A total of 40 undergraduate male volunteers, all social drinkers between the ages of 18 and 22, were randomly assigned to one of two expectation conditions in which they were led to believe that the beverage they were administered contained either vodka and tonic or tonic only. For half of the subjects in each expectation condition, the beverage actually contained vodka; the others drank only tonic. After their drinks, measures of penile tumescence were taken from subjects using a penile strain gauge during two erotic films, one depicting a heterosexual interaction, the other, a male homosexual interaction. Although 2X2 analyses of variance failed to reveal any effect of alcohol per se, there were significant effects of expectation on penile tumescence during both the heterosexual and homosexual films. Subjects who believed that they had consumed an alcoholic beverage manifested significantly greater sexual arousal than those believing they had consumed a nonalcoholic beverage, regardless of the actual contents of their drinks. Although no consistent effects were observed on additional measures of sexual arousal, including the Thematic Apperception Test, the Word Association Test, and forehead skin temperature, there was a significant positive correlation between self-report measures of sexual arousal and penile tumescence.The authors would like to express their appreciation to Elizabeth Anne Kirschling, Lorraine O'Brien, and Beverly Wojciehowski, the nursing staff of the Alcohol Behavior Research Laboratory, for their assistance in conducting this study. The authors also wish to thank Harry Heermans for his role in the data collection, Wayne Kashinsky for suggesting computer-sampling techniques and for programming the computer, and Peter E. Nathan for his support and encouragement.Requests for reprints should be sent to G.