An Experimental Investigation of the Interactive Effects of Alcohol and Sexual Arousal on Intentions to Have Unprotected Sex

Abstract

We manipulated both alcohol intoxication and sexual arousal and presented male participants (N = 79) with a video vignette in which two undergraduates decide whether to have unprotected sexual intercourse. Participants were asked what they would do if they were in a similar situation, and we found that among sober participants, those assigned to a sexually arousing or neutral condition did not differ in their intentions to engage in unprotected sexual intercourse. Among intoxicated participants (M blood alcohol level = 0.08%), however, those assigned to the sexually arousing condition reported that they were more likely to have unprotected sex than did those assigned to the neutral cue condition. These findings support alcohol myopia theory (Steele & Josephs, 1990) and are consistent with prior nonexperimental work (MacDonald, MacDonald, Zanna, & Fong, 2000). Through the use of an experimental design, this study makes a new contribution by demonstrating that sexual arousal and alcohol intoxication interact to exert a causal influence on intentions to engage in risky sexual behavior.

ACKNOWLEDGMENTS

This project was supported by a Canadian Institutes for Health Research operating grant and a Social Sciences and Humanities Research Council of Canada (SSHRC) operating grant awarded to Tara MacDonald and a SSHRC Doctoral Fellowship awarded to Anna Ebel-Lam.

We thank Steven Spencer for his comments on a draft of this article.

Notes

¹Fifteen additional participants completed the study in the intoxicated condition but evidenced relatively low BALs (i.e., less than .06). Their data were subsequently excluded from our analyses.

²We restricted the sample to men for three reasons. First, there are dangers associated with administering alcohol to sexually active female participants, who may be pregnant. Although this could be ruled out using a pregnancy test, we also had concerns about manipulating sexual arousal equally for male and female participants within the same study. A meta-analysis by Murnen and Stockton (1997) showed that male participants respond with greater sexual arousal to sexual stimuli, particularly if participants are college age (as opposed to older participants) and if they are tested in a private/small-group setting (as opposed to a large-group setting), as in our study. Finally, we originally created our video for use with men (thereby purposely choosing a male character who was average in attractiveness and a highly attractive female character), and in pretesting we have found that the mean for intentions to have unprotected sex among female participants is very low.

³This alcohol administration procedure is designed so that participants would be completing the dependent measures on the ascending limb of alcohol absorption, close to their peak BAL, when the effects of alcohol on decision making may be most pronounced (Jones & Vega, 1972; Kruse & Fromme, 2005; although conflicting evidence also exists to show that cognitive impairment is greater on the descending limb; e.g., Pihl, Paylan, Gentes-Hawn, & Hoaken, 2003).

⁴For each dependent measure, we checked for differences between the sober and placebo conditions. There were no differences on any measure, so the sober and placebo group are collapsed for all analyses that follow. The absence of an alcohol expectancy effect is consistent with research conducted in our laboratory (e.g., MacDonald et al., 1995, 1996; MacDonald, Fong, et al., 2000; MacDonald, MacDonald, et al., 2000) and other laboratories (e.g., Abbey et al., 2005; Fromme et al., 1997, 1999; Maisto et al., 2004).

⁵Although we have no direct evidence, it could be that among sober participants, those in the sexual arousal condition correct for the influence of sexual arousal by recruiting inhibiting thoughts (cf. Wegener & Petty, 1995).