Tuesday, June 23, 2009

Gadfly's take on new ABA+TMS study

I see that the NIMH hasfunded a new study on the use of transmagnetic stimulation in conjunction with ABA. TMS as it is called is the use of a tool that generates a magnetic field around a given brain area that will either knock it out or enhance it. It can be used to assess function in the area depending on what affects it has. It also has interest as a possible therapeutic tool for conditions such as depression and autism. I do not think I have to explain what ABA is to most of my readers. The grant from the NIMH is $900,000 given to researcher Manual Casanova. Casanova has done research on minicolumns which are groups of brain cells that are abnormally small in autism. The small minicolumns are believed to result in lack of inhibition of certain actions. Using TMS Casanova believes that he can artificially stimulate these inhibitory activities that the small minicolumns are not able to produce. Apparently some believe that this technique used in conjunction with ABA will produce some sort of results in autism.

One of the problems that I have with this is that Lovaas(the most celebrated of ABA practitioners and researchers) has already received quite a bit of money from the NIMH to study the adult outcomes of the participants of his subjects in his renowned 1987 study where he alleges a coin-flip probability of normal functioning in nearly half of the autistic children who as research subjects received 40 hours a week of ABA. I don't know how much money the NIMH awarded him to study these as yet still unpublished adult outcomes; however, this current study is running a tab of approximately $900,000. I don't feel that the taxpayers should foot the bill for more research into ABA therapy until we get the true story of what happened to those supposedly recovered autistic adults who are now in their late 30's, early 40's. Did they graduate college, did they become physicians, lawyers, dentists or engineers? Did they marry? Did their improved IQ scores remain stable? According to what little I have been able to gauge from my correspondence with Lovaas heir-apparent Tristram Smith, there were other psychologic tests administered to these subjects as adults as well.

Another problem is apparently only a pilot study has been done with TMS in which Casanova has gotten some results in a small group of autistic children. We still don't know what its effects are or lack of them is in autistic with just the TMS alone and no ABA other than these very preliminary findings. Of course, in a well-designed study we might be able to get an idea of which treatment affected what change.

Confounding variables were also dealt with in Lovaas(1987). What was found, however, is of chagrin to most behaviorists. Lovaas' work found that the active ingredient in ABA was the contingent use of aversives, such as hitting, slaps, water in the face and even in some cases electric shock. Without the aversives Lovaas (1987) would not have its marketing points and ABA never would have been able to achieve its current status as a cottage industry. This is in spite of the fact that the Hughes act outlawed aversives on developmentally disabled children in 1991. Therefore in jurisdictions such as my home state of California where aversives are a no-no, ABA is dishonestly marketed. Lovaas claimed in a rebuttal to his detractors written about 8 years ago or so that he had found new methods that made the use of aversives obsolete. If this is the case, to the best of my knowledge, he has never presented any evidence of this in a peer reviewed journal-the same fate that has become of his apparently NIMH funded aborted study of the adult outcomes of these children.

Transcranial magnetic stimulation may or may not be a state of the art treatment that will produce pie-in-the-sky rather than shit-from-the-sewer results in persons with autism. Interestingly enough, John Elder Robison, who wrote the bestseller Look Me in the Eye has gotten into the act This is in spite of the fact that on page 5 of his book he states that Asperger's is not a disease but a different way of being. He went on to state there was no cure and no need for one. Yet now he seems to feel the need to be a guinea pig in TMS research stating:

"I knew how much I had struggled as a young person - not knowing, being called 'retard' or 'freak.' This might help young people."

A possible complication from TMS is that it can induce seizure disorders. But supposedly, technology has improved and people have learned how to use this effectively enough to avoid giving someone a seizure.

