Perceptions of the appropriate response to norm violation in 57 societies

The study investigates metanorms—norms about the use of informal sanctions in response to norm violations—across 57 countries. While informal sanctions can sustain cooperation and social order, excessive or inappropriate sanctioning may threaten social harmony. Little is known about how preferences for specific responses (verbal or physical confrontation, social ostracism, gossip, or non-action) vary across societies and across domains of norm violations. The authors pose five preregistered hypotheses: (1) universal negative relation between perceived appropriateness of the triggering behavior and the appropriateness of using sanctions (and positive for doing nothing); (2) cross-country variation in perceived appropriateness of sanctions is robust across domains (cooperation vs. out-of-place behaviors); (3) cross-country variation may show consistency, complementarity, or independence across different sanction types; (4) higher perceived appropriateness of a sanction in a country predicts higher reported use of that sanction; (5) perceived appropriateness of direct punishment (verbal/physical confrontation) correlates with cultural, ecological, and socioeconomic factors (lower indulgence, higher power distance, lower individualism/autonomy, higher tightness, higher threat, lower emancipative moral judgments, higher pro-violence attitudes, higher pathogen prevalence, lower gender equality, lower median income). The study’s purpose is to map universal and culture-specific metanorms, test these hypotheses, and relate metanorms to broader cultural and economic indicators.

Prior work often conceptualizes sanctions generically as payoff-reducing punishment, which may not capture real-life responses. Recent scholarship distinguishes social ostracism, gossip, and direct confrontation (verbal or physical) as distinct informal sanctions that vary in directness, harm, and the likelihood the target becomes aware of being sanctioned. Cross-cultural research has typically focused on economic games, physical confrontation, or unspecified punishment, rarely distinguishing forms. Theoretical frameworks motivating cross-societal differences include Hofstede’s cultural dimensions (individualism, power distance, indulgence), tightness–looseness theory (norm strength linked to threat), behavioral immune system theory (pathogen prevalence shaping coordination needs and intolerance of deviance), and modernization theory (economic development fostering emancipative values, autonomy, and gender equality). Prior findings suggest that physical confrontation is more acceptable in less indulgent, higher power-distance, less individualistic countries; verbal confrontation of incivility is more normative in less individualistic societies; formal punitiveness is higher in tighter cultures; and modernization shifts values toward autonomy and tolerance, potentially decreasing acceptance of punishment.

Design and preregistration: Cross-cultural, preregistered study (OSF: osf.io/qg6xy) in 57 countries with convenience samples of students and non-students. Total N=22,863 (students=18,091; non-students=4,772), excluding respondents under 18. Data, materials, and analysis code available at OSF (osf.io/pm5kc/). Sampling and procedure: Targeted ≈200 college students in a major city per country; additional non-student samples in 31 countries; multiple cities in 10 countries. Survey translated into 30 languages with independent translation and back-translation. Mostly online via Qualtrics; some pen-and-paper administrations with animations shown. Ethics: Informed consent obtained; approvals from relevant IRBs/ethics committees across sites. Stimuli and measures: Ten scenarios across three domains:

