The DOI matches its Crossref record; 152 of the 152 references it cites were found in a registry or an index; 28 of the 145 sources its claims cite were checked for support.
Measuring and Understanding Trust Calibrations for Automated Systems: A Survey of the State-Of-The-Art and Future Directions
Open accessPublished version of a preprint
Abstract
Trust has been recognized as a central variable to explain the resistance to using automated systems (under-trust) and the overreliance on automated systems (over-trust). To achieve appropriate reliance, users’ trust should be calibrated to reflect a system’s capabilities. Studies from various disciplines have examined different interventions to attain such trust calibration. Based on a literature body of 1000+ papers, we identified 96 relevant publications which aimed to calibrate users’ trust in automated systems. To provide an in-depth overview of the state-of-the-art, we reviewed and summarized measurements of the trust calibration, interventions, and results of these efforts. For the numerous promising calibration interventions, we extract common design choices and structure these into four dimensions of trust calibration interventions to guide future studies. Our findings indicate that the measurement of the trust calibration often limits the interpretation of the effects of different interventions. We suggest future directions for this problem.
Overview
CoChat Summary
## TL;DR
This is a scoping review of empirical human-subject studies on trust calibration in automated systems, based on over 1000 papers screened down to 96 relevant publications. The paper finds that many interventions can influence trust, but measurement choices often make it hard to tell whether trust is actually calibrated to system capability.
## Key findings
- The search screened a literature body of 1000+ papers and included 96 empirical studies; among these, 40 were online studies and 61 were lab studies.
- Sample sizes ranged from 9 to 1224 participants in online studies (M = 232, SD = 224) and from 8 to 865 participants in lab studies (M = 72, SD = 125).
- The authors identify four dimensions of trust calibration interventions: exo vs. endo, warranted vs. unwarranted, static vs. adaptive, and capabilities vs. process-oriented.
- Results across interventions were mixed: uncertainty information, confidence displays, reliability updates, and some explanations sometimes helped calibration, but the paper also reports cases of increased workload, over-trust, or null effects; users were generally sensitive to reliability, yet decreases in reliability produced steeper trust decay than increases produced trust gain.
## Methods
- Design: Scoping review following the PRISMA 2020 reporting standard; the authors state they did not intend a systematic review.
- Data/sample: Peer-reviewed empirical studies with human subjects and a true trust calibration intervention, excluding preprints, theses, simulation-only studies, purely qualitative studies, and non-empirical theory/framework papers.
- Analysis: Two-wave search across Google Scholar, major HCI conference proceedings, the ACM library, and Scopus; papers were coded for system, task, intervention, calibration measurement, trust measures, and calibration results, then grouped into categories such as task domains and calibration strategies.
## Limitations
- The paper states that its results are limited to English-language texts.
- The paper states that much of the corpus was screened by a single coder, though it reports an inter-rater agreement of Krippendorf’s alpha = .83 in the training phase.
- The text makes evident that the review is limited by its operational definition of calibration as matching system capabilities to trust, which the authors acknowledge may leave out other trust factors such as process and purpose (stated in the paper).
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
14 Valid · 8 Not valid · 81 Not checked · 78 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“We fnd this, albeit understandable due to limited resources and recruitment possibilities, problematic as experts react diferently to systems than lay people (see, e.g., [59]).”
§ Measuring and Understanding Trust Calibrations for Automated Systems: A Survey of the State-Of-The-Art and Future Directions > Domain
What the sources say
- Supportedcites [59] 59. Stefanie Maria Jungmann, Timo Klan, Sebastian Kuhn, and Florian Jungmann. 2019. Accuracy of a Chatbot (ADA) in the diagnosis of mental disorders: com- parative case study with lay and expert users. JMIR formative research 3, 4 (2019), e13863. source
The Results (Number of Questions and Duration) report that psychotherapists took significantly longer than laypersons to complete cases using Ada. The Results (Percentage Agreement and Interrater Reliability) also report different diagnostic agreement across user groups. This supports the cited paper as an example of experts and laypeople interacting differently with a system, specifically the Ada app. (LLM, conf=0.88)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, need for trust in collaborative tasks might increase for situations of human-AI teaming, where the human team member is dependent on the performance of the system, such as [18,24,116] report.”
