Individuals with autism spectrum conditions (ASC) sometimes struggle to empathize and relate to others in social situations. Social skills training (SST) is used to help these individuals actively engage in social settings. However, because some individuals with ASC have difficulty understanding how their behaviors affect others, they might lack the motivation to complete SST with a therapist. Moreover, they may find the dynamics of human emotions and expressions in SST overstimulating, leading to a decrease in focus and learning.

To overcome these barriers, a recent study aimed to modify SST by introducing a robotic delivery format, whereas participants with ASC interacted with robots instead of therapists. This approach also has the advantage of facilitating standardized interactions, thus allowing repetitive learning. As well, the authors theorized that the participants would remain more engaged with the robots than they would with human therapists.

In this study, the researchers prepared social interactions between an android robot (i.e., a realistic human-looking robot) and a rudimentary humanoid robot. The study took place over five sessions. In the first half of each session, two participants in the experiment room observed five social interactions between the robots (these interactions were teleoperated by a research assistant in an adjacent room). Then, they were asked to report the robot’s emotions and root causes of those emotions to identify appropriate and inappropriate social responses. For the second half of the sessions, the two participants faced each other and teleoperated the robots through the same social situations they saw in the first half of the session. Prior to the first session and two weeks after the last session, the participants completed an assessment of their sociability. This allowed the researchers to determine whether their social skills improved as a result of the intervention.

Overall, the participants scored higher on the sociability scale after the intervention than prior to the intervention, although there was no change in empathy scores. All participants completed the study, therefore addressing the researchers’ concerns about a lack of motivation for SST. Moreover, despite this study taking place in a very controlled environment, the participants’ social skills generalized into the real world.

While this study offers promising advances in SST, there were some limitations. For one, there were only fourteen participants in the study, most of whom were male. Therefore, the results may not be applicable to other populations, including women with ASC. As well, the study did not include a comparison group where SST would have been conducted by human therapists.

On a final note, there has been other interesting work in this field. For instance, a 2022 study involving children with ASC found no major difference between a psychosocial intervention involving a “robot assistant therapist” and the same intervention conducted by a human therapist only. However, satisfaction was higher in the robot-assisted intervention group. In conclusion, robot-assisted interventions offer promising advancements for improving psychosocial skills in individuals with ASC.