ABSTRACT: In this talk, I will give an overview of the biology associated with the functional structure of the primary visual cortex (V1) and introduce a classical activity-dependent model for the strength of left eye/ right eye inputs. Activity-dependent competition results in the formation of ocular dominance (OD) stripes (stripe-like cortical regions driven by a single eye). Mathematically the model simplifies to a partial integro-differential equation, and a stability analysis reveals a Turing instability that yields the formation of OD stripes.

I will extend the model to present a multi-layer, activity-dependent model for the joint development of ocular dominance (OD) stripes and cytochrome oxidase (CO) blobs. CO blobs are associated with color processing, respond to stimuli with low spatial frequency, and are distributed periodically along the center of OD stripes. Using a correlation-based Hebbian learning rule with subtractive normalization, we show how the formation of an OD map in lower laminae is inherited by upper laminae via vertical projections. Competition between these feedforward projections and direct thalamic input results in the formation of CO blobs superimposed upon the OD map. The resulting CO blob distribution is shown to be consistent with experimental data with the blobs aligned in the center of the OD stripes.

BIO: Dr. Oster received his Ph.D. from the University of Utah in the field of Mathematical Biology working on mathematical models of cortical development. He’s held post-doctoral research positions at the Mathematical Biosciences Institute and at the Group for Neural Theory (Ecole Normale Superieure) where he examined computational biology models of mitochondrial calcium excitability and neural modeling of addiction, respectively. At Eastern Washington University, Dr. Oster has continued research on the dynamics of dopamine neurons and in neurodevelopment, and he’s expanded his interests to include epidemiology and on statistical models to predict risk of homelessness.