Emotions as signals of normative conduct

Abstract

Social interactions are heavily norm-based and these norms need to be learned. For this, the emotional reactions of other's in response to a norm transgression can serve as signals. We were able to show that when a group responds with anger to a norm transgressing behaviour, participants were better able to correctly infer the norm than when the group responded with sadness or emotional neutrality. We further tested a process-model showing that this inference is based on the participants' understanding of the groups' appraisals of the behaviour. That is, participants who were able to reverse engineer the underlying appraisal of norm-incompatibility from the emotion expressions inferred the norm more readily. Humans as a social species, require efficient means to quickly adapt to new situations and to perform flawlessly in social contexts. Emotion information is one of the instruments that can be used in this quest.

Keywords:

• Norm inference
• Social signals
• Appraisals
• Reverse engineering

Notes

¹ Students in Israel are generally older since most start University studies after army service. Also, since we used MBA students and since the graduate school of business accepts only students with work experience, our graduate students were also older than elsewhere.

² It should be noted that there was a significant difference in perceived intensity between the emotions, F(2, 116)=8.61, p<.001, η²=.13. Post hoc analyses revealed that whereas anger (M=4.54, SD=1.63) and emotional neutrality (M=3.68, SD=2.07) were perceived as equally intense, sadness was rated as less intense than both (M=2.71, SD=2.14). Considering this alternative explanation, we conducted all the analyses while excluding all the participants who rated the intensity of the groups' emotion in the condition to which they were assigned as 0. This left us with 102 participants and most of those who were dropped were in the sadness condition (11 out of the 17 dropped). However, whereas sadness ratings were now comparable to anger and emotional neutrality ratings (M=3.81, SD=1.47; M=4.65, SD=1.48; M=4.20, SD=1.62, for sadness, anger and neutrality, respectively), the results with regard to the level of understanding of the norm and the associated appraisals remained unchanged.