{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T21:56:17Z","timestamp":1776117377739,"version":"3.50.1"},"reference-count":61,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,8,20]],"date-time":"2025-08-20T00:00:00Z","timestamp":1755648000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"publisher","award":["GRK 2839"],"award-info":[{"award-number":["GRK 2839"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"publisher","award":["KR 5148\/3-1"],"award-info":[{"award-number":["KR 5148\/3-1"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"publisher","award":["KR 5148\/5-1"],"award-info":[{"award-number":["KR 5148\/5-1"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"publisher","award":["SCHI 1482\/3-1"],"award-info":[{"award-number":["SCHI 1482\/3-1"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>This study explores the potential for artificial agents to develop core consciousness, as proposed by Antonio Damasio's theory of consciousness. According to Damasio, the emergence of core consciousness relies on the integration of a self model, informed by representations of emotions and feelings, and a world model. We hypothesize that an artificial agent, trained via reinforcement learning (RL) in a virtual environment, can develop preliminary forms of these models as a byproduct of its primary task. The agent's main objective is to learn to play a video game and explore the environment. To evaluate the emergence of world and self models, we employ probes\u2013feedforward classifiers that use the activations of the trained agent's neural networks to predict the spatial positions of the agent itself. Our results demonstrate that the agent can form rudimentary world and self models, suggesting a pathway toward developing machine consciousness. This research provides foundational insights into the capabilities of artificial agents in mirroring aspects of human consciousness, with implications for future advancements in artificial intelligence.<\/jats:p>","DOI":"10.3389\/frai.2025.1610225","type":"journal-article","created":{"date-parts":[[2025,8,20]],"date-time":"2025-08-20T05:33:11Z","timestamp":1755667991000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":3,"title":["Probing for consciousness in machines"],"prefix":"10.3389","volume":"8","author":[{"given":"Mathis","family":"Immertreu","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Achim","family":"Schilling","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andreas","family":"Maier","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Patrick","family":"Krauss","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,8,20]]},"reference":[{"key":"B1","article-title":"Understanding intermediate layers using linear classifier probes","author":"Alain","year":"2018","journal-title":"arXiv preprint arXiv:1610.01644"},{"key":"B2","unstructured":"Andrews\n              K.\n            \n            \n              Birch\n              J.\n            \n          \n          To Understand AI Sentience, First Understand it in Animals\n          \n          2023"},{"key":"B3","first-page":"161","article-title":"\u201cA global workspace theory of conscious experience,\u201d","volume-title":"Consciousness in Philosophy and Cognitive Neuroscience","author":"Baars","year":"2013"},{"key":"B4","doi-asserted-by":"publisher","first-page":"207","DOI":"10.1162\/coli_a_00422","article-title":"Probing classifiers: promises, shortcomings, and advances","volume":"48","author":"Belinkov","year":"2022","journal-title":"Comput. Ling"},{"key":"B5","doi-asserted-by":"publisher","first-page":"503","DOI":"10.1090\/S0002-9904-1954-09848-8","article-title":"The theory of dynamic programming","volume":"60","author":"Bellman","year":"1954","journal-title":"Bull. New Ser. Am. Math. Soc"},{"key":"B6","doi-asserted-by":"publisher","first-page":"408","DOI":"10.1016\/j.tics.2019.02.006","article-title":"Reinforcement learning, fast and slow","volume":"23","author":"Botvinick","year":"2019","journal-title":"Trends Cogn. Sci"},{"key":"B7","article-title":"Exploration by random network distillation","author":"Burda","year":"2018","journal-title":"arXiv preprint arXiv:1810.12894"},{"key":"B8","doi-asserted-by":"publisher","first-page":"216","DOI":"10.1609\/aiide.v4i1.18700","article-title":"\u201cMonte-carlo tree search: a new framework for game