Non‐word repetition: An investigation of phonological complexity in children with Grammatical SLI

Abstract

We investigate whether children with Grammatical Specific Language Impairment (G‐SLI) are also phonologically impaired and, if so, what the nature of that impairment is. We focus on the prosodic complexity of words, based on their syllabic and metrical (stress) structure, and investigate this using a novel non‐word repetition procedure, the Test of Phonological Structure (TOPhS). Participants with G‐SLI (aged 12–20 years) were compared to language‐matched, typically developing children (aged 4–8 years). The results reveal that, in contrast to the controls, the accuracy with which the G‐SLI group repeated non‐words decreased as prosodic complexity increased, even in non‐words with only one‐ and two‐syllables. The study indicates that, in G‐SLI, complexity deficits in morphology and syntax can extend to prosodic phonology. The study highlights the importance of taking into account prosodic complexity in phonological assessment and the design of non‐word repetition procedures.

• SLI
• G‐SLI
• phonological complexity
• prosodic complexity
• non‐word repetition

Notes

1. On the reasons for treating a word‐internal coda as distinct from a word‐final consonant, see Harris (1994).

2. Two criteria were followed in choosing the segments used to fill out the prosodic frames of individual stimuli in TOPhS. Firstly, only unmarked segments were used; for example, none of the forms contains dental fricatives. Secondly, it was of course necessary to avoid real words. This raises the issue of whether lexical neighbourhood effects in real words might influence repetition performance in non‐words. This potential effect was not controlled for in the present study, although it is something that should certainly be investigated in future research employing the TOPhS design.

3. Scores on the GC‐ITPA and TROG tests should be taken as a general guide to the SLI participants' grammatical knowledge, since these tests assess a range of abilities, not only those that are problematic for G‐SLI participants. The tests are always supplemented with non‐standardized assessments that target specific aspects of morpho‐syntactic knowledge, which we see as crucial to the linguistic characterization of G‐SLI.