Electroencephalography (EEG) microstate analysis provides a marker of the temporal dynamics of large-scale functional brain networks. While exercise is known to influence resting-state brain activity, its relationship with dynamic network organization remains unclear. We therefore investigated associations between physical fitness and EEG microstate dynamics at rest. In 30 healthy subjects, VO2max and lower-limb strength was tested using spiroergometric testing on a cycling ergometer and isokinetic strength testing. A 32-channel resting EEG was recorded (2 min eyes open/closed) and temporal parameters for four representative microstate topographies (A – D) were extracted. Associations between physical fitness and temporal parameters of microstate dynamics were examined using partial least squares correlation (PLSC), complemented by FDR-corrected univariate regressions for features showing stable multivariate contributions (Bootstrap ratio ≥ 2. 5). The PLSC analysis shows a significant multivariate association between the microstate parameters and strength (r = 0. 71, pₚerm = 0. 028). The regression analyses on the stable features identified by the PLSC model showed that higher strength was associated with longer duration of microstates B (ß = 0. 488, qFDR = 0. 036, R 2 = 0. 36) and C (ß = 0. 493, qFDR = 0. 36, R 2 = 0. 34) and lower individual explained variance of microstate C (ß = − 0. 536, qFDR = 0. 36, R 2 = 0. 20). No such association was identified for VO2max. Our results provide first evidence that interindividual differences in physical fitness are also reflected in the temporal organization of large-scale brain networks. Future studies should determine whether longitudinal exercise interventions can induce such neurobiological adaptations. • Strength showed multivariate associations with EEG microstate parameter • Higher individual strength predicted longer durations of microstates B and C • VO₂ max showed no significant associations with microstate parameters • Interindividual fitness differences might be reflected in large-scale brain network dynamics
Dreismickenbecker et al. (Wed,) studied this question.