Of Mice and Machines: A Comparison of Learning Between Real World Mice and RL Agents

Shuo Han, German Espinosa, Junda Huang, Daniel Dombeck, Malcolm A. Maciver, Bradly C. Stadie
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:21685-21703, 2025.

Abstract

Recent advances in reinforcement learning (RL) have demonstrated impressive capabilities in complex decision-making tasks. This progress raises a natural question: how do these artificial systems compare to biological agents, which have been shaped by millions of years of evolution? To help answer this question, we undertake a comparative study of biological mice and RL agents in a predator-avoidance maze environment. Through this analysis, we identify a striking disparity: RL agents consistently demonstrate a lack of self-preservation instinct, readily risking “death” for marginal efficiency gains. These risk-taking strategies are in contrast to biological agents, which exhibit sophisticated risk-assessment and avoidance behaviors. Towards bridging this gap between the biological and artificial, we propose two novel mechanisms that encourage more naturalistic risk-avoidance behaviors in RL agents. Our approach leads to the emergence of naturalistic behaviors, including strategic environment assessment, cautious path planning, and predator avoidance patterns that closely mirror those observed in biological systems.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-han25a, title = {Of Mice and Machines: A Comparison of Learning Between Real World Mice and {RL} Agents}, author = {Han, Shuo and Espinosa, German and Huang, Junda and Dombeck, Daniel and Maciver, Malcolm A. and Stadie, Bradly C.}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {21685--21703}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/han25a/han25a.pdf}, url = {https://proceedings.mlr.press/v267/han25a.html}, abstract = {Recent advances in reinforcement learning (RL) have demonstrated impressive capabilities in complex decision-making tasks. This progress raises a natural question: how do these artificial systems compare to biological agents, which have been shaped by millions of years of evolution? To help answer this question, we undertake a comparative study of biological mice and RL agents in a predator-avoidance maze environment. Through this analysis, we identify a striking disparity: RL agents consistently demonstrate a lack of self-preservation instinct, readily risking “death” for marginal efficiency gains. These risk-taking strategies are in contrast to biological agents, which exhibit sophisticated risk-assessment and avoidance behaviors. Towards bridging this gap between the biological and artificial, we propose two novel mechanisms that encourage more naturalistic risk-avoidance behaviors in RL agents. Our approach leads to the emergence of naturalistic behaviors, including strategic environment assessment, cautious path planning, and predator avoidance patterns that closely mirror those observed in biological systems.} }
Endnote
%0 Conference Paper %T Of Mice and Machines: A Comparison of Learning Between Real World Mice and RL Agents %A Shuo Han %A German Espinosa %A Junda Huang %A Daniel Dombeck %A Malcolm A. Maciver %A Bradly C. Stadie %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-han25a %I PMLR %P 21685--21703 %U https://proceedings.mlr.press/v267/han25a.html %V 267 %X Recent advances in reinforcement learning (RL) have demonstrated impressive capabilities in complex decision-making tasks. This progress raises a natural question: how do these artificial systems compare to biological agents, which have been shaped by millions of years of evolution? To help answer this question, we undertake a comparative study of biological mice and RL agents in a predator-avoidance maze environment. Through this analysis, we identify a striking disparity: RL agents consistently demonstrate a lack of self-preservation instinct, readily risking “death” for marginal efficiency gains. These risk-taking strategies are in contrast to biological agents, which exhibit sophisticated risk-assessment and avoidance behaviors. Towards bridging this gap between the biological and artificial, we propose two novel mechanisms that encourage more naturalistic risk-avoidance behaviors in RL agents. Our approach leads to the emergence of naturalistic behaviors, including strategic environment assessment, cautious path planning, and predator avoidance patterns that closely mirror those observed in biological systems.
APA
Han, S., Espinosa, G., Huang, J., Dombeck, D., Maciver, M.A. & Stadie, B.C.. (2025). Of Mice and Machines: A Comparison of Learning Between Real World Mice and RL Agents. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:21685-21703 Available from https://proceedings.mlr.press/v267/han25a.html.

Related Material