Word Segmentations And Words Processing Style In Chinese Reading

The models of eye movment control for alphabetic writing languages were all word based. However, unlike alphabetic languages, Chinese texts consist of characters which take the same amount of horizontal extent, and there are no explicit boundaries between the words. Whether or not the models of eye movment control for western languages reading fit to Chinese reading directly was still an open question. Hence, the researches aimed on the control of eye movements in Chinese reading are theoretical and practical value.In this dissertation, we emploied three researches aimed to resolve the confuses on fundamental units (word or character) and processing stly (parallelly or sequentially) in Chinese reading.The first research embodied two experiments (Experiment1&Experiment2)which manipulated the disappearing unit (characters or word) and the locations of the word disappeared respectively. The delay time in all the disappearing conditions of both experiments was50-55ms or so. We aimed to explor the characters processing style (parallelly or sequentially) in Chinese reading. We found all the disappearing conditions do not interrupt total reading time in Experiment1. In Experiment2, there are four disappearing conditions:word n-1, word n, word n+1, and the word n+2disappearing conditions, we found that word n, word n+1, and the word n+2disappearing conditions do not interrupt reading, but word n-1disappearing condition do. Thus the results of this research supported the point of "Chinese readers parallel process the characters in perceptual span"Give that Chinese readers’perceptual span typically contains3to4words, the problem on prime unit involved word segmentations. The second research aimed to explor number, locations and the time course of word segmentations. There were four experiments in this research. The color of text related chanced changed from red to black or from black to red in these experiments, these manipulations facilitated or interrupt the word segmentations. Experiment3&Experiment4were conducted to explore the locations and the number of word segmentations. The manipulations in experiment3provided cues to facilitate word segmentations, but manipulat ions in exper iment4did not. We found that Chinese readers always segmented word n+1from its subsequent texts as they fixated on word n firstly, there was also a segmentiation of word n, and that readers continued to segment the word as it was fixated on if the first segmentation failed. Given that segmentiation of word n+1was the main form in Chinese reading, Experiment5&Experiment6were conducted to explore the time course of word segmentation on word n+1, in which we manipulated the delay time when the text changed their color. The results of these two experiments showed that there was a time delay effect when word n+1changing its’color, but not for the two-character unit.The first two researches could not clarify whether or not Chinese readers process the texts in word n+1by word unit. The third research was conducted to clarify this confuse. All of the mask conditions in Experiment7provide effects of word segmentations, beside this, they also deprive of preview benefit effects from word n+1, word n+2or both. We found that the condition which provided effects of word n+1segmentation at same time deprive of preview benefit effect from word n+2inflenced the total reading time, however, the mean gaze duration in this condition was less than the control. Experiment8was conducted which adopted similar treatments as Experiment7, but the two adjacent characters not belonging to a word were masked together. That was the manipulations did not provide cues to facilitate word segmentation. Consequently, all the masked conditions inflenced the total reading time and fixation time more seriously than those conditions in the experiment7, and mean gaze duration based on word or two characters interest in mask display conditions were also more than the control. Thus, the results obtained in Experiment7and Experiment8indicted benefit effects from the text in word n+1was based on word unit.We also found an significant previw effect on word n+2in Experiments7. In a word, the data from current researches fit better with the theoretical assumptions of parallel processing models of eye movements control in reading, the current researches confirmed that Chinese readers processe the characters in perceptual span parallelly, and that they parelle process word n and word n+1by word unit.