Visual representation of rational belief revision: Another look at the Sleeping Beauty Problem

Abstract

The coherence of probability judgments is influenced in predictable ways by people's internal representations of problems, which may be altered by the manner in which propositions are stated or "framed" (Mandel, 2008). Likewise, several studies find that probabilistic reasoning and judgment can be improved by externally representing statistical information visually (for a review, see Garcia-Retamero and Cokely, 2013). Visual representation is thought to facilitate performance by externalizing the set-subset relations among observational data. Although some studies have examined whether visual representations can improve Bayesian reasoning, they have tended to focus on the use of natural sampling trees (Sedelmeier and Gigerenzer, 2001), Euler circles (Sloman et al., 2003), or other means of representing set-subset relations.