The costs and benefits of dual language activation in bilingual processing
In this talk I present results from two studies which dig deep into lexical processing in bilinguals and the degree to which dual language activation is automatic and/or beneficial. In the first experiment, a group of 185 French-English and English-French bilingual adults of varying proficiencies took part in a fully French online masked lexical decision task which made use of stimuli pairs with a hidden morpheme repetition when the critical prime-target pairs were both translated into English. For example, a critical prime-target pair could be ‘noir-chantage’ (black-blackmail) or ‘courrier-chantage’ (mail-blackmail). The aim was to test 1) whether bilinguals automatically translate encountered words into the other language even when the task is in a single language (the vast majority of previous results had used dual language stimuli, cf. Thierry and Wu, 2004; 2007), and 2) the effect of dominance and proficiency. Results showed a significant main effect of hidden morpheme repetition in speakers who were native in both languages, and in those who were more fluent in English. In both cases the effect was that of inhibition. These findings are discussed with respect to the predictions of three prominent models of bilingual lexical access (the RHM, the BIA/BIA+, Multilink).
In the second study, we examine whether bilinguals show advantage in executive function using the directed forgetting paradigm (Basden et al, 1993). 90 participants aged 19-76 (mean age 36) participated in the study. Of these 53 were bilingual and 37 monolingual. Participants were presented with 80 mono and disyllabic English words and were asked to either remember or forget these for a later test (40 words in each condition, controlled for frequency). Greater executive function abilities were expected to allow participants to intentionally forget stimuli they do not need to remember. However, results from previous studies regarding grey matter density in the bilingual brain left the possibility that bilinguals would be more successful at overcoming inhibition of irrelevant information rather than inhibiting that information (e.g. Nowicka et al. 2009). This is indeed what was found in our study, particularly in older participants. Bilinguals exhibited greater accuracy for ‘forget’ stimuli compared to monolinguals, showing they did encode those words into long-term memory by overcoming inhibition triggered by the ‘forget’ instruction. Older participants showed greater differences in performance between linguistic groups, supporting existing evidence that bilingualism attenuates declines in memory and executive function as we age.
This leaves the question open with regards to what counts as an advantage for bilinguals in language processing when improved memory function may lead to overcoming inhibition, as was found here, or when fluency in one language may slow down access to the other, as was found in the first study.