Differential reinforcement is task D.3 of the Registered Behavior Technician (RBT) Test Content Outline, with procedures such as DRO and DRA. These 10 questions show how to tell DRO, DRA, DRI and DRL apart, how momentary and whole-interval DRO differ, what to do when a response comes too soon, and why a replacement response must be easier than the problem behavior.
Question 1
At recess, an RBT watches Jada, who spits at peers. Each 5-minute stretch with no spitting earns her a token from the RBT, regardless of how else she spent those minutes or which games she played.
Which differential reinforcement schedule is Jada on?
- A) DRO✓
- B) DRI
- C) DRL
- D) DRA
💡 Key Takeaway
DRO pays off for zero occurrences of the target behavior during the interval.
Show rationale
DRO, differential reinforcement of other behavior, delivers reinforcement when the target behavior has not occurred for a set interval, so a token after every 5 minutes with zero spitting is DRO. The clue is that Jada earns the token regardless of how else she spent the time: no specific replacement response is required, only the absence of spitting. DRA would require Jada to perform a particular alternative behavior, such as saying 'move, please' to a peer, to earn the token. DRI would require a specific behavior that physically cannot happen at the same time as spitting. DRL would still permit some spitting and reinforce only when it stays at or below a set limit, an approach not suited to behavior that should not occur at all.
Question 2
While walking through a grocery store with Omar, an RBT praises him and gives him a point each time he puts his hands in his pockets as they walk. When Omar flaps his hands, the RBT provides no praise or points.
Which type of differential reinforcement is this?
- A) DRL
- B) DRI✓
- C) DRO
- D) DRA
💡 Key Takeaway
If the replacement makes the problem behavior physically impossible, it is DRI.
Show rationale
DRI, differential reinforcement of incompatible behavior, reinforces a response that cannot physically occur at the same time as the target behavior. Omar cannot flap his hands while they are in his pockets, so reinforcing hands in pockets while flapping earns nothing is DRI. DRA is the closest distractor, but DRA reinforces an alternative behavior that could occur alongside the problem behavior, such as asking for a break while flapping; when the replacement is physically incompatible, DRI is the precise label. DRO would deliver reinforcement after a period with no flapping, without requiring any particular behavior. DRL would reinforce flapping that stays at or below a set number, which still permits some flapping during the trip.
Question 3
Ben asks, 'Is it lunch yet?' many times each morning. His spaced-responding DRL plan says the RBT answers only if at least 10 minutes have passed since his last question. Six minutes after a question, Ben asks again.
What should the RBT do?
- A) Answer the question, then restart the 10-minute timer.
- B) Withhold an answer but keep the original timer running.
- C) Withhold an answer and restart the 10-minute timer.✓
- D) Answer the question, then keep the original timer running.
💡 Key Takeaway
In spaced-responding DRL, an early response earns nothing and restarts the clock.
Show rationale
In spaced-responding DRL, a response is reinforced only when it occurs after a minimum amount of time has passed since the previous response. Because Ben asked after 6 minutes, the question is not reinforced, and the 10-minute interval starts over from that early response, so withholding an answer and restarting the timer is correct. Withholding an answer while keeping the original timer running would let a question 4 minutes later earn an answer, reinforcing questions spaced only 4 minutes apart. Answering and keeping the original timer reinforces the early question directly. Answering and then restarting the timer still reinforces the too-early response, teaching Ben that frequent asking works. Restarting after every early response is what spaces the behavior out.
Question 4
Lila's behavior plan calls for momentary DRO on a fixed 3-minute schedule, with screaming checked only at the instant each period ends. In one period, she screams 60 seconds in, yet she is calm and silent at the 3-minute check.
What is the correct way to apply the plan?
- A) Reset the timer at the scream and check again 3 minutes later.
- B) Withhold the reinforcer, since screaming occurred during the interval.
- C) Deliver the reinforcer as soon as the screaming stops at 1 minute.
- D) Deliver the reinforcer at 3 minutes and keep the schedule as set.✓
💡 Key Takeaway
Momentary DRO checks only the moment the interval ends; nothing resets.
Show rationale
Momentary DRO delivers reinforcement if the target behavior is absent at the moment each interval ends; behavior earlier in the interval does not count and does not reset the timer. Lila is silent at the 3-minute check, so the RBT delivers the reinforcer and keeps the schedule as set. Resetting the timer at the scream describes a resetting DRO, which this plan does not use. Withholding the reinforcer because screaming occurred earlier in the period applies whole-interval DRO rules, where any occurrence forfeits the interval. Delivering the reinforcer as soon as the screaming stops would follow screaming closely and could strengthen it, and it ignores the scheduled check time. Following the exact DRO variant the BCBA wrote keeps treatment consistent across staff.
Question 5
A BCBA's plan for Kai's calling out sets a lower limit for each session: 10 or fewer on Monday, 8 or fewer on Tuesday, and 6 or fewer on Wednesday. The RBT delivers a reward whenever Kai stays within that day's limit.
What is this differential reinforcement procedure called?
- A) Differential reinforcement of diminishing rates✓
- B) Differential reinforcement of alternative behavior
- C) Differential reinforcement of other behavior
- D) Differential reinforcement of incompatible behavior
💡 Key Takeaway
Diminishing rates: the allowed count steps down session by session.
Show rationale
Differential reinforcement of diminishing rates delivers reinforcement when responding falls at or below a criterion that is lowered step by step across sessions, as with Kai's limits of 10, then 8, then 6. It is a form of DRL used to reduce behavior gradually rather than eliminate it at once. Differential reinforcement of other behavior would require zero calling out during an interval, not a shrinking allowance. Differential reinforcement of alternative behavior would reinforce a specific replacement response, such as raising a hand, which the plan does not mention. Differential reinforcement of incompatible behavior would reinforce a particular response that physically cannot occur alongside calling out. The RBT's job is to compare each session's count with that day's criterion.
