https://lling.univ-nantes.fr/medias/photo/image-seminaire-synsem-last_1634558540342-JPG
-
Le 31 janvier 2024 de 14:00 à 16:00false false
- mercredi 31 janvier, 14h
Responsable : David Lobina
Titre : Conjunction and Disjunction in Infants: An Eye-movement Experiment (and the flexibility of GAMMs)
Résumé :
The question of how early in ontogeny children master logical connectives such as and and or has received much attention recently. The bulk of the evidence so far has come from experimental techniques such as the truth-value judgement task, a paradigm in which a young child is required to judge the truth/accuracy of the utterances of a playful puppet or toy, and which can be employed with children from the age of 3 or so. Such a paradigm, however, is not appropriate for use with infants and toddlers, and at three years of age children already possess rather sophisticated linguistic abilities, including a mostly fully developed conceptual system. On the other hand, the intermodal preferential looking paradigm (IPLP), involving the tracking of eye movements, has been used to unearth some of the (emergent) syntactic knowledge of infants as young as 18-month-old, and with great success. After reviewing part of the literature, I present the results of an IPLP experiment with a cohort of “terrible twos” (actually, average age: 28-month-old), the data suggesting that these children exhibit knowledge of disjunction but not of conjunction, and moreover, that they favour a true-true interpretation of disjunction instead of an exclusive (either-or) reading. I discuss how these data relate to previous results from the field, I briefly connect the discussion with the issue of the primitiveness of language’s connectives, and furthermore I describe the usefulness of generalised additive mixed models to analyse the course of time-series data such as eye movement recordings.
Titre : Conjunction and Disjunction in Infants: An Eye-movement Experiment (and the flexibility of GAMMs)
Résumé :
The question of how early in ontogeny children master logical connectives such as and and or has received much attention recently. The bulk of the evidence so far has come from experimental techniques such as the truth-value judgement task, a paradigm in which a young child is required to judge the truth/accuracy of the utterances of a playful puppet or toy, and which can be employed with children from the age of 3 or so. Such a paradigm, however, is not appropriate for use with infants and toddlers, and at three years of age children already possess rather sophisticated linguistic abilities, including a mostly fully developed conceptual system. On the other hand, the intermodal preferential looking paradigm (IPLP), involving the tracking of eye movements, has been used to unearth some of the (emergent) syntactic knowledge of infants as young as 18-month-old, and with great success. After reviewing part of the literature, I present the results of an IPLP experiment with a cohort of “terrible twos” (actually, average age: 28-month-old), the data suggesting that these children exhibit knowledge of disjunction but not of conjunction, and moreover, that they favour a true-true interpretation of disjunction instead of an exclusive (either-or) reading. I discuss how these data relate to previous results from the field, I briefly connect the discussion with the issue of the primitiveness of language’s connectives, and furthermore I describe the usefulness of generalised additive mixed models to analyse the course of time-series data such as eye movement recordings.
Mis à jour le 25 janvier 2024.