![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
Abstract
In two studies, we examined how expressions of guilt and shame affected person perception. In the first study, participants read an autobiographical vignette in which the writer did something wrong and reported feeling either guilt, shame, or no emotion. The participants then rated the writer's motivations, beliefs, and traits, as well as their own feelings toward the writer. The person expressing feelings of guilt or shame was perceived more positively on a number of attributes, including moral motivation and social attunement, than the person who reported feeling no emotion. In the second study, the writer of the vignette reported experiencing (or not experiencing) cognitive and motivational aspects of guilt or shame. Expressing a desire to apologise (guilt) or feelings of worthlessness (private shame) resulted in more positive impressions than did reputational concerns (public shame) or a lack of any of these feelings. Our results indicate that verbal expressions of moral emotions such as guilt and shame influence perception of moral character as well as likeability.
Notes
1To simplify our discussion of the results, we have omitted analyses of seven additional questionnaire items that did not provide any relevant findings.
2Ratings of universal obligation did reveal a significant Scenario×Emotion interaction, F(10, 274)=1.90, p<.05. Emotion condition had a significant effect on universal obligation within the “putting down a friend”, F(2, 45)=3.53, p<.05, and “betraying trust” scenarios, F(2, 50)=3.75, p<.05. In the “putting down a friend” scenario, ratings of universal obligation were significantly higher in the guilt condition (M=3.06, SD=0.94) than in the shame condition (M=2.07, SD=1.07, p<.05). In the “betraying trust” scenario, ratings were higher in the guilt condition (M=3.67, SD=0.49) than in the no emotion condition (M=2.94, SD=1.21, p<.05).
3The only exception to this pattern was found in the “putting down a friend” scenario, in which characters received significantly higher Negative Self-Image ratings in the shame condition (M=5.07, SD=0.83) than in the guilt condition (M=3.42, SD=1.37, p<.01), neither of which differed significantly from the no emotion condition (M=4.31, SD=1.69, both ps>.1), Scenario×Emotion F(10, 274)=3.40, p<.001. This was the only scenario and the only variable to show this pattern, however.
4The Social Anxiety/Need for Approval scale showed a significant Scenario×Emotion interaction, F(3, 185)=5.81, p<.002. The lying scenario revealed the same pattern of means as the overall emotion effect (see ). In the infidelity scenario, the character in public shame received significantly higher ratings (M=8.51, SD=0.81) on this scale than all of the other conditions (private shame M=3.96, SD=1.80; guilt M=4.40, SD=1.56; no emotion M=3.27, SD=1.51, all ps<.001 in Tukey tests). However, there were no other significant differences between the conditions, although the difference between guilt and no emotion approached significance (p<.06). In both scenarios, then, the manipulation check was successful, in that the character in public shame was seen as having a higher degree of Social Anxiety/Need for Approval than the other conditions.
5The Moral Character scale showed a significant Scenario × Emotion condition interaction, F(3, 185)=4.01, p<.01. The lying scenario showed the same pattern of means as the overall pattern (see ). In the infidelity scenario, the character in guilt (M=5.10, SD=1.71) did not differ in rated morality than the character in private shame (M=4.30, SD=1.02), but both were rated as significantly more moral than the character in public shame (M=3.01, SD=1.34, both ps<.005). The character in guilt was also rated as more moral than the character in no emotion (M=3.61, SD=1.16, p<.002). So the only difference between the scenarios was that in the lying scenario, the character in private shame was rated as significantly more moral than the character in the no emotion condition, but this difference was not significant in the infidelity scenario.
6The Empathy for the Victim scale revealed a significant Scenario × Emotion condition interaction, F(3, 185)=6.89, p<.001. In the lying scenario, the character in guilt (M=5.62, SD=1.41) and private shame (M=5.21, SD=1.28) did not differ significantly from each other, but both received significantly higher Empathy ratings than both public shame (M=4.03, SD=1.64) and no emotion (M=2.65, SD=1.29; all ps<.05). The character in public shame had significantly higher Empathy scores than the character in no emotion (p<.02). In the infidelity scenario, the character in guilt (M=5.17, SD=2.05) received higher Empathy ratings than all three of the other conditions (no emotion M=2.83, SD=1.48; private shame M=2.77, SD=1.55; public shame M=2.29, SD=1.65; all ps<.001), but there were no other significant differences between the emotion conditions. So in both scenarios, the character in guilt was rated highest in Empathy.
7The Liking scale showed a significant Scenario × Emotion condition interaction, F(3, 185)=3.87, p<.05. In the lying scenario, the character in no emotion (M=3.67, SD=1.35) was liked significantly less than those in all three of the other conditions (all ps<.02; guilt M=5.89, SD=0.94; private shame M=5.41, SD=1.20; public shame M=4.82, SD=1.47). In addition, the character in guilt was liked significantly more than the character in public shame (p<.02). In the infidelity scenario, the character in guilt (M=5.14, SD=1.78) was liked more than those in all the other conditions (public shame M=3.67, SD=1.55; private shame M=3.36, SD=1.19; no emotion M=3.44, SD=1.30; all ps<.01), but none of the remaining three conditions differed significantly from each other.