There is accumulating neuroimaging evidence for both structural and functional abnormalities in schizophrenia patients with persistent auditory verbal hallucinations (AVH). So far, the direct interrelationships between altered structural and functional changes underlying AVH are unknown. Recently, it has become possible to reveal hidden patterns of neural dysfunction not sufficiently captured by separate analysis of these two modalities. A data-driven fusion method called parallel independent component analysis (p-ICA) is able to identify maximally independent components of each imaging modality as well as the link between them. In the present study, we utilized p-ICA to study covarying components among gray matter volume maps computed from structural MRI (sMRI) and fractional amplitude of low-frequency fluctuations (fALFF) maps computed from resting-state functional MRI (rs-fMRI) data of 15 schizophrenia patients with AVH, 16 non-hallucinating schizophrenia patients (nAVH), and 19 healthy controls (HC). We found a significant correlation (r = 0.548, n = 50, p < .001) between a sMRI component and a rs-fMRI component, which was significantly different between the AVH and non AVH group (nAVH). The rs-fMRI component comprised temporal cortex and cortical midline regions, the sMRI component included predominantly fronto-temporo-parietal regions. Distinct clinical features, as measured by the Psychotic Symptoms Rating Scale (PSYRATS), were associated with two different modality specific rs-fMRI components. There was a significant correlation between a predominantly parietal resting-state network and the physical dimension of PSYRATS and the posterior cingulate/temporal cortex network and the emotional dimension of PSYRATS. These data suggest AVH-specific interrelationships between intrinsic network activity and GMV, together with modality-specific associations with distinct symptom dimensions of AVH.
Keywords: Auditory verbal hallucinations; Magnetic resonance imaging; Parallel independent component analysis; Schizophrenia.
Copyright © 2019 Elsevier Inc. All rights reserved.