Comparing AMMT vs. Control Therapy in facilitating speech output in nonverbal children with autism


Schlaug, Gottfried

Beth Israel Deaconess Medical Center


2 years




United States


State/Province Full: 
United States

Up to 25% of individuals with autism are nonverbal. At present, there are hardly any established methods that have consistently produced improvements in speech output. Recently, our laboratory has developed Auditory Motor Mapping Training (AMMT), a novel intonation-based intervention, which aims to facilitate speech output in non-verbal children with ASD. Aim: To compare the efficacy of AMMT in facilitating speech output in non-verbal children with autism, relative to a control intervention. Participants will be 24 nonverbal children with autism between 4-8 years. Participants will be assigned to two therapy groups: 1) AMMT and 2) control therapy. For both conditions, three baseline sessions will be conducted, followed by 25 treatment sessions (5 days/week at 45 mins each), and 2 follow-up sessions. The accuracy of each child's speech output will be assessed throughout the course of treatment, and also during the post-treatment phase. The data collected from videos will be coded by independent raters. During each therapy session, the child will be seated facing the therapist. For the AMMT condition, a pair of tuned drums will be used, with each drum playing a fixed pitch. The therapist introduces the target words by intoning the words while simultaneously tapping the drums on the same two pitches to facilitate auditory-motor mapping. The child is led from listening to the therapist, to unison voicing, to partially-supported voicing, to immediate repetition, and finally to producing the target word on their own. For the control condition, the same stimuli and procedure will be used, but with the intonation and motor components omitted. That is, the therapist will introduce and model the target phrases in a speech-like (i.e., non-intoned) manner, without using the tuned drums. Compared to the control intervention, AMMT is expected to lead to improvements in the accuracy of consonant and vowel productions following treatment with transfer to items that were not trained during therapy. Such findings would be followed by development of a treatment manual to make the intervention accessible to autism treatment centers.