Letter to the Canadian Senate

Honestly, if there was an award for procrastination, I'd win hands down. This should have been completed in early January. Oh well - my favourite quote is " I love deadlines - I particularly like the whooshing noise they make as they go by"

Dear Senator Eggleton and Members of the Standing Committee on Social Affairs, Science and Technology - Enquiry into Autism

I have been following the deliberations of your committee with a great deal of interest. The focus on such a relatively neglected area with the united resources of the range of world-class expertise available in Canada is, I think, an historic event. The developments here will almost certainly be followed around the world. In a parallel process, the introduction of Bill C 304 seeking to amend the Charter to mandate one treatment option for autism as medically necessary is also of interest. This controversial Bill is probably dead in the water for political reasons, mainly the setting of an unwieldy precedent in specifying a particular condition and a particular treatment in the Charter. However, the manifold reasons for rejecting this Bill say quite a lot about the current state of play in autism treatment and research and those reasons should also inform future directions.

The primary reason for rejecting Applied Behavioural Analysis (ABA or sometimes IBI) as a medically necessary treatment modality for autistic spectrum folk is that, prima facie, it is inappropriate for the majority of the spectrum. I note that Mr Murphy’s wording of the Bill specifies autistic spectrum rather than autistic disorder. A full three quarters of the spectrum have no cognitive disability. [1] So for 75% of the broad spectrum, it would be difficult to justify the use of a technique, which prides itself on teaching concepts and processes in tiny incremental steps, with this clientele. To illustrate, Professor Richard Borcherds of Cambridge University, winner of the Field Medal for Mathematics and Dr Vernon Smith Nobel Prize winner for Economics are both denizens of the spectrum. This cognitive deprivation curriculum is manifestly not the technique for the entire spectrum. These two represent an extreme to be sure, but it needs to be said, the continuous portrayal of the opposite picture of profound disability and mental retardation also represents an extreme as the true face of autism. Like the rest of humanity, the ‘truth’ is somewhere in the middle. Whether this method should be used for the remaining 25% is a separate and debatable question.

The second reason for rejecting this Bill is that the technique is in the class of ‘not proven’. ABA is essentially based on operant conditioning. It is yet to be proven that operant conditioning really works as an educational tool in any non-trivial sense with rather complicated organisms such as humans. It’s hard enough to apply it to less complicated animals under very controlled conditions.[2] I note that there seems to be some consensus among the committee and the professions that a web-based resource should be made available that specifies the evidence base of the various treatments for autism. Dr Bryson in her appearance before the Committee referred to operant conditioning as scientific fact and that is the assumption of a great many behavioural psychologists. Assumption, however, is not fact and it would be much appreciated if Dr Bryson could point to the references in mainstream psychology journals that attest to the validity of this conceptual framework. Surely if this field enjoys widespread acceptance, then there will be ample support for it beyond the sheltered workshop of the behaviourist academy. The problem appears to be that behaviorism as conceived in the analytical sense is something of an odd duck. Unlike all the Sciences and the majority of schools of thought in psychology, behaviour analysis has no theories, which is very helpful in avoiding having to pay the falsificationist piper but does not inspire confidence [3]. I refer the reader to Noam Chomsky’s famous review of ‘Verbal Behavior’ for the definitive critique of operant conditioning as applied to a complex human behaviour - language. It has never been improved on, or really answered. Alarmingly, though Verbal Behaviour (VB) programs are sprouting like mushrooms, Mark Sundberg, a well known figure in ABA circles said the following at a 2005 Behaviourist symposium.[4]

“There is much more research that needs to occur before it could be said that the necessary and sufficient empirical supports exist.”

The proliferation of VB programs points to the need for ABA/VB practitioners to have some kind of credentialing. It seems that practitioners in their VB programs have taken the roles usually assigned to speech language pathologists. Do they have the appropriate academic background and training for this? Similarly, aspects of ABA programs target the acquisition of fine and gross motor skills for various tasks. Do ABA practitioners have the necessary occupational therapy and physiotherapy backgrounds to deliver these therapies competently? The Committee has heard of one autistic who remained non verbal to middle age when it was recognised that a physical problem was a large factor in his lack of language acquisition.

