The grammatical morpheme deficit in moderate hearing impairment.

Background: Much remains unknown about grammatical morpheme (GM) acquisition by children with moderate hearing impairment (HI) acquiring spoken English.

Aims: To investigate how moderate HI impacts on the use of GMs in speech and to provide an explanation for the pattern of findings.

Methods & Procedures: Elicited and spontaneous speech data were collected from children with moderate HI (n = 10; mean age = 7;4 years) and a control group of typically developing children (n = 10; mean age = 3;2 years) with equivalent mean length of utterance (MLU). The data were analysed to determine the use of ten GMs of English. Comparisons were made between the groups for rates of correct GM production, for types and rates of GM errors, and for order of GM accuracy.

Outcome & Results: The findings revealed significant differences between the HI group and the control group for correct production of five GMs. The differences were not all in the same direction. The HI group produced possessive -s and plural -s significantly less frequently than the controls (this is not simply explained by the perceptual saliency of -s) and produced progressive -ing, articles and irregular past tense significantly more frequently than the controls. Moreover, the order of GM accuracy for the HI group did not correlate with that observed for the control group. Various factors were analysed in an attempt to explain order of GM accuracy for the HI group (i.e. perceptual saliency, syntactic category, semantics and frequency of GMs in input). Frequency of GMs in input was the most successful explanation for the overall pattern of GM accuracy. Interestingly, the order of GM accuracy for the HI group (acquiring spoken English as a first language) was characteristic of that reported for individuals learning English as a second language. An explanation for the findings is drawn from a factor that connects these different groups of language learners, i.e. limited access to spoken English input.

Conclusions: It is argued that, because of hearing factors, the children with HI are below a threshold for intake of spoken language input (a threshold easily reached by the controls). Thus, the children with HI are more input-dependent at the point in development studied and as such are more sensitive to input frequency effects. The findings suggest that optimizing or indeed increasing auditory input of GMs may have a positive impact on GM development for children with moderate HI.