"Simulation of Neocortical Epileptiform Activity Using Parallel Computing"
W. van Drongelen, H. C. Lee, M. Hereld, D. Jones, M. Cohoon, F. Elsen, M. E. Papka, and R. L. Stevens,
Neurocomputing, vol. 58, , pp. 1203-1209. Also Preprint ANL/MCS-P1170-0604
Preprint Version: [pdf]
A scalable network model intended for study of neocortical epileptiform activity was built on the pGENESIS neural simulator. The model included superficial and deep pyramidal cells plus four types of inhibitory neurons. An electroencephalogram (EEG) simulator was attached to the model to validate model behavior and to determine the contributions of inhibitory and excitatory neuronal populations to the EEG signal. We examined effects of overall excitation and inhibition on activity patterns in the network and found that the network-bursting patterns occur within a narrow range of the excitation-inhibition space. Further, we evaluated synchronization effects produced by gap junctions during synchronous and asynchronous states.