Frontoparietal connectivity and hierarchical structure of the brain's functional network during sleep

Abstract

Frontal and parietal regions are associated with some of the most complex cognitive functions,and several frontoparietal resting state networks can be observed in wakefulness. We used functional magnetic resonance imaging data acquired in polysomnographically validated wakefulness, lightsleep, and slow-wave sleep to examine the hierarchical structure of frequency functional brain network, and to examine whether frontoparietal connectivity would disintegrate in sleep. Whole brain analyses with hierarchical cluster analysis on predefined atlases were performed, as well as regression of inferior parietal lobules(IPL) seeds against all voxels in the brain, and an evaluation of the integrity of voxeltime-courses in subcortical regions-of interest. We observed that frontoparietal functional connectivity disintegrated in sleepstage1 and was absent in deeper sleep stages. Slow wave sleep was characterized by strong hierarchical clustering of local submodules. Frontoparietal connectivity between IPLandsuperior medialand rightfrontal gyrus was lower in sleep stages than in wakefulness. Moreover, thalamus voxels showed maintained integrity in sleep stage1,making in trathalamic desynchronization an unlikely source of reduced thalamocortical connectivity in this sleep stage. Our data suggest a transition from a globally integrated functional brain network in wakefulness to a disintegrated network consisting of local submodules in slow wavesleep, in which frontoparietalinter modular nodes may play a role,possibly in combination with the thalamus.