A third reason for rejecting the mandatory use of ABA as a therapy is that the Canadian Government could find itself in the unusual position for a secular democracy, of prosetylising religious dogma. Fundamental to behaviour analysis is the idea that the external environment, either the current one or the evolutionary one is responsible for what people do. What are not relevant are things like creativity, self determination, personal responsibility, free will, volition, judgement, - all the attributes that make humans agents of their own destiny. These beliefs are articles of faith rather than Science and it’s easy to see why they are necessary to the concept of behaviour analysis. The major criticism Chomsky made about behaviourism was the lack of precision in definitions, a criticism levelled as recently as 2003 [5]. Terms such as stimulus, response and reinforcer meant whatever you wanted them to mean. Add that to the post hoc attribution of significance to events under analysis and the possibility that people could behave on a whim has to be rejected or your ‘scientific’ analysis becomes so much wishful thinking. Like all materialist philosophies, there seems to be an impetus to lop off the human bits that sit uneasily within the framework rather than adjust the framework to the inconvenience of a human liking for autonomy.

The fourth reason for rejecting this therapy is that on balance the results are a poor match for the claims made for it and for around forty hours a week at a cost of up to $60,000 a year, parents have the right to expect a lot. The trouble with a therapy framed around a naïve inductivism operating in a form of Jamesian Pragmatism is that it’s a long way from what a scientist would recognise as the scientific method.[6] Not surprisingly, when confronted with the scientific method the claims made for this therapy have usually fizzled. Every enthusiastic ABA proponent points unerringly to the classic 1987 Lovaas study, showing an incredible 47% success as measured by gains in IQ and survival in mainstream classes. The critiques of this study are many and the most comprehensive of them was a product out of British Columbia [7]. I refer the reader particularly to the questionable use of IQ as an outcome measure and the sex ratios of the experimental and control groups. This was not a true scientific study because the children were not randomly assigned to each group. There have been just a few studies like the Lovaas study, which had control groups, but only one that followed the scientific method in that there was random assignment of children to the experimental and control groups and the evaluators were truly blind to that placement.[8] The results were not so impressive. 13% had as good an outcome as Lovaas’ successes, but there were no differences on measures of language, socioemotional functioning or adaptive functioning. Is this even above what a placebo such as 40 hours worth of one on one play therapy would do? Possibly not.

Dr Marianne Ofner testified before the Committee that the Sallows and Graupner study was a ‘replication’ of the 1987 Lovaas study [9]. That claim is outrageous and incidentally, the main reason for this missive. In its primary aim, to prove beyond a shadow of a doubt that it was indeed possible to get the same results that Lovaas did, Sallows et al failed and spectacularly so. Their experimental group was outperformed by the control group 2:1. When that sort of thing happens the usual interpretation is that ABA has nothing to do with the outcomes since less therapy and less professionally supervised therapy has vastly better results. I’m sure the Canadian government would be delighted to cut back on the number of hours required and the qualifications and experience of the therapists. Less is more and we can save a bundle! I doubt the parents would be so happy. There is another Canadian study, which came to the same conclusion. [10] The therapy does not seem to be the relevant factor.

The final reason for not mandating this therapy is that it is not possible to ignore the very negative effects found with a sizeable minority of children. Looking at the Sallows and Graupner study, there were some children whose progress over four years seemed to be entirely backwards. It is therefore the untested hypothesis that ABA can be harmful. Those children may have regressed anyway, but that is an assumption.

The only constant in autism intervention, as is borne out for ABA and a motley collection of less reputable therapies, is the absolute disjunction between the actual results versus the claims made for the intervention. The reason, as is the case for any other field of research, is I think an a priori assumption - this should work, this ought to work and just to make sure we get a chance to do it, let’s also throw in the terrible cost to society of not giving intervention X a chance before the child turns 4 or 5 or 6. This is not an ethical way of doing things and not an efficient or cost-saving method of doing things either. There has to be a better way and surely it cannot be controversial to get the evidence before implementing the therapy. Autistic children cannot continue to be the research fodder for successive waves of academics and therapists. They are also not the designated recipients for one therapy.

