Antipsychotic drugs (APDs) selectively disrupt conditioned avoidance responding (CAR)--a feature that distinguishes them from all other psychotropics. It is thought that this effect reflects their effect on motor initiation; however, this conclusion is questionable because most studies it relies on have often examined avoidance responding under APD treatment, and tested animals with preshock stimuli followed by the footshock. APD-induced CAR effects are confounded by APDs' motor effects and by the presence of footshock. The objective of this study was to evaluate the motor initiation hypothesis by testing animals without drug and under extinction conditions. In Experiment 1, we administered haloperidol, clozapine or chlordiazepoxide (an anxiolytic as a pharmacological control) during the acquisition phase of CAR, but tested animals 2 days later. The APD-induced CAR disruption was present even in the absence of the drug and footshock. In Experiment 2, we first trained rats to a learning criterion, and then subjected them to 4 days of CAR extinction training under drug or vehicle. In the subsequent CAR extinction tests, the rats previously treated with APDs still showed significantly lower avoidance responses. In both experiments, the effects of haloperidol and clozapine were distinct from those of chlordiazepoxide. These data suggest that APD-induced CAR decreases cannot be explained as the unconditioned motor impairment effects of APDs, but probably reflect a dopamine-blockade-mediated change in incentive motivation. Copyright 2004 Elsevier Inc.