Experience replay facilitates the formation of the hexagonal pattern of grid cells

Bo Zhang,Jia Liu
DOI: https://doi.org/10.1101/2023.02.19.529130
2023-01-01
Abstract:Abstract Reverse neuronal replay, characterized by the sequential reactivation of hippocampal place cells to represent past experiences, is proposed to organize learned knowledge into a cognitive map. Here we utilized the continuous attractor model to simulate the formation of multi-scale hexagonal patterns of entorhinal grid cells, which serve as the metric for the cognitive map, during experience replay of hippocampal place cells. We found a significant increase in grid periodicity with reverse replay compared to that without replay. This effect was exclusive to reverse replay, as neither forward nor shuffled-experience replay produced a similar enhancement. Further analysis revealed that the effect of the replay was modulated by the interaction between grid scales and the length of experience replay. That is, the replay was particularly effective for small grid scales, which helped in maintaining the excitation of formed grid bumps during pattern translation. In summary, our study illustrates a causative link between experience replay and the rapid formation of multi-scale hexagonal patterns from computational perspective, providing insights into how the experience replay by hippocampal place cells influences the formation of hexagonal patterns of entorhinal grid cells.
What problem does this paper attempt to address?