Network state transitions and connectivity alterations in attention deficit/hyperactivity disorder and schizophrenia: a dynamic fMRI study

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Abstract Background Dynamic functional network connectivity (dFNC) has become a powerful method for assessing time-varying brain dynamics in psychiatric disorders. However, it is unclear whether disruptions in network dynamics reflect disorder-specific dysfunction or shared transdiagnostic mechanisms. This work examined network state transitions and connectivity patterns in schizophrenia (SZ) and attention-deficit/hyperactivity disorder (ADHD) using resting-state functional magnetic resonance imaging (rs-fMRI) data. Methods A sliding-window approach with k-means clustering identified four distinct, recurring dFNC states. Group differences in state-specific connectivity patterns, mean dwell time (MDT), fraction of time (FOT), and state transitions were evaluated along with correlations between dFNC temporal metrics and clinical symptoms. Results Schizophrenia demonstrated increased MDT and FOT in states characterized by reduced network integration and greater segregation, compared to both ADHD and healthy control (HC). ADHD spent more time in states reflecting moderate connectivity patterns, suggesting altered but more flexible network dynamics relative to schizophrenia. Schizophrenia also showed increased network connectivity between cerebellar, control, and sensory networks, compared to ADHD. Clinically, higher MDT and FOT in state 2 were associated with physical anhedonia in SZ, while hyperactivity symptoms in SZ were associated with FOT in state 1. Conclusion These findings show that psychiatric disorders are characterized by instability in brain network dynamics, with both shared and disorder-specific features. Our results also align with frameworks emphasizing that psychopathology arises from graded disruptions in interacting neural systems rather than discrete diagnostic categories. These dFNC features may serve as potential biomarkers for symptom dimensions in schizophrenia and ADHD.