Phonological Encoding In Cantonese-English Bilingual Speech Production: An Event-Related Brain Potential Study

When one speaks, one needs to retrieve the phonological content of the utterance prior to articulation and this process has been termed phonological encoding. There is sizable evidence showing that the primary (or first retrievable) planning unit for phonological encoding is different across languages. Dissimilar to many Indo-European languages such as Dutch and English where phonemes have been found to be the primary phonological encoding units, the primary planning units are syllables in Cantonese and Mandarin. This marked contrast between Indo-European and non-Indo-European languages has posed an interesting and important question regarding how bilinguals possessing these two types of language plan for their speech.Bilingual studies have repeatedly shown that one’s first language (L1) would affect one’s acquisition of a second language (L2) (i.e., forward transfer). Likewise, one’s L2 would in turn affect one’s L1 processing (i.e., backward transfer). However, very little is known about how language transfers are realized in the mechanism of phonological encoding. Specifically, when the two languages a bilingual possesses rely on different phonological planning units (e.g., phoneme vs. syllable), how the encoding system adapts to the different demands of L1 and L2 is an issue largely open. Therefore, two picture naming experiments using the form-preparation paradigm are proposed to investigate the role of phoneme and syllable in English (Experiment 1) and Cantonese (Experiment 2) speech production among Cantonese-English bilinguals with various L2 (English) background. Meanwhile, the participants’ event related brain potentials (ERPs) will be recorded during the production task. If forward transfer exists in bilingual phonological encoding, one would expect to see a robust effect of syllable in L2 (phoneme-based English) production at least among the bilinguals with a lesser degree of L2 exposure, a pattern similar to their L1 (syllable-based Cantonese) production. If not, a robust effect of phoneme in L2 production would be expected irrespective of the bilinguals’ L2 background. Experiment 2 investigates the possibility of backward transfer in bilingual phonological encoding. If backward transfer exists, one would expect to see a robust effect of phoneme even in L1 (syllable-based Cantonese) production at least among the bilinguals with a higher degree of L2 (phoneme-based English) exposure. Otherwise, one would expect to see a robust effect of syllable only in L1 production irrespective of the bilinguals’ L2 background. Furthermore, a number of factors regarding the bilinguals’ L2 background will be measured to investigate the possible causes for language transfer in speech production.