CoChat

A foundation model to predict and capture human cognition

Open access

The DOI matches its Crossref record; 65 of the 65 references it cites were found in a registry or an index.

Abstract

Establishing a unified theory of cognition has been an important goal in psychology 1,2 . A first step towards such a theory is to create a computational model that can predict human behaviour in a wide range of settings. Here we introduce Centaur, a computational model that can predict and simulate human behaviour in any experiment expressible in natural language. We derived Centaur by fine-tuning a state-of-the-art language model on a large-scale dataset called Psych-101. Psych-101 has an unprecedented scale, covering trial-by-trial data from more than 60,000 participants performing in excess of 10,000,000 choices in 160 experiments. Centaur not only captures the behaviour of held-out participants better than existing cognitive models, but it also generalizes to previously unseen cover stories, structural task modifications and entirely new domains. Furthermore, the model’s internal representations become more aligned with human neural activity after fine-tuning. Taken together, our results demonstrate that it is possible to discover computational models that capture human behaviour across a wide range of domains. We believe that such models provide tremendous potential for guiding the development of cognitive theories, and we present a case study to demonstrate this.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41586-025-09215-4

Cites

65 of 65 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Overview of Psych-101 and Centaur. a, Psych-101 comprises trial-by-trial data from 160 psychological experiments with 60,092 participants making 10,681,650 choices in total and involving 253,597,411 text tokens. It contains domains such as multi-armed bandits, decision-making, memory, supervised learning, Markov decision processes and others (the examples shown have been stylized and abbreviated for readability). b, Centaur is a foundation of model human cognition that is obtained by adding low-rank adapters to a state-of-the-art language model and fine-tuning it on Psych-101.
Fig. 1 Overview of Psych-101 and Centaur. a, Psych-101 comprises trial-by-trial data from 160 psychological experiments with 60,092 participants making 10,681,650 choices in total and involving 253,597,411 text tokens. It contains domains such as multi-armed bandits, decision-making, memory, supervised learning, Markov decision processes and others (the examples shown have been stylized and abbreviated for readability). b, Centaur is a foundation of model human cognition that is obtained by adding low-rank adapters to a state-of-the-art language model and fine-tuning it on Psych-101.CC BY · PubMed Central
Fig. 2 Goodness-of-fit on Psych-101. a, Difference in log-likelihood of Centaur and Llama relative to a domain-specific cognitive model for each experiment. A value of zero corresponds to the goodness-of-fit of the domain-specific cognitive model and a value above zero indicates improved goodness-of-fit to human responses. Log-likelihoods are averaged over responses (n = 992,867). Error bars correspond to the standard error of the mean. Centaur outperforms both Llama and a collection of domain-specific cognitive models in almost every experiment (one-sided t-tests: t(1,985,732) = −144.22, P ≤ 0.0001; t(1,985,732) = −127.58, P ≤ 0.0001, respectively). We only included experiments for which we have implemented a domain-specific cognitive model in this graphic and merged different studies using the same paradigm. Extended Data Table 1 contains numerical results for all experiments. b, Model simulations on the horizon task. The plot shows the probability densities over reward and an information bonus parameter for both people and simulated runs of Centaur. c, Model simulations on the two-step task. The plot shows the probability densities over reward and a parameter indicating how model-based learning was for both people and simulated runs of Centaur. d, Model simulations on a social prediction game. The plot shows the probability densities over accuracies of predicting human strategies and strategies of an artificial agent, with matched statistics for both people and simulated runs of Centaur.
Fig. 2 Goodness-of-fit on Psych-101. a, Difference in log-likelihood of Centaur and Llama relative to a domain-specific cognitive model for each experiment. A value of zero corresponds to the goodness-of-fit of the domain-specific cognitive model and a value above zero indicates improved goodness-of-fit to human responses. Log-likelihoods are averaged over responses (n = 992,867). Error bars correspond to the standard error of the mean. Centaur outperforms both Llama and a collection of domain-specific cognitive models in almost every experiment (one-sided t-tests: t(1,985,732) = −144.22, P ≤ 0.0001; t(1,985,732) = −127.58, P ≤ 0.0001, respectively). We only included experiments for which we have implemented a domain-specific cognitive model in this graphic and merged different studies using the same paradigm. Extended Data Table 1 contains numerical results for all experiments. b, Model simulations on the horizon task. The plot shows the probability densities over reward and an information bonus parameter for both people and simulated runs of Centaur. c, Model simulations on the two-step task. The plot shows the probability densities over reward and a parameter indicating how model-based learning was for both people and simulated runs of Centaur. d, Model simulations on a social prediction game. The plot shows the probability densities over accuracies of predicting human strategies and strategies of an artificial agent, with matched statistics for both people and simulated runs of Centaur.CC BY · PubMed Central
