Teaching children with autism to answer novel wh-questions by utilizing a multiple exemplar strategy.
Train wh-questions with lots of shared-frame examples and demand full sentences to get kids with autism to answer new questions anywhere.
01Research in Context
What this study did
The team taught five children with autism to answer new wh-questions. They used a multiple exemplar strategy. Kids practiced many questions that shared the same sentence frame.
Each child had to give full-sentence answers. The study ran a multiple baseline across question types. Training moved from simple to harder questions.
What they found
All five children learned to answer novel wh-questions in full sentences. The skill spread to new people, rooms, and days later. No extra teaching was needed for this transfer.
How this fits with other research
Smith et al. (2016) and LaLonde et al. (2020) got the same outcome with a bingo game instead of dry drills. Their game plus praise and play modeling also created new wh-answers that lasted.
Thakore et al. (2022) pushed further. They added fluency timings to wh-question training. Kids answered faster and gave more varied responses, building on the core generalization shown here.
GBrodhead et al. (2019) used FFC prompts and saw limited transfer to truly new questions. The difference warns us: multiple exemplars or games beat simple prompting when we want broad generalization.
Why it matters
If you want flexible wh-question answering, drill-and-kill is not enough. Rotate many examples that share the same frame, or turn the task into a fun game with praise. Either way, require full sentences from the start. You will likely see the child use the skill with new teachers, in new rooms, and weeks later without retraining.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick one wh-frame (e.g., 'Where do you ___?'), write ten items, run trials, and accept only full-sentence answers.
02At a glance
03Original abstract
This study investigated the transfer and maintenance of question-answering skills in five children diagnosed with autism. A multiple baseline design across classes of questions (i.e., what, where, who and why) was applied for each child. Question-answer exemplars were selected within each class, and each class was trained separately in consecutive order. The dependent variable was the proportion of appropriate answers (i.e., complete sentences) to novel questions within each class, on first trial. The results showed that all children became able to answer novel questions with complete sentences within each of the classes that were trained, and they showed transfer of these skills across persons, settings and time. The findings support the use of analogous question-answer exemplars in order to facilitate response-transfer to novel questions. It is also suggested that this type of transfer is more likely to occur if the answers trained are in full sentence and there is a structural correspondence between the question and the answer in each single exemplar and across exemplars within a class of questions.
Research in developmental disabilities, 2001 · doi:10.1016/s0891-4222(01)00081-6