{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T12:07:37Z","timestamp":1781870857172,"version":"3.54.5"},"reference-count":41,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,1,30]],"date-time":"2026-01-30T00:00:00Z","timestamp":1769731200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Large Language Models (LLMs) have come a long way in their ability to solve a wide range of problems. Yet, LLM decision-making still relies primarily on pattern recognition, which may limit its ability to make sound decisions under uncertainty. In contrast, human reasoning often makes use of explicit causal models, allowing humans to explain, hypothesize, and extrapolate to different domains in uncertain scenarios. In this article, we explore whether human causal models can be strategically integrated with Large Language Models to improve planning outcomes under uncertainty for object assembly and troubleshooting tasks modeled as Partially Observable Markov Decision Processes (POMDPs). Our contributions consist of two parts: (1) an interactive LLM agent that plans an action at each time step by solving a POMDP targeted at an object assembly or troubleshooting task, and (2) a novel hybrid-reasoning framework that uses confidence scores in both the LLM agent's output and a human causal model to make a final decision on the most appropriate action for the current time step to achieve the task. We demonstrate the efficacy of our approach through detailed simulations and show a significant improvement in task planning reward across three different state-of-the-art LLMs when augmenting the baseline LLM planner with a human causal model.<\/jats:p>","DOI":"10.3389\/frai.2025.1730614","type":"journal-article","created":{"date-parts":[[2026,1,30]],"date-time":"2026-01-30T06:42:48Z","timestamp":1769755368000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Augmenting large language models with psychologically grounded models of causal reasoning for planning under uncertainty"],"prefix":"10.3389","volume":"8","author":[{"given":"Semanti","family":"Basu","sequence":"first","affiliation":[{"name":"Computer Science, Brown University","place":["Providence, RI, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Moon Hwan","family":"Kim","sequence":"additional","affiliation":[{"name":"Computer Science, Brown University","place":["Providence, RI, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Semir","family":"Tatlidil","sequence":"additional","affiliation":[{"name":"Cognitive and Psychological Sciences, Brown University","place":["Providence, RI, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tom","family":"Williams","sequence":"additional","affiliation":[{"name":"Computer Science, Colorado School of Mines","place":["Golden, CO, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Steven","family":"Sloman","sequence":"additional","affiliation":[{"name":"Cognitive and Psychological Sciences, Brown University","place":["Providence, RI, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ruth Iris","family":"Bahar","sequence":"additional","affiliation":[{"name":"Computer Science, Brown University","place":["Providence, RI, United States"]},{"name":"Computer Science, Colorado School of Mines","place":["Golden, CO, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,1,30]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2204.0169","article-title":"Do as I can, not as I say: grounding language in robotic affordances","author":"Ahn","year":"2022","journal-title":"arXiv [preprint]"},{"key":"B2","author":"Basu","year":"2025"},{"key":"B3","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/ICRA55743.2025.11128040","article-title":"\u201cRobot planning under uncertainty for object assembly and troubleshooting using human causal models,\u201d","volume-title":"2025 IEEE International Conference on Robotics and Automation (ICRA)","author":"Basu","year":"2025"},{"key":"B4","doi-asserted-by":"publisher","first-page":"96640","DOI":"10.52202\/079017-3064","article-title":"Unveiling causal reasoning in large language models: Reality or mirage?","volume":"37","author":"Chi","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst"},{"key":"B5","doi-asserted-by":"crossref","DOI":"10.7551\/mitpress\/4638.001.0001","volume-title":"The Mind's Arrows: Bayes Nets and Graphical Causal Models in Psychology","author":"Glymour","year":"2001"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2501.12948","article-title":"Deepseek-r1: Incentivizing reasoning capability in LLMs via reinforcement learning","author":"Guo","year":"2025","journal-title":"arXiv"},{"key":"B7","article-title":"Inner monologue: Embodied reasoning through planning with language models","author":"Huang","year":"2022","journal-title":"arXiv"},{"key":"B8","unstructured":"\u201cCan large language models infer causation from correlation?