中国英语学习者复合词加工研究

In the present study, three online experiments were conducted to examine the way Chinese EFL learners process compound words, the influence of L1(Chinese) on the processing of L2 (English) compound words, and the cross language activation of constituent morphemes in the processing of L1 and L2 compound words in Chinese EFL learners. The general research questions are how Chinese EFL learners process compound words, especially English compound words, and in particular, whether they decompose English compound words into their individual constituent morphemes, and whether there is cross language activation in both L1 and L2 compound processing. The specific research questions guiding the present study are as follows: When Chinese EFL learners process English compound words, is their performance affected by the lexical status (lexicality) of the translated compound words in Chinese (the combination of the Chinese translation equivalents of the constituents)? Lexicality effect is the effect found in L2 compound word processing when the lexical status of the translated compound in L1 affects the processing of an L2 compound word (Cheng et al., 2011; Ko et al., 2011). It is hypothesized that the processing of the L2 compound word will be facilitated if the translated compound is a real word in L1, and inhibited if the translated compound is a nonword (does not exist) in L2. For example, the processing of words like toothbrush and hotdog should be easier than that of fireman and deadline since in Chinese there exist 牙刷 [ya2shua1] and 热狗 [re4gou3], but *火人and *死线 are nonexistent. This is highly likely when we think of the common conception that an L2 word is difficult to process and acquire if it lacks an equivalent in L1. The lexicality effect is regarded as evidence that there is activation of L1 in the processing of L2 compound words, and it also in a way proves the existence of compound decomposition in L2 processing. Does semantic transparency of the English compound words affect Chinese EFL learners’ processing of English compound words? If so, how does semantic transparency interact with the lexicality of the translated compound words in Chinese? Semantic transparency refers to the consistency between the meaning of the whole compound and the meanings of its constituent morphemes. If the meanings are consistent, the compound is labeled as transparent, and conversely, the compound is said to be opaque if the meaning of the whole compound is inconsistent with (cannot be predicted from) the meanings of its constituent morphemes. Thus, toothbrush (a brush used for teeth) and 轮椅 ([lun2yi3] a chair with wheels, with the first character meaning “wheel” and the second meaning “chair”) are transparent compound words in English and Chinese respectively, whereas deadline (a time before which a task must be finished) and 吹牛 ([chui1niu2] meaning “brag” or “talk big”, with the first character meaning “blow” and the second meaning “cow”) are opaque ones in the two languages. (The topic of semantic transparency will be further elaborated in the following chapter) Semantic transparency effect refers to the effect found in compound processing when the processing of transparent compounds is faster and more accurate than that of opaque compounds. As mentioned in the previous section, semantic transparency has been assumed and proved to play an important role in L1 compound processing (e.g., Libben et al., 2003; Sandra, 1990; Zwitserlood, 1994). However, it remains controversial whether this effect also holds for L2 compound processing (e.g., Chen et al., 2009; Gan & Zhang, 2013). Does the constituent morpheme in L1/L2 facilitate the processing of a compound word in L2/L1? Specifically, this question can be further developed into two separate questions: (1) Is the Chinese translation equivalent of the second constituent of an English compound word able to prime that English compound word? In other words, does the L1 activation during the processing of L2 compound words happen on the morpheme level? (2) Is the English translation equivalent of the second constituent of a Chinese compound word able to prime that Chinese compound word? In other words, is there L2 activation on the morpheme level during the processing of L1 compound words? Constituent priming in the processing of compound words is a type of repetition priming, or partial repetition priming (the types of priming will be further elaborated in Chapter 3). It refers to the practice of using one constituent (the first or the second) of the compound word to prime that compound word. The cross-language priming paradigm employed in the present study, however, also takes on the property of semantic priming, since the prime word is semantically related to the target word in that it is the translation equivalent of a constituent of the target compound. The crosslanguage constituent priming effect, if found, could provide evidence for both morpheme-level activation and cross-language activation. If both L1-L2 priming and L2-L1 priming take place, how different is the magnitude of priming effect between the L1-L2 direction and the L2-L1 direction? The so-called priming asymmetry in cross-language priming, which predicts that the priming from L1 to L2 is much stronger than that from L2 to L1, is an intriguing hypothesis and has found evidence in previous studies of processing monomorphemic words (Jiang & Forster, 2001). It is hypothesized that the magnitude of priming effect should be different between L1-L2 priming and L2-L1 priming. Since the prime could provide even less information in the constituent translation priming paradigm used in the present study, compared with the priming paradigm in processing simple words when the prime is the translation of the target, we have a reason to suspect that the L2-L1 priming is still less likely to occur. In cross-language constituent priming (both L1-L2 and L2-L1), is the priming effect (if any) influenced by the semantic transparency of the target compound words? This research question is different from the second one though both of them focus on the same variable—semantic transparency. Unlike research question two which deals with the semantic transparency effect in processing L2 compound words, this one specifically concerns the semantic transparency effect in cross-language constituent priming, indicating an opportunity to investigate the interaction between crosslanguage activation and transparency effect. Furthermore, the transparency effect in processing Chinese compound words by Chinese EFL learners will also be tested. In cross-language constituent priming (both L1-L2 and L2-L1), do prime durations influence the magnitude of priming effect? If so, how does the influence differ between the L1-L2 direction and the L2-L1 direction? Prime duration is the time period from the beginning of the prime until the disappearance of that prime. In the present study prime duration is the same as SOA (stimulus onset asynchrony, the time period from the beginning of the prime until the beginning of the target). Sometimes SOA can be longer than prime duration when there is a blank or another stimulus (or multiple stimuli) after the disappearance of the prime. Both prime duration and SOA are the variables frequently adopted to explore the time course of processing, which is of interest because it is generally believed that different information of the word is processed at different time points in the course of processing.