The impact of AI suggestions on radiologists' decisions: a pilot study of explainability and attitudinal priming interventions in mammography examination
Open accessPublished version of a preprint
Abstract
Various studies have shown that medical professionals are prone to follow the incorrect suggestions offered by algorithms, especially when they have limited inputs to interrogate and interpret such suggestions and when they have an attitude of relying on them. We examine the effect of correct and incorrect algorithmic suggestions on the diagnosis performance of radiologists when (1) they have no, partial, and extensive informational inputs for explaining the suggestions (study 1) and (2) they are primed to hold a positive, negative, ambivalent, or neutral attitude towards AI (study 2). Our analysis of 2760 decisions made by 92 radiologists conducting 15 mammography examinations shows that radiologists’ diagnoses follow both incorrect and correct suggestions, despite variations in the explainability inputs and attitudinal priming interventions. We identify and explain various pathways through which radiologists navigate through the decision process and arrive at correct or incorrect decisions. Overall, the findings of both studies show the limited effect of using explainability inputs and attitudinal priming for overcoming the influence of (incorrect) algorithmic suggestions.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
Having a negative attitude toward AI significantly decreased CADx prediction utilization, consistent with literature.
“In contrast, having a negative attitude toward AI significantly decreased CADx prediction utilization, consistent with literature.7,15”
From Appropriate trust in artificial intelligence for the optical diagnosis of colorectal polyps: the role of human/artificial intelligence interaction (2024) · cites this paper as [7]
The Introduction (Attitude towards AI; page unavailable) says a strong negative attitude can cause under-reliance on AI suggestions, supporting the general direction of the claim. It does not establish a statistically significant decrease in CADx prediction utilization; the Findings (The moderating effect of attitudinal priming; page unavailable) report no significant difference across the four priming groups in diagnosis errors, a different outcome. (LLM, conf=0.84)
Experimental — accuracy not yet measuredProviding information that explains how algorithms come to a prediction, thereby increasing transparency, is thought to support the critical interpretation of predictions, providing decision support and facilitating appropriate trust.
“Providing information that explains how algorithms come to a prediction, thereby increasing transparency, is thought to support the critical interpretation of predictions, providing decision support and facilitating appropriate trust.7,17”
From Appropriate trust in artificial intelligence for the optical diagnosis of colorectal polyps: the role of human/artificial intelligence interaction (2024) · cites this paper as [7]
Computer-aided diagnosis (CADx) systems based on artificial intelligence (AI) can help endoscopists in the optical diagnosis of colorectal polyps by predicting the histopathology.
“Computer-aided diagnosis (CADx) systems based on artificial intelligence (AI) can help endoscopists in the optical diagnosis of colorectal polyps by predicting the histopathology.”
From Appropriate trust in artificial intelligence for the optical diagnosis of colorectal polyps: the role of human/artificial intelligence interaction (2024) · cites this paper as [7]
The Introduction discusses AI and computer-aided detection in radiology, while Methods > Experiment design and setup describes BI-RADS assessment of mammograms. Searches for colorectal polyps, endoscopy, optical diagnosis, and histopathology found no passage addressing the claim. Our copy therefore cannot verify the claimed use of AI-based CADx by endoscopists. (LLM, conf=0.76)
Experimental — accuracy not yet measuredThe stand-alone performance of CADx systems and comparisons of CADx performances with thresholds for clinical implementation have been evaluated and show promising results.
“The stand-alone performance of CADx systems and comparisons of CADx performances with thresholds for clinical implementation have been evaluated and show promising results.[1][2][3][4][5][6][7]”
From Appropriate trust in artificial intelligence for the optical diagnosis of colorectal polyps: the role of human/artificial intelligence interaction (2024) · cites this paper as [7]
The Introduction discusses how CAD assistance affects radiologists’ decisions, and the Methods assess radiologists’ decisions in response to AI suggestions. Searches for stand-alone CADx performance and clinical-implementation thresholds found no passage addressing those comparisons or reporting promising results for them; no page numbers were available in our copy. (LLM, conf=0.76)
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.
Cites
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 · cited 2× · see citing works
“Whether professionals hold a positive, negative, or ambivalent attitude towards either algorithms in general or the specific algorithm and suggestion, is considered to impact both their analytical engagement and their reliance on the AI suggestions 23 .”
BackgroundA perspective on judgment and choice: mapping bounded rationality.
Kahneman, D. (2003) · American Psychologist · doi:10.1037/0003-066x.58.9.697 · Cited by 5,540 · see citing works
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 · see citing works
Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications
Samek, W., Montavon, G., Lapuschkin, S., Anders, C. J., & Muller, K.-R. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2021.3060483 · Cited by 1,477 · cited 3× · see citing works
“Ideally, professionals engage in “reflective practices” through which they directly inspect medical images and informational inputs, and consider various medical scenarios, compared to the situation in which they do not have such suggestions 13 .”
