Rationale
This international conference brings together leading philosophers to explore foundational questions in the epistemology of medicine. How do we generate, evaluate, and integrate knowledge in medical science and practice—especially in an era shaped by big data, advanced statistics, and artificial intelligence? By examining the nature of evidence, causal reasoning, and explanatory models in health and healthcare, the event aims to clarify how different sources of knowledge can be combined to guide effective medical decision-making and policy. Through interdisciplinary dialogue, The Epistemology of Medicine seeks to advance philosophical understanding while contributing to the broader goals of improving patient care and promoting societal well-being.
Speakers
- Enno Fischer (TU Dresden)
- Ina Jäntgen (LMU Munich)
- Ulrich Mansmann (LMU Munich)
- Lauren Ross (UC Irvine)
- Jacob Stegenga (NUT Singapore)
- Jon Williamson (Manchester)
Schedule
Day 1 (Thursday, 25.6.26)
09:00-09:30 Registration and Welcome
09:30-10:20 Jacob Stegenga (Invited)
10:20-10:50 Coffee
10:50-11:25 Byron Hyde (Contributed)
11:25-12:00 Liyuan Jiao (Contributed)
12:00-13:30 Lunch
13:30-14:20 Lauren Ross (Invited)
14:20-14:30 Break
14:30-15:05 Benjamin Genta (Contributed)
15:05-15:35 Coffee
15:35-16:25 Jon Williamson (Invited)
Reception
Conference Dinner
Day 2 (Friday, 26.6.26)
09:30-10:20 Enno Fischer (Invited)
10:20-10:50 Coffee
10:50-11:25 Ethan Vorster (Contributed)
11:25-12:00 Pauline Paulik (Contributed)
12:00-13:30 Lunch
13:30-14:20 Ulrich Mansmann (Invited)
14:20-14:30 Break
14:30-15:05 Adria Segarra & Leon Assaad (Contributed)
15:05-15:35 Coffee
15:35-16:25 Ina Jäntgen (Invited)
16:25-16:30 Closing
Registration
Registration is free but required. You can register here
Organizing Committee
- Jürgen Landes (LMU Munich)
- Stephan Hartmann (LMU Munich)
Abstracts
Enno Fischer: “Excited Delirium Syndrome” or Asphyxiation? Evidential Practices in Forensic Medicine
“Excited Delirium Syndrome” (ExDS) is sometimes considered a potential cause of death. However, it has been argued that its sole purpose is to conceal excessive police violence because it is often used to explain death-in-custody. In my talk, I examine the epistemic conditions giving rise to the diagnosis by discussing the relation between causal hypotheses, evidence, and data in forensic medicine. I will critically discuss studies examining the main competing diagnosis: asphyxiation. In particular, I will address the role that experimentation plays in assessing claims about potential causes of death as opposed to actual causes.
This talk is based on joint work with Saana Jukola (University of Twente).
Ina Jäntgen: When broken is good enough: a decision-theoretic approach to adjusting effect sizes for meta-biase
In the biomedical and social sciences, amalgamated effect sizes from meta-analyses widely inform evidence-based decision-making. These effect sizes, however, are often subject to meta-biases—biases affecting the body of evidence on which an amalgamated effect size is estimated as a whole, such as publication bias. Several philosophers recommend that researchers should, whenever possible, adjust amalgamated effect sizes to account for such biases (e.g., Erasmus 2023; Stegenga 2018). In this talk, I draw on the perspective of rational decision-making to question this recommendation. Developing an expected utility model for treatment choices informed by biased effect sizes, I show that for certain groups of rational decision-makers, merely ruling out bias beyond a specific degree is just as valuable as learning the precisely adjusted effect size would be; for other groups, precise adjustment is more valuable. I illustrate these results for the mean difference, relative risk, and risk difference. Paired with the prevalent methodological challenges of adjusting, these results motivate a context-sensitive approach to adjusting effect sizes for meta-biases, rather than a blanket recommendation to adjust whenever possible.
Lauren Ross: Explanation in Medicine: A Control Element Account
Philosophical accounts of scientific explanation aim to clarify how explanations work, which exact factors are (and are not) explanatory, and the principles that guide explanatory practice. This talk provides an account of causal explanation in medicine with a focus on physical medicine disease examples. I introduce a control element account of causal explanation, which suggests that scientists explain outcomes by appealing to a set of causes that are control elements for a target of interest. This analysis suggests that explanatory causes are selected according to three control criteria—whether they provide information about (i) minimal control, (ii) ideal control, and (iii) more control (MIM) over the explanatory target. This account will outline the main steps in acquiring an explanation, standards that need to be met along the way, and considerations that guide explanatory relevance, which concerns how scientists determine what factors explain an outcome of interest.
Adria Segarra & Leon Assaad: Epistemically-Aware AI
Human — AI teams in healthcare can in principle achieve complementary performance, outperforming either humans or AI alone, but this potential is rarely realized in practice. Clinicians often misinterpret outputs and struggle to integrate recommendations. Existing models of these hybrid teams fail to explain this gap or guide effective interventions. We call for epistemically-aware AI systems that model both clinical tasks and the clinicians’ first- and second-order beliefs (beliefs about the patient outcome and about the AI’s outcome). By adapting outputs — such as uncertainty communication and framing — the AI can support better coordination and foster complementarity. Using a Bayesian model of the resulting human — AI teams, we show how this approach can improve diagnostic performance and promote complementarity, especially in complex, high-stakes clinical settings where current systems underperform.
Jon Williamson: A flexible approach to mechanism-informed evidence review
At present, some forms of evidence review (e.g., realist review) exploit mechanistic evidence. I argue that mechanistic evidence should inform evidence review much more widely. I motivate this claim by arguing that the way studies are assessed should be informed by mechanistic considerations and that a review can benefit from the inclusion of mechanistic evidence. I put forward a general and flexible account of mechanism-informed evidence review that is compatible with a broad range of review questions and types of review. I highlight an example of the use of this approach to review of the effectiveness of a cancer medicine.
(More talks and abstracts will be added soon)
This event is generously supported by the German Research Foundation (DFG) and the Carl Friedrich von Siemens Foundation.