§ Measuring and Understanding Trust Calibrations for Automated Systems: A Survey of the State-Of-The-Art and Future Directions > 4.2 The Tasks
What the sources say
- Not checked — no full text of the sourcecites [18] 18. Min Chen, Stefanos Nikolaidis, Harold Soh, David Hsu, and Siddhartha Srinivasa. 2018. Planning with trust for human-robot collaboration. In Proceedings of the 2018 ACM/IEEE international conference on human-robot interaction. 307-315. source
- Partly supportedcites [24] 24. Sylvain Daronnat, Leif Azzopardi, Martin Halvey, and Mateusz Dubiel. 2021. Inferring Trust From Users' Behaviours; Agents' Predictability Positively Af- fects Trust, Task Performance and Cognitive Load in Human-Agent Real-Time Collaboration. Frontiers in Robotics and AI 8 (2021), 194. sourceExperimental — accuracy not yet measured
The Abstract says effective human-agent collaboration requires appropriate trust and identifies agent performance as a factor influencing it. The Introduction likewise calls trust important in collaborative decision-making. Neither passage says the need for trust increases when a human teammate depends on the system’s performance. (LLM, conf=0.82)
- Not checked — no full text of the sourcecites [116] 116. Beau G Schelble, Jeremy Lopez, Claire Textor, Rui Zhang, Nathan J McNeese, Richard Pak, and Guo Freeman. 2022. Towards Ethical AI: Empirically Investi- gating Dimensions of AI Ethics, Trust Repair, and Performance in Human-AI Teaming. Human Factors (2022), 00187208221116952. source
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“Okamura and Yamada [104] implemented specifc trust calibration cues, which were displayed when participants relied too little and too much on the system.”
§ Measuring and Understanding Trust Calibrations for Automated Systems: A Survey of the State-Of-The-Art and Future Directions > 4.4 The calibrations dimensions
What the sources say
- Could not verifycites [104] 104. Kazuo Okamura and Seiji Yamada. 2020. Adaptive trust calibration for human-AI collaboration. Plos one 15, 2 (2020), e0229132. sourceExperimental — accuracy not yet measured
The Development of PRISMA 2020 and Discussion passages concern a guideline for reporting systematic reviews, not an experiment involving trust calibration. Searches for Okamura and Yamada and for cues triggered by under- or over-reliance found no passage addressing the claim. (LLM, conf=0.77)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“A third way to measure the trust calibration was to measure over-, under-, and appropriate trust as the behavioral deviation (reliance) from the ideal behavior [27,50,55,104,110,132,137].”
§ Measuring and Understanding Trust Calibrations for Automated Systems: A Survey of the State-Of-The-Art and Future Directions > System capabilities.
What the sources say
- Not checked — no full text of the sourcecites [27] 27. Ewart J De Visser, Samuel S Monfort, Ryan McKendrick, Melissa AB Smith, Patrick E McKnight, Frank Krueger, and Raja Parasuraman. 2016. Almost human: Anthropomorphism increases trust resilience in cognitive agents. Journal of Experimental Psychology: Applied 22, 3 (2016), 331. source
- Not checked — no full text of the sourcecites [50] 50. Sandy H Huang, Kush Bhatia, Pieter Abbeel, and Anca D Dragan. 2018. Es- tablishing appropriate trust via critical states. In 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 3929-3936. source
- Not checked — no full text of the sourcecites [55] 55. Theodore Jensen, Mohammad Maif Hasan Khan, and Yusuf Albayram. 2020. The role of behavioral anthropomorphism in human-automation trust calibration. In International Conference on Human-Computer Interaction. Springer, 33-53. source
- Could not verifycites [104] 104. Kazuo Okamura and Seiji Yamada. 2020. Adaptive trust calibration for human-AI collaboration. Plos one 15, 2 (2020), e0229132. sourceExperimental — accuracy not yet measured
The Discussion addresses the trustworthiness of systematic-review findings, not calibration of a person's trust or reliance behavior. Searches for over-, under-, and appropriate trust and for deviation from ideal behavior found no passage in our copy addressing that measurement method. Page numbers were not available. (LLM, conf=0.76)
