Knowing positive from negative reinforcement is the concept; using it to change behavior is the procedure. This lesson covers how you set up, deliver, and adjust reinforcement so it strengthens the target behavior.
Why it matters
Reinforcement is the core of nearly every skill-building and behavior-reduction plan: differential reinforcement, token systems, shaping, chaining, and functional communication training all depend on it. When a plan "isn't working", the delivery is often the problem: it comes too late, it is not really contingent, the learner is satiated, or the "reinforcer" was never a reinforcer at all.
Task G.1 sits in Behavior-Change Procedures (domain G, 25 scored questions). Practice items for it are mostly vignettes: a reinforcement plan is failing, and you name the most likely cause or the best next step.
Key ideas
- A procedure is named for what you do; it is confirmed by what happens. In a positive reinforcement procedure you deliver a stimulus contingent on the target response. In a negative reinforcement procedure you remove, reduce, or postpone a stimulus contingent on the response. Either way, you only know reinforcement occurred if the behavior increases.
- Start with evidence about reinforcers. Ask the learner and caregivers, observe what the person does when free to choose, and run preference assessments. A preferred item is a good bet, but only a reinforcer assessment, or the data from the program itself, shows it actually strengthens behavior.
- Keep the contingency tight. The reinforcer should follow the target response and, as much as possible, be unavailable otherwise. Free access to the same item outside the program weakens both the contingency and the learner's motivation.
- Deliver immediately. During acquisition, reinforce right after the response. A delay lets other behavior occur in the gap and get reinforced instead.
- Manage motivation. A motivating operation changes how effective a reinforcer is at a given moment. Deprivation increases effectiveness; satiation decreases it. Vary reinforcers, keep amounts small enough to avoid satiation, and pay attention to the time of day.
- Make magnitude and quality fit the effort. Harder or less preferred tasks may need a higher-quality or larger reinforcer.
- Begin dense, then thin. Reinforce every correct response (continuous reinforcement) while a skill is new. Once responding is reliable, gradually move to intermittent reinforcement and shift toward reinforcers that occur naturally in the person's environment.
- Use negative reinforcement thoughtfully. It requires a stimulus the learner will work to escape or avoid, such as task demands. Use conditions that already exist in the person's day, like a break from work, rather than adding aversive stimulation just so you can remove it.
Definitions
- Positive reinforcement procedure: presenting a stimulus contingent on a response, which increases the future occurrence of that response.
- Negative reinforcement procedure: removing, reducing, or postponing a stimulus contingent on a response, which increases the future occurrence of that response.
- Reinforcer assessment: a test of whether a stimulus, delivered contingent on a response, actually increases that response.
- Satiation: a temporary drop in a reinforcer's effectiveness after recent or repeated access to it.
- Schedule thinning: gradually increasing the response or time requirement for reinforcement so that behavior holds up under leaner, more natural schedules.
Examples from a session
Reinforcement procedures across settings
- Clinic, positive reinforcement: Each time Mateo, age 4, says "car" to request a toy car, his RBT hands him the car within a second. Vocal requests increase across sessions.
- Classroom, behavior-specific praise plus tokens: Mr. Okafor tells Jasmine, 10, "You started your reading right away, nice!" and gives her a token she later exchanges for computer time. Prompt starts increase.
- Home, negative reinforcement: After Ethan, 8, completes three spelling words, his parent lets him set the worksheet aside for a two-minute rest. Completing words increases.
- Functional communication: Assessment shows that Noor's aggression at her day program is maintained by escape from cleaning tasks. Her BCBA teaches her to hand staff a "break" card, which immediately earns a short break, while aggression no longer ends the task. Card use rises and aggression falls.
- Supervision: A BCBA adds a short, specific note of thanks in the shared drive each time an RBT uploads graphs by Friday. On-time uploads increase.
When a plan stalls, troubleshoot in order: Is the stimulus a reinforcer for this person right now? Is it delivered immediately? Is it contingent, or available for free? Is the learner satiated? Is the response requirement too high?
Common mistakes / exam traps
- Assuming that preferred means reinforcing. A favorite item can still fail to increase a specific response. The data decide.
- Letting the reinforcer leak. If a caregiver gives the tablet freely between sessions, it will not reliably strengthen the target behavior.
- Confusing time-based delivery with a reinforcement contingency. Items given on a timer or "for rapport" do not depend on the target response, so they cannot strengthen it. That is noncontingent reinforcement, a different procedure with a different purpose.
- Making the amount bigger when the problem is satiation. More of the same item usually worsens satiation. Switch reinforcers or timing.
- Thinning too fast, or never thinning. Jumping from continuous to very lean schedules can cause responding to break down, while never thinning leaves the behavior dependent on contrived reinforcers.
Exam trap
Positive vs. negative reinforcement: the label depends on whether a stimulus is added or removed after the response, not on whether the outcome feels pleasant. A break earned by finishing work, or by handing over a "break" card, removes a task demand, so it is negative reinforcement for that response even though nothing was handed over.
Exam trap
"Stopped working" vs. "never worked": if the behavior rose and then fell, look first for satiation or recent free access, a new delay, or a broken contingency before changing the whole procedure. If it never rose despite consistent, contingent delivery, the stimulus is probably not a reinforcer for that response.
Quick check
At a clinic, Rosa, an RBT, praises Eli for sharing toys with peers exactly as the plan says. After three weeks of consistent praise, sharing has not increased. Which conclusion is best supported?
Related topics
- Positive vs. negative reinforcement (B.4)
- Motivating operations in action (B.16)
- Designing and evaluating preference assessments (F.4)
- Differential reinforcement (DRA, DRO, DRL, DRH) with and without extinction (G.2)
- Establishing conditioned reinforcers and token economies (G.4)
- Maintaining change after intervention: schedule thinning and natural reinforcers (G.16)