Fig. 3 Evaluation in different held-out settings. a, Negative log-likelihoods averaged over responses (n = 9,702) for the two-step task with a modified cover story23. b, Negative log-likelihoods averaged over responses (n = 510,154) for a three-armed bandit experiment25. c, Negative log-likelihoods averaged over responses (n = 99,204) for an experiment probing logical reasoning26 with items based on the Law School Admission Test (LSAT). Centaur outperforms both Llama and domain-specific cognitive models when faced with modified cover stories, problem structures and entirely new domains. N/A, not applicable. Error bars show the s.e.m. The image in a is reproduced from ref. 23, Springer Nature Limited. The image in c is reproduced from Wikipedia.org.
Fig. 3 Evaluation in different held-out settings. a, Negative log-likelihoods averaged over responses (n = 9,702) for the two-step task with a modified cover story23. b, Negative log-likelihoods averaged over responses (n = 510,154) for a three-armed bandit experiment25. c, Negative log-likelihoods averaged over responses (n = 99,204) for an experiment probing logical reasoning26 with items based on the Law School Admission Test (LSAT). Centaur outperforms both Llama and domain-specific cognitive models when faced with modified cover stories, problem structures and entirely new domains. N/A, not applicable. Error bars show the s.e.m. The image in a is reproduced from ref. 23, Springer Nature Limited. The image in c is reproduced from Wikipedia.org.CC BY · PubMed Central
Fig. 4 Human alignment. a, Multidimensional scaling embedding of the ten behavioural metrics in CogBench33 for different models. b, Pearson correlation coefficients indicating how well human neural activity in the two-step task37 can be decoded using Centaur’s internal representations extracted from different layers. c, Pearson correlation coefficients indicating how well human neural activity in a sentence-reading task38 can be decoded using Centaur’s internal representations extracted from different layers. Control refers to a model that used representations extracted from a randomly initialized transformer model with matched architecture.
Fig. 4 Human alignment. a, Multidimensional scaling embedding of the ten behavioural metrics in CogBench33 for different models. b, Pearson correlation coefficients indicating how well human neural activity in the two-step task37 can be decoded using Centaur’s internal representations extracted from different layers. c, Pearson correlation coefficients indicating how well human neural activity in a sentence-reading task38 can be decoded using Centaur’s internal representations extracted from different layers. Control refers to a model that used representations extracted from a randomly initialized transformer model with matched architecture.CC BY · PubMed Central
Fig. 5 Model-guided scientific discovery. a, We used Psych-101 and Centaur to guide the development of a cognitive model for a multi-attribute decision-making study41. Each panel shows the AIC for the set of models considered at the given stage, starting with the models considered in the original study. b, We asked DeepSeek-R1 to generate an explanation for the human responses and formalized the resulting verbal strategy into a formal computational model. c, We refined this model through scientific regret minimization using Centaur as a reference model. Six data points are shown for which Centaur makes accurate predictions but the DeepSeek-R1-discovered model does not. We then used this information to design a domain-specific cognitive model that is as predictive as Centaur but is still interpretable. The bicycle images in a are reproduced from Flaticon.com.
Fig. 5 Model-guided scientific discovery. a, We used Psych-101 and Centaur to guide the development of a cognitive model for a multi-attribute decision-making study41. Each panel shows the AIC for the set of models considered at the given stage, starting with the models considered in the original study. b, We asked DeepSeek-R1 to generate an explanation for the human responses and formalized the resulting verbal strategy into a formal computational model. c, We refined this model through scientific regret minimization using Centaur as a reference model. Six data points are shown for which Centaur makes accurate predictions but the DeepSeek-R1-discovered model does not. We then used this information to design a domain-specific cognitive model that is as predictive as Centaur but is still interpretable. The bicycle images in a are reproduced from Flaticon.com.CC BY · PubMed Central
