: The encoding, storage, retrieval and consolidation of either episodic or other forms of memory involves temporally extended interactions of large neural populations distributed over multiple brain regions. We propose to conduct a neurophysiological analysis of memory-related neural ensemble interactions within a closely linked system of brain regions that are important in mnemonic processes. These regions include hippocampus, amygdala, medial prefrontal cortex, dorsal and ventral striatum and substantia nigra/ventral tegmentum. Our main focus is the process of spontaneous memory trace reactivation, operationally defined in neurophysiological terms as the reestablishment of neural activity patterns that match those that were imposed by recent experience. Retrieval of sequences of such previously encoded patterns can be understood as a neural reflection of episodic memory. Off-line retrieval has long been proposed to assist in the process of memory consolidation. Previously, we established several important basic characteristics of this process: it occurs coherently among hippocampal and some neocortical neural ensembles; it recreates, in a compressed form, the short-term temporal order of encoded experiences; it is strongest during a period of 30-60 minutes after an experience, but can be observed at least 24 hours later. Of particular importance for this proposal is our observation that hippocampal memory trace reactivation is expressed primarily during a specific neurophysiological event, the hippocampal sharp-wave. Sharp-waves have been proposed on theoretical grounds to reflect the convergence of hippocampal associative networks onto attractor (i.e., stored memory) states. This proposal addresses a number of key unanswered questions: Is reactivation of memory traces in neocortical and subcortical structures coordinated by the hippocampus? Is there a window of enhanced neocortical or subcortical plasticity during sharp-waves? Is hippocampal reactivation dependent on NMDA receptor activation during encoding? Is the reactivation process potentiated by positive or negative reinforcement and does it interact with the firing of dopamine neurons in a manner that may assist in learning routes to reward? Do localized drug applications (e.g., in the hippocampus or amygdala), which facilitate or impair memory consolidation, affect the reactivation process? Such analyses require large scale parallel recording methods, and we have developed technology that will enable simultaneous recording from about 150-200 single neurons distributed over multiple structures in unanesthetized, behaving rodents. These studies should clarify the neurophysiological mechanisms of memory retrieval and consolidation, and assist in understanding deficits in this process that occur under conditions such as normal and pathological aging, brain trauma, developmental disorders and substance abuse.
Showing the most recent 10 out of 45 publications
