Hippocampal phase precession may be generated by chimera dynamics

Maria MasoliverJoern DavidsenWilten Nicola^*

• University of Calgary, Calgary, Canada

The 8 Hz theta rhythm observed in hippocampal local field potentials of animals can be regarded as a "clock" that regulates the timing of spikes. While different interneuron sub-types synchronously phase lock to different phases for every theta cycle, the phase of pyramidal neurons' spikes asynchronously vary in each theta cycle, depending on the animal's position. On the other hand, pyramidal neurons tend to fire slightly faster than the theta oscillation in what is termed hippocampal phase precession. Chimera states are specific solutions to dynamical systems where synchrony and asynchrony coexist, similar to coexistence of phase precessing and phase locked cells during the hippocampal theta oscillation. Here, we test the hypothesis that the hippocampal phase precession emerges from chimera dynamics with computational modelling. We utilized multiple network topologies and sizes of Kuramoto oscillator networks that are known to collectively display chimera dynamics. We found that by changing the oscillators' intrinsic frequency, the frequency ratio between the synchronized and unsynchronized oscillators can match the frequency ratio between the hippocampal theta oscillation (≈8 Hz) and phase precessing pyramidal neurons (≈9 Hz). The faster firing population of oscillators also displays theta-sequence-like behaviour and phase precession. Finally, we trained networks of spiking integrate-and-fire neurons to output a chimera state by using the Kuramoto-chimera system as a dynamical supervisor. We found that the firing times of subsets of individual neurons display phase precession. Significance Statement Cells in the hippocampus fire spikes that either synchronize to different phases of the 8 Hz theta oscillation, or are in an asynchronous state of phase-precession, where a cell fires slightly faster than the theta oscillation. Our work shows that this co-existence of synchrony and asynchrony is well modeled by a chimera state, where synchrony and asynchrony co-exist. This was verified in computational models of chimera dynamics with Kuramoto oscillators, and in spiking neural network simulations with embedded chimera dynamics. To our knowledge, this is the first work to put forward and test the hypothesis that the hippocampal phase precession is mechanistically a chimera state.