Programming the generalization of a greeting response in four retarded children.

Four children with intellectual disability lived in a state hospital. None waved hello.

Two experimenters taught each child to raise a hand when an adult said "Hi." They used praise and small treats. The study ran a multiple-baseline across kids.

When two trainers took turns, all four kids waved to 20 new staff within two weeks. The skill stayed for three months with no extra work.

When only one trainer taught, the greeting never spread to new adults.

Conine et al. (2025) later showed the same idea: teach more than one adult and the child learns faster. They swapped experimenters for caregivers and got the same boost.

Turk et al. (2010) added an iPhone cue instead of a second person. Adults with autism kept their mascot job only after the phone joined the training. One trainer plus tech can equal two trainers.

Jenkins et al. (2016) looked at staff practice, not kids. They found one rehearsal with feedback was enough for new staff to run an assessment. Lowe et al. (1974) shows the same rule: small changes in who delivers training make big gains in real-world use.

You do not need fancy generalization plans. Just rotate a second therapist, parent, or peer tutor into greeting lessons. The child meets new faces and the skill sticks. Try it next session: swap staff halfway through and probe with a novel adult at the end.