Normative misperception in third-party punishment: An explanation from the perspective of belief in a just world

Shasha Yang,Sijing Chen

doi:10.3724/sp.j.1041.2022.00281

Abstract

<p id=C2>Punishment decisions might be guided by the norm of punishment, that is, people will implement their own punishment according to perceived prevalence of punishment in a similar social midst. However, there may be differences between an individual’s perception of norms and actual norms, which is called normative misperception. This article uses four experiments to explore the existence, the direction, and the cause of the normative misperception in third-party punishment, as well as its influence on people’s own punitive behaviors. <break/>In Experiment 1, 449 participants were randomized in a four group factorial design (punishing before estimating, estimating before punishing, punishing only, and estimating only). Experiment 1 consisted of 6 rounds of dictator game, in which participants made punishment decisions for 6 offers and/or estimated the average punishment level of other participants in each offer. Experiment 2 aimed to establish the causal relationship between the normative misperception and the punishment by directly manipulating the normative misperception. Specifically, 134 participants were randomly divided into the overestimation group and underestimation group. After receiving the feedback, participants made punishment decision for an unfair offer and estimated the level of punishment of others in this offer. The purpose of Experiment 3 was to test the model of belief in a just world (BJW)-normative misperception-punishment, as well as the moderating effect of perceived social distance (PSD), with a within-participants design involving 164 participants. The procedure was similar to that of Experiment 1, except that we measured participants’ BJW and PSD before and after the game, respectively. In Experiment 4, we manipulated participants’ BJW through reading materials to test the causal relationship between BJW and the normative misperception. <break/>The results of Experiment 1 showed that there is an underestimated normative misperception in third-party punishment, which leads to a lower level of punishment. Experiment 2 proved that there exists a causal relationship between the normative misperception and punishment by directly manipulating the independent variables. Experiment 3 demonstrated that BJW might be an underlying cause of the normative misperception, while PSD moderates the effect of BJW on the normative misperception. Finally, Experiment 4 showed the causal relationship between BJW and the normative misperception, providing additional evidence to the results of Experiment 3. <break/>To sum up, we have found evidence of normative misperception in third-party punishment through 4 experiments. This underestimated misperception might be affected by dual reference points: BJW (internal) and PSD (external). It also shows to a certain extent that third-party punishment is a norm-maintaining behavior rather than a gain-based strategic behavior.

Highlights

总体上被试的平均惩罚为 2.31 ± 1.61, 平均估计为 2.17 ± 1.53。基于性别的平均数差异检验表明, 男性和女性的惩罚水平不论是在整体上(t(2050) = 0.14, p = 0.887), 还是在不同方案中 (t(340) = −0.51~0.75, ps = 0.455~0.988), 均不存在显著差异。然而男性对他人惩罚的估计显著高于女性(t(1972) = 2.16, p = 0.031, d = 0.10, 95% CI = [0.01, 0.29]), 这种差异主要体现在 9-1 分配(t(327) = 2.42, p = 0.016, d = 0.27, 95% CI = [0.05, 0.49])和 7-3 分配中 (t(327) = 2.02, p = 0.045, d = 0.23, 95% CI = [0.005, 0.39]) 。不同教育程度和专业下惩罚 (t(340) = −1.44~0.75, ps = 0.152~0.467; F(3, 338) = 0.94~2.33, ps = 0.074~0.422)和估计(t(327) = −0.13~1.17, ps = 0.245~0.897; F(3, 325) = 0.54~2.28, ps = 0.079~0.658)的差异不显著, 且年龄与惩罚和估计之间的相关系数也不显著 (r = −0.08~0.04, ps = 0.142~0.803)。
对估计 1 和惩罚 1、估计 2 和惩罚 2、估计 4 和惩罚 3 进行平均数差异检验。估计 1 和惩罚 1 的配对样本 t 检验结果显示, 在 10-0 (t(107) = −4.14, p < 0.001, d = −0.40, 95% CI = [−0.37, −0.13])、9-1 (t(107) = −4.33, p < 0.001, d = −0.42, 95% CI = [−0.35, −0.13])和 8-2 (t(107) = −3.19, p = 0.002, d = −0.31, 95% CI = [−0.32, −0.07])这三个分配方案中, 个体对他人平均惩罚的估计显著低于实际的惩罚水平, 即存在低估的规范错觉。然而, 当分配方案为 7-3、6-4 和 5-5 时, 估计和惩罚并无显著差异, t(107) = −0.48~1.52, ps = 0.132~0.633, BF01 = 3.10~8.38。由于零假设显著性检验(null hypothesis significance testing, NHST)无法直接给出是否支持零假设的证据, 因此我们用 JASP 0.14.1 计算了相应的贝叶斯因子, 结果显示当前数据更加支持零假设(胡传鹏等, 2018):在后三种方案中不存在规范错觉。对估计 2 和惩罚 2 的检验显示了相似的结果: 方案
的相关系数也不显著(r = −0.04~0.03, ps = 0.680~ 0.914)。表 4 给出了两组被试在规范提示前后上述各变量的描述性统计。通过比较规范提示前估计和惩罚的差异, 可以为实验 1 提供稳健性检验。结果显示:在规范提示之前, 提示高估组(t(68) = −2.56, p = 0.013, d = −0.31, 95% CI = [−0.31, −0.04])和提示低估组(t(64) = −2.55, p = 0.013, d = −0.32, 95% CI = [−0.33, −0.04])中, 均存在低估的规范错觉。

Summary

Introduction

Kamei (2020)指出了感知惩罚规范(即对他人惩罚行为的估计)的重要性, 他们的研究表明, 个体的惩罚行为与感知到的惩罚强度呈正相关。但个体对规范的感知与实际的规范之间可能存在差异 (Rimal & Lapinski, 2015), 这被称为规范错觉 (normative misperception) (Cox et al, 2019)。出于对积极自我形象的追求(Leary, 2007), 个体往往存在着自我服务偏差(self-serving bias) (Zhang et al, 2018), 对他人的贬损恰好可以满足这一心理需求 (Rau et al, 2019)。这也使得规范错觉总是表现为一种系统性的认知偏差:人们倾向于低估他人的积极行为, 例如公众低估了他人的亲环境行为(Bouman et al, 2020; Bouman & Steg, 2019), 青少年低估了同龄人对校园欺凌行为的反对程度 (Dillon & Lochman, 2019), 沙特阿拉伯的已婚男性低估了其他男性对女性劳动参与的支持 (Bursztyn et al, 2020); 而倾向于高估他人的消极行为, 如食物浪费(陈思静, 濮雪丽等, 2021)和酗酒(Amialchuk et al, 2019; Dumas et al, 2019)。因此规范错觉的存在减少了人们从事积极行为的可能性(Dempsey et al, 2018)。由于积极行为(如亲环境行为)以提高群体的福利水平为目的(Sawitri et al, 2015), 但成本需要由个体承担(Keizer & Schultz, 2018), 而第三方惩罚也具备上述特征(Enge et al, 2017), 因此我们可以将其视为一种积极行为, 并推测:个体倾向于低估惩罚的社会规范, 并相应地减少了自身的惩罚行为。

Results

Conclusion