{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,5]],"date-time":"2026-11-05T04:24:53Z","timestamp":1793852693376,"version":"4.6.1"},"reference-count":67,"publisher":"National Academy of Sciences","issue":"45","license":[{"start":{"date-parts":[[2026,11,4]],"date-time":"2026-11-04T00:00:00Z","timestamp":1793750400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000006","name":"DOW | USN | Office of Naval Research","doi-asserted-by":"publisher","award":["N00014-22-1-2773)"],"award-info":[{"award-number":["N00014-22-1-2773)"]}],"id":[{"id":"10.13039\/100000006","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["2433429"],"award-info":[{"award-number":["2433429"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.pnas.org"],"crossmark-restriction":true},"short-container-title":["Proc. Natl. Acad. Sci. U.S.A."],"published-print":{"date-parts":[[2026,11,10]]},"abstract":"<jats:p>\n                    What drives exploration? Understanding the intrinsic motivation to explore is a long-standing challenge in both cognitive science and AI. In this study, we directly compare adults, children aged 8 to 12 y, and AI agents exploring a complex, open-ended environment called Crafter (\n                    <jats:xref ref-type=\"bibr\">1<\/jats:xref>\n                    ). We study how three common information-theoretic intrinsic objectives\u2014entropy, information gain, and empowerment\u2014relate to their behavior. We find that all three are positively correlated with human exploration progress for both children and adults, although entropy and empowerment are more strongly related to exploration progress in children than information gain. Furthermore, across agents and humans we observe that entropy initially increases rapidly, then plateaus, while empowerment increases continuously, suggesting that in early exploration they look for novel states, while in advanced exploration they look for states they can control. Finally, we find preliminary evidence that private speech about goals may aid exploration in children. Our data are available at\n                    <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" ext-link-type=\"uri\" xlink:href=\"https:\/\/github.com\/alyd\/humans_in_crafter_data\">https:\/\/github.com\/alyd\/humans_in_crafter_data<\/jats:ext-link>\n                    .\n                  <\/jats:p>","DOI":"10.1073\/pnas.2612366123","type":"journal-article","created":{"date-parts":[[2026,11,4]],"date-time":"2026-11-04T16:21:11Z","timestamp":1793809271000},"update-policy":"https:\/\/doi.org\/10.1073\/pnas.cm10313","source":"Crossref","is-referenced-by-count":1,"title":["Intrinsically motivated humans and agents in open-world exploration"],"prefix":"10.1073","volume":"123","author":[{"given":"Aly","family":"Lidayan","sequence":"first","affiliation":[{"name":"Berkeley AI Research Lab, Computer Science Department, University of California Berkeley","place":["Berkeley"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuqing","family":"Du","sequence":"additional","affiliation":[{"name":"Berkeley AI Research Lab, Computer Science Department, University of California Berkeley","place":["Berkeley"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Eliza","family":"Kosoy","sequence":"additional","affiliation":[{"name":"Berkeley AI Research Lab, Computer Science Department, University of California Berkeley","place":["Berkeley"]},{"name":"Psychology Department, University of California Berkeley","place":["Berkeley"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maria","family":"Rufova","sequence":"additional","affiliation":[{"name":"Psychology Department, University of California Berkeley","place":["Berkeley"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pieter","family":"Abbeel","sequence":"additional","affiliation":[{"name":"Berkeley AI Research Lab, Computer Science Department, University of California Berkeley","place":["Berkeley"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7183-0867","authenticated-orcid":false,"given":"Alison","family":"Gopnik","sequence":"additional","affiliation":[{"name":"Berkeley AI Research Lab, Computer Science Department, University of California Berkeley","place":["Berkeley"]},{"name":"Psychology Department, University of California Berkeley","place":["Berkeley"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"341","published-online":{"date-parts":[[2026,11,4]]},"reference":[{"key":"e_1_3_4_1_2","unstructured":"D. Hafner Benchmarking the spectrum of agent capabilities. arXiv [Preprint] (2021). https:\/\/doi.org\/10.48550\/arXiv.2109.06780 (Accessed 6 October 2026)."