Reconstructing Large-Scale Brain Resting-State Networks from High-Resolution EEG: Spatial and Temporal Comparisons with fMRI.

Functional magnetic resonance imaging (fMRI) studies utilizing measures of hemodynamic signal, such as the blood oxygenation level-dependent (BOLD) signal, have discovered that resting-state brain activities are organized into multiple large-scale functional networks, coined as resting-state networks (RSNs). However, an important limitation of the available fMRI studies is that hemodynamic signals only provide an indirect measure of the neuronal activity. In contrast, electroencephalography (EEG) directly measures electrophysiological activity of the brain. However, little is known about the brain-wide organization of such spontaneous neuronal population signals at the resting state. It is not entirely clear if or how the network structure built upon slowly fluctuating hemodynamic signals is represented in terms of fast, dynamic, and spontaneous neuronal activity. In this study, we investigated the electrophysiological representation of RSNs from simultaneously acquired EEG and fMRI data in the resting human brain. We developed a data-driven analysis approach that reconstructed multiple large-scale electrophysiological networks from high-resolution EEG data alone. The networks derived from EEG were then compared with RSNs independently derived from simultaneously acquired fMRI in their spatial structures as well as temporal dynamics. Results reveal spatially and temporally specific electrophysiological correlates for the fMRI-RSNs. Findings suggest that the spontaneous activity of various large-scale cortical networks is reflected in macroscopic EEG potentials.