{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T10:54:20Z","timestamp":1791543260092,"version":"4.3.4"},"reference-count":153,"publisher":"Wiley","issue":"56","license":[{"start":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T00:00:00Z","timestamp":1789344000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T00:00:00Z","timestamp":1789344000000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["advanced.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Advanced Science"],"published-print":{"date-parts":[[2026,10]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>Artificial intelligence systems are transforming scientific discovery by accelerating specific research tasks, from protein structure prediction to materials design, yet remain confined to narrow domains requiring substantial human oversight. Exponential growth of scientific literature and increasing domain specialization constrain researchers' capacity to synthesize knowledge across disciplines and develop unifying theories, motivating exploration of general\u2010purpose AI systems for science. Here we show that a domain\u2010agnostic, agentic AI Scientist system can independently navigate the scientific workflow \u2010from hypothesis generation through data collection to manuscript preparation. The system autonomously designed and executed three psychological studies on visual working memory, mental rotation, and imagery vividness, executed online data collection with 288 participants, developed analysis pipelines through 8h+ continuous coding sessions, and produced completed manuscripts. The results demonstrate the capability of AI scientific discovery pipelines to conduct research with theoretical reasoning and methodological rigor comparable to experienced researchers, though with limitations in conceptual nuance and theoretical interpretation. This is a step toward embodied AI that can test hypotheses through real\u2010world experiments, accelerating discovery by autonomously exploring regions of scientific space that human cognitive and resource constraints might otherwise leave unexplored. It raises important questions about the nature of scientific understanding and attribution of scientific credit.<\/jats:p>","DOI":"10.1002\/advs.76675","type":"journal-article","created":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T14:00:33Z","timestamp":1789394433000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Closing the Empirical Loop: Autonomous AI Agents Conduct End\u2010to\u2010end Research With Human Participants"],"prefix":"10.1002","volume":"13","author":[{"given":"Gabrielle","family":"Wehr","sequence":"first","affiliation":[{"name":"Explore Science  Brisbane Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8416-005X","authenticated-orcid":false,"given":"Reuben","family":"Rideaux","sequence":"additional","affiliation":[{"name":"Explore Science  Brisbane Australia"},{"name":"School of Psychology The University of Sydney  Sydney Australia"},{"name":"Queensland Brain Institute The University of Queensland  St Lucia Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2319-0743","authenticated-orcid":false,"given":"Amaya J.","family":"Fox","sequence":"additional","affiliation":[{"name":"School of Psychology The University of Queensland  St Lucia Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David R.","family":"Lightfoot","sequence":"additional","affiliation":[{"name":"Explore Science  Brisbane Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0649-2566","authenticated-orcid":false,"given":"Jason","family":"Tangen","sequence":"additional","affiliation":[{"name":"School of Psychology The University of Queensland  St Lucia Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jason B.","family":"Mattingley","sequence":"additional","affiliation":[{"name":"Queensland