Lin Nan, Angele Bernhard, Hua Huimin, Shen Wei, Zhou Junyi, Li Xingshan
CAS Key Laboratory of Behavioral Science, Institute of Psychology, Beijing, China.
Department of Psychology, University of Chinese Academy of Sciences, Beijing, China.
Atten Percept Psychophys. 2018 Feb;80(2):600-607. doi: 10.3758/s13414-017-1444-0.
Previous eye-movement studies have indicated that people tend to skip extremely high-frequency words in sentence reading, such as "the" in English and "/de" in Chinese. Two alternative hypotheses have been proposed to explain how this frequent skipping happens in Chinese reading: one assumes that skipping happens when the preview has been fully identified at the word level (word-based skipping); the other assumes that skipping happens whenever the preview character is easy to identify regardless of whether lexical processing has been completed or not (character-based skipping). Using the gaze-contingent display change paradigm, we examined the two hypotheses by substituting the preview of the third character of a four-character Chinese word with the high-frequency Chinese character "/de", which should disrupt the ongoing word-level processing. The character-based skipping hypothesis predicts that this manipulation will enhance the skipping probability of the target character (i.e., the third character of the target word), because the character "/de" has much higher character frequency than the original character. The word-based skipping hypothesis instead predicts a reduction of the skipping probability of the target character because the presence of the character "/de" is lexically infelicitous at word level. The results supported the character-based skipping hypothesis, indicating that in Chinese reading the decision of skipping a character can be made before integrating it into a word.
以往的眼动研究表明，人们在阅读句子时往往会跳过极高频的词汇，比如英语中的“the”和中文中的“的”。针对中文阅读中这种频繁跳过现象的发生，提出了两种不同的假设：一种假设认为，当预视词在单词层面已被完全识别时就会发生跳过（基于单词的跳过）；另一种假设认为，无论词汇加工是否完成，只要预视字易于识别，就会发生跳过（基于字符的跳过）。我们采用注视点触发显示变化范式，将四字中文词的第三个字的预视替换为高频汉字“的”，这应该会干扰正在进行的单词层面加工，以此来检验这两种假设。基于字符的跳过假设预测，这种操作将提高目标字（即目标词的第三个字）的跳过概率，因为“的”字的字频远高于原字。相反，基于单词的跳过假设预测目标字的跳过概率会降低，因为“的”字在单词层面的词汇搭配不合适。结果支持了基于字符的跳过假设，表明在中文阅读中，跳过一个字的决定可以在将其整合到一个单词之前做出。
