Gender threats and men’s antigay behaviors: The harmful effects of asserting heterosexuality

Abstract

Whereas recent work on gender role violations suggests that asserting their heterosexuality may diminish the harmful effects of gender threats (versus gender affirmations) on men’s antigay reactions, predictions derived from social identity theory suggest that asserting heterosexuality can exacerbate the negative effects of a gender threat on antigay reactions. Two studies tested these competing hypotheses. In Study 1, gender threatened versus affirmed men sent more intense noise blasts at a gay partner, but only if they asserted their heterosexuality. In Study 2, men high in sexual prejudice who underwent a gender threat sat farther from a gay confederate than gender affirmed men, but only if they asserted heterosexuality. Discussion considers the theoretical and practical implications of these findings, and highlights directions for future research.

Keywords

antigay behaviors gay men gender threats manhood sexual prejudice

Gay men constitute the majority of victims of antigay violence (Federal Bureau of Investigations, 2008), and young heterosexual men commit most documented aggressive acts against gay men (Berrill, 1992; Herek, Gillis, Cogan, & Glunt, 1997). In their efforts to illuminate the factors that underlie men’s antigay aggression, social psychologists recently began to examine the role of gender threats, or experiences that challenge one’s status as “a real man.” Accruing findings suggest that threats to their gender status heighten heterosexual men’s antigay reactions across a range of measures (Glick, Gangl, Gibb, Klumpner, & Weinberg, 2007; Parrott, 2009; Talley & Bettencourt, 2008).

A possible explanation for the link between gender threats and antigay reactions can be found by looking to cultural conceptions of masculinity. Shared beliefs about gender construe manhood as a more precarious social status than womanhood (Bosson & Vandello, 2011). Cultures around the world treat manhood as a status that is difficult to attain, and that must be earned through active demonstrations (Gilmore, 1990). At the same time, manhood status is impermanent once achieved, and can be “lost” relatively easily via shortcomings and social transgressions. Therefore, situational gender threats are experienced as particularly stressful by men, often motivating them to restore manhood via active and aggressive means (Bosson, Vandello, Burnaford, Weaver, & Wasti, 2009).

Thus, antigay reactions following gender threats may reflect men’s attempts to prove their manhood to onlookers. As noted by gender theorists (e.g., Herek, 1986; Kilianski, 2003; Kimmel, 1997), heterosexuality is an essential component of hegemonic masculinities. Men who experience challenges to their manhood status may accordingly feel the need to distance themselves from gay men, who represent violations of the masculine, heterosexual ideal. Such attempts at distancing may take the form of antigay epithets, emotional and physical avoidance, or aggression.

Two goals inspired the current work. First, we extended past work on gender threats and antigay reactions by examining two different types of behaviors—approach (direct aggression) and avoidance (physical distancing)—in the context of paradigms high in experimental realism. Specifically, we assessed heterosexual men’s behavioral responses to a gay man using the bogus stranger paradigm (Byrne, 1971) and the Taylor (1967) aggression paradigm in Study 1, and a task involving physical distancing from a confederate (Word, Zanna, & Cooper, 1974) in Study 2. Second, we tested the effects of a heterosexuality assertion—an explicit, public statement of one’s heterosexuality—on men’s antigay reactions following a gender threat. In past research, this strategy reduced men’s discomfort during gender threatening activities (Prewitt-Freilino & Bosson, 2008). Before outlining our hypotheses, we summarize the literature on gender threats and antigay reactions.

Gender threats and antigay reactions

A growing body of work points to a causal link between situational gender threats and antigay reactions among heterosexual men. By gender threat, we mean any experience that causes the self or others to question one’s status as “a real man.” Such experiences can, and often do, include gender role violations such as performing behaviors or expressing preferences that are stereotypically feminine. Thus, researchers often manipulate gender threat by inducing men to behave in a feminine manner, or by telling men that they received either a low score on a test of masculinity or a high score on a test of femininity.

To illustrate, Willer (2005) offered men and women feedback on a test of “gender identity” that indicated that they had scored like either a typical man or a typical woman. Participants then reported their support for the gay rights movement and pro-gay legal reforms. Among men, feedback that they scored like a “typical woman” reduced their support for gay rights and legal reforms relative to feedback that affirmed their manhood. Conversely, the gender identity feedback had no bearing on women’s support for gay rights. Using a similar manipulation of gender threat, Glick et al. (2007) found that gender threatened as compared to non-threatened heterosexual men displayed more negative affect toward a hypothetical, effeminate gay target, but not toward a masculine gay target. These findings demonstrate men’s efforts to distance themselves attitudinally and affectively from gay and effeminate men following gender threats.

In an investigation of antigay behavioral reactions, Talley and Bettencourt (2008) threatened heterosexual men’s gender status by giving them feedback indicating that they scored either much lower than most college-aged men in masculinity (gender threat) or right at the national average in masculinity (no threat). Men then received an opportunity to aggress against a fictional gay partner by blasting him with white noise each time he provided an incorrect response during a word association task. As expected, men chose louder noise blasts with which to punish the gay partner in the gender threat condition as compared to the no threat condition. Considered together, these findings offer converging evidence of a causal link between gender threats and antigay reactions among heterosexual men.

Does asserting heterosexuality help or hurt?

If men’s antigay reactions following a gender threat reflect a need to prove manhood to onlookers, then offering men an alternative means of communicating manhood might obviate the need for antigay demonstrations. Here, we examine the viability of this logic by testing whether an opportunity to prove their heterosexuality—a heterosexuality assertion—reduces men’s antigay reactions following a gender threat. In past research, heterosexuality assertions effectively reduced men’s discomfort when their gender was challenged. For example, men who asserted their heterosexuality to their audience, relative to those who did not assert heterosexuality, displayed less anxiety during gender threatening activities including childcare (Bosson, Haymovitz, & Pinel, 2004) and hairstyling (Bosson, Prewitt-Freilino, & Taylor, 2005; Prewitt-Freilino & Bosson, 2008). Based on these findings, we wondered whether a heterosexuality assertion would mitigate men’s antigay reactions following a gender threat. If so, gender threatening (relative to affirming) feedback should increase heterosexual men’s antigay reactions unless they prove their manhood to their audience by asserting heterosexuality.

