Setting generality and stimulus control in autistic children.

Four autistic children who had learned a new skill in one room would not do it in another room. The team suspected an accidental cue in the first room was really running the show.

They ran an ABAB reversal. First they removed the mystery cue from the teaching room. The kids stopped responding. Next they moved that same cue into the new room. Correct answers came right back.

The skill had never been under the teacher’s voice or picture cards. It was under an unnoticed lamp, shelf, or smell. Drop that tiny cue into any space and the child performs.

No extra teaching trials were needed once the real cue traveled with the child.

White et al. (1990) extends this idea. They used peer-model games to build generalization up front, so extra cues were not needed later. Rincover et al. (1975) shows the back-up plan when generalization fails.

Hawkes et al. (1974) used the same logic in reverse. They slowly moved classroom stimuli into 1:1 teaching, then faded the child into the group. Both studies move cues, but 1974 planned the trip while 1975 fixed the crash.

Levin et al. (2014) conceptually replicates the fix with blocked-trial prompting. They also saw big skill jumps in autistic preschoolers, proving prompt-based rescue still works four decades later.

Before you reteach a "lost" skill, hunt for the hidden cue. Walk the old room, note lights, textures, even air fresheners. Bring one item to the new spot and test. If the child suddenly gets it, you just saved hours of retraining. Make the cue part of the plan, then fade it once the skill is strong.