Chapter 9. The locus of cross-language activation

Two experiments were conducted to investigate the locus of cross-language activation and the lexical selection mechanism in bilingual language production. In Experiment 1, a picture naming paradigm was used to examine how far the activation of the native language (L1) was during the second language (L2) production. Unbalanced but relatively proficient Chinese-English bilinguals were asked to name pictures in English. If a Chinese character was visually presented following a picture, participants were required to name the character instead of the picture. Both the behavioral and ERP results demonstrate that during the planning of L2 speech, Chinese-English bilinguals activate their L1 to the lemma level but not to the level of the phonology. In Experiment 2, the Go/Nogo paradigm was employed to examine why the activation of L1 did not spread to the phonological level. In one block, Chinese-English bilinguals were required to classify pictures according to the initial phoneme of a picture’s Chinese name. In the other block, they were instructed to do this based on the initial phoneme of a picture’s English name. Although the behavioral data showed that it was relatively easier to perform this task in L2 than in L1, the mean amplitude of N200 elicited by L2 was much more negative, suggesting that stronger inhibition was recruited to retrieve the phonological information in L2.