Note, however, that past work only established the effectiveness of heterosexuality assertions in alleviating men’s intrapersonal feelings of discomfort, in the absence of an outgroup target. It is thus not clear whether the effects of an assertion will extend to men’s interpersonal reactions to an ostensibly real outgroup member with whom they expect to interact. Therefore, we remained mindful of other possible hypotheses. Most relevant here, social identity theory (Tajfel & Turner, 1986) suggests that factors that increase the salience of category boundaries between ingroup and outgroup members tend to promote outgroup prejudice (e.g., Brewer & Miller, 1984; Miller, Brewer, & Edwards, 1985; Zárate, Garcia, Garza, & Hitlan, 2004; but see Deffenbacher, Park, Judd, & Correll, 2009). Manipulations that focus people’s attention on category boundaries can foster stereotyping and dehumanization of outgroup members (Brown & Turner, 1981; Miller & Brewer, 1986; Turner, 1982), particularly when there are pre-existing tensions between groups (Tajfel, 1982). Such tensions often exist between heterosexual and gay men, given that heterosexual men are relatively high in sexual prejudice toward gay men (Herek, 2002; Kite & Whitely, 1996).

Some findings in the clinical literature are consistent with this logic. For example, after exposure to images of male-male sexual and non-sexual intimacy, men who held more negative attitudes toward homosexuality, and those who endorsed more traditional gender role norms, also punished a gay male target with stronger electric shocks than a heterosexual male target (Bernat, Calhoun, Adams, & Zeichner, 2001; Parrott, 2009; Parrott & Zeichner, 2005, 2008). Although this work was not framed as a test of social identity theory, it is reasonable to suppose that explicit reminders of male-male intimacy could heighten the salience of ingroup-outgroup boundaries for heterosexual men. If so, these findings support the notion that a boundary salience manipulation can heighten men’s motivation to aggress against a gay man.

Furthermore, proclaiming one’s heterosexuality may promote self-stereotyping, or the tendency to cognitively assimilate oneself to the ingroup (Brewer, 1991; Turner & Oakes, 1989). When self-stereotyping, people perceive greater similarity between the self and other ingroup members, and greater difference between the self and outgroup members. As such, gender threatening feedback may be especially distressing, and gender affirming feedback especially reassuring, when men are in a state of self-stereotyping and thereby view themselves as prototypical of their heterosexual ingroup. Thus, heightening the salience of ingroup-outgroup boundaries might increase gender threatened men’s need to establish their distinctiveness from gay men, at the same time that it decreases gender affirmed men’s need to distinguish themselves from gay men (e.g., Turner, Brown, & Tajfel, 1979). In sum, if publicly asserting their heterosexuality increases men’s self-stereotyping tendencies, then gender threatening (relative to affirming) feedback should increase heterosexual men’s antigay reactions especially if they assert their heterosexuality to their ostensible audience.

Overview and hypotheses

We conducted two studies to test competing hypotheses regarding the effects of a heterosexuality assertion on men’s antigay behaviors following a gender threat; in doing so, we utilized procedures high in experimental realism. Specifically, we tested the interactive effects of heterosexuality assertion and gender threat on men’s aggression toward and distancing from a gay man with whom they expected to interact. In both studies, heterosexual men either revealed or did not reveal their sexual orientation to a gay man, and then received bogus feedback that either threatened or affirmed their manhood. All men learned that both their gay partner and the experimenter saw their feedback, thus rendering the gender threat “public.” Finally, men had an opportunity to blast the gay partner with white noise (Study 1), and distance themselves from him by placing their chair far away from his (Study 2).

If a heterosexuality assertion effectively assuages men’s concerns about their manhood following a gender threat, then we should observe an assertion-by-threat interaction such that threat feedback elicits more aggression and distancing than affirmation feedback, but primarily among men who do not assert heterosexuality. Alternatively, if a heterosexuality assertion promotes self-stereotyping and outgroup prejudice, then we should observe an assertion-by-threat interaction such that threat feedback elicits more aggression and distancing than affirmation feedback, but especially among men who do assert their heterosexuality.

Study 1

Method

Participants and design

Eighty-three male undergraduates participated for credit toward a course requirement. To be eligible, men had to identify as “Exclusively heterosexual” (on a scale of 1 [Exclusively heterosexual] to 7 [Exclusively gay]) during an online testing session. Participants ranged in age from 18 to 42 (Md = 19), and described themselves as White (50.0%), Asian-American (17.9%), Latino (12.8%), Black (11.5%), Arabic (3.8%), and biracial (3.8%). We deleted data from one man who reported strong suspicions about the existence of the partner, leaving 82 men who were randomly assigned to condition in a 2 (Heterosexuality assertion) x 2 (Gender threat) design.

Procedure

An experimenter introduced men to a study of “the links between personality and electronic communication style,” and explained that they would soon interact via computer with another man (their partner) who was ostensibly seated in another room. To create the expectation of meeting the partner in person, the experimenter also explained that participants’ last experimental task would be a face-to-face interview with the partner. To heighten the public nature of the experience, the experimenter—who was always a man—noted that he would observe the entire session, including all of the feedback that both partners received and the contents of their computerized interactions, via a computer in another room.

Before beginning the online interaction, participants learned that they would first exchange some personal information with the partner. The experimenter gave them a “Personal Information Form” that served as the basis for administering the heterosexuality assertion manipulation. This form asked for several pieces of mundane information such as participants’ hometown and favorite food. Based on random assignment, men in the heterosexuality assertion condition were also asked to indicate their sexual orientation on this form by checking one of three options (“Heterosexual,” “Bisexual,” or “Homosexual”). Men in the no assertion condition were not asked about their sexual orientation on this form. When participants were done with the Personal Information Form, the experimenter brought it “over to show your partner,” and then returned a minute later with the partner’s completed form. The partner’s form contained handwritten responses to mundane questions (e.g., favorite sport, part-time job) that differed from the questions that participants had answered about themselves (thus, the partner’s form never revealed his sexual orientation). To explain this, the experimenter noted in an offhand manner that “we have people answer different sets of questions so you learn different things about each other.” Participants had several minutes to look over the partner’s information while the partner ostensibly looked at theirs. Note that we had men reveal (or not reveal) their heterosexuality prior to the gender threat manipulation based on past work in which the assertion manipulation always preceded the threat manipulation (e.g., Bosson et al., 2005; Prewitt-Freilino & Bosson, 2008).

