Journal of Autism

Journal of Autism

ISSN 2054-992X

Does a Computer-Assisted language training program improve social conversational skills in Autism spectrum disorder?

Lawrence Welkowitz


Department of Psychology, Keene State College, Keene, New Hampshire USA.


Study Purpose: The primary hypothesis of this study is that a computer-assisted program that provides immediate feedback about matching critical acoustical features in speech may help to shape effective conversational skills in individuals with autism spectrum disorder (ASD).

Background: Studies in psycholinguistics have demonstrated the importance of vocal congruence or “matching” in conversational speech. In typical dyadic conversation those individuals who match critical acoustical patterns, including pitch, rhythm and volume, of their conversational partners are generally viewed as more socially effective. Individuals with ASD do not match these patterns as effectively as neurotypical individuals (NT).

Presentation of hypothesis: Pilot studies utilizing an Ipad-based software program that provides visual feedback of percentage match of pitch, rhythm, and volume is presented that suggests that such strategies may improve social conversational skills.

Testing Methods: Open trials of 10 daily training sessions with cohorts of adults and adolescents with diagnosed ASD showed improved matching abilities and generalizability to real life conversations. Larger numbers of subjects should be studied utilizing web-based access to the software along with cloud-based data analytics to assess efficacy.

Implications of the hypothesis: Computer-based feedback strategies may be useful for shaping pragmatic language skills in individuals with ASD. If found to be effective access to speech and language training could be extended beyond the current limitations related to access to face-to-face speech therapy.

Keywords: Autism, software, speech

ISSN 2054-992X
Volume 6
Abstract Download