The Ultimatum–Dictator Offer Gap in the Lab

In an incentivized laboratory experiment (N = 212), we tested whether a measure capturing both the difference between offers in the Ultimatum and Dictator Games and their distance from zero may be a better predictor of intrinsic altruism than the commonly used Dictator Game offer. Participants took part in a within-subjects, dual-role Dictator Game and Ultimatum Game, followed by a modified version of the Die-under-the-cup Task, in which they could cheat either to benefit themselves or a charity. The experimental games were followed by the Self-Reported Altruism Scale and additional short surveys. The main results showed that although our measure outperformed the Dictator Game offer in predicting altruism, neither was significantly associated with task- or survey-based altruism. Perhaps the most interesting result emerged from the exploratory analysis: the Ultimatum–Dictator Gap—the difference between offers in the two games, and part of our proposed measure—appears to be the strongest predictor of survey-measured altruism. It also positively correlates with the time taken to make an Ultimatum Game offer (suggesting lower cognitive load when gauging the social norm) and with dishonesty. In contrast, it is negatively correlated with political orientation, with economically right-leaning participants showing a larger gap than left-leaning ones. This study offers preliminary support for the idea that integrating Dictator and Ultimatum Game offers—whether as a single gap or a more nuanced measure—may better capture altruistic tendencies than relying on the Dictator Game alone. Nonetheless, further research is needed to confirm and extend these findings.