Chaining vs Shaping: Two RBT Teaching Procedures
Chaining teaches a multi-step behavior by linking individual steps of an existing task analysis together in sequence, like teaching handwashing step by step until the whole routine runs independently. Shaping builds an entirely new behavior gradually, by reinforcing successive approximations that get closer and closer to a target behavior that doesn’t exist yet at all. Both are Behavior Acquisition procedures, and both get confused with each other because they sound similar, teaching something in small steps, even though they solve different problems.
The clearest way to keep them separate: chaining assumes the individual steps of a task already exist as separate skills and links them together; shaping builds a skill that doesn’t exist yet, one small improvement at a time.
(RBT HQ is not affiliated with the BACB. This page explains standard applied behavior analysis procedures in plain language; it does not reproduce any BACB exam question.)
How does chaining actually work?
It starts with a task analysis, breaking a multi-step task into its individual, observable steps. Then a chaining method links those steps together: forward chaining teaches the first step first and adds each subsequent step once the previous one is mastered; backward chaining teaches the last step first, which lets the learner experience the natural reinforcement of completing the whole task from the very first attempt; total task chaining teaches all the steps together in every session, with prompts fading on whichever steps need them.
| Chaining method | How it works |
|---|---|
| Forward chaining | Teach step 1 first, add each step in order as prior steps are mastered |
| Backward chaining | Teach the last step first, so the learner reaches task completion immediately |
| Total task chaining | Teach all steps together each session, prompting as needed on each |
How does shaping work differently?
Shaping doesn’t rely on a task analysis of existing steps; it reinforces successive approximations, responses that are gradually closer to a target behavior that hasn’t happened at all yet. If a learner has never made a specific sound needed for a word, shaping might start by reinforcing any vocalization at all, then only vocalizations closer to the target sound, gradually narrowing what earns reinforcement as the learner’s responses get closer to the actual goal.
When would a supervisor choose chaining over shaping?
When the target is a multi-step task made up of skills the learner can already perform individually, just not yet linked into a full sequence. Getting dressed, brushing teeth, and making a simple snack are typical chaining targets, since each step (picking up a shirt, putting an arm through a sleeve) is already within the learner’s ability; the teaching challenge is sequencing and independence across the whole chain, not building any single step from scratch.
When would shaping be the better fit instead?
When no version of the target behavior currently exists at all, so there’s nothing to link together yet. Building a first approximation of a new sound, a new motor movement, or a new form of communication where the learner currently has no starting point close to the goal is a shaping problem, not a chaining one, since chaining has nothing to work with until some form of the behavior exists to reinforce and refine.
Can chaining and shaping be used together?
Yes, and this happens more often than the clean textbook definitions suggest. A single step within a chained task might itself need to be shaped first, if the learner can’t yet perform that specific step at all. Once that step exists at a basic level through shaping, it gets folded back into the larger chain. An RBT implementing a plan follows whichever combination the supervisor has specified for that particular skill.
How does prompting fit into chaining and shaping?
The same prompting hierarchy principles apply to both. In chaining, prompts help the learner complete a specific step correctly, fading as that step becomes independent. In shaping, prompts (including modeling or physical guidance) can help produce an initial approximation worth reinforcing, gradually withdrawn as closer approximations emerge on their own.
Run the free RBT HQ practice exam to practice distinguishing chaining from shaping in scenario form, since exam questions typically describe a teaching situation and ask which procedure is being used. The free RBT HQ Exam Prep Cheat Sheet keeps Behavior Acquisition terms like these grouped together for faster domain-by-domain review.
FAQ
What’s the simplest way to tell chaining and shaping apart? Chaining links existing individual steps into a sequence; shaping gradually builds a behavior that doesn’t exist yet through successive approximations. One assumes the pieces exist already, the other doesn’t.
Which chaining method is fastest? None is universally fastest; the right method depends on the learner and the task. Backward chaining is often chosen because it gives the learner immediate task-completion reinforcement from the first session.
Is shaping only used for speech and communication skills? No. Shaping applies to any behavior that doesn’t yet exist in any usable form, motor skills, vocalizations, or other target behaviors, not communication specifically.
Can an RBT choose which chaining method to use? No. The specific chaining method (forward, backward, or total task) is set in the client’s plan by the supervising BCBA or BCaBA. An RBT implements the method as written.
Does chaining require a written task analysis? Typically, yes. A clear breakdown of the individual steps is what chaining is built from; without that breakdown, there’s nothing to link together in sequence.
Bottom line: chaining links existing individual steps into a full sequence; shaping builds a behavior gradually from nothing through successive approximations. Both are core Behavior Acquisition procedures, and knowing which situation calls for which is what the exam actually tests.