The last decades have seen a paradigm shift in cognitive and neurosciences, with important consequences for our understanding of multilingualism and language learning. The static, modular, localisationist models of the late 20th Century often corresponded to similarly static views of language, determined to a large extent by innate characteristics and early childhood input. In contrast, the current dynamic distributed networks, emphasising neuroplasticity and adaptation, shift the emphasis towards the importance of lifelong experience and the way it modulates brain, mind and behaviour. Accordingly, the simplistic monolingual vs. bilingual dichotomy is increasingly being replaced by approaches examining the impact of multilingual (or indeed plurilingual) experience across the whole lifespan. Patterns and contexts of language use, code switching and code mixing as well as attitudes to language learning and use are becoming relevant variables in order to understand cognitive effects of multi- and plurilingualism. This leads to an increasing convergence of psycho-, neuro- and sociolinguistics.