Back to Blog
Clinical Practice

Implementing Differential Reinforcement: DRA, DRO, DRI, DRL Guide

Differential reinforcement is one of the most frequently tested topics on the BCBA exam and one of the most useful tools in clinical practice. This guide explains how to implement DRA, DRO, DRI, DRL, and DRH, with clear definitions, examples, and step-by-step procedures. You will learn how to choose an appropriate schedule, how to set and thin intervals, and how to avoid common errors such as reinforcing the wrong behavior or thinning too fast. We compare each variation, discuss when each is most appropriate, and review how extinction components can improve effectiveness. We also cover ethical considerations including preference for reinforcement-based procedures and how to select replacement behaviors that serve the same function as problem behavior. Use this resource to sharpen your clinical decision-making and prepare for scenario-based exam questions about reducing problem behavior while teaching new skills.

October 7, 2024

What Is Differential Reinforcement?

Differential reinforcement involves reinforcing a target response or response class while withholding reinforcement for other responses. It is a reinforcement-based approach for reducing problem behavior and is generally preferred over punishment because it teaches new skills and has fewer side effects. The BCBA exam frequently asks you to identify which variation is described and which one best fits a scenario.

DRA: Differential Reinforcement of Alternative Behavior

In DRA, reinforcement is delivered for a specific alternative response, while problem behavior is placed on extinction. The alternative behavior ideally serves the same function as the problem behavior.

Steps to Implement DRA

  1. Identify the function of the problem behavior through functional assessment.
  2. Select an alternative behavior that is appropriate, easier than the problem behavior, and produces the same reinforcer.
  3. Teach the alternative behavior with prompting and reinforce it immediately and consistently.
  4. Withhold reinforcement for the problem behavior (extinction), if safe and feasible.
  5. Thin the schedule as the alternative behavior becomes fluent.

Example: A child who screams to get attention is taught to tap a teacher's arm and say "Excuse me." The teacher responds immediately to the tap and ignores screaming.

When the alternative is a communication response that matches the function, DRA becomes functional communication training (FCT).

DRI: Differential Reinforcement of Incompatible Behavior

In DRI, reinforcement is delivered for a behavior that is physically incompatible with the problem behavior. The two cannot occur at the same time.

Example: Reinforcing hands in pockets or holding a fidget toy to reduce hand flapping or hitting. The learner cannot hit while hands are occupied.

The difference between DRA and DRI is that DRI requires incompatibility, while DRA does not. All DRI procedures are DRA procedures, but not all DRA procedures are DRI.

DRO: Differential Reinforcement of Other Behavior

In DRO, reinforcement is delivered when the problem behavior has not occurred for a specified interval of time. Any other behavior is reinforced, so DRO is sometimes called omission training.

Types of DRO

  • Whole-interval DRO: Reinforcement is delivered at the end of an interval if the problem behavior did not occur at any point during the interval.
  • Momentary DRO: Reinforcement is delivered if the problem behavior is not occurring at the specific moment the interval ends.
  • Fixed-interval versus variable-interval DRO: Intervals may be constant or vary around an average.
  • Resetting DRO: The interval timer resets whenever the problem behavior occurs.

Steps to Implement DRO

  1. Collect baseline data to determine the average inter-response time of the problem behavior.
  2. Set the initial interval at or slightly below the average inter-response time so the learner can earn reinforcement frequently.
  3. Deliver a preferred reinforcer at the end of each interval with no problem behavior.
  4. Gradually lengthen the interval as the learner meets the criterion.

Caution: Because DRO reinforces the absence of behavior, it does not teach a replacement skill. It may inadvertently reinforce another undesirable behavior that occurs just before reinforcement. Pair DRO with teaching of a replacement behavior when possible.

DRL: Differential Reinforcement of Low Rates

DRL reinforces responding when it occurs at or below a specified rate. It is used for behaviors that are acceptable but occur too often, such as hand raising or asking questions.

Types of DRL

  • Full-session DRL: Reinforcement is delivered if the total responses in the session are at or below a limit.
  • Interval DRL: The session is divided into intervals and reinforcement is delivered for each interval with responses at or below the limit.
  • Spaced-responding DRL: Reinforcement follows a response only if a minimum amount of time has passed since the last response.

Example: A student who asks to leave the room 15 times per day earns reinforcement for fewer than 10 requests, then fewer than 7, then fewer than 4.

DRH: Differential Reinforcement of High Rates

DRH reinforces responding at or above a specified rate and is used to increase the frequency of a behavior, such as completing math problems or making social initiations. It can be arranged as full-session, interval, or spaced responding.

Comparing the Procedures

  • DRA: Reinforce an alternative response. Best when a replacement skill is needed.
  • DRI: Reinforce a physically incompatible response. Best for topographically defined behaviors like hitting or mouthing.
  • DRO: Reinforce the absence of the behavior for a time interval. Best for rapid reduction when no appropriate replacement is obvious, or as a component in a package.
  • DRL: Reinforce low rates. Best when the behavior is acceptable at lower levels.
  • DRH: Reinforce high rates. Best when you want to increase frequency.

Extinction and Differential Reinforcement

Differential reinforcement is more effective when problem behavior is no longer reinforced. However, extinction can produce an extinction burst, response variability, and extinction-induced aggression. Plan for these, train staff, and ensure the safety of the learner and others. Where extinction is not feasible, increase the reinforcement for the alternative behavior and consider reducing the magnitude or quality of reinforcement for problem behavior.

Schedule Thinning

After problem behavior is reduced, thin the reinforcement schedule gradually. Increase the delay, interval length, or response requirement in small steps, and monitor data. If problem behavior rebounds, return to the previous step. The goal is to reach a schedule that caregivers can sustain in the natural environment.

Common Mistakes

  • Choosing an alternative behavior that does not serve the same function.
  • Making the alternative behavior harder or less effective than the problem behavior.
  • Setting the initial DRO interval too long so the learner rarely earns reinforcement.
  • Thinning the schedule too quickly.
  • Forgetting to train all caregivers, leading to inconsistent consequences.

How This Appears on the BCBA Exam

Expect scenario questions where you must label the procedure from a description. Pay attention to the keywords: alternative for DRA, incompatible for DRI, absence of the behavior for an interval for DRO, rate at or below a criterion for DRL. You may also be asked which procedure is most appropriate for a behavior that is acceptable at low rates, which is DRL, or which is best when the function is known and a communication replacement can be taught, which is DRA or FCT.

Key Takeaways

  • Differential reinforcement is a reinforcement-based approach that is preferred over punishment.
  • Match the variation to the behavior and the function.
  • Pair with extinction when safe and feasible.
  • Thin schedules gradually and monitor data.

Ready to put this into practice?

BCBAprep has 2,000+ original flashcards, 200 timed mini exams, and 40 full-length simulations for every BCBA task list area.

Start FreeSee Plans
Content is protected