• Cooperation: animation of an agent depleting a common resource (A).
• Out-of-place behaviors: listening to music in headphones at a funeral (B); sleeping in a restaurant (C); singing in a library (D); reading a newspaper at the movies (E).
• Meta-violations: physical confrontation against the common-resource depleter (F); physical aggression after a verbal insult (G); and responses to rudeness: reprimand (H), speak negatively (gossip) (I), stay away (ostracism) (J). For each of the 10 scenarios, participants rated: (i) appropriateness of the target behavior and (ii) appropriateness of four responses: verbal confrontation, gossip, social ostracism, non-action. Ratings on a 6-point scale (0=extremely inappropriate to 5=extremely appropriate) yielding 50 appropriateness ratings per participant. Standardization and missing data: To control response styles, all 50 appropriateness ratings per participant were standardized by adding a constant so each participant’s mean across the 50 items equals the grand mean. Missing values imputed via EM in SPSS. Authors note standardization can induce slight artificial negative correlations among items. Country-level metanorm measures: For verbal confrontation, social ostracism, gossip, and non-action, country metanorms were computed using multi-level models controlling for the perceived appropriateness of the underlying norm violation at both country and individual levels. Main estimation used scenarios A–E; robustness checks used all A–J and scenario A only. Physical confrontation metanorms were computed separately from scenarios F and G. Culture and context measures: Included country-aggregated measures from the survey:
• Individual autonomy (child qualities: independence + determination − religious faith − obedience; WVS-derived; country α=0.75).
• Emancipative moral judgments (justifiability of homosexuality, divorce, abortion, suicide; 0–10; α=0.92).
• Pro-violence attitudes (justifiability of wife-beating and violence against others; α=0.78).
• Tightness (Gelfand’s 6-item scale adjusted by appropriateness-item mean; α=0.80).
• Perceived societal threat (sum of endorsed threats; α=0.89). External country indicators: Hofstede scores for indulgence (n=48), power distance (n=51), individualism (n=51); historical pathogen prevalence; Global Gender Gap Index (gender equality); Gallup median per-capita income (n=50). Use of sanctions: Single-item self-reports of frequency of using confrontation, gossip, and avoidance on 5-point scale; country means used. Quality checks and deviations: Attention and comprehension checks included. Robustness analyses excluded inattentive or low-comprehension respondents; compared students vs non-students and across cities; results stable. Due to poor reliability of in-survey Hofstede items (α<0.30), official Hofstede Insights country scores were used. Additional unregistered analyses included standardizing sanction metanorms by subtracting non-action metanorms and correlational robustness across domains.

• Universal relation between violation appropriateness and sanction appropriateness (Hypothesis 1): Within countries, across 10 scenarios, higher perceived appropriateness of the triggering behavior was associated with lower appropriateness of sanctions and higher appropriateness of non-action. Mean within-country correlations (95% BCa CI): verbal confrontation M = −0.77 [−0.80, −0.75]; gossip M = −0.67 [−0.71, −0.62]; social ostracism M = −0.39 [−0.44, −0.34]; non-action M = 0.57 [0.55, 0.60].
• Country-level metanorms: Averaged across countries, appropriateness means (95% CI): non-action 2.67 [2.63, 2.70]; verbal confrontation 2.61 [2.55, 2.66]; gossip 2.40 [2.35, 2.46]; social ostracism 2.02 [1.97, 2.07]; physical confrontation 1.80 [1.70, 1.90] (not directly comparable due to different scenarios). No global consensus on the most appropriate response: verbal confrontation rated highest in 26 countries; non-action highest in 31 countries; in 17 of these, gossip was the highest-rated sanction; in Thailand, social ostracism was the highest-rated sanction.
• Robustness across domains (Hypothesis 2): Metanorms based only on the cooperation scenario strongly correlated with main measures: verbal r = 0.79 [0.70, 0.87]; social ostracism r = 0.66 [0.41, 0.81]; gossip r = 0.92 [0.86, 0.96]; non-action r = 0.29 [0.05, 0.50]. Using all 10 scenarios yielded r > 0.9 with main measures. Physical confrontation metanorms based only on the cooperation context correlated with main physical confrontation metanorms r = 0.87 [0.81, 0.92]. Internal consistency across scenarios was generally high.
• Sanction specificity and complementarity (Hypothesis 3): Partial correlations (controlling non-action) showed: physical confrontation and social ostracism were positively associated (consistency); verbal confrontation was largely independent of physical confrontation; gossip was negatively associated with all other sanctions (complementarity), indicating societies viewing confrontation/ostracism as less appropriate tended to view gossip as more appropriate.
• Metanorms predict reported use (Hypothesis 4): Country-level metanorms for specific sanctions correlated well with country-level self-reported frequencies of using the same sanctions, supporting behavioral validity (qualitative result; specific coefficients not reported in-text).
• Cultural, ecological, and economic correlates (Hypothesis 5): For physical confrontation, correlations matched predictions; stronger relations included: power distance r = 0.69 [0.53, 0.81]; individualism r = −0.61 [−0.72, −0.49]; individual autonomy r = −0.53 [−0.71, −0.31]; emancipative moral judgments r = −0.76 [−0.84, −0.66]; tightness r = 0.50 [0.27, 0.69]; gender equality r = −0.72 [−0.81, −0.61]; median income r = −0.67 [−0.78, −0.54]. Verbal confrontation showed weaker and partly inconsistent patterns. Social ostracism resembled physical confrontation. Gossip displayed opposite associations: higher in more individualistic (r = 0.43 [0.21, 0.64]), autonomy-valuing (r = 0.35 [0.11, 0.60]), emancipative (r = 0.54 [0.34, 0.72]), gender-equal (r = 0.49 [0.26, 0.69]) and wealthier countries (median income r = 0.58 [0.40, 0.72]). Non-action tended to be more appropriate in more individualistic and wealthier countries. Scatterplots illustrated median income’s negative relation with physical confrontation metanorms (R^2 = 0.45) and positive relation with gossip metanorms (R^2 = 0.33).

