My primary research interests are in philosophy of science and epistemology. My goal is to contribute to a deeper understanding of some of the most fundamental features of science. This is not only of philosophical interest, but also touches directly on the role science plays in our society. I also do work in philosophy of physics, in which I engage with some contentious and conceptually challenging aspects of science.
Weighing Evidence Reliably
One of my key research projects investigates how we should take appropriate account of the reliability of our sources of information. This work is funded by a Veni grant from the Dutch Science Foundation (NWO). We learn from what others tell us all the time, yet others are not always reliable. The aim of the project is to determine how we should factor considerations about the reliability of our sources of information into our overall procedures for weighing evidence.
The methodology involves applying analytical and formal tools, particularly probabilistic modeling (Bayesian network models), to the problem. I also focus on two in-depth case studies; namely, how we take account of the reliability of scientific experts on climate change, and how we take account of the reliability of eyewitnesses in courtrooms. The work in this project will inform efforts to evaluate and improve current procedures for evidence assessment in scientific and legal domains. Interfacing with practitioners in these areas will be facilitated by involvement in the Kenniscentrum Filosofie Groningen, which has worked with the legal profession in the Groningen area, the Society for Risk Analysis and Glocomnet, a network with a focus on dealing with uncertainty in decision-making in a complex world.
Clarifying the role of reliability also sheds new light on recent discussions of the proper direction for the scientific realism debate. There is currently an influential new wave of realists, who argue that considerations of reliability or unreliability should have no effect on our confidence in theories, but that we should just focus on the first-order scientific evidence. This is sometimes referred to as the ‘local’ approach, in contrast with the ‘global’ approach that invokes considerations about the overall reliability of scientific method. I argue that the localists are wrong to dismiss the reliability information. Rather the challenge is to find an appropriate way to integrate it into our judgments about whether scientific theories are true.
Frameworks in science
Scientific theories can be regarded as organised into hierarchies, with higher levels sometimes called ‘paradigms’ or ‘frameworks’, and lower levels encoding more specific or concrete hypotheses. Together with cognitive scientists Josh Tenenbaum and Noah Goodman, and the philosopher of science Jim Woodward, I showed how this hierarchical picture of theories can be integrated with a Bayesian approach to confirmation. In Henderson et al. 2010, we argued that the evaluation of frameworks can be evidence-driven, despite claims to the contrary by authors such as Thomas Kuhn.
Building on this work, I have argued for a new way of understanding the evidential basis for Inference to the Best Explanation (IBE) based on hierarchical Bayesian models (Henderson 2013, Henderson (forthcoming)). The main message is that IBE is firmly based on rational procedures for evaluating evidence. This new picture of the relationship between IBE and evidential support has important implications for arguments for scientific realism. This is the subject of my ongoing research in this area.
Philosophy of physics
Prior to pursuing a career in philosophy, I did a PhD and a post-doc in quantum information theory (see my publications in this area here). Quantum information theory is a new cross-disciplinary field which brings together theoretical and experimental quantum physics with computer science, logic and information theory. It has led to the development of important new technologies such as quantum cryptography and quantum computation. It has also had a massive impact on the theoretical understanding of quantum phenomena.
My goal in my ongoing and future research is to examine what we can learn from quantum information theory about traditional foundational problems in quantum mechanics. In particular, I am analysing the idea that modified concepts of probability and information are at work in quantum theory. In earlier work, I used the analogy between the state update rule in quantum mechanics and Bayesian updating of probabilities to highlight some distinctive information-theoretic features of quantum measurement (Henderson 2010).
I also have a general interest in the process by which scientists reformulate and reaxiomatise existing physical theories. The ongoing effort to reformulate the foundations of quantum mechanics in terms of information-theoretical principles provides an example of this. Another example is the reaxiomatisation of thermodynamics by 20th century physicists. In Henderson 2014, I argued that the fundamental tension between the second law of thermodynamics and time reversal invariance of the micro-dynamics persists even in reformulated versions of thermodynamics, though it takes a somewhat different form than in the original version.
The problem of induction
I am writing a new version of the Stanford encyclopedia article on Hume’s Problem of Induction.