collins, 2009), but these studies may be outliers. Overall, this body of research indicates a consistent positive connection between exposure to media, especially sexual or romantic media content, and sexual behaviors.

Longitudinal Survey Research

To gain insight into the process of sexual socialization, many researchers have sought to explore the relation between media exposure and sexual behaviors over time through the use of longitudinal surveys. Findings regarding overall exposure to television and viewers’ sexual behaviors are mixed. In some cases, heavier screen time (two or more hours) was predictive of sexual initiation at later assessments (Ashby et al., 2006). Yet in several others, general exposure to television was not related to later behaviors (e.g., Chandra et al., 2008; Collins et al., 2004). However, many of these studies focused on pregnancy, rather than sexual initiation, which may explain these findings (e.g., Collins et al., 2004).

The majority of longitudinal studies are not concerned with overall media exposure but instead examine exposure to specific sexual media content that includes sexual behaviors, sexual talk, or other sexual content (Bleakley et al., 2008; Brown et al., 2006; Pardun et al., 2005; Price & Hyde, 2009). These studies have consistently found that heavier exposure to sexual content predicts earlier or heavier sexual activity one year later (Bleakley et al., 2008; Brown et al., 2006; Pardun et al., 2005; Price & Hyde, 2009). One study suggests that exposure to sexual content in early adolescence may advance sexual initiation by nine to seventeen months (Collins et al., 2004). However, there are studies that report null results (e.g., Gottfried et al., 2013), and some researchers have questioned commonly used measurement and analysis techniques (Collins, Martino, Elliot, Miu, & Rand Corporation, 2011; Steinberg & Monahan, 2010). Finally, in some of the longitudinal studies that have been done, evidence has emerged supporting a reciprocal, non-recursive relation between sexual content exposure and sexual activity (Bleakley et al., 2008; Frison, Vandenbosch, Trekels, & Eggermont, 2015).

Experimental Approaches

There is some limited experimental research testing contributions of media exposure to sexual behavior, but ethical concerns related to manipulating and measuring sexual behavior limit this approach. Generally, experimental work in this area examines and manipulates specific moderators, such as the likeability of media characters or genre of media content, while limiting measurable outcomes to intentions and hypothetical situations (Boot, Peter, & van Oosten, 2015; Moyer-Guse, Mahood, & Brookes, 2011; Roberts, Gibbons, Kingsbury, & Gerrard, 2013). For example, Boot, Peter, and van Oosten (2015) exposed young women to a movie clip with a likable or unlikable female character who engaged in casual sex. Their findings indicate that for single women, likeability of the character did not impact their own willingness to engage in casual sex; however, women in a relationship were more likely to report a willingness to engage in casual sex in the unlikable character condition. The authors suggest that this difference may be due to increased self-regulation in women in relationships in the likeable condition. Other experimental studies have demonstrated that humorous depictions of pregnancy in media can reduce counter-arguing and may trivialize the seriousness of the issue (Moyer-Guse et al., 2011).

Genre Specific Content

In addition to examining sexual content broadly, many studies have examined the influence of specific genres, scripts, and messages in media that might shape sexual behavior, arguing that it may not be the volume of content but the nature of the content that influences behaviors. Exposure to reality television, for example, has been associated with increased odds of engaging in one-night stands (Fogel & Kovalenko, 2013) and other risky sexual behaviors, specifically for women (Marron & Collins, 2009). In a longitudinal study, Gottfried and colleagues (2013) found that exposure to sexy content in TV comedies positively predicted intercourse initiation, but exposure to sexy content in dramas negatively predicted intercourse initiation. Exposure to music videos is associated with increased peer sexual partner estimates and increased sexual experience (Marron & Collins, 2009), but connections between exposure and behavior differ by musical genre (Wright & Qureshi, 2015). Specific sub-genres of television have also been linked to sexual behaviors. An examination of MTV’s teen pregnancy themed shows, 16 and Pregnant and Teen Mom, demonstrated that exposure to these narratives was associated with increased odds of reporting sexual intercourse within the previous ten months (Wright, Randall, & Arroyo, 2013).