Extended Data Fig. 1 Psych-101. a, Proportion of domains included in Psych-101. b, Word cloud of experimental paradigms included in Psych-101. c, We performed a data contamination analysis using the LogProber method67 for every experimental paradigm in Psych-101. LogProber fits a two-parameter exponential model to the cumulative log-likelihood of each sequence being checked for contamination. High acceleration (log B) suggests that a prompt is memorized from the pretraining data. Following the results presented in the original work67, we set a threshold for possible contamination to log B ≥ 1. This analysis indicated no evidence of contamination. d, Two-dimensional embedding of the experiments used in this paper. To obtain this embedding, we took the corresponding natural language prompts up to the point of the first human choice, extracted a vector-based representation for them using ModernBERT68, and finally projected these representations onto two dimensions using multidimensional scaling. Purple dots correspond to experiments from Psych-101, whereas the colored dots correspond to the indicated evaluation experiment.
Extended Data Fig. 1 Psych-101. a, Proportion of domains included in Psych-101. b, Word cloud of experimental paradigms included in Psych-101. c, We performed a data contamination analysis using the LogProber method67 for every experimental paradigm in Psych-101. LogProber fits a two-parameter exponential model to the cumulative log-likelihood of each sequence being checked for contamination. High acceleration (log B) suggests that a prompt is memorized from the pretraining data. Following the results presented in the original work67, we set a threshold for possible contamination to log B ≥ 1. This analysis indicated no evidence of contamination. d, Two-dimensional embedding of the experiments used in this paper. To obtain this embedding, we took the corresponding natural language prompts up to the point of the first human choice, extracted a vector-based representation for them using ModernBERT68, and finally projected these representations onto two dimensions using multidimensional scaling. Purple dots correspond to experiments from Psych-101, whereas the colored dots correspond to the indicated evaluation experiment.CC BY · PubMed Central
Extended Data Fig. 2 Negative log-likelihoods of Centaur and alternative Llama variants on Psych-101. To rule out the hypothesis that finetuning on any data aligns a model with human behavior, we compared Centaur to various Llama variants finetuned for other purposes (i.e. non-cognitive tasks). Nemotron69 is finetuned for instruction-following. Hermes70 is finetuned for various purposes, including agentic capabilities, roleplaying, reasoning, multi-turn conversation, and long context coherence. Reflection is finetuned for reasoning. None of the Llama variants captures human behavior better than the base model, ruling out the hypothesis that finetuning generally leads to models that are better at predicting human behavior. Error bars correspond to the standard error of the mean, taken over responses.
Extended Data Fig. 2 Negative log-likelihoods of Centaur and alternative Llama variants on Psych-101. To rule out the hypothesis that finetuning on any data aligns a model with human behavior, we compared Centaur to various Llama variants finetuned for other purposes (i.e. non-cognitive tasks). Nemotron69 is finetuned for instruction-following. Hermes70 is finetuned for various purposes, including agentic capabilities, roleplaying, reasoning, multi-turn conversation, and long context coherence. Reflection is finetuned for reasoning. None of the Llama variants captures human behavior better than the base model, ruling out the hypothesis that finetuning generally leads to models that are better at predicting human behavior. Error bars correspond to the standard error of the mean, taken over responses.CC BY · PubMed Central
Extended Data Fig. 3 Noise ceiling analysis. We conducted a noise ceiling analysis to better understand the capabilities of Centaur. It is not straightforward to estimate the noise ceiling for experiments with sequential dependencies, which includes the majority of Psych-101. Hence, we focused on two experiments for which such an analysis is possible: a, the choices13k data set18 and b, an intertemporal choice experiment52. In both cases, we found that Centaur substantially exceeds the estimated noise ceiling. This is possible because Centaur can pick up on context-dependent patterns that are not captured by standard noise ceiling analyses. Therefore, we have performed an additional analysis testing how well Centaur can predict human responses if we prompt it to predict each response independently. We use the suffix “ind.” to indicate this way of prompting the model. Centaur still matches the performance of domain-specific cognitive models when context-independent prompts are used, amounting to roughly half of the estimated noise ceiling.
