T135. ABERRANT EXECUTIVE CONTROL AND AUDITORY NETWORKS IN RECENT-ONSET SCHIZOPHRENIA - A GROUP ICA DUAL REGRESSION ANALYSIS

Abstract Background A great number of studies using resting-state functional MRI (rsfMRI) have been conducted for revealing altered functional connectivity in schizophrenia. Although the default mode network has been predominantly investigated, other resting-state networks (RSNs) can be networks of interest for unveiling the neurobiological basis of schizophrenia. Our study investigated changes in functional connectivity in the nine RSNs in recent-onset schizophrenia. Methods We performed a rsfMRI in 46 patients with recent-onset schizophrenia and 22 healthy controls. In the preprocessing steps, independent component analysis (ICA)-based automatic removal of motion artifacts was applied to eliminate motion-related components from an individual-level ICA. Group ICA and dual regression were performed for voxel-wise comparisons between the groups with the non-parametric permutation test using threshold-free cluster enhancement and a corrected-p < 0.05 for multiple comparisons. We used the Positive and Negative Syndrome Scale scores for the correlation analysis of symptom severity with functional connectivity in the patient group. Results The control group had areas with a significant group difference in activity compared to that of the patient group in which it existed in the left supplementary motor cortex and supramarginal gyrus (the executive control network) and the right postcentral gyrus (the auditory network). There were no significant correlations of symptom severity with the functional connectivity of these regions. Discussion The left supplementary motor cortex, supramarginal gyrus, and right postcentral gyrus having areas with a significant group difference may have a role in the pathophysiology of schizophrenia, given their language and auditory information processing functions.