Alcohol consumption as a self-handicapping strategy.

Two experiments evaluated the self-handicapping hypothesis that alcohol consumption varies directly with individuals' uncertainty of their ability to perform successfully. In a 2 X 2 X 2 factorial design, Experiment 1 manipulated (a) difficulty of an initial intellectual test (insolvable or solvable), (b) feedback regarding test performance (success or none), and (c) instructions regarding the difficulty of a retest (identical to or harder than the initial test). Ninety-six male normal drinkers then received access to an alcoholic beverage (self-handicapping option) and to study materials (performance-enchancing option). The experiment terminated before the retest. Results indicated that when a performance-enhancing option is available, subjects generally do not use alcohol to self-handicap. Experiment 2 omitted the study option and manipulated the test difficulty and retest instruction variables in a 2 X 2 factorial design. All 32 subjects received success feedback. Results showed that subjects use alcohol to self-handicap when denied access to a performance-enhancing option. With important qualifications, these data support the self-handicapping hypothesis of human drinking behavior.