Social initiations by autistic children to adults and other children.
Autistic children start far fewer peer contacts than kids with similar IQ, but brief priming or peer-PRT can fix it.
01Research in Context
What this study did
Davison et al. (1995) watched autistic children and children with mental retardation during free play. They counted how often each child walked up and started an interaction with an adult or another child.
The team also looked at the quality of each hello, question, or invitation. They wanted to see if autism itself, or thinking skills, predicted who started play.
What they found
Autistic kids started far fewer conversations with peers than the other group. Their initiations to adults looked about the same.
When autistic children did speak first, the openings were different—shorter, less flexible. The child's IQ, not autism severity, best guessed how often they approached a peer.
How this fits with other research
Martin et al. (2003) ran almost the same comparison and got the same sad score: autistic children gave fewer positive responses and shorter play bouts. The picture holds across years.
Kirkpatrick-Steger et al. (1996) flipped the script. They gave two autistic preschoolers a quick, fun "priming" warm-up right before class. Peer initiations doubled. The deficit is real, but it can move.
Harper et al. (2008) and De Korte et al. (2020) keep the ball rolling. Peers trained in PRT at recess, or a robot sidekick in PRT, both pushed autistic children to start more. The 1995 gap is now a 2020 target.
Why it matters
If you write a social goal, measure who starts the talk, not just who answers. Build cognitive supports—visual scripts, shared games—into peer time. Try a five-minute priming warm-up or train classmates to give quick PRT cues. Small setup, big jump in first words.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Open session with a 5-minute low-demand peer game before the main activity and track who speaks first.
02At a glance
03Original abstract
Social initiations made by autistic and verbal-matched retarded children were recorded in two naturalistic situations. Frequencies of initiation to adults did not differ between groups, but the retarded children initiated much more frequently to peers. Most interactions for both groups were positive, but the autistic children engaged in more ritualized, and the retarded children more playful, initiations. The autistic children monitored the social environment more when forced into proximity with peers, whereas the retarded children initiated more in the unstructured situation. Autistic initiation to peers was unrelated to severity of autism, but was related to cognitive skills, including vocabulary and comprehension of affect, whereas retarded children's initiations were unrelated to cognitive level. Results are discussed in terms of the differences between adults and children as social stimuli, prerequisite skills for initiation to peers, and the relationship between social cognition and social behavior. It is suggested that autistic and retarded children differ in the quantity of their initiations to peers, and the quality of their initiations to adults, and that initiations to peers may be a particularly useful index of social development in autistic children. Results confirm the need of autistic children for highly structured social environments, and suggest an important role for the remediation of specific cognitive skills such as comprehension of others' affects.
Journal of autism and developmental disorders, 1995 · doi:10.1007/BF02178189