Measurement of dynamic task related functional networks using MEG

- A method is developed to track dynamic electrophysiological networks using MEG.
- Method based on ICA applied to timecourses measuring evolution of connectivity.
- Method allows a unique picture of transient networks that support cognition.
- Method validated in MEG data recorded during a Sternberg working memory task.
- Sensory networks observed include visual and sensorimotor.
- Cognitive networks relate to semantic processing, pattern recognition and language.

The characterisation of dynamic electrophysiological brain networks, which form and dissolve in order to support ongoing cognitive function, is one of the most important goals in neuroscience. Here, we introduce a method for measuring such networks in the human brain using magnetoencephalography (MEG). Previous network analyses look for brain regions that share a common temporal profile of activity. Here distinctly, we exploit the high spatio-temporal resolution of MEG to measure the temporal evolution of connectivity between pairs of parcellated brain regions. We then use an ICA based procedure to identify networks of connections whose temporal dynamics covary. We validate our method using MEG data recorded during a finger movement task, identifying a transient network of connections linking somatosensory and primary motor regions, which modulates during the task. Next, we use our method to image the networks which support cognition during a Sternberg working memory task. We generate a novel neuroscientific picture of cognitive processing, showing the formation and dissolution of multiple networks which relate to semantic processing, pattern recognition and language as well as vision and movement. Our method tracks the dynamics of functional connectivity in the brain on a timescale commensurate to the task they are undertaking.