Question 6
Under a full-session DRL plan, Maya earns a sticker at the end of her 30-minute reading block if she leaves her seat 3 or fewer times. Today the RBT records 5 times out of her seat during the block.
What should the RBT do at the end of the block?
- A) Give the sticker, since she stayed seated for most of the block.
- B) Withhold the sticker, since she exceeded the limit of 3.✓
- C) Give the sticker, since some leaving is allowed under DRL.
- D) Take away a sticker she earned during an earlier block.
💡 Key Takeaway
Over the DRL limit means no reinforcer for that period, and nothing more.
Show rationale
In DRL, reinforcement is delivered only when the target behavior stays at or below a set limit for the period. Maya's limit was 3, and she left her seat 5 times, so the RBT withholds the sticker for this block and offers a fresh chance in the next block. Giving the sticker because some leaving is allowed misreads DRL: some responses are tolerated, but only up to the limit. Giving it because she stayed seated for most of the block ignores the criterion and would reinforce a rate above it. Taking away a sticker earned during an earlier block adds response cost, a punishment procedure the plan does not include, and an RBT should not add procedures without BCBA direction.
Question 7
In the first week of a DRA program, Noah is learning to hand the RBT a 'break' card instead of pushing his materials off the table. Pushing materials no longer produces a break.
How often should the RBT honor the break card at this early stage?
- A) About every fifth card, on a variable schedule
- B) Every third time Noah hands over the card
- C) Every time Noah hands over the card✓
- D) Only after Noah has also finished one task
💡 Key Takeaway
New replacement behaviors are reinforced every time (FR1); thinning comes later.
Show rationale
When a replacement behavior is first taught in DRA, it should be reinforced every time it occurs, a continuous or FR1 schedule. Honoring every break card makes the new response more efficient than pushing materials and helps it compete with the problem behavior quickly. Reinforcing every third card, or about every fifth card on a variable schedule, are thinning steps that come later, only after the card is used reliably and the BCBA directs the change; starting there risks the problem behavior returning. Requiring Noah to finish a task before the card works is also a later step, often called demand fading, and introducing it now could make the card seem ineffective. The RBT thins reinforcement only as the BCBA directs.
Question 8
Tara screams to get her tablet. Her DRA program requires her to say, 'May I please have my tablet now,' to request it, which she rarely manages, and after three weeks her screaming has not decreased.
What is the BEST action for the RBT?
- A) Keep requiring the full sentence until she says it more often.
- B) Give the tablet after screaming, since the sentence is too hard.
- C) Switch to a one-word request without first consulting the BCBA.
- D) Tell the BCBA the request may take more effort than screaming.✓
💡 Key Takeaway
A replacement must be as easy as, or easier than, the problem behavior.
Show rationale
A replacement behavior in DRA should require the same or less effort than the problem behavior; otherwise screaming stays the easier way to get the tablet. A long sentence Tara rarely completes is likely harder than screaming, so the best action is to tell the BCBA, who can revise the replacement. Switching to a one-word request without consulting the BCBA may match what the BCBA eventually chooses, but changing a treatment plan is outside RBT scope. Giving the tablet after screaming reinforces the problem behavior and undoes the DRA. Continuing to require the full sentence without reporting the data ignores a clear sign that the replacement is too effortful and lets screaming keep working better than asking.
Question 9
Zoe asks for help even on tasks she has mastered. Under her interval DRL plan, she earns a token at the end of each 10-minute interval if she asks for help 2 or fewer times. In the current interval, she asks twice.
What should the RBT do when the interval ends?
- A) Give the token, since 2 requests meets the criterion.✓
- B) Withhold the token, since she must ask fewer than 2 times.
- C) Give the token only if she also finishes the task alone.
- D) Withhold the token, since any request ends the interval.
💡 Key Takeaway
At or below the DRL limit means the interval is reinforced.
Show rationale
In interval DRL, the session is divided into intervals, and reinforcement is delivered at the end of any interval in which the target behavior stays at or below the criterion. Zoe's criterion is 2 or fewer requests, and she asked exactly twice, so she has met it and the RBT gives the token. Withholding the token because any request ends the interval confuses DRL with DRO, which requires zero occurrences. Withholding it because she must ask fewer than 2 times misreads the criterion, since at or below includes 2. Giving the token only if she also finishes the task alone adds a requirement the plan does not include, and RBTs implement the plan exactly as written.
Question 10
Before a session, an RBT reviews the differential reinforcement plan for Aria, who hits peers during group play. The plan reinforces a replacement behavior that could still occur at the same moment as hitting.
Which replacement behavior fits this DRA procedure?
- A) Clasping her hands behind her back
- B) Saying 'stop, please' to the peer✓
- C) Keeping both hands folded in her lap
- D) Sitting on both of her hands
💡 Key Takeaway
DRA alternatives can co-occur with the problem behavior; DRI behaviors cannot.
Show rationale
DRA reinforces an alternative behavior that serves as a replacement but can physically occur at the same time as the problem behavior. Aria could say 'stop, please' while hitting, so that request is an alternative, not incompatible, behavior, which fits DRA. Keeping both hands folded in her lap, clasping her hands behind her back, and sitting on both of her hands all make hitting physically impossible while they continue, so reinforcing any of them would be DRI, differential reinforcement of incompatible behavior. DRI is a narrower type of DRA, and the defining test for an incompatible behavior is that the two cannot occur together. The RBT should know which version the plan specifies so the correct response is reinforced.