Lucas Pinto, M.D, Ph.D. - US grants

DESCRIPTION (provided by applicant): The prefrontal cortex (PFC) is a crucial region involved in goal-directed behavior and top-down modulation of sensory processing. Although much work has been done to study task-related activity in the PFC, the principles underlying its functional organization remain unclear. Here I propose to investigate the intriguing possibility that the functional similarity between PFC neurons is closely related to their shared projection targets, using the PFC projection to the basal forebrain (BF) as a case in point. This pathway is thought to partly mediate PFC's top-down modulation of sensory processing, but this idea still lacks experimental support. Using state-of-the- art retrograde viral tracing techniques, I will firt perform a detailed anatomical characterization of the PFC-BF pathway. The next step will be to combine such techniques with two-photon calcium imaging in awake, behaving mice to identify BF- projecting cells in the PFC and measure their task-related activity, contrasting it with other cell populations within the region. The results will shed light on the functional organization of te PFC, as well as the circuit underpinnings of top-down modulation of sensory processing. Such knowledge is of great potential clinical relevance, given the involvement of PFC and BF in several neurological disorders, such as Alzheimer's disease, schizophrenia, and Parkinson's disease-related dementia.

Large- and meso-scale cortical dynamics underlying evidence accumulation and decision-making Abstract The accumulation of sensory evidence is a crucial aspect of perceptual decision-making, and it involves complex neural computations requiring sensory processing, weighing of sensory evidence for or against a decision, as well as short-term memory of accrued evidence. Given this complexity, it is likely that many cortical areas are causally involved in evidence accumulation. However, little is known about which areas are necessary for evidence accumulation; moreover, we do not understand the neuronal circuit mechanisms underlying this important phenomenon. To answer these questions, we will use a novel virtual navigation-based visual accumulation task for head-fixed mice, and combine it with a suite of optical techniques to manipulate and record neural activity on meso- to large spatial scales, informed by detailed computational models of cognitive behavior. First, we propose to use optogenetic inactivation through the intact skull rendered optically transparent to systematically map cortical areas involved in accumulating visual evidence as the animal navigates the virtual maze. Taking advantage of this transparent skull preparation, we will also perform large-scale Ca2+ imaging from the entire dorsal portion of the cortex to map the large-scale spatiotemporal dynamics and flow of information during evidence accumulation and decision-making. Importantly, we will analyze the data using sophisticated quantitative models of evidence accumulation recently developed in the Brody laboratory, which will allow us to precisely quantify which aspect(s) of evidence accumulation depend on which cortical areas, and how dynamic patterns of neural activity map onto individual computations during decision-making. We thus aim use a unique combination of state-of-the-art techniques to provide a detailed and causal account of how cortical circuits underlie evidence accumulation for decision-making during spatial navigation. Beyond its importance for basic research, elucidating these processes will be crucial for understanding and treating the deficits in evidence accumulation and decision-making that are a hallmark of many psychiatric and neurological disorders such as obsessive-compulsive disorder, attention deficit hyperactivity disorder and drug abuse.

It has become increasingly clear that both spontaneous and trained behaviors engage activity throughout the cortex. However, at least in the case of perceptual decisions, task complexity critically modulates the underlying large- and mesoscale cortical dynamics. When decisions are simple sensorimotor mappings, cortical activity is correlated, and behavioral effects of inactivation are essentially restricted to the relevant sensory areas. Conversely, when decisions are complex and demanding, e.g. when accumulating evidence over seconds, cortical activity becomes decorrelated and behavioral effects of inactivation are widespread, indicating distributed processes. The present proposal seeks to understand two important problems related to these observations. In the postdoctoral (K99) stage, I will use a combination of virtual-reality based behavior, temporally-specific optogenetic inactivations from many cortical regions, simultaneous two-photon calcium imaging from groups of cortical areas with cellular resolution, and a dynamical model of large-scale cortical activity to understand how distributed processes support complex perceptual decisions, particularly evidence accumulation. In the independent research (R00) phase, I will seek to understand the mechanisms underlying task-induced changes in large-scale cortical dynamics. In particular, I will use task switching in virtual reality, large-scale calcium imaging at mesoscale or cellular resolution, pharmacogenetics, optogenetics and modeling, in isolation or combined, to test the hypothesis that the basal forebrain cholinergic system is part of the mechanism that induces large-scale decorrelations with increased task complexity. I thus aim to use a unique combination of state-of-the-art techniques to provide a detailed and causal account of how distributed cortical process underlie complex decision-making, and how task-specific cortical states are influenced by neuromodulation. Beyond its importance for basic research, elucidating these processes will be crucial for understanding and treating the deficits in decision-making that are a hallmark of many brain disorders such as obsessive-compulsive disorder, attention deficit hyperactivity disorder and drug abuse.