This is an experimental study to explore how different forms of presenting information (described or experienced information) to decision makers affect their confidence in choice and confidence in the realiability/trustworthiness of the information.
Decision makers use estimates of the reliability of the information they encounter to make a decision under uncertainty (Aitchison et al. 2015). For example, a manager will wait until the information she consults is deemed sufficiently reliable and trustworthy to make an important strategic decision. These perceptions of the reliability of the information they use can be communicated to third parties to reflect confidence in their beliefs or opinions, which can be useful in group discussions or leadership roles. The information used by decision makers to make a choice can typically be accessed in two broad ways: either it is access through their own experience (a sample of previously encountered choices) or through convenient descriptions of the probabilities and outcomes of the choice problem. In this study, we ask whether the source of information (experience or convenient descriptions) affects decision makers’ confidence level. More precisely, we ask whether the source of information affects two aspects of confidence: (1) confidence in the reliability of the information they used to make the choice, and (2) confidence in having made the right choice. Previous studies have emphasized the idea that experience breeds confidence even in settings where skills or expertise do not increase with experience (superstitious learning account). Yet, these changes in confidence have not been compared to other ways of learning about a task (e.g. descriptions of risky choices). We wonder whether some of the differences in choice behavior observed in the ‘DE gap’ literature may relate to differences in confidence brought about by the source of information.