In this rodent study, we evaluated the effects of different time periods (7, 14, 45, and 90 days) of oral treatment with haloperidol (HAL; 2.0 mg/kg/day) or ziprasidone (ZIP; 12.0 mg/kg/day) on nerve growth factor (NGF) and choline acetyltransferase (ChAT) levels in the hippocampus, and we subsequently assessed water maze task performance, prepulse inhibition (PPI) of the auditory gating response, and several NGF-related proteins and cholinergic markers after 90 days of treatment. Seven and 14 days of treatment with either HAL or ZIP resulted in a notable increase in NGF and ChAT immunoreactivity in the dentate gyrus (DG), CA1, and CA3 areas of the hippocampus. After 45 days, NGF and ChAT immunoreactivity had abated to control levels in ZIP-treated animals, but it was markedly reduced in HAL-treated subjects. After 90 days of treatment, NGF and ChAT levels were substantially lower than controls in both antipsychotic groups. Furthermore, after 90 days of treatment and a drug-free washout period, water maze performance (but not PPI) was impaired in both antipsychotic groups, although the decrement was greater in the HAL group. Several NGFrelated and cholinergic proteins were diminished in the brains of subjects treated with either neuroleptic as well. These data support the premise that, although ZIP (given chronically) seems somewhat superior to HAL due to less pronounced behavioral effects and a more delayed appearance of neurochemical deficits, both antipsychotics produce time-dependent deleterious effects on NGF, cholinergic markers (i.e., important neurobiological substrates of memory), and cognitive function.The newer pharmacological treatments for schizophrenia, now commonly referred to as second generation antipsychotics (SGAs), offer several advantages over first generation antipsychotics (FGAs) such as greater improvements in negative symptoms, prevention of relapse, increased functional capacity and quality of life, and fewer movement-related side effects (for review, see Miyamoto et al., 2005). It is also generally believed that SGAs are superior to FGAs when their effects on cognition are considered (for reviews, see Keefe et al., 1999;Purdon, 1999), and some studies suggest that SGAs improve cognition in schizophrenia. This suggestion is of particular importance given that the degree of cognitive impairment in schizophrenia is recognized as an important predictor of social functioning, unemployment, and even relapse of psychiatric symptoms (for review, see Castner et al., 2004). It should be noted, however, that such conclusions regarding antipsychotics and cognitive function rely primarily on meta-analyses and short clinical trials (i.e., they rarely exceed a few months to 1 year in length). There-