18.03.2025 Gabriella Vigliocco

Most studies of language use investigate individuals listening to spoken words, or reading text, grounded in traditional theoretical views of language as abstract and symbolic. Listening to speech or reading, however, are only two among many communicative contexts in which we use language, they permit a high degree of experimental control, but they do not represent the communicative context in which language is learnt and very often processed in adulthood. As infants, we learn language in face-to-face interactions with caregivers and much of our language use in adulthood continues to be in face-to-face contexts.
Face-to-face language is multimodal (in addition to speech, listeners also process non-linguistic information such as prosody, gestures, etc) and social (it implies interacting individuals). In this talk I will present work that my lab has carried out in the past 5 years that primarily asks whether and how multimodal, non-verbal information affects language (word) learning and processing. I will present results from analyses of a naturalistic corpus (ECOLANG) of caregiver-child dyadic interaction, from electrophysiological experiments using (more) naturalistic stimuli that show how so-called non-verbal behaviours support word learning and are dynamically integrated with speech during language comprehension. While most of the evidence presented in the talk will concern multimodality, I will end discussing the importance of social interaction and present some initial results that show how learning of semantic information from language can benefit from being embedded in a dialogue.

Different time slot: 12:15-1:30 PM, Geneva zoom