中文复合远距离联想测验的编制
详细信息    查看官网全文
摘要
根据Mednick提出的语义联想理论,创造性过程被定义为一种将不同概念通过远距离联想联系在一起的能力。基于该理论,Mednick编制了不依赖与特定领域知识的远距离联想测验(Remote Association Test),该测验要求受测者找出一个能与这三个词组成关联字对的词语。由于语言和文化背景的差异,国外编制的远距离联想测验无法直接翻译使用。而且其题目数量较少,难以满足需要大量实验试次的认知神经科学实验。本研究建立了适合我国语言文化的中文复合远距离联想测试(CCRAT),为国内相关领域研究者进一步探索远距离联想测验的认知神经机制提供标准化数据。中文复合远距离联想测试题目由三个汉字组成,这三个汉字能够组成(如:板、洞、色),有且只有一个能够与前面三个字组成合法双字词的字作为答案字(如:黑)。其中,所有词语均来源于《现代汉语词频词典》中的高频词,以排除语言水平干扰。指导语和题目使用E-prime程序在计算机上呈现和收集被试的答案和答题时间,并计算正确率和反应时。经过预测验筛选,选出135道中文复合远距离测验题目。随后,将135道中文复合远距离测验题目与《托兰斯创造力测验(成人简版)》同时施测,测量结果显示,135道中文复合远距离联想题目信度为0.897(Cronbach's α),与托作为校标比较,与其总分有显著相关(r=0.465,p<0.01)。该测验难度为0.66,属于中等偏偏下难度,平均回答所需时间为3785毫秒,该测验难度适合进行认知神经科学研究或作为标准化题库。
Based on Mednick's associate theory, the process of creativity was defined as an ability to combine different concepts together by association. Mednick developed the Remote Association Test(RAT) as a means of measuring creativity without requiring knowledge specific to any field. Each item consists of three words that can be associated with a solution word. However, the RAT test developed by Mednick cannot be directly translated for the different linguistic and cultural backgrounds between countries. In order to provide applicable Remote Association test items to Chinese and its normative data, the 135 Chinese Compound Remote Association Test(CCRAT) is developed. Participates were instructed to find an answer word which could be a meaningful words with other words. Each item consists of three words which were chosen from The Modern Chinese word frequency dictionary. The high frequency word is preferred for the reason that participates would understand them easily. Based on the difficulty and discrimination parameters in a pretest, 135 quality items(Difficulty was 0.66) were selected and used to test the screening samples together with Abbreviated Torrance Test for Adults. The result showed that CCRAT has a good internal consistency(Cronbach's α = 0.897) and the test has a good criterion validity, with a significance correlation to TTCT(r=0.465, p<0.01).
引文