{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T14:28:30Z","timestamp":1783434510052,"version":"3.54.6"},"reference-count":75,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,7,28]],"date-time":"2025-07-28T00:00:00Z","timestamp":1753660800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Comput. Sci."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>As AI systems integrate into high-stakes domains, effective human-AI collaboration requires users to be able to assess when and why to trust model predictions. This study investigates whether combining uncertainty estimates with explanations enhances human-AI interaction effectiveness, particularly examining the interplay between model uncertainty and users' self-confidence in shaping reliance, understanding, and trust.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We conducted an empirical study with 120 participants across four experimental conditions, each providing increasing levels of model assistance: (1) prediction only; (2) prediction with corresponding probability; (3) prediction with both probability and class-level recall rates; and (4) all prior information supplemented with feature importance explanations. Participants completed an income prediction task comprising of instances with varying degrees of both human and model confidence levels. In addition to measuring prediction accuracy, we collected subjective ratings of participants' perceived reliance, understanding, and trust in the model. Finally, participants completed a questionnaire evaluating their objective model understanding.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Uncertainty estimates were sufficient to enhance accuracy, with participants showing significant improvement when they were uncertain but the model exhibited high confidence. Explanations provided complementary benefits, significantly increasing both subjective understanding and participants' performance with respect to feature importance identification, counterfactual reasoning, and model simulation. Both human confidence model confidence played a role in shaping user's reliance, understanding, and trust toward the AI system. Finally, the interaction between human and model confidence determined when AI assistance was most beneficial, with accuracy gains occurring primarily when human confidence was low but model confidence was high, across three of four experimental conditions.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>These findings demonstrate that uncertainty estimates and explanations serve complementary roles in human-AI collaboration, with uncertainty estimates enhancing predictive accuracy, and explanations significantly improving model understanding without compromising performance. Human confidence acts as a moderating factor influencing all aspects of human-AI interaction, suggesting that future AI systems should account for user confidence levels. The results provide a foundation for designing AI systems that promote effective collaboration in critical applications by combining uncertainty communication with explanatory information.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fcomp.2025.1560448","type":"journal-article","created":{"date-parts":[[2025,7,28]],"date-time":"2025-07-28T05:31:06Z","timestamp":1753680666000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":6,"title":["Why not both? Complementing explanations with uncertainty, and self-confidence in human-AI collaboration"],"prefix":"10.3389","volume":"7","author":[{"given":"Ioannis","family":"Papantonis","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Vaishak","family":"Belle","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2025,7,28]]},"reference":[{"key":"B1","author":"Adams","year":"2003","journal-title":"Trust in Automated Systems"},{"key":"B2","volume-title":"Understanding Attitudes and Predicting Social Behavior","author":"Ajzen","year":"1980"},{"key":"B3","article-title":"In AI we trust? Factors that influence trustworthiness of ai-infused decision-making processes","author":"Ashoori","year":"2019","journal-title":"arXiv preprint arXiv:1912.02675"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i13.17359","article-title":"\u201cIs the most accurate ai the best teammate? Optimizing AI for teamwork,\u201d","author":"Bansal","year":"","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1145\/3411764.3445717","article-title":"\u201cDoes the whole exceed its parts? The effect of AI explanations on complementary team performance,\u201d","author":"Bansal","year":"","journal-title":"Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.2139\/ssrn.3872711","author":"Bauer","year":"2022","journal-title":"Expl (AI) ned: The impact of explainable artificial intelligence on cognitive processes"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1145\/3461702.3462571","article-title":"\u201cUncertainty as a form of transparency: Measuring, communicating, and using uncertainty,\u201d","author":"Bhatt","year":"2021","journal-title":"Proceedings of the 2021 AAAI\/ACM Conference on AI, Ethics, and Society"},{"key":"B8","volume-title":"UCI Repository of Machine Learning Databases 1998","author":"Blake","year":"1998"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1260","DOI":"10.1016\/j.ssci.2009.03.015","article-title":"Does projection into use improve trust and exploration? An example with a cruise control system","volume":"47","author":"Cahour","year":"2009","journal-title":"Safety Sci"},{"key":"B10","doi-asserted-by":"publisher","first-page":"947","DOI":"10.1177\/0018720815582261","article-title":"The role of trust as a mediator between system characteristics and response behaviors","volume":"57","author":"Chancey","year":"2015","journal-title":"Hum. Factors"},{"key":"B11","first-page":"285","article-title":"\u201cThe role of trust as a mediator between signaling system reliability and response behaviors,\u201d","volume-title":"Proceedings of the Human Factors and Ergonomics Society Annual Meeting","author":"Chancey","year":"2013"},{"key":"B12","doi-asserted-by":"publisher","first-page":"1170","DOI":"10.3390\/make6020055","article-title":"Uncertainty in xAI: human perception and modeling approaches","volume":"6","author":"Chiaburu","year":"2024","journal-title":"Mach. Learn. Knowl. Extr"},{"key":"B13","doi-asserted-by":"publisher","first-page":"719","DOI":"10.1016\/S1071-5819(03)00039-9","article-title":"The effects of errors on system trust, self-confidence, and the allocation of control in route planning","volume":"58","author":"De Vries","year":"2003","journal-title":"Int. J. Hum. Comput. Stud"},{"key":"B14","first-page":"225","article-title":"\u201cExecutive functions,\u201d","volume-title":"Handbook of Clinical Neurology","author":"Diamond","year":"2020"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1145\/3301275.3302310","article-title":"\u201cExplaining models: an empirical study of how explanations impact fairness judgment,\u201d","author":"Dodge","year":"2019","journal-title":"Proceedings of the 24th International Conference on Intelligent User Interfaces"},{"key":"B16","doi-asserted-by":"publisher","first-page":"54","DOI":"10.1214\/ss\/1177013815","article-title":"Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy","volume":"1","author":"Efron","year":"1986","journal-title":"Stat. Sci"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3359152","article-title":"The principles and limits of algorithm-in-the-loop decision making","volume":"3","author":"Green","year":"2019","journal-title":"Proc. ACM Hum. Comput. Inter"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1145\/3025453.3025683","article-title":"\u201cSharevr: enabling co-located experiences for virtual reality between HMD and non-HMD users,\u201d","author":"Gugenheimer","year":"2017","journal-title":"Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems"},{"key":"B19","doi-asserted-by":"publisher","first-page":"8562","DOI":"10.1080\/10447318.2023.2285640","article-title":"Improving trust in ai with mitigating confirmation bias: effects of explanation type and debiasing strategy for decision-making with explainable AI","volume":"40","author":"Ha","year":"2024","journal-title":"Int. J. Hum. Comput. Interact"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1145\/3290605.3300577","article-title":"\u201cRealitycheck: blending virtual environments with situated physical reality,\u201d","author":"Hartmann","year":"2019","journal-title":"Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems"},{"key":"B21","doi-asserted-by":"publisher","first-page":"407","DOI":"10.1177\/0018720814547570","article-title":"Trust in automation: integrating empirical evidence on factors that influence trust","volume":"57","author":"Hoff","year":"2015","journal-title":"Hum. Factors"},{"key":"B22","article-title":"Metrics for explainable AI: challenges and prospects","author":"Hoffman","year":"2018","journal-title":"arXiv preprint arXiv:1812.04608"},{"key":"B23","doi-asserted-by":"publisher","first-page":"1237","DOI":"10.1177\/0018720819879273","article-title":"Trust mediating reliability-reliance relationship in supervisory control of human-swarm