AI,\u201d","author":"Chaslot","year":"2008","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment"},{"key":"B9","first-page":"52","article-title":"\u201cOracle-sage: planning ahead in graph-based deep reinforcement learning,\u201d","volume-title":"Joint European Conference on Machine Learning and Knowledge Discovery in Databases","author":"Chester","year":"2022"},{"key":"B10","doi-asserted-by":"publisher","first-page":"718","DOI":"10.1038\/nrn.2016.113","article-title":"Mind-wandering as spontaneous thought: a dynamic framework","volume":"17","author":"Christoff","year":"2016","journal-title":"Nat. Rev. Neurosci"},{"key":"B11","doi-asserted-by":"publisher","first-page":"2231","DOI":"10.1093\/brain\/awac194","article-title":"Homeostatic feelings and the biology of consciousness","volume":"145","author":"Damasio","year":"2022","journal-title":"Brain"},{"key":"B12","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1016\/B978-0-12-374168-4.00001-0","article-title":"\u201cConsciousness: an overview of the phenomenon and of its possible neural basis,\u201d","author":"Damasio","year":"2009","journal-title":"The Neurology of Consciousness: Cognitive Neuroscience and Neuropathology"},{"key":"B13","doi-asserted-by":"publisher","first-page":"414","DOI":"10.1038\/s41586-021-04301-9","article-title":"Magnetic control of tokamak plasmas through deep reinforcement learning","volume":"602","author":"Degrave","year":"2022","journal-title":"Nature"},{"key":"B14","article-title":"Go-explore: a new approach for hard-exploration problems","author":"Ecoffet","year":"2019","journal-title":"arXiv preprint arXiv:1901.10995"},{"key":"B15","doi-asserted-by":"publisher","first-page":"106852","DOI":"10.1016\/j.jobe.2023.106852","article-title":"Comparative study of model-based and model-free reinforcement learning control performance in hvac systems","volume":"74","author":"Gao","year":"2023","journal-title":"J. Build. Eng"},{"key":"B16","doi-asserted-by":"publisher","first-page":"7193","DOI":"10.1523\/JNEUROSCI.0151-18.2018","article-title":"The successor representation: its computational logic and neural substrates","volume":"38","author":"Gershman","year":"2018","journal-title":"J. Neurosci"},{"key":"B17","doi-asserted-by":"publisher","first-page":"3389","DOI":"10.1109\/ICRA.2017.7989385","article-title":"\u201cDeep reinforcement learning for robotic manipulation with asynchronous off-policy updates,\u201d","author":"Gu","year":"2017","journal-title":"2017 IEEE International Conference on Robotics and Automation (ICRA)"},{"key":"B18","article-title":"World models","author":"Ha","year":"2018","journal-title":"arXiv preprint arXiv:1803.10122"},{"key":"B19","first-page":"1861","article-title":"\u201cSoft actor-critic: off-policy maximum entropy deep reinforcement learning with a stochastic actor,\u201d","volume-title":"International Conference on Machine Learning","author":"Haarnoja","year":"2018"},{"key":"B20","article-title":"Mastering diverse domains through world models","author":"Hafner","year":"2023","journal-title":"arXiv preprint arXiv:2301.04104"},{"key":"B21","first-page":"41","article-title":"\u201cInsights from the neurips 2021 nethack challenge,\u201d","volume-title":"NeurIPS 2021 Competitions and Demonstrations Track","author":"Hambro","year":"2022"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1615","DOI":"10.1145\/3715275.3732108","article-title":"\u201cPeople cannot distinguish gpt-4 from a human in a turing test,\u201d","author":"Jones","year":"2024","journal-title":"Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency"},{"key":"B23","article-title":"Motif: Intrinsic motivation from artificial intelligence feedback","author":"Klissarov","year":"2023","journal-title":"arXiv preprint arXiv:2310.00166"},{"key":"B24","doi-asserted-by":"publisher","first-page":"556544","DOI":"10.3389\/fncom.2020.556544","article-title":"Will we ever have conscious machines?","volume":"14","author":"Krauss","year":"2020","journal-title":"Front. Comput. Neurosci"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/8404.001.0001","article-title":"\u201cRepresentational similarity analysis of object population codes in humans, monkeys, and models,\u201d","author":"Kriegeskorte","year":"2012","journal-title":"Visual population codes: towards a common multivariate framework for cell recording and functional imaging"},{"key":"B26","doi-asserted-by":"publisher","first-page":"28","DOI":"10.1016\/j.pbiomolbio.2023.12.003","article-title":"A