Fast updating of stimulus history reveals weak internal representations of faces in autism.
Autistic adults don’t form stable ‘average’ face pictures, so we must teach facial features explicitly.
01Research in Context
What this study did
The team showed adults a long stream of faces on a screen. Each face morphed slightly toward the ‘average’ face the viewer had seen so far.
Autistic and non-autistic adults watched the same sequence. The researchers tracked how fast each group updated their mental ‘average’ face.
What they found
Non-autistic adults quickly shifted their idea of the average face after only a few new images.
Autistic adults updated much more slowly. Their internal face template stayed weak and fuzzy, even after many faces.
How this fits with other research
Hartston et al. (2023) already showed that autistic adults score lower on standard face-memory tests. The new study tells us why: their brains don’t build a stable face ‘average’ to compare against.
O'Connor et al. (2008) found weaker activity in face-selective brain areas in autism. The slow updating seen here gives a behavioural reason for that low brain activity—less reliable input templates.
Spriggs et al. (2015) tracked babies who were later diagnosed with ASD. These babies looked less at eyes from infancy. The current data link that early gaze gap to still-fragile face templates in adulthood.
KAgiovlasitis et al. (2025) recently repeated the gaze-reduction finding in Indian adults with high autistic traits. Together the papers show the face-input problem is consistent across cultures and ages.
Why it matters
If your learner has ASD, don’t assume repeated exposure to faces will create a strong memory. Instead, teach specific features—ear shape, eyebrow curve, glasses style—one at a time. Use flashcards that highlight these parts and ask the learner to name them. This gives the brain the clear anchors it misses when it tries to average whole faces.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick one distinctive feature (e.g., glasses) and run 10-trial matching drills with your client—point, name, and praise each correct match.
02At a glance
03Original abstract
Atypical perception has been widely reported in autism spectrum disorders, and deficits in face recognition, specifically, are argued to be closely associated with social impairment experienced by these individuals. However, it is still debated (a) whether deficits are perceptually based, and (b) what the role is of experience-based refinements of perceptual face representations in autism. We investigated the effect of short- and long-term experienced stimulus history on face processing. Autistic and non-autistic individuals performed same-different judgments in a serial discrimination task where two consecutive faces were drawn from a distribution of morphed faces. Use of stimulus statistics was measured by testing the gravitation of face representations towards, the mean of a range of morphed faces around which they were sampled (regression-to-the-mean). The results show that unlike non-autistic individuals, representations of own- and other-race faces were equally biased by stimulus statistics in autistic individuals. Moreover, autistic individuals used the most recently exposed faces without forming a strong internal representation based on the overall experienced faces, indicating a weaker internal model of the "typical" averaged face. This accumulated history of faces may underlie typical face specialization, and thus may account for the reduced specialization for own-race faces shown in autism. The results shed light on the way autistic people process and recognize faces, and on the basic mechanisms underlying atypical face perception.
Autism research : official journal of the International Society for Autism Research, 2024 · doi:10.1002/aur.3236