},{"key":"e_1_3_4_2_2","unstructured":"B. Matusch J. Ba D. Hafner Evaluating agents without rewards. arXiv [Preprint] (2020). https:\/\/doi.org\/10.48550\/arXiv.2012.11538 (Accessed 6 October 2026)."},{"key":"e_1_3_4_3_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41562-023-01662-1"},{"key":"e_1_3_4_4_2","first-page":"20190502","article-title":"Childhood as a solution to explore\u2013exploit tensions","volume":"2020","author":"Gopnik A.","year":"1803","unstructured":"A. Gopnik, Childhood as a solution to explore\u2013exploit tensions. Philos. Trans. R. Soc. Lond. B Biol. Sci. 2020, 20190502 (1803).","journal-title":"Philos. Trans. R. Soc. Lond. B Biol. Sci."},{"key":"e_1_3_4_5_2","unstructured":"E. Hughes Open-endedness is essential for artificial superhuman intelligence. arXiv [Preprint] (2024). https:\/\/doi.org\/10.48550\/arXiv.2406.04268 (Accessed 6 October 2026)."},{"key":"e_1_3_4_6_2","unstructured":"A. Lidayan M. D. Dennis S. Russell BAMDP shaping: A unified framework for intrinsic motivation and reward shaping. arXiv [Preprint] (2025). https:\/\/doi.org\/10.48550\/arXiv.2409.05358 (Accessed 6 October 2026)."},{"key":"e_1_3_4_7_2","doi-asserted-by":"publisher","DOI":"10.3390\/e25020327"},{"key":"e_1_3_4_8_2","first-page":"222","article-title":"A possibility for implementing curiosity and boredom in model-building neural controllers","author":"Schmidhuber J.","year":"1991","unstructured":"J. Schmidhuber, A possibility for implementing curiosity and boredom in model-building neural controllers. Proc. Int. Conf. Simul. Adapt. Behav., 222\u2013227 (1991).","journal-title":"Proc. Int. Conf. Simul. Adapt. Behav."},{"key":"e_1_3_4_9_2","doi-asserted-by":"publisher","DOI":"10.1017\/9781009026949.004"},{"key":"e_1_3_4_10_2","first-page":"20190503","article-title":"The elaboration of exploratory play","volume":"2020","author":"Pelz M.","year":"1803","unstructured":"M. Pelz, C. Kidd, The elaboration of exploratory play. Philos. Trans. R. Soc. B 2020, 20190503 (1803).","journal-title":"Philos. Trans. R. Soc. B"},{"key":"e_1_3_4_11_2","doi-asserted-by":"publisher","DOI":"10.1109\/IRDS.2002.1041446"},{"key":"e_1_3_4_12_2","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177728069"},{"key":"e_1_3_4_13_2","first-page":"128","volume-title":"IEEE Congress on Evolutionary Computation","author":"Klyubin A. S.","year":"2005","unstructured":"A. S. Klyubin, D. Polani, C. L. Nehaniv, \u201cEmpowerment: A universal agent-centric measure of control\u201d in IEEE Congress on Evolutionary Computation (IEEE, 2005), Vol. 1, pp. 128\u2013135."},{"key":"e_1_3_4_14_2","doi-asserted-by":"publisher","DOI":"10.1037\/11193-000"},{"key":"e_1_3_4_15_2","first-page":"1","volume-title":"Thinking and Speech","author":"Vygotsky L.","year":"1987","unstructured":"L. Vygotsky, Thinking and Speech (Problems of General Psychology, The collected works of LS Vygotsky, 1987), p. 1."},{"key":"e_1_3_4_16_2","doi-asserted-by":"publisher","DOI":"10.1016\/0361-476X(86)90030-5"},{"key":"e_1_3_4_17_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-78252-y"},{"key":"e_1_3_4_18_2","doi-asserted-by":"publisher","DOI":"10.1080\/03004430.2010.494253"},{"key":"e_1_3_4_19_2","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-devpsych-070120-014806"},{"key":"e_1_3_4_20_2","first-page":"3369","article-title":"Creativity, compositionality, and common sense in human goal generation","volume":"44","author":"Davidson G.","year":"2022","unstructured":"G. Davidson, T. M. Gureckis, B. Lake, Creativity, compositionality, and common sense in human goal generation. Proc. Annu. Meet. Cognit. Sci. Soc. 44, 3369\u20133375 (2022).","journal-title":"Proc. Annu. Meet. Cognit. Sci. Soc."},{"key":"e_1_3_4_21_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.tics.2023.08.011"},{"key":"e_1_3_4_22_2","doi-asserted-by":"crossref","unstructured":"R. Portelas C. Colas L. Weng K. Hofmann P. Y. Oudeyer Automatic curriculum learning for deep RL: A short survey. arXiv [Preprint] (2020). https:\/\/doi.org\/10.48550\/arXiv.2003.04664 (Accessed 6 October 2026).","DOI":"10.24963\/ijcai.2020\/671"},{"key":"e_1_3_4_23_2","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1800"},{"key":"e_1_3_4_24_2","doi-asserted-by":"publisher","DOI":"10.1613\/jair.1.13554"},{"key":"e_1_3_4_25_2","unstructured":"Y. Du Guiding pre-training in reinforcement learning with large language models. arXiv [Preprint] (2023). https:\/\/doi.org\/10.48550\/arXiv.2302.06692 (Accessed 6 October 2026)."