Neural correlates of phonological and semantic priming in Korean-English bilinguals

Abstract

This study investigated bilingual lexical non-selectivity with respect to the two bilingual word recognition models. The BIA+ model assumes non-selective access to an integrated bilingual lexicon whereas the RHM assumes independent lexicons with shared conceptual representations. To determine whether the processing of one language automatically activates processing in the other language, balanced Korean-English bilinguals were compared with English monolinguals in an event-related potential (ERP) masked priming word naming task that manipulated phonological and semantic primes. For Korean- English bilinguals, no phonological difference was observed behaviourally, although a marginal P2 pre-lexical phonological inhibition followed by a weak N400 lexical facilitation was found in the ERP findings. Bilinguals were also faster to respond to semantically related prime, which was reflected as a pre-lexical P2 inhibition followed by strong N400 lexical facilitation in the ERP. Given that pre-lexical phonological effects were observed in bilinguals, results generally supported the BIA+ model where pre-lexical and lexical representations of the two languages in bilinguals are integrated in a shared mental lexicon. However, the current BIA+ model is still insufficient to explain pre-lexical semantic and phonological interference effect in different-script bilinguals. Incorporation of elements in the BLINCS model would be appropriate to improve the BIA+ model.