Follow the white robot: Efficacy of robot-assistive training for children with autism spectrum disorder

TLDR: The results support the idea that robot-assisted interventions can be combined with the standard treatment plan to improve clinical outcomes, and show that the combination of robot- assisted training with standard therapy was more effective than the standard therapy alone.

About: This article is published in Research in Autism Spectrum Disorders.The article was published on 2021-08-01 and is currently open access. It has received 15 citations till now. The article focuses on the topics: Social skills & Autism spectrum disorder.

Figures

Figure 2 – On the left: experimental setup. On the right: the control panel of the application provided to the therapists to run the training. Using the panel, the therapists were able to start the trial (section B), annotate the answer and provide the feedback through the robot (section D). Additionally, therapists were able to provide intra-trial encouragements using additional feedbacks (section A) and to make the robot move again toward the target cube (section C).

Figure 3 – Violin plot summarizing the fixed effect on social interaction dimensions due to the robotassisted intervention. Red dots indicate mean values. Red bars indicate +/- 1 standard error. The asteriskdenotes a significant difference (** p < .01; * p < .05). The black dot denotes marginally significant difference (.05 < p < .055).

Figure 4 – Violin plot summarizing the fixed effect on behavioral requests dimensions due to the robotassisted intervention. Red dots indicate mean values. Red bars indicate +/- 1 standard error. Asterisks denote significant differences (** p < .01; * p < .05). The black dot denotes marginally significant comparison (.05 < p < .055).

Table 1. Grouping based on the ADOS levels and the child's age (n = 36)

Figure 1 – Crossover design adopted in the current study

Figure 5 – Violin plot summarizing the fixed effect on joint attention dimensions due to the robot-assisted intervention. Red dots indicate mean values. Red bars indicate +/- 1 standard error. Asterisks denote significant comparisons (** p < .01; * p < .05). Black dots denote marginally significant comparisons (.05 < p < .055).