Capturing Humans’ Mental Models of AI: An Item Response Theory Approach
Open accessThe DOI matches its Crossref record; 20 of the 20 references it cites were found in a registry or an index.
Abstract
Improving our understanding of how humans perceive AI teammates is an important foundation for our general understanding of human-AI teams. Extending relevant work from cognitive science, we propose a framework based on item response theory for modeling these perceptions. We apply this framework to real-world experiments, in which each participant works alongside another person or an AI agent in a question-answering setting, repeatedly assessing their teammate’s performance. Using this experimental data, we demonstrate the use of our framework for testing research questions about people’s perceptions of both AI agents and other people. We contrast mental models of AI teammates with those of human teammates as we characterize the dimensionality of these mental models, their development over time, and the influence of the participants’ own self-perception. Our results indicate that people expect AI agents’ performance to be significantly better on average than the performance of other humans, with less variation across different types of problems. We conclude with a discussion of the implications of these findings for human-AI interaction.
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.
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
(2015) · doi:10.1007/s11222-016-9696-4 · Cited by 5,611 · cited 1×
“dimensional model with the multidimensional model using three widely-used statistical model selection techniques: (i) held-out loglikelihood, (ii) WAIC score, and (iii) LOO score [46, 66].”
MethodExplanation in Artificial Intelligence: Insights from the Social Sciences
Miller, T. (2017) · arXiv · doi:10.48550/arxiv.1706.07269 · Cited by 5,338
Task-Dependent Algorithm Aversion
Castelo, N., Bos, M. W., & Lehmann, D. R. (2019) · Journal of Marketing Research · doi:10.1177/0022243719851788 · Cited by 1,500 · cited 1×
“, tasks that involve retrieving factual information, when compared to humans [16, 44].”
BackgroundHuman-computer collaboration for skin cancer recognition
Tschandl, P., Rinner, C., Apalla, Z., Argenziano, G., Codella, N., Halpern, A., Janda, M., Lallas, A., Longo, C., Malvehy, J., et al. (2020) · Nature Medicine · doi:10.1038/s41591-020-0942-0 · Cited by 1,024
Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance
Bansal, G., Wu, T., Zhou, J., Fok, R., Nushi, B., Kamar, E., Ribeiro, M. T., & Weld, D. (2021) · Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411764.3445717 · Cited by 757 · cited 1×
“One important goal for human-AI systems is complementarity— achieving better performance than either the human(s) or the AI agent(s) acting independently [8, 22, 30].”
BackgroundDoes the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance
Bansal, G., Wu, T., Zhou, J., Fok, R., Nushi, B., Kamar, E., Ribeiro, M. T., & Weld, D. S. (2020) · arXiv · doi:10.48550/arxiv.2006.14779 · Cited by 754
Cai, C. J., Winter, S., Steiner, D., Wilcox, L., & Terry, M. (2019) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3359206 · Cited by 594 · cited 1×
“a “primer” or onboarding process [15] or prediction explanations [2, 52].”
BackgroundBeyond Accuracy: The Role of Mental Models in Human-AI Team Performance
Bansal, G., Nushi, B., Kamar, E., Lasecki, W. S., Weld, D. S., & Horvitz, E. (2019) · Proceedings of the AAAI Conference on Human Computation and Crowdsourcing · doi:10.1609/hcomp.v7i1.5285 · Cited by 466 · cited 4×
“More specifically, our work differs from that of [6], who also experimentally investigated the capacity of humans to understand AI agents.”
BackgroundResultInfluentialBayesian Item Response Modeling: Theory and Applications
Fox, J.-P. (2010) · doi:10.1007/978-1-4419-0742-4 · Cited by 448
Updates in Human-AI Teams: Understanding and Addressing the Performance/Compatibility Tradeoff
Bansal, G., Nushi, B., Kamar, E., Weld, D. S., Lasecki, W. S., & Horvitz, E. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33012429 · Cited by 284 · cited 2×
“In contrast, prior work generally has not investigated OMMs directly, instead analyzing a proxy such as team performance or humans’ predictions of AI behavior (e.g., [2, 7]).”
BackgroundResultTell me more?: the effects of mental model soundness on personalizing an intelligent agent
Kulesza, T., Stumpf, S., Burnett, M., & Kwan, I. (2012) · Proceedings of the SIGCHI Conference on Human Factors in Computing Systems · doi:10.1145/2207676.2207678 · Cited by 225
A Case for Humans-in-the-Loop: Decisions in the Presence of Erroneous Algorithmic Scores
De-Arteaga, M., Fogliato, R., & Chouldechova, A. (2020) · Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3313831.3376638 · Cited by 221 · cited 1×
“Humans can act as a safeguard for unpredictable or undesirable behavior in AI algorithms, and can incorporate the type of contextual information and common sense reasoning that AI often lacks [17, 19, 20].”
