Ma Guojie, Li Xingshan, Rayner Keith
Key Laboratory of Behavioral Science.
Department of Psychology.
J Exp Psychol Hum Percept Perform. 2015 Oct;41(5):1409-19. doi: 10.1037/xhp0000072. Epub 2015 Jul 13.
We performed 2 eye movement studies to explore whether readers can extract character or word frequency information from nonfixated-target words in Chinese reading. In Experiments 1A and 1B, we manipulated the character frequency of the first character in a 2-character target word and the word frequency of a 2-character target word, respectively. We found that fixation durations on the pretarget words were shorter when the first character of a 2-character target word was presented with high frequency. Such effects were not observed for word frequency manipulations of a 2-character target word. In particular, further analysis revealed that such effects only occurred for long pretarget fixations. These results for character and word frequency manipulations were replicated in a within-subjects design in Experiment 2. These findings are generally consistent with the notion that characters are processed in parallel during Chinese reading. However, we did not find evidence that words are processed in parallel during Chinese reading.
我们进行了两项眼动研究,以探究读者在中文阅读中是否能够从非注视目标词中提取字频或词频信息。在实验1A和1B中,我们分别操纵了双字目标词中首字的字频和双字目标词的词频。我们发现,当双字目标词的首字高频呈现时,目标前词的注视时长更短。对于双字目标词的词频操纵,未观察到此类效应。特别地,进一步分析表明,此类效应仅在目标前的长注视时出现。在实验2的被试内设计中重复了字频和词频操纵的这些结果。这些发现总体上与中文阅读中字是并行加工的观点一致。然而,我们没有找到证据表明中文阅读中词是并行加工的。