2020/01/11 by Nilanjan Das · 34 citations
Arts and Humanities · Neuroscience · Psychology · #Computer science #Economics #Epistemology #Epistemology, Ethics, and Metaphysics #Free Will and Agency #Mathematical economics #Philosophy #Philosophy and Theoretical Science #Positive economics #Rationality #Value (mathematics) #Value of information
paper · open access · doi:10.1093/bjps/axaa003
published in The British Journal for the Philosophy of Science 74(1), 25-55 (Oxford University Press)
openalex publication_date 2020/01/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In this article, I cast doubt on an apparent truism, namely, that if evidence is available for gathering and use at a negligible cost, then it’s always instrumentally rational for us to gather that evidence and use it for making decisions. Call this ‘value of information’ (VOI). I show that VOI conflicts with two other plausible theses. The first is the view that an agent’s evidence can entail non-trivial propositions about the external world. The second is the view that epistemic rationality requires us to update our credences by conditionalization. These two theses, given some plausible assumptions, make room for rationally biased inquiries where VOI fails. I go on to argue that this is bad news for defenders of VOI.