Mirror, Mirror on the Wall: Algorithmic Assessments, Transparency, and Self-Fulfilling Prophecies
Open accessThe DOI matches its Crossref record; 34 of the 34 references it cites were found in a registry or an index.
Abstract
Predictive algorithmic scores can significantly impact the lives of assessed individuals by shaping decisions of organizations and institutions that affect them, for example, influencing the hiring prospects of job applicants or the release of defendants on bail. To better protect people and provide them the opportunity to appeal their algorithmic assessments, data privacy advocates and regulators increasingly push for disclosing the scores and their use in decision-making processes to scored individuals. Although inherently important, the response of scored individuals to such algorithmic transparency is understudied. Inspired by psychological and economic theories of information processing, we aim to fill this gap. We conducted a comprehensive empirical study to explore how and why disclosing the use of algorithmic scoring processes to (involuntarily) scored individuals affects their behaviors. Our results provide strong evidence that the disclosure of fundamentally erroneous algorithmic scores evokes self-fulfilling prophecies that endogenously steer the behavior of scored individuals toward their assessment, enabling algorithms to help produce the world they predict. Our results emphasize that isolated transparency measures can have considerable side effects with noticeable implications for the development of automation bias, the occurrence of feedback loops, and the design of transparency regulations.
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
Perceived usefulness, perceived ease of use, and user acceptance of information technology
Davis, F. D. (1989) · MIS Quarterly · doi:10.2307/249008 · Cited by 68,483
Tversky, A., & Kahneman, D. (1974) · Science · doi:10.1126/science.185.4157.1124 · Cited by 28,810 · cited 1×
“Anchoring effects describe a cognitive bias whereby attitudes, behaviors, and beliefs are skewed toward a provided reference point, that is, an anchor (Tversky and Kahneman 1974).”
BackgroundThe case for motivated reasoning.
Kunda, Z. (1990) · Psychological Bulletin · doi:10.1037/0033-2909.108.3.480 · Cited by 9,022
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 1×
“…adoption of information systems (Davis 1989, Ben-lian et al. 2012), trustworthiness beliefs determine trusting behaviors in online environments (McKnight et al. 2002, Kim and Benbasat 2006), reciprocity beliefs influence the willingness to contribute to and use knowledge management systems…”
BackgroundTrust, Reciprocity, and Social History
Berg, J., Dickhaut, J., & McCabe, K. (1995) · Games and Economic Behavior · doi:10.1006/game.1995.1027 · Cited by 5,548
Mathieson, K. (1991) · Information Systems Research · doi:10.1287/isre.2.3.173 · Cited by 4,216
Fehr, E., & Fischbacher, U. (2003) · Nature · doi:10.1038/nature02043 · Cited by 3,475 · cited 2×
“…closely mirrors sequential human transactions that require both trust by the first mover (e.g., loan officer, HR manager, supplier) and reciprocity by the second mover (e.g., borrower, worker, buyer), especially in incomplete contract situations (Fehr and Fischbacher 2003, Johnson and Mislin 2011).”
BackgroundMethodA Room with a Viewpoint: Using Social Norms to Motivate Environmental Conservation in Hotels
Goldstein, N. J., Cialdini, R. B., & Griskevicius, V. (2008) · Journal of Consumer Research · doi:10.1086/586910 · Cited by 3,409 · cited 1×
“These “second-order beliefs” are integral to individuals’ decision-making processes (Battigalli and Dufwenberg 2007, Goldstein et al. 2008) and components of popular psychological models such as the Theory of Planned Behavior (Ajzen 1989, Mathieson 1991).”
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×
“Algorithm aversion refers to human decision makers’ reluctance to rely on superior yet imperfect algorithms (and prefer human advice) (Dietvorst et al. 2015), whereas algorithm appreciation describes the inclination to adjust more toward advice from algorithms than humans (Logg et al. 2019).”
BackgroundResultThe nature and function of self-esteem: Sociometer theory
Leary, M. R., & Baumeister, R. F. (2000) · Advances in Experimental Social Psychology · doi:10.1016/s0065-2601(00)80003-9 · Cited by 2,521
European Union Regulations on Algorithmic Decision Making and a “Right to Explanation”
Goodman, B., & Flaxman, S. (2017) · AI Magazine · doi:10.1609/aimag.v38i3.2741 · Cited by 2,131 · cited 1×
“…for example, not being hired or not being released on bail, researchers, activists, and regulators increasingly advocate disclosing the outcome, use, and workings of algorithmic scoring processes to the public, especially to scored individuals themselves (Goodman and Flaxman 2017, Cabral 2021).”
