Probabilistic Representation Differences between Decisions from Description and Decisions from Experience.

Abstract

For mathematically identical risky decisions, different choices can be made depending on whether information about outcomes and their probabilities is learned by description or by experience, known as the description-experience gap. However, it is unclear whether different ways of obtaining information lead to different representation forms of probability, resulting in a description-experience gap. The current study investigates the representation formats of the alternative options' probability for decisions from description and decisions from experience. The experiments measured the relative error of probability estimation in percentage and frequency forms for the two types of decisions in low and medium-to-large probability situations. The results demonstrate that for decisions from description in medium-to-large probability scenarios, the estimation error was lower in percentage form than in frequency form, with equally near-perfect estimates in small-probability scenarios. Nevertheless, in decisions from experience, the accuracy of estimation in percentage form was lower than in frequency form in both low and medium-to-large probability situations. This suggests that decision makers in decisions from description tend to represent the probability information in percentage form. However, in decisions from experience, they tend to represent probability in frequency form. The utilization of different formats for probability representation is one of the factors that contribute to the description-experience gap.