record

Thesis Info

LABS ID
00836
Thesis Title
Social Robot Videos: A New Task to Measure Theory of Mind Towards Humanoid Robots in Children with Autism Spectrum Disorder
Author
Ziggy O'Reilly
2nd Author
3rd Author
Degree
Master of Biological Arts
Year
Number of Pages
147
University
University of Western Australia
Thesis Supervisor
Ionat Zurr
Supervisor e-mail
ionat.zurr AT uwa.edu.au
Other Supervisor(s)
David Silvera-Tawil, Diana Tan
Language(s) of Thesis
English
Department / Discipline
School of Human Sciences
Copyright Ownership
Languages Familiar to Author
URL where full thesis can be found
Keywords
animation, social robots, human-robot interaction, socially-assistive robots, autism, social cognition, Theory of Mind,
Abstract: 200-500 words
The social communication and interaction difficulties experienced by individuals with autism spectrum disorder (ASD) are thought to be related to impairments in Theory of Mind (ToM)—the ability to attribute mental states to others which is crucial in understanding social behaviour. Furthermore, individuals with ASD are said to have a preference for systems. Consequently, computer-mediated interventions that aim at improving ToM ability were developed to leverage this preference for systems in ASD. However, computer-mediated interventions are limited in their representativeness of real-life social interactions, thus may lead to restricted generalisability. In contrast, interventions with humanoid robots are argued to be more representative of real-life social interactions and could have greater therapeutic outcomes if applied to teach recognition of mental states. However, there is currently no appropriate measure of ToM involving humanoid robots to assist in examining the degree to which these interventions are effective at achieving generalisation. This project aimed to (1) use animation techniques to develop a sensitive and practical ToM task using humanoid robots (the Social Robot Videos task), based on the Frith-Happé Animations task; and (2) investigate the face validity of the Social Robot Videos task in a preliminary study. Both ToM tasks were administered to seven children with ASD and nine typically-developing children aged between 8 and 11 years. The differences between children with ASD and typically-developing children were comparable between the two tasks and the ToM condition of the Social Robot Videos task could discriminate between the groups. Preliminary results suggest that the Social Robot Videos task has face validity and that the task could be a useful measure of ToM intervention outcomes using humanoid robots.