Acquisition of sign language by autistic children. I: Expressive labelling.
Prompt, fade, and rotate toys and every nonverbal autistic child in the study learned to sign object names.
01Research in Context
What this study did
Four nonverbal kids with autism got their first sign language words. The team used three moves: hand-over-hand prompt, fade the help, and swap the toys fast so no item got stale.
A multiple baseline across objects proved the teaching caused the gains. New therapists later tested if the signs stuck with strangers.
What they found
Every child signed the correct item name after a handful of trials. The signs held up with new adults and new table toys.
How this fits with other research
Wearden (1983) and Hanley et al. (1997) flipped the direction: they used the child’s own echo as an auditory prompt to teach receptive labels. Same prompting DNA, but ears instead of hands.
Delamater et al. (1986) kept the 1978 design but aimed later in the journey—replacing echoed sentences with real answers. Together the trio shows prompting-fading works for first words, then for smarter language.
Fuqua et al. (2025) leap to 2025: an iPad self-monitoring app boosted initiations for non-speaking students. The old sign frame still matters, yet tech can now carry the load.
Why it matters
If you have a silent client, start here: prompt the sign, fade your hands, rotate the stimuli. Once signs are echoing back, borrow H’s trick and let the child say the word before giving the item. Later, move to Amelia’s self-monitoring if you want initiations without adult hovering.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick two objects, give a full physical prompt for the sign, fade to a touch, then swap the objects every few trials.
02At a glance
03Original abstract
There has been growing interest in teaching sign language to autistic children who have failed to develop speech. However, controlled experimentation in this area is nonexistent. In the present study, four nonverbal autistic children were taught expressive sign labels for common objects, using a training procedure that consisted of prompting, fading, and stimulus rotation. The efficacy of the procedure was demonstrated in a multiple-baseline design across objects. The results were reliable, replicable across children, and generalizable across therapists. A stimulus control analysis demonstrated that, for three of the children, correct signing was controlled solely by the visual cues associated with the presentation of a given object and was independent of respect to the known perceptual and linguistic deficits of autistic children.
Journal of applied behavior analysis, 1978 · doi:10.1901/jaba.1978.11-489