This paper studies stochastic choice in social contexts, and explores the difference between inequity-averse and image-conscious preferences, in particular, shame aversion. This paper studies a class of additive perturbed utility (APU) introduced in Fudenberg et al. (2015). We find that the general class of inequity-averse preferences correspond to the case of item-invariant APU, and that image-conscious preferences correspond to the case of menu-invariant APU. Under some conditions, the difference between inequity aversion and shame aversion is discussed through the violation of Regularity, one of the most well-known properties of stochastic choice.

Bénabou and Tirole (2004) explain that people in the face of self-control problems could commit to rational choices with concern for self-reputation. The key of the mechanism is uncertainty of one's willpower and induced signaling effect of willpower-related actions. We suggest that this mechanism with self-reputation potentially explains a wide range of the decision making under biases. One of the example is procrastination. In a certain situation, procrastination works as a signal of low self-control power and induces precaution of doing the job early in the next time. Conversely, this strategic pretending to be low willpower might be used as an excuse for procrastination and then concern for self-reputation may induce ambivalent effects on one's welfare, contrasted with the result in Bénabou and Tirole (2004). We suggest that patience amid the craving determines the effect on one's welfare and welfare may worsen for strategic but impatient people.

In this study of economic decision-making, we conducted a laboratory experiment to investigate the relationship between dishonest behaviors and their reaction times. Dishonest behavior is defined here as the decision to lie to increase own monetary payoff. We divided participants into three groups to examine the heterogeneity of the extent of lying: Group 1= the Honest group, comprising people who always made honest decisions; Group 2= the Big Liars’ group, comprising people who always made dishonest decisions; and Group 3= the Liars’ group, comprising people who sometimes made honest decisions and sometimes made dishonest decisions. The reaction times for all decisions, including honest decisions, were significantly longer in the Liars’ group than in the Honest group, but there was no statistically significant difference between the mean reaction times of the Big liars’ group and Liars’ group. Panel data analysis of each group revealed two results. First, the Honest group made honest decisions even when a dishonest decision would result in a large payoff, but their reaction times for such decisions were long. Second, in the Liars’ group, there was a non-linear relationship between dishonest behaviors and reaction times; that is, the reaction times for honest decisions and those for dishonest decisions receiving maximal payoff were lower than those for dishonest decisions receiving medium payoff.

Consider a decision-maker who has an opportunity to wait for information before making a choice. He can obtain more information by waiting more, but this is costly. As a result, he endogenously determines the length of time to choose an alternative, which is called response time. The Aoyama (2019) models such a decision-maker as if he solves an optimal stopping problem. The present paper summarizes the main result in that paper. The model incorporates dynamic information structure formalized as evolving information partition, which we call subjective filtration. We axiomatically characterize the model using the behavioral data consisting of choice and response time conditional on choice situations. From the data, we partially identify subjective filtration as well as other parameters, by which we explain the data. This result implies that using response time somewhat helps us understand the human cognitive process.