In the first experiment, the conditioned place preference (CPP) paradigm was used to examine the rewarding properties of bilateral microinfusions of cocaine HCl into the nucleus accumbens (0, 12.5, 25, 50, or 100 micrograms). No dose of intra-accumbens cocaine induced a significant CPP. However, bilateral intra-accumbens infusions of d-amphetamine sulfate (10 micrograms) or intraperitoneal administration of cocaine HCl (5 or 10 mg/kg) both produced a significant preference for the drug-paired compartment. In the second experiment, the ability of bilateral intra-accumbens infusions of cocaine HCl (50 micrograms) to elicit conditioned locomotor activity (CLA) was examined. During the conditioning trials, intra-accumbens cocaine significantly increased locomotor activity. On the test day, when no drug was administered, the group that had previously received cocaine in the activity chamber showed significantly greater locomotor activity than the vehicle control group. This demonstration of CLA indicates that rats are able to associate the effects of intra-accumbens infusions of cocaine with environmental stimuli; however, these infusions are not rewarding as measured by the CPP paradigm. In addition, these results may indicate important differences between the neural substrates for cocaine and amphetamine reward and reveal a dissociation between CPP and CLA.