Abstract:This study investigates brain dynamic connectivity patterns in psychosis and their relationship with psychopathological profile and cognitive functioning using a novel dynamic connectivity pipeline on resting-state EEG. Data from seventy-eight individuals with first-episode psychosis (FEP) and sixty control subjects (CTR) were analyzed. Source estimation was performed using eLORETA, and connectivity matrices in the alpha band were computed with the weighted phase-lag index. A modified k-means algorithm was employed to cluster connectivity matrices into distinct brain network states (BNS), from which metrics were extracted. The segmentation revealed five distinct BNSs. FEP exhibited significantly lower connectivity power in BNS 2 and 5 and a greater duration dispersion in BNS 1 than CTR. Negative correlations were identified between BNS metrics and negative symptoms in FEP. In CTR, correlations were found between BNS metrics and cognitive domains. This analysis method highlights the variability of neural dynamics in psychosis and their relationship with negative symptoms.

Resting-State Electroencephalography Alpha Dynamic Connectivity: Quantifying Brain Network State Evolution in Individuals with Psychosis