An adult draws on paper with a pen and smiles while doing so. Shortly afterwards a child is given the same pen and is allowed to play with it. Whether and how exactly they imitate the movement apparently also depends on how the action was shown beforehand. A new study of 102 preschool children shows that playful demonstrations can improve imitation. The differences between autistic children who use little spoken language and those who already use more are particularly interesting.
The study was published on 28 October 2026 in the Journal of Autism and Developmental Disorders. Researchers from the USA wanted to find out how children observe other people’s actions and under what conditions they spontaneously imitate them. The results provide new indications for early support, but also show how differently autistic children learn and communicate.
New autism study examines 102 preschool children
The research team examined a total of 102 children of preschool age. The average age was about two years and eight months. The participants were divided into three groups.
44 children had an autism diagnosis and used very little spoken language. A further 43 autistic children already had more developed language skills. The third group consisted of 15 children with no identified developmental concerns.
The researchers wanted to examine whether the children spontaneously imitated simple actions. The aim was not to expressly teach the children a particular movement. Rather, they were to observe what an adult demonstrated and then play with the same materials themselves.
In addition, a special eye-tracking system was used to examine what the children paid attention to during the demonstration. This allowed the researchers to determine how long the children looked at the adult’s face or at the action being demonstrated.
Playful demonstration led to more accurate imitation
In the study, the children watched short videos. In them, an adult showed simple actions with pens. For example, they drew a line or put several dots on a sheet of paper.
The actions were shown in two different situations. In one, the adult behaved neutrally. In the other version, they showed clearly positive feelings and carried out the action in a playful way.
The children were then given the same materials. They could decide for themselves what they wanted to do with them. The researchers observed whether the children imitated the action shown beforehand and how exactly they carried it out.
The result was remarkable. Across all three groups, imitation was more accurate in the playful situation than after the neutral demonstration.
The playful presentation also influenced where the children looked. While in the neutral demonstration the children looked longer at the action itself, in the playful version they paid more attention to the adult’s face.
Clear differences between the autistic children
At the same time, the study shows how much children with the same autism diagnosis can differ.
Autistic children who already used more spoken language imitated the actions shown about as often and as accurately as the children without developmental concerns.
It looked different for the minimally speaking autistic children. They watched the actions shown for a shorter time and imitated them less often and less accurately overall.
In the neutral demonstration, 40 percent of the minimally speaking children showed imitation. In the group of speaking autistic children it was 69 percent, in the comparison group 75 percent.
In the playful demonstration, the corresponding values were 46, 71 and 77 percent. The figures refer to the observations that could be evaluated in each comparison and must not be understood as general probabilities for autistic children.
What matters is that the minimally speaking children also benefited from the playful presentation. Their imitation became more accurate on average, although the differences from the other two groups remained.
Little spoken language does not mean a lack of interest
The study makes clear why it is important to look at language abilities and other areas of development in a differentiated way.
If a child speaks only a few words, this does not automatically mean that they have no interest in other people or cannot learn by observation in principle.
In the minimally speaking autistic children, less attention to the action demonstrated was seen. Attention to the adult’s face, by contrast, did not differ statistically significantly between the three groups.
The researchers therefore suspect that the way attention is distributed could play a role in the differences observed.
However, the data do not clearly show whether the lower imitation is directly related to language abilities. The groups also differed in other developmental features, including cognitive abilities.
Limited spoken language is also not the same as a lack of language comprehension. Which communication options are suitable has to be considered individually.
Why positive feelings can matter when playing together
The researchers had expected that the playful presentation would not make a difference equally in all groups. In fact, however, the accuracy of imitation improved regardless of whether a child had autism and how pronounced their spoken language was.
This suggests that positive emotional signals can play a role even in children with very different developmental profiles.
It is important that the study does not prove a general effect of cheerful facial expressions on learning. A few simple actions were examined in a short, standardised test situation.
The results do argue, however, for not paying attention only to the correct execution of a movement when playing together. The design of the situation and the kind of interaction could also influence how children take in other people’s behaviour.
What parents can take from the new study
For parents of autistic children it is particularly interesting that imitation can change depending on how a situation is designed.
