During wakefulness, hippocampal networks generate spiking sequences that encode sensory cues and internally tile the gaps in space or time between them, forming memory maps of temporally related experiences. During sleep, the same networks generate sharp wave ripples (SWRs), i.e. synchronous high frequency oscillations that harbor fast-scale replays of those sequences, involved in memory consolidation. What population dynamics and excitation-inhibition interactions govern the generation of awake sequences or their SWR replays? I will first present a network model of hippocampal areas CA3-CA1 that accurately captures SWR features and reveals that strong, fast-decaying, recurrent inhibition is crucial for generating these oscillations. This network coupled with a model of cortical slow oscillations (i.e. <1 Hz transitions between activated UP and silent DOWN states during sleep) demonstrates that UP states regulate SWR generation, through the excitation-inhibition balance they induce via multiple synaptic pathways. Can these reactivations be detected extracellularly? Extending the CA3-CA1 network into a biophysically realistic model of local field potentials (LFP) during SWRs, combined with silicon probe recordings in rats demonstrate that the spiking patterns of CA1 spiking ensembles are reflected in spatiotemporal patterns of multi-site LFP. Finally, through recent in vivo two-photon calcium imaging on CA1 of head-fixed mice, I will show that CA1 networks combine stable and flexible representations with different learning-related dynamics which may be a crucial feature for the hippocampus to construct maps of fixed external cues as well as their changing temporal relationships.
Title: Excitation-inhibition interactions govern the dynamics of spiking sequences and their replays: Combining modeling, electrophysiology and calcium imaging.Abstract: During wakefulness, hippocampal networks generate spiking sequences that encode sensory cues and internally tile... Read More →
