Seminar: John M. Beggs

The Spring 2009 Biocomplexity Institute Seminar Series presents John M. Beggs, Biocomplexity Institute, Indiana University Bloomington.

Tuesday, Jan. 27, 2009
4:00-5:00 p.m.
Swain Hall West 238
Tea preceding the seminar at 3:30 p.m., Swain Hall West 238

Professor Beggs will discuss, "Information Flow Through Local Cortical Networks Is Not Democratic."

Abstract:
The average cortical neuron makes and receives 1,000-10,000 synaptic contacts, suggesting that cortical neurons are connected in a fairly equal manner. The pattern of information flow in such networks, however, is virtually unknown and can not be inferred from anatomy alone. Theory indicates that the presence of even a few hubs, or nodes with many more connections than average, can greatly improve the efficiency and robustness of a network. Accordingly, we tested for the presence of hubs of information flow in local cortical networks. Information flow was very unequally distributed among neurons, and the data were significantly more likely to have been drawn from networks with hubs than networks without hubs. Simulations indicate the observed cortical information flow networks are significantly more efficient in routing signals, can form more associations, and are more robust than control networks without hubs. To our knowledge, this is the fist demonstration of information flow hubs in local cortical networks. We conclude that the highly unequal distribution of information flow among cortical neurons contributes to the efficiency and robustness of information processing in the cortex.

Work done in collaboration with Alan Litke of UC Santa Cruz, Olaf Sporns of Indiana University, and Duane Nykamp of The University of Minnesota.