REVIEW 1 major objections 1 minor 38 references
Scientific Explanations in Health Sciences: Causality, Trust, and Epistemic Adequacy
T0 review · 1 major / 1 minor · reviewed 2026-07-01 · grok-4.3
Pith's one-line read Prevailing accounts of explanation in the health sciences supply necessary conditions for adequate explainability in medical AI.
desk verdict This is a literature synthesis arguing that philosophy of science supplies necessary conditions for medical XAI explanations, with no new technical results or tests. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The three central axes of analysis—the role of causality in medical reasoning, the epistemic and relational dimensions of medical trust, and the criteria of explanatory adequacy shaped by pragmatic stakeholder needs—which carry the argument for integrating philosophical accounts into XAI design.
What would settle it
A concrete case in which an XAI system produces explanations that fully satisfy clinicians' requirements for causal insight, trust, and practical adequacy while making no reference to the philosophical accounts reviewed in the paper would challenge the necessity claim.
Extended reading notes
Core claim
The paper develops a critical review at the intersection of philosophy of science and XAI. It examines prevailing accounts of what counts as an explanation in the health sciences and assesses their adequacy for informing XAI in medicine, arguing that they provide necessary conditions for a philosophically grounded approach to explainability in this domain. Building on this, the discussion identifies three central axes: the role of causality in medical reasoning, the epistemic and relational dimensions of medical trust, and the criteria of explanatory adequacy as shaped by the pragmatic needs of diverse stakeholders. The paper outlines principles for designing XAI systems that offer explanati
Load-bearing premise
The philosophical literature on explanation, causality, and trust in medicine can be mapped onto the design requirements of XAI systems to supply necessary conditions for adequate medical explanations.
Editorial extensions
If this is right
- XAI systems in medicine must incorporate causal relationships as they figure in medical reasoning rather than purely statistical associations.
- Explanations must address both the epistemic reliability and the relational dimensions of trust between clinicians, patients, and the AI.
- Criteria for what counts as an adequate explanation should be evaluated against the distinct pragmatic needs of different clinical stakeholders.
- Design choices in medical XAI should draw from established philosophical accounts instead of relying solely on ad hoc technical metrics.
- Debates on medical XAI would shift toward these conceptual foundations rather than remaining focused on purely computational solutions.
Reading between the lines
- XAI development teams would benefit from including philosophers to translate the identified conditions into implementable system features.
- The same philosophical requirements might apply to explainability demands in other regulated high-stakes domains such as finance or autonomous systems.
- Empirical studies could test whether existing XAI techniques already meet or fall short of the philosophical criteria in real clinical workflows.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper develops a critical review at the intersection of philosophy of science and XAI. It examines prevailing accounts of explanation in the health sciences and argues that they provide necessary conditions for a philosophically grounded approach to explainability in medicine. It identifies three central axes: the role of causality in medical reasoning, the epistemic and relational dimensions of medical trust, and the criteria of explanatory adequacy as shaped by pragmatic needs of stakeholders. Building on this, it outlines principles for designing XAI systems that are epistemically robust and aligned with clinical decision-making.
Significance. This synthesis of philosophical literature on explanation, causality, and trust has the potential to address gaps in current XAI research by providing conceptual foundations. The paper's approach of identifying necessary conditions from established philosophy is a strength, as it grounds the discussion in prior work rather than ad hoc assumptions. If the mapping to XAI design holds, it could shape debates toward more adequate explanations in high-stakes medical contexts.
major comments (1)
- [Discussion of three central axes and outline of design principles] The central claim that philosophical accounts supply necessary conditions for medical XAI (stated in the abstract and developed in the discussion of the three axes) requires explicit justification. The manuscript should demonstrate necessity by showing that omitting any axis produces explanations that fail core epistemic or clinical requirements, rather than treating the axes as relevant desiderata.
minor comments (1)
- [Abstract] The abstract contains an overly long sentence listing the three axes; splitting it would improve readability.
Simulated Author's Rebuttal
We thank the referee for the constructive comment and positive overall assessment. We agree that the necessity of the three axes for adequate medical XAI requires more explicit demonstration and will revise the manuscript accordingly.
read point-by-point responses
-
Referee: [Discussion of three central axes and outline of design principles] The central claim that philosophical accounts supply necessary conditions for medical XAI (stated in the abstract and developed in the discussion of the three axes) requires explicit justification. The manuscript should demonstrate necessity by showing that omitting any axis produces explanations that fail core epistemic or clinical requirements, rather than treating the axes as relevant desiderata.