BackgroundAutomation bias: a systematic review of frequency, effect mediators, and mitigators
Goddard, K., Roudsari, A., & Wyatt, J. C. (2012) · Journal of the American Medical Informatics Association · doi:10.1136/amiajnl-2011-000089 · Cited by 1,162 · cited 1× · see citing works
“This results in a more heuristic-based usage of AI suggestions in order to compensate for the increased cognitive effort instead of systematically and critically examining the AI suggestion 31 .”
MethodBuçinca, Z., Malaya, M. B., & Gajos, K. Z. (2021) · arXiv · doi:10.48550/arxiv.2102.09692 · Cited by 1,124 · see citing works
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× · see citing works
“In the second mechanism, the very presence and complexity of the explainability inputs can create the impression of soundness of AI and its suggestion, hence causing users to follow the AI suggestions without examining their veracity 32 .”
BackgroundAlgorithmic management in a work context
Jarrahi, M. H., Newlands, G., Lee, M. K., Wolf, C. T., Kinder, E., & Sutherland, W. (2021) · Big Data & Society · doi:10.1177/20539517211020332 · Cited by 532 · cited 1× · see citing works
“The second category of factors moderating the effect of AI suggestions on human decisions are attitudes towards AI 33 .”
BackgroundAutomation bias and verification complexity: a systematic review
Lyell, D., & Coiera, E. (2017) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocw105 · Cited by 451 · cited 1× · see citing works
“Where over-reliance and under-reliance pose antithetical examples of inappropriate reliance 23 , the paradox of their occurrence has been one of the still unsolved dilemmas 24 .”
BackgroundLanglotz, C. P., Allen, B., Erickson, B. J., Kalpathy-Cramer, J., Bigelow, K., Cook, T. S., Flanders, A. E., Lungren, M. P., Mendelson, D. S., Rudie, J. D., et al. (2019) · Radiology · doi:10.1148/radiol.2019190613 · Cited by 368 · see citing works
The explainability paradox: Challenges for xAI in digital pathology
Evans, T., Retzlaff, C. O., Geißler, C., Kargl, M., Plass, M., Müller, H., Kiehl, T.-R., Zerbe, N., & Holzinger, A. (2022) · Future Generation Computer Systems · doi:10.1016/j.future.2022.03.009 · Cited by 135 · see citing works
Flagged first, then by citations
Research works printed but not linked: 31
- [1]Borch C Hee MB Toward a sociology of machine learning explainability: Human–machine interaction in deep neural network-based automated trading Big Data Soc. 2022 9 20539517221111361 10.1177/20539517221111361 · DOI
- [2]Reyes M Meier R Pereira S Silva CA Dahlweid F-M von Tengg-Kobligk H On the interpretability of artificial intelligence in radiology: Challenges and opportunities Radiol. Artif. Intell. 2020 2 e190043 10.1148/ryai.2020190043 32510054 PMC7259808 · DOI · Paper page
- [4]Alberdi E Povykalo A Strigini L Ayton P Effects of incorrect computer-aided detection (CAD) output on human decision-making in mammography Acad. Radiol. 2004 11 909 918 10.1016/j.acra.2004.05.012 15354301 · DOI · Paper page
- [5]Povyakalo, A. A., Alberdi, E., Strigini, L. & Ayton, P. Evaluating “Human+ Advisory computer” systems: A case study. In HCI2004, 18th British HCI Group Annual Conf British HCI Group, Leeds. researchgate.net, 93–96. https://www.researchgate.net/profile/Andrey-Povyakalo/publication/254291567_EVALUATING_HUMAN_ADVISORY_COMPUTER’'_SYSTEMS_A_CASE_STUDY/links/53f324f00cf256ab87b079d7/EVALUATING-HUMAN-ADVISORY-COMPUTER-SYSTEMS-A-CASE-STUDY.pdf (2004).
