Resting State Networks spatio-spectral fingerprints: the Default Mode Network case study

The resting state functional magnetic resonance imaging (fMRI) approach has allowed to investigate the large-scale organization of processing systems in the human brain, revealing that it can be viewed as an integrative network of functionally interacting regions. However, to date, the neuronal basis of the fMRI signal dynamics at rest are not fully understood, weakening the fMRI capability to explain brain activity. In this scenario, the integration with information derived from electroencephalography (EEG) is very useful, since conversely from fMRI, EEG represents a direct measure of neuronal activity. EEG-fMRI resting state studies investigating the correlation between fMRI signals and corresponding global EEG spectral characteristics in single spectral bands have provided a certain degree of inconsistency in the results. This may be due to the fact that the distinct functional networks involve more than a single frequency band, and therefore analysis of simultaneous EEG/fMRI data should consider the whole frequency spectrum. A couple of studies have been performed in this direction but they did not investigate the impact of the scalp distribution of EEG spectral metrics. To overcome this gap, this study aims to identify the spatio-spectral fingerprints of distinct networks by using an analytical approach that takes into account the interplay between the different EEG frequency bands and the corresponding topographic distribution within each network. This approach was applied to four sub-components of the Default Mode Network (DMN), revealing for the first time the distinctive subcomponent-specific spatial-frequency patterns of correlation between the fMRI signal and EEG rhythm.