Extended Data Fig. 3 Noise ceiling analysis. We conducted a noise ceiling analysis to better understand the capabilities of Centaur. It is not straightforward to estimate the noise ceiling for experiments with sequential dependencies, which includes the majority of Psych-101. Hence, we focused on two experiments for which such an analysis is possible: a, the choices13k data set18 and b, an intertemporal choice experiment52. In both cases, we found that Centaur substantially exceeds the estimated noise ceiling. This is possible because Centaur can pick up on context-dependent patterns that are not captured by standard noise ceiling analyses. Therefore, we have performed an additional analysis testing how well Centaur can predict human responses if we prompt it to predict each response independently. We use the suffix “ind.” to indicate this way of prompting the model. Centaur still matches the performance of domain-specific cognitive models when context-independent prompts are used, amounting to roughly half of the estimated noise ceiling.CC BY · PubMed Central
Extended Data Fig. 4 Further out-of-distribution evaluations. Each subplot shows negative log-likelihoods for a different experiment. None of these paradigms were included in Psych-101, hence they provide a stress test for a model’s generalization capabilities. Centaur robustly captured human behavior in all of these settings, while smaller and non-finetuned models did not do so consistently. Error bars correspond to the standard error of the mean, taken over responses. We state one-sided t-tests comparing the negative log-likelihoods of Centaur to those of Llama in brackets. a, Negative log-likelihoods on moral decision-making27 (t(181388) = −103.54, p ≤ 0.0001). b, Negative log-likelihoods on economic games1 (t(7798) = −11.69, p ≤ 0.0001). c, Negative log-likelihoods on naturalistic category learning1 (t(21838) = −14.05, p ≤ 0.0001). d, Negative log-likelihoods on behavioral propensities30 (t(156230) = −11.06, p ≤ 0.0001). e, Negative log-likelihoods on naturalistic reward learning1 (t(9838) = −12.63, p ≤ 0.0001). f, Negative log-likelihoods on a deep sequential decision task31 (t(6092) = −1.06, p = 0.144).
Extended Data Fig. 4 Further out-of-distribution evaluations. Each subplot shows negative log-likelihoods for a different experiment. None of these paradigms were included in Psych-101, hence they provide a stress test for a model’s generalization capabilities. Centaur robustly captured human behavior in all of these settings, while smaller and non-finetuned models did not do so consistently. Error bars correspond to the standard error of the mean, taken over responses. We state one-sided t-tests comparing the negative log-likelihoods of Centaur to those of Llama in brackets. a, Negative log-likelihoods on moral decision-making27 (t(181388) = −103.54, p ≤ 0.0001). b, Negative log-likelihoods on economic games1 (t(7798) = −11.69, p ≤ 0.0001). c, Negative log-likelihoods on naturalistic category learning1 (t(21838) = −14.05, p ≤ 0.0001). d, Negative log-likelihoods on behavioral propensities30 (t(156230) = −11.06, p ≤ 0.0001). e, Negative log-likelihoods on naturalistic reward learning1 (t(9838) = −12.63, p ≤ 0.0001). f, Negative log-likelihoods on a deep sequential decision task31 (t(6092) = −1.06, p = 0.144).CC BY · PubMed Central
Extended Data Fig. 5 metabench and CogBench results. a, Results for metabench34, a sparse benchmark containing several canonical benchmarks from the machine learning literature. We find that Centaur maintains the level of performance of Llama, indicating that finetuning on human behavior did not lead to deterioration in other tasks (ARC: z = −0.126, p = 0.9, GSM8K: z = −0.529, p = 0.597, HellaSwag: z = 0.0, p = 1.0, MMLU: z = 0.0, p = 1.0, Winogrande: z = −0.556, p = 0.578). Performance on TruthfulQA71 – which measures how models mimic human falsehoods – even improved significantly with finetuning (z = 2.312, p = 0.021; all z-test were two-sided). b, Performance-based metrics from CogBench33, a benchmark that includes ten behavioral metrics derived from seven cognitive psychology experiments. We find that – relative to Llama – Centaur’s performance improves in all experiments (Probabilistic reasoning: z = 6.371, p ≤ 0.0001, Horizon task: z = 22.176, p ≤ 0.0001, Restless bandit: z = 7.317, p ≤ 0.0001, Instrumental learning: z = 0.126, p = 0.45, Two-step task: z = 1.458, p = 0.072, Balloon analog risk task: z = 1.496, p = 0.067; all z-test were one-sided). c, Behavioral metrics from CogBench. We observe that Centaur becomes more similar to human subjects in all ten behavioral metrics (Prior weighting: z = 2.176, p = 0.015, Likelihood weighting: z = 1.131, p = 0.129, Directed exploration: z = 0.525, p = 0.3, Random exploration: z = 2.014, p = 0.022, Meta-cognition: z = 2.206, p = 0.014, Learning rate: z = 0.477, p = 0.317, Optimism bias: z = 0.78, p = 0.218, Model-basedness: z = 9.608, p ≤ 0.0001, Temporal discounting: z = 2.594, p = 0.005, Risk taking: z = 1.612, p = 0.053; all z-test were one-sided).
