Objective: We evaluated the self-fulfilling prophecy hypothesis that belief in an "alcohol enhances/stimulates sex" expectancy fosters heightened sexual behavior-via effects on intermediate sexual perceptions. To test this notion, we investigated the effects of self-reported alcohol expectancies, alcohol expectancy set and a co-participant's gender and apparent drinking status on dyadic viewing of erotica. Method: Sex-related alcohol expectancies were assessed in 100 male moderate social drinkers. In a subsequent session, participants were led to believe they were consuming either alcoholic or nonalcoholic drinks. No alcohol was actually administered. Each participant rated his sexual arousal, rated an alcohol-drinking or nondrinking co-participant on sexual disinhibition and then viewed erotic slides with the co-participant. Slide viewing times were assessed unobtrusively. Results: Path analysis revealed support for the self-fulfilling prophecy hypothesis: Expectancy score (moderated by alcohol expectancy set heightened viewing indirectly via effects on sexual arousal (beta = .26) and perceived disinhibition (beta = .25). Sexual arousal in turn predicted perceived disinhibition (beta = .37), which in turn predicted viewing (beta = .23). Co-participant drinking had direct (beta = .21) and indirect (beta = .40 via perceived disinhibition) effects on viewing. Conclusions: Alcohol expectancy variables-apart from alcohol-interactively determined men's responding in a dyadic sexual situation. Consistent with psychosocial formulations, predrinking expectancy steered postdrinking perceptions along an expectancy-congruent course to shape subsequent behavior. Thus, alcohol's role in stimulating men's sexual responding cannot be construed as occurring through a strictly pharmacological mechanism. Speculations about the comparative domains of expectancy versus alcohol explanations are discussed.