What to Do When a Client Withdraws Assent (After You Back Off)

The client says no, walks away, or shuts down. Do you push, redirect, or stop for the day? That single call is what this page is about. Most assent training stops at "back off." That is step one. It is not the plan.

This page picks up where the back-off ends. It is for BCBAs (Board Certified Behavior Analysts, the supervising clinicians) and RBTs (Registered Behavior Technicians, the people running session) who already know what assent withdrawal looks like and want a clear set of next steps. If you are still figuring out what assent is, start with the basics page linked at the bottom. If you are trying to remember when an RBT should stop and call you, the 3-strike rule page covers that.

"Everyone talks about ascent withdrawal means back off, ascent withdrawal means back off. But what about after you back off? What happens then?" From the talk — Matt Harrington

That is the whole question. Let us walk it.

Step 1: Pause and check for safety#

The first thing you do when a kid says no, pushes the materials away, or walks out of the room is stop the trial. Not five seconds from now. Now.

Then you scan. Is the kid safe? Are you safe? Is anything in the room about to become a problem? If the answer is yes to any of those, you handle that first. Safety beats programming every single time.

Once the room is calm, you have a window. The window is not long. Use it to think, not to run the same trial again.

Step 2: Ask "what is worth moving away from?"#

This is the single question that drives every next decision.

"As soon as I see a scent withdrawal away from something, I ask why. What is worth moving away from?" From the talk — Matt Harrington

Not "what is the function." Not "is this escape-maintained." Just: what about the current setup is the kid trying to get away from? Could be the noise. The chair. The smell of the room. The cards. The person sitting across the table. The length of the task. The fact that the iPad is in view but not in hand. The MO (motivating operation, which just means whatever is making something more or less desirable right now, like being tired or hungry).

Write the answer down. You will need it for the next step. If you have no idea, that is fine. Guess. Pick the most likely one and move on. You can test more guesses if the first one is wrong.

Step 3: Shrink the bad (and what "bad" usually means)#

Picture two circles. One holds all the bad stuff about the task. One holds all the good stuff. The kid is choosing where to spend behavior based on which circle is bigger.

"Bad" almost never means the task itself is evil. It means the arrangement around the task is heavier than the payoff. Common bads worth shrinking:

• Duration. Going from fifteen minutes to one minute is a real change.
• Difficulty. Switching from writing the full word to tracing one letter is a real change.
• Sensory load. Headphones, dimmer lights, no clicking pen.
• Format. Writing with a highlighter instead of a pencil. Doing the same drill on a tablet instead of flashcards.
• The person. Sometimes the kid is fine with the task and not fine with who is delivering it. That is a pairing problem, not a programming problem.

You do not need to fix all of them. Pick the heaviest one and cut it.

"We could shrink the bad things and make the bad circle smaller, or we can expand the good things. Or, preferably, we can do both." From the talk — Matt Harrington

Step 4: Grow the good (without dropping the FR)#

Growing the good does not mean handing over a bigger reinforcer for less work. You do not need to drop the FR (fixed ratio, the number of correct responses required before a reward) from ten to one. That is a different conversation and usually the wrong move.

Growing the good means making the task itself worth more to the kid. Three ways that actually work in session:

1. Embed an interest. If the kid loves trains, the counting cards have trains on them. Same drill, same FR, more pull.
2. Hook to a value. For older or more verbal kids, line the task up with something they already want. A teenager who hates personal hygiene might care a lot about having friends. Hygiene is the path to that. Now the task carries weight it did not have before.
3. Change the setting. A beanbag chair, music on low, a different room. The trial is the same. The wrapper is different.

The point is not to bribe the kid into participating. The point is to make the choice between participating and leaving a real choice instead of a foregone conclusion.

Step 5: Re-present with the new arrangement#

You changed something. Now you bring the task back. One trial. See what happens.

If the kid assents, great. Run the trial, deliver reinforcement, log it, keep going. Track which change worked so you can use it again.

If the kid withdraws assent again, you do not run the same modified trial five more times. You change something else. Same intervention, same withdrawal rate. That is the whole reason step three and step four exist.

So change something. Then re-present. Loop.

A worked example: the kid who would not ride the bus#

A real case from the talk. The kid had finally gotten to where he could attend school. Now the team had to get him on the bus to get there.

Starting state: lots of bads, almost no goods. The bus was loud. The ride was an hour long. He did not know anyone on it. There was nothing to do.

First change, shrink the bad and grow the good at the same time. Noise-canceling headphones killed the loudest part of the sensory load. A tablet filled the boredom. One change, two improvements.

Second change, shrink the bad. The team looked at the route. He was last on, last off, which is why the ride was an hour. They moved his pickup down the route so he became the last on, first off. The ride dropped to ten minutes.

Third change, grow the good. They did pairing sessions with the bus aide. After a couple weeks the aide was someone he liked seeing in the morning. Now the bus had a person on it who was a positive.

"First we got a pair of noise-canceling headphones and we gave that sucker a tablet... So the bus ride went from an hour to 10 minutes." From the talk — Matt Harrington

Nobody made him ride the bus three months in a row hoping he would eventually quit fighting. Nobody added an FR-1 reinforcer for getting on. The team changed the arrangement until the choice flipped. Then they kept going.

When to call the BCBA (the 3-withdrawal rule)#

If you are an RBT running session, here is the line. Three withdrawals on the same target, you stop and call the BCBA. Not five. Not ten. Three.

The reason is in step two. Picking what to change after assent withdrawal is a real clinical decision. It involves looking at antecedents, the reinforcer schedule, the shaping plan, and sometimes the whole goal. That is BCBA work. It is unfair to the RBT and unsafe for the kid to leave that call sitting in the tech's lap for an hour.

Faster pivots when the risk is higher. If the kid's assent withdrawal tends to escalate into headbanging or biting, you are not waiting for three. You are calling after one. Severity raises the urgency, not the threshold.

Frequently asked questions#

How long should I wait before re-presenting?

Long enough to be calm. Not so long that the moment is gone. Usually that is between thirty seconds and a couple of minutes. The clock is not the point. The state of the kid is the point. If the kid is regulated and the room is quiet, you can re-present. If the kid is still elevated, you wait.

What if the client withdraws assent again on the modified trial?

You change something else. You do not re-run the same modified trial. The whole point of changing the arrangement is to test a guess about what was "worth moving away from." If the kid still moves away, your guess was wrong or incomplete. Pick a different bad to shrink or a different good to grow. After the third withdrawal on the same target, the RBT calls the BCBA.

Should I document each withdrawal?

Yes. Track assent withdrawal rate the same way you track any other clinical data, by trial. You want to be able to look at the week and see whether withdrawals are going down, staying flat, or climbing. That number tells you whether your changes are working. It is also the number that makes caregiver conversations easy, because you can show progress on something the parent can see.

Where to go next#

If your kid is stuck in a loop where every modified trial still ends in withdrawal, the bottleneck is usually not the arrangement. It is a missing skill. Toleration, communication, or cooperation skills are the thing that lets the kid stay in a context that used to be too heavy. That is a separate plan, but it starts the same way. What is worth moving away from? What skill would make that thing not worth moving away from?

If you have not watched the full session yet, the bus case and the shrink-bad-grow-good walkthrough are worth the hour. The CEU credit is one BACB unit.