Recognition of face and non-face stimuli in autistic spectrum disorder.
Autistic teens remember fewer faces because they use general memory, not special face software.
01Research in Context
What this study did
Arkush et al. (2013) asked autistic teens to remember faces and houses. They showed pictures on a screen, then later asked which ones the kids had seen.
Some faces were cropped to remove hair and clothes. The team wanted to know if autistic teens use special face skills or general memory tricks.
What they found
The autistic teens remembered fewer faces than typical peers. They remembered houses just as well, so the trouble was face-specific.
Cropping the faces did not hurt them more. That tells us they were not using special face-configuring skills. They treated faces like any other object.
How this fits with other research
Hartston et al. (2023) saw the same face-memory drop in adults with autism. They added eye-tracking and pinned the problem on early seeing, not later memory.
Sparaci et al. (2015) seems to disagree. They ran the Thatcher illusion test and found autistic teens showed normal holistic face effects. The tasks differ: Leo used memory for new faces, Laura used upside-down distortion. Memory needs stable templates; illusions test quick configural checks. Both can be true.
Wallace et al. (2008) and Hartston et al. (2024) back Leo. They show weaker face templates and less stable internal averages, explaining why memory fails.
Why it matters
Do not spend hours drilling face-feature spacing. Instead, teach broad visual memory skills like chunking, rehearsal, and distinctive feature tags. Use clear, consistent photos for staff IDs, class seating charts, and social stories. Pair names with salient non-face cues such as hair clips, shirts, or backpacks. These domain-general tactics help autistic learners build reliable person files without relying on elusive face expertise.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a distinctive visual cue (hat, sticker) to each peer’s photo during social-skills drills and teach the learner to label that cue.
02At a glance
03Original abstract
The ability to remember faces is critical for the development of social competence. From childhood to adulthood, we acquire a high level of expertise in the recognition of facial images, and neural processes become dedicated to sustaining competence. Many people with autism spectrum disorder (ASD) have poor face recognition memory; changes in hairstyle or other non-facial features in an otherwise familiar person affect their recollection skills. The observation implies that they may not use the configuration of the inner face to achieve memory competence, but bolster performance in other ways. We aimed to test this hypothesis by comparing the performance of a group of high-functioning unmedicated adolescents with ASD and a matched control group on a "surprise" face recognition memory task. We compared their memory for unfamiliar faces with their memory for images of houses. To evaluate the role that is played by peripheral cues in assisting recognition memory, we cropped both sets of pictures, retaining only the most salient central features. ASD adolescents had poorer recognition memory for faces than typical controls, but their recognition memory for houses was unimpaired. Cropping images of faces did not disproportionately influence their recall accuracy, relative to controls. House recognition skills (cropped and uncropped) were similar in both groups. In the ASD group only, performance on both sets of task was closely correlated, implying that memory for faces and other complex pictorial stimuli is achieved by domain-general (non-dedicated) cognitive mechanisms. Adolescents with ASD apparently do not use domain-specialized processing of inner facial cues to support face recognition memory.
Autism research : official journal of the International Society for Autism Research, 2013 · doi:10.1002/aur.1318