interactions","volume":"62","author":"Hussein","year":"2020","journal-title":"Hum. Factors"},{"key":"B24","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1207\/S15327566IJCE0401_04","article-title":"Foundations for an empirically determined scale of trust in automated systems","volume":"4","author":"Jian","year":"2000","journal-title":"Int. J. Cogn. Ergon"},{"key":"B25","volume-title":"Thinking, Fast and Slow","author":"Kahneman","year":"2011"},{"key":"B26","doi-asserted-by":"publisher","first-page":"1147","DOI":"10.1037\/bul0000160","article-title":"The unity and diversity of executive functions: a systematic review and re-analysis of latent variable studies","volume":"144","author":"Karr","year":"2018","journal-title":"Psychol. Bull"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1145\/3313831.3376219","article-title":"\u201cInterpreting interpretability: understanding data scientists' use of interpretability tools for machine learning,\u201d","author":"Kaur","year":"2020","journal-title":"Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems"},{"key":"B28","volume-title":"Guidelines for Trust in Future ATM Systems-Principles","author":"Kelly","year":"2003"},{"key":"B29","doi-asserted-by":"publisher","first-page":"291","DOI":"10.1016\/j.jmva.2015.05.001","article-title":"Parametric and nonparametric bootstrap methods for general manova","volume":"140","author":"Konietschke","year":"2015","journal-title":"J. Multivar. Anal"},{"key":"B30","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3486950","article-title":"Towards balancing VR immersion and bystander awareness","volume":"5","author":"Kudo","year":"2021","journal-title":"Proc. ACM Hum. Comput. Interact"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1145\/3313831.3376873","author":"Lai","year":"2020","journal-title":"Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.1145\/3287560.3287590","article-title":"\u201cOn human predictions with explanations and predictions of machine learning models: a case study on deception detection,\u201d","author":"Lai","year":"2019","journal-title":"Proceedings of the Conference on Fairness, Accountability, and Transparency"},{"key":"B33","doi-asserted-by":"publisher","first-page":"1243","DOI":"10.1080\/00140139208967392","article-title":"Trust, control strategies and allocation of function in human-machine systems","volume":"35","author":"Lee","year":"1992","journal-title":"Ergonomics"},{"key":"B34","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1006\/ijhc.1994.1007","article-title":"Trust, self-confidence, and operators' adaptation to automation","volume":"40","author":"Lee","year":"1994","journal-title":"Int. J. Hum. Comput. Stud"},{"key":"B35","doi-asserted-by":"publisher","first-page":"50","DOI":"10.1518\/hfes.46.1.50.30392","article-title":"Trust in automation: designing for appropriate reliance","volume":"46","author":"Lee","year":"2004","journal-title":"Hum. Factors"},{"key":"B36","doi-asserted-by":"publisher","first-page":"104","DOI":"10.1037\/\/1076-898X.6.2.104","article-title":"The dynamics of trust: comparing humans to automation","volume":"6","author":"Lewandowsky","year":"2000","journal-title":"J. Exper. Psychol"},{"key":"B37","doi-asserted-by":"publisher","first-page":"1382693","DOI":"10.5772\/intechopen.111293","article-title":"Developing trustworthy artificial intelligence: insights from research on interpersonal, human-automation, and human-AI trust","volume":"15","author":"Li","year":"2024","journal-title":"Front. Psychol"},{"key":"B38","first-page":"2482","article-title":"\u201cHuman-automation collaboration in dynamic mission planning: a challenge requiring an ecological approach,\u201d","volume-title":"Proceedings of the Human Factors and Ergonomics Society Annual Meeting","author":"Linegang","year":"2006"},{"key":"B39","first-page":"4768","article-title":"\u201cA unified approach to interpreting model predictions,\u201d","volume-title":"Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS'17","author":"Lundberg","year":"2017"},{"key":"B40","doi-asserted-by":"publisher","first-page":"749","DOI":"10.1038\/s41551-018-0304-0","article-title":"Explainable machine-learning predictions for the prevention of hypoxaemia during surgery","volume":"2","author":"Lundberg","year":"2018","journal-title":"Nat. Biomed. Eng"},{"key":"B41","first-page":"6","article-title":"\u201cMeasuring human-computer trust,\u201d","volume-title":"11th Australasian Conference on Information Systems","author":"Madsen","year":"2000"},{"key":"B42","doi-asserted-by":"publisher","DOI":"10.1002\/9780470479216.corpsy0524","article-title":"\u201cMann-whitney u test,\u201d","author":"McKnight","year":"2010","journal-title":"The Corsini