The dynamic reorganization of the default-mode network during a visual classification task

The default-mode network has been reported to possess highly versatile and even contrasting functions but the underlying functioning mechanism remains elusive. In this study, we adopt a dynamic view of the default-mode network structure and hypothesize that it could potentially contribute to different functions through dynamic reorganization of its functional interaction pattern within and across network boundaries depending on the ongoing cognitive demands. With four experimental states and functional connectivity magnetic resonance imaging, we show that the default-mode network is characterized by within-network desynchronization and outside-network integration during the transition from resting state to an external visual classification task. Such default-mode network dynamics are task demand-dependent and return to their original status during the transition back to resting. More importantly, the degree of within-network desynchronization correlates with reaction time while the level of outside-network integration indexes task performance accuracy. Overall, the documented dynamic reorganization of the default-mode network and the significant behavioral correlations provide new insights into our understanding of this complex network and emphasize a dynamic view in future studies of its functioning mechanism.