Psychology Ph.D. Dissertations

The Dictator Game as a Test of the Social Affiliative Function of Counterfactual Expression

Date of Award


Document Type


Degree Name

Doctor of Philosophy (Ph.D.)



First Advisor

Richard Anderson (Advisor)

Second Advisor

Bonnie Berger (Other)

Third Advisor

Dara Musher-Eizenman (Committee Member)

Fourth Advisor

Carolyn Tompsett (Committee Member)


The Personal Experience and Expression of Regret (PEER) model of counterfactual thinking proposes that public expression of regretful counterfactuals should be motivated by a goal of increasing social closeness, which is demonstrated in a series of studies. The present research investigates whether expressions of counterfactual regret can influence the behavior, thoughts, and feelings of others, which is a necessary component of functional emotion expression (Cosmides & Tooby, 2000).

Whereas previous research demonstrated that the motivation to share regretful counterfactuals is to increase social closeness (Summerville & Buchanan, 2014), it did not address how these expressions affect others. The present study addresses these questions in an experimental design where the participant plays the role of someone hearing an expression of regret. Participants in the present study played a variant of the dictator’s game. In Round 1, a simulated player unfairly divided a pot of money, choosing to keep almost all of it and leaving just a small portion for the participant. Next, the simulated player sent the participant a regretful note: either a low-regret counterfactual thought or a high regret counterfactual thought. Then Round 2 began, with a new pot of money-this time with the participant being asked to divide the money between himself (or herself) and the simulated player. The initial independent variable failed to find significant results that supported the hypothesis and predictions. Using a different independent variable, participants’ report of their perceived regret of the regretful note, results demonstrated limited support for the hypothesis and predictions. Participants who perceived more regret felt socially closer to the simulated player, trusted the simulated player more, were more willing to play another game with the same player, felt the dictator game was more fair, and felt less unfairly treated by the simulated player, compared to participants who perceived less regret in the note. Of note, the behavioral measure, returning money to the simulated player, was not significantly predicted by perceived regret. The present research provides some support for the PEER model as a method of evaluating regret in emotional expression with significant limitations highlighted in the discussion.