Research investigating specific scripts and messages in sexual media content often focuses on the ways in which gender and sexuality are presented. As noted earlier, much media sexual content endorses traditional gender roles and gendered norms concerning sexual relationships, such as components of the heterosexual script (Kim et al., 2007). Among undergraduate men, music video exposure has been found to predict greater endorsement that men are sex driven; in turn, this belief is associated with men’s inconsistent condom use (Ward et al., 2011). For adolescent girls, seeing women represented as passive sexual objects has been linked to increased sexual experience but decreased sexual agency (Tolman, Kim, Schooler, & Sorsoli, 2007). Similarly, regular exposure to music lyrics that describe degrading sexual encounters in which women are objectified is associated with the advancement of non-coital sexual activity and intercourse initiation (Martino et al., 2006; Primack et al., 2009).

Mediators of Links Between Media and Sexual Behavior

The research presented so far has largely focused on direct associations between media exposure and sexual behavior. However, social cognitive theory posits that behaviors are the result of cognitions gleaned from models. Hence, it is argued that media influence beliefs, attitudes, and cognitions, which then influence behavior. Applications of this specific mediation model to sexual socialization have sought to explore the mechanisms by which media may influence behavior. Several cognitions have been examined related to sexual behaviors, including cognitive susceptibility to engage in intercourse (L’Engle & Jackson, 2008; Martino et al., 2005), sexual self-efficacy (Martino et al., 2005; Schooler & Ward, 2006), body consciousness (Vandenbosch & Eggermont, 2014), perceived peer norms (Bleakley et al., 2008, Brown et al., 2006; Chia, 2006, Ward et al., 2011), attitudes about sex (Bleakley et al., 2011; Martino et al., 2005), and sexual scripts (Ward & Friedman, 2006; Ward et al., 2011).

One of the most influential cognitions related to sexual behavior is the perception of peer norms related to sex (Bleakley et al., 2008; Brown et al., 2006; Chia, 2006, Ward et al., 2011). Perceiving that one’s peers approve of sex is associated with increased sexual activity independent of media exposure (Bleakley et al., 2008; Brown et al., 2006); once media use is added to the equation, evidence from longitudinal research indicates that exposure to sexual media content predicts perceived normative pressure (from peers), which then predicts intention to have sex (Bleakley et al., 2011). Chia (2006) proposes an alternate model for the influence of peer norms, arguing that adolescents may infer the exposure to and effects of media on their peers and base their own peer norms on estimates of these media effects.

Moderators of Links Between Media and Sexual Behavior

There are also several key moderators that impact the presence and strength of the link between media exposure and sexual behavior. Demographic factors such as race and gender have been found to play a key role. In terms of race, several studies report that although media effects were found for white participants, there was no evidence of an effect for black participants (Brown et al., 2006; Jackson et al., 2008; Somers & Tynan, 2006). This racial difference may be due to earlier sexual initiation in that population (Brown et al., 2006); however, some scholars have argued that there may be differences in interpretation of media, the context of media viewing, or levels of identification or engagement with media (Ashby et al., 2006; Hennessy et al., 2009). Black youth may be less inclined to identify with media characters who do not look like them. In terms of gender, Somers and Tynan (2006) found that sexual content on TV was positively related to the frequency of kissing and intercourse for men only, and O’Hara et al. (2012) found a significantly stronger effect of media sex exposure on sexual behaviors for men. Another study demonstrated that specific sexual scripts may impact men and women differently (Tolman et al., 2007).

Beyond demographics, researchers on sex and the media have examined the moderating effects of parental mediation (Bleakley et al., 2008; Radanielina-Hita, 2014; Schooler et al., 2006; Wright, Randall, & Arroyo, 2013). In general, more involved parenting related to media exposure is associated with lower levels of sexual experience (Schooler et al., 2006). Active mediation—discussion of media content—can serve as a protective factor, minimizing the effects of media on risky sexual behavior (Brown et al., 2006; Fisher et al., 2009; Guo & Nathanson, 2011; Radanielina-Hita, 2014; Wright et al., 2012). Restrictions on media content, termed restrictive mediation, produced the most consistently pro-social effects, specifically relating to lower reports of oral sex or intercourse and lower intentions to engage in intercourse (Fisher et al., 2009). Parental co-viewing led to mixed results, relating to lower expectations of positive health outcomes (Fisher et al., 2009), but to earlier sexual initiation (Guo & Nathanson, 2011).