Next, the experimenter explained that both partners would take different personality tests and then share their scores with each other. Participants selected their test by picking a folded slip of paper from a cup (all papers read “Test A”). To begin the test, participants clicked a button labeled “Test A” on the computer screen. Written instructions explained that “Personality Test A is a measure of people’s gender identity, and it consists of questions about common gender related knowledge. After you complete the test, it will calculate your score and give you feedback.” Before leaving the participant to complete the test, the experimenter noted that “I’ll be at my computer in the main room, watching to make sure the program runs smoothly.”

The test consisted of 32 difficult forced-choice questions pertaining to stereotypically masculine (sports, cars, home repair) and feminine (childcare, cooking, fashion) topics. Sample items include: “Karate originated in which country: Japan or China?” and “A roux is best described as which: sauce or cake?” (Rudman & Fairchild, 2004; Vandello, Bosson, Cohen, Burnaford, & Weaver, 2008). Upon completing the test, men waited several seconds for the computer to generate their score. Based on random assignment, men learned that they scored at either the 37th percentile (gender threat condition) or the 83rd percentile (gender affirmation condition) compared to other men at their university. This feedback was bolstered by a continuum anchored with endpoints labeled “Feminine Gender Identity” and “Masculine Gender Identity”; an arrow labeled “Your score” pointed toward either the feminine (threat) or masculine (affirmation) end of the continuum.

Men then learned that their score was being sent to their partner. In exchange, they received the partner’s feedback on a different personality test. The partner’s test, Test B, was described as “a test of relationship styles.” His feedback stated that, because he had indicated that he was gay and romantically uninvolved, his personality test scores were being compared to those of other single gay men. Thus, all participants learned that their partner was gay in the context of reading his test feedback. The partner’s feedback described him as scoring at the 54^th percentile on the test, and an arrow labeled “Your score” pointed toward the midpoint of a continuum with endpoints labeled “Intimacy Avoidant Style” and “Intimacy Seeking Style.”

After participants read the partner’s personality feedback, the experimenter introduced them to the next task, which served as the measure of aggression. This task, borrowed from Harmon-Jones (2007) and based on the Taylor Aggression Paradigm (1967), was described as an interactive, competitive, online game. The purpose of the game was to press a key more quickly than the partner each time a prompt appeared on the screen. The game consisted of 20 reaction time trials, and was rigged so that participants won on 10 randomly selected trials and lost on the remaining 10 trials. Before every trial, men selected the decibel level of “white noise” with which to blast their partner if they won; the options included 60dB, 70dB, 80dB, 90dB, and 100dB. After each of the 10 winning trials, men pressed the spacebar to blast their partner at the selected decibel level for up to 10 seconds. After each of the 10 losing trials, men received (via headphones worn throughout the game) a 5- to 7-second, 102-decibel blast of white noise, ostensibly delivered by the partner. This type of aggression paradigm is both widely-used and well-validated, and is ideal for our purposes because it captures spontaneous aggression while minimizing demand characteristics (for a review, see Giancola & Chermack, 1998).

After the game, participants completed a manipulation check item that queried them about their score on the gender identity test, and then provided some demographic data. Only four men (5%) failed the manipulation check; because including versus excluding these men’s data did not alter our findings, we retained them in analyses. Finally, men underwent an exit interview during which they were probed for suspicion about the false feedback and the existence of the gay partner, and then received a full debriefing.

Results

To compute the measure of aggression we first standardized both the decibel levels selected before and the length of spacebar presses after each winning trial. We then summed the standardized decibels and spacebar presses for each winning trial, and averaged these sums across the 10 trials to create an internally consistent index of aggression (α = .95).

We submitted this aggression index to a 2 (Heterosexuality assertion) x 2 (Gender threat) ANOVA, which yielded an assertion-by-threat interaction, F(1, 77) = 4.43, p < .04, $η_{partial}^{2}$ = .054, and no other effects, Fs < 1. As shown in Figure 1, among men who first asserted their heterosexuality, those who experienced a manhood threat displayed significantly more antigay aggression than those who received gender affirming feedback, F(1, 77) = 4.07, p < .05, $η_{partial}^{2}$ = .050. Among men who did not assert heterosexuality, the gender feedback did not significantly affect their antigay aggression, F < 1. Note, though, that this interaction pattern was driven primarily by low levels of aggression among gender affirmed men in the assertion condition rather than by high levels of aggression among gender threatened men who asserted heterosexuality: Men in the assertion/affirm condition were less aggressive than those in the no assertion/affirm condition, F(1, 77) = 3.81, p = .055, $η_{partial}^{2}$ = .047, whereas the assertion manipulation had no effect on aggression among gender threatened men, F < 1.¹

Figure 1.

Average decibels delivered after ten winning trials as a function of heterosexuality assertion and gender threat conditions.

To determine whether the observed interaction pattern characterized men’s antigay behavior across the entire game, we also created aggression composites based on the first five (α = .90) and last five (α = .93) winning trials separately. A 2 (Heterosexuality assertion) x 2 (Gender threat) x 2 (Game half) ANOVA, with repeated measures on the last factor, revealed no main or interactive effects of game half, all Fs < 1.78, ps > .18. Thus, men’s pattern of antigay aggression across game trials remained relatively constant.

