Improving generalization of peer socialization gains in inclusive school settings using initiations training.

Three elementary students with autism played alone at recess. The researchers taught each child how to start games with classmates. They used a multiple-baseline design across kids. Sessions happened on the playground with no adult help after training.

All three children began inviting peers to play. Their unprompted initiations doubled. They stayed in games longer and smiled more. Gains held four weeks later, even when staff stepped back.

Jason et al. (1985) showed similar peer-start training works for blind, multihandicapped preschoolers. Lowe et al. (1995) got big gains using peer-delivered PRT instead of plain initiation drills.

Gotham et al. (2015) looks like a contradiction — they used scripted phrases while Syriopoulou-Delli et al. (2012) used natural play. Both worked because scripts are just a stepping stone; the 2012 study proves you can skip scripts and still get spontaneous talk.

Leung et al. (2014) updated the idea by adding teacher briefings. Their preschoolers only generalized after teachers knew the goals. Syriopoulou-Delli et al. (2012) kept staff blind and still saw carry-over, showing older kids may need less adult support.

If you run recess groups, spend a few sessions teaching kids with ASD to invite peers. Skip the scripts. Use simple cues like “Ask to join.” Fade yourself out. The play should keep going without you, saving staff time and building real friendships.