Enhancing theory of mind in autism through humanoid robot interaction in a randomized controlled trial

Abstract

Autism Spectrum Disorder presents significant challenges in social cognition, particularly in understanding others' thoughts, emotions, and intentions. Traditional interventions often rely on role-playing games with human therapists or inanimate objects, but these approaches may lack consistency and ecological validity. This study integrated Applied Behavior Analysis principles with robot-assisted training to improve social cognition in children with autism. A randomized, two-period crossover trial involving 32 children (mean age = 7.53 ± 1.32 years, 7 females) compared robot-assisted training using the humanoid robot iCub with standard therapy and an active human-assisted control condition. During robot-assisted sessions, children engaged in structured social role-play scenarios, practicing essential social skills such as perspective-taking, joint attention, and recognizing intentions. The robot's human-like appearance and adaptive behavior provided an engaging, predictable learning environment. Results indicated that robot-assisted training significantly improved social cognition, in contrast to traditional therapy and the human-assisted control group, where no improvements were found. Importantly, the active human control confirmed that these improvements were driven by the robot's presence rather than the protocol itself. These findings demonstrate the potential of humanoid robots as effective therapeutic tools for enhancing social skills in children with autism, offering a scalable and engaging complement to existing clinical practices. Clinical Trial Registration: ISRCTN15341724, registered on May 6, 2025. Available at: https://www.isrctn.com/ISRCTN15341724 .

Keywords: Applied behavior analysis (ABA); Autism spectrum disorder (ASD); Humanoid robots; Robot-Assisted training; Social cognition training; Theory of Mind (ToM).

Fig. 1

Box-plot summarizing the fixed effects of the robot-assisted trainings in comparison with the Standard Therapy (ST) on NEPSY-II subscale Theory of Mind. The plot on the left (A) represents pre–post delta scores for each condition; the plot on the right (B) shows performance across three assessment phases (T0–T1-T2). The group that first received RAT improved significantly after the first training phase (pink bars; T0 vs. T1), while the group that initially received ST showed no improvement during that phase (turquoise bars; T0 vs. T1) but improved following RAT (T1 to T2). Horizontal lines indicate statistically significant comparisons (***p < .001).Crossover Effects and Training Retention.

Fig. 2

Box-plot illustrating pre–post NEPSY-II Theory of Mind scores in the Human-Control (HC) group. Children in this group received the structured training protocol delivered by a trained clinician, without robotic assistance. No statistically significant improvement was observed between T0 and T1.

Fig. 3

The figure illustrates two examples of the setups employed for the activity. Stickers and backgrounds were altered in each session to introduce novel stimuli, thereby mitigating potential boredom among the children throughout the sessions.

Fig. 4

The figure depicts a flowchart representing the various phases of the role-playing game carried out by the child and the robot. On the left (A), an example of the phase where the child is asked to respond to a behavioral request, which is executed by the robot and repeated if the child fails to respond; if successful, the robot reinforces the response with positive feedback (i.e. the robot raises its arms and hands in a gesture of victory, reinforcing the child with verbal prompts such as ‘let’s keep it up’ or ‘you’re doing great’). On the right (B), an example of the phase where the child is asked to perform a behavioral response, which may be prompted by the robot. If the child immediately complies with the request, the trial proceeds and the robot promptly responds, otherwise, the prompt to execute the request is given a second time.