Facial Expression Production in Autism: A Meta-Analysis.
Autistic people make fewer, shorter, and lower-quality facial expressions, especially in real talk, so social reinforcement may need extra clarity.
01Research in Context
What this study did
The team pooled 48 studies that filmed or photographed autistic people while they made facial expressions.
They counted how often, how long, and how well the expressions matched the moment.
All ages and IQ levels were included, giving 1,500 autistic faces to compare with 1,500 non-autistic faces.
What they found
Autistic faces moved less often and for shorter time than non-autistic faces.
The quality of the smile or frown was also rated lower, but the strength and speed of the movement stayed the same.
The gap was largest when people were just talking naturally, not when an adult said "show me happy."
How this fits with other research
Taylor et al. (2017) saw the same pattern in the voice: autistic adults spoke emotional sentences louder and slower, sounding odd even though the words were clear.
Tonnsen et al. (2016) found the same thing in tone-language speakers: pitch swung more in autistic adults, again showing expressive signals that are "off" even when language is intact.
Together, the three papers show the problem is not mouth or voice alone—it is the whole emotion broadcast system.
Why it matters
If the face and voice send weak signals, social rewards like praise may not land. Check whether your client’s smile looks and sounds natural before you use it as reinforcement. When you teach emotions, practice in real conversations, not just in drill blocks, because that is where the biggest gap shows up.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Film a 2-minute free-play clip, count genuine smiles, and add extra visual or vocal cues if the rate is below peer level.
02At a glance
03Original abstract
This meta-analysis summarized studies that examined group differences on the production of facial expressions in participants with ASD compared to typically developing or nonautistic clinical comparison groups. The overall summary effect from 67 effect sizes representing the average ASD-comparison group differences in facial expressions was -0.481, indicating a moderate effect size. We conducted subgroup analyses to group effect sizes according to separate facial expression abilities identified in the literature. These analyses revealed that participants with ASD display facial expressions less frequently and for less amount of time, and they are less likely to share facial expressions with others or automatically mimic the expressions of real faces or face stimuli. Their facial expressions are also judged to be lower in quality and are expressed less accurately. However, participants with ASD do not express emotions less intensely, nor is their reaction time of expression onset slower in response to odors, startling sensations, or in response to face stimuli in mimicry studies. ASD-comparison group differences were moderated by matching procedures, age, and intellectual functioning of the ASD participants suggesting that persons with higher IQ and larger number of accumulated life experiences are better able to produce facial expressions that are more consistent with "neurotypical" norms. Group differences were also stronger for "covertly elicited" than "explicitly elicited" facial expressions suggesting individuals with ASD may naturally produce facial expressions differently from other populations, but are less impaired in expressing emotions typically when prompted to do so in a laboratory setting. Autism Research 2018, 11: 1586-1601. © 2018 International Society for Autism Research, Wiley Periodicals, Inc. LAY SUMMARY: We reviewed studies that compared facial expressions in people with and without autism. Results revealed that facial expressions of people with autism are atypical in appearance and quality and are used atypically to regulate social interactions. The magnitude of these differences was influenced by participant characteristics (e.g. age and intellectual functioning), and by how facial expressions were measured and analyzed in various studies.
Autism research : official journal of the International Society for Autism Research, 2018 · doi:10.1002/aur.2037