We design an experiment to test the hypothesis that, in violation of Bayes Rule, some people respond more forcefully to the strength of information than to its weight. We provide incentives to motivate effort, use naturally occurring information, and control for risk attitude. We find that the...