Summary

Among men who asserted their heterosexuality to their gay partner, those who received gender threatening feedback exhibited more antigay aggression than those who received gender affirming feedback. This pattern is more consistent with hypotheses derived from social identity theory than it is with recent work on gender role violations (e.g., Bosson et al., 2005), because the harmful effects of a gender threat on men’s antigay reactions only emerged when ingroup-outgroup boundaries were especially salient. Unexpectedly, however, men in the no assertion condition who received gender affirming feedback also displayed relatively high levels of antigay aggression. Indeed, the interaction was driven primarily by low levels of aggression among men who both asserted their heterosexuality and received gender affirming feedback. Thus, rather than being fully consistent with either hypothesis—both of which predicted more aggression among gender threatened than gender affirmed men in the no assertion condition—this pattern points to a possibility that we did not consider earlier: Perhaps men show the least antigay reactions when they are “double-buffered” from gender threats. After all, men in the assertion/affirm condition could feel confident knowing that both their prototypicality as men and their adherence to the heterosexual ideal were conveyed to their audience (the experimenter and gay partner). Thus, it is possible that this combination of factors allowed men in the assertion/affirm condition to be especially nonaggressive toward the gay partner, whereas men whose gender status was questionable—due to threatening feedback or an inability to assert heterosexuality—behaved relatively aggressively in an effort to prove their manhood.

Even if the double-buffer hypothesis is viable, however, it does not explain why we failed to replicate past work that found greater aggression among gender threatened versus gender affirmed men in the no assertion condition. As noted earlier, several studies found that manhood threats increased men’s negative reactions to gay men, and these effects emerged in the absence of a boundary salience manipulation (e.g., Talley & Bettencourt, 2008). One possible reason for our failure to find such an effect is that men in our study—unlike those in past studies—expected to meet the gay man against whom they played the competitive game. Another possible reason is that gender affirming feedback may, in some circumstances, heighten sexual prejudice to the same degree as gender threatening feedback does (e.g., Lehmiller, Law, & Tormala, 2010; Rivera, 2009). We return to these ideas in the General Discussion; for now, we replicate and extend Study 1’s findings using a different measure of antigay behavior.

Study 2

Study 1’s findings indicate that gender threatened men display more antigay aggression than gender affirmed men only when their heterosexual status is salient. In Study 2, we followed up on Study 1 with two primary goals in mind. First, we tested the generalizability of the effect observed in Study 1 to a different measure of antigay behaviors. Whereas in Study 1 we examined men’s approach behavior (aggression), here we examined their avoidance behavior; specifically, we used a physical distancing task in which participants placed a chair somewhere near the chair of their gay partner, who was played by a confederate. With this task, larger distances are assumed to reflect stronger motivation to avoid the gay man (Macrae, Bodenhausen, Milne, & Jetten, 1994; Word et al., 1974).

Second, we tested whether the joint effects of gender threat and heterosexuality assertion were moderated by men’s pre-existing sexual prejudice levels. As noted earlier, factors that heighten the salience of group boundaries should be especially likely to promote undesirable intergroup dynamics when there is a history of tension between groups (e.g., Hamilton & Bishop, 1976; Tajfel, 1982). Although heterosexual men, on average, hold relatively negative attitudes toward gay men (Herek, 2002), there is substantial variability in their sexual prejudice levels (e.g., Herek, 1988). Thus, whereas some heterosexual men are high in sexual prejudice, others hold benign or even favorable attitudes toward gay men. Not surprisingly, people high (vs. low) in sexual prejudice show both greater avoidance of gay persons (Pryor, Reeder, Yeadon, & Hesson McInnis, 2004), and stronger activation of avoidance goals in response to gay cues (Wyer, 2010).

Given this, we expected a three-way interaction of sexual prejudice, assertion, and threat on men’s physical distancing from the gay confederate. Among men high in sexual prejudice, for whom intergroup relations with gay men are relatively tense, we should observe an assertion-by-threat interaction. First, among high prejudice men who assert their heterosexuality, those in the gender threat condition should sit farther from the gay man than those in the affirmation condition. Next, among high prejudice men in the no assertion condition, two patterns seem possible: The social identity hypothesis still predicts more distancing from the gay man among gender threatened relative to gender affirmed men, but this difference should be smaller than that observed in the assertion conditions. Alternately, the double-buffer hypothesis predicts that both threatened and affirmed men in the no assertion condition should exhibit similarly high levels of distancing such that the interaction is driven by men in the assertion/affirm condition sitting much closer to the gay man as compared to men in all other conditions. Among men low in sexual prejudice, we did not anticipate an assertion-by-threat interaction. These men should display moderately low levels of antigay distancing regardless of our manipulations.

Method

Participants and design

Sixty-two heterosexual male undergraduates participated in exchange for credit toward a course requirement. To be eligible, men had to identify as “Exclusively heterosexual” (see Study 1) and complete a measure of sexual prejudice during an online testing session. Participants ranged in age from 18 to 37 (Md = 20), and described themselves as White (60.0%), Asian-American (8.0%), Latino (16.0%), Black (6.0%), Arabic (4.0%), biracial (2.0%) and “other” (4.0%). Seven participants expressed strong suspicions about the gender test feedback or the confederate and were thus deleted from analyses. This left a total of 55 men who were randomly assigned to condition in a 2 (Heterosexuality assertion) x 2 (Gender threat) design, with sexual prejudice scores as a continuous moderator.

Procedure

Before signing up for the study, students completed online a modified version of the Gay Male subscale of Raja and Stokes’ (1998) Modern Homophobia Scale. This scale consists of 13 statements that assess an aversion to gay men (e.g., “I welcome new male friends who are gay” [reverse-coded]) and a belief that homosexuality is curable (e.g., “Gay men should undergo therapy to change their sexual orientation”). In addition, students completed five items that assess beliefs about the immutability of homosexuality (e.g., “People cannot change their sexual orientation” [reverse-coded]; see Haslam & Levy, 2006). These 18 items were highly internally consistent (α = .90), so we averaged them to create sexual prejudice scores.

Upon arrival at the laboratory, men were greeted by a male experimenter who introduced them to a study of “the links between personality and electronic communication style.” As in Study 1, participants first either asserted or did not assert their heterosexuality to their partner, and then took the gender identity test and received threatening or affirming feedback. Next, men learned that their partner saw their test score, and they saw the partner’s test score (describing him as a gay man). From this point on, the procedure differed from that of Study 1.

