Overview
The hippocampus is a limbic structure that plays a critical role in the formation, consolidation and recall of various forms of memory, including episodic and spatial memory. How do such complex cognitive functions emerge from the activity of hippocampal neurons? One key property of the hippocampal network that has gained tremendous interest in recent years is its ability to encode and reactivate sequential activity patterns: specific subgroups of neurons become successively active, either in response to the ongoing behavioral and cognitive context, or spontaneously during recall, decision making and sleep. In the spatial domain, fast sequential activation of hippocampal 'place cells' anticipates future trajectories, associates traveled paths with rewarding or aversive stimuli, and appears to underlie certain forms of spatial learning. During subsequent sleep, reactivation of the same sequences plays a critical role in memory consolidation via a hippocampo-cortical dialogue. This versatile capacity to manipulate and memorize abstract information in the form of neural sequences may provide a unifying conceptual framework to reconcile the numerous functions of the hippocampus.
The goal of our research is to decipher the various roles of hippocampal dynamics, including their interactions with cortical and subcortical activity (in e.g. prefrontal cortex, medial and lateral entorhinal cortices, ventral and dorsal striatum, locus cœruleus, ventral tegmental area, etc.), and dissect their network mechanisms in freely moving rodents. We combine several cutting-edge technologies, including massively parallel electrophysiology across multiple brain areas, an innovative optical imaging approach, and optogenetics. We perform state-of-the-art electrophysiological recordings and are developing optical imaging and targeting approaches for real-time optical excitation or inhibition of single, identified neurons, as well as long-term monitoring of vast neural populations - all in freely moving rodents performing complex behavioral tasks in large scale mazes and sleeping. Advanced analytical and statistical methods allow us to unravel the mechanisms and roles of hippocampal sequences in the formation, consolidation and recall of memory.
Our team includes researchers with backgrounds in neurophysiology, animal behavior, physics and engineering.
Formation and Consolidation of Episodic-Like Memory Traces
The 'two-stage' theory of memory posits that memory consolidation involves a dialogue during sleep between the hippocampus, where memory traces are initially formed, and the neocortex, where they are stored for long term retention. A prominent target is the medial prefrontal cortex, which over days becomes progressively involved in memory recall, concomitantly with a gradual hippocampal disengagement. During sleep, task-related neural activity patterns are replayed in both structures, orchestrated by various brain oscillations related to memory consolidation, that are often observed in temporal proximity.
We have shown that fast sequential activation of hippocampal assemblies at the theta time scale ('theta sequences') are crucial to encode new memories, which are later replayed during sleep for memory consolidation ('sleep replay'). We further demonstrated that the formation of these sequences requires fine timescale coordination between hippocampal cell assemblies. We have also shown that subsequent replay during sleep ripples is instrumental for memory consolidation, a long-standing hypothesis that had never received experimental confirmation. Our subsequent work established that hippocampal replay underlies a hippocampo-cortical dialogue, involving enhanced coupling between ripples, cortical delta waves and spindles. More recently, we have shown that, contrary to a generally accepted tenet in systems neuroscience, an ever-changing minority of cortical neurons remain active during delta waves, forming assemblies between neurons involved in coding memories. Delta waves thus shut most of the cortical network off to isolate critical computations involved in memory consolidation.
Navigation and Spatial Memory
Our focus is on cognitive functions such as spatial navigation learning, memory and decision-making and how they are linked with neuro-electrical activity, ensembles of neurons and neural networks. With chronically implanted multiple electrodes, we record from rats as they perform tasks requiring specific types of cognitive processing. Much of the work is centered on a popular experimental model for abstract representations in the brain and the place and head direction cells in the hippocampal system and related areas. We have studied how self-movement cues are engaged for this activity and also how brain areas downstream from the hippocampus (ventral striatum, prefrontal cortex) exploit this for navigation, orienting behavior and spatial memory. The role of sleep and associated brain oscillations in off-line memory consolidation is another focus. Furthermore coherence between oscillatory local field potentials in multiple structures is studied as a potential mechanism of selection of active pathways within the massively interconnected brain. This work is carried out in collaboration with roboticians and computational modelers to facilitate creation of bio-inspired automatons.
