Shape complex behaviors through successive approximations

Reinforce progressively closer approximations to the target behavior rather than waiting for the full behavior to appear.

Why it works

Shaping works because complex behaviors cannot be reinforced until they occur, and they rarely occur spontaneously. By reinforcing steps toward the target — each one a closer approximation — the person (or animal) is guided through a behavior that would never emerge through waiting. This is the principle behind all graded task assignment in therapy and progressive overload in training: you reward the direction, not the destination.

How to do it

1. Define the terminal behavior precisely (what would "fully there" look like?).
2. Work backward to identify the step just below the current level of performance.
3. Reinforce that step consistently until it is fluent, then raise the criterion to the next approximation.
4. Move the criterion up only when the current step is stable — never when it is still being acquired.

Evidence

Shaping is a foundational operant technique with extensive experimental support in animal learning, applied behavior analysis, and behavioral therapy (graded task assignment). (clinical)

Setting the criterion steps requires judgment; moving too fast stalls progress, moving too slowly creates fixation at a sub-target level.

Sources

• Cooper, Heron & Heward, "Applied Behavior Analysis" — standard reference for shaping procedures

Common mistake

Setting the initial criterion too high so reinforcement never occurs, which extinguishes the behavior before it can be shaped.