Authors: We accept this point. The current manuscript grounds the necessity claim in the philosophical literature on explanation, causality, and trust, but does not include explicit counterexamples showing failure when an axis is omitted. In the revised version we will add a short subsection (likely in Section 4 or 5) that supplies three targeted illustrations: (1) a non-causal feature-attribution explanation that misleads on intervention targets in a diagnostic model; (2) an explanation that satisfies local accuracy yet erodes clinician trust because it omits relational context; and (3) an explanation calibrated only to model developers that is unusable by bedside clinicians. These cases will be drawn from existing medical-AI literature and will directly link each omission to a concrete epistemic or clinical shortfall, thereby converting the axes from desiderata into demonstrated necessary conditions. revision: yes
Circularity Check
No significant circularity; philosophical synthesis relies on external sources
full rationale
The paper is a critical review synthesizing external philosophy-of-science literature (e.g., on explanation, causality, and trust) to identify necessary conditions for medical XAI. No equations, derivations, fitted parameters, or self-referential definitions appear. Central claims rest on cited external sources rather than self-citation chains or renamings that reduce to the paper's own inputs. The interpretive argument is self-contained against external benchmarks and does not exhibit any of the enumerated circularity patterns.
Assumptions & free parameters
assumptions (1)
- domain assumption Insights from philosophy of science on explanation have been largely overlooked in XAI research
Cite this review
Pith. "Pith review of Scientific Explanations in Health Sciences: Causality, Trust, and Epistemic Adequacy." pith.science (2026). https://pith.science/paper/YRIIXTOD
@misc{pith2026260631616,
author = {Pith},
title = {Pith review of: Scientific Explanations in Health Sciences: Causality, Trust, and Epistemic Adequacy},
year = {2026},
howpublished = {\url{https://pith.science/paper/YRIIXTOD}},
note = {Machine review of arXiv:2606.31616}
}
read the original abstract
Medical Artificial Intelligence (AI) is widely expected to transform clinical practice, yet the decision-making processes of many Machine Learning (ML) models remain opaque. Explainability has been advanced as a partial remedy to clarify why AI generates predictions, particularly in high-stakes contexts. Despite ongoing efforts, debates on what constitutes an adequate medical explanation remain unsettled. Yet, explanation has long been a central topic of inquiry in the philosophy of science and medicine. The insights developed in these fields, however, have been largely overlooked in contemporary explainable AI (XAI) research, leaving its foundational assumptions insufficiently examined. To address this gap, this paper develops a critical review at the intersection of philosophy of science and XAI. It examines prevailing accounts of what counts as an explanation in the health sciences and assesses their adequacy for informing XAI in medicine, arguing that they provide necessary conditions for a philosophically grounded approach to explainability in this domain. Building on this foundational philosophical literature, the discussion identifies three central axes of analysis: the role of causality in medical reasoning, the epistemic and relational dimensions of medical trust, and the criteria of explanatory adequacy as shaped by the pragmatic needs of diverse stakeholders. By integrating philosophical analysis with current developments in medical AI, the paper outlines principles for designing XAI systems that offer explanations that are not only epistemically robust but also aligned with the epistemic and practical requirements of clinical decision-making, shaping ongoing debates in medical XAI toward underexplored conceptual foundations.
Figures
Reference graph
Works this paper leans on
-
[1]
plausibility: On the (un) reliability of explanations from large language models , author=
Chirag Agarwal, Sree Harsha Tanneru, and Himabindu Lakkaraju. Faithfulness vs. plausibility: On the (un) reliability of explanations from large language models.arXiv preprint arXiv:2402.04614, 2024
-
[2]
Mohammed Ameen. Explainable mammogram analysis with EfficientNetV2 and Grad-CAM++ for robust cancer diagnosis.Diagnostics, 16(1):105, 2025
work page 2025
-
[3]
Jens Christian Bjerring and Jacob Busch. Artificial intelligence and patient-centered decision-making.Philosophy & Technology, 34(2):349–371, 2021