Brain Institute The University of Queensland  St Lucia Australia"},{"name":"School of Psychology The University of Queensland  St Lucia Australia"},{"name":"Canadian Institute of Advanced Research (CIFAR)  Toronto Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4944-3673","authenticated-orcid":false,"given":"Shane E.","family":"Ehrhardt","sequence":"additional","affiliation":[{"name":"Explore Science  Brisbane Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,9,14]]},"reference":[{"key":"e_1_2_9_2_1","volume-title":"The Beginning of Infinity: Explanations That Transform the World","author":"Deutsch D.","year":"2011"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1063\/1.3050617"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.7312\/badi16016"},{"key":"e_1_2_9_5_1","volume-title":"Posterior Analytics; Topica","author":"Aristotle","year":"1960"},{"key":"e_1_2_9_6_1","volume-title":"The Fabric of Reality: The Science of Parallel Universes\u2013 and Its Implications","author":"Deutsch D.","year":"1997"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1063\/1.3060577"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1002\/asi.23329"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1038\/nj7612-457a"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1162\/qss_a_00327"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1087\/20150203"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1629\/2048-7754.25.2.130"},{"key":"e_1_2_9_13_1","volume-title":"The Structure of Scientific Revolutions","author":"Kuhn T. S.","year":"1962"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1175\/1520-0469(1963)020<0130:DNF>2.0.CO;2"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.1515\/9781400882618-003"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/B978-0-934613-03-3.50026-X"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.ads0018"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-021-03819-2"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06735-9"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41524-022-00765-z"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06734-w"},{"key":"e_1_2_9_22_1","first-page":"251","article-title":"On Automated Scientific Theory Formation: A Case Study Using the AM Program","volume":"9","author":"Lenat D. B.","year":"1977","journal-title":"Machine intelligence"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06924-6"},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-022-05172-4"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-021-03828-1"},{"key":"e_1_2_9_26_1","unstructured":"A.Vaswani N.Shazeer N.Parmar et\u00a0al. \u201cAttention Is All You Need \u201darXiv(2017): 1706.03762 https:\/\/doi.org\/10.48550\/ARXIV.1706.03762."},{"key":"e_1_2_9_27_1","unstructured":"OpenAI J.Achiam S.Adler et\u00a0al. \u201cGpt\u20104 Technical Report \u201darXiv(2023): 2303.08774 https:\/\/doi.org\/10.48550\/ARXIV.2303.08774."},{"key":"e_1_2_9_28_1","unstructured":"OpenAI. OpenAI O3 and O4\u2010Mini System Card https:\/\/cdn.openai.com\/pdf\/2221c875\u201002dc\u20104789\u2010800b\u2010e7758f3722c1\/o3\u2010and\u2010o4\u2010mini\u2010system\u2010card.pdf(2025)."},{"key":"e_1_2_9_29_1","unstructured":"DeepSeek\u2010AI D.Guo D.Yang et\u00a0al. \u201cDeepSeek\u2010R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning \u201darXiv(2025): 2501.12948 https:\/\/doi.org\/10.48550\/ARXIV.2501.12948."},{"key":"e_1_2_9_30_1","unstructured":"Anthropic. System Card: Claude Opus 4 & Claude Sonnet 4 https:\/\/www\u2010cdn.anthropic.com\/4263b940cabb546aa0e3283f35b686f4f3b2ff47.pdf(2025)."},{"key":"e_1_2_9_31_1","unstructured":"Google. Gemini 2.5 Pro Model Card https:\/\/storage.googleapis.com\/model\u2010cards\/documents\/gemini\u20102.5\u2010pro.pdf(2025)."