Rad N Bad: The Stereotypy Standoff — Autonomy, Function, and Contingency Design

How the Hosts Defined Stereotypy Before Arguing About It

The hosts set the terms before they argued. They defined stereotypy (repeated motor or vocal behavior) as fixed patterns that repeat again and again. Their examples were rocking, spinning, and repeated sounds or words. They left function out of the definition on purpose. To name a function you would need a functional analysis (a test of why behavior happens). So the debate ran on form, not on cause.

The format shapes how you should read the rest. A coin toss picked the sides. Mike won the toss and chose to argue for treating stereotypy. Sean took the side against it. Neither host knew his side before the show started. Both said the points were not their personal beliefs. They were the arguments you hear across the field.

That matters for practice. You get each side at full strength with no hedging. The hosts banned "it depends" as an answer. Then they hunt for the middle ground together. Their landing spot is not a compromise for comfort. It is a different question than the one most people fight about.

But we are defining stereotypy as repetitive and fixed patterns of motor or vocal behavior that are repeated consistently.

From the talk — Sean Yocum, BCBA, LBA & Michael Carrero, BCBA, LBA

The Case Against Intervening: Autonomy, Regulation, and Masking

Sean led with bodily autonomy. Many critics hold that non-injurious stereotypy is part of normal self-regulation. It is not a disease that needs a fix. On that view, treating it without clear harm steps outside our ethical scope. A behavior can be easy to see and easy to change. That alone does not make it ours to change.

His second point was emotional regulation. Many neurodivergent people report that stimming lowers stress and sensory overload. Suppress it and you may get more stress, not less. Sean warned the behavior can shift into less safe forms instead. He also raised masking (hiding who you really are). He said self-reports and research tie long-term masking to mental health risk.

Then came the quiet hands critique. Goals like quiet hands or still body often serve the observer, not the learner. They treat typical bodies as the standard everyone should meet. Sean said that choice attacks the social validity (whether goals matter to people) of the goal itself. It turns a human rights question into a procedure question.

His last point here was communication. Repetitive behavior often moves with an internal state: excitement, frustration, pain, or overload. For a learner with little speech and no strong AAC (devices that speak for users) repertoire, the behavior may be the message. Cut the message before you read it and you lose real information.

I think removing it without a replacement is functionally equivalent to removing a coping strategy.

From the talk — Sean Yocum, BCBA, LBA & Michael Carrero, BCBA, LBA

Why Extinction Is a Weak Tool for Automatically Reinforced Stereotypy

Sean closed his side with a technical problem. Say the behavior is automatically reinforced (the behavior rewards itself). The payoff sits inside the person. It is always available to them. It is not under your environmental control. Extinction (cutting off a behavior's payoff) has nothing left to cut off.

So plans fall back on other tools. Response blocking (physically stopping the behavior) is one. High-effort redirection, punishment procedures, and nonstop competing demands are others. Each one can build an aversive setting around the learner. Each one raises questions about restriction, assent (the learner's ongoing okay), and quality of life.

There is also a fidelity problem (doing a plan exactly as written). You cannot block vocal or motor stereotypy every single time. Miss some and the behavior now pays off sometimes. That is intermittent reinforcement (a payoff that comes sometimes). Sean's warning is blunt: the rate can go up, not down.

It is impossible to implement extinction on a behavior like this.

From the talk — Sean Yocum, BCBA, LBA & Michael Carrero, BCBA, LBA

The Case For Intervening: Learning, Social Access, and Independence

Mike agreed with much of that, then pushed on scope. He said he loves the idea of autonomy. His question was where one person's autonomy starts to limit their own future options. His first case was access to learning. Picture a learner with very high rate motor stereotypy and no spoken words. You want to teach a speech device. The behavior and the teaching compete for the same hands and minutes.

He was clear about the goal, and the goal was not zero. He used rough numbers: move the rate from 100 to 80 so teaching can land. The aim is room for instruction, not a still body.

His second case was social access. Shape when, where, and how instead of erasing the behavior. He described a learner who talks well and also stims. Peers mock him, and he stops going out in the future. Mike asked whether that outcome warrants teaching some context rules.

