Associations between negative symptoms and resting-state functional connectivity within social brain networks among individuals with early psychosis
摘要
Negative symptoms of psychotic disorders are best represented within a hierarchical structure comprising two broad dimensions—Motivation and Pleasure (MAP) and Diminished Expressivity (EXP)—and five lower-level domains. The validity of these two dimensions and five domains is supported by associations with cognitive, psychological, and clinical outcomes. However, few studies have examined whether they are differentiated by distinct neural mechanisms. The current study examined the specificity of associations between the two dimensions and five domains and resting-state functional connectivity (RS-FC) within five large-scale brain networks critical for social behavior. Participants included 125 early psychosis (EP) patients and 58 healthy controls (CN) from the Human Connectome Project-Early Psychosis who completed resting-state functional magnetic resonance imaging (rsfMRI) scans. RS-FC was quantified in five social brain networks: affiliation network, aversion network, perception network, mentalizing network, and mirror network. Early psychosis patients exhibited significantly reduced RS-FC in social brain networks compared to CN, but no specific network was responsible for this effect. Reduced RS-FC in the mirror network was significantly associated with greater asociality, anhedonia, and avolition, while reduced RS-FC in the mirror and mentalizing networks was associated with more severe blunted affect. Findings suggest that some RS-FC networks align with the broader higher-order dimensions, while others align with the lower-level domains. Overall, the pattern of findings suggests that abnormal patterns of resting-state social brain network activation are broadly associated with the MAP dimension and not more selectively related to anhedonia, avolition, or asociality. Thus, findings suggest that the neurobiology of negative symptoms in EP is best captured at the level of the broader higher-order dimensions than the specific lower-level domains that make them up.