Philosophers have argued that people ought to change their graded beliefs via Bayes’ rule. Recent work in psychology indicates sometimes violate rule by attending explanatory factors. Results from computational modeling suggest such violations may actually be rational.
