Emotions as guardians of group norms: expressions of anger and disgust drive inferences about autonomy and purity violations

Figures & data

Table 1. Descriptive statistics and simple effects for the main dependent measures depending on condition in Study 1.

	No statement		Universal statement		Personal statement		Correlations
	Anger	Disgust	Anger	Disgust	Anger	Disgust	1	2	3	4
Anger MC (1)	6.07 (0.83)	4.81 (1.20)	5.63 (0.89)	4.84 (1.27)	6.07 (0.77)	5.68 (0.65)	1.00
Disgust MC (2)	5.04 (1.32)	6.65 (0.75)	4.63 (1.45)	5.72 (1.20)	4.86 (1.51)	5.58 (1.15)	0.02	1.00
Autonomy NV (3)	5.42 (0.69)	5.18 (0.93)	5.41 (0.39)	4.71 (0.79)	5.54 (0.67)	5.32 (0.62)	0.29*	−0.11	1.00
Purity NV (4)	3.15 (1.33)	4.28 (1.05)	3.45 (1.35)	3.82 (1.27)	3.79 (1.40)	3.75 (1.25)	−0.04	0.44*	−0.01	1.00

Note: MC = Manipulation check. NV = Norm violation. Reported descriptive statistics are means (SDs).

*p < .050.

Figure 1. Estimated strength of inferred autonomy-based and purity-based injunctive norms depending on emotion expressed about a behaviour (Study 1). Prediction is based on model coefficients; error bars represent 95% confidence intervals.

Table 2. Descriptive statistics for the main variables in Study 2 and their zero-order inter-correlations.

	Low Power			High Power			Correlations
	None	Anger	Disgust	None	Anger	Disgust	1	2	3	4	5
Anger MC (1)	2.81 (1.80)	5.57 (1.59)	4.55 (1.36)	2.47 (1.50)	5.13 (1.58)	4.20 (1.43)	1.00
Disgust MC (2)	4.13 (1.67)	6.13 (1.14)	6.73 (0.78)	3.00 (1.94)	4.57 (1.83)	6.63 (0.55)	0.43*	1.00
Power MC (3)	2.89 (1.12)	3.79 (1.14)	3.54 (1.06)	4.72 (0.99)	5.11 (0.57)	4.76 (1.16)	0.18*	−0.05	1.00
Autonomy NV (4)	3.65 (1.29)	4.01 (1.07)	4.03 (1.06)	3.71 (1.29)	4.20 (1.07)	3.90 (0.99)	0.27*	0.17*	0.25*	1.00
Purity NV (5)	3.33 (1.58)	3.55 (1.35)	4.37 (1.41)	3.24 (1.34)	3.26 (1.04)	4.07 (1.58)	0.08	0.30*	0.04	0.29*	1.00

Note: Reported descriptive statistics are means (SDs). MC = Manipulation Check. NV = Norm Violation.

*p < .050.

Figure 2. Estimated strength of inferred autonomy-based and purity-based injunctive norms depending on emotion expressed about a behaviour (Study 2). Prediction is based on model coefficients; error bars represent 95% confidence intervals.

Table 3. Means, SDs (in brackets) and zero-order correlations of the main measures in Study 3.

	Expression			Correlations
	None	Anger	Disgust	1	2	3	4	5
Anger MC (1)	2.27 (1.16)	4.27 (0.86)	3.62 (1.13)	1.00
Disgust MC (2)	2.26 (1.15)	3.48 (1.00)	4.15 (1.06)	0.56*	1.00
Autonomy NV (3)	2.13 (1.01)	3.15 (0.70)	3.20 (0.74)	0.58*	0.58*	1.00
Purity NV (4)	2.10 (1.02)	3.07 (0.78)	3.39 (0.83)	0.54*	0.61*	0.87*	1.00
Inappropriateness (5)	2.30 (1.04)	3.27 (0.76)	3.40 (0.88)	0.57*	0.56*	0.86*	0.87*	1.00

Note: MC = Manipulation check. NV = Norm violation. The reported inter-correlations between dependent measures are generally large because the no expression condition led to much lower scores on all dependent variables than both emotional expression conditions.

*p < .050.

Figure 3. Estimated strength of inferred autonomy-based and purity-based injunctive norms depending on emotion expressed about a behaviour (Study 3). Prediction is based on model coefficients; error bars represent 95% confidence intervals.

Figure 4. Confirmatory Factor Analysis in which moral violations are subtypes of injunctive norm violations (standardized coefficients) for Study 3. MFQ: Moral Foundations Questionnaire (Graham et al., 2011). NVQ: Norm violation questionnaires as developed for this article. Model fit statistics: χ²(2) = 1.02, χ²/df = 0.51, CFI = 1.00, TLI = 1.01, RMSEA = 0.00.

Graham, J., Nosek, B. A., Haidt, J., Iyer, R., Koleva, S., & Ditto, P. H. (2011). Mapping the moral domain. Journal of Personality and Social Psychology, 101, 366–385. doi: 10.1037/a0021847

 PubMed Web of Science ®Google Scholar