The effects of chronic desmethylimipramine (DMI) treatment on measures of incentive motivation for cocaine were assessed in order to investigate the predictive validity of the extinction/reinstatement model of drug craving. Rats were trained to respond for cocaine infusions (0.75 mg/kg per 0.1 ml i.v.) or received yoked-saline infusions during daily 3-h sessions. A light and tone were presented with the infusions. Following self-administration training, each group received daily injections of either saline or DMI (10 mg/kg, i.p.) for 21 days of withdrawal from the self-administration regimen. On days 12-21 of withdrawal, rats were allowed to respond in the absence of cocaine reinforcement (extinction phase). After reaching an extinction criterion of no responses for 1 h, the cocaine-paired stimuli were repeatedly presented to reinstate responding (reinstatement phase). In the control group, DMI treatment did not alter responding during either test phase, but increased the response latency during the extinction phase. In contrast, DMI treatment in the cocaine group decreased responding and increased the response latency during both test phases, and decreased the extinction latency during the extinction phase. Overall, the effects of DMI were consistent with a reduction of incentive motivation for cocaine, lending support for the predictive validity of the extinction/reinstatement model of drug craving.