TY - CONF

T1 - Nonword repetition is also difficult for Danish-speaking children with developmental language disorder

AU - Christensen, Rikke Vang

PY - 2018/6/26

Y1 - 2018/6/26

N2 - This presentation focuses on the characteristics of a Danish nonword repetition task and on its potential as a clinical marker of developmental language disorder in Danish. During the last decades, processing dependent language tasks have been promoted in the field of developmental language disorder. These tasks activate cognitive mechanisms involved in general language skills without building on children's experience to the same extent as traditional language tasks. One type of language processing task is nonword repetition (NWR) which has shown diagnostic potential in a number of languages such as English (e.g. Conti-Ramsden, Botting & Faragher, 2001), Italian (Dispaldro, Leonard & Deevy, 2013) and Swedish (Sahlén et al., 1999). Graf Estes, Evans, and Else-Quest (2007) found that the mean difference in NWR performance between children with developmental language disorder and typically developing controls had a large effect size, d = 1.27, in a meta-analysis of 23 studies of English-speaking children. The meta-analysis also showed that NWR tests were not interchangeable and pointed to word length, articulatory complexity and wordlikeness as factors, which potentially influence NWR performance and the magnitude of group differences. The present study includes data from nonwords repeated by children with developmental language disorder (DLD; n = 26) aged 5;9 (years;months) to 14;1 and children with typical development (TD; n = 88) aged 5;6 to 13;4. These two groups differed on language measures, but not on chronological age. The NWR task developed for Danish includes twenty nonword, two to five syllables long (e.g. pusame and nofelybati). All syllables included in the nonwords are CV combinations that do not correspond to real Danish words or affixes. The wordlikeness is low as assessed by Danish adults (Sandberg, 2013). This probably relates to (1) the stressed syllable of many of the nonwords occurring medially or finally rather than initially as in most Danish words, and (2) the use of full vowels as final segments in all nonwords rather than the unstressed schwa, which is highly frequent in word final position in Danish. The results show group differences with respect to both the number of nonwords accurately repeated and the number of phonemes accurately repeated: Participants with DLD performed markedly below the level of their typically developing peers. In the presentation, the accuracy of NWR relative to the length and stress pattern of the nonwords is also addressed, as is the identification accuracy (sensitivity, specificity and likelihood ratios) of the NWR task. The results are discussed in the light of previous NWR studies and views on the nature of DLD. The research was funded by the Danish Agency for Science, Technology and Innovation.

AB - This presentation focuses on the characteristics of a Danish nonword repetition task and on its potential as a clinical marker of developmental language disorder in Danish. During the last decades, processing dependent language tasks have been promoted in the field of developmental language disorder. These tasks activate cognitive mechanisms involved in general language skills without building on children's experience to the same extent as traditional language tasks. One type of language processing task is nonword repetition (NWR) which has shown diagnostic potential in a number of languages such as English (e.g. Conti-Ramsden, Botting & Faragher, 2001), Italian (Dispaldro, Leonard & Deevy, 2013) and Swedish (Sahlén et al., 1999). Graf Estes, Evans, and Else-Quest (2007) found that the mean difference in NWR performance between children with developmental language disorder and typically developing controls had a large effect size, d = 1.27, in a meta-analysis of 23 studies of English-speaking children. The meta-analysis also showed that NWR tests were not interchangeable and pointed to word length, articulatory complexity and wordlikeness as factors, which potentially influence NWR performance and the magnitude of group differences. The present study includes data from nonwords repeated by children with developmental language disorder (DLD; n = 26) aged 5;9 (years;months) to 14;1 and children with typical development (TD; n = 88) aged 5;6 to 13;4. These two groups differed on language measures, but not on chronological age. The NWR task developed for Danish includes twenty nonword, two to five syllables long (e.g. pusame and nofelybati). All syllables included in the nonwords are CV combinations that do not correspond to real Danish words or affixes. The wordlikeness is low as assessed by Danish adults (Sandberg, 2013). This probably relates to (1) the stressed syllable of many of the nonwords occurring medially or finally rather than initially as in most Danish words, and (2) the use of full vowels as final segments in all nonwords rather than the unstressed schwa, which is highly frequent in word final position in Danish. The results show group differences with respect to both the number of nonwords accurately repeated and the number of phonemes accurately repeated: Participants with DLD performed markedly below the level of their typically developing peers. In the presentation, the accuracy of NWR relative to the length and stress pattern of the nonwords is also addressed, as is the identification accuracy (sensitivity, specificity and likelihood ratios) of the NWR task. The results are discussed in the light of previous NWR studies and views on the nature of DLD. The research was funded by the Danish Agency for Science, Technology and Innovation.

M3 - Poster

Y2 - 25 June 2018 through 26 June 2018

ER -