A statistical note on the detection of autoassociative dynamics in hippocampal place-cell sequences

Abstract

The sequential activity of hippocampal place cells observed during sleep and awake resting is widely viewed as a neural correlate of memory processes. While recent work has advanced our understanding of the content represented during rest-related place-cell sequences, the nature of hippocampal population dynamics during sequential activity remains poorly understood. A recent experimental study has reported that place-cell sequences show a pattern of step-like movement, reminiscent of transitions between discrete attractor states (Pfeiffer and Foster, 2015). By contrast, previous theoretical models predict that the spatiotemporal structure of place-cell sequences should reflect the distribution of place fields, typically observed to be spatially smooth. Motivated by this discrepancy between models and experimental data, we performed a quantitative comparison between these results and the spike trains generated by a network model for the generation of place-cell sequences (Gönner et al., 2017). Although the model is based on continuous attractor network dynamics, we observed that the movement of sequential place representations was phase-locked to the population oscillation, highly similar to the experimental data interpreted as evidence for discrete attractor dynamics. To resolve this potential contradiction, we performed a detailed analysis of the methodology used to identify discrete attractor dynamics. Our results show that a previous 1 approach to step size decoding is prone to a decoding artefact. We propose a modified approach to estimate step sizes which may help to characterize the underlying circuit dynamics in vivo.