BackgroundAlgorithm appreciation: People prefer algorithmic to human judgment
Logg, J. M., Minson, J. A., & Moore, D. A. (2019) · Organizational Behavior and Human Decision Processes · doi:10.1016/j.obhdp.2018.12.005 · Cited by 2,061
The robust beauty of improper linear models in decision making.
Dawes, R. M. (1979) · American Psychologist · doi:10.1037/0003-066x.34.7.571 · Cited by 1,966
Guilt: An interpersonal approach
Baumeister, R. F., Stillwell, A. M., & Heatherton, T. F. (1994) · Psychological Bulletin · doi:10.1037/0033-2909.115.2.243 · Cited by 1,884
Global Evidence on Economic Preferences*
Falk, A., Becker, A., Dohmen, T., Enke, B., Huffman, D., & Sunde, U. (2018) · The Quarterly Journal of Economics · doi:10.1093/qje/qjy013 · Cited by 1,724
Artificial Intelligence in Human Resources Management: Challenges and a Path Forward
Tambe, P., Cappelli, P., & Yakubovich, V. (2019) · California Management Review · doi:10.1177/0008125619867910 · Cited by 1,609
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×
“…underuse algorithmic advice depends on their task experience (Prahl and Van Swol 2017), unmet expectations about the algorithm’s characteristics (Castelo et al. 2019), the perceived loss of decision autonomy (Scherer et al. 2015), and the presence of a “human-in-the-loop” both during the…”
BackgroundExploiting moral wiggle room: experiments demonstrating an illusory preference for fairness
Dana, J., Weber, R. A., & Kuang, J. X. (2007) · Economic Theory · doi:10.1007/s00199-006-0153-z · Cited by 1,445 · cited 6×
“In their seminal paper, Dana et al. (2007) show that an option to remain ignorant about the consequences of one’s decisions serves as moral wiggle room for behaving selfishly at the expense of others while maintaining a positive self-image.”
BackgroundInfluentialThe Effects of Personalization and Familiarity on Trust and Adoption of Recommendation Agents1
Komiak, S. Y. X., & Benbasat, I. (2006) · MIS Quarterly · doi:10.2307/25148760 · Cited by 1,374
Dietvorst, B. J., Simmons, J. P., & Massey, C. (2018) · Management Science · doi:10.1287/mnsc.2016.2643 · Cited by 1,230
Positive reciprocity and intentions in trust games
McCabe, K. A., Rigdon, M. L., & Smith, V. L. (2003) · Journal of Economic Behavior & Organization · doi:10.1016/s0167-2681(03)00003-9 · Cited by 659
Englich, B., Mussweiler, T., & Strack, F. (2006) · Personality and Social Psychology Bulletin · doi:10.1177/0146167205282152 · Cited by 554 · cited 2×
“For instance, anchoring biases occur in the domain of legal decisions (Englich et al. 2006), purchasing decisions (Ariely et al. 2003), valuation decisions (Northcraft and Neale 1987), and even self-efficacy beliefs (Cervone and Peake 1986).”
BackgroundExplanations From Intelligent Systems: Theoretical Foundations and Implications for Practice1
Gregor, S., & Benbasat, I. (1999) · MIS Quarterly · doi:10.2307/249487 · Cited by 535 · cited 2×
“While focusing mainly on system users, not data subjects, previous research on explainable AI (XAI) provides ample evidence that explainability can influence human attitudes toward and perceptions of algorithmic systems (Gregor and Benbasat 1999, Poursabzi-Sangdeh et al. 2021).”
BackgroundJussupow, E., Spohrer, K., Heinzl, A., & Gawlitza, J. (2021) · Information Systems Research · doi:10.1287/isre.2020.0980 · Cited by 422
The Mechanics of Motivated Reasoning
Epley, N., & Gilovich, T. (2016) · Journal of Economic Perspectives · doi:10.1257/jep.30.3.133 · Cited by 360 · cited 1×
“Research on motivated reasoning consistently reveals that individuals selectively avoid, distort, or misinterpret information to form or maintain a particular set of self-serving beliefs (Dunning 1999, Balcetis and Dunning 2006, Uhl-mann et al. 2009, Epley and Gilovich 2016).”
BackgroundUnderstanding algorithm aversion: When is advice from automation discounted?