BackgroundDesigning for human–AI complementarity in K‐12 education
Holstein, K., & Aleven, V. (2022) · AI Magazine · doi:10.1002/aaai.12058 · Cited by 156
Mental Models of AI Agents in a Cooperative Game Setting
Gero, K. I., Ashktorab, Z., Dugan, C., Pan, Q., Johnson, J., Geyer, W., Ruiz, M., Miller, S., Millen, D. R., Campbell, M., et al. (2020) · Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3313831.3376316 · Cited by 121 · cited 3×
“While we use perceptions of behavior to estimate these quantities, all three of [28]’s components of mental”
BackgroundMethodWang, Q., Saha, K., Gregori, E., Joyner, D., & Goel, A. (2021) · Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411764.3445645 · Cited by 120 · cited 1×
“Given the importance of human perceptions of AI agents in humanAI collaboration, an emerging body of work aims to understand these perceptions [18, 68, 70].”
BackgroundWilder, B., Horvitz, E., & Kamar, E. (2020) · Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2020/212 · Cited by 118
Anchoring Bias Affects Mental Model Formation and User Reliance in Explainable AI Systems
Nourani, M., Roy, C., Block, J. E., Honeycutt, D. R., Rahman, T., Ragan, E., & Gogate, V. (2021) · 26th International Conference on Intelligent User Interfaces · doi:10.1145/3397481.3450639 · Cited by 107 · cited 3×
“stochastic [6] and can be biased by first impressions of the agent [51].”
BackgroundResultBeat the Machine: Challenging Humans to Find a Predictive Model's “Unknown Unknowns”
Attenberg, J., Ipeirotis, P., & Provost, F. (2015) · Journal of Data and Information Quality · doi:10.1145/2700832 · Cited by 103
Human-Algorithm Collaboration: Achieving Complementarity and Avoiding Unfairness
Donahue, K., Chouldechova, A., & Kenthapadi, K. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533221 · Cited by 33 · cited 1×
“One important goal for human-AI systems is complementarity— achieving better performance than either the human(s) or the AI agent(s) acting independently [8, 22, 30].”
Background
Cited by
Design Principles for Generative AI Applications
Weisz, J. D., He, J., Muller, M., Hoefer, G., Miles, R., & Geyer, W. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642466 · Cited by 232
“The use of all AI systems, including those that incorporate generative capabilities, may un-fortunately lead to diverse forms of harms, especially for people in vulnerable situations.”
BackgroundThree Challenges for AI-Assisted Decision-Making
Steyvers, M., & Kumar, A. (2024) · Perspectives on Psychological Science · doi:10.1177/17456916231181102 · Cited by 217
“Furthermore, a recent study conducted by M. Kelly et al. (2023) examined the comparative assessments made by individuals regarding the abilities of AI systems and other humans in trivia-related tasks.”
Self-citationSegbenya, M., Bervell, B., Frimpong-Manso, E., Otoo, I. C., Andzie, T. A., & Achina, S. (2023) · Computers and Education Artificial Intelligence · doi:10.1016/j.caeai.2023.100188 · Cited by 52
Building Appropriate Mental Models: What Users Know and Want to Know about an Agentic AI Chatbot
Brachman, M., Kunde, S., Miller, S., Fucs, A., Dempsey, S., Jabbour, J., & Geyer, W. (2025) · Proceedings of the 30th International Conference on Intelligent User Interfaces · doi:10.1145/3708359.3712071 · Cited by 18
From testbeds to high-stakes work: a review of human-AI teaming domains and teaming factors
Kargarnovin, S., Hernandez, C. I., Reiners, D., Cruz-Neira, C., Bochenek, G., & Karwowski, W. (2026) · Frontiers in Robotics and AI · doi:10.3389/frobt.2026.1733942 · Cited by 2
“This matters for performance because inaccurate mental models can push teams toward the wrong coordination strategy, either over-delegating when the AI is weak in a subtask or underusing it when it is strong, reducing complementarity even when the AI is capable (Kelly et al., 2023).”
Metadata
- Authors
- Kelly, M., Kumar, A., Smyth, P., & Steyvers, M.
- Venue
- 2023 ACM Conference on Fairness Accountability and Transparency
- Publisher
- ACM
- Published
- 2023-06-12
- Pages
- 1723-1734
- OpenAlex
- W4376988629
- Cited by
- 26
- References
- 20
- FWCI
- 13.35
- Type
- conference paper
- Language
- en
- Open access
- hybrid
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
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.