A Biologically Interpretable Cognitive Architecture for Online Structuring of Episodic Memories into Cognitive Maps

Evgenii Dzhivelikian, Nikita Bainaev-Mangilev, Aleksandr Panov
Proceedings of the First Workshop on NeuroAI Multimodal Intelligence @ AAAI 2026, PMLR 308:67-75, 2026.

Abstract

Cognitive maps provide a powerful framework for understanding spatial and abstract reasoning in biological and artificial agents. While recent computational models link cognitive maps to hippocampal-entorhinal mechanisms, they often rely on global optimization rules (e.g., backpropagation) that lack biological plausibility. In this work, we propose a novel cognitive architecture for structuring episodic memories into cognitive maps compatible with neural substrate constraints. Our model integrates the Successor Features framework with episodic memories, enabling incremental, online learning through agent-environment interaction. We demonstrate its efficacy in a partially observable gridworld, where the architecture autonomously organizes memories into structured representations without centralized optimization. This work bridges computational neuroscience and AI, offering a biologically grounded approach to cognitive map formation in artificial adaptive agents.

Cite this Paper

BibTeX
@InProceedings{pmlr-v308-dzhivelikian26a, title = {A Biologically Interpretable Cognitive Architecture for Online Structuring of Episodic Memories into Cognitive Maps}, author = {Dzhivelikian, Evgenii and Bainaev-Mangilev, Nikita and Panov, Aleksandr}, booktitle = {Proceedings of the First Workshop on NeuroAI Multimodal Intelligence @ AAAI 2026}, pages = {67--75}, year = {2026}, editor = {Abbasi-Asl, Reza and Iqbal, Asim and Ito, Shinya and Arkhipov, Anton and Sanborn, Sophia}, volume = {308}, series = {Proceedings of Machine Learning Research}, month = {27 Jan}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v308/main/assets/dzhivelikian26a/dzhivelikian26a.pdf}, url = {https://proceedings.mlr.press/v308/dzhivelikian26a.html}, abstract = {Cognitive maps provide a powerful framework for understanding spatial and abstract reasoning in biological and artificial agents. While recent computational models link cognitive maps to hippocampal-entorhinal mechanisms, they often rely on global optimization rules (e.g., backpropagation) that lack biological plausibility. In this work, we propose a novel cognitive architecture for structuring episodic memories into cognitive maps compatible with neural substrate constraints. Our model integrates the Successor Features framework with episodic memories, enabling incremental, online learning through agent-environment interaction. We demonstrate its efficacy in a partially observable gridworld, where the architecture autonomously organizes memories into structured representations without centralized optimization. This work bridges computational neuroscience and AI, offering a biologically grounded approach to cognitive map formation in artificial adaptive agents.} }
Endnote
%0 Conference Paper %T A Biologically Interpretable Cognitive Architecture for Online Structuring of Episodic Memories into Cognitive Maps %A Evgenii Dzhivelikian %A Nikita Bainaev-Mangilev %A Aleksandr Panov %B Proceedings of the First Workshop on NeuroAI Multimodal Intelligence @ AAAI 2026 %C Proceedings of Machine Learning Research %D 2026 %E Reza Abbasi-Asl %E Asim Iqbal %E Shinya Ito %E Anton Arkhipov %E Sophia Sanborn %F pmlr-v308-dzhivelikian26a %I PMLR %P 67--75 %U https://proceedings.mlr.press/v308/dzhivelikian26a.html %V 308 %X Cognitive maps provide a powerful framework for understanding spatial and abstract reasoning in biological and artificial agents. While recent computational models link cognitive maps to hippocampal-entorhinal mechanisms, they often rely on global optimization rules (e.g., backpropagation) that lack biological plausibility. In this work, we propose a novel cognitive architecture for structuring episodic memories into cognitive maps compatible with neural substrate constraints. Our model integrates the Successor Features framework with episodic memories, enabling incremental, online learning through agent-environment interaction. We demonstrate its efficacy in a partially observable gridworld, where the architecture autonomously organizes memories into structured representations without centralized optimization. This work bridges computational neuroscience and AI, offering a biologically grounded approach to cognitive map formation in artificial adaptive agents.
APA
Dzhivelikian, E., Bainaev-Mangilev, N. & Panov, A.. (2026). A Biologically Interpretable Cognitive Architecture for Online Structuring of Episodic Memories into Cognitive Maps. Proceedings of the First Workshop on NeuroAI Multimodal Intelligence @ AAAI 2026, in Proceedings of Machine Learning Research 308:67-75 Available from https://proceedings.mlr.press/v308/dzhivelikian26a.html.

Related Material