Do Empathic Individuals Behave More Prosocially? Neural Correlates for Altruistic Behavior in the Dictator Game and the Dark Side of Empathy

Abstract

Do empathic individuals behave more prosocially? When we think of highly empathic individuals, we tend to assume that it is likely that those people will also help others. Most theories on empathy reflect this common understanding and claim that the personality trait empathy includes the willingness to help others, but it remains a matter of debate whether empathic individuals really help more. In economics, a prominent demonstration that our behavior is not always based on pure self-interest is the Dictator Game, which measures prosocial decisions in an allocation task. This economic game shows that we are willing to give money to strangers we do not know anything about. The present study aimed to test the relationship between dispositional empathy and prosocial acting by examining the neural underpinnings of prosocial behavior in the Dictator Game. Forty-one participants played different rounds of the Dictator Game while being scanned with functional magnetic resonance imaging (fMRI). Brain activation in the right temporoparietal junction area was associated with prosocial acting (number of prosocial decisions) and associated with empathic concern. Behavioral results demonstrated that empathic concern and personal distress predicted the number of prosocial decisions, but in a negative way. Correlations with the amount of money spent did not show any significant relationships. We discuss the results in terms of group-specific effects of affective empathy. Our results shed further light on the complex behavioral and neural mechanisms driving altruistic choices.

Keywords: altruistic; dictator game; empathy; fMRI; rTPJ; social.

Figure 1

Experimental design. See text for further details.

Figure 2

Scatterplots of behavioral responses (prosocial minus selfish decisions) and empathy subscales (Pearson correlations). The figure demonstrates that individuals with high affective empathy (empathic concern and personal distress) tend to be less prosocial in the DG. Cognitive empathy (perspective taking and fantasy) was not related to prosocial behavior.

Figure 3

Statistical maps showing brain activation for trials with social relative to non-social decisions. (A): Areas of significant fMRI signal change are shown as color overlays on the T1-MNI reference brain (at p < 0.005, uncorrected, for picture purpose only). (B): Glass brain depicting same contrast as in A (at p < 0.005, uncorrected). Even at the very lenient threshold of p < 0.005 (uncorrected) we found no anterior insula activation.

Figure 4

(A): Brain responses for trials with social relative to trials with non-social decisions predicted by the empathy subscale empathic concern. Results revealed brain activation in rTPJ-posterior STS (at p < 0.001, uncorrected, for picture display only). (B): Glass brain depicting same contrast as in A (at p < 0.001, uncorrected). (C): Scatterplot showing empathic concern correlated with peak activation in rTPJ-posterior STS brain region (Pearson’s correlation: r = 0.14).