The experimenter explained that the next part of the experiment involved a face-to-face interview with the partner, in an adjoining room. When the experimenter led the participant into the next room, a male confederate was already seated at a chair in the center of the room. Four men played the role of the gay confederate. To standardize their appearance and behavior, they all wore jeans and a solid color t-shirt, carried a black backpack adorned with a gay pride button and a rainbow ribbon, and enacted the same routine. Specifically, as the experimenter and participant entered the room the confederate stood up, placed his backpack on his chair, and stated that he needed to “use the restroom.” He then hurried out of the room, claiming that he would “be right back.” At this, the experimenter turned to the participant and said “While we’re waiting for him to return, you can set up your chair for the interview. Just grab a chair from that pile [at this, he gestured toward a stack of chairs] and set it near your partner’s.”

As soon as the participant positioned his chair the experimenter ended the study, led the participant to sit down at a separate table (to prevent him from inadvertently shifting the position of his chair), probed him for suspicion, and debriefed him. After the participant left the room, the experimenter measured the distance between the two front legs of the participant’s chair and the two front legs of the confederate’s chair; the shorter of these two distances served as the dependent measure.

Results

In a simultaneous multiple regression analysis, we regressed the distance (in inches) that men placed their chair from the gay partner’s chair onto: (a) sexual prejudice scores (centered), (b) heterosexuality assertion condition (coded as 0, 1), (c) gender threat condition (coded as 0, 1), (d-f) all two-way interactions among variables, and (g) the three-way interaction. This analysis yielded a significant three-way interaction, β = .49, t(47) = 2.43, p < .02, f² = .12, that qualified a sexual prejudice-by-assertion interaction, β = -.45, t(47) = -2.43, p < .02, and a marginally significant main effect of threat, β = -.33, t(47) = -1.81, p < .08. No other effects were significant, ps > .10.

We decomposed the three-way interaction by examining the assertion-by-threat interaction separately among men who were high versus low in sexual prejudice. Among men high in sexual prejudice, the predicted assertion-by-threat interaction emerged, β = -1.13, t(47) = -2.87, p < .01, f² = .17. As shown in Figure 2, when men asserted their heterosexuality, those who underwent a manhood threat sat farther away from the gay partner relative to those whose gender was affirmed, β = .75, t(47) = 2.06, p < .05, f² = .09. Interestingly, the opposite pattern reached significance among men who did not assert heterosexuality, β = -.54, t(47) = -2.05, p< .05, f² = .09. That is, gender threatened men sat closer to the gay target than gender affirmed men when they did not communicate their heterosexuality to him. Moreover, unlike in Study 1, the heterosexuality assertion (vs. no assertion) marginally significantly increased distancing among gender threatened men, β = .69, t(47) = 1.85, p = .07, f² = .08, and it decreased distancing among gender affirmed men, β = -.57, t(47) = -2.44, p < .02, f² = .13. Among men low in sexual prejudice, the assertion-by-threat interaction did not approach significance, t < 1, p > .37; across conditions, low prejudice men sat moderately close to the gay man (M ₌ 39.71 inches).

Figure 2.

Distance (in inches) from gay partner’s chair as a function of heterosexuality assertion and gender threat conditions, among men high in sexual prejudice.

Summary

As in Study 1, only when men asserted their heterosexuality did those who received gender threatening feedback exhibit more antigay behavior than those who received affirming feedback. Moreover, in this study we found some (albeit weak) evidence that increasing the salience of ingroup-outgroup boundaries heightens the negative effects of a gender threat on men’s antigay behaviors: Gender threatened men showed marginally significantly more aversion to a gay man when they asserted their heterosexuality than they did in the no assertion condition. Notably, both of these effects were limited to men who held negative attitudes toward gay men; among men low in sexual prejudice, the assertion-by-threat interaction did not emerge. Thus, it was when intergroup tensions were already high that a category salience manipulation yielded more avoidance of a gay man among gender threatened versus gender affirmed men. Note that these findings, like those in Study 1, are more consistent with social identity theory than they are with recent work on gender role violations and heterosexuality assertions. However, the curious tendency for high sexual prejudice, gender threatened men to sit closer to the gay target than gender affirmed men in the no assertion condition is not consistent with any hypotheses considered thus far. We discuss possible explanations for this finding in the next section.

General discussion

The goals of the current studies were twofold. First, we extended past research by examining the effects of a gender threat on heterosexual men’s spontaneous, antigay behaviors, using paradigms that are high in experimental realism: a bogus stranger and a confederate. Second, we tested competing hypotheses regarding the effects of a heterosexuality assertion on men’s antigay behaviors following a manhood challenge versus a manhood affirmation. In what follows, we summarize the findings and, where relevant, note lingering questions and directions for future research. Given the complexity of our results, we consider the patterns first among men who asserted heterosexuality and then among men who did not.

Asserting heterosexuality: Increasing the salience of ingroup-outgroup boundaries

Across both studies, the data more strongly supported social identity theory than the alternative hypothesis, which was derived from recent work on gender role violations (e.g., Bosson et al., 2005). Moreover, although we considered (briefly) a third hypothesis to explain the findings from Study 1, this double-buffer hypothesis failed to explain the pattern that emerged in Study 2. Instead, and consistent with social identity theory, gender threatened men in both studies exhibited more antigay behavior than gender affirmed men, but only if they first asserted their heterosexuality. In Study 1, gender threatened (vs. affirmed) men who asserted their heterosexuality delivered louder and longer noise blasts to a gay man. In Study 2, gender threatened men who were high in sexual prejudice placed their chair farther from a gay man than did gender affirmed men but, again, only if they asserted heterosexuality.

It is noteworthy that this pattern emerged on two very different measures of spontaneous behavior that assess distinct motivational systems. In Study 1, our manipulations increased heterosexual men’s aggression toward a gay man, in the form of noise blasts; in Study 2, the same manipulations increased physical distancing away from a gay man, in the form of chair placement. Thus, we observed the same pattern of responses across measures which captured both approach and avoidance motivations. The findings from Study 1 are consistent with research linking approach-motivated responses to anger (Carver & Harmon-Jones, 2009; Harmon-Jones, 2003), which is an important affective precursor to antigay aggression (Bernat et al., 2001; Parrott & Peterson, 2008). At the same time, the findings from Study 2 offer a partial replication of recent work demonstrating that priming people with the category label “gay” activates avoidance goals among individuals high in sexual prejudice (Wyer, 2010). The fact that we observed nearly identical effects of our manipulations on behaviors that reflect both approach and avoidance gives us confidence that these effects should generalize to a variety of antigay reactions.

