Circuit dynamics supporting associative learning in the dentate gyrus

Project Summary The brain transforms experiences into patterns of activity that control emotions, decisions and behaviors. Discriminating these patterns of activity allow these experiences to be stored as distinct entities, separating important stimuli from unimportant ones, and catalogued into memory. Modern techniques in neuroscience such as large-scale recording, computational tools for analysis of these datasets and circuit-based manipulations provide an opportunity for developing a deeper understanding the mechanisms of learning and discrimination. These efforts are of profound importance to human health, as the inability to encode experiences appropriate precision is a hallmark of cognitive disorders associated with aging. One way that neural circuits may discriminate experiences is by, at the population level, separating the neural representations of these experiences with learning, allowing a readout area to better decode the stimulus from the patterns of activity. A locus of this computation is the hippocampus (HPC), which, with learning, encodes the relationships and distinctions between behaviorally relevant variables. We have recently found the dentate gyrus subregion of the hippocampus classifies cortical representations of olfactory stimuli, increasing the distance between odor representations with learning. In this proposal, we aim to understand the mechanism by which stimulus representations in the DG change with learning. In Aim 1, we will use viral, electrophysiological and imaging tools to map the cell-types and networks that generate odor representations in the DG. In Aim 2 we will determine the local circuit mechanisms that control the flexibility of odor representations with learning, with a focus on dopamine-dependent modulation of encoding dynamics in DG GCs. In Aim 3, we will determine how aging impacts the flexibility neural representations in the DG, and how circuit-based manipulations can reverse neural discrimination deficits in aged mice. Understanding the mechanisms that support learning-induced flexibility of neural ensembles will facilitate the development of therapeutics for the treatment of age-related cognitive decline.