Thirty-two male social drinkers were arranged into two tolerance groups, based on changes in standing stability after ingestion of alcohol. Subjects consumed either a large (1.0 g/kg) or small (0.5 g/kg) dose of alcohol. On finishing their drinks, subjects were requested to interact with a female confederate whose continued silence induced anxiety. Heart rate, skin conductance, overt behavior, and self-report measures were taken. Heart rate increased more at the small than the large dose, consistent with the tension-reduction hypothesis. Further, heart rate of high-tolerance subjects increased significantly more than that of low-tolerance subjects, which suggests that alcohol was less effective at tension reduction for the high-tolerance group. Finally, measures of both skin conductance and heart rate showed significant dose-by-tolerance interactions. High-tolerance subjects were more aroused than were low-tolerance subjects at the small but not at the large dose, suggesting that high-tolerance subjects must consume more alcohol to achieve the same autonomic effect experienced by the low-tolerance subjects.