Final Report Summary - UNDER CONTROL (Mechanisms of cognitive control and language learning) How do humans manage to learn language? Most research in this field has focused on how the different types of linguistic knowledge (sounds, words, grammar...) are acquired. However, to achieve this goal, humans must sort out what information is relevant, and this is not always trivial. Even worse, some information that may be important in one language may be irrelevant in another. For instance, in some languages a change in the position of the stress changes the meaning of words, such as with the Spanish words “sábana” (meaning bed sheet) and “sabána” (meaning savannah).However, other languages, such as French do not bear this property, and stress is meaningless at the word level, as all words are stressed on the last syllable. Given that the acoustic properties characterizing stress (duration, intensity and frequency) are present in any speech signal, how does the brain of the listener manage to extract and organize the relevant information? The present project has tackled the relationship between general-domain mechanisms, in particular attention, and language learning. To this end we have investigated infant and adults across a set of tasks tapping at different attention and language processing mechanisms. We have studied how language learning is shaped by attention in infants who are learning their first language(s) and also in adults who have learned a second language. We have compared individuals who grow up in a monolingual environment with bilinguals, as the latter have been claimed to develop specific attentional mechanisms to deal with the need of sorting and classifying the two languages.The research with adults has investigated what makes some individuals particularly good or particularly poor at learning the sounds of a second language. At the neural level, we have identified differences in patterns of oscillatory activity and structural brain differences. The pooled analysis of a battery of language and attentional tasks has shown that it is the capacity to perceive and produce speech, rather than general-domain attentional factors what underlies individual differences in L2 learning.The research with infants has shown that already at 4-5 months of age, infants growing up in bilingual environments process the speech signal in a different way than monolingual ones. One critical difference between monolingual and bilingual firest language learning is the need for the latter to successfully differentiate languages. We have observed differences between monolingual and bilinguals in oscillatory changes in the theta band range (4-8Hz), when they hear sentences in a native language (but not when they hear an unfamiliar language). The bilinguals’ higher values in theta band synchronization could be reflecting higher cognitive demands, that we link to phonological processing, when compared to monolinguals. Research with older infants (from 6 months of age on) has shown that during the first two years of life, bilingual infants pay more attention to the mouth of people than to their eyes, even more, they are less able to shift their attention from the mouth to the eyes when new information appears in the eye zone of a character. This increase of attention to the mouth is not restricted to linguistic stimuli, as bilingual infants will look more to the mouth than monolingual ones, even if they see non-speaking faces, for instance, laughing or crying ones. Finally, several of our studies provide converging evidence that from very early in life, infants are sensitive to the social (group membership) cues that language provides. Infants seem to treat in a preferential way actions and information that individuals speaking their own language perform or provide. Language seems to be an important cue triggering increase in attention to incoming information and therefore likely increasing learning from individuals with whom we share culture in general, and language in particular.
