To elucidate the "atypicality" of ziprasidone, its striatal and extrastriatal D2/D3-receptor binding was characterized in patients with schizophrenia under steady-state conditions. These data were compared with striatal receptor occupancy values after single-dose ziprasidone ingestion in healthy controls. [F]fallypride positron emission tomography (PET) recordings were obtained in 15 patients under steady-state ziprasidone treatment at varying time points after the last dose. Binding potentials were calculated for striatal and extrastriatal regions. D2/D3-receptor occupancies were expressed relative to binding potentials in 8 unmedicated patients. In a parallel [C]raclopride-PET study, striatal D2/D3-receptor occupancy was measured in healthy subjects after single oral doses of 40 mg ziprasidone or 7.5 mg haloperidol. Ziprasidone plasma concentrations correlated significantly with D2/D3-receptor occupancies in all volumes of interests. Occupancy in extrastriatal regions was approximately 10% higher than in striatal regions. Half maximal effective concentration values were consistently higher in striatal than in extrastriatal regions (temporal cortex: 39 ng/mL; putamen: 64 ng/mL), irrespective of the time between last dosing and scan. Single ziprasidone doses resulted in higher occupancies exceeding the 95% prediction limits of the occupancy versus plasma concentrations for chronic dosing. Ziprasidone shares moderate preferential extrastriatal D2/D3-receptor binding with some other atypicals. D2/D3-receptor occupancy is rapidly attuning to the daily course of ziprasidone plasma levels, suggesting relatively high intraday variations of D2/D3-receptor binding. The discrepancies between single-dose and steady-state results are important for the future design of dose-finding PET occupancy studies of novel antipsychotics. Single-dose studies may not be totally relied on for final dose selection.