Autism and Behaviorism

When a common practice isn't necessary or useful even under presumably optimal conditions, it's time to question whether that practice makes sense at all.

For example, if teachers might not need to give grades, even in high school. And if eliminating grades clearly benefits their students, how can we justify grading younger children? If research shows there may be little or no benefit to assigning homework even in math, which is the discipline that proponents assume makes the clearest case for its value, why would we keep assigning it in other subjects?

And if it turns out that, contrary to widespread assumptions, behavior modification techniques aren't supported by solid data even when used with autistic kids. Why would we persist in manipulating anyone with positive reinforcement? A rigorous new meta-analysis challenges the claim that applied behavior analysis (ABA) therapy is the only intervention for children with autism that's "evidence-based." In fact, it raises serious questions about whether ABA merits that description at all.

Before exploring the new report (in part 3 of this series), let's consider what we know about rewards and positive reinforcement more generally. In 2018, I reviewed two decades of recent research for the 25th-anniversary edition of my book Punished by Rewards. These studies strongly confirm the original findings: Carrots, like sticks, are not merely ineffective over the long haul but often actively counterproductive—at work, at school, and at home—and these negative effects are found across ages, genders, and cultural settings.

As a rule, the more you reward people for doing something, the more they tend to lose interest in whatever they had to do to get the reward. And they often end up being less successful at a task they're completing than are people who weren't offered any reward for doing it. (Even more damaging, according to the research, is an arrangement where people are offered a reward for doing something well.)

In the face of such evidence, which has been accumulating for about half a century, the last refuge of behaviorists has been to claim that rewards must be used on people with special needs and challenges. Heavy-handed controlling tactics, and rewards in particular, are most pervasively applied to children who carry a label that sets them apart. They are often subjected to a relentless regimen of Skinnerian manipulation, complete with elaborate charts, point systems, and reinforcement schedules. Even teachers and clinicians who would hesitate to treat other children that way assume it's justified, or even necessary, to do so with, well, you know, those kids.

But that claim has always been hard to defend based on research. Even older studies showed, for example, that (a) teachers act in a more controlling way with children who are thought to have a learning disability than they do with other students, (b) moral objections aside, the use of control almost always backfires, (c) children identified as learning disabled are just as intrinsically motivated to learn as their peers are in the early grades, but (d) they soon come to be "more dependent on external sources of evaluation" such as rewards and praise, "whereas regular students [feel more] capable of making decisions on their own." (I'm quoting here from a study in the Journal of Clinical Child Psychology).

Much the same is true of children whose diagnosis would now be ADHD.[1] In fact, regardless of whether we're talking about kids with emotional issues, problems with learning or attention, intellectual disabilities, or behavior challenges, offering rewards (including praise) for doing what the adult wants can sometimes buy temporary compliance but rarely does the intended effect generalize to other situations. And not uncommonly it is actually worse than doing nothing.

However, like economists with their axiomatic commitment to using incentives to change people's behavior, "behavior analysts" have set up an unfalsifiable belief system: When behavioral manipulation fails, the blame is placed on the specific reinforcement protocol being used or on the adult who implemented it or on the child—never on behaviorism itself. The underpinnings of that ideology include: a focus only on observable behaviors that can be quantified, a reduction of wholes to parts, the assumption that everything people do can be explained as a quest for reinforcement, and the creation of methods for selectively reinforcing whichever behaviors are preferred by the person with the power. Behaviorists ignore, or actively dismiss, subjective experience—the perceptions, needs, values, and complex motives of the human beings who engage in behaviors.

The late Herb Lovett used to say that there are only two problems with "special education" in America: It's not special and it sure as hell isn't education. The field continues to be marinated in behaviorist assumptions and practices despite the fact that numerous resources for teachers, therapists, and parents offer alternatives to behavior control.[2]

Behaviorism Essential Reads

Teaching and the New Behaviorism

My Dog Is Quite a Character

These alternatives are based on a commitment to care and to understand. By "care," I mean that our relationship with the child is what matters most. He or she is not a passive object to be manipulated but a subject, a center of experience, a person with agency, with needs and rights. And by "understand," I mean that we have an obligation to look beneath the behavior, in part by imaginatively trying to adopt that person's point of view, attempting to understand the whys rather than just tabulating the frequency of the whats. As Norm Kunc and Emma Van der Klift urged us in their "Credo for Support": "Be still and listen. What you define as inappropriate may be my attempt to communicate with you in the only way I can... [or] the only way I can exert some control over my life... Do not work on me. Work with me."

[continued in Part 2]

Notes:

1. In one early experiment, rewards used to motivate hyperactive children led them to respond more impulsively. In another, rewards either had a negative effect on performance from the beginning or else they had beneficial effects that disappeared as soon as the rewards stopped coming. In a third study, with children who had short attention spans but were not hyperactive, the use of rewards failed to improve reaction times as the experimenters had predicted; rather, they caused children to make more mistakes. Citations for these studies as well as those referenced in the preceding paragraph can be found in Punished by Rewards.

2. Among them: a terrific book by Lovett himself called Learning to Listen: Positive Approaches and People with Difficult Behavior; The Explosive Child, and several subsequent books by Ross Greene; When Slow Is Fast Enough by Joan Goodman; and Engaging Troubling Students: A Constructivist Approach by Scot E. Danforth and Terry Jo Smith.