How are long-term memories formed? One prominent theory proposes a two-stage process: memories are encoded in the hippocampus during active behavior and then consolidated across neocortical circuits during sleep. At the level of neuronal assemblies, the main evidence for this conjecture is that the hippocampus replays portions of its experience-specific awake activity during both REM and slow-wave sleep (SWS), even though the electrical and chemical profiles of these stages differ drastically. This raises several key questions: What differential roles, if any, do these two sleep stages play in memory consolidation? How are these roles supported by the neural activity patterns of SWS and REM? And how does plasticity contribute to shaping these patterns? We will employ a combination of large-scale electrophysiology in freely behaving rats, computational modeling, and pharmacological manipulations to test the hypothesis that REM and SWS differentially alter coordinated firing in the hippocampus in an experience-dependent manner. In particular, our preliminary data strongly suggest that REM increases coordinated firing in the hippocampus (Aim 1), while SWS has the opposite effect (Aim 2). Building on the observation that hippocampal cells exhibit place specific firing (place cells), we will use repeated linear track traversals to generate consistent activation of hippocampal patterns. By changing the environment and the number of traversals, we will create multiple experience-specific traces of parametrically varying strengths and measure their evolution over several SWS/REM sleep stages. Changes in coordinated firing play a critical role in controlling the ability of the hippocampus to drive its post-synaptic targets and engage plasticity mechanisms.
In Aim 3, we will characterize the role of synaptic plasticity in controlling the level of coordinated firing within the hippocampus by: (1) developing a computational framework for testing whether plasticity rules can account for changes in correlated firing produced by REM and SWS;(2) by repeating the experimental measurements of Aims 1,2 after pharmacological blockade of NMDA receptors;and (3) by using electrical stimulation to probe the mean synaptic weight of CA3 recurrent connections and track its evolution during sleep.
The proposed studies integrate experimental and computational approaches to quantify how synchrony alters hippocampal activity patterns during sleep, and how these changes depend on waking experience. Misregulation of sleep activity is observed in many psychiatric disorders, such as schizophrenia and depression. The proposed studies may provide a framework for understanding the origins and consequences of such misregulation. In addition, we investigate the interactions between burst firing and plasticity in recurrent networks. Abnormal interactions between these processes may underlie paroxysmal states in the hippocampus such as epileptic seizures.
| Wierzynski, Casimir M; Lubenov, Evgueniy V; Gu, Ming et al. (2009) State-dependent spike-timing relationships between hippocampal and prefrontal circuits during sleep. Neuron 61:587-96 |
