Sleep disorders are a common and important part of schizophrenia's clinical Picture; however, the number of polysomnography-based studies of schizophrenia is limited and there is a lack of consensus regarding a specific sleep pattern in schizophrenia patients. As such, the aim of the present study was to investigate the sleep architecture in non-medicated schizophrenia patients.The study included 13 adult male inpatients with schizophrenia, undifferentiated type, (based on DSM-IV-TR criteria) and an age- and sex-matched group of normal controls. The participants were studied during 2 consecutive nights in the sleep laboratory. The Brief Psychiatric Rating Scale (BPRS), Scale for Negative Symptoms (SANS), and Scale for Positive Symptoms (SAPS) were used for clinical assessment. Polysomnographic recordings obtained on the second night were used for analysis.Compared to the controls, the schizophrenic patients had less total sleep time, lower sleep efficiency, longer sleep latency, more awakenings, and increased duration of awakenings after falling asleep. In terms of sleep architecture, the schizophrenia patients showed no evidence of abnormal-slow wave sleep, but the percentage of REM sleep was reduced. REM sleep measures, including REM latency and density, did not significantly differ between the 2 groups. Based on correlation analysis between the sleep parameters and clinical symptoms, slow-wave sleep was inversely correlated with formal thought disorder.The findings indicate that in addition to decreased REM sleep time, disturbances in sleep initiation and maintenance were prominent in the non-medicated schizophrenia patients. The correlation between decreased decreased REM sleep and, slow-wave sleep, and formal thought disorder we observed in the patients might have been related to the underlying pathophysiology of schizophrenia.