},{"key":"e_1_3_4_26_2","unstructured":"J. Zhang J. Lehman K. Stanley J. Clune OMNI: Open-endedness via models of human notions of interestingness. arXiv [Preprint] (2023). https:\/\/doi.org\/10.48550\/arXiv.2306.01711 (Accessed 6 October 2026)."},{"key":"e_1_3_4_27_2","first-page":"2681","volume-title":"International Conference on Machine Learning","author":"Hazan E.","year":"2019","unstructured":"E. Hazan, S. Kakade, K. Singh, A. Van Soest, \u201cProvably efficient maximum entropy exploration\u201d in International Conference on Machine Learning, (PMLR, 2019), pp. 2681\u20132691."},{"key":"e_1_3_4_28_2","first-page":"18459","article-title":"Behavior from the void: Unsupervised active pre-training","volume":"34","author":"Liu H.","year":"2021","unstructured":"H. Liu, P. Abbeel, Behavior from the void: Unsupervised active pre-training. Adv. Neural Inf. Process. Syst. 34, 18459\u201318473 (2021).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"e_1_3_4_29_2","first-page":"11920","volume-title":"International Conference on Machine Learning","author":"Yarats D.","year":"2021","unstructured":"D. Yarats, R. Fergus, A. Lazaric, L. Pinto, \u201cReinforcement learning with prototypical representations\u201d in International Conference on Machine Learning, (PMLR, 2021), pp. 11920\u201311931."},{"key":"e_1_3_4_30_2","first-page":"1109","article-title":"Variational information maximizing exploration","volume":"29","author":"Houthooft R.","year":"2016","unstructured":"R. Houthooft , Variational information maximizing exploration. Adv. Neural Inf. Process. Syst. 29, 1109\u20131117 (2016).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"e_1_3_4_31_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW.2017.70"},{"key":"e_1_3_4_32_2","first-page":"25217","article-title":"NovelD: A simple yet effective exploration criterion","volume":"34","author":"Zhang T.","year":"2021","unstructured":"T. Zhang , NovelD: A simple yet effective exploration criterion. Adv. Neural Inf. Process. Syst. 34, 25217\u201325230 (2021).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"e_1_3_4_33_2","doi-asserted-by":"publisher","DOI":"10.1007\/11553090_75"},{"key":"e_1_3_4_34_2","doi-asserted-by":"publisher","DOI":"10.3390\/e16052789"},{"key":"e_1_3_4_35_2","unstructured":"K. Gregor D. J. Rezende D. Wierstra Variational intrinsic control. arXiv [Preprint] (2016). https:\/\/doi.org\/10.48550\/arXiv.1611.07507 (Accessed 6 October 2026)."},{"key":"e_1_3_4_36_2","unstructured":"D. Hafner J. Pasukonis J. Ba T. Lillicrap Mastering diverse domains through world models. arXiv [Preprint] (2023). https:\/\/doi.org\/10.48550\/arXiv.2301.04104 (Accessed 6 October 2026)."},{"key":"e_1_3_4_37_2","unstructured":"Y. Burda H. Edwards A. Storkey O. Klimov Exploration by random network distillation. arXiv [Preprint] (2018). https:\/\/doi.org\/10.48550\/arXiv.1810.12894 (Accessed 6 October 2026)."},{"key":"e_1_3_4_38_2","doi-asserted-by":"publisher","DOI":"10.1037\/0012-1649.43.4.1045"},{"key":"e_1_3_4_39_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cognition.2011.03.003"},{"key":"e_1_3_4_40_2","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-8624.2011.01691.x"},{"key":"e_1_3_4_41_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.tics.2012.06.004"},{"key":"e_1_3_4_42_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.dcn.2019.100733"},{"key":"e_1_3_4_43_2","doi-asserted-by":"publisher","DOI":"10.1007\/s00221-014-3907-z"},{"key":"e_1_3_4_44_2","first-page":"68","article-title":"Novelty and curiosity as determinants of exploratory behaviour","volume":"41","author":"Berlyne D. E.","year":"1950","unstructured":"D. E. Berlyne, Novelty and curiosity as determinants of exploratory behaviour. Br. J. Psychol. 41, 68 (1950).","journal-title":"Br. J. Psychol."},{"key":"e_1_3_4_45_2","doi-asserted-by":"publisher","DOI":"10.2307\/1126663"},{"key":"e_1_3_4_46_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cognition.2022.105119"},{"key":"e_1_3_4_47_2","first-page":"1360","article-title":"Developmental change in what elicits curiosity","volume":"43","author":"Liquin E. G.","year":"2021","unstructured":"E. G. Liquin, F. Callaway, T. Lombrozo, Developmental change in what elicits curiosity. Proc. Annu. Meet. Cogn. Sci. Soc. 43, 1360\u20131366 (2021).","journal-title":"Proc. Annu. Meet. Cogn. Sci. Soc."