Consolidation and interference of hippocampal spatial maps - Project Summary Neurons throughout the brain remain highly active during sleep and produce complex patterns of action potentials. In unit recordings from the hippocampus, by comparing these patterns against earlier behavioral templates, numerous groups have shown that during hippocampal sharp-wave ripple (SWR) events these sleep patterns can correspond to replays of earlier experience. Additional research has further indicated that these SWR patterns play an important role in long term retention of memory. Developing and using advanced machine learning techniques and large-scale population recordings, we have recently demonstrated two remarkable phenomena during SWRs: that individual SWR-associated replays correspond to memories of specific individual experiences within a context, and that the fine scale representations of individual neurons can be detected and dynamically tracked across SWRs. Including our own studies, the study of SWRs has largely focused on the relationship between sleep and experiences within individual contexts. However, numerous experiments report that sleep and the hippocampus are also critically involved in isolating memories of experiences in distinct contexts, a process which fails in the case of interference. In this proposal, we describe a series of experiments which will allow us to test the hypothesis that hippocampal ensemble activity during sleep SWR can either support the stabilization of the memories of experiences within different contexts or mechanistically underly the failures of context-specific memory in interference. In Aim 1, we will further develop our machine learning techniques for studying the evolution of hippocampal codes during sleep SWRs – providing new approaches to quantifying the tuning and retuning of single neurons following specific experiences and measuring the interactions of context-specific manifolds within SWRs. In Aim 2, we will carry out the key experiments in different behavioral paradigms designed to optimize for ensemble manifold detection, isolation of individual behavioral experiments, and context-specific behavior. Finally, in Aim 3, we will implement closed-loop signal processing to test the necessity of sleep SWRs for the expression of context- specific memories. Collectively, our proposed work will provide a critical foundation to understanding how hippocampal activity during sleep underlies our properly functioning memories.
