Monday, January 26, 2009

The origins of ABA-based autism interventions
When it comes to behaviour analytic approaches to autism, the paper that started it all, and set the tone for what was to come, was Fuller (1949), the first published report of operant conditioning in a human being.

Fuller opens his paper by noting that classical conditioning had been tested in "normal and subnormal" humans. For example, it had been found that "subnormal" children formed conditioned responses to shock faster than "normal" children. But so far, operant conditioning had only been tested in animals ("infra-humans"). Fuller was given permission--no mention of by whom--to experiment on a person he described as an 18 year old "vegetative idiot" incarcerated in a "feeble-minded institution."

This person, referred to as "S," is reported never to move his trunk or legs, though no reason is given for this. According to Fuller, S had limited movement of his head, shoulders and arms, but couldn't roll over or change his position from where he was left lying on his back all day. S was never given solid food; liquids and semi-solids were "stuffed into his mouth," Fuller reports, while S coughed and choked. While Fuller writes that S never made any sounds, "in the course of the experiment vocalizations were heard."

Fuller's experiment started by depriving S of food for 15 hours. A syringe full of warm milk-sugar solution, which was squirted into the corner of S's mouth, was the reinforcer. A bit of this solution was given to S when he raised his right arm.

Eventually, S was conditioned such that after being deprived of food for five hours, he raised his right arm 19 times in 16 minutes. After this he fell asleep. The next morning, he raised his right arm and opened his mouth simultaneously. And while S had earlier also raised his left arm sometimes, by this point, he only raised his right arm.

Fuller declared success at this point, and then extinguished this response by removing the food reinforcer. The extinction procedure took 70 minutes, at which point S stopped raising his right arm. This completed the experiment.

Fuller wrote about S:

"An interesting feature of this study is the example it affords of phylogenetic overlap. While of normal human parentage, this organism was, behaviorally speaking, considerably lower in the scale than the majority of infra-human organisms used in conditioning experiments--dogs, rats, cats."
The physicians in the institution in which S had been an inmate (this word is used) for a year had reported to Fuller that S had not, in all his 18 years, learned anything at all. Fuller simply accepted this as true, even though he had no relevant information about S's history and minimal information about S's present. This allowed Fuller, and many behaviour analysts to follow, to be very impressed with what Fuller accomplished in conditioning S. After all, Fuller is reported to have trained a vegetative organism to move, a stunning and unprecedented achievement.

Along these lines, here is Fuller's conclusion:

"For years, many psychologists have experimented exclusively with infra-human [subjects], and they have expressed a preference for the simple, less variable behavior of the lower organisms in the laboratory. Perhaps by beginning at the bottom of the human scale the transfer from rat to man can be effected."
Fuller's "classic" paper, a landmark in the history of ABA, resonates with the much later words of Ivar Lovaas and his colleagues, about autistic human beings:

"You see, you start pretty much from scratch when you work with an autistic child. You have a person in the physical sense—they have hair, a nose and a mouth—but they are not people in the psychological sense. One way to look at the job of helping autistic kids is to see it as a matter of constructing a person. You have the raw materials, but you have to build the person." (from an interview with Chance, 1974)
"In any case, what one usually sees when first meeting an autistic child who is 2, 3, or even 10 years of age is a child who has all the external physical characteristics of a normal child—that is, he has hair, and he has eyes and he has a nose, and he may be dressed in a shirt and trousers—but who really has no behaviors that one can single out as distinctively ‘human’. The major job then, for a therapist—whether he's behaviorally oriented or not—would seem to be a very intriguing and significant one, namely, the creation or construction of a truly human behavioral repertoire where none exists." (Lovaas & Newsom, 1976)
"To use another analogy, at the beginning of treatment, the children may be regarded as having close to a tabula rasa. In this sense they can be considered very young persons, as persons with little or no experience, presenting the teacher with the task of building a person where little had existed before." (Lovaas & Smith, 1989)
"Instead, the fascinating part for me was to observe persons with eyes and ears, teeth and toenails, walking around yet presenting few of the behaviors one would call social or human. Now, I had the chance to build language and other social and intellectual behaviors where none had existed, a good test of how much help a learning-based approach could offer." (Lovaas, 1993)
The currently predominant autism advocacy policy position that autistics can't learn, develop, communicate, progress, etc., or become human, except via ABA programs, reminds me of assumptions behaviour analysts have made about the nature of Fuller's S. They just know, like Fuller did, that S had never learned anything at all until Fuller came along. Fuller needed no evidence for this, or any information about S's past. He just knew there were subhumans in human form, including vegetative human organisms who were lower in the scheme of things than rats (and were treated as such). So all he had to do was to never, ever question any of this, and then go out and find one of these vegetative organisms and experiment on him. And so Fuller became famous, just like Dr Lovaas did later for "building" human behaviours in autistic children who he claimed had none at all, for "building" a person where he claimed none existed.


