Neurocognitive Mechanisms of Language Processing and Control in Early Bilinguals and Late Language Learners

The ability to speak more than one language is nowadays commonly the rule, rather than the exception. Many forms of bilingualism exist, varying from early bilingualism in which more than one language is acquired in early childhood, to late bilingualism, where another language is not learned until adulthood. Although a large amount of research has been devoted to the question of how these multiple languages are processed and controlled, conclusive answers have not yet been given. Especially regarding language perception, there have not been many studies to investigate whether active language control is needed in case one does not need to actively select and produce lexical items. The first aim of this thesis was thus to study language control during language perception in more detail, and secondly, to investigate to what extent lexical and semantic processing differs between early-acquired and later-learned languages. The first study examines the effects of language switching on semantic processing (Study I). The second study focuses on trilingual language switching, taking a closer look at the control mechanisms that play a role in this (Study II), while the third study investigates whether early bilingualism leads to possible disadvantages or qualitatively different lexical processing (Study III). This PhD work used various research methods, namely magneto- encephalography (MEG) and encephalography (EEG), as well as behavioural methods and extensive language background questionnaires. The findings of this thesis suggest that language control differs according to the strength of the language network. Even though language switching from L2 to L1 is costly at the neurocognitive level, evidenced by enhanced N400 effects, semantic processing remains unhindered (Study I). However, there is no apparent cost of switching between native languages (Study II) whereas an increase in N400(m) is seen after switches from a later-learned language to the native one (Study I and II). Furthermore, the acquisition of two languages at an early age does not notably affect the speed or accuracy with which lexical processing in either language occurs (Study III), as early bilinguals performed worse on only 1 out of 12 data sets, compared to monolingually raised native speakers. Taken together, the results of this thesis show that bilingual language processing and control is modulated by various language background factors, such as the experience and skills in each particular language, as well as the frequency of use. Provided that the language network is sufficiently strong, lexical and semantic processing of a second language will look similar to that of monolingual native speakers. This thesis proposes that bilinguals use their full linguistic knowledge to make sense of the linguistic input around them, while they are at the same time constantly aware of language membership.