关键词: 防御性自尊, 攻击性, 注意偏向训练, 注意偏向, 记忆偏向

Abstract: Abstract Individuals with defensive self-esteem is associated with aggressive and defensive behavior. Defensive self-esteem is characterized by processing biases for aggressive information. Processing biases for aggressive information may be maintenance factors for aggressive behavior. Despite a wealth of studies examining the processing biases (at the level of attention and memory) for aggressive information in individuals with defensive self-esteem, little is known about relationships between different processing biases. The purpose in this study was to explore whether attention bias has an impact on memory bias in college students with defensive self-esteem. Participants were asked to respond to Rosenberg’s (1965) Self-Esteem Scale and completed the implicit association task (IAT). According to the results of the tests, the defensive self-esteem group (a combination of high explicit and low implicit self-esteem) was selected. And a total of 61 participants were selected. 61 participants were randomly assigned to a attention training (n = 31) or control group (n = 30). The attention modification program (AMP) was followed by a memory task during which novel aggressive and neutral words were presented. Memory for words presented in memory task was tested using incidental free recall. In attention task, attention bias index was subjected to a 2 (group: training group vs. control group) × 2 (measurement time: pre-training vs. post-training) mixed-factors ANOVA. The interaction between group and measurement time reached significance [F(1,59) = 10.23，p = .002，ηp2 = .148]. Further analysis indicated that at pre-training there was no difference in attention bias index between the two groups. But there was a significant difference in attention bias index between the two groups at post-training. Specifically, the attention bias index of the training group was significantly lower than that of the control group. In addition, the training group showed a significant reduction in attention bias index from pre- to post-training, while the difference of attention bias index of the control group from pre- to post-training was not significant. For the memory task, the number of recalled items were subjected to a 2 (group: training group vs. control group) × 2 (word valence: aggressive vs. neutral) mixed-factors ANOVA. The interaction between group and word valence reached significance [F(1,59) = 8.34，p = .005，ηp2 = .124]. Further analysis indicated that training group had significantly worse recall than control group for aggressive words, but not for netural words. In addition, control group showed enhanced memory for aggressive words compared with netural words, whereas training group showed no memory bias. These results indicate that AMP could directly change automatic attention process that is hypothesized to contribute to the development and maintenance of aggressive behavior. Attention bias could be one of the causes of trait-congruent memory in college students with defensive self-esteem. Further researches should examine whether the training generalizes over time and can also be applied successfully in college students with defensive self-esteem.

Key words: Defensive self-esteem, Aggression, Attentional bias modification, Attention bias, Memory bias