"Sequencing Matters": Investigating Suitable Action Sequences in Robot-Assisted Autism Therapy

Area of Science:

• Robotics
• Developmental Psychology
• Special Education

Background:

• Social robots show potential for Autism Spectrum Disorder (ASD) therapy.
• Current robot interventions lack flexibility for individual needs.
• Personalization is key to enhancing robot-assisted therapeutic efficacy.

Purpose of the Study:

• Investigate the impact of varied robot action sequences on children with ASD.
• Target attention skills through structured, socially interactive robot tasks.
• Explore personalization in robot-assisted interventions for ASD.

Main Methods:

• Developed a robotic prompting scheme on a NAO robot using an autism diagnostic tool.
• Integrated the scheme into an interactive storytelling scenario with screens.
• Programmed robot in three modes: Assess (diagnostic-inspired), Therapy (personalized), and Explore (random).

Main Results:

• Exploratory study involved 11 young children with ASD.
• Each robot mode demonstrated unique usefulness and limitations.
• Results varied based on individual child profiles and interaction goals.

Conclusions:

• Findings highlight the potential of personalized robot action sequences.
• Suggests a need for adaptive strategies to balance short-term and long-term therapeutic goals.
• Paves the way for more sophisticated personalized robot-assisted therapies for ASD.