Washing hands and getting dressed are not single behaviors. Each is a series of smaller steps in a set order. This lesson covers how those steps are written down in a task analysis and three ways to teach them as a chain.
Why it matters
Many meaningful goals in a learner's plan are self-care routines, and doing them independently means more privacy and choice. Your supervisor writes the task analysis and picks the chaining method; you run it session after session, prompt the right steps, and record what the learner did on each step.
Practice items on this task are usually short scenarios: an RBT teaches some steps and not others, and you name the procedure or pick which step comes next. Know which step is taught first in each method and how the chain grows.
Key ideas
- A task analysis comes first. It breaks a complex skill into small, observable steps, listed in the order they happen. The supervisor builds it (often by watching a skilled person do the task) and sizes the steps to fit the learner.
- The learner's starting point is assessed. Before teaching, you may be asked to check which steps the learner can already do. In the single-opportunity method, the assessment stops at the first error and the remaining steps are scored incorrect. In the multiple-opportunity method, you complete a missed step for the learner, score it incorrect, and keep assessing the rest.
- Each step links to the next. In a behavior chain, finishing one step changes the environment (the water is on, the soap is on the hands). That change acts as a conditioned reinforcer for the step just finished and signals the next one. The last step produces the terminal reinforcer, such as clean hands or a finished snack.
- Three ways to teach the chain:
- Forward chaining teaches the first step first, with reinforcement right after that step. Once it meets the mastery criterion, the learner must do step 1 and step 2, then steps 1 to 3, and so on.
- Backward chaining starts at the end. The earlier steps are completed for or with the learner, and the learner does the last step on their own. Then the last two, then the last three, back to the start.
- Total-task chaining teaches every step on every trial, in order, with prompts on any step the learner cannot yet do alone.
- Mastered steps stay in the chain. In forward and backward chaining, the learner keeps doing earlier mastered steps independently as each new step is added.
- Data is step by step. Score each step as independent or prompted, then summarize the percentage of steps done independently.
Definitions
- Task analysis: a written list of the individual, teachable steps that make up a complex skill, in the order they are performed.
- Behavior chain: a sequence of responses in which each response produces a stimulus change that reinforces it and serves as the discriminative stimulus (SD) for the next response.
- Forward chaining: teaching the steps in their natural order, starting with the first, and adding the next step after each one is mastered.
- Backward chaining: teaching the last step first, then adding steps in reverse order, so the learner finishes the chain and contacts the terminal reinforcer on every trial.
- Total-task chaining: teaching all steps on each trial, prompting as needed, until the whole chain is independent.
- Terminal reinforcer: the reinforcer that follows the final step of the chain.
Examples from a session
Three learners, three methods
- Forward chaining: Mateo, 24, is learning to make a smoothie at his day program. He has mastered "get the blender out," so now he must do that step and "add the fruit" before he earns praise and a break. Later steps are completed with help.
- Backward chaining: Zara, 5, is learning to put on her socks at home. Her RBT guides her through every step until the sock is over her heel, and Zara pulls it up the rest of the way herself. She gets a high-five and gets to go play right away.
- Total-task chaining: Jordan, 14, is learning to use the school vending machine. He already does most steps, so on every trial he runs through all seven. His RBT prompts only "press the item number," the one step he still misses.
Non-example: an RBT reinforces "ba," then "bah-l," then only "ball." That is shaping, not chaining: one response moves closer to a target, rather than several different steps being linked in order.
In each case, the RBT follows the plan's order, prompts, and mastery criterion. If the learner keeps getting stuck on the same step, record what you see and tell your supervisor rather than changing the method yourself.
Common mistakes / exam traps
- Confusing chaining with shaping. Chaining links several different responses into a sequence. Shaping reinforces closer versions of one behavior (see the non-example above).
- Reinforcing at the wrong moment. Reinforcement follows the step the learner is working on. In backward chaining, that means right after the final step, not during the earlier steps you helped with, which would reinforce being helped instead of finishing.
- Dropping mastered steps. Once a step is learned, the learner keeps doing it independently while the next step is taught.
Exam trap
Forward vs backward chaining. Both add one step at a time; they start at opposite ends. Ask which step the learner does on their own first. Step 1 means forward. The final step means backward, and the learner contacts the terminal reinforcer on every trial.
Exam trap
Total-task vs backward chaining. The RBT prompts some steps in both, so "the RBT helps" does not decide it. Ask which steps are taught each trial. In total-task chaining the learner works on all of them, with prompts wherever needed. In backward chaining only the steps at the end are taught; earlier steps are completed for or with the learner.
Quick check
Using a 6-step task analysis for brushing teeth, an RBT helps Liam with steps 1 to 5. Liam rinses the brush and puts it in the cup (step 6) on his own and then earns his favorite song. Which chaining procedure is this?