Search Papers | Poster Sessions | All Posters
Poster C22 in Poster Session C - Friday, August 9, 2024, 11:15 am – 1:15 pm, Johnson Ice Rink
A comprehensive large-scale model of primary visual cortex (V1)
Shivang Rawat1,2 (), David J. Heeger3,4, Stefano Martiniani1,2,5; 1Courant Institute of Mathematical Sciences, New York University, New York 10003, USA, 2Center for Soft Matter Research, Department of Physics, New York University, New York 10003, USA, 3Department of Psychology, New York University, New York 10003, USA, 4Center for Neural Science, New York University, New York 10003, USA, 5Simons Center for Computational Physical Chemistry, Department of Chemistry, New York University, New York 10003, USA
We introduce a comprehensive retinotopic model of V1 based on ORGaNICs, a stochastic recurrent circuit framework implementing divisive normalization via modulation of recurrent amplification. Specifically, we simulate the membrane potentials and firing rates of complex V1 neurons driven by the outputs of a steerable pyramid, thus capturing the retinotopy, spatial frequency, receptive field size, and orientation-tuning selectivity of the neurons. We further implement a Gaussian-Rectification (GR) model for the generation of spiking activity that takes into account the time-correlations of the membrane potentials. We demonstrate that this GR model accurately captures the dependence of the Fano factor and noise correlations as a function of stimulus contrast. The spike process is then filtered and fed back as input to the dynamical variables simulating the membrane potential of the neurons. Thus, using the theory of stochastic LTI systems, we demonstrate that, for a grating response, the circuit exhibits gamma frequency oscillations and accurately captures the contrast dependence of gamma activity and LFP coherence, measured across neuronal populations tuned to different spatial locations, orientation, and spatial frequency. Finally, we predict these quantities in the context of plaids, and natural images. Therefore, our framework offers a versatile tool for understanding the dynamics and noise correlation of V1 activity.
Keywords: gamma oscillations visual cortex spiking activity