An SPITM Research On Semantic Representation Of Mental Lexicon In Bilinguals

The semantic representation of bilingual mental lexicon is becoming an important area of inquiry. Though semantic priming paradigms provide a powerful tool for the investigation of the structure of bilingual representation, the results of which are far from unanimous. Some support the independence model; some of the results, however, are consistent with the interdependence model. De Groot et al. argue that language similarity or between-language distance can affect bilingual semantic representation. As is well known, the Chinese language differs, on important orthographic and linguistic dimensions, from the languages of the Indo-European family. Thus research on Chinese-English bilingual semantic representation may provide useful information about linguistic universals. At least, if the results of such research do support the interdependence model, it will forcefully corroborate the lexical decision task is universal, and independent of between-language distance. Unfortunately, research on this issue is on the threshold of the embryonic stage.This study attempts to use the new method (SPITM) to investigate the same issue further. The thesis falls into four chapters. In Chapter One, the author gives the definitions of key terms in this thesis, and then introduces the historical studies on bilingual semantic representation. Chapter Two presents the hypotheses and test designs. Chapter Three is a presentation of the detailed process of theexperiments. And then, the experiment results are elaborated and discussed in this chapter. Finally, Chapter Four presents the conclusions derived from this experimental endeavor. To promote the future research on the same issue, the implications for future research are also put forward at the end of this chapter.Specifically, the conclusions reached by this study are:1. Within- and cross-language semantic priming effects do exist, so both Chinese and English share one common semantic system, which strongly supports the interdependence model and forcefully corroborates the lexical decision task is universal, and independent of between-language distance.2. The time interval from 70msec to 1,500msec can be roughly divided into two subintervals. One is automatic semantic priming interval from 70msec to approximately 250-350msec, with the gradual increasing of the effect till reaching its climax at SOAs around 250-350msec. The other is automatic semantic priming and strategic processes superposed interval from about 350-1,500msec, during which automatic semantic priming effect decreases as SOAs extend, strategic processes begin at SOAs around 350msec and increase as SOAs become longer.3. The magnitude of semantic priming effects is as follows-thegreatest in the C-C type, next in the E-C type, then in the C-E type, and the smallest in the E-E type.4. The production of automatic semantic priming effects depends simultaneously on the following important factors: the extent of the primes being accessed, the distance between the concept of the prime and that of the target, the English learning method, the degree of bilinguals' fluency in English and the SOA(s).5. The results of this study support the interdependence model of bilingual mental lexicon.Being objectively obtained from experiments conducted in this study, the conclusions presented in this thesis are findings scientifically well-grounded, with little research bias or contamination by randomerrors. These significant findings are of academic values for promoting the researches on psychology, psycholinguistics, brain science and foreign language teaching.