The DOI matches its Crossref record; 187 of the 188 references it cites were found in a registry or an index, and 1 are not checked yet.
When Can AI Models Explain Learning? Validity Criteria for AI as Cognitive Models in Education
Open accessAbstract
Traditional verbal theories in educational psychology often remain underspecified at the mechanistic level. While they offer rich descriptive constructs and conceptual insights, they provide limited accounts of how learning processes unfold dynamically and causally. Here, “verbal” denotes theories stated in prose and qualitative relations rather than as formal, computational process models. This lack of mechanistic precision constrains rigorous theory testing, limits integration across levels of analysis, and reduces the potential to design interventions grounded in explanatory understanding of learning processes. In this Review, we synthesize an emerging paradigm that treats artificial intelligence (AI) systems not merely as predictive tools or instructional technologies, but as cognitive models of learners, explicit, runnable instantiations of theoretical assumptions about cognition and learning. Our central question is not whether AI systems can serve as cognitive models, but when they should be allowed to count as such. We therefore organize the Review around explicit validity criteria, theoretical grounding, construct validity, mechanistic transparency, alignment with human learning trajectories, error-signature matching, causal-intervention tests, ecological validity, and instructional usefulness, that an AI system must satisfy before its cognitive-model status is granted rather than assumed. We examine how major families of AI models, including neural networks, reinforcement learning agents, cognitive architectures, and large language models, have been used to operationalize core educational constructs such as memory, strategy use, motivation, self-regulation, and social learning. Across these approaches, we highlight how mechanistic transparency, interpretability, and alignment with human learning trajectories and error patterns are essential for explanatory validity. We further discuss methodological tools, such as representation analysis, ablation, and trajectory-level comparison, that enable causal inference about learning mechanisms within models. Finally, we outline key challenges and future directions, including construct validity, ecological realism, individual differences, and ethical accountability. By positioning AI as a theoretical instrument rather than solely an engineering solution, this Review argues that AI-based cognitive models can, when they satisfy these criteria, help transform abstract learning theories into precise, testable, and educationally actionable accounts of how students learn.
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.
Integrity signals
Registry
Cites
Statistics
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.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“A further distinction, developed for neural models of cognition more broadly, separates failures that depend on a model's current operating regime from failures that reflect its architectural and representational capacity [50,51].”
From Hallucination as epistemic pollution: epistemic injustice and credibility in futures-oriented science education (2026) · cites this paper as [51]
Experimental — accuracy not yet measured
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
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404 · cited 1×
“Neural networks have a long history in cognitive modeling, from early connectionist work in the 1980 s and 1990 s (Rumelhart et al., 1986 ) to modern deep learning architectures such as recurrent networks (Ji-An et al., 2025 ) and Transformers (Vaswani et al., 2017 ).”
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Ryan, R. M., & Deci, E. L. (2000) · American Psychologist · doi:10.1037//0003-066x.55.1.68 · Cited by 54,599 · cited 1×
“…well as non-cognitive constructs, such as working memory (Baddeley, 1992 ), learning strategies (Mayer, 2002 ), metacognition (Flavell, 1979 ), motivation (Ryan & Deci, 2000 ), self-efficacy (Bandura, 1997 ), and emotions (Pekrun et al., 2011 ) that are used to explain how students think and learn.”
Reinforcement Learning: An Introduction
Sutton, R. S., & Barto, A. G. (1998) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.1998.712192 · Cited by 27,225
Parallel Distributed Processing: Explorations in the Microstructure of Cognition: Foundations
Rumelhart, D. E., McClelland, J. L., & AU (1986) · doi:10.7551/mitpress/5236.001.0001 · Cited by 15,686
Metacognition and cognitive monitoring: A new area of cognitive–developmental inquiry
Flavell, J. H. (1979) · American Psychologist · doi:10.1037/0003-066x.34.10.906 · Cited by 11,282 · cited 1×
“…set of cognitive as well as non-cognitive constructs, such as working memory (Baddeley, 1992 ), learning strategies (Mayer, 2002 ), metacognition (Flavell, 1979 ), motivation (Ryan & Deci, 2000 ), self-efficacy (Bandura, 1997 ), and emotions (Pekrun et al., 2011 ) that are used to explain how…”
Elman, J. L. (1990) · Cognitive Science · doi:10.1207/s15516709cog1402_1 · Cited by 11,212 · cited 1×
“Work in the 1980 s and 1990 s extended this program with symbolic production-rule systems and early neural networks to simulate how skills are acquired and how language develops (Elman, 1990 ; Laird et al., 1987 ).”
Rudin, C. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0048-x · Cited by 10,561 · cited 2×
“In this context, interpret - ability requires two steps: (a) identifying what information internal variables carry and (b) demonstrating a causal role in producing behavior, not mere correlation (Lip - ton, 2018; Rudin, 2019 ).”
Estimating the reproducibility of psychological science
Open Science Collaboration (2015) · Science · doi:10.1126/science.aac4716 · Cited by 8,515 · cited 2×
“These constructs are often presented as static categories, supported mainly by discursive argumentation or correlational evidence (Open Science Collaboration, 2015 ), rather than as a dynamic, complex system whose interactions unfold over time and can be instantiated, observed, and evaluated at each step of a learning…”
ChatGPT for good? On opportunities and challenges of large language models for education
Kasneci, E., Sessler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., Gasser, U., Groh, G., Günnemann, S., Hüllermeier, E., et al. (2023) · Learning and Individual Differences · doi:10.1016/j.lindif.2023.102274 · Cited by 7,326 · cited 2×
“…pos - sibility for educational cognitive modeling: a single model that can perform a wide variety of language-mediated tasks that resemble classroom activity, including reading, explaining, generating feedback, and participating in dialogue (Baidoo-Anu & Owusu Ansah, 2023; Kasneci et al., 2023 ).”
Research works printed but not linked: 24
- Aleven, V. A., & Koedinger, K. R. (2002). An effective metacognitive strategy: Learning by doing and explaining with a computer-based cognitive tutor. Cognitive Science, 26(2), 147–179.
- Anderson, J. R., & Schunn, C. D. (2000). Implications of the ACT-R learning theory: No magic bullets. In R. Glaser (Ed.), Advances in instructional psychology: Educational design and cognitive science (Vol. Vol. 5, pp. 1–33). Lawrence Erlbaum Associates Publishers.
- Bandura, A. (1997). Self-efficacy: The exercise of control (pp. ix–604). W H Freeman/Times Books/ Henry Holt & Co.
- Breiman, L. (2003). Statistical modeling: The two cultures. Quality Control and Applied Statistics, 48(1), 81–82.
- Chi, M. T., De Leeuw, N., Chiu, M. H., & LaVancher, C. (1994). Eliciting self-explanations improves understanding. Cognitive Science, 18(3), 439–477.
- Conati, C., Porayska-Pomsta, K., & Mavrikis, M. (2018). AI in Education needs interpretable machine learning: Lessons from Open Learner Modelling. arXiv.Org. https://arxiv.org/abs/1807.00154v1 · arXiv
- Dziri, N., Lu, X., Sclar, M., L Li, X., Jiang, L., Y Lin, B., West, P., Bhagavatula, C., Le Bras, R., D Hwang, J., Sanyal, S., Welleck, S., Ren, X., Ettinger, A., Harchaoui, Z., & Choi, Y. (2023). Faith and fate: Limits of transformers on compositionality. Proceedings of the 37th International Conference on Neural Information Processing Systems, NIPS ’23, 70293–70332.
- Elhage, N., Nanda, N., Olsson, C., Henighan, T., Joseph, N., Mann, B., Askell, A., Bai, Y., Chen, A., & Conerly, T. (2021). A mathematical framework for transformer circuits. Transformer Circuits Thread, 1(1), 12.
- Gal, Y., & Ghahramani, Z. (2016). Dropout as a bayesian approximation: Representing model uncertainty in deep learning. International Conference on Machine Learning, 1050–1059.
- Graves, A., Wayne, G., & Danihelka, I. (2014). Neural Turing Machines. arXiv.Org. https://arxiv.org/abs/1410.5401v2 · arXiv
- Kharchenko, J., Roosta, T., Chadha, A., & Shah, C. (2025). How well do LLMs represent values across cultures? Empirical analysis of LLM responses based on Hofstede cultural dimensions. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '25). Association for Computing Machinery.
- MacLellan, C. J., Harpstead, E., Marinier, I. I. I., R. P., & Koedinger, K. R. (2018). A framework for natural cognitive system training interactions. Advances in Cognitive Systems, 6, 1–16.
- Messick, S. (1989). Validity. In R. L. Linn (Ed.), Educational measurement (3rd ed, pp. 13–103). American Council on Education.
