ABA's Secret to Success
Part of the "secret" has to do with the way that autism is diagnosed. Even the most well-respected tests of autism rely on external behaviors of the person being tested. As a result, anything that alters those behaviors can alter the results of the test.
ABA is no different - at its core - from all of the other behavioral modification techniques used on humans and other animals. The desired behaviors are rewarded and the undesired behaviors are punished. Now, I'm sure that ABA's supporters will argue that punishment ("negative reinforcement") is no longer a part of ABA. However, the continuous repetition of the command or "physical prompting" (especially to people who are averse to physical contact) has to be considered a punishment.
So, the ABA therapist uses a slight modification of classic conditioning to get an autistic person (usually a child) to either stop doing behaviors that are considered "autistic" or get them to do behaviors that are considered "normal". If ABA manages to get an autistic person to make eye contact, respond to their name or stop flapping, their scores on the various autism rating scales will move toward less autistic (i.e. "improve").
The big question is this - does ABA actually "treat" autism - or just treat the "symptoms"? Does reducing the behaviors unique to autism equate to reducing autism? Or is it just putting on a different coat of paint?
Reducing the automatisms (e.g. "flapping") will generate improvement on standard autism rating scales, so placing the autistic person in a strait jacket should result in a similar "improvement" - right?
A trained psychologist would not be fooled by a strait jacket, but apparently at least a few have been fooled by its behavioral equivalent.
So, what exactly are parents accomplishing by "treating" their children with ABA? Undoubtably, some of the behavioral manifestations of autism that they (the parents, not the child) find most bothersome may be eliminated. This is good for the parents, I suppose, but is there a corresponding benefit to the child?
Not sure about that.
Again, the child may be considered to benefit because the behaviors that mark them as "different" are reduced. However, since these behaviors don't appear to be bothering the child, the "benefits" would seem to be reaped by the other members of the community who are not "disturbed". And, since many autistic children (and adults) report that their "autistic" behaviors (e.g. flapping, avoiding eye contact, etc.) are comforting, the community's "benefit" is realized at a cost to the child.
But, at least ABA improves something about the autistic person's ability to interact with the outside world - right?
Not sure about that.
Let's say I train my goldfish to maintain eye contact with me by only feeding him when he is looking right at me. Has this made him more able to "relate" to people? I doubt it. The same thing applies to ABA and autism. It may alter the behaviors that other people find objectionable, but it is unlikely to change anything fundamental about the autistic person, any more than training me to drool when I hear a bell ring will turn me into a dog.
Woof. Woof, woof!
I suppose that some people will argue that ABA "therapy" (I prefer to call it "training" - that seems more honest) allows autistic people to "get beyond" the obstacles that keep them "locked in" to their autistic world. That's an interesting idea, but completely without any data to support it.
Others will argue that ABA's "successes" are proof enough that it works. Well, let's look at that argument.
The original work by Ivar Lovaas seemed pretty impressive, with nineteen autistic children improved and eight "recovered". However, subsequent studies have failed to reach that level of success. The scenario where early reports (and subsequent reports from the same author or lab) are favorable but other authors (and labs) are unable to demonstrate the same results is depressingly familiar in science. It is usually a solid indicator that the early reports were in error.
Part of the problem is that a lot of the early studies of ABA compared it to doing nothing. Heck, you don't need to be a PhD psychologist to know that spending more time with any child will make them more sociable. As more people (and, especially important, people who did not train under Ivar Lovaas or his graduates) study the results of ABA, I expect that the effects on IQ and speech will continue to diminish.
So, who is benefitting from ABA? Perhaps only the therapists.