Behavioral and neural reactions to emotions of others in the distribution of resources

Abstract

This study investigated the neural mechanisms involved in the interpersonal effects of emotions—i.e., how people are influenced by other people's emotions. Participants were allocators in a version of the dictator game and made a choice between two offers after receiving written emotional expressions of the recipients. The results showed that participants more often made a self-serving offer when dealing with an angry recipient than when dealing with a happy or disappointed recipient. Compared to disappointment, expressions of anger increased activation in regions associated with self-referential thinking (anterior medial prefrontal cortex, aMPFC) and (emotional) conflict (anterior cingulate cortex). We found increased activation in temporoparietal junction for receiving happy reactions in comparison with receiving angry or disappointed reactions. This study thus emphasizes that distinct emotions have distinct effects on people in terms of behavior and underlying neurological mechanisms.

Keywords:

• Interpersonal effects of emotions
• fMRI
• Temporoparietal junction
• Medial prefrontal cortex
• Dictator game

Acknowledgments

We thank Félice van Nunspeet and Margot Schel for their help with data collection and analysis. E.A.C. is funded by a VIDI grant from the Dutch Science Foundation (NWO).

Notes

¹Although there is evidence that suggests that men and women process emotions differently (Lithari et al., 2010; Wager, Luan Phan, Liberzon, & Taylor, 2003), we observed no significant gender effects in this study.

²Because we only included participants who chose the 6–4 option and not the 5–5 option, one might wonder whether we, thereby, may have excluded individuals who have a tendency to act prosocially. However, because we created a business setting where participants were focused on profits and on maximizing the outcomes for the company, even individuals who act prosocially may have chosen the 6–4 option. This is supported by the results from a social value orientation measure, which measured whether participants were prosocials (who maximize joint outcomes and minimize differences in outcomes for the self and another person) or proselfs (who maximize individual outcomes). We assessed participants' social value orientation with the nine-item version of the decomposed games measure (for more information, see Van Lange, De Bruin, Otten, & Joireman, 1997)—a measure that has been demonstrated to have good internal consistency (Liebrand & Van Run, 1985), test–retest reliability (Van Lange & Semin-Goossens, 1998), and construct validity (Parks, 1994). Using the criterion of at least six consistent choices, 12 participants were classified as prosocial (46.2%) and 13 as proself (50.0%). One participant (3.8%) did not make at least six consistent choices and was therefore unclassifiable. These results show that we had an almost equal number of prosocials and proselfs in our experiment.