Though this might be a good line of research at some point in time, one problem I have is that not enough is known about the etiology of the brain dysfunction responsible for ASD's. I suppose if Dr. Casanova were available to me he could explain the rationale to me based on what is known of the neurophysiology of autism aside from the work he has done on minicolumns. It may be that the autistic brain has various things wrong with it in various regions spread out diffusely within the brain. There may be an elaborate mirror neuron system with a variety of connections and it might be hard to find anything that is wrong. I guess Dr. Casanova being the recipient of the grant may have to show some evidence that this is a promising line of research and possible treatment of autism. Of course the same cannot be said of Dr. Lovaas and his cronies. If the work is replicated by a researcher besides Dr. Casanova perhaps studies could be made of TMS in lieu of ABA rather than as an adjunct with ABA. As regular gadfly readers (of which I know there are not very many)I am a cynic when it comes to anything in autism research or the new autism treatments that come up not infrequently as variable flavors of the month. This work may or may not yield some promise for autistic people, but it may not generate anything other than a bitter harvest for years or decades to come. Also work in neurofeedback such as what Jaime Pineda and Lindsay Oberman and their ilk are doing may also be promising avenues. This work may generate figurative diamonds in the rough at some point in time. Much work needs to be done though.


Marius Filip said...

Come on Jonathan, when you say aversives are the active ingredient of ABA, you are copy-pasting from "The Misbehavior of Behaviorists". You should do better than that.

What Michelle Dawson does there and I hope you won't do, is to grossly misrepresent the role of harsh aversives (which are illegal in some juristictions, as you mentioned) and to exhibit a basic misunderstanding of behaviorism - plus a ton of wrong intention.

Let me re-interpret the findings on harsh aversives for the sake of your readers.

What harsh aversives do is to accelerate early learning of simple elements. This has been well known to educators and parents for centuries, who used education by the rod, and is still known in correctional facilities with truly disobedient subjects.

So, if you want to teach someone something simple, the most efficient way is to beat the crap out of him - and out of fear he'll ultimately do what you want.

Aversives, by any means, CANNOT be the active ingredient of ABA because, as it is well known to behaviorists, aversive stimuli BLOCK GENERALIZATION - which is crucial to any recovery program from autism, since it is well known that autistics have difficulty with generalization.

Had aversives been the active ingredient of his treatment, Lovaas could not have obtained any recovered kid - only kids doing quite well some things without being able to transfer their knowledge from one context to another.

But, as long as he did obtain recovered kids - at least up to the age of 13 - then aversives did not play a crucial role in the treatment.

What Lovaas did prove though by applying aversives both on the control group and on the treated group, was to show that aversives do indeed accelerate learning in autistics, in some areas which belong to the early stages of skill acquisition.

What I am saying does not come from me or Lovaas. This is known since the 30s, the Skinner's era.

In fact, it was Skinner who advocated among the first for a public educational system that uses less punishment and more reward.

He did that based not on moral grounds, but as a behaviorist who understood that punishments block creative learning despite accelerating simple, unilateral learning.

jonathan said...

Marius, I am not copy pasting from Michelle Dawson's article. I am only citing the same study that she cites. You can read Lovaas 1987 which in fact does say that aversives were the active ingredient. Actually, if I remember correctly aversives were applied only to the experimental group and not to the controls and they did some sort of within subjects design of the lovaas 1987 study. It has been a bit since I read it, but that is my memory. Certainly there is no question at all that he did say so much in his own study that aversives were what made the difference in the recovery of the children. In jurisdictions where aversives are outlawed ABA is falsely advertised when the Lovaas (1987) study is used as an example.

Unknown said...


I respect your right to disagree with agencies such as the US Surgeon General, the NYSDOH, the MADSEC Autism Task Force etc but I ask you to tell me is the AAP simply wrong when it states:

The effectiveness of ABA-based intervention in ASDs has been well documented through 5 decades of research by using single-subject methodology21,25,27,28 and in controlled studies of comprehensive early intensive behavioral intervention programs in university and community settings.29–40 Children who receive early intensive behavioral treatment have been shown to make substantial, sustained gains in IQ, language, academic performance, and adaptive behavior as well as some measures of social behavior, and their outcomes have been significantly better than those of children in control groups.31–4

I know I have posted this citation before in response to your critiques of ABA but can you please tell me if you think the statements made by the AAP above are just plain wrong? If so why?

For example do you think the gains have not been made by the participants in the single subject studies for example?



Stephanie said...

All I know is that when the dog pissed on the carpet my brother beat the crap out of him and he never did it again.