When playing together, it can make sense to follow the child’s interests and to demonstrate simple actions with familiar objects. This can be, for example, rolling a toy car, stacking building blocks or drawing with pens.
The child does not have to repeat every movement. Watching, an idea for their own play or another form of taking part can also be part of a shared activity.
Some children need more time, others want to explore the materials themselves first. It is important to respect the child’s interests, means of communication and limits of tolerance.
No recommendation follows from the study to push children to imitate or to carry out a certain number of exercises. It shows only that the design of a playful situation can influence the imitation observed.
Augmentative and alternative communication remains important for minimally speaking children
For children who communicate little or not at all through spoken language, suitable augmentative and alternative communication can be important alongside shared play.
Depending on the stage of development, this includes, for example, picture cards, visual symbols or electronic communication aids with speech output. These options can make it easier for children to express wishes, needs and thoughts.
A child does not have to develop particular imitation skills first before gaining access to a suitable communication aid. Support for communication should be guided by their individual abilities and needs.
Which options exist is explained in our knowledge article on augmentative and alternative communication in autism.
What the results could mean for early intervention
Various play-based approaches are already used in the early support of autistic children. The aim can be, for example, to support joint attention, communication and taking part in everyday activities.
The British NICE guideline recommends including play-based strategies, involving parents and professionals, in suitable support services. The communication and developmental level of the individual child should be at the centre.
The new study adds to these recommendations findings on how emotional signals and attention to an action are related to imitating it.
It is not, however, an effectiveness study of a particular therapy. The children were observed only during a short experimental task. Whether a comparable design leads to progress in communication or independence over weeks or months was not examined.
The study has important limitations
The results should be interpreted cautiously also because the three groups were of different sizes. While 44 minimally speaking autistic children took part, the comparison group without developmental concerns consisted of only 15 children.
In addition, boys were represented in the sample markedly more often than girls. Whether the same results would appear in a larger and more diverse group is therefore open.
The researchers used short video recordings of an unfamiliar adult. In everyday life, however, children often learn together with familiar people and in situations that are much more complex.
The researchers also could not clearly distinguish what proportion of the differences observed was due to language development and what to other cognitive or developmental features.
The study thus shows associations within a particular test situation. It allows no general statement about how well autistic children learn through observation overall.
Why autistic children should be supported individually
The new study makes clear that the diagnosis of autism alone says little about how a child takes in or imitates a particular action.
In the task examined, the minimally speaking children differed markedly from the speaking autistic children. At the same time, all groups reacted in a similar way to the playful design of the demonstration.
For professionals, this can be a pointer to look more closely in support at what information a child takes in and under what conditions they engage in a shared activity.
For parents, too, it is decisive that support is not based on a diagnosis or general expectations alone. The child’s individual interests and means of communication should be taken into account.
How differently autism is identified and classified even in childhood is explained in our article on autism assessment in children.
Playful situations could open up new possibilities
The study published on 28 October provides new insights into imitation in autistic children. It is particularly remarkable that a playful design of an action was associated with more accurate imitation in all groups studied.
At the same time, it becomes clear that minimally speaking autistic children can show different developmental profiles from children who already use more spoken language.
Further studies now have to examine whether the results also appear when playing together with familiar people and whether effective support approaches can emerge from them in the long term.
An important message can be drawn from the results above all for families. When playing together, it is not only a matter of what a child imitates. It is equally important under what conditions they can take part and what opportunities they have to bring in their own interests and abilities.
Sources
Cavalli, G. et al. Spontaneous Imitation and Affective Modulation in Minimally Speaking Autistic Preschoolers. Journal of Autism and Developmental Disorders, 28 October 2026.
National Institute for Health and Care Excellence. Autism spectrum disorder in under 19s, support and management. Guideline CG170, section 1.3.1.
Deniz, E. et al. Parent-Mediated Play-Based Interventions to Improve Social Communication and Language Skills of Preschool Autistic Children. Review Journal of Autism and Developmental Disorders, 2024.
Found a mistake?
If something is factually wrong, out of date or unclear, let us know. You do not have to give any personal details.



