Using Tiffany's (1990) cognitive model of drug use and craving as a theoretical basis, the present experiments investigated whether cravings for food expend limited cognitive resources. Cognitive performance was assessed by simple reaction time (Experiment 1) and an established measure of working memory capacity, the operation span task (Experiment 2). In each experiment, female undergraduate students were randomly assigned to a craving or control condition. Participants in the craving condition abstained from eating chocolate for 24 hours prior to testing and performed the cognitive task in the presence of chocolate, a manipulation that successfully elicited chocolate cravings. In both experiments, there was no main effect of craving condition on cognition, but there was a significant interaction between condition and trait chocolate craving. In support of Tiffany's model, our results show that habitual food cravers direct limited cognitive resources to craving-related cues, at the cost of competing cognitive demands. (Contains 2 tables and 2 figures.)