My brother has his BS in psychology and explained his reasoning to me (paraphrased):

"It's behavioral therapy. When he does something wrong I punish him and then he never does it again. If I didn't punish him that would be rewarding him and he would just continue to do it. My friends don't understand behavioral therapy which is why their dogs still piss on the floor."

I realize this isn't the same thing as ABA, but I just think it's a funny story anyway.

jonathan said...

Hi Harold: I see you have not deleted your macro still.

jonathan said...

Harold, I have been over a lot of these things before. The single subject designs don't mean much to me because there is no control in the study, it is just the report of one subject.

I have dealt with these other issues in various places as has Michelle Dawson. Though there are many things on which I disagree with Michelle on, her analysis of some of the problems with the ABA research is one thing I agree with her on. The inequivalency in control and experimental groups which could not have come from anything resembling true random assignment, except with a probability of being struck by lightening in afew times in a year. Different pre and post assessment tests and not knowing which subject got which test. Also, the lack of adult outcomes which lovaas and company were given tax $$$ to study and publish but have still not published. Also the question of a Rett's person in the control group and possibly a second Rett's person in the control group based on the evidence from other studies. I have already gone through all of these things.

Unknown said...


I know you have been over ABA issues before and with this post ... you raised them again.

The AAP and other credible agencies do not agree with your rejection of studies such as single subject studies. They may mean nothing to you but they are evidence of ABA effectiveness in the eyes of Pediatric and other autism treatment professionals ... and in the eyes of parents.

Whether you arrived at your conclusions separately from Ms Dawson, who also shares the views of Dr. Mottron is irrelevant. All three of you reject the evidence based approach to autism interventions that is utilized by credible health and research agencies in the US. An evidence based approach calls for treatment based on the best evidence currently available and by that standard only ABA has prevailed over the past years.

I appreciate that anecdotal evidence is the lowest quality of evidence but having said that as a parent whose son did NOT receive 40 hours of Lovass Discrete Trial Training ABA from 2-5 I have still seen HUGE gains from the ABA he received sparsely from 2-5 and regularly in school for the past 5 years 8-13.

Prior to learning ABA techniques we had not way of communicating with each other. We were able, with ABA, to immediately begin teaching words instead of screams as a primary method of communication. Toilet training was gained through ABA principles. Self injury was reduced.

By all means hold out for a few more double blind studies of autistic children from 2 through adult hood. It can't happen and won't happen. It would be unethical to withhold for a lifetime evidence based treatment shown to the minds of the ovewhelming majority of involved professionals to provide substantial real life gains.

I make these comments with respect for your differing opinion and ... your different perspective.

Anonymous said...

Here's testimony from a Ph.D. that autism is treatable -


it's from the Autism Reasearch Institute here -

John Robison said...

Well, since you have brought me into this . . .

I do believe TMS has tremendous promise for neuroscience in general, and for certain aspects of autism in particular.

Dr. Casanova is a very smart researcher with some good ideas. Your argument about ABA is truly perhiperal to the thrust of his study, which is to use TMS with some minimal behavoral therapy to produce a result that would not be available otherwise.

If he does not succeed at this, another TMS researcher will. I believe that strongly.

I have seen that TMS in certain areas can improve mirror neuron function, which strikes right at the heart of my inability to read facial expressions, eyes, and body language.

You are right that I am opposed to medication that targets the whole brain and produces a host of side effects. That's the weakness TMS promises to address. We can use emerging therpies like TMS to reach specific areas of the brain, and enhance or inhibit them, while leaving other areas untouched.

If there was a technology that would help me minimize certain weaknesses, would I do it? Sure. Most people would. TMS may offer such promise.

I wish there were more in the public domain for folks like you to read. But the work is just too new. I've been very luck to see it from the inside; something that no doubt explains much of our differing viewpoints.

I am certainly all for acceptance of Aspergians as equal but different. Yet at the same time, I am open to ideas that allow us to minimize our weaknesses while building upon our strengths.

ABA isn't neuroscience, and I wish it had never been dragged into this because it just confuses matters.

And yes, I too am an opponent of ABA for high functioning Asperger people.

Anonymous said...

The point of the study isn't to focus on ABA. It simply uses ABA as part of the control, which is a common research practice.

Have written a brief summary of rTMS and autism on my website, if you're interested in reading: