Brain network dynamics in the alpha band during a complex postural control task

Abstract

Objective. To decipher brain network dynamic remodeling from electroencephalography (EEG) during a complex postural control (PC) task combining virtual reality and a moving platform. Approach. EEG (64 electrodes) data from 158 healthy subjects were acquired. The experiment is divided into several phases, and visual and motor stimulation is applied progressively. We combined advanced source-space EEG networks with clustering algorithms to decipher the brain networks states (BNSs) that occurred during the task. Main results. The results show that BNS distribution describes the different phases of the experiment with specific transitions between visual, motor, salience, and default mode networks coherently. We also showed that age is a key factor that affects the dynamic transition of BNSs in a healthy cohort. Significance. This study validates an innovative approach, based on a robust methodology and a consequent cohort, to quantify the brain networks dynamics in the BioVRSea paradigm. This work is an important step toward a quantitative evaluation of brain activities during PC and could lay the foundation for developing brain-based biomarkers of PC-related disorders.