Extended Data Fig. 5 metabench and CogBench results. a, Results for metabench34, a sparse benchmark containing several canonical benchmarks from the machine learning literature. We find that Centaur maintains the level of performance of Llama, indicating that finetuning on human behavior did not lead to deterioration in other tasks (ARC: z = −0.126, p = 0.9, GSM8K: z = −0.529, p = 0.597, HellaSwag: z = 0.0, p = 1.0, MMLU: z = 0.0, p = 1.0, Winogrande: z = −0.556, p = 0.578). Performance on TruthfulQA71 – which measures how models mimic human falsehoods – even improved significantly with finetuning (z = 2.312, p = 0.021; all z-test were two-sided). b, Performance-based metrics from CogBench33, a benchmark that includes ten behavioral metrics derived from seven cognitive psychology experiments. We find that – relative to Llama – Centaur’s performance improves in all experiments (Probabilistic reasoning: z = 6.371, p ≤ 0.0001, Horizon task: z = 22.176, p ≤ 0.0001, Restless bandit: z = 7.317, p ≤ 0.0001, Instrumental learning: z = 0.126, p = 0.45, Two-step task: z = 1.458, p = 0.072, Balloon analog risk task: z = 1.496, p = 0.067; all z-test were one-sided). c, Behavioral metrics from CogBench. We observe that Centaur becomes more similar to human subjects in all ten behavioral metrics (Prior weighting: z = 2.176, p = 0.015, Likelihood weighting: z = 1.131, p = 0.129, Directed exploration: z = 0.525, p = 0.3, Random exploration: z = 2.014, p = 0.022, Meta-cognition: z = 2.206, p = 0.014, Learning rate: z = 0.477, p = 0.317, Optimism bias: z = 0.78, p = 0.218, Model-basedness: z = 9.608, p ≤ 0.0001, Temporal discounting: z = 2.594, p = 0.005, Risk taking: z = 1.612, p = 0.053; all z-test were one-sided).CC BY · PubMed Central
Extended Data Fig. 6 Finegrained neural alignment results in the two-step task. a, Pearson correlation coefficients between the predicted activity from Centaur’s representations and the BOLD data shown on a surface brain (image created with nilearn66). Centaur achieves the most accurate predictions in the left motor cortex. As participants performed the task with their right hand in the scanner, this effect may be explained by Centaur’s strong performance in predicting choices. b, Predictive performance of Centaur’s representations against alternatives for ROIs that have been identified as behaviorally relevant in previous work. Cortical scores are averaged over the corresponding bilateral parcels in the Schaefer atlas. The accumbens is defined based on the Harvard-Oxford atlas. Pearson correlation coefficients are shown for layer 20 but exhibit a similar pattern across all layers. Centaur outperformed Llama and the cognitive model in predicting activity in accumbens, the ROI from the original study that showed a reward prediction error effect19,37. We found a similar pattern in the medial PFC, another region that showed an effect in the original article37, as well as in the sensory and motor cortices.
Extended Data Fig. 6 Finegrained neural alignment results in the two-step task. a, Pearson correlation coefficients between the predicted activity from Centaur’s representations and the BOLD data shown on a surface brain (image created with nilearn66). Centaur achieves the most accurate predictions in the left motor cortex. As participants performed the task with their right hand in the scanner, this effect may be explained by Centaur’s strong performance in predicting choices. b, Predictive performance of Centaur’s representations against alternatives for ROIs that have been identified as behaviorally relevant in previous work. Cortical scores are averaged over the corresponding bilateral parcels in the Schaefer atlas. The accumbens is defined based on the Harvard-Oxford atlas. Pearson correlation coefficients are shown for layer 20 but exhibit a similar pattern across all layers. Centaur outperformed Llama and the cognitive model in predicting activity in accumbens, the ROI from the original study that showed a reward prediction error effect19,37. We found a similar pattern in the medial PFC, another region that showed an effect in the original article37, as well as in the sensory and motor cortices.CC BY · PubMed Central
Extended Data Fig. 7 Log-likelihood comparison between Centaur and Minitaur on the analyses from the main text. a, Negative log-likelihoods relative to the domain-specific cognitive models on held-out participants from Psych-101. Error bars correspond to the standard error of the mean, taken over responses. b, Negative log-likelihoods for the two-step task with a modified cover story. c, Negative log-likelihoods for a three-armed bandit experiment. d, Negative log-likelihoods for an experiment probing logical reasoning with items based on the Law School Admission Test (LSAT).
Extended Data Fig. 7 Log-likelihood comparison between Centaur and Minitaur on the analyses from the main text. a, Negative log-likelihoods relative to the domain-specific cognitive models on held-out participants from Psych-101. Error bars correspond to the standard error of the mean, taken over responses. b, Negative log-likelihoods for the two-step task with a modified cover story. c, Negative log-likelihoods for a three-armed bandit experiment. d, Negative log-likelihoods for an experiment probing logical reasoning with items based on the Law School Admission Test (LSAT).CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 65 references

