语境中语块的加工及其影响因素——以中级汉语学习者为例

The Processing of the Contextualized Formulaic Sequences:A Case Study of Intermediate Chinese L2 Learners

本文通过两个语言加工实验,对比汉语母语者和学习者在有/无语境条件下的语块加工,讨论语境对多词结构加工的影响。在无语境的词汇判断任务中,两组被试语块反应时(RT)都明显快于匹配的非语块;在有语境的自控步速阅读任务中,通过n-gram运算将语境量化,与语块/非语块RT做协方差分析。两组结果均显示语境对RT的显著贡献,但同时,语块RT的速度优势却从L1组消失,且在L2组的显著性也有所降低。此外,通过对语境可能性、多词结构在语料库的频次、t-值、互信息值、笔画数、字符数等诸多参数与RT的回归分析发现,影响两组被试加工速度的核心因素均是语境;与二语组RT最显著的相关因素是多词结构在学习者语料库中的频次。研究结果提示二语教学:语块整体教学或应与语境法相结合。

This paper presents the results of two experiments designed to investigate the processing of contextualized formulaic sequences by Chinese native speakers（NSs）and non-native intermediate L2speakers（NNSs）,aiming to discuss the context effect in the on-line processing of multi-word expressions.In the lexical decision task,the selected formulaic sequences（FSs）were processed significantly faster than the matched non-formulaic sequences（non-FSs）by both the NS and the NNS groups.In the self-paced reading task,the context probability,which is the conditional probability for a FS/non-FS to occur given its pre-context,was transformed into a continuous variable through a corpus-based method（n-gram）,and was analyzed as covariate with response time（RT）.The results demonstrated significant correlations between the context probability and the RTs of the two types of sequences for both groups.By contrast,the processing advantage of the formulaic sequences found in the grammatical judgment task using non-contextualized measures disappeared from the NS group.Multiple linear regressions also showed that context was the most significant contributor to the NSs＇ RT and the second most significant one to the NNSs＇ RT.The overall results suggest that in Chinese second language teaching,the introduction to the typical contexts of FSs should never be neglected even when the target FSs are taught under holistic treatment.