Abstract: Social interaction is significant for the sound development of personality and the successful socialization of people. As a social cognitive phenomenon, shyness is closely related to social contexts. Clark and Wells proposed a cognitive model of social anxiety, and their proposal that the negative interpretation bias plays an important role in the maintenance of the disorder is widely supported by studies later on. However, comparing with the relatively numerous researches on the interpretation bias of social anxiety and depression in the past few decades, researches on shyness in the field of psychology seem very weak. In this study, we aim to explore whether the shy have this kind of bias .What’s more, we will try to distinguish the online interpretation bias and the offline ones. In experiment1, the text material paradigm was used to explore the shy group’s off-line interpretation bias when faced with the ambiguous situational information. A sample of 60 shy individuals and 61 nonshy individuals went in for the experiment as subjects. They were asked to assess the possibility of interpretations given after the ambiguous situational information from 1-5.The two interpretations are of one positive and one negative. Ascensions of the subjects were recorded. The SPSS 16.0 was used to analyze the data.In experiment2, the modified word sentence association paradigm was used to explore the shy group’s on-line interpretation bias of the ambiguous situational information. A sample of 45 shy and 42 nonshy individuals went in for the experiment. Participants reacted to the words after the ambiguous situational information. They were required to distinguish whether the word given was approving or derogatory. Their RT to give the judgments was recorded. What’s more, the baseline RT was also measured. The baseline RT was the time they react to the word after a blanking screen. It was shown by t test that the difference between groups was found on positive interpretation in experiment 1.While when referring to the negative interpretation, that the difference is not significant. Further analysis showed the shy have no offline positive interpretation bias, also they don’t show more offline negative interpretations bias compared with the nonshy. In experiment 2, we first subtracted the RT and the baseline RT to get the ΔRT .We found that there’s no difference between the ΔRT to the negative words of the shy and the nonshy. Towards the positive words, the effect is significant. Further comparison shows that the ΔRT of the nonshy is smaller than the shy . The results of the two experiments showed: (1) the shy lack both online and offline positive interpretation bias to the ambiguous situational information. (2) The shy has neither offline nor online negative interpretation bias to the ambiguous situational information. In experiment1, the text material paradigm was used to explore the shy group’s off-line interpretation bias when faced with the ambiguous situational information. A sample of 60 shy individuals and 61 nonshy individuals went in for the experiment as subjects. They were asked to assess the possibility of interpretations given after the ambiguous situational information from 1-5.The two interpretations are of one positive and one negative. The sequence of which is randomized. Ascensions of the subjects were recorded. The SPSS 16.0 was used to analyze the data. In experiment2, the modified word sentence association paradigm was used to explore the shy group’s on-line interpretation bias of the ambiguous situational information. A sample of 45 shy and 42 nonshy individuals went in for the experiment. Participants reacted to the words after the ambiguous situational information. They were required to distinguish whether the word given was approving or derogatory. Their RT to give the judgments was recorded. What’s more, the baseline RT was also measured. The baseline RT was the time they react to the word after a blanking screen. SPSS 16.0 was used to analyze the data. It was shown by t test that the difference between groups was found on positive interpretation in experiment 1（t=11.38，df=119，p<.001，rpb2=.521）.While when referring to the negative interpretation, that the difference is not significant（t=-1.59，df=97.824，p>.05，rpb2=.025）. Further analysis showed the shy have no offline positive interpretation bias, also they don’t show more offline negative interpretations bias compared with the nonshy. In experiment 2, we first subtracted the RT and the baseline RT to get the ΔRT .We found that there’s no difference between the ΔRT to the negative words of the shy and the nonshy（t=-.43，df=85，p>.05，rpb2=.002)）.Towards the positive words，the effect is significant（t=2.17，df=85，p<.05，rpb2=.05）.Further comparison shows that the ΔRT of the nonshy is smaller than the shy . The results of the two experiments showed: (1) the shy lack offline positive interpretation bias to the ambiguous situational information. (2) The shy lack online positive interpretation bias to the ambiguous situational information. (3)The shy has neither offline nor online negative interpretation bias to the ambiguous situational information. In this article, we first explore both the offline and the online interpretation bias of the shy. Also, we tried to exclude the social anxiety from the shy before experiment through more specific standard. In addiction，we discussed the possible means to classify the shy from the social anxiety. In experiment1, the text material paradigm was used to explore the shy group’s off-line interpretation features of the ambiguous information. A sample of 60 shy individuals and 61 nonshy individuals went in for the experiment as subjects. SPSS 16.0 was used to analyze the data. In experiment2, the modified word sentence association paradigm was used to explore the shy group’s on-line interpretation features of the ambiguous information. A sample of 45 shy and 42 nonshy individuals went in for the experiment. SPSS 16.0 was used to analyze the data. It was shown by t test that the difference between groups was found on positive interpretation in experiment 1（t=11.38，df=119，p=.000， rpb2=.521）.While when referring to the negative interpretation, that the difference is not significant（t=-1.59，df=97.824，p=.115）. Further analysis showed the shy have less positive interpretations, but they don’t show more negative interpretations compared with the nonshy. In experiment 2, we found that there’s no difference between the ΔRT to the negative words of the shy and the nonshy（t=-.43，df=85，p=.666）.Towards the positive words，the effect is significant（t=2.17，df=85，p=.030， rpb2=.05）.Further comparison shows that the ΔRT of the nonshy is smaller than the shy . The results of the two experiments showed: (1) The shy lack offline positive interpretation. (2)The shy lack online positive interpretation. (3)The shy has neither offline nor online negative interpretation.

Key words: shyness, ambiguous situational information, offline interpretation bias, online interpretation bias

摘要： 以模糊情境句子和形容词为刺激材料，分别采用文本材料范式和改编的词汇情境联接范式，对害羞者的延时解释和即时解释偏向特点进行探讨。结果发现：害羞者对模糊情境信息不仅缺乏积极的即时解释偏向特点，而且也缺乏积极的延时解释偏向特点；害羞者未表现出消极的即时和延时解释偏向特点。研究表明害羞者可能是社交焦虑的前期阶段，对害羞者的干预可以通过引导其积极的解释偏向的方式来实现。

关键词: 害羞者 模糊情境信息 延时解释 即时解释