Bayesian reasoning, updating subjective probability in light of new information, is notoriously difficult. One factor that may contribute to this difficulty is lack of a mental model for how to combine the three key parameters in any Bayesian problem. An experiment was conducted contrasting four representations of Bayesian problems: three types of diagrams and a two by two contingency table. All four representations led to extremely good performance on a Bayesian task. This advantage also extended to a superficially dissimilar task and also persisted beyond the day of training, suggesting that graphic and tabular representation can lead to flexible and durable changes in the way people think about such problems.
CITATION STYLE
Cole, W. G. (1989). Understanding Bayesian reasoning via graphical displays. ACM SIGCHI Bulletin, 20(SI), 381–386. https://doi.org/10.1145/67450.67522
Mendeley helps you to discover research relevant for your work.