Long-term stress shapes dynamic reconfiguration of functional brain networks across multi-task demands

Exposure to sustained stress can have a profound impact on the brain, emotion and cognition, with either adaptive or maladaptive effects. Human functional brain networks are dynamically organized to enable rapid and flexible adaptation to meet ever-changing task demands. Yet, little is known about how long-term stress alters the dynamic reconfiguration of functional brain networks across multi-task demands. Here we show prominent changes in the dynamic reconfiguration of large-scale brain networks during resting-state, emotional and working-memory processing under long-term stress. Hidden Markov Model analysis detected several latent brain states and switching processes involving the default mode, emotional salience and executive-control networks that are dominant to rest, emotion and working memory, respectively. Critically, long-term stress increased persistent time on brain states relevant to goal-directed demands and cognitive control, with more frequent transitions to these brain states when compared to controls. Furthermore, long-term stress led to higher correlations of the occupancy and persistency of brain states linked to psychological distress and behavioral performance. Our findings provide a neurocognitive framework whereby long-term stress shapes the way the brain adapts to varying task demands and increases the sensitivity of functional brain networks to psychological and behavioral responses. These changes can be both adaptive and maladaptive, reflecting the complex effects of long-term stress on brain function.