Topological Variations of Large-Scale Brain Functional Networks Based on Mutual Information

Recent resting-state fMRI studies have suggested that inter-regional functional connectivity shows a small-world topology combined with a high level of global connectivity, which indicates an organization of brain in highly clustered sub-networks. However, little is known about the visual-activating functional connectivity in the human brain. In this study, we derive the brain functional networks under resting and visual-activation state out of all cortical regions of interest based on mutual information. The connections among regions reflect inter-region functional connectivity. Specifically, we investigate dynamical evolution of topology of brain functional networks realized by moving window technique. Robust small-world property with higher clustering coefficient and shorter mean distance is found in the brain functional networks under both resting and visual-activation state. More importantly, our results show that the topological structures of brain functional networks display obvious difference during the transition from visual-activation state to resting state. The diversity of topology of brain functional networks under visual-activation state may provide some insights to deeper understanding the mechanism of interaction between the bran functional configuration and cognitive activation.