ObjectiveThis study was conducted to investigate the cognitive factors that can longitudinally predict the response to treatment in patients with schizophrenia.MethodsThe subjects were 49 patients with schizophrenia who were newly hospitalized in an acute psychiatry ward and had not been treated with medication for at least 8 weeks prior to the study. The symptoms and cognitive functions of the patients were evaluated at baseline before treatment (T0), at eight weeks after treatment (T1), and one year after treatment (T2). Clinical symptoms were assessed using the PANSS, and cognitive functions were estimated using the Vigilance Test, Cognitrone Test, Wisconsin Card Sorting Test (WCST), and the Korean version of the Memory Assessment Scales (K-MAS).ResultsThe patient group showed marked impairments in cognitive function when compared to the normal group, but the patients' clinical symptoms and cognitive functions improved after drug treatment. The patients also showed consistent improvement in verbal and nonverbal memory function as time progressed. Furthermore, there was a significant correlation between clinical symptoms and cognitive functions in the patient group. The cognitive variables that best predicted treatment response and prognosis were total errors on the WCST and immediate list recall component of the K-MAS. It was also shown that the number of total errors on the WCST was a better cognitive predictor than the number of errors in immediate recall.ConclusionThe results of the present study show that the neurocognitive functions of patients with schizophrenia can be stabilized with treatment intervention, that treatment response is related to improvement in cognitive function, and that cognitive domains, especially executive function, can predict treatment response and prognosis in patients with schizophrenia.