Frequency-specific structure-function decoupling underlies core symptoms in schizophrenia

Abstract

A major challenge in schizophrenia is its clinical heterogeneity, which has obstructed biomarker identification. Recent research indicates that frequency-specific structural eigenmodes, which govern distinct spatiotemporal propagation patterns, are linked to the functions of unimodal and transmodal regions. Notably, differential dysfunction in these regions has previously been associated with specific symptoms of schizophrenia. Based on this, we hypothesized that frequency-specific SC–FC decoupling underlies the core symptoms in schizophrenia. To test this hypothesis, our study examined 42 patients with schizophrenia versus 51 healthy controls. Low- and high-frequency components were extracted from structural connectivity and fed into a deep attention network to predict functional connectivity. Frequency-specific coupling values were computed to assess their associations with clinical symptom scores. Additionally, relationships between coupling differences and neurotransmitter distributions as well as gene expression profiles were examined. Our results showed that patients exhibited frequency-specific SC–FC decoupling. Low-frequency coupling between the default mode and sensorimotor networks correlated with hallucination severity, while intra-sensorimotor low-frequency coupling was associated with delusions and insufficient spontaneity and conversational fluidity; high-frequency intra-sensorimotor coupling was also linked to conversational deficits. Between-group coupling differences were significantly associated with specific neurotransmitter distributions and gene expression patterns. Taken together, these findings demonstrate that frequency-specific SC–FC decoupling may underlie the core symptoms of schizophrenia, providing mechanistic insight and a potential foundation for precision-targeted interventions.