Using Social Robots to Improve Directed Eye Gaze of Children with Autism Spectrum Disorders

Background: Children with Autism Spectrum Disorders (ASD) experience deficits in appropriate verbal and nonverbal communication skills including motor control, emotional facial expressions, and coordinated eye gaze. Recent research suggests that children with ASD exhibit positive social behaviors when interacting with robots compared to their peers that do not interact with robots [1-3]. These positive behaviors include directing emotional facial expressions (e.g., smiling towards social partner), imitating gestures and coordinating eye gaze with other forms of communication. These investigations suggest that interaction with robots may be a promising approach for rehabilitation of children with ASD.

Objectives: The relatively emerging field of robot-assisted therapy for autism aims to utilize robotic systems to develop novel interventions for improving the quality of life for children with ASD and their families. The main objective of our research study is to investigate whether a humanoid robot can successfully be used to improve directed eye gaze skill in children with high functioning autism (HFA) or Asperger syndrome. The current project is a preliminary evaluation of how individuals with HFA interact with a humanoid robot.

Methods: Participants in this study are 8 male children ages 7-17 (M=11.5 years) diagnosed with ASD or AS.  The study uses NAO, an autonomous, programmable humanoid robot from Aldebaran Robotics. Participants interacted with NAO during a series of conversations and interactive games (i.e., guessing games) across 3 sessions. During the conversations and games, NAO asked participants questions, asked them to comply with simple requests, and provided feedback during the conversations and games. Sessions were video-recorded using cameras installed in the session room as well as through NAO’s front-facing camera. Videos were later scored to obtain measures of the duration and frequency of gazes directed to the robot.

Results: Participants spent, on average, approximately 50% of the session directing their gaze towards NAO (M=49%, range= 31-82%). Participants also engaged in frequent gaze shifts towards and away from NAO during the session (M=39, range=27-79) indicating that they were directing their attention towards the robot and modulating their gaze during the sessions.

Conclusions: Overall, these preliminary findings support the aim of utilizing humanoid robots as possible therapeutic agents for individuals with ASD. These results show that participants were engaged with the robot and directed their attention to the robot during a large portion of the sessions. Data collection is on going and parallel measures of participants interacting with the examiner during identical interactions will allow a direct comparison of the interaction to human and humanoid interactive partners. These results will serve as an important basis to significantly advance the emerging field of robot-assisted therapy.

References:

[1]      B. Robins, et al., "Robot-Mediated Joint Attention in Children with Autism", A Case Study in Robot-Human Interaction, Interaction Studies, vol.5(2), pp.161-198, 2004.

[2]        B. Scassellati, "How Social Robots Will Help Us to Diagnose, Treat, and Understand Autism",  Procs. Int. Symp. on Robotics Research, San Francisco, USA, 2005.

[3]        D. J. Ricks, M. B. Colton, "Trends and considerations in robot-assisted autism therapy", IEEE International Conference on Robotics and Automation, pp. 4354-4359, 2010.