Modeling cortical networks for SSVEP simulations

Synchronous activation of large neural populations generates coherent electromagnetic signals which are detectable on the scalp or cortical surface and are the primary signal modality for some of the most well known brain-machine interface (BMI) applications – including cursor control and item selection. However, the neural dynamics that generate synchronous signals, as well as changes in the amount of synchrony in different frequency bands, are not clearly understood. In this project, we will take the approach of simulating neural activity through computational models and direct comparisons to recorded neural activity, specifically data collected using electroencephalography (EEG). While models abound, they tend to operate under either implicit or explicit assumptions about the underlying neural network connectivity. Recent developments in dynamical systems have provided possible theoretical underpinnings explaining the relationship between a network’s explicit structure and the synchronous activities it may undergo. In particular, the steady state visual evoked potential (SSVEP) will serve as a test-case for explaining the neural connectivity mediating the two-part effect found as a result of observation of a driving visual input. Specifically, low-level neural connectivity analyses will be used to investigate how a large neural network can maintain one stable synchronous state in the absence of visual input (e.g., the alpha rhythm) then switch to another synchronous state in the presence of visual input (e.g., SSVEP) without changing the connectivity structure.

Main personnel Rob Law Funding:
Collaborators Jon Brumberg, Mike Cohen, Frank Guenther CELEST: Investigating properties of neural synchrony with applications for BMI control