Latent inhibition (LI) is the proactive interference of inconsequential preexposure to a stimulus with its ability to signal significant events, and disrupted LI is considered to model positive symptoms of schizophrenia. We have recently shown that lesions of the nucleus accumbens core (NACc), basolateral amygdala (BLA) and orbitofrontal cortex (OFC) produce abnormally persistent LI, and suggested that this phenomenon may model negative symptoms. Here we tested whether NACc, BLA and OFC lesion-induced persistent LI would be reversed by the atypical antipsychotic drug (APD) clozapine but not by the typical APD haloperidol. Because clozapine's action is likely reflecting its 5HT2A receptor antagonism, we also tested whether NACc lesion-induced persistent LI would be reversed by the selective 5HT2A antagonist M100907. LI was measured in a conditioned emotional response procedure by comparing suppression of drinking in response to a tone in rats receiving 0 (non-preexposed) or 40 tone presentations (preexposed) followed by five tone-shock pairings. Under these conditions, control rats did not show LI but all lesioned rats persisted in exhibiting LI, and this was reversed by clozapine but not by haloperidol. In addition, M100907 reversed NACc lesion-induced persistent LI. These two novel phenomena, abnormally persistent LI and its selective reversal by an atypical APD, suggest a novel index of schizophrenia relevant behavioral abnormality and of atypical antipsychotic activity in the LI model. The identification of brain regions whose damage leads to persistent LI in the rat may provide valuable cues on dysfunctional brain circuits involved in negative symptoms and in the action of atypical APDs.