摘要
实验采用EyeLinkⅡ眼动仪,以30名大学生为被试,采用2(预测性:高、低)×3(空格:正常无空格、词间空格、非词空格)被试内设计,考察了预测性和空格对中文阅读影响,探讨了模块化理论和相互作用理论的争论。结果发现:词的预测性和空格两因素的主效应显著,但二者之间无交互作用,符合模块化理论的预期。
There is a dispute between the modularity theory and the interactive theory in language process.The modularity theory holds that language is composed of low-level factors（including word frequency,word length,phonetic information,prefix,postfix,space and so on） and high-level factors（containing syntax and semantics）.They don＇t interact with each other in processing.Contrarily,the interactive theory maintains that they are interactive.Predictability（a high-level factor） and space（a low-level factor） are important factors in reading,but there aren＇t any studies discussing the theory controversy in terms of these two factors.According to the contents of the theories,we hypothesize that if predictability and space main effects are all significant,but their interactive effect is not significant, the results will favor the modularity theory;if predictability and space main effects are all significant,and their interactive effect is significant, too,the results will favor the interactive theory. T rty Chinese undergraduates participated in the experiment.Their eye movements were recorded with an SR Research EyelinkⅡeyetracker（sampling rate=500 Hz,refresh rate=150Hz） that monitored the position of the right eye every two milliseconds.The experiment was a 2（predictability：high,low）×3（space：normal unspaced,word spaced,nonword spaced） within-subject design. The participant read 90 experimental sentences and 6 practice sentences.The target words were all two-character ones.For the target words,we controlled the character frequency,word frequency and character strokes（see Table1,Table2）. In the data analysis,global analysis and local analysis were conducted.The local analysis contained two parts.In the first local analysis, we compared the target words under normal unspaced and word spaced conditions.In the second local analysis,we compared the regions that comprised a target word and a space under word and nonword spaced conditions. In the global analysis,the main predictability main effects were significant in total fixation counts and total reading time,ps.01. The space main effects were significant in all indicators,ps.05.The total fixation counts indicated processing difficulty.In the total fixation counts,the nonword spaced sentences were more than the spaced sentences and the normal unspaced text.Therefore,the spaced condition was the same as the normal unspaced condition in processing difficulty. In the first local analysis,the predictability main effects were significant in total reading time and probability of skipping,ps.05, and high predictability words＇ fixation duration was shorter and skip probability higher.The space main effects were significant in all indicators, ps.01.In the second local analysis,the predictability and main space effects were significant in gaze time and total reading time,ps.05.Besides,there was no interaction between predictability and space in global and local analyses,ps.05. To conclude,the predictability and space main effects are all significant,and their interaction is not significant.Therefore,the experimental results support the modularity theory.However,we don＇t consider launch distance.So,we think that a more productive approach is needed to discuss the present findings in relation to language processing models of eye movement control.
出处
《心理科学》
CSSCI
CSCD
北大核心
2011年第6期1282-1288,共7页
Journal of Psychological Science
基金
全国教育科学"十一五"规划教育部重点项目(DBA090290)
国家自然科学基金项目(30870781)的资助
关键词
模块化理论
相互作用理论
预测性
空格
眼动
modularity theory
interactive theory
predictability
space
eye movement
作者简介
通讯作者：曹玉肖。E-mail：cyxhope@163．com
