ChatGPT-Delivered Physical Activity Intervention for Children With Autism Spectrum Disorder: Pre-Post Feasibility Study

Area of Science:

• Pediatric Health
• Autism Spectrum Disorder Research
• Digital Health Interventions

Background:

• Digital technologies are increasingly used to promote physical activity in children with autism spectrum disorder (ASD).
• Existing research lacks studies utilizing advanced AI tools like ChatGPT for this purpose.
• This study addresses the gap by exploring ChatGPT's potential in ASD physical activity interventions.

Purpose of the Study:

• To assess the feasibility of using ChatGPT to deliver physical activity recommendations to children with ASD.
• To evaluate the potential effectiveness of these AI-driven interventions on physical activity levels.
• To gather parent feedback on the usability and value of ChatGPT-generated activities.

Main Methods:

• A randomized controlled trial involving 26 families (parent-child dyads) with children diagnosed with ASD.
• The intervention group received physical activities recommended by ChatGPT, implemented by parents.
• Data collected via the Leisure Time Exercise Questionnaire (LTEQ) and a feasibility questionnaire.

Main Results:

• Parents reported ChatGPT-delivered physical activities as feasible and beneficial for increasing children's exercise.
• The AI-generated content was perceived as interesting and useful by parents.
• LTEQ data showed a significant increase in physical activity levels in the intervention group post-intervention.

Conclusions:

• ChatGPT-delivered physical activities show promise as an effective intervention for enhancing physical activity in children with ASD.
• The findings support further research into AI-driven health interventions for pediatric populations.
• This approach offers a novel, accessible method for promoting exercise in children with ASD.