},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11704-024-40231-1"},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2025.113280"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1002\/adma.202413523"},{"key":"e_1_2_9_35_1","unstructured":"J.Gottweis W.\u2010H.Weng A.Daryin et\u00a0al. \u201cAccelerating Scientific Discovery with Co\u2010Scientist \u201darXiv(2025): 2502.18864 https:\/\/doi.org\/10.48550\/ARXIV.2502.18864."},{"key":"e_1_2_9_36_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-2442-2"},{"key":"e_1_2_9_37_1","unstructured":"A.GhafarollahiandM. J.Buehler \u201cAtomAgents: Alloy Design and Discovery through Physics\u2010aware Multi\u2010Modal Multi\u2010Agent Artificial Intelligence \u201darXiv(2024): 2407.10022 https:\/\/doi.org\/10.48550\/ARXIV.2407.10022."},{"key":"e_1_2_9_38_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-024-00832-8"},{"key":"e_1_2_9_39_1","unstructured":"C.Lu C.Lu R. T.Lange et\u00a0al. \u201cThe AI Scientist: towards Fully Automated Open\u2010Ended Scientific Discovery \u201darXiv(2024): 2408.06292 https:\/\/doi.org\/10.48550\/ARXIV.2408.06292."},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ipm.2025.104225"},{"key":"e_1_2_9_41_1","unstructured":"Y.Yamada R. T.Lange C.Lu et\u00a0al. \u201cThe AI Scientist\u2010v2: Workshop\u2010Level Automated Scientific Discovery via Agentic Tree Search \u201darXiv(2025) 2504.08066 https:\/\/doi.org\/10.48550\/ARXIV.2504.08066."},{"key":"e_1_2_9_42_1","unstructured":"Intology. Zochi Technical Report. Zochi Technical Reporthttps:\/\/www.intology.ai\/blog\/zochi\u2010tech\u2010report(2025)."},{"key":"e_1_2_9_43_1","unstructured":"S.Schmidgall Y.Su Z.Wang et\u00a0al. \u201cAgent Laboratory: Using LLM Agents as Research Assistants \u201darXiv(2025): 2501.04227 https:\/\/doi.org\/10.48550\/ARXIV.2501.04227."},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586\u2010025\u201009442\u20109"},{"key":"e_1_2_9_45_1","unstructured":"Y.Weng M.Zhu G.Bao et\u00a0al. \u201cCycleResearcher: Improving Automated Research via Automated Review \u201darXiv(2024): 2411.00816 https:\/\/doi.org\/10.48550\/ARXIV.2411.00816."},{"key":"e_1_2_9_46_1","doi-asserted-by":"publisher","DOI":"10.1056\/AIoa2400555"},{"key":"e_1_2_9_47_1","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511803161"},{"key":"e_1_2_9_48_1","doi-asserted-by":"publisher","DOI":"10.1016\/0004-3702(91)90053-M"},{"key":"e_1_2_9_49_1","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/1552.001.0001"},{"key":"e_1_2_9_50_1","volume-title":"How the Body Shapes the Way We Think: A New View of Intelligence","author":"Pfeifer R.","year":"2007"},{"key":"e_1_2_9_51_1","unstructured":"S.Pagel M.Jirasek andL.Cronin \u201cValidation of the Scientific Literature via Chemputation Augmented by Large Language Models \u201darXiv(2024): 2410.06384 https:\/\/doi.org\/10.48550\/ARXIV.2410.06384."},{"key":"e_1_2_9_52_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06792-0"},{"key":"e_1_2_9_53_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature06860"},{"key":"e_1_2_9_54_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.171.3972.701"},{"key":"e_1_2_9_55_1","first-page":"153","article-title":"New Directions for Mental Imagery Research","volume":"19","author":"Marks D. F.","year":"1995","journal-title":"Journal of Mental Imagery"},{"key":"e_1_2_9_56_1","doi-asserted-by":"publisher","DOI":"10.3758\/s13428-017-0935-1"},{"key":"e_1_2_9_57_1","doi-asserted-by":"publisher","DOI":"10.5195\/jmla.2017.88"},{"key":"e_1_2_9_58_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.matt.2024.06.003"},{"key":"e_1_2_9_59_1","doi-asserted-by":"publisher","DOI":"10.1039\/D3DD00223C"},{"key":"e_1_2_9_60_1","unstructured":"W.Zhang L.Zeng Y.Xiao et\u00a0al. \u201cAgentOrchestra: A Hierarchical Multi\u2010Agent Framework for General\u2010Purpose Task Solving \u201darXiv(2025): 2506.12508 https:\/\/doi.org\/10.48550\/ARXIV.2506.12508."