Chance, P. (1974). "After you hit a child, you can't just get up and leave him; you are hooked to that kid": A conversation with O. Ivar Lovaas about self-mutilating children and why their parents make it worse. Psychology Today, 7, 76-84.

Fuller, P. R. (1949). Operant conditioning of a vegetative human organism. American Journal of Psychology, 62, 587-590

Lovaas, O.I. (1993). The development of a treatment-research project for developmentally disabled and autistic children. Journal of Applied Behavior Analysis, 26, 617-30.

Lovaas, O.I., & Newsom, C.D. (1976). Behavior modification with psychotic children. In H. Leiteberg (ed.), Handbook of Behavior Modification and Behavior Therapy. Englewoood Cliffs, NJ: Prentice-Hall.

Lovaas, O.I., & Smith, T. (1989). A comprehensive behavior theory of autistic children: Paradigm for research and treatment. Journal of Behavior Therapy and Experimental Psychiatry, 20, 17-29.


Unknown said...

"The currently predominant autism advocacy policy position that autistics can't learn, develop, communicate, progress, etc., or become human, except via ABA programs"

Thank you for, once again, undermining your own credibility with silly, false comments like that.

By the way, what is "an autistic"?

Jannalou said...

Okay, so what is the actual autism advocacy policy position? Because from where I sit, Michelle's summation is pretty accurate.

Michelle Dawson said...

In response to Mr Doherty, in addition to the sources I provided, see the Auton and Wynberg trial decisions, both of which are fully supported by Canada's leading autism advocates, including Mr Doherty, who has also fully supported the work of Ivar Lovaas.

Mr Doherty, a lawyer, may see the jurisprudence (involving Canada's highest law, which is a human rights law) and statements by Canada's law makers (which Mr Doherty has quoted himself) as "silly" but autistics have to live the consequences of this, every day.

See more quotes from famous autism advocates here. Then there's the more recent claim that autistic children who aren't in ABA programs are rotting.

In my view, Mr Doherty's question, "what is "an autistic"?" speaks for itself.

Alyric said...

Could be just me, but Fuller's view of the grades of humanity is identical to that of Peter Singer. Or, there is nothing original about Singer's view.

Michelle Dawson said...

Having thought about it, in further response to Mr Doherty, I'm going to pull a quote from one of the sources I provided in the original post, which of course Mr Doherty read before he made his comment.

In this quote, "treatment" refers to ABA-based autism interventions starting early in life, and "they" refers to autistic children:

"Without treatment, they are deprived of the chance to learn how to communicate with the world and learn from it, and of the means to avoid injuries, thus engaging both their psychological and physical integrity."

This is the kind of statement that Mr Doherty, in his comment above, has called "silly" and "false" and so on. It is from paragraph 63 of the Auton parents' written argument at the Supreme Court of Canada.

Jannalou said...

It's really very strange (and somewhat disturbing) to read a statement from the ABA-only camp that echoes the declarations of Greenspan in The Growth of the Mind, which I am finally back to reading. Greenspan doesn't say that autistic children need ABA, of course, as he is a developmentalist, but he does use the same basic argument (i.e., they will be denied the opportunity to develop certain skills without the treatment he espouses).

