Distributional learning of novel visual object categories in children with and without developmental language disorder

  • Iris Broedelet (University of Amsterdam)
  • Paul Boersma (University of Amsterdam)
  • Judith Rispens (University of Amsterdam)


It has been proposed that a deficit in statistical learning contributes to problematic language acquisition in children with developmental language disorder (DLD), but at the same time the nature and extent of this relationship is not clear. This paper focuses on the role of statistical learning in lexical-semantic development by investigating visual distributional learning of novel object categories in children with and without DLD and its relation to vocabulary knowledge. Distributional learning is a form of statistical learning and entails the learning of categories based on the frequency distribution of variants in the environment. Fifty children (25 DLD, 25 TD) were tested on a visual distributional learning task. Results indicate that children can learn novel object categories on the basis of distributional information. We did not find evidence for a deficit in visual distributional learning in children with DLD. To investigate whether visual distributional learning ability is related to vocabulary knowledge, the children with DLD were tested on different measures of vocabulary. Phonological processing ability and non-verbal intel-ligence were taken into account as control variables. Multiple linear regression analyses did not reveal evidence for a relationship between distributional learning and vocabulary in DLD. 

Keywords: developmental language disorder, statistical learning, distributional learning, vocabulary

Published on
11 Jan 2023
Peer Reviewed