Affiliation:
1. University of Helsinki
2. University of Glasgow
3. Aalto University
Abstract
Abstract
Visual working memory (VWM) comprises of distinct short-term storage of sensory information coupled with executive processes. It has remained untested whether long-range synchronization of neuronal oscillations in the human brain would reflect sensory contents of VWM. We used a VWM task in which memorized contents were precisely controlled to test the sensitivity of synchronization to VWM contents using concurrent magneto- and electroencephalography (M/EEG) combined with source-connectivity analysis. We show here that alpha-band phase-synchronization and theta-band desynchronization track VWM contents and correlates with performance. The alpha-band network was found in visual areas corresponding to the memorized features coupled to the fronto-parietal network (FPN) such that key connections of the FPN network were shared across contents. Machine-learning analysis established that only alpha-band networks contained content-information and allowed decoding of VWM contents. Our results establish that alpha-band phase synchronization reflects remembered VWM contents and provides a novel perspective for the organization of VWM.
Publisher
Research Square Platform LLC