Visual word recognition in bilinguals and monolinguals: behavioural and ERP investigations of the role of word frequency, lexicality and repetition

This thesis investigated visual word recognition in bilinguals and monolinguals through the effects of word frequency, lexicality and repetition. The first series of experiments focused on whether bilinguals can suppress lexical access in a non-target language and the role of task demands in L1 and L2 lexical access. The next series of experiments further investigated the role of task demands, list composition and repetition in monolinguals. The ERP data show, for the first time, that in bilinguals, lexical access occurs in the non-target language, supporting the notion of a nonselective lexical access. Delayed lexicality effects in ERPs for L2 compared to L1 suggested a delay in lexical access for L2 in late bilinguals, although behavioural data showed a similar word frequency effect to L1. These conflicting responses have not been anticipated by current models of bilingual visual word recognition. However, monolingual data make clear that lexical effects can be modulated by task demands and list composition in behavioural responses. In monolinguals, the slower processing of less familiar items was enhanced by item repetition only when the task required a lexical decision and words and word-like letter strings were presented. Lastly, this thesis has demonstrated that current models of visual word recognition have not fully implemented these elements and have not predicted response times distribution or ERPs. Future models of visual word recognition should incorporate these elements to be able to characterise lexical access in bilinguals and monolinguals