ISSN 0439-755X
CN 11-1911/B
主办:中国心理学会
   中国科学院心理研究所
出版:科学出版社

心理学报 ›› 1990, Vol. 22 ›› Issue (2): 31-38.

• • 上一篇    下一篇

汉字的视知觉——知觉任务效应和汉字属性效应

喻柏林;冯玲;曹河圻;李文玲   

  1. 中国科学院心理研究所,中国科学院心理研究所,中国科学院心理研究所,北京师范大学心理系 北京 ,北京 ,北京
  • 发布日期:1990-06-25 出版日期:1990-06-25
  • 通讯作者: 喻柏林

VISUAL PERCEPTION OF CHINESE CHARACTERS-EFFECT OF PERCEPTUAL TASK AND CHINESE CHARACTER ATTRIBUTES

Yu Bolin;Feng Ling;Cao Heqi;Li Wenling Institute of Psychology, Academia Sinica, Beijing Beijing Normal University, Department of Psychology   

  • Online:1990-06-25 Published:1990-06-25

摘要: 本研究通过三个实验发现:1.在三种非速示的特定知觉任务下,认知汉字反应时的长短依次是;识别汉字两个部件的反应时>识别整体字的反应时>辨认字的结构方式的反应时,在错误率指标上也获得了同样的结果模式。这表明,在汉字的视知觉中存在知觉任务效应,而这些任务标志着不同复杂程度的加工水平。2.对于左右和上下结构的字,在辨认其结构方式上无差异。3.在命名整体字中,汉字的表层属性,如结构方式和表音特点对整体字的识别没有影响,这表明整体字是正常阅读条件下的知觉单位。4.在知觉分解整体字和命名识别其部件中,(1)字的结构方式有相当大的影响:上下结构字的反应时远远长于左右结构字的反应时。(2)字的表音特点对部件识别也极有影响,如形声字的反应时和错误率都远远短于和少于非形声字。(3)在误操作中,存在着大量整体字的命名现象。以上表明的汉字属性效应,都是发生在汉字知觉的分解过程中,它们反映了整体对部分的强烈影响。

关键词: NULL

Abstract: Three experiments were carried out to study the visual perception of Chinese characters. The results show that: 1) the order of response time (RT) in the recognition of Chinese cha- racters was: identifying the two components of a character>identifying the whole character>identifying the construction of the character. A si- milar results was obtained for the error rate. These results suggest that there was a perceptual task effect on the recognition of Chinese chara- cters, suggesting that these tasks represented different levels of processing complexity. 2) There was no difference between discriminating the left-rightco- nstruction characters and the up-down construction characters. 3) The surface attributes of a Chinese character such as its constructive and phonetic features did not have an influence on the recognition of the whole character, suggesting that the whole character was the perceptual unit in normal reading. 4) In decomposing the whole character and naming its components, the constructive and/on the phonetic features of a character had a great effe- ct: RT for the up/down characters was much longer than that for the right/left characters; RT for the pictophonetic characters was shorter than that for the unpictophonetic characters and the error rate for pictophon- etic characters was much smaller than for the unpictophonetic characters. The typical error was to name the whole character when the task was to name the components of the character

Key words: NULL