Meta-stereotype refers to a person’s beliefs regarding the stereotype that out-group members hold about their own group. The model of intergroup anxiety argues that negative meta-stereotype generates threat by creating negative expectations on the behavior of out-group members. Previous research has demonstrated that intergroup anxiety following stereotype threat contribute to the depletion of working memory and then impairs the performance in the related domain, but the mechanism of the meta-stereotype threat effects on cognition remains unclear. Accordingly, it was common that migrant children have the negative meta-stereotype, however, the influence of negative meta-stereotype on migrant children’s cognition has not been thoroughly investigated. The current study aimed to explore, firstly, the negative meta-stereotype effects on working memory among migrant children, secondly, the mediated role of the intergroup anxiety between meta-stereotype and working memory, thirdly, the moderated role of difficulty of working memory task on negative meta-stereotype effects.

A total of 90 migrant children participated in the present study. Participants were instructed to write descriptive adjectives to evoke negative meta-stereotype or not according to different instructions. Then, participants were assigned to either the meta-stereotype threat (MST) condition (25 males, 24 females, aged 11~13 years, M = 12.05, SD = 0.83) or the non-MST condition (21 males, 20 females, aged 11~13 years, M = 12.24, SD = 0.79). This study was organized into a 2×3 design. The first factor was the type of meta-stereotype, consisting of 2 levels, MST condition and non-MST condition. The second factor was the type of working memory, including 3 levels, 0-, 1-, and 2-back working memory tasks. The participants also completed the intergroup-anxiety scale. Analysis of variance (ANOVA) and mediation analysis were used to analyze the data.

The following results were observed: (1) The intergroup anxiety under the MST condition was higher than the case of non-MST condition. (2) The accuracy of the MST condition was significantly lower than that of the non-MST condition across the 0-, 1-, 2-back working memory tasks. Moreover, the accuracy in the MST and the non-MST condition both decreased significantly with the task difficulty growing. (3) In the 0-, 1-, 2-back working memory tasks, the reaction time of participants in the MST condition was significantly higher than those in the non-MST condition, and in the MST condition, the reaction time significantly increased with the increase of difficulty, the interaction between the group and the type of working memory was significant. (4) The relationship between the meta-stereotype and the accuracy of working memory (medium difficulty) was entirely mediated by the intergroup anxiety. The relationship between the meta-stereotype and the reaction time of working memory (medium difficulty) was partially mediated by the intergroup anxiety.

Overall, it was suggested that the deleterious effects of negative meta-stereotype on working memory and the intergroup anxiety among migrant children contributed to the explanation of the decline of working memory at medium level of difficulty. Further study still needs to precisely assess other factors (i.e., prejudice, social identity, and so on) that are involved in the threat effects of the meta-stereotype.