Autistic children with imaginary friends may have stronger social skills and be better able to understand others’ minds than their peers without imaginary friends, according to a new study.

While previous research has shown that neurotypical children with imaginary companions (ICs) have significantly more developed theory of mind (ToM) and social understanding, this article is the first to consider whether the same applies to autistic children.

Almost half of participants said their autistic child had an imaginary friend

The research included 124 parents of autistic children aged five to 12 years old. They were recruited through Facebook support groups and through a participant sourcing database called Prolific.

The participants were then asked to fill out an online questionnaire about their child’s age, sex, IC status, communication abilities, ToM (their ability to understand what someone else is feeling) and social skills.

Almost half of the participants said their child had an imaginary friend, a much larger number than previous research has suggested.

These parents rated their children significantly higher than those who did not create ICs on both ToM and social skills regardless of communication ability.

Boys were less likely to have an IC than girls, and those with an IC had significantly more interest in making and spending time with real friends, according to their parents. However, IC status did not relate to the number of real friends the child already had made.

Why are autistic children with ICs more likely to have improved social skills?

The authors suggest a number of explanations for these results. As suggested in previous research, it is thought that ICs could improve children’s social skills and ToM understanding because the creation of another mind gives them the opportunity to think about how other minds work.

It also gives the child a chance to practice social interactions in the absence of a real individual. This could be particularly useful to an autistic child who may not have as many chances to interact socially, due in part to misinterpretation of autistic behaviour as indicating lack of social motivation.

However, they note that children already excelling in socio-cognitive areas are more likely to create ICs, so the direction of causality is not clear.

Another possibility is that ToM, social skills and ICs are caused by another variable that has not been identified. This would mean that all three areas were impacted by this unidentified variable.

Further research needed

For this reason, the authors of the paper say further research is necessary to examine the direction of causality of the relationship between ICs and social skills among autistic children.

Nevertheless, the results add to the evidence that with respect to the creation imaginary friends and their potential benefits, the play profiles of autistic children are similar to the general population.

“It also provides more evidence that the understanding of others’ minds is not ‘all or nothing’ in autism and gives reason for researchers to investigate whether the causes of these differences are the same or different for autistic children,” the authors conclude.