},{"key":"e_1_2_9_61_1","unstructured":"A.Fourney G.Bansal H.Mozannar et\u00a0al. \u201cMagentic\u2010One: A Generalist Multi\u2010Agent System for Solving Complex Tasks \u201darXiv(2024): 2411.04468 https:\/\/doi.org\/10.48550\/ARXIV.2411.04468."},{"key":"e_1_2_9_62_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-022-10246-w"},{"key":"e_1_2_9_63_1","unstructured":"S.Bubeck V.Chandrasekaran R.Eldan et\u00a0al. \u201cSparks of Artificial General Intelligence: Early Experiments with GPT\u20104 \u201darXiv(2023): 2303.12712 https:\/\/doi.org\/10.48550\/arXiv.2303.12712."},{"key":"e_1_2_9_64_1","unstructured":"K.Stechly K.Valmeekam andS.Kambhampati \u201cOn the Self\u2010Verification Limitations of Large Language Models on Reasoning and Planning Tasks \u201darXiv(2024): 2402.08115 https:\/\/doi.org\/10.48550\/ARXIV.2402.08115."},{"key":"e_1_2_9_65_1","unstructured":"K.Valmeekam K.Stechly andS.Kambhampati \u201cLLMs Still Can't Plan; Can LRMs? A Preliminary Evaluation of OpenAI's o1 on PlanBench \u201darXiv(2024): 2409.13373 https:\/\/doi.org\/10.48550\/arXiv.2409.13373."},{"key":"e_1_2_9_66_1","doi-asserted-by":"publisher","DOI":"10.1111\/tops.12278"},{"key":"e_1_2_9_67_1","doi-asserted-by":"publisher","DOI":"10.1207\/s15327809jls1401_4"},{"key":"e_1_2_9_68_1","unstructured":"J.Liu A.Liu X.Lu et\u00a0al. \u201cGenerated Knowledge Prompting for Commonsense Reasoning \u201darXiv(2021): 2110.08387 https:\/\/doi.org\/10.48550\/ARXIV.2110.08387."},{"key":"e_1_2_9_69_1","unstructured":"O.Press M.Zhang S.Min et\u00a0al. \u201cMeasuring and Narrowing the Compositionality Gap in Language Models \u201darXiv(2022): 2210.03350 https:\/\/doi.org\/10.48550\/ARXIV.2210.03350."},{"key":"e_1_2_9_70_1","unstructured":"Y.Zhou A. I.Muresanu Z.Han et\u00a0al. \u201cLarge Language Models Are Human\u2010Level Prompt Engineers \u201darXiv(2022): 2211.01910 https:\/\/doi.org\/10.48550\/ARXIV.2211.01910."},{"key":"e_1_2_9_71_1","doi-asserted-by":"publisher","DOI":"10.1093\/oxfordhb\/9780199734689.013.0035"},{"key":"e_1_2_9_72_1","doi-asserted-by":"publisher","DOI":"10.1037\/rev0000045"},{"key":"e_1_2_9_73_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.tics.2014.01.006"},{"key":"e_1_2_9_74_1","doi-asserted-by":"crossref","unstructured":"J.Wei X.Wang D.Schuurmans et\u00a0al. \u201cChain\u2010of\u2010Thought Prompting Elicits Reasoning in Large Language Models \u201darXiv(2022): 2201.11903 https:\/\/doi.org\/10.48550\/ARXIV.2201.11903.","DOI":"10.52202\/068431-1800"},{"key":"e_1_2_9_75_1","doi-asserted-by":"crossref","unstructured":"S.Yao D.Yu J.Zhao et\u00a0al. \u201cTree of Thoughts: Deliberate Problem Solving with Large Language Models \u201darXiv(2023): 2305.10601 https:\/\/doi.org\/10.48550\/ARXIV.2305.10601.","DOI":"10.52202\/075280-0517"},{"key":"e_1_2_9_76_1","unstructured":"M.Zhuge C.Zhao D.Ashley et\u00a0al. \u201cAgent\u2010as\u2010a\u2010Judge: Evaluate Agents with Agents \u201darXiv(2024): 2410.10934 https:\/\/doi.org\/10.48550\/ARXIV.2410.10934."},{"key":"e_1_2_9_77_1","unstructured":"M.RenzeandE.Guven \u201cSelf\u2010Reflection in LLM Agents: Effects on Problem\u2010Solving Performance \u201darXiv(2024): 2405.06682 https:\/\/doi.org\/10.48550\/ARXIV.2405.06682."},{"key":"e_1_2_9_78_1","volume-title":"The Architecture of Cognition. Xi","author":"Anderson J. R.","year":"1983"},{"key":"e_1_2_9_79_1","volume-title":"Human Problem Solving. xiv","author":"Newell A.","year":"1972"},{"key":"e_1_2_9_80_1","unstructured":"D.Zhou N.Sch\u00e4rli L.Hou et\u00a0al. \u201cLeast\u2010to\u2010Most Prompting Enables Complex Reasoning in Large Language Models \u201darXiv(2022): 2205.10625 https:\/\/doi.org\/10.48550\/ARXIV.2205.10625."