Maximal flexibility in dynamic functional connectivity with critical dynamics revealed by fMRI data analysis and brain network modelling.

OBJECTIVE: The exploration of time-varying functional connectivity (FC) through human neuroimaging techniques provides important new insights on the spatio-temporal organization of functional communication in the brain's networks and its alterations in diseased brains. However, little is known about the underlying dynamic mechanism with which such a dynamic FC is flexibly organized under the constraint of structural connections. In this work, we explore the relationship between critical dynamics and FC flexibility based on both functional magnetic resonance imaging data and computer models.

APPROACH: First, we proposed the connectivity number entropy (CNE), which was an entropy measure for the flexibility of FC. Through an analysis of resting-state fMRI (rs-fMRI) data from 95 healthy participants, we explored the correlation between CNE and long-range temporal correlations (LRTCs), which can represent the critical dynamics. Then, we employed a whole-brain computer model based on diffusion tensor imaging (DTI) to further demonstrate this relationship.

MAIN RESULTS: We found that the most flexible FC is present when the brain is operating close to the critical point of a phase transition. Additionally, around this point, our model can yield the best prediction for the regional distribution of CNE because structural information is reflected the most by the CNE through critical dynamics.

SIGNIFICANCE: Our results not only reveal the underlying dynamic mechanism for the organization of time-dependent FC but also provide a possible pathway to model the flexible functional organization in the human brain and may have potential application in the analysis of altered dynamic FC in diseased brains.