Identifying and manipulating behavioral and neural correlates of contextual learning - Project Summary/Abstract The influence of context on behavior is a major topic in contemporary learning theory and neuroscience and the ability to transfer or generalize learning from one context to another is necessary for navigating the world. Deficits in generalization, or inappropriate generalization, are hallmarks of many disorders, including autism, schizophre- nia, and post-traumatic stress disorder. The hippocampus (HPC) and downstream regions, such as the lateral septum (LS), are important structures for learning, memory, and navigation. Damage to either region can disrupt the way in which context influences behavior. However, little is known about how the HPC and related structures encode tasks or information across varying contexts. This proposal leverages cutting edge approaches from theoretical, computational, systems, and behavioral neu- roscience, as well as learning theory, to address the involvement of the HPC in the representation of context and the way in which context interacts with learning. This proposal addresses three major open questions: 1) how HPC representations of discrete associated stimuli are affected by context manipulation, 2) how disruptions in HPC activity during contextual change affects learning generalization, and 3) how representations of behavioral responses are integrated with representations of context during REM sleep. To answer these questions, in Aim 1, I will use calcium imaging (CaImg) to investigate how neuronal firing in the HPC that occurs during a classically conditioned response is altered by shifts in spatial context. I will supplement traditional analytic techniques by analyzing population level cellular activity using low dimensional neuronal manifold analysis. In Aim 2, I will use optogenetics to disrupt periods of high HPC activity (sharp wave ripples) to determine how these periods con- tribute to task generalization. Finally, in Aim 3, I will use a combination of CaImg, electrophysiology, and opto- genetic manipulations to test the hypothesis that, during REM, HPC-derived spatial information is combined with movement information in the LS to inform task-relevant decisions across different contexts. These studies have the potential to lead to breakthroughs in the understanding of how cellular and behavioral correlates of learning are affected by and interact with context, and of the role of sharp wave ripples and REM sleep in influencing these interactions. My goal is to become the principal investigator of a research laboratory where I combine cutting-edge electro- physiology, imaging, and optogenetic techniques to answer questions about the role of circuit dynamics at the single cell and population level in tasks involving high level sensorimotor integration. The K99 period will include training on additional advanced mathematical, behavioral, and optogenetic techniques and prepare me to be a highly competitive candidate for tenure-track faculty positions.
