Is moral beauty different from facial beauty? Evidence from an fMRI study

Abstract

Is moral beauty different from facial beauty? Two functional magnetic resonance imaging experiments were performed to answer this question. Experiment 1 investigated the network of moral aesthetic judgments and facial aesthetic judgments. Participants performed aesthetic judgments and gender judgments on both faces and scenes containing moral acts. The conjunction analysis of the contrasts 'facial aesthetic judgment > facial gender judgment' and 'scene moral aesthetic judgment > scene gender judgment' identified the common involvement of the orbitofrontal cortex (OFC), inferior temporal gyrus and medial superior frontal gyrus, suggesting that both types of aesthetic judgments are based on the orchestration of perceptual, emotional and cognitive components. Experiment 2 examined the network of facial beauty and moral beauty during implicit perception. Participants performed a non-aesthetic judgment task on both faces (beautiful vs common) and scenes (containing morally beautiful vs neutral information). We observed that facial beauty (beautiful faces > common faces) involved both the cortical reward region OFC and the subcortical reward region putamen, whereas moral beauty (moral beauty scenes > moral neutral scenes) only involved the OFC. Moreover, compared with facial beauty, moral beauty spanned a larger-scale cortical network, indicating more advanced and complex cerebral representations characterizing moral beauty.

Keywords: OFC; fMRI; facial beauty; moral beauty; putamen.

Fig. 1

Examples of stimuli in Experiment 1. Figure (A) and (B) demonstrate two examples of scene drawings. Figure (A) depicts a morally beautiful scene (SB), in which the main character is performing a morally beautiful (i.e. altruistic) behaviour, and Figure (B) demonstrates a morally neutral scene (SN), in which the main character is performing a neutral behavior. Figure (C) is an example of a beautiful face (FB) and Figure (D) is an example of a common face (FC).

Fig. 2

Experimental design and procedure in Experiment 1. Four types of tasks were performed in separate blocks: FA judgments (beautiful vs common), facial gender (FG) judgments (male vs female), moral aesthetic judgments of scenes (SA; morally beautiful vs neutral) and gender judgments of individuals in scenes (SG; male vs female). The neural substrates underlying the aesthetic judgments for facial beauty were identified by the contrasts of ‘FA-FG’. The neural substrates underlying the aesthetic judgments for moral beauty were identified by the contrasts of ‘SA-SG’. Each block lasted for 15 s and was followed by 10 s of fixation. An instruction for task type appeared for 2 s before block onset.

Fig. 3

The conjunction results of FA judgments and moral aesthetic judgments. Left ITG, left inferior OFC and bilateral superior medial frontal gyrus were found to exhibit a conjunction effect in both contrasts of ‘aesthetic judgment > gender judgment’ for faces and scenes. Mean parameter estimates in left inferior OFC, left ITG and bilateral superior medial frontal gyrus are shown in four conditions: scene aesthetic judgment (SA), scene gender judgment (SG), face aesthetic judgment (FA), and face gender judgment (FG). A LOOCV analysis was performed (see ‘fMRI data analysis’ for details) to ensure that the data examined within each ROI are independent from the procedure used to define the ROI. *P < 0.05.

Fig. 4

Examples of stimuli in Experiment 2. Figure (A) and (B) demonstrate two examples of scene drawings, which are matched for visual complexity, familiarity and artistry with identical main character and background. Figure (A) depicts moral beauty scenes, in which the main character is performing a morally beautiful behaviour, and Figure (B) demonstrates morally neutral scene, in which the main character is performing a neutral behaviour. Figure (C) is an example of beautiful face, and Figure (D) is an example of common face.

Fig. 5

Experimental design and procedure in Experiment 2. Participants viewed stimuli in four different types of blocks: FB, FC, SB and SN. The neural substrates of facial beauty were identified by the contrast of ‘FB-FA’. The neural substrates of moral beauty were identified by the contrast of ‘SB-SN’. Each block lasted for 15 s and was followed by 10 s of fixation.

Fig. 6

The conjunction results of perception of physical and moral beauty. Left inferior OFC was found to exhibit a conjunction effect in both contrasts of ‘beautiful faces > common faces’ and ‘moral beauty scenes > neutral scenes’. Mean parameter estimates in left inferior OFC are shown in four conditions: SB, SN, FB, and FC.

Fig. 7

The correlation analysis between brain activity and aesthetic ratings. (A) Correlation results of the activation of OFC with the rating difference of ‘beautiful face > common face’ (FB-FC). (B) Correlation results of the activation of putamen with the rating difference of ‘beautiful face > common face’ (FB-FC). (C) Correlation results of the activation of OFC with the rating difference of ‘moral beauty scene > neutral scene’ (SB-SN). (D) Correlation results of the activation of putamen with the rating difference of ‘moral beauty scene > neutral scene’ (SB-SN). *P < 0.05; **P < 0.01.