Generalized imitation and the mand: inducing first instances of speech in young children with autism.

It has been suggested that the use of a generalized motor imitation sequence before a vocal model may be an effective procedure for teaching nonvocal children with autism to speak. However, the tactic has rarely been empirically demonstrated. The purpose of this experiment was to test the effects of presenting a rapid generalized motor imitation sequence before an opportunity to imitate on the vocal speech of nonvocal children with autism. Participants emitted no vocal imitations during a mand training baseline. During the intervention, a rapid motor imitation sequence was presented before an opportunity to imitate a model's vocalizations. The teacher's presentation of the rapid motor imitation sequence was then faded by presenting an opportunity to vocally imitate without the sequence followed by an opportunity to independently mand. Results of the intervention phase indicated that all of the participants began to vocalize with the generalized motor imitation sequence and that mands were maintained during a follow-up phase and 3-month follow-up probes.