There are other promising avenues out there. Aldred et al showed the excellent results that can be obtained from manipulating environmental variables, in this case the parent not the autistic child and no operant conditioning required.[12] It is apparent that autism research has recently joined the cognitive revolution, about forty years behind everybody else but better late than never. Mottron et al of the University of Montreal are probably the first to objectively look at what autism is, what makes it tick without reflexively trying to stamp it out.[13] Surely if there are to be interventions they should be based on what Mottron et al have turned up. Sallows and Graupner may not have been able to replicate Lovaas but they used the vast amount of data they acquired to try to find out what factors in the child led to what outcomes. This is the natural starting position – what is autism, before deciding what should be done about it.

Brigitte Harrisson testified before your Committee that ABA could be good if it was adjusted to an autistic way of thinking “if there was a national will to do so.” For this to happen there would have to be some sort of gathering together of therapy providers and the sorts of researchers who know something about autistic thinking, including those who know all about autistic thinking without needing a translator - autistic adults. It is fortunate that Canada has such ready access to the ideal candidate. Ms Michelle Dawson is the person I would like to see on any group making decisions about future directions in autism funding and research. With due deference to Dr Szatmari and Dr Fombonne, it’s likely across the very broad array of sub-specialities constituting autism research , Ms Dawson knows more about the current state of play than any single person anywhere in the world. She is herself a researcher and I note that she is principal author of a paper titled “The level and nature of autistic intelligence” to be published in Psychological Science. No doubt some of the researchers you have heard from would be ecstatic to have their work published in such a prestigious journal. Knowledge aside, I trust her objectivity. In the internecine squabbles that characterise the redoubts and battlefields among the various fiefdoms in psychology, it would be positively advantageous to have someone there who would impartially temper the enthusiasms of a Marianne Ofner, or confront the touters of psychological ‘instruments’ of dubious validity. Above all it is absolutely necessary that there be as little translation of the autistic perspective as possible. Ms Kathleen Seidel of www.neurodiversity.com described this problem as follows:

“I think that one major roadblock that many NT educators and service providers have to get past, is the fact that they tend to be "people people" by nature. Many go into the business out of a desire to do good and use their skills productively and connect with others in a way that has meaning to them. When "people people" find themselves in an interaction with someone who isn't the same kind of "people person," the default response is to wonder what that person's problem is. Such a default response, if not acknowledged as such and examined critically, is a source of disrespect.”

I suppose I am being rather forward writing to a Senate Committee, which is not of my Senate. I will, however, have a ringside seat in Ottawa to further developments a little later in the year and for a number of years. This issue transcends both political and national boundaries. To paraphrase Lincoln, the world will note what you do here.

Yours sincerely

1. Morton Ann Gernsbacher, Michelle Dawson, and H. Hill Goldsmith, Three reasons not to believe in an autism epidemic, Current Directions in Psychological Science, Volume 14, Number 2, April 2005, pp. 55-58(4)

2. Timberlake, William, Is The Operant Contingency Enough For A Science Of Purposive Behavior? Behavior and Philosophy, 2004

3.Noam Chomsky, A Review of B. F. Skinner's Verbal Behavior, Language, 35, No. 1 (1959), 26-58.

4. http://www.abainternational.org/ConvArchive/conv2005/program.asp Paper # 383 Symposium Paper

5. Palmer, Daniel K, Specifying Psychology's Observable Units: Toward An Integration Of Kantor's Behavior Segment, Skinner's Operant, And Lee's Deed, Behavior and Philosophy, 2003

6. Steven C. Hayes, Nicholas M. Berens, Why Relational Frame Theory Alters the

Relationship between Basic and Applied Behavioral, International Journal of Psychology and Psychological Therapy 2004, Vol. 4, Nº 2, pp. 341-353

7. Centre for Health Services and Policy Research, British Columbia Office of Health Technology Assessment, Autism and Lovaas treatment: A systematic review of

effectiveness evidence, July 2000

8. Smith, T., Groen, A. D., & Wynn, J. W. (2000). Randomized trial of intensive early intervention for children with pervasive developmental disorder. American Journal on Mental Retardation, 105, 269-285.

9. Glen O. Sallows and Tamlynn D. Graupner, Wisconsin Early Autism Project (Madison) Intensive Behavioral Treatment for Children With Autism: Four-Year Outcome and Predictors Volume 110, Number 6: 417–438 Z November 2005 American Journal On Mental Retardation

10. Eaves LC, Ho HH. The very early identification of autism: outcome to age 4 1/2-5, .J Autism Dev Disord. 2004 Aug;34(4):367-78.