Phonological processing skills in children with specific language impairment

Abstract

There is much debate in the literature about the cause, presentation, diagnosis and treatment of specific language impairment (SLI). Research has been hampered by the heterogeneity evident within the diagnostic group as well as a paucity of tasks to measure specific skills and thus increase our understanding of the underlying deficit. One prominent theory is that children with SLI have an underlying deficit with phonological processing skills although the role of phonology in the establishment of accurate, well specified phonological representations is still unclear.This program of research aimed to add to the body of evidence by addressing these key methodological issues and exploring the phonological processing skills of children with SLI. In the initial phase of the research, two silent measures of phonological representations were designed and developed to fill a recognised gap. The Quality of Phonological Representations task aims to explore the accuracy of a child’s stored phonological representation of a multisyllabic word. The Silent Deletion of Phonemes task aims to explore how well specified a stored phonological representation is, and requires a child to perform a silent deletion task on a stored phonological representation.The Quality of Phonological Representations and Silent Deletion Of Phonemes were then used as part of a comprehensive battery of phonological processing measures to explore the phonological processing skills of a well-defined group of children with SLI (n=21), typically developing children matched for age (n=21) and typically developing children matched for receptive language skills (n=21). The task battery also included measures of phonological awareness, short-term and working memory and rapid automatised naming.Children with SLI had generally weaker phonological processing skills than typically developing children matched for age. The profile was more varied when compared to typically developing children matched for language. Despite employing tight selection criteria, there was a wider spread of scores for children with SLI than for typically developing peers. The children with SLI demonstrated weaker performance on both short-term and working memory tasks, as well as a measure of quality of phonological representations.Overall, the children with SLI demonstrated an interesting pattern of phonological processing skills, with particular difficulty observed in phonological and working memory. Children with SLI also evidenced lower quality stored phonological representations of multisyllabic words. Performance on measures of phonological awareness was strong indicating that such skills can be taught, but that improvement in this area does not necessarily improve the quality of the underlying phonological representation.The research provided some support for a specific processing account of SLI. It also highlighted the importance of phonological and working memory in the development of accurate phonological representations.