In 'Instrumental Rationality, Epistemic Rationality, and Evidence-Gathering', I address the question of whether gathering additional evidence is always rationally required, both from the point of view of instrumental rationality and of epistemic rationality. It is shown that in certain situations, it is not instrumentally rational to look for more evidence before making a decision. These are situations in which the risk of “misleading” evidence – a concept that has both instrumental and epistemic senses – is not offset by the gains from the possibility of non-misleading evidence. Thus, our epistemic and our practical goals sometimes point us in different directions, because of the nature of rational action itself.
In 'Belief, Credence, and Norms', I present a puzzle for those who think that there is a formal reduction of a rational agent’s beliefs to her credences. I argue that if our traditional understanding of our practices of holding each other responsible is correct, then belief has a distinctive role to play, even for ideally rational agents, that cannot (unless the puzzle is resolved) be played by credence.
In In 'Groupthink', Jeffrey Sanford Russell, John Hawthorne, and I extend the established work on credence aggregation with two new results. We take conditionalization as a basic constraint, and explore two kinds of rules that allow credence aggregation: rules that use fixed prior credences and rules that use geometric averaging.