Analog language teaching versus natural language teaching: generalization and retention of language learning for adults with autism and mental retardation.

Elliott et al. (1991) compared two ways to teach new words to adults with autism and severe intellectual disability.

One group learned in a quiet room with flash cards and repeated trials. The other group learned while doing everyday tasks like setting a table.

Both groups had the same number of teaching minutes. The team later checked if the adults still used the words in new places and with new people.

Both teaching styles worked equally well. Adults said the new words during lessons and still used them weeks later.

No method beat the other. Even when the test room looked like the flash-card room, the natural-teaching group kept up.

Bondy et al. (1976) first showed that children with autism can learn language through tight prompting and fading. O et al. moved that question forward by asking, "Does the setting matter for adults?" Their answer: once the skill is in place, the setting does not matter.

Petursdottir et al. (2009) later found that no single verbal-operant drill gives full vocabulary mastery with typical kids. That lines up with O’s null result: one format is not enough; mixing or matching may be fine.

Newman et al. (2021) showed that reinforcer choice changes outcome even within the same ABA procedure. O’s study hints at a similar idea: the teaching procedure matters less than simply providing good teaching.

You can stop worrying that table-top drills are the only path to lasting language. If your adult client learns better while packing groceries or folding laundry, go for it. Use the same data sheets, the same prompts, just move them into real tasks. You will get the same generalization and save sterile table time for other goals.