Dynamic organization of visual cortical networks inferred from massive spiking datasets

Abstract

We introduce a probabilistic model to discover the dynamic connectivity structure in a network of spiking cortical neurons. The model is evaluated on ground truth synthetic data and compared to alternative methods to ensure quality and quantification of model predictions. When applied to large-scale recordings of spiking activity in a mammalian visual cortex, the model reveals the latent dynamic functional interactions in cortical networks that is typically omitted from analysis.