Also consistent across studies was our failure to find support for the notion that a heterosexuality assertion will reduce men’s antigay behaviors following a manhood challenge. This is somewhat surprising, given past work showing that heterosexuality assertions mitigated men’s discomfort during public gender role violations. If asserting their heterosexuality alleviates gender threatened men’s discomfort, why does it not similarly attenuate men’s antigay actions? One possible answer lies in the divergent nature of the outcomes measured here and in past work. Whereas prior work on heterosexuality assertions measured men’s intrapersonal reactions following a manhood threat (Bosson et al., 2005), the current work focused on men’s reactions in interpersonal and, more specifically, intergroup contexts. Moreover, the very nature of the interpersonal exchange used here—an interaction with a gay man—may have raised manhood concerns above and beyond those elicited by the gender threat feedback. Thus, the anxiety-assuaging effects of heterosexuality assertions on gender threats may be limited to contexts in which men perform solo activities; when the context involves an interaction with an outgroup member, heterosexuality assertions appear to exaggerate antigay behaviors among gender threatened (vs. affirmed) men.

As such, our findings suggest that drawing attention to their heterosexual status may not reduce men’s antigay reactions in real-world encounters and, in fact, may have the opposite effect, especially if men are already feeling uncertain about their gender status. This points to a potential conundrum: In real-life contexts, when men undergo gender threats and experience motivation to prove their masculinity, they may spontaneously rely on heterosexuality assertions. Indeed, asserting heterosexuality is a relatively easy and effective means of alleviating anxiety about gender threats (Bosson et al., 2005), and it is one that many men utilize. Williams (1989) noted, for example, that in her interviews with men in gender-atypical careers (e.g., nursing), interviewees often spontaneously noted their heterosexual status without prompting. Thus, men who undergo public challenges to their manhood may be especially likely to rely on heterosexuality assertions to assuage their own discomfort. Unfortunately, in doing so, they may also heighten intergroup tensions.

Of course, it is possible that spontaneously generated heterosexuality assertions will reduce men’s antigay reactions following gender threats in real-world contexts. Although we took measures to maximize the experimental realism of our procedures, there are at least two ways in which our assertion manipulation most likely differed from assertions that occur naturally. First, we induced some men to reveal their sexual orientation to observers whereas most spontaneous heterosexuality assertions probably occur as freely chosen communications. Second, we had men in assertion conditions reveal their heterosexuality prior to the gender threat/affirmation manipulation whereas, in reality, spontaneous assertions often follow rather than precede manhood threatening experiences. That is, men may use heterosexuality assertions in the wake of a gender threat as a means of demonstrating their manhood. Thus, a more ecologically valid manipulation might involve threatening versus affirming men’s gender status and only then allowing them to assert their heterosexuality. Given the importance of understanding and reducing antigay behavior and attitudes, future research should continue to examine if and when heterosexuality assertions can reduce heterosexual men’s antigay reactions under different conditions.

No assertion conditions

Unexpectedly, we observed a trend across both studies for gender threatened men to exhibit less antigay behavior than gender affirmed men when they did not assert their heterosexuality to their partner. Although this pattern did not even approach significance in Study 1, a glance at the white bars in Figures 1 and 2 suggests a roughly similar tendency for gender affirmed men in the no assertion condition to behave in a relatively antigay manner. Indeed, had we measured men’s sexual prejudice in Study 1, this difference might have reached statistical significance among high prejudice men as it did in Study 2. Of course, this supposition is purely speculative, but the similarity of the interaction patterns across studies seems to suggest that gender affirming feedback may actually heighten heterosexual men’s antigay reactions when ingroup-outgroup boundaries are not at the fore of consciousness.

Note, of course, that this pattern is opposite to what others have found using a similar gender threat manipulation (Glick et al., 2007; Talley & Bettencourt, 2008; Willer, 2005). Although our failure to replicate other work is puzzling, we can think of plausible reasons why we observed this pattern. One possibility is that our efforts to heighten the realism of our procedures, by giving the gay male partner an identity and leading participants to believe that they were going to interact with him face-to-face, motivated men to be on their best behavior. If gender threatened men—especially those who are high in sexual prejudice—are aware of their motivation to punish or reject gay men, then those who anticipate an upcoming interaction may overcompensate by suppressing their antigay urges. This should occur particularly among men who feel either external or internal pressure to avoid prejudice (e.g., Devine, Plant, Amodio, Harmon-Jones, & Vance, 2002; Plant & Devine, 2009). Concerned about appearing bigoted, these men may behave in a more pro-gay fashion than they would otherwise. Conversely, gender affirmed men may have an altogether different experience. Those who are high in sexual prejudice and receive feedback that affirms their ingroup prototypicality might be focused more on punishing or avoiding a disliked outgroup member than on appearing unbigoted. If so, then unlike gender threatened men, they may be (ironically) free to display antigay behaviors.

On this note, our use of gender affirming feedback in the current studies may have created unanticipated problems for interpreting our data. Our logic in choosing this feedback was that, by affirming men’s gender status, we would effectively free them from concerns about their manhood. However, recent findings in the self-affirmation literature suggest that gender affirming feedback may paradoxically heighten sexual prejudice. For instance, Rivera (2009) found that offering men positive feedback about their masculinity led them to evaluate gay men more negatively, presumably because the feedback encouraged men to denigrate those who are viewed as “deficient” in masculinity. Similarly, Lehmiller and colleagues demonstrated that self-affirmations that remind people of “family values” do not ameliorate sexual prejudice, because they make salient aspects of the self that are inconsistent with tolerance toward gays and lesbians (Lehmiller et al., 2010). Thus, by offering men in the gender affirmation condition positive feedback about their manhood status, we may have activated self-views (e.g., “I am very masculine”) that promote sexual prejudice rather than tolerance. Among men already high in sexual prejudice, such feedback may be just the impetus needed to motivate antigay behaviors. Unfortunately, the lack of a neutral or no feedback control condition in the current studies is a weakness that precludes us from drawing firm conclusions about this issue. Inclusion of a true, neutral control condition thus remains an important direction for future research.

