The distortion of information to support an emerging evaluation of risk

A persistent problem in the assessment of the risk of an event is a bias driven by the desirability of different outcomes. However, such a desirability bias should not occur in the absence of prior dispositions toward those outcomes. This assumption is tested in an experiment designed to track the evaluation of information during an emerging evaluation of risk. Results confirm the presence of a substantial desirability bias even when there is no prior disposition toward any outcome. These findings bear implications for the assessment of risk not only in the presence of prior desirability, but also in situations currently considered benign.