Tuesday, 16 March 2010
The ‘When’ of Inference: A Variation on a Hempelian Theme
David Danks, Carnegie Mellon University, Department of Philosophy
817R Cathedral of Learning
Abstract: Inference is a major topic of research in cognitive science, but the focus is almost always on how inference occurs, rather than when. In this talk, I will first argue that the latter issue is crucial to understanding the nature of inference, and (through a generalization of Hempel's Raven Paradox) that it is a deeply challenging problem. There are several possible solutions in the psychological literature, but I argue that they all fail (at least, individually) to solve the problem. I will finish by outlining a possible solution to the problem of the "when" of inference, and draw several morals from my account.