西班牙語第三人稱受詞代名詞與指涉名詞資訊處理之眼動研究

This study investigated the relationship between Spanish third-person object pronouns and referents in the input processing of Taiwanese learners through the analysis of the eye-movement indices in eye-tracking research. It took advantage of the modern technology of an eye-tracking device and expanded upon previous research on acquiring Spanish object pronouns by Taiwanese learners of Spanish. In the experiment, participants' attention to the interest areas related to the target forms in a sentence and their comprehension of Spanish object pronouns and referents was examined and analyzed. The results of this study are concluded as follows: (1) The distance between the object pronoun and its referent in a sentence could affect information processing capacity. When the distance between the clitic pronoun and its referent was larger, the eye regression to the referent facilitated the comprehension of the target forms. When the distance was shorter, the regression suggested participants' hesitation, which was associated with a lower comprehension rate than those who did not demonstrate eye regression. When the distance was shorter, the regression to the object pronouns and frequency would lead to a higher comprehension rate. When the distance was farther, accurate comprehension was related to single or multiple regressions. (2) Regression to the interest areas of the object pronoun and its referent in various positions in a sentence showed different significances. When the distance was closer, the amounts of fixations and fixation duration of object pronouns reflected participants' checking frequency and degree and also affected comprehension. Most importantly, this study establishes the importance of the distance between the clitic pronouns and their referents and how it might affect comprehension of the target forms in sentences. This finding has not been discussed in Spanish second language acquisition research. Future research and instruction on Spanish object pronouns should not overlook this issue.