A multi-region brain model to elucidate the role of hippocampus in spatially embedded decision tasks

26 Sept 2024 (modified: 05 Feb 2025)Submitted to ICLR 2025EveryoneRevisionsBibTeXCC BY 4.0
Keywords: place cell, grid cell, cognitive map, multi-region interactions, decision making, neuroscience
TL;DR: We propose a multi-region brain model revealing conjunctive tuning in grid cells and sensory inputs as crucial for spatial decision-making and hippocampal cognitive maps, offering insights into the interactions underlying cognitive functions.
Abstract: We present a multi-region brain model exploring the role of structured memory circuits in spatially embedded decision-making tasks. We simulate decision-making processes that involve the cognitive maps formed within the CA1 region of the hippocampus during an evidence integration task, which animals learn through reinforcement learning (RL). Our model integrates a bipartite memory scaffold architecture that incorporates grid and place cells of the entorhinal cortex and hippocampus, with an action-selecting recurrent neural network (RNN) that integrates hippocampal representations. Through RL-based simulations, we demonstrate that joint encoding of position and evidence within medial entorhinal cortex, along with sensory projection to hippocampus, replicates experimentally observed place cell representations and promotes rapid learning and efficient spatial navigation relative to alternative circuits. Our findings predict conjunctive spatial and evidence tuning in grid cells, in addition to hippocampus, as essential for decision-making in space.
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: 8300
Loading

OpenReview is a long-term project to advance science through improved peer review with legal nonprofit status. We gratefully acknowledge the support of the OpenReview Sponsors. © 2025 OpenReview