Phencyclidine (PCP) administration in rats acutely in high doses or chronically in lower doses causes neurotoxicity characterized by neuronal vacuolization and apoptotic neuronal death, respectively. The purpose of this study was to determine whether drugs that previously had been reported to prevent either type of neurotoxicity were also able to prevent locomotor sensitization following chronic PCP administration. PCP (5 or 20 mg/kg) was administered once a day for 5 days following drug pretreatment. After withdrawal, rats were challenged with 3.2 mg/kg PCP and locomotor activity was assessed. Haloperidol and clozapine significantly attenuated sensitization elicited by PCP (20 mg/kg). The D(1)-like antagonist SCH23390 was much less effective than clozapine, showing a marginal inhibition. Risperidone, a D(2)/serotonin (5-HT(2)) antagonist, also resulted in a marginal attenuation of 15%. Ketanserin, a 5-HT(2) antagonist, had no effect. Atropine retarded sensitization by 35% and (+)-sulpiride caused a 50% reduction. The AMPA/kainate antagonist, 6,7-dinitroquinoxaline-2,3-dione, had no effect, but barbital sodium reduced sensitization by 54%. These data suggest that gamma-aminobutyric acid A, D(2), and muscarinic receptors play a major role in the complex pathway underlying sensitization to PCP, whereas D(1), 5-HT(2) and AMPA receptors have little or no relevance in the behavioral sensitization produced by 20 mg/kg PCP. In a model using 5 mg/kg PCP, the effects of sulpiride and SCH23390 replicated those observed with 20 mg/kg PCP and further showed that acute locomotor activation is not a strict requirement for the development of sensitization. These data argue that there is overlap, but nonidentity, between the mechanisms underlying PCP-induced sensitization and neurotoxicity.