},{"key":"e_1_2_9_81_1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-naacl.264"},{"key":"e_1_2_9_82_1","doi-asserted-by":"publisher","DOI":"10.1146\/annurev.psych.52.1.1"},{"key":"e_1_2_9_83_1","doi-asserted-by":"publisher","DOI":"10.1037\/0003-066X.55.1.68"},{"key":"e_1_2_9_84_1","doi-asserted-by":"crossref","unstructured":"A.Madaan N.Tandon P.Gupta et\u00a0al. \u201cSelf\u2010Refine: Iterative Refinement with Self\u2010Feedback \u201darXiv(2023): 2303.17651 https:\/\/doi.org\/10.48550\/ARXIV.2303.17651.","DOI":"10.52202\/075280-2019"},{"key":"e_1_2_9_85_1","unstructured":"Z.Gou Z.Shao Y.Gong et\u00a0al. \u201cCRITIC: Large Language Models Can Self\u2010Correct with Tool\u2010Interactive Critiquing \u201darXiv(2023): 2305.11738 https:\/\/doi.org\/10.48550\/ARXIV.2305.11738."},{"key":"e_1_2_9_86_1","doi-asserted-by":"crossref","unstructured":"N.Shinn F.Cassano E.Berman A.Gopinath K.Narasimhan andS.Yao \u201cReflexion: Language Agents with Verbal Reinforcement Learning \u201darXiv(2023): 2303.11366 https:\/\/doi.org\/10.48550\/ARXIV.2303.11366.","DOI":"10.52202\/075280-0377"},{"key":"e_1_2_9_87_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0079-7421(08)60452-1"},{"key":"e_1_2_9_88_1","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-psych-010814-015031"},{"key":"e_1_2_9_89_1","doi-asserted-by":"publisher","DOI":"10.1146\/annurev.ne.18.030195.001205"},{"key":"e_1_2_9_90_1","doi-asserted-by":"publisher","DOI":"10.1038\/nrn755"},{"key":"e_1_2_9_91_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.tics.2016.07.002"},{"key":"e_1_2_9_92_1","unstructured":"W.Su Y.Tang Q.Ai Z.Wu andY.Liu \u201cDRAGIN: Dynamic Retrieval Augmented Generation Based on the Information Needs of Large Language Models \u201darXiv(2024): 2403.10081 https:\/\/doi.org\/10.48550\/ARXIV.2403.10081."},{"key":"e_1_2_9_93_1","unstructured":"P.Lewis E.Perez A.Piktus et\u00a0al. \u201cRetrieval\u2010Augmented Generation for Knowledge\u2010Intensive NLP Tasks \u201darXiv(2020): 2005.11401 https:\/\/doi.org\/10.48550\/ARXIV.2005.11401."},{"key":"e_1_2_9_94_1","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3703155","article-title":"A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions","volume":"43","author":"Huang L.","year":"2025","journal-title":"ACM Transactions on Information Systems"},{"key":"e_1_2_9_95_1","unstructured":"R.Kinney C.Anastasiades R.Authur et\u00a0al. \u201cThe Semantic Scholar Open Data Platform \u201darXiv(2023): 2301.10140 https:\/\/doi.org\/10.48550\/arXiv.2301.10140."},{"key":"e_1_2_9_96_1","unstructured":"J.Priem H.Piwowar andR.Orr \u201cOpenAlex: a Fully\u2010Open Index of Scholarly Works Authors Venues Institutions and Concepts \u201darXiv(2022): 2205.01833 https:\/\/doi.org\/10.48550\/ARXIV.2205.01833."},{"key":"e_1_2_9_97_1","unstructured":"M. D.Skarlinski S.Cox J. M.Laurent et\u00a0al. \u201cLanguage Agents Achieve Superhuman Synthesis of Scientific Knowledge \u201darXiv(2024): 2409.13740 https:\/\/doi.org\/10.48550\/ARXIV.2409.13740."},{"key":"e_1_2_9_98_1","unstructured":"J.Wang J.Wang B.Athiwaratkun C.Zhang andJ.Zou \u201cMixture\u2010of\u2010Agents Enhances Large Language Model Capabilities \u201darXiv(2024): 2406.04692 https:\/\/doi.org\/10.48550\/arXiv.2406.04692."},{"key":"e_1_2_9_99_1","unstructured":"R.Bommasani D. A.Hudson E.Adeli et\u00a0al. \u201cOn the Opportunities and Risks of Foundation Models \u201darXiv(2022): 2108.07258 https:\/\/doi.org\/10.48550\/arXiv.2108.07258."},{"key":"e_1_2_9_100_1","unstructured":"J.Wei Y.Tay R.Bommasani et\u00a0al. \u201cEmergent Abilities of Large Language Models \u201darXiv(2022): 2206.07682 https:\/\/doi.org\/10.48550\/ARXIV.2206.07682."