Resistance of grammatical impairment to computerized comprehension training in children with specific and non-specific language impairments

Background: Receptive language impairments in school-age children have a poor prognosis, yet there is a dearth of research on effective interventions. Aims: Children's responses to a computerized grammatical training program were evaluated to consider whether repeated responding to spoken sentences with variable semantic content and the same syntactic structure would lead to consistent and fluent comprehension. Methods & Procedures: Children with receptive language impairments aged from 8 to 13 years were randomly assigned to three groups: Group S (n = 12) responded to reversible sentences in a computerized game, using speech stimuli with pauses before critical phrases. Group M (n = 12) had the same stimuli acoustically modified to lengthen and amplify dynamic portions of the signal. Group U (n = 9) was an untrained control group. On average, children in groups S and M completed over 1000 training trials, focusing on training comprehension of reversible sentences. Outcomes & Results: Although responses speeded up over the course of training, and most children performed well above chance, accuracy typically remained below 95% correct for constructions such as above/below and reversible active/passive. Trained groups did not differ from untrained children on language or auditory outcomes. There was no evidence that acoustically modified speech input enhanced comprehension. Conclusions: Rote training of comprehension of reversible sentences does not seem to be an effective approach to remediating such problems. For most children, the pattern of performance suggested that the problem was not a lack of syntactic knowledge, bur rather limited processing capacity that led to failures of on-line computation of meaning.