Autism and Behaviorism
Part 2 of 3: ABA may lead to lasting damage to children.
Posted January 28, 2020 | Reviewed by Devon Frye
This is Part 2 of a series on autism. You can read Part 1 here.
Here is potentially a very serious set of challenges: a child who can't communicate the way other children do and doesn't seem to understand boundaries, who is exquisitely sensitive to (or, conversely, in constant need of) sensory stimulation, who may explode when there's even a slight deviation from rigid routines, who engages in repetitive behaviors and might even be at risk of injuring himself or those around him. All of this is unsettling at best, and often downright frightening, for a neurotypical adult.
Enter ABA: an intensive training regimen consisting of an elaborate system of rewards to make children comply with external directives, and to memorize and engage in very specific behaviors. An expert promises to train the child to make eye contact or point at an object on command, to stop fluttering his hands or rocking—in short, to make him act like a normal kid. ABA is the accepted, expected, even mandated system for dealing with autistic children.
But what do the children themselves say? It is nothing short of stunning to learn just how widely and intensely ABA is loathed by autistic adults who are able to describe their experience with it. Frankly, I'm embarrassed that, until about a year ago, I was completely unaware of all the websites, articles, scholarly essays, blog posts, Facebook pages, and Twitter groups featuring the voices of autistic men and women, all overwhelmingly critical of ABA, and eloquent in describing the trauma that is, in their view, its primary legacy.
How is it possible that their voices have not transformed the entire discussion? Suppose you participated in implementing a widely used strategy for dealing with homelessness, only to learn that the most outspoken critics of that intervention were homeless people. Would that not stop you in your tracks? What would it say about you if it didn't? And yet the consistent, emphatic objections of autistic people don't seem to trouble many ABA practitioners at all.
So why are autistic people so opposed? For many, the underlying assumption that they have a disease that needs to be cured is misconceived and offensive. Resistance to this premise led to the founding of the Autistic Self-Advocacy Network and has been described in such mainstream periodicals as Salon, the Atlantic, and the New York Times.
From the last of those three articles: "Autism has traditionally been seen as a shell from which a normal child might one day emerge. But some advocates contend that autism is an integral part of their identities, much more like a skin than a shell, and not one they care to shed. The effort to cure autism, they say, is not like curing cancer, but like the efforts of a previous age to cure left-handedness." Or like curing homosexuality: In the autism community, ABA is often compared to gay conversion therapy. Many argue that its goal is to force these children to stop being who they are.
While this objection presumably would apply to any method used to "cure" people of autism, ABA is uniquely problematic. Here's why:
1. It's dehumanizing and infantilizing.
Should we really be surprised that people chafe at spending hour after hour being promised the equivalent of a doggie biscuit for jumping through hoops? (Actually, one professional dog trainer rejects the comparison. After investigating ABA, she exclaimed, "I would never treat a dog like that!")
2. It ignores internal realities.
ABA is rooted in an ideology that proudly stays on the surface, committed to reinforcing whatever behaviors the people who control the reinforcements endorse and extinguishing those they don't. This focus on behavior—on that which can be seen and quantified—isn't just problematic theoretically (reflecting a truncated understanding of human psychology) and ethically; it also fails from a practical perspective, as has been demonstrated repeatedly. If you train an autistic kid to stop rocking or squealing or flapping his hands, you have done exactly nothing to address what elicited that self-regulating or self-stimulating behavior and its emotional significance to him. Kids need to feel safe; ABA just eliminates the (unusual) ways he tries to attain that safety—for example, by elaborately praising him for "quiet hands."
3. It undermines intrinsic motivation.
Rewards fail to help people (of any age) develop an underlying commitment to whatever value or action is being reinforced. Worse, they actively impede the development of intrinsic interest. Thus, several studies have shown that when children are rewarded or praised for doing something nice for someone else, they are likely to be less generous in the future as a result. Students who are led to focus on getting good grades become less interested in the learning itself. Employees who are promised bonuses for meeting certain criteria come to find their work less satisfying.
In a therapeutic context, the fact that rewards don't promote, and may undermine, an interest in doing x often means x won't be generalized to other situations. As one writer explained, "A child may learn to make eye contact in response to 'How are you?' and to reply, 'Fine, how are you?' But such rote memorization does not give the child the intuition to know when a stranger is to be greeted warmly and when to be avoided, and it does not enable him to meet his grandmother with greater warmth than the grocer."
Behaviorists insist this can be solved by fiddling with the reinforcement protocol—the type or size of the rewards, the schedule on which they're meted out, or the specificity or difficulty of the target behavior. But decades of research, along with real-life experience, suggest that the problem inheres in the whole idea of extrinsic inducements.
4. It's all about compliance.
ABA does not exist to do what's best for the children themselves, to meet their needs and honor their preferences. Its goal is to extinguish behaviors that make the people around them uncomfortable. What's candidly called "compliance training"—often preceded by the adjective "errorless"—is an integral feature of ABA.
If their theory collapses the richness of human experience into measurable behaviors and their practice relies on objectifying children, is it really surprising that the widespread antipathy for ABA expressed by people who have had it done to them doesn't seem to faze its practitioners and proponents one bit? Behaviorists see only behaviors. The experience of those to whom they're doing things is, if you'll excuse the expression, outside the spectrum of what they've been trained to detect and address.
5. It creates dependence.
If you devote tens of hours each week to exhaustively and exhaustingly teaching a small child that he gets a reward when he suppresses his own preferences and does exactly what he's told, you have grievously compromised that child's nascent sense of autonomy. The more you control him—even with the sugar-coated control of positive reinforcement—the more he may come to rely on being controlled. This is why many autistic people view ABA as not only distasteful but dangerous.
A peer-reviewed report in the Journal of Cogent Psychology discussed the inappropriateness of using ABA on children with low verbal skills given the likelihood that they will be left with lifelong passivity and, in particular, "prompt dependence"—that is, an inability to initiate activities after having been trained to wait for a nudge or command.
6. It communicates conditional acceptance.
What defines a reward isn't just its desirability—the chance to play with a favorite toy, go to the playground, watch a video—but the fact that it's offered contingently. What someone finds enjoyable has been turned into a lever with which to control the child. (As one critic puts it, "ABA therapists are trained to find out what your child loves the most and hold it [for] ransom.") Now the child must earn the right to do what gives her pleasure by obeying the adult—and that right can be snatched away at any time. (The latter reminds us of the punitive underbelly of "positive reinforcement"; every carrot contains a stick.)
But there's also a deeper and more disturbing kind of conditionality at work here, particularly when affection and attention are treated not as something that all kids should receive but as additional goodies to be dangled and withdrawn. With endless "Good jobs!" and other expressions of enthusiasm, offered only conditionally, ABA makes care transactional and leads children to infer that they're worthwhile only when they do what is demanded of them. This message is toxic regardless of whether it succeeds at (temporarily) buying the desired behavior.
And to the extent it teaches autistic children that they must suppress or mask their impulses in order to pass as "normal," it can create both shame and anxiety. One suggestive, though not definitive, investigation found that children with autism were significantly more likely to display posttraumatic stress symptoms if they had been treated with ABA.
This series is continued in Part 3.