Shaping procedures
Shaping builds a new behavior by reinforcing successive approximations: each response a little closer to the goal is reinforced, and earlier, less accurate versions stop being reinforced. It is used when the target behavior does not yet exist in a form that can be prompted or reinforced directly.
What is shaping, and what does this task cover?
Shaping is the differential reinforcement of responses that gradually come closer to a target behavior. Task C.10 on the RBT Test Content Outline (3rd ed.) asks technicians to implement shaping plans: reinforce the current approximation, stop reinforcing earlier ones once the plan raises the criterion, and record which step the learner is performing reliably.
Shaping combines two processes. Reinforcement strengthens the current approximation, and extinction of older approximations creates the natural variation from which a slightly better response can be caught and reinforced.
When do plans use shaping?
Shaping fits goals where the learner cannot yet produce the target, and prompting cannot easily create it. Vocal sounds are the classic example, since you cannot physically guide someone's speech. It also suits changing a dimension of an existing behavior: louder, longer, faster, more accurate. It suits multi-step routines less well; those usually call for chaining.
| Goal | Possible shaping steps |
|---|---|
| Saying ball | Any vocal sound near the ball, then ba, then bah-l, then ball |
| Speaking audibly in class | Whisper, then quiet voice, then voice audible across one desk, then across the room |
| Sitting for circle time | 2 minutes, then 3, then 5, then the full 10 |
| Writing a letter within lines | Any mark, then a shape near the letter, then the letter within wide lines, then standard lines |
How do you run a shaping program day to day?
Start each session at the step the data say the learner has mastered, reinforce responses that meet the current criterion right away, and withhold reinforcement for responses below it. When the learner meets the plan's rule for moving on, raise the criterion one step. If responding breaks down, follow the plan's rule for dropping back.
- Know the current step's exact criterion before the session starts
- Reinforce immediately, so the reinforcer follows the closer approximation and not whatever comes next
- Stop reinforcing the previous approximation once the step has changed
- Record which step was in effect and how often the learner met it
- Report steps where the learner stays stuck or responding falls apart
What happens if you move too fast or too slow?
Raising the criterion too quickly means the learner rarely earns reinforcement, so responding drops, frustration rises and the behavior can fall apart. Moving too slowly lets an approximation become so strongly reinforced that the learner gets stuck on it. The plan's movement rules exist to balance those risks, which is why the technician follows them rather than going by feel.
Clinic example: sitting was raised from 2 minutes straight to 6, and the learner began leaving the chair after a minute. Under the plan's backup rule, the technician returned to 3 minutes and reported the change.
How is shaping different from chaining?
Shaping changes one behavior over time, with each step replacing the last until only the final form is reinforced. Chaining links several different behaviors in a sequence, and every step stays part of the finished routine. If the scenario describes one response getting closer to a goal, think shaping. If it describes a series of steps, think chaining.
Which shaping mistakes and scope issues come up on the exam?
Watch for scenarios where a technician keeps reinforcing an old approximation after the criterion changed, which stalls progress, or reinforces late so the wrong response is strengthened. On scope, the BCBA defines the target, the steps and the rules for moving between them. The technician implements those rules and reports steps that need splitting or adjusting.
Worked scenario: a learner now says bah-l reliably, but the technician still hands over the ball for ba because the learner seems eager. Each time ba pays off, the closer form has less reason to appear. The fix is to reinforce only bah-l or better, as the current step requires, and expect a little frustration while the old form extinguishes.
Another scenario: a quiet student is reinforced for answers audible across one desk. If the technician also praises whispered answers to be kind, the plan's criterion has quietly dropped back a step, and the data no longer show what the student can do.
Practice questions
What is shaping?
Theo, 2, says 'ba' when he wants his ball, and the technician gives it to him. Once 'ba' is reliable, she gives the ball only for 'bah-l,' and later only for 'ball.' Which procedure is this?
During shaping, Theo now says 'bah-l' reliably, but his technician still gives him the ball when he says only 'ba.' What is the problem?
Rina, 10, speaks so softly that her classmates can't hear her. Her technician reinforces answers that are slightly louder than her usual volume, then raises the requirement step by step. Which procedure is this?
A technician is shaping longer sitting for Aiden, 4. After the requirement was raised from 2 minutes to 6 minutes in one session, Aiden stopped sitting at all. What is the best next step?
Questions people ask
Does shaping use prompts?
Shaping relies mainly on differential reinforcement, though some plans combine it with prompts where a prompt can help.
Is shaping a form of differential reinforcement?
Yes. It differentially reinforces closer approximations and withholds reinforcement for earlier ones.
Can shaping reduce a behavior?
Shaping builds or refines behavior. Reducing behavior is handled through other procedures, such as differential reinforcement of other or lower rates of behavior.
How ready are you?
Ten questions, about 3 minutes, no account. Get your readiness score and your weakest domains.
Check my readiness