A meta-analysis of imitation abilities in individuals with autism spectrum disorders.
Imitation deficits in ASD are real and large—use tasks that score form, not just success, to spot them.
01Research in Context
What this study did
The team pooled every paper they could find on imitation in autism. They compared people with ASD to matched groups without ASD. Only studies that scored both the shape of the movement and the end result were kept.
The final pool covered many labs, ages, and tasks. This gives a big-picture view instead of one small sample.
What they found
Across all studies, people with ASD showed large imitation gaps. The gap vanished when tests only checked if the final goal was reached, not how the hands or face moved.
In plain words: clients can copy the end point, but they struggle to copy the exact form.
How this fits with other research
Diemer et al. (2023) seems to disagree. They saw no group difference when kids watched and then reported timing without moving. The trick is they removed the need to actually move. Once the body must act, the deficit shows up, matching the 2014 review.
Schulte-Rüther et al. (2017) also found no gap in quick, automatic facial mimicry. Again, the task was reflex-like, not a full goal-directed copy. These null results line up once you see the meta-analysis targets planned imitation.
McAuliffe et al. (2020) tracked kids learning new gestures via video. The ASD group’s curve was flatter and slower. This dynamic view extends the 2014 snapshot: the gap grows as trials add up.
Wang et al. (2021) zoomed in on speech and song. Only absolute pitch and duration were off; relative timing was fine. This finer grain confirms the broad deficit while showing where to aim therapy.
Why it matters
When you test imitation, score both form and end point. If you only record success, you will miss the problem. Use tasks that show the exact hand path, lip shape, or voice pitch. Target those details in intervention instead of just the final product.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add one form-based probe to your next session—score how the client shapes their hand or mouth, not just whether the puzzle is finished.
02At a glance
03Original abstract
Although imitation impairments are often reported in individuals with autism spectrum disorders (ASD), previous work has not yet determined whether these impairments are significant, specific to ASD, and present across the entire spectrum. This report of 53 studies on imitation in ASD seeks to determine whether individuals with ASD show significant imitation deficits, the magnitude of these deficits, and whether they are specific to ASD. Using standard meta-analytic techniques in a random-effects model, the data reviewed suggest that individuals with ASD show deficits in imitation, performing on average 0.81 SDs below individuals without ASD on imitation tasks. This deficit was specific to the condition of having ASD. Moderator analyses revealed that the average Autism Diagnostic Observation Schedule (ADOS) scores of groups of ASD participants were significantly and strongly negatively associated with the imitation abilities of these subjects, but average participant IQ was not associated with imitation abilities. Study setting, novelty of actions, format of imitation tasks (live vs. not), number of actions to imitate, or verbal prompts were not found to significantly affect the sizes of the imitation differences between individuals with and without ASD. The manner in which imitation was operationalized, however, had significant effects on whether imitation deficits were found between individuals with and without ASD. In tests that measured imitation of both form and end points, participants with ASD showed significant deficits compared with those without ASD; on tests of end point emulation only, individuals with ASD showed no deficits.
Autism research : official journal of the International Society for Autism Research, 2014 · doi:10.1002/aur.1379