Regardless of why we failed (in no assertion conditions) to replicate past work, one thing seems clear: Researchers ought not assume that gender threats will consistently and reliably increase antigay reactions. Our findings suggest instead that, under some conditions, gender threats versus affirmations may decrease, or have no effect on, antigay reactions.

Summary and conclusion

Sexual prejudice and antigay aggression are, to a large degree, “problems of manhood.” Young, heterosexual men hold especially negative attitudes toward gay men, and they are the primary perpetrators of hate crimes based on sexual orientation. At present, much remains unknown about the situational factors that heighten and ameliorate men’s antigay reactions. The current findings are thus important because they indicate that a strategy shown to effectively minimize men’s intrapersonal concerns about gender threats does little to minimize their spontaneous negative behaviors toward gay men following a gender threat. Although it may be discouraging to learn that heterosexuality assertions are not a panacea for men’s antigay reactions, it is only through research such as this that we can hope to uncover the situational predictors of sexual prejudice.

Footnotes

Notes

References

Bernat

J. A.

Calhoun

K. S.

Adams

H. E.

Zeichner

(2001). Homophobia and physical aggression toward homosexual and heterosexual individuals. Journal of Abnormal Psychology, 110, 179–187.

Berrill

K. T.

(1992). Anti-gay violence and victimization in the United States: An overview. In Herek

G. M.

Berrill

K. T.

(Eds.), Hate crimes: Confronting violence against lesbians and gay men (pp. 19–45). Newbury Park, CA: Sage.

Bosson

J. K.

Haymovitz

E. L.

Pinel

E. C.

(2004). When saying and doing diverge: The effects of stereotype threat on self-reported versus nonverbal anxiety. Journal of Experimental Social Psychology, 40, 247–255.

Bosson

J. K.

Prewitt-Freilino

J. L.

Taylor

J. N.

(2005). Role rigidity: A problem of identity misclassification? Journal of Personality and Social Psychology, 89, 552–565.

Bosson

J. K.

Vandello

J. A.

(2011). Precarious manhood and its links to action and aggression. Current Directions in Psychological Science, 20, 82–86.

Bosson

J. K.

Vandello

J. A.

Burnaford

R. M.

Weaver

J. R.

Wasti

S. A.

(2009). Precarious manhood and displays of physical aggression. Personality and Social Psychology Bulletin, 35, 623–634.

Brewer

M. B.

(1991). The social self: On being the same and different at the same time. Personality and Social Psychology Bulletin, 17, 475–482.

Brewer

M. B.

Miller

(1984). Beyond the contact hypothesis: Theoretical perspectives on desegregation. In Miller

Brewer

M. B.

(Eds.), Groups in contact: The psychology of desegregation (pp. 281–302). Orlando, FL: Academic Press.

Brown

R. J.

Turner

(1981). Interpersonal and intergroup behavior. In Turner

Giles

(Eds.), Intergroup behavior (pp. 33–65). Oxford: Basil Blackwell.

10.

Byrne

(1971). The attraction paradigm. New York, NY: Academic Press.

11.

Carver

C. S.

Harmon-Jones

(2009). Anger is an approach-related affect: Evidence and implications. Psychological Bulletin, 135, 183–204.

12.

Deffenbacher

D. M.

Park

Judd

C. M.

Correll

(2009). Category boundaries can be accentuated without increasing intergroup bias. Group Processes and Intergroup Relations, 12, 175–193.

13.

Devine

P. G.

Plant

E. A.

Amodio

D. M.

Harmon-Jones

Vance

(2002). The regulation of explicit and implicit race bias: The role of motivations to respond without prejudice. Journal of Personality and Social Psychology, 82, 835–848.

14.

Federal Bureau of Investigations. (2008). Table 1: Incidents, offenses, victims, and known offenders by bias motivation. In Uniform crime reporting program: Hate crime statistics 2008. Retrieved August 26, 2010, from http://www.fbi.gov/ucr/hc2008/data/table_01.html

15.

Giancola

P. R.

Chermack

S. T.

(1998). Construct validity of laboratory aggression paradigms: A response to Tedeschi and Quigley (1996). Aggression and Violent Behavior, 3, 237–253.

16.

Gilmore

D. D.

(1990). Manhood in the making. New Haven, CT: Yale University Press.

17.

Glick

Gangl

Gibb

Klumpner

Weinberg

(2007). Defensive reactions to masculinity threat: More negative affect toward effeminate (but not masculine) gay men. Sex Roles, 57, 55–59.

18.

Hamilton

D. L.

Bishop

G. D.

(1976). Attitudinal and behavioral effects of initial integration of white suburban neighborhoods. Journal of Social Issues, 32, 47–67.

19.

Harmon-Jones

(2003). Anger and the behavioral approach system. Personality and Individual Differences, 35, 995–1005.

20.

Harmon-Jones

(2007). Social emotive neuroscience lab at Texas A&M University: Resources.

Retrieved December 19, 2006

from http://www.socialemotiveneuroscience.org/

21.

Haslam

Levy

S. R.

(2006). Essentialist beliefs about homosexuality: Structure and implications for prejudice. Personality and Social Psychology Bulletin, 32, 471–485.

22.

Herek

G. M.

(1986). On heterosexual masculinity: Some psychical consequences of the social construction of gender and sexuality. American Behavioral Scientist, 29, 563–577.

23.

Herek

G. M.

(1988). Heterosexuals’ attitudes toward lesbians and gay men: Correlates and gender differences. Journal of Sex Research, 25, 451–477.

24.

Herek

G. M.

(2002). Gender gaps in public opinion about lesbians and gay men. Public Opinion Quarterly, 66, 40–66.

25.

Herek

G. M.

Gillis

J. R.

Cogan

J. C.

Glunt

E. K.

(1997). Hate crime victimization among lesbian, gay, and bisexual adults: Prevalence, psychological correlates, and methodological issues. Journal of Interpersonal Violence, 12, 195–215.

