Insights Into Electrophysiological Brain States Dynamics

The human brain is a marvel of complex and dynamic networks that enable us to move, sense, think, and remember. How do these networks constantly reconfigure themselves in real-time to process information? To answer this question, we need reliable methods able to capture the dynamics of the dominant functional 'states'. However, this task is challenging due to the diversity of the available pipelines that are difficult to evaluate. Here, we present an overview of a computational framework that combines electrophysiological data with dynamic functional connectivity approach (dFC) followed by dimensionality reduction methods. We highlighted the utility of such framework in tracking the dynamics of key brain network states. Then, we provide a practical example on real MEG data during fast motor task. Our analysis evaluated the effectiveness of the proposed framework including several instances of decomposition and clustering techniques in capturing relevant spatial and temporal patterns of brain network reconfiguration.