Generalization and transfer between comprehension and production: a comparison of retarded and nonretarded persons.
Teaching receptive labels first cuts training time by two-thirds when later teaching expressive labels.
01Research in Context
What this study did
Researchers worked with the adults. Six had intellectual disability. Six were neurotypical.
They taught everyone the names of 12 coins. Half learned comprehension first, then production. Half learned production only.
The team counted how many trials each person needed to master all coin names.
What they found
Both groups learned all coin names. The comprehension-first group needed only one-third as many trials.
After training, everyone could both point to coins when named AND say coin names when shown.
Skills lasted for weeks. IQ level did not change the result.
How this fits with other research
Cordova et al. (1993) later showed most people with moderate ID can pass identity-matching tests. This supports the idea that stimulus-equivalence training works well across ID levels.
Davison et al. (2002) found neurotypical adults needed extensive training to control equivalence-based transformations. This seems to contradict our finding that training was quick. The difference is focus: Michael studied complex contextual control, while we taught simple coin names.
Laugeson et al. (2014) extended our approach to adults with dementia. They used similar transfer procedures to teach picture-based requests. Their mixed results remind us that individual differences matter, even when methods work for most.
Why it matters
You can save hours of teaching time by starting with receptive labels. Try teaching a child to point to 'cup' before asking them to say 'cup.' This works for both neurotypical learners and those with ID. Track trials to see the difference yourself.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick one target item. Teach the learner to point to it when named. Only after mastery, teach them to name it aloud.
02At a glance
03Original abstract
The study compared the acquisition, generalization, transfer, and maintenance of language comprehension and production responses by persons at two IQ levels: mentally retarded (N=10) and nonretarded (N=10). The two levels of the IQ Level factor were combined factorially with two levels of a Training Condition factor: Comprehension-Production and Production Only. Participants in the former groups were trained sequentially to (a) comprehend coin labels by pointing and then (b) produce verbally the correct coin label. Participants in the Production Only groups were trained on the latter response only. A three-factor mixed design with one repeated measure plus a multiple baseline across coin responses was employed. Results indicated that both mentally retarded and nonretarded subjects attained a high level of acquisition and maintained their performance on 1- and 4-week follow-up tests. No difference occurred between mentally retarded and nonretarded participants in magnitude of acquisition, but the mentally retarded groups took approximately three times as many trials to complete training. Data also suggested, contrary to past research, that generalization from comprehension to production was bidirectional, with no difference in magnitude between mentally retarded and nonretarded subjects. Transfer from comprehension to production occurred in both nonretarded and retarded subjects; comprehension training facilitated a savings of trials in production training. These results show that language differences between retarded and nonretarded persons are quantitative rather than qualitative as some past research may have suggested.
Journal of applied behavior analysis, 1980 · doi:10.1901/jaba.1980.13-315