work page 2021
-
[4]
Causality and scientific explanation of artificial intelligence systems in biomedicine
Florian Boge and Axel Mosig. Causality and scientific explanation of artificial intelligence systems in biomedicine. Pflügers Archiv-European Journal of Physiology, 477(4):543–554, 2025
work page 2025
-
[5]
Florian J. Boge and Michael Poznic. Machine learning and the future of scientific explanation.Journal for General Philosophy of Science, 52(1):171–176, March 2021
work page 2021
-
[6]
Defining explanation and explanatory depth in XAI.Minds and Machines, 32(3):563–584, 2022
Stefan Buijsman. Defining explanation and explanatory depth in XAI.Minds and Machines, 32(3):563–584, 2022
work page 2022
-
[7]
Chung Ying Cheng. Model of causality in Chinese philosophy: A comparative study.Philosophy East and West, 26(1):3–20, 1976
work page 1976
-
[8]
Trust in medicine.The Journal of Medicine and Philosophy, 27(1):11–29, 2002
Chalmers C Clark. Trust in medicine.The Journal of Medicine and Philosophy, 27(1):11–29, 2002
work page 2002
Show all 38 references
-
[9]
Causation in Chinese Philosophy.A Companion to World Philosophies, pages 165–173, 2017
Carine Defoort. Causation in Chinese Philosophy.A Companion to World Philosophies, pages 165–173, 2017
2017
-
[10]
Juan M. Durán. Dissecting scientific explanation in AI (sXAI): A case for medicine and healthcare.Artificial Intelligence, 297:103498, 2021
2021
-
[11]
On the ethical and epistemological utility of explicable AI in medicine.Philosophy & Technology, 35(2):50, 2022
Christian Herzog. On the ethical and epistemological utility of explicable AI in medicine.Philosophy & Technology, 35(2):50, 2022
2022
-
[12]
Trust and its role in the medical encounter.Health Care Analysis, 25(3):260– 274, 2017
Stephen Holland and David Stocks. Trust and its role in the medical encounter.Health Care Analysis, 25(3):260– 274, 2017
2017
-
[13]
Causability and explainability of artificial intelligence in medicine.Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 9(4):e1312, 2019
Andreas Holzinger, Georg Langs, Helmut Denk, Kurt Zatloukal, and Heimo Müller. Causability and explainability of artificial intelligence in medicine.Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 9(4):e1312, 2019
2019
-
[14]
AI in medicine must be explainable.Nature Medicine, 27(8):1328–1328, 2021
Shinjini Kundu. AI in medicine must be explainable.Nature Medicine, 27(8):1328–1328, 2021
2021
-
[15]
Artificial intelligence and black-box medical decisions: Accuracy versus explainability
Alex John London. Artificial intelligence and black-box medical decisions: Accuracy versus explainability. Hastings Center Report, 49(1):15–21, 2019
2019
-
[16]
Understanding XAI through the philosopher’s lens: A historical perspective
Martina Mattioli, Antonio Emanuele Cinà, and Marcello Pelillo. Understanding XAI through the philosopher’s lens: A historical perspective. InECAI 2024, volume 392 ofFrontiers in Artificial Intelligence and Applications, pages 987–994. IOS Press, 2024. 7 APREPRINT- JULY1, 2026
2024
-
[17]
Ganterfac- tual—counterfactual explanations for medical non-experts using generative adversarial learning.Frontiers in Artificial Intelligence, 5:825565, 2022
Silvan Mertes, Tobias Huber, Katharina Weitz, Alexander Heimerl, and Elisabeth André. Ganterfac- tual—counterfactual explanations for medical non-experts using generative adversarial learning.Frontiers in Artificial Intelligence, 5:825565, 2022
2022
-
[18]
The human factor in explainable artificial intelligence: Clinician variability in trust, reliance, and performance.NPJ Digital Medicine, 8(1):658, 2025
Angus Nicolson, Elizabeth Bradburn, Yarin Gal, Aris T Papageorghiou, and J Alison Noble. The human factor in explainable artificial intelligence: Clinician variability in trust, reliance, and performance.NPJ Digital Medicine, 8(1):658, 2025
2025
-
[19]
Models of causation in epidemiology.Health, Disease and Causal Explanations in Medicine, 16:129–36, 1984
Staffan Norell. Models of causation in epidemiology.Health, Disease and Causal Explanations in Medicine, 16:129–36, 1984
1984
-
[20]
Explanatory pragmatism: A context-sensitive framework for explainable medical AI.Ethics and Information Technology, 24(1):13, 2022
Rune Nyrup and Diana Robinson. Explanatory pragmatism: A context-sensitive framework for explainable medical AI.Ethics and Information Technology, 24(1):13, 2022
2022
-
[21]
The pragmatic turn in eXplainable Artificial Intelligence (XAI).Minds and Machines, 29(3):441–459, Sep 2019
