Positive practice overcorrection combined with additional procedures to teach signed words to an autistic child.
Positive-practice overcorrection with reinforcement can teach manual signs to non-verbal autistic children—add brief contingent exercise when progress stalls.
01Research in Context
What this study did
The team worked with one autistic child who had no spoken words.
They used positive-practice overcorrection plus treats for praise.
When the first sign stalled, they added short exercise after errors.
A changing-criterion design showed step-by-step progress.
What they found
The child learned the first sign with practice and praise alone.
The second sign needed the extra exercise to move forward.
Both signs later showed up in a new matching game without teaching.
How this fits with other research
Huguenin et al. (1980) proved the same practice-plus-reinforcement pair gives perfect spelling scores in typical kids.
Wilkinson et al. (1998) and Barall et al. (2026) later used simpler prompting and praise to teach gestures and pointing to many autistic children, showing the package scales.
Périkel et al. (1974) did it first: prompts plus treats built real sentences in one autistic child, foreshadowing this sign method.
Why it matters
If a non-verbal client is stuck on a first sign, run positive-practice with strong praise.
If progress halts, add a brief contingent exercise—five chair stands or wall pushes—right after each error.
Drop the exercise once the sign is solid and test generalization with a matching game or new listener.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick one target sign, run five positive-practice trials after each error, deliver praise for correct responses, and add three quick star jumps if the data flat-line for two days.
02At a glance
03Original abstract
The present study evaluated the effectiveness of using positive practice overcorrection in combination with other techniques to teach two manual signs ("milk" and "cookie") to an autistic boy. This boy had a great deal of difficulty in forming any type of discrimination and often became confused in learning the most simple simultaneous discrimination. Intervention primarily consisted of positive practice overcorrection in which the subject was physically guided to form a required hand sign 10 times when he responded incorrectly and was positively reinforced when he signed correctly. The study used a changing criterion within a multiple-baseline design across responses. The results indicated that overcorrection plus positive reinforcement was effective in teaching one sign (milk), however, and added contingent exercise (having to stand up and sit down 10 times for an incorrect response) was required to teach the second sign (cookie). Once the two signs were learned to a criterion level, it was a relatively easy task for the subject to respond correctly with the signs in a matching-to-sample task.
Journal of autism and developmental disorders, 1982 · doi:10.1007/BF01531371