Encyclopedia of Psychology"},{"key":"B43","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3457607","article-title":"A survey on bias and fairness in machine learning","volume":"54","author":"Mehrabi","year":"2021","journal-title":"ACM Comput. Surv"},{"key":"B44","doi-asserted-by":"publisher","first-page":"194","DOI":"10.1518\/001872008X288574","article-title":"Not all trust is created equal: dispositional and history-based trust in human-automation interactions","volume":"50","author":"Merritt","year":"2008","journal-title":"Hum. Factors"},{"key":"B45","first-page":"1849","article-title":"\u201cBehavioral measurement of trust in automation: the trust fall,\u201d","volume-title":"Proceedings of the Human Factors And Ergonomics Society Annual Meeting","author":"Miller","year":"2016"},{"key":"B46","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1037\/\/1076-898X.6.1.44","article-title":"Adaptive automation, trust, and self-confidence in fault management of time-critical tasks","volume":"6","author":"Moray","year":"2000","journal-title":"J. Exper. Psychol"},{"key":"B47","first-page":"7076","article-title":"\u201cConsistent estimators for learning to defer to an expert,\u201d","volume-title":"International Conference on Machine Learning","author":"Mozannar","year":"2020"},{"key":"B48","doi-asserted-by":"publisher","first-page":"862322","DOI":"10.3389\/fsurg.2022.862322","article-title":"Legal and ethical consideration in artificial intelligence in healthcare: who takes responsibility?","volume":"266","author":"Naik","year":"2022","journal-title":"Front. Surg"},{"key":"B49","doi-asserted-by":"publisher","first-page":"47","DOI":"10.3390\/brainsci15010047","article-title":"AI chatbots and cognitive control: enhancing executive functions through chatbot interactions: a systematic review","volume":"15","author":"Pergantis","year":"2025","journal-title":"Brain Sci"},{"key":"B50","doi-asserted-by":"publisher","DOI":"10.1145\/3411764.3445315","article-title":"\u201cManipulating and measuring model interpretability,\u201d","author":"Poursabzi-Sangdeh","year":"2021","journal-title":"Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems"},{"key":"B51","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/S0001-6918(99)00050-5","article-title":"Optimal number of response categories in rating scales: reliability, validity, discriminating power, and respondent preferences","volume":"104","author":"Preston","year":"2000","journal-title":"Acta Psychol"},{"key":"B52","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11491","article-title":"\u201cAnchors: High-precision model-agnostic explanations,\u201d","author":"Ribeiro","year":"2018","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"B53","article-title":"\u201cCan we do better explanations? A proposal of user-centered explainable AI,\u201d","author":"Ribera","year":"2019","journal-title":"IUI Workshops"},{"key":"B54","doi-asserted-by":"publisher","DOI":"10.1145\/3126594.3126638","article-title":"\u201cOne reality: augmenting how the physical world is experienced by combining multiple mixed reality modalities,\u201d","author":"Roo","year":"2017","journal-title":"Proceedings of the 30th Annual ACM Symposium on User Interface Software and Technology"},{"key":"B55","first-page":"231","article-title":"Parametric measures of effect size","volume":"621","author":"Rosenthal","year":"1994","journal-title":"Handb. Res. Synth"},{"key":"B56","doi-asserted-by":"publisher","first-page":"105846","DOI":"10.1016\/j.ijmedinf.2025.105846","article-title":"Explainability and uncertainty: two sides of the same coin for enhancing the interpretability of deep learning models in healthcare","volume":"197","author":"Salvi","year":"2025","journal-title":"Int. J. Med. Inform"},{"key":"B57","volume-title":"Quantifying the User Experience: Practical Statistics for User Research","author":"Sauro","year":"2016"},{"key":"B58","doi-asserted-by":"crossref","first-page":"427","DOI":"10.1016\/B978-0-08-036226-7.50076-4","article-title":"\u201cTrustworthiness of command and control systems,\u201d","volume-title":"Analysis, Design and Evaluation of Man-Machine Systems 1988","author":"Sheridan","year":"1989"},{"key":"B59","doi-asserted-by":"publisher","first-page":"102551","DOI":"10.1016\/j.ijhcs.2020.102551","article-title":"The effects of explainability and causability on perception, trust, and acceptance: implications for explainable AI","volume":"146","author":"Shin","year":"2021","journal-title":"Int. J. Hum. Comput. Stud"},{"key":"B60","doi-asserted-by":"publisher","DOI":"10.1145\/3379337.3415827","article-title":"\u201cTransceivr: bridging