Tom Hummer1,2, Matthew Yung1,2, Joaquin Goni3, Michael Francis1,2, Susan Conroy1,2, Nikki Mehdiyoun1,2, Alan Breier1,2; 1Department of Psychiatry, Indiana University School of Medicine, 2Indiana University Psychotic Disorders Program, 3Purdue University
Characterizing dysconnectivity in psychosis patients can be advanced by understanding the interactions of functional networks. This investigation aims to distinguish network characteristics of functional connectivity in patients with early-phase psychosis (EPP) both within and between distinct brain networks. EPP subjects (n=56, within four years of treatment initiation) and matched controls (n=32) underwent resting-state fMRI. Time-series data underwent advanced processing and scrubbing methods to minimize noise. For each subject, gray matter was segmented into 278 regions, and organized into nine distinct a priori networks: visual, somatomotor, dorsal attention, ventral attention, limbic, frontoparietal, default mode, subcortical, and cerebellum networks. Network connectivity properties were defined by the mean and variance of correlations of all regions within the same network (e.g. visual-visual) and between distinct networks (e.g. visual-limbic). Connectome measures, including global efficiency and within-module z-scores, were also calculated. Global efficiency was higher in the control group, reflecting consistently stronger connectivity. The control group had higher connectivity between limbic and default mode networks and multiple other networks, but no within-network connectivity differences were found. The variance of between-network connections was generally higher in patients, particularly for connections with ventral attention network regions. In addition, patients with longer illness duration had greater variability in their network connectivity patterns. Thus, lower functional connectivity in EPP, relative to controls, was largely present between distinct networks, suggesting poor inter-network communication. Higher variance of between-network connectivity in patients suggests disorganized communication patterns that align inconsistently with typical functional networks, disruptions that may increase as the disorder progresses.
