Bilingual children perform differently from children with PLI on standardized tests
medwireNews: Bilingual children achieve higher nonword repetition (NWR) scores and are unaffected by the length of the nonwords compared with children with primary language impairment (PLI), research shows.
In addition, NWR and sentence imitation (SI) correctly identified children with PLI from children with normal development regardless of bilingualism.
"The results of this study suggest that language-processing measures are less affected by amount of exposure to a language than are measures of language knowledge," say researchers Elin Thordardottir and Myrto Brandeker (McGill University, Quebec, Canada).
Published in the Journal of Communication Disorders, the researchers conducted two assessments, the first including 84 5-year-old children acquiring French and English simultaneously with different amounts of exposure to the two languages.
In the second test, monolingual and bilingual children with and without PLI were assessed for NWR, SI, and receptive vocabulary in French to determine their diagnostic accuracy.
In the first test, which evaluated the overall performance on SI and NWR in terms of items repeated correctly, the processing measures, particularly NWR, was less affected by previous exposure to language than vocabulary measures.
"Overall, these findings indicate that SI is more highly influenced by previous exposure than is NWR," report Thordardottir and Brandeker. "This was expected given the fact that SI involves meaningful material."
They add that with a critical exposure level of approximately 35% to 40% of waking hours since birth, 5-year-old children can be expected to perform similarly to native speakers on the English NWR.
The researchers also observed that the level of previous bilingual exposure had little impact on the children's ability to repeat nonwords of increasing length.
In the second assessment, NWR and SI were able to distinguish children with PLI from children with typical development regardless of bilingualism.
Bilingual and monolingual PLI children did not differ significantly from each other, nor did bilingual and monolingual children with typical development. There was a significant difference between PLI and typical development in NWR regardless of bilingual exposure.
NWR and SI had a high level of sensitivity for detecting PLI, 85% and 92%, respectively, regardless of whether the children were bilingual or monolingual. Specificity for NWR and SI was high in monolingual children, but declined in bilingual children (specificities for NWR and SI were 79% and 57%, respectively).
NWR and SI tasks are frequently used as standardized protocols for the assessment of language ability. The identification of PLI in bilingual children, however, is often difficult because these children tend to score lower than monolingual children on standardized language tests.
medwireNews (www.medwire-news.md) is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2012
By medwireNews Reporters