Facial Expression Databases and Autism Spectrum Disorder: A Scoping Review.
Most ASD emotion-recognition studies reuse the same few Western face sets—audit your stimuli for cultural fit before the next session.
01Research in Context
What this study did
Pandey et al. (2025) read 80 papers that used photos of faces to test emotion skills in kids and adults with autism. They listed every face database, the country it came from, and how well people with ASD did on the tasks.
The team wanted to see if the same white, Western faces keep popping up and whether that matters for kids who are not white or Western.
What they found
Most studies lean on only a handful of databases made in the US or Europe. The faces are mostly white and show big, clear emotions. Kids with ASD often struggle with these tasks, but not always.
Interventions that teach kids to read these faces can help, yet the review warns the practice may not travel well to other cultures.
How this fits with other research
Golan et al. (2008) and Root et al. (2017) already showed kids with ASD miss subtle or complex emotions. The new review says the faces used in those studies come from the same small pool, so the tasks may overstate the problem for non-white children.
Wang et al. (2023) found kids with ASD learn better when emotions start weak and grow stronger. The review adds that most databases only show high-intensity faces, so programs may skip the gradual step that helps.
Nuebling et al. (2024) meta-analysis shows autistic people have more emotion dysregulation overall. The scoping review points out we measure that dysregulation with tools that may not fit every client’s culture.
Why it matters
Before you run a social-skills lesson, check where the face cards came from. If every face is white and the child is not, swap in photos that match the child’s community or use video of local faces. This small step can make your emotion program fairer and more effective.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Open your emotion deck, note the source, and replace any mismatched faces with local photos or clips.
02At a glance
03Original abstract
Emotion recognition (ER) deficit in autism spectrum disorder (ASD) is widely accepted, and a variety of research was done to assess the deficit in ER under various conditions and intervention research to augment emotion recognition in mostly children with ASD. This scoping review attempted to summarize the diverse research in the context of ER in ASD with a focus on the facial expression databases used for research and intervention. PRISMA-ScR guidelines were followed, and 80 studies were selected after a search from electronic research databases like Web of Science and PubMed with predefined inclusion and exclusion criteria. The trends indicate a small sample size with mostly children as a sample and the ASD group with their matched counterparts. The results show mixed findings; the majority of work indicates a deficit in facial emotion recognition in ASD, but under certain stimuli and conditions, ASD performs well in facial emotion recognition. The interventions showed promising results with increased ER ability in children with ASD. The review also focused on the moderating factors in the study of ER, such as age, IQ, comorbidity, task paradigm, and cultural factors. A small subset of facial expression databases was widely used in the research and intervention, and it has been developed and validated in Western countries. Research gaps were highlighted, and recommendations for future directions were mentioned.
Autism research : official journal of the International Society for Autism Research, 2025 · doi:10.1002/aur.70030