\u201d\n          \n          \n            \n              Jin\n              Z.\n            \n            \n              Liu\n              J.\n            \n            \n              LYU\n              Z.\n            \n            \n              Poff\n              S.\n            \n            \n              Sachan\n              M.\n            \n            \n              Mihalcea\n              R.\n            \n          \n          Vienna\n          The Twelfth International Conference on Learning Representations (ICLR)\n          \n          2024"},{"key":"B9","doi-asserted-by":"publisher","author":"Kambhampati","year":"2025","DOI":"10.48550\/arXiv.2504.09762"},{"key":"B10","unstructured":"\u201cThe role of causal models in reasoning under uncertainty,\u201d\n          \n          \n            \n              Krynski\n              T. R.\n            \n            \n              Tenenbaum\n              J. B.\n            \n          \n          UC Merced\n          Proceedings of the Annual Meeting of the Cognitive Science Society. Vol. 25\n          \n          2003"},{"key":"B11","doi-asserted-by":"publisher","first-page":"430","DOI":"10.1037\/0096-3445.136.3.430","article-title":"The role of causality in judgment under uncertainty","volume":"136","author":"Krynski","year":"2007","journal-title":"J. Exp. Psychol.: General"},{"key":"B12","doi-asserted-by":"publisher","first-page":"e253","DOI":"10.1017\/S0140525X16001837","article-title":"Building machines that learn and think like people","volume":"40","author":"Lake","year":"2017","journal-title":"Behav. Brain Sci"},{"key":"B13","doi-asserted-by":"publisher","first-page":"105382","DOI":"10.1016\/j.cognition.2023.105382","article-title":"Drawing conclusions: Representing and evaluating competing explanations","volume":"234","author":"Liefgreen","year":"2023","journal-title":"Cognition"},{"key":"B14","unstructured":"\u201cLet's verify step by step,\u201d\n          \n          \n            \n              Lightman\n              H.\n            \n            \n              Kosaraju\n              V.\n            \n            \n              Burda\n              Y.\n            \n            \n              Edwards\n              H.\n            \n            \n              Baker\n              B.\n            \n            \n              Lee\n              T.\n            \n          \n          Vienna\n          The Twelfth International Conference on Learning Representations (ICLR)\n          \n          2024"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.891","article-title":"Explica: Evaluating explicit causal reasoning in large language models","author":"Miliani","year":"2025","journal-title":"arXiv"},{"key":"B16","unstructured":"Gpt-4o System Card"},{"key":"B17","unstructured":"Introducing SWE-Bench Verified."},{"key":"B18","unstructured":"Learning to Reason with LLMs"},{"key":"B19","unstructured":"OpenAI o3 and o4-Mini System Card"},{"key":"B20","unstructured":"OpenAI o3-Mini System Card"},{"key":"B21","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511803161","volume-title":"Causality: Models, Reasoning and Inference","author":"Pearl","year":"2009"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv:2501.14249","article-title":"Humanity's last exam","author":"Phan","year":"2025","journal-title":"arXiv [preprint]"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv:2311.12022","article-title":"GPQA: a graduate-level google-proof Q&A benchmark","author":"Rein","year":"2023","journal-title":"arXiv [preprint]"},{"key":"B24","unstructured":"\u201cRobots that ask for help: uncertainty alignment for large language model planners,\u201d\n          \n          \n            \n              Ren\n              A. Z.\n            \n            \n              Dixit\n              A.\n            \n            \n              Bodrova\n              A.\n            \n            \n              Singh\n              S.\n            \n            \n              Tu\n              S.\n            \n            \n              Brown\n              N.\n            \n          \n          Atlanta, GA\n          7th Annual Conference on Robot Learning (CoRL)\n          \n          2023"},{"key":"B25","doi-asserted-by":"publisher","author":"Saklad","year":"2025","DOI":"10.48550\/arXiv.2505.18931"},{"key":"B26","unstructured":"\u201cReflexion: language agents with verbal reinforcement learning,\u201d\n          \n          \n            \n              Shinn\n              N.