Building individualised models of language development

In this work package, we constructed computational models that explained the performance of a subset of children from the Language 0-5 cohort at an individual level across a range of language tasks administered at different points in development.

A unique strength of the language 0-5 dataset is that it includes multiple detailed measures of language and language-related abilities from the same children across a wide developmental range. In this work package, we exploited the constraints provided by this unique feature of the dataset to build models that explained the behaviour of particular children across tasks and across development, and thereby developed a theory of how different language abilities interacted over the course of development.

Project Team: Fernand Gobet (Lead), Evan Kidd, Padraic Monaghan, Julian Pine, Andrew Jessop and Caroline Rowland

Duration: 3 years, starting October 2020

Project Number: 3.2

Key Outputs

Jessop, A., Pine, J., & Gobet, F. (2025). Chunk-based incremental processing and learning: An integrated theory of word discovery, implicit statistical learning, and speed of lexical processing. Psychological Review. Advance online publication.

Rowland, C. F., Bidgood, A., Jones, G., Jessop, A., Stinson, P., Pine, J. M., Durrant, S., & Peter, M. S. (2025). Simulating the Relationship Between Nonword Repetition Performance and Vocabulary Growth in 2-Year-Olds: Evidence From the Language 0–5 Project Language Learning, 75: 379-423.