Altered Global and Local Network Organization in Exercise Dependence: Evidence from Graph Theory Analysis of Resting-state EEG

Abstract

Exercise dependence is a behavioral disorder characterized by an obsessive and uncontrolled compulsion to exercise, which ultimately leads to detrimental outcomes for both physical health (e.g., injury, exhaustion) and mental state (e.g., anxiety, impaired social functioning). To investigate whether exercise-dependent individuals show altered brain network properties and if these properties relate to specific addictive behavioral tendencies. Resting-state EEG signals were collected from two groups of adult participants: individuals with a high risk of exercise dependence and those with a low risk, as classified using the Chinese version of the Exercise Dependence Scale (EDS). Source analysis and brain network graph-theory analysis were applied to examine key global and local network measures. The results show that group differences were observed in local graph measures (nodal degree of the right ventral prefrontal cortex and left temporal pole), which survived FDR correction, as well as in global graph measures (global clustering coefficient and small-worldness), which did not reach FDR-corrected significance but showed moderate effect sizes. Furthermore, the nodal degree of the left temporal pole, global clustering coefficient, and small-worldness were independently associated with the total score of the Chinese version of the EDS and correlated significantly with scores on specific sub-scales. The findings suggest that high EDS individuals demonstrate altered brain functional networks that are significantly associated with their specific addictive behavioral tendencies.

Keywords: Behavioral addiction; Exercise dependence; Graph theory analysis; Resting-state EEG.