- Could not verifycites [110] 110. Dalai Dos Santos Ribeiro, Gabriel Diniz Junqueira Barbosa, Marisa Do Carmo Silva, Hélio Lopes, and Simone Diniz Junqueira Barbosa. 2021. Exploring the impact of classifcation probabilities on users' trust in ambiguous instances. In 2021 IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC). IEEE, 1-9. sourceExperimental — accuracy not yet measured
Our copy of Ribeiro et al. (2021) is a literature review of RPA tools with AI support (Sections 1-2 define RPA, Sections 4-5 survey UiPath, Kofax, Automation Anywhere, AssistEdge, Table 1 compares AI techniques per tool, Section 6 conclusions); no passage addresses trust, trust calibration, over-/under-trust, reliance or ideal-performance reference behaviour. Three searches (trust calibration / over-trust under-trust reliance ideal behavior; trust measurement human-agent interaction metrics; trust in automation reliance human robot interaction) returned only RPA-tool and reference-list passages, the closest being Section 2 on task replacement and cost reduction. The claim therefore cannot be checked against this source, and nothing in our copy either supports or contradicts it. (LLM, conf=0.70)
- Not checked — no full text of the sourcecites [132] 132. Xinru Wang and Ming Yin. 2022. Efects of Explanations in AI-Assisted Decision Making: Principles and Comparisons. ACM Transactions on Interactive Intelligent Systems (TiiS) (2022). source
- Not checked — no full text of the sourcecites [137] 137. Fumeng Yang, Zhuanyi Huang, Jean Scholtz, and Dustin L Arendt. 2020. How do visual explanations foster end users' appropriate trust in machine learning?. In Proceedings of the 25th International Conference on Intelligent User Interfaces. Wischnewski et al. 189-201. source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“Okamura and Yamada [104] could show that participants who received adaptive trust calibration cues outperformed those who continuously received information about the system's performance.”
§ Measuring and Understanding Trust Calibrations for Automated Systems: A Survey of the State-Of-The-Art and Future Directions > 4.4 The calibrations dimensions
What the sources say
- Could not verifycites [104] 104. Kazuo Okamura and Seiji Yamada. 2020. Adaptive trust calibration for human-AI collaboration. Plos one 15, 2 (2020), e0229132. sourceExperimental — accuracy not yet measured
The Discussion addresses strategies for improving adherence to reporting guidelines, not trust-calibration cues or participant performance. Searches for Okamura and Yamada, adaptive trust calibration, and system-performance information found no passage addressing the claimed comparison. (LLM, conf=0.78)
hedgedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
70 more claims · 7 with their cited sources checked
Create a free account to see all 181 claims and to verify the 11 references still open.
- It may state two separate claims as one. 65
- Probably not a substantial claim. 53
- The quote only partly states this claim. 1
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?
In the citing paper
“Because goals vary by task and context [75], aligning AI behavior with an individual's ongoing goals is crucial for effective human-AI collaboration [148].”
From What Needs Attention? Prioritizing Drivers of Developers' Trust and Adoption of Generative AI (2025) · cites this paper as [148]
In the citing paper
“Developers trusted genAI more when its actions aligned with their goals, reducing verification effort and cognitive load.”
From What Needs Attention? Prioritizing Drivers of Developers' Trust and Adoption of Generative AI (2025) · cites this paper as [148]
In the citing paper
“Trust is central to explaining resistance toward automated systems [148, 150, 151].”
From What Needs Attention? Prioritizing Drivers of Developers' Trust and Adoption of Generative AI (2025) · cites this paper as [148]
In the citing paper
“Trust is bolstered when genAI actions stay congruent with developers' objectives, reinforcing its credibility as a cognitive collaborator [148].”
From What Needs Attention? Prioritizing Drivers of Developers' Trust and Adoption of Generative AI (2025) · cites this paper as [148]
In the citing paper
“It is more consistent with the broader logic of the trust-calibration literature, in which the depth of user processing is treated as a condition for appropriately differentiated reliance (Lee and See 2004;Wischnewski et al. 2023;Kaplan et al. 2023).”
