Segmentation and integration of experience in hippocampal neuronal representations

An episodic memory is a set of moments recalled together; these moments often share a unique combination of time, location, and content. Despite our experience with memories as falling into these segmented categories, it is not well-understood how the brain categorizes moments into episodes. It is theorized that the

hippocampus responds to event boundaries in the environment – specific stimuli, changes in location or time, that trigger movement into the next represented episode. In humans, room boundaries and task changes are two candidate event boundaries that seem to separate distinct episodes, demonstrated by decreased memory

performance across these boundaries.1 Despite the importance of episodic memory to our everyday experiences, the neural mechanisms underlying attention to event boundaries is not well understood due to resolution limitations of fMRI. Rodents provide a window into these potential neural mechanisms of event

boundaries through large scale in vivo electrophysiology recordings of single neurons and local field potentials (LFP) in the hippocampus. The hippocampus is known to play a role in episodic memory consolidation and retrieval, as well as in planning and learning. Planning and episodic memory retrieval appear to have similar

mechanisms displayed through replay of place cell sequences representing the track both in moments preceding and following a run during sharp-wave ripples (SWRs), transient bursts of activity in the 150-250 Hz LFP band.2 Place cell sequences matching possibilities of the rat's upcoming trajectory also often occur during

theta (5-12Hz) LFP oscillations,3 a phenomenon termed “theta sequences.” Assessing how the brain represents upcoming runs during theta and SWRs, alongside determining how it replays past runs during SWRs, together brings us closer to understanding how the brain forms specific episodes. Importantly, reward

locations and environmental changes have been shown to modulate hippocampal place cells and LFPs. We and others4-5 theorize that these landmarks and content shifts in rodent tasks act in a similar way to event boundaries in human episodic memory. The proposed study introduces a novel experimental design to test two

candidate event boundaries in the rat, reward locations and task rule transitions paired to separately colored sides of the maze. Aim 1 will focus on discerning how hippocampal place cells and theta sequences represent space between pairs of reward locations and at environmental boundary transitions, and specifically how these

representations change upon introduction of new and conflicting reward locations. Aim 2 will focus on awake SWR replays, also focusing on how segments between reward locations are represented and how these representations are together integrated with newly introduced reward locations. Completion of this proposed

project will provide important insight into the neural mechanisms underlying episodic memory. This project will provide rigorous practice in experimental and computational methods and will take place in Kamran Diba's Neural Circuits and Memory lab as part of the Neuroscience Graduate Program at the University of Michigan.