Larry Arnold PhD FRSA said...

Come off it the Spanish Inquisition were there first!

MalchowMama said...

Ms. Dawson, I have been reading some of your very well-written work and am grateful to you for providing this perspective. I believe that my 28 month old son, who does not use words, is on the autism spectrum. While I wish to help him communicate effectively in whatever way suits him, my immediate response upon learning the basics of ABA/VB was repulsion and dread. Due to the overwhelming push in favour of ABA, however, I had begun to become resigned to it. As we live in Germany, where the ABA industry does not seem to have taken hold as strongly as it has in English-speaking countries, I was worried there would be little ABA help available, especially since we have no money for this very expensive therapy. I felt I was already failing Sam. Our first appointment with a speech therapist is scheduled for tomorrow, and I am now hoping that she does not take an ABA approach. I need to read more of your blogs (there is so much!), but wonder also if you would consider responding to me personally if you have the time. My email address is I have no interest whatsoever in changing Sam, all I want is to be able to communicate with him. Thank you again for this under-represented and important perspective.

Unknown said...

Ms Dawson here is what I say about ABA for autism. And yes they are not my words. They are the words of a responsible authority which has reviewed the available research objectively:

"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

American Academy of Pediatrics, Management of Children with Autism Spectrum Disorders"

Unknown said...

Ms Dawson

I note that you did not answer my question asking you what you mean by an "autistic".

Is that because you do not want to answer it or because you can not answer it? Is it just a term you made up yourself?

Michelle Dawson said...

In belated response to Janna, the well-known highly-marketed autism interventions (ABA in all its forms, RDI, DIR, Son-Rise...) have all been promoted in the same way. The future of autistics who have the misfortune of not receiving these interventions is painted as grim, if not horrific. The horrific future of autistics who have not received ABA-based interventions starting early in life (that's most autistics in Canada--too late for us) was front and center at the Auton Supreme Court of Canada hearings.

In response to Mr Rex, I agree that forceful (or something) objections to accurate reporting from primary sources are all the rage, in this current era of autism advocacy.

In response to MalchowMama, re communication, you might be interested in this paper. In my view, everyone changes through development and learning and so on, but I'm concerned when it's assumed that autistic traits and abilities, the way we develop and learn, and so on, are simply wrong. I'm also concerned about the poor quality of research in the area of ABA-based autism interventions.

Michelle Dawson said...

In response to Mr Doherty's 2 messages:

1. Here is a response to Mr Doherty's boilerplate re ABA-based interventions, and here is a bit more. In case this is unclear, I didn't make up the term "systematic review."

2. No, I didn't make up the term "autistic." I suggest reading the relevant literature for its origins. Also, see my first response to this query, above.

jypsy said...

Here are Mr. Doherty's words on ABA, as he speaks them

Anonymous said...

listening to the youtube video refered by Jypsy, Mr. Doherty use the words "An autistic child" several times; I find it funny that he is asking what is an autistic...


Anonymous said...

FYI: The noun "autistic" is defined in the Merriam-Webster online medical dictionary.

Anonymous said...
This comment has been removed by a blog administrator.
jypsy said...

"FYI: The noun "autistic" is defined in the Merriam-Webster online medical dictionary."
Main Entry: 2autistic
Function: noun
: an individual affected with autism

Anonymous said...

I have just stumbled upon this chain of comments. Forgive me if 1) my thoughts have already been addressed, or 2) I'm missing the point ... in that there were several diverging conversations within these postings. I'm curious & concerned about the usage of "ABA" as an intervention. ABA is not an intervention. Applied Behavior Analysis is the science; the application (applied) of behavior principles to socially significant (every day) behavior in which interventions are continually analyzed providing critical information to the interventionist for further treatment work. Its kind of in error when people say, "I'm doing ABA". What they are doing is applying behavior principles (reinforcement, extinction, differential reinforcement, etc.) to target behaviors. This could be skill based behaviors that are targeted for increase. This could be problem behaviors in which the family and the interventionist are decreasing.

