Abstract: For late second language (L2) learners, some researchers argued that L2 proficiency plays an important role on the L2 syntactic processing. However, some other researchers argued that first language-based experience determines the L2 syntactic processing. Until to now, few studies strictly manipulated both L2 proficiency and language similarity simultaneously, so that the effects of L2 proficiency and language similarity may be confused with each other. Therefore, it is still open if the modulating effect from L2 proficiency is different for different L2 structures which may have different relationship with L1. In the present study, we constructed two types of syntactic violations based on the similarity of Mandarin to English syntactic structures: 1) a similar syntactic violation, which exists in both Chinese and English and is expressed in a similar manner, such as predicate verb category violation, where the main word order in both languages is SVO (subject-verb-object) and a subject must be followed by a verb or verb phrase; 2) a different syntactic violation, which only exits in English, but not in Mandarin, such as subject-verb agreement violation, as this does not exist in Chinese. A group of high-proficiency English learners and a group of low-proficiency L2 learners with Chinese as native language were presented with English sentences including above mentioned syntactic violations. The EEG data were collected when the participants performed a semantic judgment task, which was designed to make the task more natural reading comprehension. The repeated ANOVAs were run on the mean amplitudes computed on three consecutive time windows (150-300ms, 300-550ms, 550-800ms). ERP data analysis showed that similar violation elicited a positive-going left-anterior distributed ERP response lasting from 150 to 800ms in high-L2 proficiency learners, but there was no any ERP effect for different violation both in high and low-proficiency learners. Behavioral data showed that there were significant differences between different L2 proficiency participants and different syntactic violations, namely, the accuracy of semantic judgment of the high-L2 proficiency learners was higher than low-L2 proficiency learners, and the accuracy was significantly lower when the sentences included similar syntactic violation. These results suggest that high-L2 proficiency learners are more sensitive to the similar syntactic violation than the different syntactic violation during sentence reading, and similar syntactic violation has a stronger effect on sentence comprehension. The interacting effects between L2 proficiency and language similarity reveal that L2 proficiency plays different modulating roles on different L2 syntactic structures, i.e., its modulating effect was significant on similar structures, but not on different structures. The effects of syntactic similarity revealed the mechanism that native language-specific experience has an effect on L2 leaning and processing. L2 proficiency also plays an important role during L2 syntactic processing, but its effect is confined by the L1 experience. Taken together, the results of the present experiment support the proposal that L1 experience influences L2 learning and processing.

摘要： 采用事件相关电位（ERP）技术，考察了汉、英句法结构相似性与第二语言（英语）熟练度对第二语言句法加工的影响。实验材料包含与汉语句法结构相似和不同的两种英语句法结构违反的句子。结果发现：与正确句子相比，相似句法违反句在高熟练组被试中，引发正向的ERP波形偏转（150-800ms），而在低熟练组被试中，只引发了降低的N400（300-550ms）效应；对于不同句法违反句，无论高低熟悉程度的被试，均未出现显著的ERP效应。研究结果表明，第二语言句法与母语的相似性越高，加工起来更容易；第二语言熟练度水平对第二语言句法加工具有一定的促进作用，但受到两种语言句法结构相似性的影响，相似性越高，熟练度的促进作用越大。

关键词: 句法加工, 句法相似性, 第二语言的熟练度, Syntactic processing, syntactic similarity, L2 proficiency