Prahl, A., & Van Swol, L. (2017) · Journal of Forecasting · doi:10.1002/for.2464 · Cited by 308 · cited 1×
“Whether human decision makers underuse algorithmic advice depends on their task experience (Prahl and Van Swol 2017), unmet expectations about the algorithm’s characteristics (Castelo et al. 2019), the perceived loss of decision autonomy (Scherer et al. 2015), and the presence of a…”
BackgroundA Newer Look: Motivated Social Cognition and the Schematic Representation of Social Concepts
Dunning, D. (1999) · Psychological Inquiry · doi:10.1207/s15327965pli1001_1 · Cited by 264 · cited 1×
“Research on motivated reasoning consistently reveals that individuals selectively avoid, distort, or misinterpret information to form or maintain a particular set of self-serving beliefs (Dunning 1999, Balcetis and Dunning 2006, Uhl-mann et al. 2009, Epley and Gilovich 2016).”
BackgroundKim, D., & Benbasat, I. (2006) · Information Systems Research · doi:10.1287/isre.1060.0093 · Cited by 239
Falk, A., Becker, A., Dohmen, T., Huffman, D., & Sunde, U. (2016) · doi:10.2139/ssrn.2725035 · Cited by 195
Complacency and Automation Bias in the Use of Imperfect Automation
Wickens, C. D., Clegg, B. A., Vieane, A. Z., & Sebok, A. L. (2015) · Human Factors: The Journal of the Human Factors and Ergonomics Society · doi:10.1177/0018720815581940 · Cited by 176
Strategic Ignorance and the Robustness of Social Preferences
Grossman, Z. (2014) · Management Science · doi:10.1287/mnsc.2014.1989 · Cited by 176 · cited 1×
“Researchers first studied the tendency to construct and subsequently exploit opportunistic beliefs in the context of honesty and generosity (Dana et al. 2007, Lar-son and Capra 2009, Grossman 2014).”
BackgroundAutomobile insurance classification ratemaking based on telematics driving data
Huang, Y., & Meng, S. (2019) · Decision Support Systems · doi:10.1016/j.dss.2019.113156 · Cited by 119 · cited 2×
“…during hiring processes (Tambe et al. 2019), insurance agents who use machine-learning models to compute prospective customers’ risk premiums (Huang and Meng 2019), investors leveraging credit or profit scores to estimate borrowers’ default risks (Serrano-Cinca and Guti ´ errez-Nieto…”
BackgroundLi, X., Grahl, J., & Hinz, O. (2022) · Information Systems Research · doi:10.1287/isre.2021.1074 · Cited by 93
Trust in deliberation: The consequences of deliberative decision strategies for medical decisions.
Scherer, L. D., de Vries, M., Zikmund-Fisher, B. J., Witteman, H. O., & Fagerlin, A. (2015) · Health Psychology · doi:10.1037/hea0000203 · Cited by 41 · cited 1×
“…(Prahl and Van Swol 2017), unmet expectations about the algorithm’s characteristics (Castelo et al. 2019), the perceived loss of decision autonomy (Scherer et al. 2015), and the presence of a “human-in-the-loop” both during the development (Jago 2019) or production stage (Palmeira and Spassova…”
Background
Cited by
Large Language Model in Creative Work: The Role of Collaboration Modality and User Expertise
Chen, Z., & Chan, J. (2024) · Management Science · doi:10.1287/mnsc.2023.03014 · Cited by 204
Zhou, Y., Liu, Q., Huang, J., & Li, G. (2026) · Technology in Society · doi:10.1016/j.techsoc.2025.103087 · Cited by 24
Sikhondze, B., Ye, H. J., Wang, W., Zhan, X., & Santhanam, R. (2025) · Journal of Management Information Systems · doi:10.1080/07421222.2025.2487314 · Cited by 18
“This perception leads to greater trust in human agents, which is a key driver of user valuation [12].”
High-risk artificial intelligence
Sunyaev, A., Benlian, A., Pfeiffer, J., Jussupow, E., Thiebes, S., Maedche, A., & Gawlitza, J. (2025) · Business & Information Systems Engineering · doi:10.1007/s12599-025-00942-6 · Cited by 16
“Further, as explanations are provided for the AI advice, they influence which factors decision-makers consider and can reinforce prior incorrect beliefs of individuals about a specific decision (Bauer and Gill 2024).”
Metadata
- Authors
- Bauer, K., & Gill, A.
- Venue
- Information Systems Research
- Published
- 2024-03
- Volume
- 35
- Issue
- 1
- Pages
- 226-248
- ISSN
- 1047-7047 (ISSN-L)
- 1526-5536
- OpenAlex
- W4367842162
- Cited by
- 64
- References
- 34
- FWCI
- 15.63
- Type
- journal article
- Language
- en
- Open access
- green
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.