Parental recasts and production of copulas and articles by children with specific language impairment and typical language

In intervention, children with specific language impairment (SLI) have been shown to develop productive use of morphemes in response to target-specifi c recasts at rates generally equivalent to younger, language-matched childr en with typical language development (TL). Our previous work demonstrated t hat in conversation, the overall recast rates produced by parents of childr en with SLI and those with TL are similar. Still, despite their apparently typical ability to use recast input in intervention and their equivalent en vironmental exposure to recasts, children with SLI continue to demonstrate grammatical delays in comparison to children with TL, The purpose of this study was to examine three possible resolutions to this paradox. We examined target-specific copula and article recast usage by 10 parents of children with SLI acid 10 parents of younger language-matched c hildren with TL, and we examined their children's productions of these same forms at three points across an 8-month period. The results provide strong support only for the third of the proposed hypotheses. Contrary to the pre dictions of the first hypothesis, a strong, positive relation was observed between the copula recasts used by parents of children with TL at Time 1 an d their children's use of copulas 8 months later. On the other hand, correl ations between recasts of articles by parents and later production of artic les by their children were not statistically reliable. Contrary to the seco nd hypothesis, parents of children with SLI and those with TL produced equi valent rates of article and copula recasts. The third hypothesis received s upport on two essential counts. First, although significant correlations we re found between parental recasts of copulas and production of this form 8 months later for the children with TL, no such relations were observed for the group with SLI. Second, the rate of parental target-specific recasts wa s less than a quarter of the rate provided in the successful intervention o f Camarata, Nelson, and Camarata (1994). We conclude that children with SLI can benefit substantially from the grammar-facilitating properties of reca sts, but only when the recasts are presented at rates that are much greater than those available in typical conversations with young children.