asymmetrical communication between VR users and external collaborators,\u201d","author":"Thoravi Kumaravel","year":"2020","journal-title":"Proceedings of the 33rd Annual ACM Symposium on User Interface Software and Technology"},{"key":"B61","doi-asserted-by":"publisher","first-page":"100049","DOI":"10.1016\/j.patter.2020.100049","article-title":"Rapid trust calibration through interpretable and uncertainty-aware AI","volume":"1","author":"Tomsett","year":"2020","journal-title":"Patterns"},{"key":"B62","doi-asserted-by":"publisher","first-page":"841","DOI":"10.2139\/ssrn.3063289","article-title":"Counterfactual explanations without opening the black box: Automated decisions and the GDPR","volume":"31","author":"Wachter","year":"2018","journal-title":"Harvard J. Law Technol"},{"key":"B63","doi-asserted-by":"publisher","first-page":"281","DOI":"10.1177\/0018720809338842","article-title":"Trust and reliance on an automated combat identification system","volume":"51","author":"Wang","year":"2009","journal-title":"Hum. Factors"},{"key":"B64","doi-asserted-by":"publisher","DOI":"10.1145\/3397481.3450650","article-title":"\u201cAre explanations helpful? A comparative study of the effects of explanations in ai-assisted decision-making,\u201d","author":"Wang","year":"2021","journal-title":"26th International Conference on Intelligent User Interfaces"},{"key":"B65","article-title":"Learning to complement humans","author":"Wilder","year":"2020","journal-title":"arXiv preprint arXiv:2005.00582."},{"key":"B66","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-26633-6_7","article-title":"\u201cNonparametric statistics in human-computer interaction,\u201d","author":"Wobbrock","year":"2016","journal-title":"Modern statistical methods for HCI"},{"key":"B67","doi-asserted-by":"publisher","DOI":"10.1002\/9780471462422.eoct979","article-title":"\u201cWilcoxon signed-rank test,\u201d","author":"Woolson","year":"2007","journal-title":"Wiley encyclopedia of clinical trials"},{"key":"B68","article-title":"\u201cOn the (in) fidelity and sensitivity of explanations,\u201d","author":"Yeh","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B69","doi-asserted-by":"publisher","first-page":"179","DOI":"10.1207\/s15516709cog2102_3","article-title":"The nature of external representations in problem solving","volume":"21","author":"Zhang","year":"1997","journal-title":"Cogn. Sci"},{"key":"B70","doi-asserted-by":"publisher","DOI":"10.1145\/3491102.3517791","article-title":"\u201cYou complete me: human-AI teams and complementary expertise,\u201d","author":"Zhang","year":"2022","journal-title":"Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems"},{"key":"B71","doi-asserted-by":"publisher","DOI":"10.1145\/3351095.3372852","article-title":"\u201cEffect of confidence and explanation on accuracy and trust calibration in AI-assisted decision making,\u201d","author":"Zhang","year":"2020","journal-title":"Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency"},{"key":"B72","doi-asserted-by":"publisher","first-page":"593","DOI":"10.3390\/electronics10050593","article-title":"Evaluating the quality of machine learning explanations: a survey on methods and metrics","volume":"10","author":"Zhou","year":"2021","journal-title":"Electronics"},{"key":"B73","doi-asserted-by":"publisher","first-page":"663","DOI":"10.1177\/0272989X07303824","article-title":"Validation of the subjective numeracy scale: effects of low numeracy on comprehension of risk communications and utility elicitations","volume":"27","author":"Zikmund-Fisher","year":"2007","journal-title":"Med. Dec. Mak"},{"key":"B74","volume-title":"In the Age of the Smart Machine: The Future of Work and Power","author":"Zuboff","year":"1988"},{"key":"B75","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1162\/jocn_a_02237","article-title":"Working memory guides action valuation in model-based decision-making strategy","volume":"37","author":"Zuo","year":"2025","journal-title":"J. Cogn. Neurosci"}],"container-title":["Frontiers in Computer Science"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fcomp.2025.1560448\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,28]],"date-time":"2025-07-28T05:31:09Z","timestamp":1753680669000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fcomp.2025.1560448\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,28]]},"references-count":75,"alternative-id":["10.3389\/fcomp.2025.1560448"],"URL":"https:\/\/doi.org\/10.3389\/fcomp.2025.1560448","relation":{},"ISSN":["2624-9898"],"issn-type":[{"value":"2624-9898","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,28]]},"article-number":"1560448"}}