BROADband brain Network Estimation via Source Separation (BROAD-NESS)

This study presents BROADband brain Network Estimation via Source Separation (BROAD- NESS), a novel method tailored for event-related designs, leveraging magnetoencephalography’s (MEG) high temporal and spatial resolution to identify dynamic brain networks without predefined regions. By applying principal component analysis (PCA) to source-reconstructed MEG data from 83 participants in a long-term musical sequence recognition task, BROAD-NESS captured event-related networks more effectively than traditional approaches, revealing two main networks that explained 88% of the variance. The first network, involving the auditory cortices and medial cingulate gyrus, was associated with continuous auditory processing. The second network, encompassing prefrontal and hippocampal regions, inferior temporal cortex, and insula, was linked to memory, confirmed predictions and prediction error processing. With limited computational demands and minimal assumptions, BROAD-NESS offers a powerful tool for studying event-related brain dynamics, enhancing understanding of memory-related networks in recognition of temporal sequences.