},{"key":"e_1_2_9_101_1","volume-title":"The Way We Think: Conceptual Blending and the Mind's Hidden Complexities. Xvii","author":"Fauconnier G.","year":"2002"},{"key":"e_1_2_9_102_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41583-022-00653-x"},{"key":"e_1_2_9_103_1","doi-asserted-by":"publisher","DOI":"10.1037\/10528-000"},{"key":"e_1_2_9_104_1","doi-asserted-by":"publisher","DOI":"10.1037\/h0074898"},{"key":"e_1_2_9_105_1","first-page":"317","volume-title":"Remembering: A Study in Experimental and Social Psychology. Xix","author":"Bartlett F. C.","year":"1932"},{"key":"e_1_2_9_106_1","doi-asserted-by":"publisher","DOI":"10.1037\/11494-000"},{"key":"e_1_2_9_107_1","doi-asserted-by":"crossref","unstructured":"T.Kojima S. S.Gu M.Reid Y.Matsuo andY.Iwasawa \u201cLarge Language Models Are Zero\u2010Shot Reasoners \u201darXiv(2022): 2205.11916 https:\/\/doi.org\/10.48550\/ARXIV.2205.11916.","DOI":"10.52202\/068431-1613"},{"key":"e_1_2_9_108_1","unstructured":"Y.Zhou H.Liu T.Srivastava H.Mei andC.Tan \u201cHypothesis Generation with Large Language Models \u201darXiv(2024): 2404.04326 https:\/\/doi.org\/10.48550\/ARXIV.2404.04326."},{"key":"e_1_2_9_109_1","doi-asserted-by":"publisher","DOI":"10.1287\/stsc.2024.0189"},{"key":"e_1_2_9_110_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbef.2017.12.004"},{"key":"e_1_2_9_111_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jesp.2017.01.006"},{"key":"e_1_2_9_112_1","doi-asserted-by":"publisher","DOI":"10.3758\/s13428-021-01694-3"},{"key":"e_1_2_9_113_1","unstructured":"P.Gauthier Aider: AI Pair Programming in your Terminal(2024) https:\/\/github.com\/Aider\u2010AI\/aider."},{"key":"e_1_2_9_114_1","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.39238"},{"key":"e_1_2_9_115_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-41032-5"},{"key":"e_1_2_9_116_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-025-58551-6"},{"key":"e_1_2_9_117_1","volume-title":"An Introduction to Cybernetics. Ix","author":"Ashby W. R.","year":"1956"},{"key":"e_1_2_9_118_1","volume-title":"Cybernetics; or Control and Communication in the Animal and the Machine","author":"Wiener N.","year":"1948"},{"key":"e_1_2_9_119_1","unstructured":"P. T. J.Kon J.Liu Q.Ding et\u00a0al. \u201cCurie: Toward Rigorous and Automated Scientific Experimentation with AI Agents \u201darXiv(2025): 2502.16069 https:\/\/doi.org\/10.48550\/ARXIV.2502.16069."},{"key":"e_1_2_9_120_1","doi-asserted-by":"publisher","DOI":"10.3758\/s13423-024-02602-4"},{"key":"e_1_2_9_121_1","doi-asserted-by":"publisher","DOI":"10.1167\/9.10.7"},{"key":"e_1_2_9_122_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cortex.2017.10.014"},{"key":"e_1_2_9_123_1","doi-asserted-by":"publisher","DOI":"10.1080\/00031305.2019.1583913"},{"key":"e_1_2_9_124_1","doi-asserted-by":"publisher","DOI":"10.1080\/00031305.2016.1154108"},{"key":"e_1_2_9_125_1","doi-asserted-by":"publisher","DOI":"10.1177\/0013164496056005002"},{"key":"e_1_2_9_126_1","doi-asserted-by":"publisher","DOI":"10.1037\/0003-066X.49.12.997"},{"key":"e_1_2_9_127_1","doi-asserted-by":"publisher","DOI":"10.4300\/JGME-D-12-00156.1"},{"key":"e_1_2_9_128_1","doi-asserted-by":"publisher","DOI":"10.1038\/533452a"},{"key":"e_1_2_9_129_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41562-018-0399-z"},{"key":"e_1_2_9_130_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.aac4716"},{"key":"e_1_2_9_131_1","doi-asserted-by":"publisher","DOI":"10.1177\/20966083221091867"},{"key":"e_1_2_9_132_1","unstructured":"T.Kwa B.West J.Becker et\u00a0al. \u201cMeasuring AI Ability to Complete Long Tasks \u201darXiv(2025): 2503.14499 https:\/\/doi.org\/10.48550\/ARXIV.2503.14499."},{"key":"e_1_2_9_133_1","unstructured":"M.Zhu Q.Xie Y.Weng et\u00a0al. \u201cAI Scientists Fail without Strong Implementation Capability \u201darXiv(2025): 2506.01372 https:\/\/doi.org\/10.48550\/ARXIV.2506.01372."