An analysis of individual differences in generalization between receptive and productive language in retarded children.

The team worked with four children who had intellectual disability. They taught plural rules in two ways at once: receptive (point to the picture of "cats") and productive (say "cats" when shown the picture).

The goal was to see if learning the rule in one form would automatically show up in the other form without extra teaching.

Three kids learned the receptive task but could not say the plurals on their own. Only one child used the rule both ways after the paired lessons.

In short, most kids needed separate practice for listening and speaking even when the rule was the same.

Zeiler (1969) saw the same block: kids mastered receptive plurals yet produced zero correct expressive forms until they got direct expressive drills. Wilson et al. (1973) adds a twist by running both tracks together and still finding the gap in three of four cases.

Clayborne et al. (2024) looks like a contradiction at first. Their preschoolers with autism formed new equivalence classes and generalized listener responses after only brief teaching. The difference is population and content: the 2024 study used category names, not grammar rules, and the children were younger with ASD, not ID.

Cameron et al. (1996) also used equivalence training, but with reading and spelling. They got broad generalization when they added a hands-on word-building step, showing that extra production practice can unlock the transfer that D et al. found missing.

For BCBAs, the message is clear: do not trust that a child who follows the receptive instruction "Touch the dogs" can also say "dogs" when asked "What are these?" Probe both modalities every session. If the expressive side is weak, add direct productive trials right away instead of waiting for generalization that may never come.