- [6]Brown MS Goldin JG Rogers S Kim HJ Suh RD McNitt-Gray MF Computer-aided lung nodule detection in CT: Results of large-scale observer test1 Acad. Radiol. 2005 12 681 686 10.1016/j.acra.2005.02.041 15935966 · DOI
- [7]Petrick N Haider M Summers RM Yeshwant SC Brown L Iuliano EM CT colonography with computer-aided detection as a second reader: Observer performance study Radiology 2008 246 148 156 10.1148/radiol.2453062161 18096536 · DOI
- [8]Jorritsma W Cnossen F van Ooijen PMA Improving the radiologist–CAD interaction: Designing for appropriate trust Clin. Radiol. 2015 70 115 122 10.1016/j.crad.2014.09.017 25459198 · DOI · Paper page
- [9]Zheng B Ganott MA Britton CA Hakim CM Hardesty LA Chang TS Soft-copy mammographic readings with different computer-assisted detection cuing environments: Preliminary findings Radiology 2001 221 633 640 10.1148/radiol.2213010308 11719657 · DOI
- [10]Zheng B Swensson RG Golla S Hakim CM Shah R Wallace L Detection and classification performance levels of mammographic masses under different computer-aided detection cueing environments1 Acad. Radiol. 2004 11 398 406 10.1016/S1076-6332(03)00677-9 15109012 · DOI
- [11]Beyer F Zierott L Fallenberg EM Juergens KU Stoeckel J Heindel W Comparison of sensitivity and reading time for the use of computer-aided detection (CAD) of pulmonary nodules at MDCT as concurrent or second reader Eur. Radiol. 2007 17 2941 2947 10.1007/s00330-007-0667-1 17929026 · DOI
- [12]Jussupow E Spohrer K Heinzl A Gawlitza J Augmenting medical diagnosis decisions? An investigation into physicians’ decision-making process with artificial intelligence Inf. Syst. Res. 2021 10.1287/isre.2020.0980 · DOI · Paper page
- [13]Abdel-Karim BM Pfeuffer N Carl V Hinz O How AI-based systems can induce reflections: The case of AI-augmented diagnostic work Manag. Inf. Syst. Q. 2022 10.25300/MISQ/2022/16773 · DOI
- [15]Stanovich KE SIX. The Cognitive Miser: Ways to Avoid Thinking. What Intelligence Tests Miss 2009 Yale University Press 70 85
- [16]Goddard K Roudsari A Wyatt JC Automation bias: Empirical results assessing influencing factors Int. J. Med. Inform. 2014 83 368 375 10.1016/j.ijmedinf.2014.01.001 24581700 · DOI · Paper page
- [17]Anthony C When knowledge work and analytical technologies collide: The practices and consequences of black boxing algorithmic technologies Adm. Sci. Q. 2021 66 1173 1212 10.1177/00018392211016755 · DOI · Paper page
- [18]Fiske ST Taylor SE Social Cognition 1984 Addison-Wesley
- [20]Gigerenzer G Gaissmaier W Heuristic decision making Annu. Rev. Psychol. 2011 62 451 482 10.1146/annurev-psych-120709-145346 21126183 · DOI · Paper page
- [21]Mahmud H Islam AKMN Ahmed SI Smolander K What influences algorithmic decision-making? A systematic literature review on algorithm aversion Technol. Forecast Soc. Change. 2022 175 121390 10.1016/j.techfore.2021.121390 · DOI · Paper page
- [22]Allen R Choudhury P Algorithm-augmented work and domain experience: The countervailing forces of ability and aversion Organ. Sci. 2022 33 149 169 10.1287/orsc.2021.1554 · DOI
- [25]Galsgaard A Doorschodt T Holten A-L Müller FC PlougBoesen M Maas M Artificial intelligence and multidisciplinary team meetings; a communication challenge for radiologists’ sense of agency and position as spider in a web? Eur. J. Radiol. 2022 10.1016/j.ejrad.2022.110231 35361507 · DOI · Paper page
- [26]Dratsch T Chen X RezazadeMehrizi M Kloeckner R Mähringer-Kunz A Püsken M Automation bias in mammography: The impact of artificial intelligence BI-RADS suggestions on reader performance Radiology 2023 10.1148/radiol.222176 37129490 · DOI · Paper page
- [27]Gunning D Aha D DARPA’s explainable artificial intelligence (XAI) program AI Mag. 2019 40 44 58 10.1609/aimag.v40i2.2850 · DOI · Paper page
- [28]Tonekaboni, S., Joshi, S., McCradden, M. D. & Goldenberg, A. What clinicians want: contextualizing explainable machine learning for clinical end use. In Proceedings of the 4th Machine Learning for Healthcare Conference. PMLR; 09–10 Aug 2019 (eds Doshi-Velez, F. et al.) 359–380 https://proceedings.mlr.press/v106/tonekaboni19a.html.
- [35]Andolina, V. & Lille, S. Chapter 7: Mammographic Imaging: A Practical Guide. 3rd ed. (2011).
