Individual differences in moral judgment competence influence neural correlates of socio-normative judgments

Abstract

To investigate how individual differences in moral judgment competence are reflected in the human brain, we used event-related functional magnetic resonance imaging, while 23 participants made either socio-normative or grammatical judgments. Participants with lower moral judgment competence recruited the left ventromedial prefrontal cortex and the left posterior superior temporal sulcus more than participants with greater competence in this domain when identifying social norm violations. Moreover, moral judgment competence scores were inversely correlated with activity in the right dorsolateral prefrontal cortex (DLPFC) during socio-normative relative to grammatical judgments. Greater activity in right DLPFC in participants with lower moral judgment competence indicates increased recruitment of rule-based knowledge and its controlled application during socio-normative judgments. These data support current models of the neurocognition of morality according to which both emotional and cognitive components play an important role.

Fig. 1

Main effect of task. Left panel: Brain regions showing a main effect of task (socio-normative judgment vs grammatical judgment). Yellow-red regions responded more during socio-normative than during grammatical judgments, blue regions showed the reverse pattern (greater responses during grammatical than during socio-normative judgments). Results of group analysis (mixed effects analysis, n = 23) thresholded at Z = 3.09. Right panel: BOLD responses [mean and standard error of the mean in arbitrary units (a.u.), n = 23] during the five conditions (NormJ/v = violations of social norms, NormJ/nv = non-violations of social norms, GramJ/v = violations of grammatical rules, GramJ/nv = non-violations of grammatical rules, Reading = reading control task) in these regions. Analyses of BOLD responses in left VMPFC, right temporal pole and left PSTS were based on functional ROIs thresholded at Z = 3.09. Analyses in left OFC and left temporal pole were based on functional ROIs thresholded at Z = 3.89. All ROIs were clusters with a size of at least 28 voxels (cluster size corresponding to a corrected cluster threshold of P < 0.05).

Fig. 2

Moral judgment competence reflected in BOLD responses in right DLPFC. (A) Covariation of C-scores with BOLD responses in right DLPFC during socio-normative vs grammatical judgments [activation from higher level analysis thresholded at Z = 3.09 (P < 0.05, corrected)]. (B) Left panel: Negative correlation of C-scores and BOLD responses in right DLPFC during socio-normative judgments [r = −0.45; P = 0.03; C-scores plotted against BOLD responses in arbitrary units (a.u.) with regression line and 95% confidence limits]. Right panel: The subgroup with lower moral judgment competence (median split, n = 12) showed significantly greater activation in right DLPFC (mean and standard error of the mean) during socio-normative judgments than the subgroup with greater moral judgment competence (n = 11). (C) Left panel: No correlation of C-scores and BOLD responses in right DLPFC during grammatical judgments [r = −0.04; P = 0.64; C-scores plotted against BOLD responses in arbitrary units (a.u.) with regression line and 95% confidence limits]. Right panel: No difference in BOLD responses (mean and standard error of the mean) between the two subgroups during grammatical judgments.

Fig. 3

Moral judgment competence modulates BOLD responses in regions involved in socio-normative judgments. BOLD responses in (A) left VMPFC and (B) left PSTS during identification of social norm violations [activation from higher level analysis thresholded at Z = 3.09 (P < 0.05, corrected) based on functional ROIs thresholded at Z = 3.09]. Participants with lower moral judgment competence (median split, n = 12) showed significantly greater activation (mean and standard error of the mean) than participants with higher moral judgment competence (n = 11).