A Mechanism for Eliciting Second-Order Beliefs and the Inclination to Choose

Abstract

This paper describes a direct revelation mechanism for eliciting decision makers' introspective beliefs on sets of subjective prior or posterior probabilities. The proposed scheme constitutes a revealedpreference procedure for measuring the inclination of decision makers to choose one alternative over another modeled by Minardi and Savochkin (2015).Keywords: Knightian uncertainty; second-order beliefs; probability elicitation; random choice; introspective beliefs; graded preferences.JEL classification numbers: D80, D82, D83 * Department of Economics, Johns Hopkins University, Baltimore, MD 21218, USA. e-mail address: karni@jhu.edu. I am grateful to two anonymous referees for their useful comments and suggestions.

Dynamic Belief Elicitation

“…The distinction is crucial: subject B has no ability to manipulate the reports of subject A , which are, from the viewpoint of subject B , states that the experimenter can observe, so that standard probability scoring rules apply. In elegant recent works, Karni (2018, 2020) uses a similar structure to elicit the second‐order beliefs of the same subject. Karni argues that this structure is useful when the subject's behavior conforms to nonstandard decision models.…”

“…In elegant recent works, Karni (2018, 2020) uses a similar structure to elicit the second‐order beliefs of the same subject. Karni argues that this structure is useful when the subject's behavior conforms to nonstandard decision models. In this case, however, the mechanism is not incentive compatible, because the subject would manipulate his future reports (see Appendix A).…”

Belief Formation and Belief Updating Under Ambiguity: Evidence from Experiments

“…These two aspects work very differently in shaping the final decisions. A sizable number of theoretical literature studies ambiguity by separating these two aspects (Ahn 2008;Brennan 1998;Cao et al 2005;Chen and Epstein 2002;Epstein and Schneider 2007;Galaabaatar and Karni 2013;Karni 2018, to name a few). However, empirically it is difficult to distinguish beliefs from attitudes, and especially to elicit beliefs.…”