His third case was functional development: ADLs (daily living skills like dressing) and work. Learning any new skill means giving up some older response. He used a baseball swing to make it plain. If he swings the bat like an axe, a good coach fixes the swing. You do not leave it alone just because it is his. Mike framed every call as a cost-benefit talk with the learner.

I'm not saying get rid of the stereotypic behaviors, but maybe we teach how, when, where, and shape those behaviors to have more meaningful outcomes in certain situations, right?

From the talk — Sean Yocum, BCBA, LBA & Michael Carrero, BCBA, LBA

The Forms People Forget: Safety, Risk, and Stereotypy That Hurts

Mike then named a blind spot in the whole debate. Most of us picture rocking, hand flapping, and scripting. Those are the easy cases to leave alone. He asked about the harder topographies (the forms a behavior takes).

What if the behavior is high-intensity aggression, or aggression is a byproduct of it? What if it is excessive running that can knock a person down? What about high-magnitude public self-stimulatory sexual behavior? What about self-injury? For cases like these, Mike said reduction belongs on the table.

He paired that with a careful point about function. He does not decide the function of a behavior. He only works to understand it. And one function in one learner does not set the function for everyone. The same form can serve very different ends across people. That is why he refused to answer the question by topography alone.

just because it serves one function for one individual or multiple individuals doesn't inherently mean that that's the function for everybody

From the talk — Sean Yocum, BCBA, LBA & Michael Carrero, BCBA, LBA

The Reframe: Behavior Reduction Versus Contingency Design

Both hosts called the standard debate a false binary. Either you cut stereotypy to raise learning, or you never touch it in the name of autonomy. Neither host would stand on either pole. Sean said the field has to grow past that choice.

Then he offered the reframe that carries the whole episode. Our science is not about behavior reduction. It is about contingency design (arranging what follows a behavior). He applied the same logic to self-injury. You do not set out to reduce the behavior. You design contingencies that shape behavior instead. While one rate falls, skill acquisition (learning new useful skills) and contact with reinforcement rise.

Run stereotypy through that frame and the target moves. Stereotypy is almost never the target. Access is the target. That single move settles most of the fights in the two sections above.

It's about contingency design with everything.

From the talk — Sean Yocum, BCBA, LBA & Michael Carrero, BCBA, LBA

A Decision Guide: The Five Access Domains

The decision guide rests on five access domains. The first four are instruction, environments, reinforcement, and safety. The fifth is autonomy. Note that reinforcement includes access to stereotypy itself. For any proposed goal, ask whether the behavior blocks one of these domains.

If it blocks none of them, you have no case for intervention. If it blocks one, you have work to do. The work is not suppression. It is teaching discrimination (telling one setting from another), contextual choice, and functional replacement skills.

Stereotypy stays available in this model. It is not banned anywhere. The environment teaches the choice, so the plan does not run on compliance. In some settings the behavior fits fine. In others it competes with a goal the learner values. The learner keeps the ability to engage either way. Nothing here involves mechanical restraint.

Mike added the piece most plans skip. If all you do is drop a rate, you taught nothing at all. Teach goals that are functional and meaningful and stereotypy often falls on its own. He also pushed for honest targets. Aggression plans aim at near zero, not zero, and stereotypy plans should be just as honest.

If stereotypy does not interfere with any of those domains, then it doesn't warrant the intervention.

From the talk — Sean Yocum, BCBA, LBA & Michael Carrero, BCBA, LBA

Social Validity, Redefined by the Learner

Both hosts expected a fight about social validity. Their answer is that this frame does not abandon it. It reclaims it and redefines it. The question is not whether the learner now looks typical to a watcher.

They offered three tests instead. Does this raise quality of life as the learner, the family, and the community define it? Does it widen choice or narrow it? Does it cut suffering or the inability to learn without erasing the person's identity? Those three tests make a plan client-centered rather than observer-centered.

The last thread was professional nerve. Mike said too many analysts weigh how they will look online over what the learner needs. He pushed for critical thinking and for asking a colleague before you decide. Run the cost-benefit both ways on paper. State what happens if you do nothing at all. State what happens at the other extreme. Then pick the point that protects both access and autonomy.

Does this expand choice rather than constrain it?

From the talk — Sean Yocum, BCBA, LBA & Michael Carrero, BCBA, LBA