Incomplete-Information Models of Guilt Aversion in the Trust Game

What is it about?

Most economic models assume that agents maximize their expected monetary payoff. However, subjects in the lab exhibit persistent and significant deviations from this self-interested maximizing behavior. A reasonable explanation for this behavior is that players can be motivated not only by monetary payoffs but also by what are sometimes referred to as 'psychological utilities'. When we deal with intention-based feelings, emotions and social norms, i.e., belief-dependent motivations, we need to turn to psychological game theory. This new framework focuses on strategic settings where at least one player has belief-dependent motivations or believes that one of his opponents has belief-dependent motivations. In the theory of psychological games it is assumed that players' preferences on material consequences depend on endogenous beliefs (games with belief-dependent motivations). Most of the applications of this theoretical framework assume that the psychological utility functions representing such preferences are common knowledge. However, this is often unrealistic. In particular, it cannot be true in experimental games where players are subjects drawn at random from a population. Therefore, an incomplete-information methodology is needed. We take a first step in this direction, focusing on guilt aversion in the Trust Game. In our models, agents have heterogeneous belief hierarchies. Our analysis illustrates the incomplete-information approach to psychological games and can help to organize experimental results in the Trust Game.

Featured Image

Why is it important?

To our knowledge, this is the first paper offering a fully-fledged Bayesian equilibrium analysis of guilt aversion. Furthermore, we highlight that belief-dependent motivations may or may not depend on the player's role: truster or trustee. To show the consequences of each modeling choice, we consider both alternatives. In one model, we make the standard assumption that guilt sensitivity and beliefs about it do not depend on the role played in the game (role-independent guilt). In the other model, only the trustee can feel guilty (role-dependent guilt). Even when players are drawn from the same population, it is not implausible to assume that sensitivity to guilt is triggered only when playing in the role of trustee. This assumption resonates with (i) the evolutionary psychology of emotions, which suggests that, when a single emotion (guilt) operates in a variety of different domains, its effects are moderated by contextual cues; and with (ii) the conceptual act theory of emotion, which posits that people experience an (in our case, anticipated) emotion by categorizing an instance of affective feeling (anticipated disappointment of the other); with this, it is plausible that the role played in the interaction is part of the categorization process. Finally, the different responses to oxytocin of trusters and trustees, and findings in the animal literature on the reactivity of oxytocin to social cues provide some indirect evidence supporting the role-dependent model of guilt aversion. The theoretical insights of our models can be used by experimental economists to extend the elicitation and analysis of players' beliefs, design new experiments, and explain previous experimental results.

Perspectives