Examining clinical symptoms in schizophrenia based on visual-spatial working memory and resting-state EEG

Schizophrenia (SZ) is a serious and disabling mental illness characterized by compromised brain network interactions. The current study investigated the potential associations between the resting-state electroencephalogram (EEG) and individual cognitive traits/clinical recordings in SZ patients. Furthermore, the positive and negative syndrome scale (PANSS) scores based on the connectivity revealed in the resting-state EEG and working memory (WM) when SZ patients participated in the visual-spatial WM task were predicted. The results demonstrated that stronger frontal-parietal linkages for the SZ group were associated with lower response accuracy compared to the HC group in the alpha and beta frequency bands during the resting-state process. There were significant relationships between PANSS scores and the resting-state network properties; increased PANSS scores were accompanied by a larger clustering coefficient, as well as global and local efficiency, and shorter characteristic path length. Moreover, based on the built multivariable linear regression model, the PANSS scores were predicted accurately, as indicated by a rather large correlation coefficient between predicted and actual PANSS scores in the SZ group. Current findings may deepen our knowledge of SZ and accelerate progress in early diagnosis and intervention approaches.