Alcohol intoxication, social influence, and self-aggressive behavior in men

Michael S. McCloskey


The importance of determining factors contributing to self-aggressive behavior cannot be overstated. Non-experimental studies have shown that alcohol use and social influence are associated with many forms of self-aggression, including suicide. However, no published studies employing an experimental methodology have tested the hypothesis that alcohol or social influence facilitates self-aggressive behavior. The current study employed an experimental paradigm called the Self-Aggression Paradigm (SAP task) to examine the effect of alcohol intoxication and social influence on self-aggressive behavior. Forty-seven males, aged 21 to 45, were told they were participating in an experiment on alcohol and visual-motor performance. Participants were assigned to one of two alcohol conditions (alcohol vs. no alcohol), and one of two social influence conditions (social influence vs. no social influence). All participants were administered a series of questionnaires and interviews to assess variables related to self-aggressive behavior (e.g., depression, past self-aggressive behaviors). Participants then drank 700 ml of diet Sprite that, for participants in the alcohol condition, also contained a high dose of vodka. Twenty minutes after drinking the liquid, participants engaged in a "competitive reaction time task" (SAP task) against a fictitious opponent in which participants determined the level of electrical shock they would receive if they were slower on a given trial. The level of self-selected shock served as a measure of self-aggression. Participants in the social influence condition were given visual feedback indicating that their opponent was becoming increasingly self-aggressive. Participants in the no social influence condition received identical visual feedback, but were not told that it was related to their opponent's self-shock setting. Results indicated that alcohol intoxication facilitated self-aggression, including the selection of shock levels twice the amount the participant believed to be highly unpleasant. Social influence also appeared to facilitate self-aggression, but only when shock selection did not exceed a level they believed was highly unpleasant. These results occurred independent of any risk factors, and suggest that alcohol use and peer influences are important variables to examine when assessing self-aggression in populations believed to be at risk for such behavior. Results also provide construct validity for the use of the SAP task as a self-aggression paradigm, and possible experimental analogue for suicidal behaviors.