This humanoid robot is called Kaspar and he can perform human-like features like talking, singing and even playing the tambourine. There are sensors inside Kaspar that are sensitive to human’s touch and that trigger a pre-programmed response to cast inappropriate behaviors down. Kaspar has been developed by the Adaptive Systems Research Group in the University of Hertfordshire to support children with autism in schools, homes and hospitals to improve their social interaction skills. As of today, only little research has been done on using autonomous robots in autism therapy. Therefore Robins, Dautenhahn, te Boekhorst and Billard (2005) decided to study the effects of interactions between humanoid robots and children with autism.
The authors hypothesized that repeated exposure to an interactive humanoid robot would increase basic social interaction skills of children with autism. Studies have shown that children with autism interact better with robots than with humans. Indeed, people with autism feel more comfortable in front of a computer than in front of a person. Moreover, it has also been shown that children with autism reacted more positively to robots having a simple appearance than robots having a humanized appearance, hence the authors decision to control the robots’ appearance in their study. To monitor the children’s reactions, they used a robot with two different appearances: one with a robotic appearance and one with a “pretty doll” appearance.
Robins et al. (2005) chose to conduct a longitudinal study to give the children enough time to express their initial reactions as well as progress to the experiment. The study took place in a school in the UK and four children, aged five to ten, with autism participated to the study. Each of them had different types of issues and received help by a carer in school. The trial procedure was composed of three phases of approximately three minutes, during which one researcher, one child and his carer were present in a room. During the first phase the children had to get used to the robot, then they were encouraged to interact with it and, during the final phase, they were left to interact with the robot on their own. The carer and the researcher only had to observe the children’s reactions, they should only interfere if the children was to harm the robot.
The quantitative and qualitative analysis showed that the children got familiar with the robot and expressed some progress in social interaction skills, such as imitation games. The results could support the initial hypothesis and some children even used the robot as an object of shared attention and included it in their own world, opening themselves up to it, talking to it and sharing experiences with it.
Nevertheless, in these results have to be seen in a global way because each child is in a different stage of development and their ages are different. The diversity of autism is also an important point to keep in mind and makes it difficult to generalize the results. The authors are aware of the limitations of their study and were able to suggest practical involvement for future studies, such as developing new interaction games.
In conclusion, this study supports the usefulness of robots like Kaspar in the everyday lives of children with autism and bring opportunities in the field of social interactions. On the contrary, they require a certain amount of money and are not financially affordable for every child. Plus, even if robots in therapy have been studied a lot over the past few years, researchers still do not agree on the benefits and we cannot assert that the therapeutic effects of the studies are linked to the interactions with the robot.
Bibliography
Robins, B., Dautenhahn, K., Te Boekhorst, R., & Billard, A. (2005). Robotic assistants in therapy and education of children with autism: can a small humanoid robot help encourage social interaction skills? Universal Access in the Information Society, 4(2), 105-120.
–
Written by: Jessica Couzinié, Arielle Dunais and Coraline Virot