landscape of consciousness: toward a taxonomy of explanations and implications","volume":"190","author":"Kuhn","year":"2024","journal-title":"Prog. Biophys. Mol. Biol"},{"key":"B27","article-title":"\u201cThe NetHack learning environment,\u201d","author":"K\u00fcttler","year":"2020","journal-title":"Proceedings of the Conference on Neural Information Processing Systems (NeurIPS)"},{"key":"B28","article-title":"Emergent world representations: Exploring a sequence model trained on a synthetic task","author":"Li","year":"2022","journal-title":"arXiv preprint arXiv:2210.13382"},{"key":"B29","first-page":"3053","article-title":"\u201cRLlib: abstractions for distributed reinforcement learning,\u201d","volume-title":"Proceedings of the 35th International Conference on Machine Learning, volume 80 of Proceedings of Machine Learning Research","author":"Liang","year":"2018"},{"key":"B30","doi-asserted-by":"publisher","first-page":"446","DOI":"10.1038\/s42256-019-0103-7","article-title":"Homeostasis and soft robotics in the design of feeling machines","volume":"1","author":"Man","year":"2019","journal-title":"Nat. Mach. Intell"},{"key":"B31","article-title":"\u201cSkillhack: a benchmark for skill transfer in open-ended reinforcement learning,\u201d","author":"Matthews","year":"2022","journal-title":"ICLR Workshop on Agent Learning in Open-Endedness"},{"key":"B32","doi-asserted-by":"crossref","first-page":"2671","DOI":"10.23919\/ECC.2007.7068926","article-title":"\u201cConvergence of q-learning with linear function approximation,\u201d","volume-title":"2007 European Control Conference (ECC)","author":"Melo","year":"2007"},{"key":"B33","article-title":"Playing atari with deep reinforcement learning","author":"Mnih","year":"2013","journal-title":"CoRR, abs\/1312.5602"},{"key":"B34","doi-asserted-by":"publisher","first-page":"680","DOI":"10.1038\/s41562-017-0180-8","article-title":"The successor representation in human reinforcement learning","volume":"1","author":"Momennejad","year":"2017","journal-title":"Nat. Hum. Behav"},{"key":"B35","article-title":"Learning to query internet text for informing reinforcement learning agents","author":"Nottingham","year":"2022","journal-title":"arXiv preprint arXiv:2205.13079"},{"key":"B36","volume-title":"Affective Neuroscience: The Foundations of Human and Animal Emotions","author":"Panksepp","year":"2004"},{"key":"B37","first-page":"17473","article-title":"\u201cEvolving curricula with regret-based environment design,\u201d","volume-title":"International Conference on Machine Learning","author":"Parker-Holder","year":"2022"},{"key":"B38","first-page":"2778","article-title":"\u201cCuriosity-driven exploration by self-supervised prediction,\u201d","volume-title":"International Conference on Machine Learning","author":"Pathak","year":"2017"},{"key":"B39","first-page":"705","article-title":"\u201cCora: benchmarks, baselines, and metrics as a platform for continual reinforcement learning agents,\u201d","volume-title":"Conference on Lifelong Learning Agents","author":"Powers","year":"2022"},{"key":"B40","doi-asserted-by":"publisher","first-page":"676","DOI":"10.1073\/pnas.98.2.676","article-title":"A default mode of brain function","volume":"98","author":"Raichle","year":"2001","journal-title":"Proc. Nat. Acad. Sci"},{"key":"B41","article-title":"\u201cMinihack the planet: a sandbox for open-ended reinforcement learning research,\u201d","author":"Samvelyan","year":"2021","journal-title":"Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 1)"},{"key":"B42","doi-asserted-by":"crossref","DOI":"10.7551\/mitpress\/3115.003.0030","article-title":"\u201cA possibility for implementing curiosity and boredom in model-building neural controllers,\u201d","volume-title":"From Animals to Animats: Proceedings of the First International Conference on Simulation of Adaptive Behavior","author":"Schmidhuber","year":"1991"},{"key":"B43","doi-asserted-by":"publisher","first-page":"604","DOI":"10.1038\/s41586-020-03051-4","article-title":"Mastering Atari, go, chess and Shogi by planning with a learned model","volume":"588","author":"Schrittwieser","year":"2020","journal-title":"Nature"},{"key":"B44","article-title":"High-dimensional continuous control using generalized advantage estimation","author":"Schulman","year":"2015","journal-title":"arXiv preprint arXiv:1506.02438"},{"key":"B45","article-title":"Proximal policy optimization