Do bilinguals have a cognitive advantage? Examining effects of bilingualism and language use on executive control

Abstract

The daily practice of bilingual language control has been argued to affect both lexical processing and non-verbal executive control in bilingual speakers.

On the one hand, bilingualism may slow down lexical processing in both languages.

On the other hand, bilinguals have been said to show cognitive advantages compared to monolinguals, for example on inhibition and switching tasks. However, this ‘bilingual advantage’ is hotly debated, can often not be replicated, and language groups have been poorly matched on background variables in previous studies.

Furthermore, I examined the reliability of the literature and found evidence for the existence of a publication bias (Chapter 3). This over-representation of positive studies compared to studies with null or negative findings hinders a reliable interpretation of the actual effects of bilingualism.

The current thesis therefore aimed to examine possible effects of bilingualism on both lexical processing and executive control. Specifically, I investigated the effects of an understudied, but important feature of bilingualism: language use.

Effects of bilingualism have been argued to be largest in older adults. Chapter 4 presents a study discussing inhibition and possible effects of age across various tasks. I show that inhibitory control and age effects depend on task-specific features, including the type of interference, type of stimuli, and processing speed.

Next, I present a study (Chapter 5 and 6) examining the relation between bilingualism and both lexical processing and executive control in older adults. Importantly, bilingual and monolingual groups were matched on background variables including immigrant status. I furthermore compared a group of active to inactive bilinguals to assess effects of language use. On a lexical processing task, bilinguals had a disadvantage compared to monolinguals. This effect was modulated by language use, implying that not only language proficiency but also actual language use are needed to explain lexical effects of bilingualism. However, the non-verbal executive control tasks showed no consistent effects of bilingualism or language use on inhibition or task switching. Thus, this study did not replicate positive effects on executive control in older adults.

Between-subject comparisons remain problematic as groups can never be matched perfectly. Furthermore, these designs cannot assess a causal effect of bilingualism. Therefore, I conducted another study using behavioural and EEG measurements to test for causal effects of language switching on task switching (Chapter 7). When young bilinguals completed a language-switching task prior to a verbal task-switching paradigm, they showed larger switching costs than after a monolingual naming task. However, this effect of language switching was not found for non-verbal task switching. Language switching may thus have a negative impact on verbal switching, but these effects did not extend to non-verbal executive control. Together, these studies suggest that bilingualism and language use affect lexical processing, but there was no evidence for effects of bilingualism and language use on non-verbal executive control in younger or older adults. In combination with other failed replications and the biased literature, this questions the reliability of cognitive benefits associated with bilingualism. However, executive control is not a unity and its manifestation depends on task-specific features. This task impurity, together with the degree to which participant groups are matched, may explain the inconsistency with which effects of bilingualism on executive control have been observed.