The important question is how people process probabilistic information
, not whether they process it in accordance with a normative model tha
t we never should have expected them to be capable of following. Exper
ience is not the cure, as widely thought, to problems with utilizing b
ase rate information.