algorithms","author":"Schulman","year":"2017","journal-title":"arXiv preprint arXiv:1707.06347"},{"key":"B46","doi-asserted-by":"publisher","first-page":"417","DOI":"10.1017\/S0140525X00005756","article-title":"Minds, brains, and programs","volume":"3","author":"Searle","year":"1980","journal-title":"Behav. Brain Sci"},{"key":"B47","doi-asserted-by":"publisher","first-page":"439","DOI":"10.1038\/s41583-022-00587-4","article-title":"Theories of consciousness","volume":"23","author":"Seth","year":"2022","journal-title":"Nat. Rev. Neurosci"},{"key":"B48","doi-asserted-by":"publisher","first-page":"484","DOI":"10.1038\/nature16961","article-title":"Mastering the game of go with deep neural networks and tree search","volume":"529","author":"Silver","year":"2016","journal-title":"Nature"},{"key":"B49","doi-asserted-by":"publisher","DOI":"10.53765\/20512201.28.11.153","author":"Solms","year":"2021","journal-title":"The Hidden Spring: A Journey to the Source of Consciousness"},{"key":"B50","doi-asserted-by":"publisher","first-page":"1643","DOI":"10.1038\/nn.4650","article-title":"The hippocampus as a predictive map","volume":"20","author":"Stachenfeld","year":"2017","journal-title":"Nat. Neurosci"},{"key":"B51","first-page":"391","article-title":"\u201cConceptual cognitive maps formation with neural successor networks and word embeddings,\u201d","volume-title":"2023 IEEE International Conference on Development and Learning (ICDL)","author":"Stoewer","year":""},{"key":"B52","article-title":"Multi-modal cognitive maps based on neural networks trained on successor representations","author":"Stoewer","year":"","journal-title":"arXiv preprint arXiv:2401.01364"},{"key":"B53","doi-asserted-by":"publisher","first-page":"3644","DOI":"10.1038\/s41598-023-30307-6","article-title":"Neural network based formation of cognitive maps of semantic spaces and the putative emergence of abstract concepts","volume":"13","author":"Stoewer","year":"","journal-title":"Sci. Rep"},{"key":"B54","doi-asserted-by":"publisher","first-page":"11233","DOI":"10.1038\/s41598-022-14916-1","article-title":"Neural network based successor representations to form cognitive maps of space and language","volume":"12","author":"Stoewer","year":"2022","journal-title":"Sci. Rep"},{"key":"B55","doi-asserted-by":"crossref","first-page":"1481","DOI":"10.1109\/ICMLA58977.2023.00223","article-title":"\u201cWord class representations spontaneously emerge in a deep neural network trained on next word prediction,\u201d","volume-title":"2023 International Conference on Machine Learning and Applications (ICMLA)","author":"Surendra","year":"2023"},{"key":"B56","volume-title":"Reinforcement Learning: An Introduction","author":"Sutton","year":"2018"},{"key":"B57","article-title":"\u201cPolicy gradient methods for reinforcement learning with function approximation,\u201d","author":"Sutton","year":"1999","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B58","doi-asserted-by":"publisher","first-page":"450","DOI":"10.1038\/nrn.2016.44","article-title":"Integrated information theory: from consciousness to its physical substrate","volume":"17","author":"Tononi","year":"2016","journal-title":"Nat. Rev. Neurosci"},{"key":"B59","doi-asserted-by":"publisher","first-page":"674","DOI":"10.1109\/9.580874","article-title":"An analysis of temporal-difference learning with function approximation","volume":"42","author":"Tsitsiklis","year":"1997","journal-title":"IEEE Trans. Automat. Contr"},{"key":"B60","doi-asserted-by":"publisher","first-page":"433","DOI":"10.1093\/mind\/LIX.236.433","author":"Turing","year":"1950","journal-title":"I\u2013Computing Machinery and Intelligence"},{"key":"B61","doi-asserted-by":"publisher","first-page":"241","DOI":"10.1080\/09540099108946587","article-title":"Function optimization using connectionist reinforcement learning algorithms","volume":"3","author":"Williams","year":"1991","journal-title":"Conn. Sci"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1610225\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,20]],"date-time":"2025-08-20T05:33:18Z","timestamp":1755667998000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1610225\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,20]]},"references-count":61,"alternative-id":["10.3389\/frai.2025.1610225"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1610225","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8,20]]},"article-number":"1610225"}}