Research works printed but not linked: 10

  1. Anderson, J. The Architecture of Cognition (Harvard Univ. Press, 1983).
  2. Newell, A. Unified Theories of Cognition (Harvard Univ. Press, 1990).
  3. Binz, M. & Schulz, E. Turning large language models into cognitive models. In Proc. 12th International Conference on Learning Representations (ICLR, 2024).
  4. Dettmers, T., Pagnoni, A., Holtzman, A. & Zettlemoyer, L. QLORA: efficient finetuning of quantized LLMs. In Proc. Advances in Neural Information Processing Systems 36 (eds Oh, A. et al.) (NeurIPS, 2023).
  5. Demircan, C. et al. Evaluating alignment between humans and neural network representations in image-based learning tasks. In Proc. Advances in Neural Information Processing Systems 37 (eds Globerson, A. et al.) (NeurIPS, 2024).
  6. Kipnis, A., Voudouris, K., Schulze Buschoff, L. M. & Schulz, E. metabench - a sparse benchmark of reasoning and knowledge in large language models. In Proc. 13th International Conference on Learning Representations (ICLR, 2025).
  7. Huben, R., Cunningham, H., Smith, L. R., Ewart, A. & Sharkey, L. Sparse autoencoders find highly interpretable features in language models. In Proc. 12th International Conference on Learning Representations (ICLR, 2024).
  8. Loshchilov, I. & Hutter, F. Decoupled weight decay regularization. In Proc. 7th International Conference on Learning Representations (ICLR, 2019).
  9. Friston, K. J., Ashburner, J. T., Kiebel, S. J., Nichols, T. E. & Penny, W. D. (eds) Statistical Parametric Mapping: The Analysis of Functional Brain Images (Elsevier, 2011).
  10. Teknium, R., Quesnelle, J. & Guang, C. Hermes 3 technical report. Preprint at https://arxiv.org/abs/2408.11857 (2024). · arXiv

