Clear signals or mixed messages: inter-individual emotion congruency modulates brain activity underlying affective body perception

Abstract

The neural basis of emotion perception has mostly been investigated with single face or body stimuli. However, in daily life one may also encounter affective expressions by groups, e.g. an angry mob or an exhilarated concert crowd. In what way is brain activity modulated when several individuals express similar rather than different emotions? We investigated this question using an experimental design in which we presented two stimuli simultaneously, with same or different emotional expressions. We hypothesized that, in the case of two same-emotion stimuli, brain activity would be enhanced, while in the case of two different emotions, one emotion would interfere with the effect of the other. The results showed that the simultaneous perception of different affective body expressions leads to a deactivation of the amygdala and a reduction of cortical activity. It was revealed that the processing of fearful bodies, compared with different-emotion bodies, relied more strongly on saliency and action triggering regions in inferior parietal lobe and insula, while happy bodies drove the occipito-temporal cortex more strongly. We showed that this design could be used to uncover important differences between brain networks underlying fearful and happy emotions. The enhancement of brain activity for unambiguous affective signals expressed by several people simultaneously supports adaptive behaviour in critical situations.

Keywords: amygdala; body perception; emotion; fMRI; occipito-temporal cortex; parietal lobe.

Fig. 1

Elementary model of visual emotion processing. Th, Thalamus; V1–V2, primary and secondary visual cortex; PFC, prefrontal cortex, BLA, basolateral complex amygdala; CeA, central amygdala; OTC, occipital temporal cortex.

Fig. 2

Task design. An illustration of the six different conditions is displayed (not true size). HF, happy faces; HB, happy bodies; FF, fearful faces; FB, fearful bodies; DF, different faces; DB, different bodies.

Fig. 3

Same-emotion vs different-emotion body pairs. The maps show significantly activated voxels [P_corr < 0.05] for three different linear contrasts, superimposed on an inflated representation of the cortical sheet: HB > DB (yellow), FB > DB (blue) and FB + HB > DB (red). The functional cluster in the amygdala (bottom right, in red) is shown in a coronal slice in isolation (left) or superimposed on the SF (in green) and BLA (in purple) subdivisions of the amygdala as defined by the anatomical probability maps (right). HB, happy bodies; FB, fearful bodies; DB, different bodies; SF, superficial group amygdala; BLA, basolateral complex amygdala.

Fig. 4

Faces vs bodies. The top (A) shows significantly activated voxels (P_corr < 0.05) for four linear contrasts that compare object category over all emotions, superimposed on an inflated representation of the cortical sheet: DF > DB (orange), FF + HF > FB + HB (red outline), DB > DF (blue), FB + HB > FF + HF (dark blue outline). The bottom (B) shows four linear contrasts that compare object category for each emotion separately: FF > FB (yellow), FB > FF (green), HF > HB (orange), HB > HF (blue). DF, different faces; DB, different bodies; FF, fearful faces; HF, happy faces; FB, fearful bodies; HB,  happy bodies.

Fig. 5

Fearful vs happy bodies. The map shows significantly activated voxels (P_corr < 0.05) for three linear contrasts that compare emotion conditions superimposed on an inflated representation of the cortical sheet: FB > HB (orange), HB > FB (blue) and FF + FB > HF + HB (red outline). FB,  fearful bodies, HB, happy bodies; FF, fearful faces; HF, happy faces.