The present study aimed to investigate whether bilinguals access phonological representations in both of their languages simultaneously, when required to produce speech in their second language (L2) only. Theories of bilingual processing such as that of Poulisse and Bongaerts (1994) are based on Levelt's (1989) influential model of lexicalisation in the monolingual literature. This would lead us to predict that processing of bilingual speech should be strictly serial, and that only after the appropriate lemma has been selected in the appropriate language, does phonological encoding begin.
However, Peterson and Savoy (1998) found evidence against the strict serial stages proposed by Levelt (1989), by using a picture naming task reliant on synonyms. Their findings lead Levelt et al (1999) to concede that synonyms are something of a special case and that processing may indeed occur for both synonyms before lexical selection.
The current experiment extends Peterson and Savoy's (1998) methodology for use with bilinguals. 20 French-English bilinguals named a series of pictures in English. On critical trials, 300ms after presentation of the picture, a word was presented for reading, which was phonologically related to the French picture name. (e.g. picture of an apple, "pomme" in French, with probe word POND). No effect of phonological priming on reaction time was found ( t(19) = 0.201, p=0.843 by participants, and t(59)=0.75, p=0.94 by items) although slower responses appeared to be somewhat facilitated when primed. There was a non-significant (p=0.15) positive correlation between the frequency of the experimental picture name and the size of the experimental effect, suggesting that the French picture name may have been accessed for low frequency items where the English name may have been unavailable. This indicates that the methodology may have been sensitive enough to pick up activation of French phonology, but it does not support the cascaded processing argument. The effect of the manipulation on reading durations will also be analysed.
Costa et al (2000) found evidence of parallel activation of phonology in bilinguals, but their materials were cognates, which may be stored differently from other words in the bilingual lexicon (e.g. de Groot & Nas, 1991). The present findings were not restrained in this way, and therefore add to the developing literature on modelling bilingual language production. The results may also have implications for the current debate in the monolingual psycholinguistics literature over whether lexicalisation is a serial or cascaded process (e.g. Peterson and Savoy, 1998, Levelt, Roelofs and Meyer, 1999). A follow-up study is planned to verify the effectiveness of the priming technique.
References
Costa, A., Caramazza, A., & Sebastian-Galles (2000) The Cognate Facilitation Effect: Implications for Models of lexical Access. Journal of Experimental Psychology: Learning, Memory and Cognition 26(5) 1283-1296
De Groot, A.M.B., & Nas, G. L. J. (1991) Lexical Representation of Cognates and Non-Cognates in Compound Bilinguals. Journal of Memory and Language 30 90-123
Levelt, W. J. M. (1989) Speaking : from intention to articulation. Cambridge,Mass. London : MIT Press
Levelt, W.J.M., Roelofs, A., & Meyer, A.S. (1999) A Theory of Lexical Access in Speech Production. Behavioural and Brain Sciences 22 1-75
Peterson, R.R., & Savoy, P. (1998) Lexical Selection and Phonological Encoding During Language Production: Evidence for Cascaded Processing. Journal of Experimental Psychology: Learning, Memory and Cognition 24(3) 539-557
Poulisse, N., & Bongaerts, T. (1994) First Language Use in Second Language Production. Applied Linguistics 15(1) 36-57