- Newell, A., & Simon, H. A. (1972). Human problem solving (Vol. 104, Number 9). Prentice-hall Englewood Cliffs, NJ.
- Newell, A., Shaw, J. C., & Simon, H. A. (1959). Report on a general problem solving program. IFIP Congress, 256(1), 1959.
- Piech, C., Bassen, J., Huang, J., Ganguli, S., Sahami, M., Guibas, L. J., & Sohl-Dickstein, J. (2015). Deep knowledge tracing. Advances in Neural Information Processing Systems, 28.
- Rose, T. (2016). The end of average: How to succeed in a world that values sameness. Penguin UK.
- Selwyn, N. (2019). Should robots replace teachers? AI and the future of education. John Wiley & Sons.
- Singh, S., Lewis, R., & Barto, A. (2009). Where Do Rewards Come From? In Proceedings of the 31st Annual Meeting of the Cognitive Science Society.
- Sun, R. (2006). Cognition and multi-agent interaction: From cognitive modeling to social simulation. Cambridge University Press.
- Sun, R. (2008). The Cambridge handbook of computational psychology. Cambridge University Press.
- Vanlehn, K., Ohlsson, S., & Nason, R. (1998). Applications of Simulated Students: An Exploration. Journal of Artificial Intelligence in Education, 5(2), 135–175.
- Winne, P. H., & Hadwin, A. F. (1998). Studying as self-regulated learning. In D. J. Hacker, J. Dunlosky, & A. C. Graesser (Eds.), Metacognition in educational theory and practice (pp. 277–304). Lawrence Erlbaum Associates Publishers.
- Yudkowsky, E. (2016). The AI alignment problem: Why it is hard, and where to start. Symbolic Systems Distinguished Speaker, 4(1) , 1–20.
Cited by
Küçükuncular, A. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-02318-5 · Cited by 1
“5.1 [50, 51], so that the residual risk does not fall to zero as the tooling improves.”
InfluentialZhang, Z., Jiang, C., & Shen, Y. (2026) · Global Education Bulletin · doi:10.71052/grb2025/tjtu6716
Fronzi, M. (2026) · Cognitive Computation · doi:10.1007/s12559-026-10668-6
“Drawing on criteria recently articulated for computational models in cognitive science [39], a genuinely explanatory claim requires at minimum: (a) construct validity — the model’s latent variables must correspond to mea - surable empirical constructs; (b) trajectory correspondence — not only…”
Metadata
- Authors
- Wang, P., Viberg, O., Law, E. L.-C., & Greiff, S.
- Affiliations
- University of Liège (BE)
- Digital Futures (SE)
- Durham University (GB)
- Technical University of Munich (DE)
- Corresponding author
- Wang, Peng
- Venue
- Educational Psychology Review
- Published
- 2026-07-18
- Volume
- 38
- Issue
- 1
- ISSN
- 1040-726X (ISSN-L)
- 1573-336X
- OpenAlex
- W7169613397
- Cited by
- 3
- References
- 172
- Citations by year
- 2026: 3
- FWCI
- 10.25
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- Educational Psychology Review (publishedVersion, cc-by)
- EUR Research Repository (Erasmus University Rotterdam) (publishedVersion, cc-by)
- Durham Research Online (Durham University) (acceptedVersion, other-oa)
- Durham Research Online (Durham University) (submittedVersion, cc-by)
- VU Research Portal (submittedVersion, other-oa)
- Funders
- China Scholarship Council (202208060121)
- Canada Excellence Research Chair in Neuroplasticity (Canada Excellence Research Chair in Neuroplasticity)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Intelligent Tutoring Systems and Adaptive Learning
- Social Sciences › Education › Artificial Intelligence in Education
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Visual and Cognitive Learning Processes
- Keywords
- validity criteria, construct validity, ecological validity, causal inference, large language models, reinforcement learning agents, cognitive architecture, artificial neural networks, memory, strategy use, motivation, self-regulation, social learning, individual differences, ethical accountability, theory testing
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: When Can AI Models Explain Learning? Validity Criteria for AI as Cognitive Models in Education. https://app.cochat.ai/papers/10.1007/s10648-026-10202-y
- The paper
- Wang, P., Viberg, O., Law, E. L.-C., & Greiff, S. (2026). When Can AI Models Explain Learning? Validity Criteria for AI as Cognitive Models in Education. Educational Psychology Review. https://doi.org/10.1007/s10648-026-10202-y
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.