},{"key":"e_1_2_9_134_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2012.01.010"},{"key":"e_1_2_9_135_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-020-00257-z"},{"key":"e_1_2_9_136_1","doi-asserted-by":"crossref","unstructured":"M.Ballon A.Algaba andV.Ginis \u201cThe Relationship between Reasoning and Performance in Large Language Models \u2013 o3 (mini) Thinks Harder Not Longer \u201darXiv(2025): 2502.15631 https:\/\/doi.org\/10.48550\/ARXIV.2502.15631.","DOI":"10.1038\/s41598-026-50923-2"},{"key":"e_1_2_9_137_1","unstructured":"B. Y.Lin R. L.Bras K.Richardson et\u00a0al. \u201cZebraLogic: on the Scaling Limits of LLMs for Logical Reasoning \u201darXiv(2025): 2502.01100 https:\/\/doi.org\/10.48550\/ARXIV.2502.01100."},{"key":"e_1_2_9_138_1","doi-asserted-by":"crossref","unstructured":"J.LouandY.Sun \u201cAnchoring Bias in Large Language Models: An Experimental Study \u201darXiv(2024): 2412.06593 https:\/\/doi.org\/10.48550\/arXiv.2412.06593.","DOI":"10.32388\/O2N5J6"},{"key":"e_1_2_9_139_1","unstructured":"Y.Li X.Shen Y.Miao et\u00a0al. \u201cBeyond Single\u2010Turn: A Survey on Multi\u2010Turn Interactions with Large Language Models \u201darXiv(2025): 2504.04717 https:\/\/doi.org\/10.48550\/arXiv.2504.04717."},{"key":"e_1_2_9_140_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.socec.2010.10.008"},{"key":"e_1_2_9_141_1","doi-asserted-by":"publisher","DOI":"10.1037\/0022-3514.73.3.437"},{"key":"e_1_2_9_142_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.185.4157.1124"},{"key":"e_1_2_9_143_1","doi-asserted-by":"publisher","DOI":"10.1101\/2025.05.30.656746"},{"key":"e_1_2_9_144_1","volume-title":"Experiential Learning: Experience as the Source of Learning and Development","author":"Kolb D. A.","year":"1984"},{"key":"e_1_2_9_145_1","doi-asserted-by":"publisher","DOI":"10.1037\/0033-295X.111.1.3"},{"key":"e_1_2_9_146_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.1165620"},{"key":"e_1_2_9_147_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-937X.2008.00531.x"},{"key":"e_1_2_9_148_1","doi-asserted-by":"publisher","DOI":"10.1037\/0033-2909.86.3.638"},{"key":"e_1_2_9_149_1","unstructured":"Y.Liu G.Deng Y.Li et\u00a0al. \u201cPrompt Injection Attack against LLM\u2010Integrated Applications \u201darXiv(2023): 2306.05499 https:\/\/doi.org\/10.48550\/ARXIV.2306.05499."},{"key":"e_1_2_9_150_1","volume-title":"Risk Society: Towards a New Modernity","author":"Beck U.","year":"1992"},{"key":"e_1_2_9_151_1","unstructured":"States of Knowledge: The Co\u2010Production of Science and Social Order. (Routledge London 2010)."},{"key":"e_1_2_9_152_1","doi-asserted-by":"publisher","DOI":"10.1525\/9780520908116"},{"key":"e_1_2_9_153_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11229-008-9435-2"},{"key":"e_1_2_9_154_1","doi-asserted-by":"publisher","DOI":"10.7208\/chicago\/9780226416502.001.0001"}],"container-title":["Advanced Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/advs.76675","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/advs.76675","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/advs.76675","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T10:23:25Z","timestamp":1791541405000},"score":1,"resource":{"primary":{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/10.1002\/advs.76675"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,14]]},"references-count":153,"journal-issue":{"issue":"56","published-print":{"date-parts":[[2026,10]]}},"alternative-id":["10.1002\/advs.76675"],"URL":"https:\/\/doi.org\/10.1002\/advs.76675","archive":["Portico"],"relation":{},"ISSN":["2198-3844","2198-3844"],"issn-type":[{"value":"2198-3844","type":"print"},{"value":"2198-3844","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,14]]},"assertion":[{"value":"2026-01-13","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-07-07","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-09-14","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e76675"}}