以熟练汉-英双语者为被试,考察了在短语水平上语码切换的机制及切换代价的来源。结果表明:(1)汉语与英语表示空间方所的短语的差异和表示量的短语的对应与否影响语码切换过程,但对于两种语言的切换代价的影响不显著,表明语码切换代价未受特定语言的语法特征影响;(2)熟练汉.英双语者在短语水平上表现出语码切换代价的不对称性:切换至熟练语言的代价比切换至非熟练语言的代价大;(3)熟练汉-英双语者的切换代价源于两种语言短语的激活和竞争,支持非特定语言选择假说,可以用抑制控制模型来解释。整个研究表明,熟练双语者的语码切换代价源于语言表征系统之外。
Code-switching has attracted much attention from researchers with different perspectives. It has widely accepted that when bilinguals switch from one language to another, there is generally a transient cost to performance, namely the so-called switching cost. Switching cost is the main aspect when we explore the essence of the language switching mechanism. The recent study focused on where these costs arise from. The costs could be caused by different mechanisms. It may arise as a result of processes within the bilingual lexicon and/or because of processes outside the lexicon associated with task performance. Although the investigator have done a lot of researches with different experimental paradigms and materials, non-controversial conclusion has not been arrived. Until now, most conclusions were driven from lexical level or sub-lexical level and most researches was involved by two alphabetic characters and their orthographic specificity. If we select some language units (phrases or sentences)as experimental materials which are involved with grammatical pattern and/or syntax, this will help us to ascertain the source of switching cost in language representation system more than in lexicon or mental dictionary. For phrases, we mean prepositional phrases and noun phrases which represent spatial orientation and quantity separately. This study focused on ideogram characters such as Mandarin characters and phrase level to explore the mechanism of language switching with proficient Chinese-English bilinguals in picture naming task. 29 undergraduates or post-graduates whose major are English participated in two experiments. They were asked to name the picture in either Chinese or English as fast and accurately as possible according to different background color in two experiments. The material of experiment 1 was prepositional phrases. The material of experiment 2 consisted two kinds of noun phrases, either having quantifier or not. All the participants were tested individually on computers. Reaction times for c