Cognitive predictors of language development in children with specific language impairment (SLI)

Abstract

BACKGROUND: Language development is generally viewed as a multifactorial process. There are increasing indications that this similarly holds for the problematic language development process. AIMS: A population of 97 young Dutch children with specific language impairment (SLI) was followed over a 2-year period to provide additional evidence for the existence of underlying language factors. Furthermore, the children's language development was related to their non-verbal intellectual reasoning capacity. METHODS & PROCEDURES: The language abilities were assessed via the administration of an extensive battery of language tests. Cognition was assessed via administration of the Raven progressive matrices, short-term memory capacity via administration of the Digit Span task. OUTCOMES & RESULTS: The results provide empirical support for distinct language factors for children with language problems. The detected factors were labelled: phonology, lexical-semantics, syntax, and speech production, and were found to be stable and interrelated. Short-term memory showed strong relations with the language factor syntax and medium relations with the other language factors. Intellectual capacity showed weak to medium relations with three language factors but no relation with the factor speech. CONCLUSIONS & IMPLICATIONS: The language development of 5- and 6-year-old Dutch children with SLI appears to be divided into four language factors. The language factors are very stable in this sample and correspond with language factors found in recent studies on language development in children with SLI. Short-term auditory memory turns out to play an important role in the problematic language acquisition of children with SLI. It is recommended that children with SLI should be assessed on possible deficits in information processing and/or short-term memory. The existence of such deficits calls for specific neuropsychological intervention.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 24 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.