Exploring brain efficiency during exercise: A fNIRS graph theory analysis

Abstract

This study seeks to elucidate the reduction in brain functional network connectivity during exercise compared to rest, utilizing graph theory techniques to analyze data from resting and movement phases across various exercise modalities. This study employed a graph theory approach to examine differences in brain network functions across various motor phases. The participants engaged in upper limb rehabilitation exercises, including passive, active, and resistance exercises, while functional near-infrared spectroscopy was used to monitor the motor-related cortex. Functional connectivity was reduced during exercise compared with rest, particularly during active and resistance exercises. Small-world network properties did not vary significantly between the two phases, although these properties were higher during movement under conditions of high sparsity. Both global and local efficiencies remained largely unchanged between phases. However, local efficiency increased during the active and resistance exercises in the movement phase. Node efficiency analysis indicated that the motor and supplementary motor areas played critical roles during exercise, with the movement phase exhibiting shorter path lengths. While the overall brain functional connectivity decreased during exercise, there was an improvement in the efficiency of specific brain nodes, suggesting a network mechanism that supports movement execution. During exercise, there was a decrease in whole-brain functional connectivity, yet an enhancement in brain efficiency was observed. This enhancement in functionality of specific nodes may signify the network mechanism responsible for movement execution.