\n            \n            \n              Cassano\n              F.\n            \n            \n              Gopinath\n              A.\n            \n            \n              Narasimhan\n              K. R.\n            \n            \n              Yao\n              S.\n            \n          \n          New Orleans, LA\n          Thirty-seventh Conference on Neural Information Processing Systems (NeurIPS)\n          \n          2023"},{"key":"B27","article-title":"\u201cMonte-carlo planning in large POMDPs,\u201d","author":"Silver","year":"2010","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B28","doi-asserted-by":"crossref","DOI":"10.1093\/acprof:oso\/9780195183115.001.0001","volume-title":"Causal Models: How People Think About the World and its Alternatives","author":"Sloman","year":"2005"},{"key":"B29","first-page":"853","volume-title":"Causal Bayes Nets as Psychological Theory","author":"Sloman","year":"2022"},{"key":"B30","unstructured":"\u201cScaling LLM test-time compute optimally can be more effective than scaling parameters for reasoning,\u201d\n          \n          \n            \n              Snell\n              C. V.\n            \n            \n              Lee\n              J.\n            \n            \n              Xu\n              K.\n            \n            \n              Kumar\n              A.\n            \n          \n          Singapore\n          The Thirteenth International Conference on Learning Representations (ICLR)\n          \n          2025"},{"key":"B31","doi-asserted-by":"crossref","first-page":"14054","DOI":"10.1109\/ICRA57147.2024.10610981","article-title":"\u201cInteractive planning using large language models for partially observable robotic tasks,\u201d","volume-title":"2024 IEEE International Conference on Robotics and Automation (ICRA)","author":"Sun","year":"2024"},{"key":"B32","doi-asserted-by":"publisher","first-page":"1544387","DOI":"10.3389\/fcogn.2025.1544387","article-title":"A comparison of methods to elicit causal structure","volume":"4","author":"Tatlidil","year":"2025","journal-title":"Front. Cognit"},{"key":"B33","doi-asserted-by":"crossref","DOI":"10.1093\/oxfordhb\/9780199399550.001.0001","volume-title":"The Oxford Handbook of Causal Reasoning","author":"Waldmann","year":"2017"},{"key":"B34","article-title":"\u201cChain-of-thought prompting elicits reasoning in large language models,\u201d","volume-title":"Proceedings of the 36th International Conference on Neural Information Processing Systems, NIPS '22","author":"Wei","year":"2022"},{"key":"B35","doi-asserted-by":"crossref","first-page":"1819","DOI":"10.18653\/v1\/2024.naacl-long.102","article-title":"\u201cReasoning or reciting? exploring the capabilities and limitations of language models through counterfactual tasks,\u201d","volume-title":"Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)","author":"Wu","year":"2024"},{"key":"B36","article-title":"\u201cTree of thoughts: Deliberate problem solving with large language models,\u201d","author":"Yao","year":"","journal-title":"Thirty-seventh Conference on Neural Information Processing Systems"},{"key":"B37","article-title":"\u201cReact: Synergizing reasoning and acting in language models,\u201d","author":"Yao","year":"","journal-title":"International Conference on Learning Representations (ICLR)"},{"key":"B38","author":"Yu","year":"2025","journal-title":"Causaleval: Towards Better Causal Reasoning in Language Models"},{"key":"B39","article-title":"\u201cReST-MCTS*: LLM self-training via process reward guided tree search,\u201d","author":"Zhang","year":"2024","journal-title":"The Thirty-eighth Annual Conference on Neural Information Processing Systems"},{"key":"B40","doi-asserted-by":"publisher","author":"Zhang","year":"2025","DOI":"10.48550\/arXiv.2503.24235"},{"key":"B41","article-title":"\u201cLarge language models as commonsense knowledge for large-scale task planning,\u201d","author":"Zhao","year":"2023","journal-title":"Thirty-seventh Conference on Neural Information Processing Systems"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1730614\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,30]],"date-time":"2026-01-30T06:42:53Z","timestamp":1769755373000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1730614\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,30]]},"references-count":41,"alternative-id":["10.3389\/frai.2025.1730614"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1730614","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,30]]},"article-number":"1730614"}}