TAGteach – Using Clicker Training with Humans, continued
In my book, I had added:
If you take the time to teach a “generalized” secondary reinforcer, e.g., a sound such as a “snap,” “tap,” or “click,” or a light such as can be provided by a flashlight, you will find that it can be used very unobtrusively and effectively to “mark out” for the child those behaviours that you would like him or her to learn, i.e., those behaviours which may lead to reinforcement.
This is referred to as “charging the clicker.” I have since been told by a well-seasoned TAGteacher that she no longer bothers with “charging the clicker,” presumably since the TAG seems to function more as an indicator of success – sort of like biofeedback – than as reinforcement per sē.
When there is a complex behaviour that we want a student to learn, good ABA practice demands that we start small and work our way up. That is, we choose a goal to be achieved, break it down into easily learned steps and/or components, and shape it into the response that we want to get, one small part at a time. Here are a few examples:
- In teaching waiting quietly, we might begin by rewarding the learner for a single second of quiet waiting. Since, as Karen Pryor has pointed out, there is always a certain amount of variability in any behavior, on occasion the learner will wait quietly for two or even three seconds, and that is what you will start to reward. Again, once the learner has learned to wait quietly for 2-3 seconds, he will sometimes wait quietly for an even longer time, and you will reward that longer period of waiting, and so on. Gradually stop reinforcing less satisfactory responses in favour of those close to your end goal, being careful to ensure that the amount of reinforcement never drops so low that the learner stops responding.
- In teaching a golfer how to improve his putting skill, we might train for distance and direction separately, and then train the golfer how to do both of them at the same time.
- In teaching a skill that has a number of sequential components, you might think of “backward chaining”: teach the last step in the process, then the second last step, and so on until all the steps have been mastered.