From Epistemic Trust in Generative AI as an Information Source: Development and Validation of the Trust in AI-Epistemic Scale (TAI-E) (2026) · cites this paper as Wischnewski et al. 2023
Citations
Cites
The PRISMA 2020 statement: An updated guideline for reporting systematic reviews
Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., et al. (2021) · BMJ · doi:10.1136/bmj.n71 · Cited by 108,283
An integrative model of organizational trust
Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995) · The Academy of Management Review · doi:10.2307/258792 · Cited by 10,736
The PRISMA 2020 statement: an updated guideline for reporting systematic reviews
Page, M. J., McKenzie, J., Bossuyt, P., Boutron, I., Hoffmann, T., mulrow, C., Shamseer, L., Tetzlaff, J., Akl, E., Brennan, S. E., et al. (2020) · doi:10.31222/osf.io/v7gm2_v1 · Cited by 8,090
Affect and Cognition Based Trust as Foundations for Interpersonal Cooperation in Organizations
McAllister, D. J. (1995) · Academy of Management Journal · doi:10.2307/256727 · Cited by 7,346
Trust in Automation: Designing for Appropriate Reliance
Lee, J. D., & See, K. A. (2004) · Human Factors: The Journal of the Human Factors and Ergonomics Society · doi:10.1518/hfes.46.1.50_30392 · Cited by 6,605
Developing and validating trust measures for e-commerce: an integrative typology
McKnight, D. H., Choudhury, V., & Kacmar, C. (2002) · Information Systems Research · doi:10.1287/isre.13.3.334.81 · Cited by 5,550 · cited 2×
“However, we also noticed that only a few studies diferentiated between diferent subconstructs of trust (performance, process, purpose) which is inherently included in many scales (e.g., [91]) but authors did not make use of this information.”
BackgroundMethodExplanation in artificial intelligence: Insights from the social sciences
Miller, T. (2019) · Artificial Intelligence · doi:10.1016/j.artint.2018.07.007 · Cited by 5,302
The PRISMA 2020 statement: an updated guideline for reporting systematic reviews
Page, M. J., McKenzie, J., Bossuyt, P., Boutron, I., Hoffmann, T., mulrow, C., Shamseer, L., Tetzlaff, J., Akl, E., Brennan, S. E., et al. (2020) · doi:10.31222/osf.io/v7gm2 · Cited by 5,177
Humans and automation: Use, misuse, disuse, abuse
Parasuraman, R., & Riley, V. (1997) · Human Factors: The Journal of the Human Factors and Ergonomics Society · doi:10.1518/001872097778543886 · Cited by 4,554 · cited 2×
“Hence, many studies aim to adjust trust bestowed in a system to refect the trustworthiness of systems, arriving at the concept of calibrated trust [74, 106].”
BackgroundAlgorithm aversion: People erroneously avoid algorithms after seeing them err.
Dietvorst, B. J., Simmons, J. P., & Massey, C. (2015) · Journal of Experimental Psychology: General · doi:10.1037/xge0000033 · Cited by 3,232 · cited 2×
“sity to trust and their attitudes towards automation, both factors which afect trust [30, 41].”
Background
Research works printed but not linked: 7
- Na Du, Kevin Y Huang, and X Jessie Yang. 2020. Not all information is equal: effects of disclosing different types of likelihood information on trust, compliance and reliance, and task performance in human-automation teaming. Human factors 62, 6 (2020), 987–1001.
- Alexandra D Kaplan, Theresa T Kessler, J Christopher Brill, and PA Hancock. 2021. Trust in artificial intelligence: Meta-analytic findings. Human Factors (2021), 00187208211013988.
- M. Madsen and S. Gregor. 2000. Measuring Human-Computer Trust. In Proceedings of 11th Australasian Conference on Information Systems. 6e8.
- Bonnie Marlene Muir. 2002. Operators’ trust in and use of automatic controllers in a supervisory process control task.(2002).
