Naturalistic language intervention leads to successful use of a graphic mode communication system among four children with Down syndrome or cerebral palsy

Design: A single-subject multipleprobe design across three activities. The design was administered to each of four participants. Maintenance was examined during a two-week period following intervention. Maintenance was addressed by displaying pre-post probe data on the same targets that had been the focus of intervention. Probes were accumulated during each of three consecutive weeks by obtaining probe data three times per week. Generalization was not addressed in this investigation. Allocation: All participants were taught the same vocabulary items during a simultaneous group intervention activity. Order of stimulus presentation during intervention was not directly addressed. During probe opportunities (which took the form of a non-identity matching task), the interventionists randomized the presentation of stimuli. Blinding: Although not explicitly stated, it is very unlikely that interventionists and the individuals analyzing the data were blind to the purpose of the study. Study duration: The investigation was conducted over a time span of approximately three weeks for each of the four participants. Setting: The experiment was conducted in the participants' school building in a small room other than the children's classroom. All four children received instruction in the room simultaneously. Participants: Four children with cognitive impairment diagnosed with Down syndrome or cerebral palsy served as participants. None of the participants had previously had experience with augmentative and alternative communication (although the evidentiary base for this statement was not disclosed in the article). All participants were learning English as a second language. Each participant (1) had fewer than 15 intelligible spoken words as judged by a speechlanguage pathologist, (2) was between 8 and 12 years of age, (3) had successfully passed a hearing screening, otoscopic examination, and tympanometry, (4) was able to select a correct line-drawn choice from an array of four given a spoken sample, and (5) did not give evidence of comprehending any of the target vocabulary. Parental report confirmed inclusion criteria. Intervention: Prior to intervention, two pretests were carried out. In one an oddity task was implemented to ensure that each participant discriminated among all objects used in the investigation. A second pretest examined each participant's ability to match an object to the picture of the referent. An aided language stimulation strategy was implemented. The interventionist pointed to a referent and then, one second later, produced the spoken label that corresponded to the referent item/activity. Each 15-25-min session was focused around one of three activities, which were introduced sequentially. Each participant was exposed to the same eight vocabulary items in each activity. Two of the three activities required collaborative engagement among study participants. Data were collected via probes implemented on three days a week throughout the three-week period following stimulation sessions. There were five consecutive days of instruction for the eight vocabulary items in each activity, using a communication board comprised of 16 line-drawn symbols. Treatment integrity data were obtained on 20% of the sessions. Integrity was between 90% and 100%. Outcomes: Dependent measures included (1) total instances of aided language stimulation divided by the total number of opportunities, (2) ratio of statements to questions, (3) instances of interventionist's use of target vocabulary paired with simultaneous pointing, and (4) number of words correctly identified during probes by child participants. Reliability on the dependent measures, collected from 20% of the data, was 95-100%. Reliability on the transcription of probe data was between 90% and 95%. Attrition: All four of the original participants completed the experimental procedures.