Developmental dyslexia and specific language impairment: distinct syntactic profiles?

ABSTRACT

Recent work exploring syntax in developmental dyslexia (DD) has identified morphosyntactic deficits, striking parallelisms between children with DD and specific language impairment (SLI). The question remains open if the underlying causes for such deficits are related to difficulties in phonology, which is affected in DD, or to working memory, as has been previously reported for SLI. We focus on the production of third person accusative clitic pronouns (ACC3) and of homophonous definite determiners in French-speaking children with DD and SLI as well as typically developing (TD) controls. If syntactic complexity modulates performance of DD children, as has already been shown for SLI, we predict children with DD to perform significantly worse on ACC3 compared to definite determiners, which are homophonous but syntactically simpler. In addition, if impairment in ACC3 stems from phonology or working memory difficulties, we expect ACC3 performance in both clinical groups to relate to performance on non-word repetition or forward/backward digit spans. We studied 2 groups of 21 children and adolescents, with DD and SLI (7–15 years) and age-matched TD controls. Results reveal significant weaknesses with ACC3 in DD and SLI groups compared to TD controls, but no difficulty for homophonous definite determiners, confirming a deficit relating specifically to syntactic complexity. As for links to phonology and working memory, a single correlation emerged between ACC3 and the backward digit span in SLI, but not in DD, suggesting different underlying sources for syntactic deficits in these populations. Clinical implications of these results are discussed.

KEYWORDS:

• Developmental dyslexia
• specific language impairment
• syntax
• French

Acknowledgements

We would like to thank the children and their parents, the schools and the speech–language pathologists who agreed to participate in the study. Special thanks go to the master’s students in speech–language therapy for their help with testing the participants, namely Léa Chape, Fanny Chapelet and Sophie Tissot.

Declaration of interest

The authors report no declaration of interest.

Notes

¹. Memory loads were manipulated by presenting the sentences after or before the pictures to select, the last case being intended to increase working memory demand.

². We chose to include adolescents in the SLI group since we know that they may display persistent difficulties in both working memory and accusative clitic production (Montgomery, Magimairaj, & Finney, 2010; Tuller et al., 2011).

³. The particularity of this measure is that it is essentially non-verbal, including 36 items focusing on visual problem solving.

⁴. It is worth noting that there were no statistical differences between the monolingual and bilingual children for performance on accusative clitics (p = 0.4 for DD; p = 0.3 for SLI), definite determiners (p = 0.2 for DD; p = 0.9 for SLI), non-word repetition (p = 0.7 for DD; p = 0.2 for SLI) or digit-span tasks (p = 0.5 and 0.9 for DD; p = 0.5 and 0.9 for SLI).

⁵. We ensured that these controls did not differ from our clinical groups for chronological age (p = 0.4 for DD and p = 0.7 for SLI).

⁶. Note that the difference between ACC1 and ACC3, in favour of ACC1, has been shown in younger typically developing children, aged 4–6, by Tuller and collaborators (2011) and Delage et al. (2016).

⁷. We ensured that these controls did not differ from our clinical groups for chronological age (p = 0.6 for DD and p = 0.2 for SLI).

⁸. This result can seem unexpected but a series of observations may be relevant here. First, we only included participants for which speech and language therapists had confirmed expressive language disorders, as assessed by French standardized tests. Second, all participants with SLI had a level of language difficulty that continued to justify their being included in remediation programs, which is where they were recruited. Third, the ‘BILO’ standardized test comprises items which are very frequently used by speech therapists during their language evaluations and remediation programs. Due to this, participants with SLI may have received a little ‘training’ on these items, thus boosting their performance. However, amongst the five participants with SLI who managed to perform well on the BILO, two present difficulties on ACC3 production (with scores <2 SD compared to age-matched controls) and the other three display severe phonological disorders (with SD between −4 and −5 on non-word repetition).

⁹. We ensured that these controls did not differ from our clinical groups for chronological age (p = 0.9 for DD and p = 0.7 for SLI).

¹⁰. Our clinical groups were compared to the same control children as in the previous section (N = 38, Age range = 6;2–12;9, M = 9;10).

¹¹. Ex: Il les lui donne: ‘He is giving them to him/her’.

¹². Ex: Il est poussé par la fille: ‘He is pushed by the girl’.

¹³. This is in line with other reports claiming that the phonological deficit associated with DD is milder than that attested in children with SLI (Guasti, 2013; Talli, 2010).