The mammalian hippocampal formation is crucial to the storage and consolidation of 'episodic'memories: memories for experiences that unfold in space and time. It is now clear that the hippocampus accomplishes its role in memory by generating a unique code reflecting the spatio-temporal context of experiences. This code provides a tag or 'index'that links together components of a given experience stored in a distributed form throughout the neocortex. The long-term goal of this program is to further our understanding of how this unique tag is generated. The generation of the hippocampal code is founded on internal mechanisms for keeping track of spatial location and for appending information about external events onto an internal spatial coordinate system or 'cognitive map'. Sets of neurons in hippocampus are selective for spatial location ('place cells'), neurons in related thalamic and midbrain structures are selective for head orientation in the horizontal plane ('head-direction cells'), and recently discovered cells in medial entorhinal cortex establish a spatial coordinate system ('grid cells'). These neurons are the core of a network that updates its spatial coordinate primarily on the basis of self-motion information ('path-integration'), and then creates and stores a code that is unique to current external and internal events at the current spatial coordinate. This capability appears to depend critically on the generation of precise spike timing relationships relative to the local EEG oscillations (theta rhythm). The project focuses on understanding the mechanisms underlying the grid-like firing patterns of entorhinal neurons and their spike timing dynamics, where the essential linear self-motion information that enables this system to perform path integration comes from, how neuronal firing characteristics of hippocampal cells are synthesized from entorhinal inputs, and how the entorhinal grid-cell network is wired up by a self- organizing process in early post-natal development. We also propose a specific experimental test that can, in principle, distinguish between two leading theoretical models for the mechanism of grid cells. Finally, we will test the hypothesis that a subregion of the hippocampus (the dentate gyrus) adds a temporal tag to episodic memories. In pursuing our long-term goal, we combine computer modeling and theoretical analysis with neurophysiological methods (largely pioneered by the PI) for recording the activity of large ensembles of single neurons in behaving animals.
Impairment of normal hippocampal function is strongly linked to memory impairments associated with normal and pathological aging, brain trauma and disease, developmental disorders and substance abuse. Thus, the neuro-physiological and neuro-computational principles underlying hippocampal function provide a framework for understanding memory processes, which are so important to quality of life, and disorders of which have such a major impact on our health care system. A well characterized and well understood rodent model of hippocampal function will also provide a platform for drug discovery and the development of genetic intervention for treatment of human memory disorder.
Showing the most recent 10 out of 60 publications
