CRCNS: Dissecting Directed Interactions Amongst Multiple Neuronal Populations

Nearly all brain functions involve activity that is distributed across multiple areas. To understand these functions, it is critical to understand the flow of signals across this distributed network. To date approaches to understanding inter-areal signaling have been limited in several critical ways. First, they often focus on single neurons or a few voxels to summarize an area’s activity—an impoverished sample of the intricate neuronal population activity patterns that are known to represent and transmit information. Second, prior approaches often consider only pairwise inter-areal interactions, though the relevant network of areas is often much larger. Third, they rarely consider the concurrent flow of signals both from and to any given node in the network. In this project, we aim to overcome all three of these limitations. In Aim 1, we will develop and validate statistical methods that allow us to assess the directed, multi-dimensional flow of neuronal population signals among multiple (more than two) brain areas. We will identify directed interactions based on if the activation of a population activity pattern in one brain area tends to reliably precede the activation of a population activity pattern in another brain area with a consistent time delay. In Aim 2, we will refine and deploy the methods we develop to assess signal flow across multiple stages of the macaque visual system, an ideal testbed given a great deal of prior work on the anatomical and functional properties of the sampled areas. Specifically, we will record hundreds of neurons distributed across different layers of primary visual cortex (V1), V2, and V3. We will determine how columnar interactions within each area interact with feedforward and feedback processes, at a laminar level. Our project aims to provide insights that will strongly advance understanding of fundamental aspects of cortical function—how neuronal populations communicate with each other and how that communication relates to cortical processing. We expect the understanding we gain, and the analytic and conceptual tools we develop, will be broadly applicable across different brain systems. Our ambitious goals will be accomplished by pooling complementary expertise of three PIs, building on a successful collaboration that has extended over many years.