Shaping is the incremental building of a skill by adding more parts or more difficulty one step at a time. As a learner becomes better at the skill, the teacher gradually increases the requirement to get a tag.

At first when you start to shape a new behaviour, you will tag just attempts that are fairly close to the goal. You will tag every attempt.

Once the learner gets a bit better, you will tag only those responses that are generally closer to the goal than the first attempts were. Stick to the 3-try rule. So that if you don’t tag three attempts in a row, lower your criteria back to a point of success and then work up again.

You may have heard that it is beneficial to move to a intermittent schedule of reinforcement. This means that you don’t reinforce every time, rather you only reinforce sometimes. An old-fashioned way to do this was to reinforce every 3rd or 4th response and gradually make reinforcement more and more intermittent. The trouble with this is that this type of reinforcement schedule is not based on how good the performance was. What if the 3rd or 4th try that you had decided to reinforce was a sub-par response? Then you would have reinforced something that is sub-par and this will reduce the effectiveness of your teaching and could also confuse the learner.

With TAGteach there is no intermittent reinforcement in the teaching phase. The tag point is reinforced every time it is accurately performed and the tag point should be set so that the learner is getting it right every time (or almost every time).

Intermittent reinforcement may come into play during the maintenance phase. That is, when the behavior has been learned in a teaching drill situation and now is part of a real-life situation. Examples could be social skills during daily family life or soccer skills during a scrimmage at practice. Sometimes you may want to tag if you see a really exceptional example of a skill or behavior, but you’re not tagging specifically as part of a formal teaching session.

The best way to think about intermittent reinforcement is to think about gold star performances. At any level of learning, there will be some responses that are better than most of the the others. If your intermittent reinforcement is based on those gold star performances, then you will be reinforcing only the best responses and you will be reinforcing intermittently, since these only come along every so often.