Examining correlates of cooperation in autism: Imitation, joint attention, and understanding intentions.

Colombi et al. (2009) watched preschoolers with autism play short games with an adult. They scored how well each child copied actions, shared eye contact around toys, and read the adult’s intentions. The team compared these scores to a group of children with developmental delays matched on language level.

Kids with autism showed weaker imitation and joint attention than the delay group. When the researchers ran the numbers, only imitation and joint attention predicted how cooperatively the child played. Surprisingly, understanding the adult’s intentions added no extra power.

Warreyn et al. (2005) saw the same joint-attention gap in mother-child play, but they found imitation looked typical—an apparent contradiction explained by different play tasks and slightly younger kids.

Iao et al. (2024) later tracked Taiwanese toddlers for 18 months and showed that early imitation and responding to joint attention forecast later language gains, backing the idea that these two skills are developmental engines.

Chen et al. (2001) went a step further: they taught typical peers to run play groups and autistic preschoolers quickly gained joint attention, pretend play, and new words. The correlational deficit Costanza found became a teachable target.

If a child is slow to cooperate, probe imitation and joint attention first. Quick checks—like having the child copy a novel action or follow your point to a far toy—tell you whether these cores are in place. When they’re weak, use peer play or adult modeling to build them; intention-reading drills can wait. Strengthening these two levers often lifts cooperation, play, and later language all at once.