},{"key":"e_1_3_4_48_2","doi-asserted-by":"publisher","DOI":"10.1111\/desc.13411"},{"key":"e_1_3_4_49_2","doi-asserted-by":"publisher","DOI":"10.1111\/cdev.12060"},{"key":"e_1_3_4_50_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41562-023-01661-2"},{"key":"e_1_3_4_51_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-022-29742-2"},{"key":"e_1_3_4_52_2","doi-asserted-by":"publisher","DOI":"10.1098\/rsta.2025.0003"},{"key":"e_1_3_4_53_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cognition.2017.12.014"},{"key":"e_1_3_4_54_2","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2014.00317"},{"key":"e_1_3_4_55_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-021-26196-w"},{"key":"e_1_3_4_56_2","unstructured":"E. Kosoy Exploring exploration: Comparing children with RL agents in unified environments. arXiv [Preprint] (2020). https:\/\/doi.org\/10.48550\/arXiv.2005.02880 (Accessed 6 October 2026)."},{"key":"e_1_3_4_57_2","unstructured":"R. Dubey P. Agrawal D. Pathak T. L. Griffiths A. A. Efros Investigating human priors for playing video games. arXiv [Preprint] (2018). https:\/\/doi.org\/10.48550\/arXiv.1802.10217 (Accessed 6 October 2026)."},{"key":"e_1_3_4_58_2","first-page":"1","article-title":"Toddler-inspired visual object learning","volume":"31","author":"Bambach S.","year":"2018","unstructured":"S. Bambach, D. Crandall, L. Smith, C. Yu, Toddler-inspired visual object learning. Adv. Neural Inf. Process. Syst. 31, 1\u201310 (2018).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"e_1_3_4_59_2","first-page":"390","volume-title":"Conference on Causal Learning and Reasoning","author":"Kosoy E.","year":"2022","unstructured":"E. Kosoy , \u201cLearning causal overhypotheses through exploration in children and computational models\u201d in Conference on Causal Learning and Reasoning, (PMLR, 2022), pp. 390\u2013406."},{"key":"e_1_3_4_60_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-53734-9_4"},{"key":"e_1_3_4_61_2","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT.2004.1365218"},{"key":"e_1_3_4_62_2","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2308950121"},{"key":"e_1_3_4_63_2","first-page":"1","article-title":"A path towards autonomous machine intelligence version 0.9. 2, 2022\u201306-27","volume":"62","author":"LeCun Y.","year":"2022","unstructured":"Y. LeCun, A path towards autonomous machine intelligence version 0.9. 2, 2022\u201306-27. Open Review 62, 1\u201362 (2022).","journal-title":"Open Review"},{"key":"e_1_3_4_64_2","unstructured":"H. Zhou F. Dai A. Gopnik C. Wu What is worth representing? Representational empowerment for continual model construction (Neurips Main Track Sydney 2026). Workshop on Bridging Language Agent and World Models for Reasoning and Planning."},{"key":"e_1_3_4_65_2","first-page":"301","article-title":"Nearest neighbor estimates of entropy","volume":"23","author":"Singh H.","year":"2003","unstructured":"H. Singh, N. Misra, V. Hnizdo, A. Fedorowicz, E. Demchuk, Nearest neighbor estimates of entropy. Am. J. Math. Manag. Sci. 23, 301\u2013321 (2003).","journal-title":"Am. J. Math. Manag. Sci."},{"key":"e_1_3_4_66_2","doi-asserted-by":"crossref","unstructured":"M. Hessel Rainbow: Combining improvements in deep reinforcement learning. Proc. AAAI Conf. Artif. Intell. 32 3215\u20133222 (2018).","DOI":"10.1609\/aaai.v32i1.11796"},{"key":"e_1_3_4_67_2","unstructured":"A. Lidayan Humans in Crafter dataset. https:\/\/github.com\/alyd\/humans_in_crafter_data. Deposited 4 September 2026."}],"container-title":["Proceedings of the National Academy of Sciences"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pnas.org\/doi\/pdf\/10.1073\/pnas.2612366123","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,11,4]],"date-time":"2026-11-04T16:21:25Z","timestamp":1793809285000},"score":1,"resource":{"primary":{"URL":"https:\/\/pnas.org\/doi\/10.1073\/pnas.2612366123"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,11,4]]},"references-count":67,"journal-issue":{"issue":"45","published-print":{"date-parts":[[2026,11,10]]}},"alternative-id":["10.1073\/pnas.2612366123"],"URL":"https:\/\/doi.org\/10.1073\/pnas.2612366123","relation":{},"ISSN":["0027-8424","1091-6490"],"issn-type":[{"value":"0027-8424","type":"print"},{"value":"1091-6490","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,11,4]]},"assertion":[{"value":"2026-04-07","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-24","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-11-04","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e2612366123"}}