Finally, aspects of engagement with the media may moderate the influence of media on sexual behavior. Specifically, given the importance of attractive models and identification to social cognitive theory, researchers have examined the role of identification as a moderator (Ward & Friedman, 2006). For girls, identification with same sex media figures is associated with more dating and sexual experience (Schooler et al., 2006; Ward & Friedman, 2006). Some work has also been done examining the role of multi-tasking. Jeong, Hwang, and Fishbein (2010) found that the effect of sexual content on sexual behavior was highest for light multi-taskers who were more focused on one medium at a time. In contrast, there is also some limited evidence to suggest that multi-tasking with Internet use as one of the tasks may increase the impact of sexual content on sexual behaviors (Collins, 2008). More research is needed in this area and in the role of digitization in sexual socialization.

Attitudinal Effects

Behavioral Effects

Nature of Sexual Media Content

Several large-scale analyses of television’s sexual content were conducted at the start of the new millennium. Although these data provide a useful foundation, the analyses need to be updated, in light of today’s current media market. These analyses did not focus heavily on reality programming, which has proliferated in the past decade. Also, television content is no longer restricted to the major networks or to live programming. First, studies are needed that focus on media content created for premium channels like HBO and Showtime, for Netflix, and for Amazon Prime. Second, greater analysis is needed of the nature of the sexual messages communicated. All sexual references or sexual content are not equivalent. There are some data documenting the presence of the heterosexual script on some types of programming. This work is excellent and needs to be updated to included current media content and extended to include other cultural sexual scripts. Finally, study is needed on the nature of sexual content in other media formats, such as feature films and music lyrics.

Effects of Mainstream Media on Sexual Attitudes

Research in this area has focused predominantly on the influence of a) traditional media forms, such as television and music videos on b) white adolescents or emerging adults in Western contexts, using c) cross-sectional surveys or laboratory experiments. Future research should aim to expand upon each of these by examining the influence of new media technologies on sexual attitudes and norms; broadening our understanding of sexual media influences on adults beyond their mid-twenties, those who do not identify as white, and/or those who live in non-Western contexts; and employing longitudinal designs that could track the influence of media on the development of sexual attitudes and norms over time. Additionally, future research should work to identify the myriad moderators and mediators that increase or decrease our susceptibility to sexual media effects (Valkenburg & Peter, 2013). Several such mediating variables have been identified in the 3AM model (Wright, 2011), and work building on this model needs to continue.

Effects of Mainstream Media on Sexual Behavior

Initially, research on media and sexual behavior was focused on determining whether or not any media effects existed. Once it was established that media do influence sexual behaviors and intentions (including sexual initiation, non-coital sexual activity, pregnancy, and risky sexual behaviors), focus shifted to variables impacting the strength of these effects, specifically medium, genre, and demographic and contextual factors shaping media exposure and engagement. Most recently, researchers have begun to consider the changing media environment and its impact on viewer engagement, interpretation, and internalization of media content. Methodologically, research in this area continues to rely on survey and longitudinal data, although some experiments involving the manipulation of moderators have been completed. In the future, research on the media’s influence on sexual behavior will need to address issues related to the inclusion of non-white, non-heterosexual, non-U.S. populations, methodological challenges related to establishing causality, and the shifting nature of the digital media landscape.

Effects of Pornography

A large number of studies within the last decade have examined connections between pornography consumption and viewers’ sexual attitudes, gender attitudes, sexual risk behaviors, and sexual aggression. However, research on viewers’ sexual experiences (e.g., sexual satisfaction, sexual practices) remains limited. Additionally, a large portion of the adult literature has focused on men, specifically. Thus, further research with samples of adult women is needed.