Latest Hits in Cognitively Inpsired AI

4 minute read

This paper collection reflects some of the most exiting trends in computational cognitive science and cognitively inspired AI, in my opinion.


How can AI and cognitive science shed light on each other’s biggest challenges?

  • Lake, B. M., Ullman, T. D., Tenenbaum, J. B., & Gershman, S. J. (2017). Building machines that learn and think like people. Behavioral and Brain Sciences, 40. (PDF)
  • Ullman, T. D., & Tenenbaum, J. B. (2020). Bayesian models of conceptual development: Learning as building models of the world. Annual Review of Developmental Psychology, 2, 533-558. (PDF)
  • Hamrick, J. B. & Mohamed, S. (2020). Levels of Analysis for Machine Learning. In Proceedings of the ICLR 2020 Workshop on Bridging AI and Cognitive Science. (PDF)


New research on how people reason about causality, solve problems, make decisions, etc.

  • Rule, J. S., Tenenbaum, J. B., & Piantadosi, S. T. (2020). The child as hacker. Trends in Cognitive Sciences, 24(11), 900-915. (PDF)
  • Adams, G. S., Converse, B. A., Hales, A. H., & Klotz, L. E. (2021). People systematically overlook subtractive changes. Nature, 592(7853), 258-261. (link, behind paywall)
  • Dasgupta, I., Schulz, E., Tenenbaum, J. B., & Gershman, S. J. (2020). A theory of learning to infer. Psychological Review, 127(3), 412-441. (PDF)


It’s fascinating how the lines are blurred between “using” vs. “understanding” a language by gigantic models such as BERT, GPT-3, and the likes.

  • Lake, B. M., & Murphy, G. L. (2020). Word meaning in minds and machines. arXiv preprint arXiv:2008.01766. (PDF)
  • Dasgupta, I., Guo, D., Stuhlmüller, A., Gershman, S. J., & Goodman, N. D. (2018). Evaluating compositionality in sentence embeddings. arXiv preprint arXiv:1802.04302. (PDF)
  • Mollica, F., & Piantadosi, S. T. (2019). Humans store about 1.5 megabytes of information during language acquisition. Royal Society Open Science, 6(3), 181393. (PDF)

Mental Simulation

Josh Tenenbaum once said in a talk: “What Pixar engineers spend millions of dollars doing, my 4-year-old daughter can do intuitively” — The mind simulates the physical world effortlessly with uncannily efficient physical engines.

  • Ullman, T. D., Spelke, E., Battaglia, P., & Tenenbaum, J. B. (2017). Mind games: Game engines as an architecture for intuitive physics. Trends in Cognitive Sciences, 21(9), 649-665. (PDF)
  • Ludwin-Peery, E., Bramley, N., Davis, E., & Gureckis, T. (2021). Limits on simulation approaches in intuitive physics. Cognitive Psychology, 127, 1-25. (PDF)
  • Bates, C. J., Yildirim, I., Tenenbaum, J. B., & Battaglia, P. (2019). Modeling human intuitions about liquid flow with particle-based simulation. PLoS Computational Biology, 15(7), e1007210. (link)
  • Smith, K., Mei, L., Yao, S., Wu, J., Spelke, E., Tenenbaum, J., & Ullman, T. (2019). Modeling expectation violation in intuitive physics with coarse probabilistic object representations. Advances in Neural Information Processing Systems, 32, 8985-8995. (PDF)

Social Cognition

New agent-based research on how we learn from others (imitation, teaching…) and about others (social evaluation, action understanding…)

  • Jara-Ettinger, J., Schulz, L. E., & Tenenbaum, J. B. (2020). The naive utility calculus as a unified, quantitative framework for action understanding. Cognitive Psychology, 123, 101334. (PDF)
  • Yang, S. C. H., Vong, W. K., Yu, Y., & Shafto, P. (2019). A unifying computational framework for teaching and active learning. Topics in Cognitive Science, 11(2), 316-337. (PDF)
  • Gershman, S.J. & Cikara, M. (2020). Social-structure learning. Current Directions in Psychological Science, 29, 460-466. (PDF)


How can machines solve Hume’s “is-ought” problem by going from how the world is to how it should be? How does learning to be moral differ from other statistical learning problems?

  • Levine, S., Kleiman-Weiner, M., Schulz, L., Tenenbaum, J., & Cushman, F. (2020). The logic of universalization guides moral judgment. Proceedings of the National Academy of Sciences, 117(42), 26158-26169. (PDF)
  • Bear, A., Bensinger, S., Jara-Ettinger, J., Knobe, J., & Cushman, F. (2020). What comes to mind?. Cognition, 194, 104057. (PDF)
  • De Freitas, J., Censi, A., Smith, B. W., Di Lillo, L., Anthony, S. E., & Frazzoli, E. (2021). From driverless dilemmas to more practical commonsense tests for automated vehicles. Proceedings of the National Academy of Sciences, 118(11). (PDF, possibly behind paywall)

Deep Reinforcement Learning

Can the mind be modeled by or seen as implementing deep RL?

  • Ma, W. J., & Peters, B. (2020). A neural network walks into a lab: Towards using deep nets as models for human behavior. arXiv preprint arXiv:2005.02181. (link)
  • van Opheusden, B., & Ma, W. J. (2019). Tasks for aligning human and machine planning. Current Opinion in Behavioral Sciences, 29, 127-133. (PDF)
  • Gershman, S.J. (2019). The generative adversarial brain. Frontiers in Artificial Intelligence, 2, 18. (PDF)
  • Hamrick, J. B. (2019). Analogues of mental simulation and imagination in deep learning. Current Opinion in Behavioral Sciences, 29, 8-16. (PDF)
  • Gershman, S. J., & Uchida, N. (2019). Believing in dopamine. Nature Reviews Neuroscience, 20(11), 703-714. (PDF)
  • Tuli, S., Dasgupta, I., Grant, E., & Griffiths, T. L. (2021). Are Convolutional Neural Networks or Transformers more like human vision? arXiv preprint arXiv:2105.07197. (PDF)