Cited by

See all 339

Metadata

Authors
Binz, M., Akata, E., Bethge, M., Brändle, F., Callaway, F., Coda-Forno, J., Dayan, P., Demircan, C., Eckstein, M. K., Éltető, N., Griffiths, T. L., Haridi, S., Jagadish, A. K., Ji-An, L., Kipnis, A., Kumar, S., Ludwig, T., Mathony, M., Mattar, M., Modirshanechi, A., Nath, S. S., Peterson, J. C., Rmus, M., Russek, E. M., Saanum, T., Schubert, J. A., Schulze Buschoff, L. M., Singhi, N., Sui, X., Thalmann, M., Theis, F. J., Truong, V., Udandarao, V., Voudouris, K., Wilson, R., Witte, K., Wu, S., Wulff, D. U., Xiong, H., & Schulz, E.
Affiliations
  • Helmholtz Munich (DE)
  • University of Tübingen (DE)
  • University of Oxford (GB)
  • New York University (US)
  • Max Planck Institute for Biological Cybernetics (DE)
  • Google DeepMind (United Kingdom) (GB)
  • Princeton University (US)
  • University of California San Diego (US)
  • Boston University (US)
  • Technische Universität Darmstadt (DE)
  • University of Cambridge (GB)
  • Georgia Institute of Technology (US)
  • University of Basel (CH)
Corresponding author
Binz, Marcel
Venue
Nature
Published
2025-07-02
Volume
644
Issue
8078
Pages
1002-1009
ISSN
  • 0028-0836 (ISSN-L)
  • 1476-4687
OpenAlex
W4411932262
Cited by
119
References
57
Citations by year
2026: 84 · 2025: 34
FWCI
66.94
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • MPG.PuRe (Max Planck Society) (submittedVersion, cc-by-nd)
  • Apollo (University of Cambridge) (submittedVersion, cc-by)
  • MPG.PuRe (Max Planck Society) (submittedVersion)
Funders
  • DeepMind
  • Volkswagen Foundation
  • Max-Planck-Gesellschaft
  • NOMIS Stiftung
  • Helmholtz Zentrum München
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Cognitive Science and Education Research
  • Social Sciences › Decision Sciences › General Decision Sciences › Decision-Making and Behavioral Economics
  • Physical Sciences › Computer Science › Artificial Intelligence › Psychiatry, Mental Health, Neuroscience
Keywords
centaurs, computational cognitive model, human behavior prediction, language model fine-tuning, generalization to unseen tasks
MeSH terms
Choice Behavior, Cognition, Computer Simulation, Humans, Language, Male, Models, Psychological

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: A foundation model to predict and capture human cognition. https://app.cochat.ai/papers/10.1038/s41586-025-09215-4
The paper
Binz, M., Akata, E., Bethge, M., Brändle, F., Callaway, F., Coda-Forno, J., Dayan, P., Demircan, C., Eckstein, M. K., Éltető, N., Griffiths, T. L., Haridi, S., Jagadish, A. K., Ji-An, L., Kipnis, A., Kumar, S., Ludwig, T., Mathony, M., Mattar, M., . . . Schulz, E. (2025). A foundation model to predict and capture human cognition. Nature. https://doi.org/10.1038/s41586-025-09215-4

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.