We evaluated the effect of antipsychotic dose-reduction on the neurocognitive function of 17 schizophrenic patients (11 male and 6 female, mean age=42.4+/-11.3) who have been taking high-doses of multiple conventional antipsychotics. The mean (+/-SD) of total daily antipsychotic doses (in mg/day, chlorpromazine-equivalent) was 2,253 (+/-668) at baseline, which was reduced to 1,315 (+/-276). Possible changes in neurocognitive function were assessed using Wisconsin card sorting test (WCST) and continuous performance test (CPT). As controls, we examined WCST and CPT in 6 schizophrenic patients (4 male and 2 female, mean age=47.7+/-14.2) who had been taking high-doses of multiple antipsychotics (mean daily antipsychotic dose=1,753+/-165 mg) and declined to change their antipsychotic regimen. In WCST, the mean number of total correct answers significantly increased (53.2+/-16.3 vs. 63.8+/-19.6, P=0.035, Wilcoxon signed rank test); perseverative errors significantly decreased (54.4+/-27.3 vs. 35.4+/-20.1, P=0.013, Wilcoxon signed rank test) after the antipsychotic dose-reduction. In contrast, the control group showed no significant difference between the two WCST performances conducted with a three-month interval. The improvements in WCST performance significantly correlated with the decreases in PANSS negative syndrome score in the subject patients. No significant change was observed in CPT performances in either group. Our preliminary data shows that, in schizophrenic patients taking high-doses of multiple conventional antipsychotics, dose-reduction might lead to improvements in cognitive functions.