Sommerfeld, Linda; Staudte, Maria; Mani, Nivedita; Kray, Jutta

Children and adults integrate complex visual contexts in language prediction

Architectures and Mechanisms for Language Processing AMLAP Annual Meeting, 2021.

Not only adults, but already children anticipate upcoming input in sentences with semantically constraining verbs. In the visual world, the sentence context is used to anticipatorily fixate the only object matching potential sentence continuations. Adults process multiple visual cues in parallel when predicting language. Here, we examined whether young children can also process multiple visual objects fitting the constraining verb of a sentence. We also examined if their processing of multiple potential referents is affected by language skills, given that vocabulary size modulates children’s prediction.