Complementary Coding of Space with Coupled Place Cells and Grid Cells

Tianhao Chu, Wentao Qiu, Zihao Jiang, Si Wu

Keywords: Place cells, Grid cells, Complementary Coding of Space, Coupled Attractor Networks

Abstract:

Spatial coding is a fundamental function of the brain. Place cells in the hippocampus (HPC) and grid cells in the medial entorhinal cortex (MEC) are two primary types of neurons accounting for spatial representation in the brain. These two types of neurons employ different spatial coding strategies and process environmental and motion cues, respectively.
In this work, we develop a computational model to elucidate how place and grid cells can complement each other to integrate information optimally and overcome their respective shortcomings. Specifically, we build a model with reciprocally coupled continuous attractor neural networks (CANNs), in which a CANN with location coordinate models the place cell ensemble in HPC, and multiple CANNs with phase coordinate model grid cell modules with different spacings in MEC, and the coupling between place and grid cells conveys the correlation prior between sensory cues. We theoretically derive that the dynamics of our model effectively implements the gradient-based optimization of the posterior. Using simulations, we demonstrate that our model achieves Bayesian optimal integration of the environmental and motion cues, and avoids the non-local error problem in phase coding of grid cells. We hope that this study gives us insights into understanding how place and grid cells complement each other to improve spatial representation in the brain.

Primary Area: applications to neuroscience & cognitive science

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 3418