Neural correlates of experienced moral emotion: An fMRI investigation of emotion in response to prejudice feedback

Abstract

Guilt, shame, and embarrassment are quintessential moral emotions with important regulatory functions for the individual and society. Moral emotions are, however, difficult to study with neuroimaging methods because their elicitation is more intricate than that of basic emotions. Here, using functional MRI (fMRI), we employed a novel social prejudice paradigm to examine specific brain regions associated with real-time moral emotion, focusing on guilt and related moral-negative emotions. The paradigm induced intense moral-negative emotion (primarily guilt) in 22 low-prejudice individuals through preprogrammed feedback indicating implicit prejudice against Black and disabled people. fMRI data indicated that this experience of moral-negative emotion was associated with increased activity in anterior paralimbic structures, including the anterior cingulate cortex (ACC) and anterior insula, in addition to areas associated with mentalizing, including the dorsomedial prefrontal cortex, posterior cingulate cortex, and precuneus. Of significance was prominent conflict-related activity in the supragenual ACC, which is consistent with theories proposing an association between acute guilt and behavioral inhibition. Finally, a significant negative association between self-reported guilt and neural activity in the pregenual ACC suggested a role of self-regulatory processes in response to moral-negative affect. These findings are consistent with the multifaceted self-regulatory functions of moral-negative emotions in social behavior.

Keywords:

• Anterior cingulate cortex
• Behavioral inhibition
• fMRI
• Guilt
• Moral emotions.

This work was supported by the National Research Foundation (NRF) of South Africa, the Oppenheimer Memorial Trust, the AW Mellon Foundation, the University of Cape Town, and the University of the Free State.

Notes

¹ The large number of excluded individuals reflects both the multicultural and prejudiced nature of the South African population.

² These means resemble the average IMS and EMS scores typically seen for US samples, however (Devine et al., 2002).

³ IAT variants of the same topic contained different elaborated feedback sentences, and the order of presentation of congruent and incongruent trials in these variants was counterbalanced.

⁴ One participant did not complete within-scan emotion ratings correctly, and so her data were excluded from all subsequent analyses of emotion ratings.

⁵ The degrees of freedom were corrected using Greenhouse–Geisser estimates of sphericity (ε = .37).