TOPOLOGY AND DYNAMICS OF BRAIN INTRINSIC CONNECTIVITY NETWORKS REFLECT BEHAVIORAL CHANGES IN HEALTH AND DISEASE

Abstract

Spontaneous fluctuations of functional Magnetic Resonance Imaging (fMRI) signals in subjects at resting state are synchronized over anatomically distributed brain regions, forming intrinsic connectivity networks (ICNs). Neuroimaging studies have investigated ICNs to understand the role of the large-scale functional network in supporting cognition and to characterize brain dysconnectivity in neuropsychiatric disorders. Nevertheless, there is a lack of study elucidating the relationship between the topology of ICNs and the symptomatology in the psychosis. Moreover, although rich spatial-temporal features can be extracted from ICNs, the physiological and behavioral significance of ICN dynamics is largely unknown. Here, we first examined the strength of insular connectivity in subjects at-risk for psychosis to test the integrity of salience processing ICN in pre-clinical psychosis. Further, when applying graph theory, we found altered topological measures and community structures in ICN that predicted clinical outcomes in at-risk individuals. Next, by extracting time-varying features in ICNs, we found distinct arousal-associated dynamic connectivity states that were related to individual differences in vigilance task performance and fluctuations in intra-individual vigilance over time. These findings demonstrate the utility of ICN as a neuroimaging tool that links organization and dynamics of neural networks to behavioral changes in the diseased and healthy brain.