.png){kind=icon}
.png){kind=icon}
Return
Guidelines for Emotion Recognition in Robot-Supported Interventions in Autism
.png){kind=icon}
OA
AI
M
K
T
D
H
M
B
B
A
DOI：10.1007/s12369-026-01401-2.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Therapy for children with autism spectrum disorder benefits from the use of diverse technologies, as most children with the condition are eager to interact with them. The integration of emotion recognition technologies with social robots offers promising opportunities to make therapeutic interactions more adaptive, personalized, and responsive to a child’s emotional state, potentially improving engagement and learning outcomes. The purpose of this study was to evaluate the feasibility of combining robot-based interventions with emotion recognition technologies for therapeutic use in autism. Specifically, we sought to identify best practices and obstacles in the integration of these technologies. To this end, we developed a comprehensive set of guidelines, which includes recommendations on observation channels, reliable acquisition of emotional expressions, technologies and devices, data processing, study design and reporting, and the specificity of emotional expressions in individuals with autism. They were evaluated using a triad of methods: questionnaires, focus groups, and a formal expert review using the AGREE II instrument. The results indicate that the guidelines are understandable and adequately described, and the focus groups helped identify the most relevant recommendations for researchers, therapists, and technology developers. The guidelines presented in this paper constitute the final version, which was refined based on the evaluation of preceding iterations.
Keywords:
Autism
Robot-enhanced therapy
Social robots
Automatic emotion recognition
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
3.7
Papers:
1.4K
Citations:
5.6K
Organization
Cited Papers