<p>The present study investigated the effects of a history of alcohol abuse upon behavior variation and repetition. Thirty participants were distributed into two groups: Alcoholics and Control. Participants had to emit eight-response sequences according to a multiple schedule with three components. For the PROB component, sequences were reinforced according to a probability of 0.5, regardless of the level of variation (or repetition) in the emission of the sequences. For the REP component, a single sequence, among the 70 possible ones, produced reinforcers. For the VAR component, to be reinforced a sequence had to differ from the 15 previous ones. For both groups, U-values were lower in the REP component and higher in the PROB and VAR components. Alcoholics showed lower levels of variation, greater sensitivity to contingency changes, and more systematic patterns in the emission of sequences compared to control participants. These results suggest that although individuals with a history of alcohol abuse exhibit more repetitive responding than individuals without such a history, their behavioral repertoires may become more varied under contingencies that explicitly reinforce variation.</p>