26.

Kilianski

S. E.

(2003). Explaining heterosexual men’s attitudes toward women and gay men: The theory of exclusively masculine identity. Psychology of Men & Masculinity, 4, 37–56.

27.

Kimmel

M. S.

(1997). Masculinity as homophobia: Fear, shame, and silence in the construction of gender identity. In Gergen

M. M.

Davis

S. N.

(Eds.), Toward a new psychology of gender (pp. 223–242). New York, NY: Routledge.

28.

Kite

M. E.

Whitely

B. E.

Jr. (1996). Sex differences in attitudes toward homosexual persons, behaviors, and civil rights: A meta-analysis. Personality and Social Psychology Bulletin, 22, 336–353.

29.

Lehmiller

J. J.

Law

A. T.

Tormala

T. T.

(2010). The effect of self-affirmation on sexual prejudice. Journal of Experimental Social Psychology, 46, 276–285.

30.

Macrae

C. N.

Bodenhausen

G. V.

Milne

A. B.

Jetten

(1994). Out of mind but back in sight: Stereotypes on the rebound. Journal of Personality and Social Psychology, 67, 808–817.

31.

Miller

Brewer

M. B.

(1986). Categorization effects on in-group and out-group perception. In Dovidio

J. F.

Gaertner

S. L.

(Eds.), Prejudice, discrimination and racism (pp. 209–230). San Diego, CA: Academic Press.

32.

Miller

Brewer

M. B.

Edwards

(1985). Cooperative interaction in desegregated settings: A laboratory analogue. Journal of Social Issues, 41, 63–75.

33.

Parrott

D. J.

(2009). Aggression toward gay men as gender role enforcement: Effects of male role norms, sexual prejudice, and masculine gender role stress. Journal of Personality, 77, 1137–1166.

34.

Parrott

D. J.

Peterson

J. L.

(2008). What motivates hate crimes based on sexual orientation? Mediating effects of anger on antigay aggression. Aggressive Behavior, 34, 306–318.

35.

Parrott

D. J.

Zeichner

(2005). Effects of sexual prejudice and anger on physical aggression toward gay and heterosexual men. Psychology of Men & Masculinity, 6, 3–17.

36.

Parrott

D. J.

Zeichner

(2008). Determinants of anger and physical aggression based on sexual orientation: An experimental examination of hypermasculinity and exposure to male gender role violations. Archives of Sexual Behavior, 37, 891–901.

37.

Plant

E. A.

Devine

P. G.

(2009). The active control of prejudice: Unpacking the intentions guiding control efforts. Journal of Personality and Social Psychology, 96, 640–652.

38.

Prewitt-Freilino

J. L.

Bosson

J. K.

(2008). Defending the self against identity misclassification. Self and Identity, 7, 168–183.

39.

Pryor

J. B.

Reeder

G. D.

Yeadon

Hesson McInnis

(2004). A dual process model of reactions to perceived stigma. Journal of Personality and Social Psychology, 87, 436–452.

40.

Raja

Stokes

J. P.

(1998). Assessing attitudes toward lesbians and gay men: The Modern Homophobia Scale. Journal of Gay, Lesbian, and Bisexual Identity, 3, 113–134.

41.

Rivera

L. M.

(October, 2009). Not all self-affirmations alleviate prejudice: The effect of affirming masculinity on derogating gay men. Society of Experimental Social Psychology Meeting, Portland, ME.

42.

Rudman

L. A.

Fairchild

(2004). Reactions to counterstereotypic behavior: The role of backlash in cultural stereotype maintenance. Journal of Personality and Social Psychology, 87, 157–176.

43.

Tajfel

(1982). Social psychology of intergroup relations. Annual Review of Psychology, 33, 1–39.

44.

Tajfel

Turner

(1986). The social identity theory of intergroup behavior. In Worchel

Austin

(Eds.), Psychology of intergroup relations (pp. 7–24). Chicago, IL: Nelson-Hall.

45.

Talley

A. E.

Bettencourt

B. A.

(2008). Evaluations and aggression directed at a gay male target: The role of threat and antigay prejudice. Journal of Applied Social Psychology, 38, 647–683.

46.

Taylor

S. P.

(1967). Aggressive behavior and physiological arousal as a function of provocation and the tendency to inhibit aggression. Journal of Personality, 35, 297–310.

47.

Turner

J. C.

(1982). Towards a cognitive redefinition of the social group. In Tajfel

(Ed.), Social identity and intergroup relations (pp. 15–40). Cambridge: Cambridge University Press.

48.

Turner

J. C.

Brown

R. J.

Tajfel

(1979). Social comparison and group interest in ingroup favoritism. European Journal of Social Psychology, 9, 187–204.

49.

Turner

J. C.

Oakes.

P. J.

(1989). Self-categorization theory and social influence. In Paulus

P. B.

(Ed.), The psychology of group influence (pp. 233–275). Hillsdale, NJ: Erlbaum.

50.

Vandello

J. A.

Bosson

J. K.

Cohen

Burnaford

R. M.

Weaver

J. R.

(2008). Precarious manhood. Journal of Personality and Social Psychology, 95, 1325–1339.

51.

Willer

(August, 2005). Overdoing gender: A test of the masculine overcompensation thesis. American Sociological Association Meeting, Philadelphia, PA.

52.

Williams

C. L.

(1989). Gender differences at work. Berkeley, CA: University of California Press.

53.

Word

C. O.

Zanna

M. P.

Cooper

(1974). The nonverbal mediation of self-fulfilling prophecies in interracial interaction. Journal of Experimental Social Psychology, 10, 109–120.

54.

Wyer

N. A.

(2010). Salient egalitarian norms moderate activation of out-group approach and avoidance. Group Processes and Intergroup Relations, 13, 151–165.

55.

Zárate

M. A.

Garcia

Garza

A. A.

Hitlan

R. T.

(2004). Cultural threat and perceived realistic group conflict as dual predictors of prejudice. Journal of Experimental Social Psychology, 40, 99–105.