But now a research project headed by Dr Paige Davis a lecturer in Psychology at the University of Huddersfield confirms that children diagnosed with autism are able to create and play with imaginary friends. Further research is to be conducted and could eventually help to develop new therapies.

The research described in the Journal of Autism and Developmental Disorders by Dr Davis and her three co-authors is based on evidence gathered in the USA and UK from 215 questionnaires completed by approximately equal numbers of parents of children with typical development and of children diagnosed with autistic spectrum disorder.

The findings do indicate that fewer children on the autistic spectrum create an imaginary companion — 16.2 per cent as opposed to 42 per cent of other children. Also children with autism began playing with their imaginary friends at a significantly later age and were proportionately more likely to play with a “personified object” such as a stuffed toy or doll.

But the argument of the new article from Dr Davis is that while there is a quantitative difference between the developments of imaginary friends between the two categories of children, there is no difference in the quality of the play.

The article includes examples of some of the imaginary companions created by children with autism whose parents took part in the project. They include Ghosty Bubble, an invisible bubble person who slept on a bubble bed next to the child; Mikey, an invisible Ninja who lived in a sewer; and Pretend Ada, an invisible version of a real school pal who played with the child when she needed a friend.

“The finding that children diagnosed with autism even spontaneously create such imaginary companions refutes existing beliefs that they are not imagining in the same way as typically developing children,” said Dr Davis.

“Imaginary companions are special because they are social in nature and children with autism have issues with social development and communication. So if you are actually creating a mind for an imaginary person you are involving yourself in a range of social activities that the autism diagnosis itself would say you couldn’t do.”

Dr Davis argues that if children with ASD are showing the same positive social developments as TD youngsters from the creation of ICs, then that could have implications for future intervention and lead to new therapies based on the imagination.

Her collaborators on the research and co-authors of the article were Elizabeth Meins of the University of York, Haley Simon of Drexel University in Philadelphia, USA, and Diana Robins of the AJ Drexel Autism Institute in Philadelphia. The article — titled Imaginary Companions in Children with Autism Spectrum Disorder — describes the research methodology and the findings in detail.

Now there are plans for further research into the benefits of imaginary companions to typically developing children and whether the same applies to autistic youngsters.