- Mona SharifHeravi, John R Taylor, Christopher J Stanton, Sandra Lambeth, and Christopher Shanahan. 2020. It’sa Disaster! Factors Affecting Trust Development and Repair Following Agent Task Failure. In Proceedings of the 2020 Australasian Conference on Robotics and Automation (ACRA 2020), 8-10 December 2020, Brisbane, Queensland.
- Yihan Wu and Ryan M Kelly. 2020. Online Dating Meets Artificial Intelligence: How the Perception of Algorithmically Generated Profile Text Impacts Attractiveness and Trust. In 32nd Australian Conference on Human-Computer Interaction. 444–453.
- Jin Xu and Ayanna Howard. 2022. Evaluating the Impact of Emotional Apology on Human-Robot Trust. In 2022 31st IEEE International Conference on Robot and Human Interactive Communication (RO-MAN). IEEE, 1655–1661.
Cited by
Psychological factors underlying attitudes toward AI tools
De Freitas, J., Agarwal, S., Schmitt, B., & Haslam, N. (2023) · Nature Human Behaviour · doi:10.1038/s41562-023-01734-2 · Cited by 221
“Consequently, research has sought to determine the psychological factors driving attitudes toward AI tools, and how to overcome AI resistance, so that user trust is calibrated to the system’s capabilities 8 .”
BackgroundThe role of explainable AI in the context of the AI Act
Panigutti, C., Hamon, R., Hupont, I., Llorca, D. F., Yela, D. F., Junklewitz, H., Scalzo, S., Mazzini, G., Sánchez, I., Garrido, J. S., et al. (2023) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3593013.3594069 · Cited by 270
“Many factors play a role in the relationship between trust and AI explanations [91, 119].”
BackgroundHow Do We Assess the Trustworthiness of AI? Introducing the Trustworthiness Assessment Model (TrAM)
Schlicker, N., Baum, K., Uhde, A., Sterz, S., Hirsch, M. C., & Langer, M. (2022) · doi:10.31234/osf.io/qhwvx · Cited by 79
A Systematic Review of Trust Assessments in Human–Robot Interaction
Campagna, G., & Rehm, M. (2024) · ACM Transactions on Human-Robot Interaction · doi:10.1145/3706123 · Cited by 102
Ma, S., Wang, X., Lei, Y., Shi, C., Yin, M., & Ma, X. (2024) · arXiv · doi:10.48550/arxiv.2403.09552 · Cited by 56
Metadata
- Authors
- Wischnewski, M., Krämer, N., & Müller, E.
- Affiliations
- Research Center for Trustworthy Data Science and Security, Germany
- University of Duisburg-Essen (DE)
- TU Dortmund University (DE)
- Venue
- Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems
- Publisher
- ACM
- Published
- 2023-04-19
- Pages
- 1-16
- OpenAlex
- W4366548373
- Cited by
- 146
- References
- 111
- Citations by year
- 2026: 73 · 2025: 43 · 2024: 23 · 2023: 7
- FWCI
- 72.97
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- CHI Conference on Human Factors in Computing Systems (publishedVersion, cc-by)
- Universitätsbibliographie, Universität Duisburg-Essen (submittedVersion)
- OSF Preprints (OSF Preprints) (submittedVersion)
- PsyArXiv (OSF Preprints) (submittedVersion, cc-by)
- PsyArXiv (OSF Preprints) (submittedVersion)
- Topics
- Social Sciences › Human Factors and Ergonomics › Human-Automation Interaction and Safety
- Social Sciences › Psychology › Social Psychology › Social Robot Interaction and HRI
- Social Sciences › Decision Sciences › Information Systems and Management › Technology Adoption and User Behaviour
- Keywords
- trust calibration, trust in automation, trust measurement, human-automation interaction
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).
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: Measuring and Understanding Trust Calibrations for Automated Systems: A Survey of the State-Of-The-Art and Future Directions. https://app.cochat.ai/papers/10.1145/3544548.3581197
- The paper
- Wischnewski, M., Krämer, N., & Müller, E. (2023). Measuring and Understanding Trust Calibrations for Automated Systems: A Survey of the State-Of-The-Art and Future Directions. Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems. https://doi.org/10.1145/3544548.3581197
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.