Low-dimensional dynamics of resting-state cortical activity

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• Endogenous brain activity, crucial for cognition, is studied using resting-state functional magnetic resonance imaging (fMRI).
• fMRI reveals robust spatial networks but has limited temporal resolution.
• Electroencephalography (EEG) offers complementary temporal insights into brain dynamics.

Purpose of the Study:

• To develop a novel method for reconstructing resting-state brain network dynamics from human EEG data.
• To analyze the temporal characteristics and stability of these brain networks.
• To identify the principal networks underlying spontaneous human thought.

Main Methods:

• Utilized a novel multivariate time-frequency interdependence method for EEG data analysis.
• Reconstructed principal resting-state network dynamics.
• Employed resampling techniques to assess network expression stability across subjects.

Main Results:

• Identified seven robust resting-state brain networks with distinct topographic organizations.
• These networks possess unique high-frequency (∼ 5-45 Hz) fingerprints.
• Networks are nested within slow temporal sequences, showing concurrent expression with temporal asymmetries in formation and dissolution.

Conclusions:

• The study uncovers the complex temporal character of endogenous cortical fluctuations.
• A novel method allows reconstruction of brain network dynamics from EEG.
• Findings provide insights into the low-dimensional subspace governing spontaneous brain activity.