Abstract: Abstract The purpose of this study is to investigate the inhibition of return (IOR) on self-twist face, which might further look into the processing on self-threat stimuli. Cue-target tasks were conducted in both two experiments. In Experiment 1, 28 college students are presented self-twist face, other’s twist or neutral face as the cue each time. In Experiment 2, 22 college students were presented self-twist face, other’s twist or neutral face as the target each time. The participants were asked to react to the target (a black square in Experiment 1 and a face in Experiment 2) via pressing the related keys followed by the different cue (a face in Experiment 1 and a black square in Experiment 2) as quickly and precisely as possible. At the end of the experiment, the participants were asked if the presenting twist faces or neutral faces were their own or not. ANOVAs were conducted on reaction time for each experiment. In Experiment 1, the results revealed significant main effect on cue location [F(1,27)=50.629，p<.000，η2=.652] and the reaction time on uncued location (393.637±6.692) is significantly lower than cued location (406.687±6.789), which indicates significant IOR effect. The main effect on SOA [F(2,54)=127.434, p<.001, η2=.825] was significant and with the increase of SOA the reaction time decreased significantly. There was significant main effect on face configurations [F(2,54)=3.884，p=.027，η2=.126] either and the paired comparison showed that the reaction time on self-twist face ((366.2008±8.868)) was marginally faster than others’ neutral face (370.028±9.446), which means that attentional bias occurred on self-twist face. Furthermore, the reaction time on self-twist face was the fastest on cued location and the slowest on uncued location, which indicates the facilitated attention to self-twist face. As for the tests on the amounts of IOR among three face configurations, there was significant main effect on face configuration [F(2,54)=5.742，p=.005,η2=.175] and the paired comparison showed that the effect of IOR on self-twist was the most least among the three kinds of faces which means that self-twist may seize the resource of attention dramatically. In Experiment 2, there were also significant IOR main effect [F(1,21)=27.321，p<.000， η2=.565] and significant SOA main effect [F(2,42)=109.072,p<.000,η2=.839] on the whole. There was significant face configuration main effect [F(2,42)=14.493,p<.000,η2=.408] and the reaction time on self-twist face was also the fastest. There was significant interact effect between face configuration and cue location [F(2,42)=11.317,p<.000,η2=.350]. Simple effect test showed that the reaction time for self-twist target on cued location was significantly faster than uncued location which indicates that the IOR disappeared when the self-twist face was on target location. As for the tests on the amounts of IOR, there was no significant effect between other’s twist and neutral face. The results about this research indicate that self-twist face is different form other’s twist face and the former means the self-threat stimuli. Compared to the face cues, there comes stable attention control when the face is arranged as the target. The face configuration has significant effect on IOR both as cue and as target, which means that IOR is ascribed evolutionary signi?cance and should not reduce our chances of noticing event information that could be relevant for our well-being or survival, and in particular, human faces or threatening events. According to the two experiments in this study, it could be speculated that the self-threat stimuli activates the threat to self-concept (i.e. the negative attributes on self-concept) and therefore arouses the motive to self-protect which result in attention bias. No matter the button-up unconscious activation or the top-down active processing, the threatening self-concept always narrows the attentional space and the visual searching field which caused the boundary of self to be narrowed. This study has revealed that the threat to self-concept could be activated easily and even is impossible to defend effectively. The susceptibility on self-threat is probably the potential mechanism on self-cognition, which causes the human relationship issues and other social adaption problems. Future study is going to be focused on the self-threat environment to see the effect on social cognitive processing.

Key words: self-face, self-threat, visual search, inhibition of return, attention bias

摘要： 使用线索-靶子范式，分别将面孔刺激作为线索（实验1）和靶子（实验2）呈现，考察自我扭曲面孔对返回抑制的影响。结果发现：自我扭曲面孔为线索时，与他人扭曲、他人中性面孔产生了实验性分离，自我扭曲面孔的返回抑制效应量最小；自我扭曲面孔为靶子时，返回抑制消失。研究结果揭示：以自我扭曲面孔为代表的自我威胁刺激激活了威胁性自我概念，缩小了注意搜索范围，缩小了自我边界，这可能是导致人际紧张的潜在自我认知机制。

关键词: 自我面孔, 自我威胁, 视觉搜索, 返回抑制, 注意偏向