Blind source separation can recover systematically distributed neuronal sources from 'resting' EEG

Abstract

Blind source separation algorithms have been increasingly applied to electroencephalographic (EEG) and magnetoencephalographic (MEG) signals from the human brain. Second-order blind identification (SOBI) [1] is one of the emerging algorithms which enable the extraction of functionally distinct, neuro-physiologically, and anatomically meaningful components. SOBI's ability to extract activity associated with a variety of brain sources during visual, auditory, and somatosensory stimulation has been well documented [2]-[6] . Here we demonstrate that SOBI is able to extract a set of neuronal components distributed within the visual, and somatosensory systems, as well as the frontal cortices, from resting EEG data obtained in the absence of explicit sensory stimulation or overt behavioral responses.