Nederlandse Organisatie voor Wetenschappelijk Onderzoek
Children from a young age on use the pronoun 'him' in an adult-like and systematic way when they speak. Yet they often do not understand this word correctly when it is uttered by someone else. Why would these children use their knowledge of grammar in production, but not in comprehension? In other cases, children understand a form before they use it themselves. Obviously, such asymmetries in acquisition cannot be explained by a traditional rule-based grammar, according to which speakers who obey a rule in production know this rule, and hence should obey this rule in comprehension too (and vice versa for hearers). This program seeks a unified explanation for these production-comprehension asymmetries. It tests the hypothesis that the grammar is an asymmetrical system of constraints on form and meaning. Hence, production may yield other results than comprehension. Mature language users have learned to integrate the two directions of language use by taking into account their conversational partner's options as well. This results in a symmetrical matching between forms and meanings. The ability to consider other people's options in communication may be limited in children and impaired in autistic individuals. The proposed program thus explores a novel view on the organization of the grammar. It models the way children produce and comprehend utterances, and the way adults consider their conversational partner's perspective in communication. Psycholinguistic experiments are conducted which test this model by studying production and comprehension in the same subjects. These experiments reveal which part of our mature linguistic competence arises from the grammar, and which part arises from implicit reasoning about our conversational partner?s options. The program also applies these new linguistic insights to the field of autism and may improve the diagnosis and treatment of the language deficits within autistic spectrum disorder (ASD), which are still poorly understood.