Dynamic functional network connectivity alterations in obesity

Obesity involves both metabolic and neural dysfunction, yet the temporal dynamics of brain connectivity remain unclear. This study applied dynamic functional network connectivity (dFNC) analysis to reveal time-varying brain network patterns in obesity.

Eighty-three individuals with obesity and 40 normal-weight controls underwent resting-state functional magnetic resonance imaging. After preprocessing and group independent component analysis, dFNC was estimated using a sliding-window approach and clustered into distinct connectivity states. Temporal metrics (fraction time, dwell time and transitions) were compared between groups, and correlations with clinical characteristics were analysed.

Three recurring connectivity states were identified. Compared with controls, individuals with obesity showed enhanced coupling among the default mode, attention and visual networks, with reduced network flexibility-manifested as prolonged dwell time and fewer transitions. Uncontrolled eating correlated positively with time spent in maladaptive states, whereas cognitive restraint was negatively associated with participation in integrative states.

Obesity is characterised by state-dependent reorganisation of large-scale brain networks and diminished temporal flexibility. These dynamic connectivity alterations are closely related to eating behaviour and metabolic characteristics, suggesting that dFNC provides a valuable neuroimaging framework for understanding impaired self-regulation in obesity and for guiding future intervention studies.