One of the greatest challenges of supporting autistic learners lies at the intersection of two facts about most autistic brains:
1) Predictability helps keep the autistic brain in a regulated and ready-to-learn state.
2) Flexible thinking is difficult for the autistic brain.
Often, instructional approaches cater primarily to Fact #1, using repetition and routine to establish an optimal context for learning.
But research has shown that too much repetition actually makes it harder for autistic brains to learn, because then you run straight into Fact #2. Too much sameness in a teaching approach makes it harder to take what's learned and use it in a different context.
When we drill requesting again and again before moving on to other reasons to use language (e.g. commenting, describing, asking questions, etc.), are we accidentally teaching autistic learners that requesting is the only thing that language is good for?
Are we inadvertently establishing the rule seen below?
language = getting stuff
Once a rule is learned, it's hard to unlearn it if you have trouble with flexible thinking. Realizing that language can serve a huge variety of purposes can be especially hard because an inflexible mind might not easily make room for more than one answer on the right side of that equation.
So what do we do about this?
The research summary linked above proposes a solution to the dilemma of teaching vocabulary with generalization in mind:
"For example, in the context of learning what a dog is, using a full range of examples of dogs — and even of animals, more generally — incorporates variability from the beginning and promotes learning a broad concept rather than a specific example."
The learning-language-beyond-requesting equivalent? Exposing the learner to a variety of examples and opportunities to practice the full range of communicative purposes:
