任务转换和目标变化下的自我加工优势

本研究旨在通过联结学习任务和线索转换任务探讨自我关联刺激在任务转换和目标变化下的加工优势。结果发现：（1）相比于朋友和陌生人，自我更容易学习联结；（2）在线索转换任务中，所有关联刺激均未出现任务转换成本；（3）在目标变化条件下，自我关联刺激的目标结转效率最佳。结果表明自我关联刺激能够调节任务转换成本，并且在目标变化条件下能快速聚焦注意，表现出基于感知的特异性加工优势。

The Prioritization of Self-associated Stimuli on Task Switching and Target Shifting

Information associated with the self is prioritized relative to information associated with others and is therefore processed more quickly and accurately. Previous researches have found that self-associated stimuli would rapidly activate self-concept and enhance social attentional orientation in the limited cognitive resources, thus showing the prioritization at the level of perceptual processing. However, the other researchers proposed that the source of the self-bias was more consistent with a response bias rather than a perception bias. Although the conclusions of the above researches are not consistent, it is undeniable that self-associated stimuli have a self-processing bias in attention, memory, and response. Here we used an improved task-switching paradigm to investigate the self-prioritization in the condition of task-switching and target shifting. Based on the task-set reconstruction theory, we investigated the prioritization of self-associated stimuli through the moderating effect of self-associated stimuli on task-switching costs. We further explored the perception bias of self-associated stimuli by comparing the carry-over rate under the condition of target shifting. The present study reported an experiment to assess the prioritization of the self-associated stimuli on task switching and target shifting. We combined an associative learning task with a task-switching task (categorizing judgment task). 27 participants participated in the experiment. In the associative learning task, three geometric shapes (circle, square, & triangle) were randomly assigned to three labels (self, friend, and stranger). Having formed a personal association, participants performed the categorizing judgment task. Participants were predictably informed on the rule of each task by pre-cues. The rules were either "self/ non-self" or "familiar/ unfamiliar". Participants needed to judge which option the shape belongs to. Reaction time and accuracy were measured when participants conducted the tasks and analyzed using repeated-measures analyses of variance (ANOVAs) with 2(judgment: familiarity, ego) ×2(task type: switch, repetition) ×3(association type: self, friend, stranger) as within-subjects variables. Based on the study of Huber-Huber and Anorge (2017), the carry-over rate can be measured by the relative percentage deviation between 〖"RT" 〗_"n-1" and 〖" RT" 〗_"n" , i.e., (〖RT〗_(n-1)-〖RT〗_n)/〖RT〗_n ×100%. Considering the speed-accuracy tradeoff, we performed repeated-measures ANOVA of 2(task type: switch, repetition) ×2(association type: self, stranger) ×2(change type: stick, shift) before and after the self-associated stimulus and the stranger associated stimulus.The above tasks mainly obtained the following results: in the associative learning task, the performance for self-associated shape is superior to that for friend-associated shape (accuracy: p = .001; reaction times: p < .001) and other-associated shape (accuracy: p< .01; reaction times: p< .001). In the categorizing judgment task, none of the associated stimuli showed the task-switching effect. Specifically speaking, no matter what kind of associated stimuli, there is no difference in search performance between switch task and repetition task. But for the carry-over rate, a significant interaction between association type and change type. Specifically speaking, for self-associated stimuli, the carry-over rate with the target sticking is smaller than that with target shifting; for the stranger-associated stimuli (p= .001), the carry-over rate with target sticking was significantly higher than that with target shifting (p< .001). It can draw the following conclusions: self-associated stimuli can regulate the cost of task switching, and quickly focus attention under the condition of target shifting, showing a perception-based specific processing advantage.