Examining and reexamining the structure of Chinese-English bilingual memory.

The asymmetry model of bilingual memory proposed by Kroll and Stewart assumes that translation from a first language (L1) into a second language (L2), or forward translation, is mediated by an underlying conceptual memory, whereas L2-to-L1, or backward, translation is lexical and direct. Lexical links from L2 to L1 are hypothesized to be stronger than those from L1 to L2, but conceptual links are stronger for L1 than L2. These assumptions were evaluated in this study, which used stimulus items with similar recognition thresholds. Proficient Chinese-English bilinguals were tested on picture naming, word naming, word translation and category matching. The translation asymmetry effect predicted by the model (i.e., backward being faster than forward translation) was demonstrated; matching an L1 item to a category name was also faster than matching an L2 item. This pattern of results, nevertheless, did not completely hold after an adjustment procedure which sought to control for variations in response production and concept retrieval times due to stimulus types. The present findings are consistent with previous ones which show increased L2 conceptual processing for the proficient bilingual. They also suggest that an extended version of the asymmetry model should take into account possible biasing effects associated with some conventionally used bilingual tasks.