The Structure of Hippocampal Activity During REM Sleep

Author: Hoenselaar, Andreas

Year: 2015

Degree: Dissertation (Ph.D.)

Advisor: Siapas, Athanassios G.

Committee Members: Perona, Pietro; Elowitz, Michael B.; Roukes, Michael Lee; Dickinson, Michael H.; Siapas, Athanassios G.

Option: Computation and Neural Systems

DOI: 10.7907/Z98W3B8H

Abstract

The hippocampus is a brain structure critical for the formation of long-term episodic memories. The current predominant theory is that memories are gradually established across neocortical networks under the influence of hippocampal activity. This process of memory consolidation is conjectured to occur during sleep, which is characterized by two different modes of activation: slow-wave sleep (SWS) and rapid eye movement (REM) sleep. The functional roles of these two different sleep states remain unknown. Paradoxically, REM sleep exhibits the main features of awake activity, and is the stage of sleep when most dreams occur. Despite decades of study, the organization and function of REM sleep activity remains poorly understood. The goal of this thesis is to achieve a deeper quantitative understanding of the patterns of firing in area CA1 of the hippocampus during REM sleep using chronic multi-tetrode recordings from freely behaving and naturally sleeping rats. Our analysis shows that CA1 neurons significantly elevate their firing rate for periods that are short in relation to the duration of the REM sleep episode. Furthermore, for the majority of neurons, there is exactly one such burst per REM episode. This leads to lower overall firing rates and sparser population activity in CA1 compared to SWS. The time of onset of these bursts defines a natural order of firing across the population of recorded neurons within each REM episode. We demonstrate that this order does not repeat across REM episodes. Our results suggest that CA1 neurons are activated in random sequences across REM episodes, resulting in sparse patterns with only a small fraction of neurons active at any given time.

Files