......................................................................

NJIT Mathematical Biology Seminar

Tuesday, November 7, 2006, 4:00pm
Cullimore Hall 611
New Jersey Institute of Technology

......................................................................

Signal Gating and Detailed Balance in Networks of Model Neurons

Tim Vogels

Columbia University School of Medicine

Cognitive processing involves communication both within and between brain regions and requires precise dynamic control over which sub- circuits respond at any given time. Although considerable work has been done on signal propagation in networks, it is still unclear how propagation is gated to control the flow of information between specific sub-regions. In our work, we extend the idea of globally balanced network states to what we call "detailed balance", meaning that the levels of excitation and inhibition along specific signaling pathways are similarly balanced. This means that, in the default state, signals fail to propagate because of cancellation between excitation and inhibition generated in the target region for each signaling pathway. Signal flow can be turned on by modulating the gain of some of the transmitting neurons such that gain is increased for neurons that generate net excitation and decreased for those generating net inhibition (through interneurons in the target region). This allows the flow of information to be gated between regions. We illustrate this concept in large networks of integrate-and-fire neurons.