关键词: 社会阶层, 普遍信任, 风险来源, 背信规避, 社会认知理论

Abstract: With the continuous solidification of social class, people have fair bias and implicit attitude towards high and low social class, which is of great significance to social governance and development. Therefore, two experiments were conducted to explore the fair bias and implicit attitude of individuals towards different social classes. In experiment 1, we used the Implicit Association Test (IAT) to examine the individual's fair bias, that is, whether individuals classified high social class and fair words into the same category, and low social class and unfair words into the same category. Experiment 2 used the same method to examine the influence of social class on the implicit attitude of individuals, that is, whether individuals classified high social class with positive words and low social class with negative words. Used the questionnaire to screen 5 target words (high social class vocabulary / low social class vocabulary) for Implicit Association Test (IAT), 6 attribute words (fair vocabulary / unfair vocabulary), positive vocabulary / negative vocabulary 5 related words from the classic implicit association test study were used. The experimental procedure was adapted from the standard IAT experimental procedure of Greenwald (1998). Before the experiment began, the subjects were told that "this is a test of reaction speed." At the beginning of each task, the instructions for the task were presented on the computer screen, and the participants were required to press the space bar to carry out the formal experiment after they fully understand the instructions. The results showed that :(1) in the implicit association task of compatible tasks (high class-fairness, low class-unfairness) and incompatible tasks (high class-unfairness, low class-fairness), the response time of the compatible tasks (M = 799.89, SD = 203.05) were significantly faster than the incompatible tasks (M = 887.77, SD = 251.11), t(35) = -2.39, p = 0.022, d = -0.40. Meanwhile, according to the calculation method of the effect value D proposed by Greenwald (2003), 72.22% of the participants of the D value is greater than 0, the participants of the IAT effect value D distributed in 0.01~ 0.94, the D value with 0 for single sample t-test results showed that the mean (M = 0.17, SD = 0.41) than 0, t (35) = 2.41, p = 0.021, d = 0.40, with IAT effect. (2) in the implicit association task of compatible tasks (high class-positive, low class-negative) and incompatible tasks (high class-negative, low class-positive), the response time of the compatible tasks (M = 688.80, SD = 124.98) were significantly faster than the incompatible tasks (M = 1023.67, SD = 241.81), t(37) = -10.63, p < 0.001, d = -1.72. In addition, the participants' IAT effect value D was distributed between 0.17~1.18, and the results of single sample t test of D value at 0 showed that its mean value (M = 0.64, SD = 0.27) was significantly greater than 0, t(37) = 14.67, p < 0.001, d = 2.38, with IAT effect. On the implicit level, individuals were more likely to associate high social class with fair and positive words, and low social class with unfair and negative words. This study finds that people do not have a "hatred" of rich people, but an implicit cognitive evaluation of their fairness and positive. Therefore, people have a social class preference, that is, to hold a more positive and fair implicit cognition to the higher social class, which further enriches the social class psychology.

Key words: social class, generalized trust, source of risk, betrayal aversion, the socio-cognitive theory