Findings demonstrate both universal and culture-specific aspects of metanorms. Universally, people deem sanctions (confrontation, ostracism, gossip) less appropriate as the triggering behavior is seen as more appropriate, and vice versa for non-action, confirming these responses function as expressions of disapproval. Cross-culturally, metanorms generalize across domains: similar rules apply to cooperative norm violations and to uncivil/out-of-place behaviors, suggesting a broad psychology of informal sanctions not limited to cooperation-specific adaptations. Non-action was often seen as the most appropriate response to relatively minor violations, challenging assumptions that metanorms require punishment. Sanction-specificity emerged clearly. Physical confrontation and social ostracism covaried positively across countries, while verbal confrontation was largely independent. Gossip contrasted sharply, showing complementarity: where confrontation and ostracism were less appropriate, gossip was more appropriate. Metanorms were reflected in self-reported sanctioning behavior, supporting their behavioral relevance. Linking metanorms to culture and economy suggests that cultural values and socioeconomic development shape preferred sanctioning modes. In countries with higher power distance, tighter norms, lower individualism, lower emancipative values, and lower gender equality/income, direct punishment and ostracism were more acceptable. Conversely, in more individualistic, emancipative, gender-equal, and affluent societies, gossip was more acceptable, and direct punishment less acceptable. This pattern is consistent with modernization processes and with theories emphasizing the roles of cultural dimensions and ecological pressures, though perceived threat showed limited associations. The substitution-like relationship between gossip and more punitive sanctions raises questions about how sanction mode affects norm compliance and social harmony.

The study contributes a comprehensive cross-cultural map of metanorms regarding responses to norm violations. It identifies a universal link between perceived violation appropriateness and sanction appropriateness, shows that metanorms generalize across domains, and reveals sanction-specific patterns marked by complementarity between gossip and more punitive responses (physical confrontation, ostracism). Metanorms correlate systematically with cultural values and socioeconomic indicators, suggesting that as societies modernize and adopt emancipative values, preferences may shift away from direct punishment toward gossip. Future research should assess the causal dynamics of these associations (e.g., longitudinal change with economic development), the relative effectiveness of different sanctions for achieving compliance in naturalistic contexts, potential consequences for social cohesion and conflict, and within-country heterogeneity (e.g., urban–rural, socioeconomic strata) in metanorms.

• Stimuli limited to 10 scenarios; although diverse, they cannot capture all violation contexts or severities.
• Appropriateness judgments are hypothetical; while validated against self-reported behavior, real-world sanctioning may differ.
• Convenience sampling focused on urban populations and university-affiliated sites; potential within-country variation (e.g., urban–rural, socioeconomic) was not fully captured.
• Standardization of ratings may introduce slight artificial negative correlations between items.
• Cross-sectional design precludes causal inference; many cultural, ecological, and economic variables are intercorrelated.
• In-survey Hofstede scales showed poor reliability; reliance on external country scores may not reflect recent cultural changes.
• Some measures (e.g., perceived threat) may not comprehensively capture ecological pressures.