Davies, P. , Deuchar, M. , Hoshino, N. , Parafita-Couto, M. C. & Thierry, G.
ESRC Centre for Research on Bilingualism in Theory and Practice, Bangor Univeristy
A recent corpus-based study of code-switching in nominal constructions (Herring et al., in press) attempted to evaluate the Matrix Language Frame (MLF) approach (Myers-Scotton, 2002) in relation to a competitor model. However, because of the paucity of relevant naturally occurring data we designed a study to evaluate the MLF approach experimentally by using neurocognitive measures. In the present study, we investigated the acceptability of code-switching by using a sentence verification task with event-related potentials (ERPs). Welsh-English bilinguals read sentences presented one word at a time at a fixed speed. At the end of each sentence, two pictures were presented and the bilinguals were asked to select the picture which matched the sentence. The language of the matrix verb, the adjective, and the noun and the word order of the nominal construction (the adjective + the noun) were manipulated. According to the MLF theory, code-switching is acceptable when the language of the matrix verb is the same as the language of the word order. For example, I saw one black ci (‘I saw one black dog’) is acceptable for the MLF because the verb saw is from English as is the adjective-noun order. In contrast, I saw one ci black would not be acceptable because the noun-adjective order is not normal in English even though the verb saw is from English. It was predicted that the latter type of sentences should produce an enhanced negativity between 300 and 500 ms (left anterior negativity: LAN) and an increased positivity between 450 and 850 ms (late positive component: LPC) relative to the former type of sentences. The results will be discussed in the light of these predictions.
* Authors are listed alphabetically.