Abstract: Bilingual memory representation has received considerable attention from many researchers for several years. How the conceptual representation is shared is an important issue these days. Previous studies have found evidence that the proficiency of second language influences the relation between lexical representation and conceptual representation of bilinguals. Thus, the representation models should be different for highly proficient bilinguals and less proficient bilinguals. This hypothesis has been supported by some experiments based on explicit memory tasks. However, less is clear about the cross-language repetition priming which is based on implicit memory tasks. Therefore, we conducted the present study to further test whether there would be cross-language repetition priming effect for less proficient bilinguals.
Method
In the present study, we used the task of cross-language repetition priming. Participants included 140 students from South China Normal University who were all Chinese natives age18-20 years, with English as their second language. the students were non-English-majors who did not pass CET Band 4. All four experiments were carried out on computers using the E-Prime software. There were study phase and test phase during each experiment in this study. Participants were instructed to complete the animacy decision tasks in experiment 1, the lexical decision tasks in experiment 2 in each phase. In experiment 3, participants were instructed to complete the animacy decision tasks in study phase in English and lexical decision tasks in test phase in Chinese, the animacy decision tasks in study phase in Chinese and lexical decision tasks in test phase in English. The response time and accurate rates were recorded and SPSS 10.0 software was used to analyze the data.
Results
The response time to the words studied was significantly faster than that to the words unstudied in conceptual implicit memory tasks both in same language condition and cross language condition. The same results were found for the studied words and the unstudied words when conceptual tasks were presented in English in study phase and lexical tasks were presented in Chinese in test phase. But there was no cross language repetition priming effect in lexical implicit memory tasks in cross language condition. For the accurate rates, the results of the experiments were consistent with that on response time.
Conclusion
Consistent with the current theories of bilingual memory representation, we obtained reliable cross-language repetition priming effects when conceptual implicit memory tasks were used and no priming effects when lexical implicit memory tasks were used. The present results based on implicit memory tasks indicate that for less proficient bilinguals, conceptual representations for translation equivalents are shared and lexical representations are separate. In particular, our results support the Word Association Model

Key words: bilingual memory representation, conceptual representation, lexical representation, cross-language repetition priming