中国科学院心理健康重点实验室知识库

摘要：本研究旨在归纳中文里可以用于测量解释偏差的多义词,并使用多义词造句任务,在广泛性焦虑症状人群中评估其测量有效性.经过对多义词负性与非负性含义常用程度的评分以及多义词造句任务项目区分度、信度、效度和目的隐蔽性的检验,筛选出两含义常用程度大致相同且对焦虑较为敏感的34个题目构成多义词造句任务,该任务信度良好,可以有效测得高低广泛性焦虑和状态焦虑人群中解释偏差的显著差异,且与同类的乱句重排任务相比,能猜出与猜对实验目的的被试显著更少.因此,中文多义词造句任务信效度良好,具有便捷、隐蔽的优势,可以用于解释偏差的相关研究.

This study aimed to search for usable Chinese homographs for interpretation bias measurement and test their measuring validity in individuals with generalized anxiety symptoms using Sentence Constructing Task (SCT). By rating and comparing usage frequency of the negative and benign meanings of the homographs, as well as testing reliability, validity and ability of hiding experimental intentions, a 34-item SCT was generated, in which the homograph items have meanings of approximately identical usage frequencies and anxiety-specific. Results showed that the 34-item SCT had high internal constancy and structural validity in measuring the interpretation bias of generalized anxiety and state anxiety. Compared to the scrambled sentence test, there were significantly fewer participants who figured out the intentions of the SCT, showing that the SCT is a more implicit measurement. In conclusion, the 34-item SCT has good reliability and validity, and can be used as a convenient and implicit measurement in future research.