Individual Differences in Multilingualism and Language Learning: Effects on Cognitive Flexibility and Structure Learning

Abstract

The effects of bilingualism on cognitive control have been significantly debated, with studies reporting both positive and null/negative findings. The lack of consensus is due to inconsistent experimental instantiation of key constructs. Researchers usually employ binary approaches, reducing linguistic diversity to a dichotomous and mono-factorial categorisation, viz., monolinguals vs. bilinguals. Such comparisons fail to capture the multifactorial nature of the bilingualism continuum. Cognitive control is assessed with limited tasks, which cannot reliably measure the underlying processes. Research typically concentrates on selected groups of participants, that is, it is restricted to the Global North, even though most of the world’s population resides in the Global South, and there is also an intense focus on young university students. This generates bias and limits the interpretability of existing literature. The present dissertation seeks to advance the field from an unfruitful debate on the existence of a “bilingual advantage”. Bilingualism in this thesis is viewed as a complex multifactorial phenomenon, comprising numerous continuous variables. Similarly, multiple cognitive tasks have been used to fractionate and accurately measure Executive Function (Cognitive Flexibility, Working Memory, Inhibition) and Structure Learning, a novel statistical learning framework involving learning under uncertainty. Furthermore, various age groups, including children, young and middle-aged adults have been assessed, as well as diverse sociolinguistic contexts (United Kingdom, Singapore, Greece). Finally, apart from cross-sectional designs measuring the effects of individual differences in bilingual experience on cognitive control, longitudinal or training studies have been employed, evaluating the effects of language-switching training and second-language (L2) learning. Our cross-sectional studies in the UK and Singapore showed that context mediates the relation between bilingualism and cognition. Effects were more apparent in the multilingual environment of Singapore. Proficiency and frequency of language switching were found to be two of the most influential factors, so we further investigated them with longitudinal/training studies. Language-switching training in an experimental setting with Singaporean adults led to benefits outside the language domain and “penetrated” executive task-switching aspects of Cognitive Flexibility, but not more abstract aspects such as rule learning, nor Structure Learning. Conversely, Greek children learning an L2 in a naturalistic setting (school) for seven months showed enhanced performance in Structure Learning and abstract Cognitive Flexibility aspects. These tasks entail the uncertainty, rule discovery, and adaptation to change that is present in real-life language learning. By breaking down both bilingualism and cognitive ability into composing factors, we managed to identify precise links between the two. This thesis does not speak of an ill-defined, overall “bilingual advantage” but demonstrates how different linguistic practices can influence different aspects of cognition.