左右结构对称汉字的认知神经加工机制研究

发布时间：2018-05-04 08:54

本文选题：结构对称汉字 + 家族效应　；参考：《西南大学》2014年硕士论文

【摘要】：结构对称汉字具有视觉上的对称性,这种特殊的对称结构给汉字带来一种知觉加工优势。汉字作为表意文字,以形表意,它词汇特征(如结构方式、部件、笔画等)在汉字识别加工过程中起着重要作用。根据以往的行为实验研究,汉字的结构对称效应即与非对称字相比,结构对称汉字的识别速度更快,正确率更高。为了更加突出汉字结构的对称性,本研究对结构对称汉字的定义更严谨,即要求结构对称汉字是以字的中心线为轴折叠左右部分可以完全重叠的汉字,也叫左右结构对称汉字(如,黄等)。很多关于词汇认知加工过程的ERP研究中,都会发现N400成分的出现,所以普遍认为N400与词汇的语义加工有关,但是也有很多研究发现词汇特征(如词频、词的具体性、词的真假性和正字法家族规模大小)可以影响N400成分的波幅大小。汉字的家族大小,一般是指改变汉字的某些笔画和部件所能构成的汉字个数。汉字的结构对称效应提示我们结构信息在视觉汉字认知过程中起着重要作用,所以本研究认为结构对称汉字的一般家族成员里还有一类次家族成员,即基于汉字的这一特殊结构信息组成的左右结构也对称的家族成员,本研究也将其称之为真家族,那么非对称汉字的真家族就是结构也不对称的家族成员。通过对《现代汉语字频统计表》中的汉字进行统计分析,发现在0.0015%-0.0418%的频率范围内左右结构对称汉字的比例只占3.8%,所以本研究认为左右结构对称汉字的真家族成员比非对称字的真家族成员少。实验一采用语义分类任务,记录和分析左右结构对称汉字和非对称汉字两种水平下的脑电数据,考察结构对称汉字的ERP效应。基于汉字结构信息的真家族假设,对左右结构对称汉字和非对称汉字两种实验材料的真家族成员进行统计分析,发现左右结构对称汉字真家族成员显著少于非对称汉字的真家族成员[t=5.43,p0.05]。假如实验结果诱发出了与家族效应有关的几个ERP成分P150、P200和N400,则表明基于汉字的这一结构信息组成的真家族成员假设是可以被考虑的,否则表明并不存在这样的对称结构家族。实验二采用押韵判断任务,记录和分析左右结构对称汉字和非对称汉字两种水平下的脑电数据。实验材料与实验一同。一些ERP成分对不同的实验任务不敏感,本实验的目的是考察使用押韵判断任务,结构对称汉字的ERP效应是否与实验一的结果一致,如果是,则进一步为基于结构信息的汉字真家族假设提供了证据,也进一步表明汉字的结构对称性不仅在视觉特征加工阶段起作用,也参与了语义信息的提取加工阶段。结果表明：(1)在ERP研究中汉字的结构对称效应依然寻在,具体表现为一系列的ERP效应(N170、P150、P200和N400),结构对称汉字比非对称汉字的真家族成员少,在认知加工中受到的干扰更少,从而更容易被识别。即对结构对称汉字的加工是通过早期的家族成员激活和后期词汇的干扰作用来完成的；(2)汉字的结构对称特征在汉字加工的早期正字法加工阶段和后期语义加工阶段都起着重要作用；(3)N170对汉字的结构信息也比较敏感,说明N170可能参与了视觉信息的加工,P200和P150成分是两个相似的成分,都反映了词汇早期正字法加工,N400成分也反映词汇表征信息的加工。
[Abstract]:Symmetrical Chinese characters have visual symmetry, and this special symmetrical structure brings a perceptual processing advantage to Chinese characters. As a ideographic character, Chinese characters are ideographic, and its lexical features (such as structure, components, strokes, etc.) play an important role in the process of Chinese character recognition and processing.
According to the previous study of behavioral experiments, the structural symmetry effect of Chinese characters is faster and more correct than asymmetric words. In order to highlight the symmetry of Chinese characters, the definition of symmetrical Chinese characters is more rigorous in this study, that is, the symmetrical Chinese characters are required to fold the center line of the word as the axis to fold the left The Chinese characters that can be completely overlapped on the right part are also called symmetrical Chinese characters (such as yellow).
A lot of ERP studies on cognitive processing of vocabulary will find the appearance of N400 components, so it is generally believed that N400 is related to the semantic processing of vocabulary, but there are many studies that have found that lexical features (such as word frequency, word concreteness, true and false word, and family rule size of orthography) can affect the amplitude of N400 components. Family size, generally refers to the number of Chinese characters that change some strokes and parts of Chinese characters. The structural symmetry of Chinese characters suggests that our structural information plays an important role in the cognitive process of visual Chinese characters. Therefore, this study holds that there is a class of family members in the general family members of the symmetrical Chinese characters, that is, Chinese characters. This special structure information consists of family members with symmetrical structure and symmetrical structure. This study also calls it true family. Then the true family of asymmetric Chinese characters is a family member with asymmetric structure. By statistical analysis of the Chinese characters in modern Chinese word frequency statistics, it is found in the frequency range of 0.0015%-0.0418%. The proportion of symmetrical Chinese characters only accounts for 3.8%. Therefore, this study holds that the true family members of the left and right symmetrical Chinese characters are less than the true family members of the asymmetric characters.
In Experiment 1, the semantic classification task was used to record and analyze the EEG data of two levels of symmetrical Chinese characters and asymmetric Chinese characters, and to investigate the ERP effect of structural symmetrical Chinese characters. Based on the true family hypothesis of Chinese character structure information, the statistics of the true family members of the two experimental materials of symmetrical and asymmetrical Chinese characters and asymmetric Chinese characters were statistically divided. It is found that the true family members of the left and right symmetric Chinese characters are significantly less than the true family members [t=5.43 of the asymmetric Chinese characters. P0.05]., if the experimental results lure several ERP components related to the family effect P150, P200 and N400, indicates that the true family members based on the structure information of the Chinese character can be considered, otherwise the table can be considered. There is no family of such symmetrical structure in Ming Dynasty.
In experiment two, the rhyme judgment task was used to record and analyze the EEG data under two levels of symmetrical and asymmetrical Chinese characters. Experimental materials were together with experiments. Some ERP components were not sensitive to different experimental tasks. The purpose of this experiment is to investigate the use of rhyme judgment, whether the ERP effect of the structure of symmetrical Chinese characters and the experiment one is one. The result is consistent. If it is, it further provides evidence for the hypothesis of Chinese character real family based on structural information, and further indicates that the structural symmetry of Chinese characters not only plays a role in the processing stage of visual features, but also participates in the extraction and processing stage of semantic information.
The results are as follows: (1) the structural symmetry effect of Chinese characters is still found in the ERP study, which is characterized by a series of ERP effects (N170, P150, P200 and N400). The structural symmetry Chinese characters are less than the true family members of the asymmetric Chinese characters, and the interference in cognitive processing is less, and it is easier to be identified. Early family members activation and the interference of later vocabulary; (2) the structural symmetry of Chinese characters plays an important role in the early orthographic and semantic processing stages of Chinese character processing; (3) N170 is also sensitive to the structural information of Chinese characters, indicating that N170 may be involved in the processing of visual information, P200 and P150 components are two similar components, which reflect the early orthographic processing of words. N400 components also reflect the processing of lexical representation information.

【学位授予单位】：西南大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：B842.1

【参考文献】