Andrés Páez. The pragmatic turn in eXplainable Artificial Intelligence (XAI).Minds and Machines, 29(3):441–459, Sep 2019
2019
-
[22]
Axe the X in XAI: A plea for understandable AI.arXiv preprint arXiv:2403.00315, 2024
Andrés Páez. Axe the X in XAI: A plea for understandable AI.arXiv preprint arXiv:2403.00315, 2024
2024
-
[23]
Causal inference and counterfactual prediction in machine learning for actionable healthcare.Nature Machine Intelligence, 2(7):369–375, 2020
Mattia Prosperi, Yi Guo, Matt Sperrin, James S Koopman, Jae S Min, Xing He, Shannan Rich, Mo Wang, Iain E Buchan, and Jiang Bian. Causal inference and counterfactual prediction in machine learning for actionable healthcare.Nature Machine Intelligence, 2(7):369–375, 2020
2020
-
[24]
Models of explanation and explanation in medicine.International Studies in the Philosophy of Science, 3(2):199–212, 1989
Ren Zong Qiu. Models of explanation and explanation in medicine.International Studies in the Philosophy of Science, 3(2):199–212, 1989
1989
-
[25]
AI in health and medicine.Nature Medicine, 28(1):31–38, 2022
Pranav Rajpurkar, Emma Chen, Oishi Banerjee, and Eric J Topol. AI in health and medicine.Nature Medicine, 28(1):31–38, 2022
2022
-
[26]
Causality for trustworthy artificial intelligence: Status, challenges and perspectives.ACM Computing Surveys, 57(6):1–30, 2025
Atul Rawal, Adrienne Raglin, Danda B Rawat, Brian M Sadler, and James McCoy. Causality for trustworthy artificial intelligence: Status, challenges and perspectives.ACM Computing Surveys, 57(6):1–30, 2025
2025
-
[27]
Why should I trust you?
Marco T. Ribeiro, Sameer Singh, and Carlos Guestrin. “Why should I trust you?”: Explaining the predictions of any classifier. InProceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’16, page 1135–1144, New York, USA, 2016. ACM
2016
-
[28]
Causal reasoning and the diagnostic process.Theoretical Medicine, 15(3):315–333, 1994
Dominick A Rizzi. Causal reasoning and the diagnostic process.Theoretical Medicine, 15(3):315–333, 1994
1994
-
[29]
Explanation as a social practice: Toward a conceptual framework for the social design of AI systems.IEEE Transactions on Cognitive and Developmental Systems, 13(3):717–728, 2020
Katharina J Rohlfing, Philipp Cimiano, Ingrid Scharlau, Tobias Matzner, Heike M Buhl, Hendrik Buschmeier, Elena Esposito, Angela Grimminger, Barbara Hammer, Reinhold Häb-Umbach, et al. Explanation as a social practice: Toward a conceptual framework for the social design of AI ...
2020
-
[30]
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead.Nature Machine Intelligence, 1(5):206–215, 2019
Cynthia Rudin. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead.Nature Machine Intelligence, 1(5):206–215, 2019
2019
-
[31]
Interpreting causality in the health sciences.International Studies in the Philosophy of Science, 21(2):157–170, 2007
Federica Russo and Jon Williamson. Interpreting causality in the health sciences.International Studies in the Philosophy of Science, 21(2):157–170, 2007
2007
-
[32]
Salmon.Four Decades of Scientific Explanation
Wesley C. Salmon.Four Decades of Scientific Explanation. University of Pittsburgh Press, 1990
1990
-
[33]
Schaffner
Kenneth F. Schaffner. Explanation and causation in biomedical sciences. In Larry Laudan, editor,Mind and Medicine: Problems of Explanation and Evaluation in Psychiatry and the Biomedical Sciences. University of California Press, Berkeley, 1983
1983
-
[34]
University of Chicago Press, 1993
Kenneth F Schaffner.Discovery and Explanation in Biology and Medicine. University of Chicago Press, 1993
1993
-
[35]
Explaining disease: Correlations, causes, and mechanisms.Minds and Machines, 8(1):61–78, 1998
Paul Thagard. Explaining disease: Correlations, causes, and mechanisms.Minds and Machines, 8(1):61–78, 1998
1998
-
[36]
A survey on eXplainable Artificial Intelligence (XAI): Toward medical XAI.IEEE Transactions on Neural Networks and Learning Systems, 32(11):4793–4813, 2020
Erico Tjoa and Cuntai Guan. A survey on eXplainable Artificial Intelligence (XAI): Toward medical XAI.IEEE Transactions on Neural Networks and Learning Systems, 32(11):4793–4813, 2020
2020
-
[37]
The pragmatics of explanation.American Philosophical Quarterly, 14(2):143–150, 1977
Bas C Van Fraassen. The pragmatics of explanation.American Philosophical Quarterly, 14(2):143–150, 1977
1977
-
[38]
Cambridge University Press, 2019
Markus Wolfensberger and Anthony Wrigley.Trust in Medicine. Cambridge University Press, 2019. 8
2019
Reviewed July 1, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.