Resting-state functional dynamic connectivity and healthy aging: A sliding-window network analysis

Abstract

Background: Graph theory has been widely used to study structural and functional brain connectivity changes in healthy aging, and occasionally with clinical samples; in both cases, during task-related and resting-state experiments. Recent studies have focused their interest on dynamic changes during a resting-state fMRI register in order to identify differences in non-stationary patterns associated with the aging process. The objective of this study was to characterize resting-state fMRI network dynamics in order to study the healthy aging process. Method: 114 healthy older adults were measured in a resting-state paradigm using fMRI. A sliding-window approach to graph theory was used to measure the mean degree, average path length, clustering coeffi cient, and smallworldness of each subnetwork, and the impact of age and time in each graph measure was assessed. Results: A combined effect of age and time was detected in mean degree, average path length, and small-worldness, where participants aged 75 to 79 showed a curvilinear trend with reduced network density and increased small-world coeffi cient in the middle of the register. Conclusion: An effect of age was observed on average path length, with younger participants showing slightly lower scores.