Stimulus control induced by (-)-2,5-dimethoxy-4-methylamphetamine (DOM) is believed to be mediated by agonism at 5-HT2A receptors. We hypothesized that blockade of (-) DOM-induced stimulus control may thus prove useful in the pre-clinical characterization of novel antipsychotic agents by providing an in vivo index of antagonism at that receptor. A previous study (Fiorella et al., 1995a) observed no antagonism by typical agents such as haloperidol and thioridazine, partial antagonism by the atypical agent, clozapine, and full antagonism by risperidone, a second atypical antipsychotic. The present investigation extends these observations to include seven additional drugs: SCH 23390, sulpiride, amperozide, melperone, octoclothepin, tiospirone and ritanserin. Of the drugs tested in rats in which (-) DOM-induced stimulus control had reliably been established, only tiospirone and ritanserin produced complete antagonism of the (-) DOM stimulus. Intermediate levels of antagonism were observed following treatment with amperozide, melperone, and octoclothepin. Finally, SCH 23390 and sulpiride yielded no evidence of antagonistic activity in (-) DOM-trained animals. Because clozapine and risperidone are both classified as atypical antipsychotics yet yield different degrees of antagonism of (-) DOM-induced stimulus control, we tested the substitution of risperidone for clozapine in rats trained with clozapine as a discriminative stimulus. No significant substitution was observed. In conclusion it appears that complete or partial antagonism of the (-) DOM stimulus serves as an effective pre-clinical means of identifying antipsychotics with significant 5-HT2A antagonist properties. However, the failure of risperidone to substitute for clozapine in pigeons (Hoenicke et al., 1992) and in rats (present study) suggests that despite their shared 5-HT2A antagonist properties, clozapine and risperidone differ with respect to their stimulus effects.