- [36]Eberl MM Fox CH Edge SB Carter CA Mahoney MC BI-RADS classification for management of abnormal mammograms J. Am. Board Fam. Med. 2006 19 161 164 10.3122/jabfm.19.2.161 16513904 · DOI
- [38]Magny, S. J., Shikhman, R. & Keppke, A. L. Breast imaging reporting and data system. StatPearls. (StatPearls Publishing, 2022) https://www.ncbi.nlm.nih.gov/pubmed/29083600. 29083600
- [39]Baldelli P Keavey E Manley M Power G Phelan N Investigation of detector uniformity issues for Siemens Inspiration systems Phys. Med. 2020 69 262 268 10.1016/j.ejmp.2019.12.021 31927263 · DOI
- [40]Ragin CC The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies 1989 University of California Press
- [41]Haygood TM Wang J Atkinson EN Lane D Stephens TW Patel P Timed efficiency of interpretation of digital and film-screen screening mammograms AJR Am. J. Roentgenol. 2009 192 216 220 10.2214/AJR.07.3608 19098202 PMC4124527 · DOI
- [43]Huisman M Ranschaert E Parker W Mastrodicasa D Koci M Pinto de Santos D An international survey on AI in radiology in 1,041 radiologists and radiology residents part 1: Fear of replacement, knowledge, and attitude Eur. Radiol. 2021 10.1007/s00330-021-07781-5 33744991 PMC8379099 · DOI
- [44]RezazadeMehrizi MH van Ooijen P Homan M Applications of artificial intelligence (AI) in diagnostic radiology: A technography study Eur. Radiol. 2021 31 1805 1811 10.1007/s00330-020-07230-9 32945967 PMC7979626 · DOI
Cited by
Sorted by citations
Exploring automation bias in human–AI collaboration: a review and implications for explainable AI
Romeo, G., & Conti, D. (2026) · AI & SOCIETY · doi:10.1007/s00146-025-02422-7 · Cited by 209 · see citing works
“On this point, Rezazade Mehrizi et al. (2023) investigated how AI explainability tools (like heatmaps) and attitudinal priming (positive or negative information about AI) influence radiologists’ diagnostic decisions when interpreting mammograms.”
InfluentialMacnamara, B. N., Berber, I., Çavuşoğlu, M. C., Krupinski, E. A., Nallapareddy, N., Nelson, N. E., Smith, P. J., Wilson-Delfosse, A. L., & Ray, S. (2024) · Cognitive Research: Principles and Implications · doi:10.1186/s41235-024-00572-8 · Cited by 97 · see citing works
Shifa, N., Saleh, M., Akbari, Y., & Al Maadeed, S. (2025) · Clinical Imaging · doi:10.1016/j.clinimag.2025.110492 · Cited by 45 · see citing works
van der Zander, Q. E. W., Roumans, R., Kusters, C. H. J., Dehghani, N., Masclee, A. A. M., de With, P. H. N., van der Sommen, F., Snijders, C. C. P., & Schoon, E. J. (2024) · Gastrointestinal Endoscopy · doi:10.1016/j.gie.2024.06.029 · Cited by 27 · see citing works
Bellahsen-Harrar, Y., Lubrano, M., Lépine, C., Beaufrère, A., Bocciarelli, C., Brunet, A., Decroix, E., El-Sissy, F. N., Fabiani, B., Morini, A., et al. (2025) · PLOS One · doi:10.1371/journal.pone.0323270 · Cited by 11 · see citing works
- Authors
- Rezazade Mehrizi, M. H., Mol, F., Peter, M., Ranschaert, E., Dos Santos, D. P., Shahidi, R., Fatehi, M., & Dratsch, T.
- Affiliations
- Vrije Universiteit Amsterdam (NL)
- Ghent University (BE)
- Bushehr University of Medical Sciences (IR)
- University of Cologne (DE)
- Venue
- Scientific Reports
- Published
- 2023-06-07
- Volume
- 13
- Issue
- 1
- Pages
- 9230-9230
- ISSN
- 2045-2322 (ISSN-L)
- PMC
- PMC10247804
- OpenAlex
- W4379769053
- Cited by
- 54
- References
- 11
- Citations by year
- 2026: 18 · 2025: 19 · 2024: 16 · 2023: 1
- FWCI
- 5.07
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- Scientific Reports (publishedVersion, cc-by)
- Research Square (acceptedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Ghent University Academic Bibliography (Ghent University) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- VU Research Portal (submittedVersion, other-oa)
- Data Archiving and Networked Services (DANS) (submittedVersion)
- Funders
- Strong
- Nederlandse Organisatie voor Wetenschappelijk Onderzoek (40.5.22865.041)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Family Practice › Clinical Reasoning and Diagnostic Skills
- Keywords
- radiologists, explainability, attitudes toward AI, mammography, diagnostic decision-making, algorithmic bias, trust in AI, human-AI interaction
- MeSH terms
- Radiologists, Algorithms, Artificial Intelligence, Breast Neoplasms, Female, Humans, Mammography, Pilot Projects
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