Evidenced Based Practices are so important for parents to understand. Unfortunately, parents of children with Autism are most susceptible to the "newest" and "greatest" treatments that seem to pop up every few months or so. Does anyone blame them? They want the best for their child ... an accusation that most every parent is guilty of. Behavioral scientists are charged with determining if the claims of any given method or intervention are valid. Examples of the scientific method in action can be seen with statements made by "facilitated communication", "Secretin therapy", "Chelation therapy", "Auditory Integration Therapy" ... the list goes on and on.

Intervention methods such as Discrete Trial Teaching, Pivotal Response Training, PECS, Picture Activity Schedules, Functional Communication Training are all evidenced based practices that utilize the principles of behavior.

I hope this explanation helps to understand that saying things like, "ABA is the only way to go" is somewhat misleading. Effective treatment programs will utilize several different methods. For example, a therapist might utilize Discrete Trial Teaching methods to teach stimulus discrimination skills, Verbal analysis for language development, and Picture Activity Schedules to help the child organize his day. All of these methods take from the science of Behavior referred to as Applied Behavior Analysis.

With ENORMOUS respect for parents of children with Autism!

Michelle Dawson said...

Catching up... the deleted message from Konnie was (badly place) spam from an autism service provider.

Many thanks to Anonymous and jypsy for looking in the dictionary...

Michelle Dawson said...

Pete Molino is a BCBA and ABA service provider who is promoting the services he provides. I wasn't sure whether to leave his advertisement up here, but there it is, speaking for itself.

From what he writes, Mr Molino has no use for RCTs (or in some cases, any controlled trials at all) when assessing whether ABA-based autism interventions are evidence-based. Like Mr Doherty (see my responses to his boilerplate, above; here and here they are again, and I'll throw in some ethics), he is denying autistics the scientific standards that are recognized as essential for the well-being of nonautistics.

Mr Molino quotes from the current edition of the major ABA textbook (Cooper et al., 2007). This highly influential textbook recommends and attempts to justify the use of many punishment procedures in ABA-based autism interventions, including strong aversives like electric shock.

Likewise, the ABA literature in the area of functional communication training includes the use of strong aversive procedures such as basket holds (e.g., Hagopian et al., 1998; this study and others similar are published in a journal Mr Molino recommends). Basket holds have been fatal.

Mr Molino's "Recommended Websites" include the National Autism Association, the Autism Research Institute (home of DAN!), and the Autism Collaboration, a group of organizations which agree that there has been an epidemic of autism caused at least in part by vaccines.

RAJ said...

The most effective treatments are behavioral interventions.

"Although Autism Spectrum Disorders (ASD) are generally assumed to be lifelong, we review evidence that between 3% and 25% of children reportedly lose their ASD diagnosis and enter the normal range of cognitive, adaptive and social skills. Predictors of recovery include relatively high intelligence, receptive language, verbal and motor imitation, and motor development, but not overall symptom severity. Earlier age of diagnosis and treatment, and a diagnosis of Pervasive Developmental Disorder-Not Otherwise Specified are also favorable signs. The presence of seizures, mental retardation and genetic syndromes are unfavorable signs, whereas head growth does not predict outcome. Controlled studies that report the most recovery came about after the use of behavioral techniques. Residual vulnerabilities affect higher-order communication and attention. Tics, depression and phobias are frequent residual co-morbidities after recovery. Possible mechanisms of recovery include: normalizing input by forcing attention outward or enriching the environment; promoting the reinforcement value of social stimuli; preventing interfering behaviors; mass practice of weak skills; reducing stress and stabilizing arousal. Improving nutrition and sleep quality is non-specifically beneficial".

Michelle Dawson said...

In responses to RAJ, he has scanned in the abstract from a single narrative (subjective) review (Helt et al., 2008). Helt et al. (2008) assume that an "optimal" outcome in autism is a nonautistic person who has lost all autistic traits and abilities.

The authors' evidence from a controlled trial re "recovery" comes from a study where very young autistic children were systematically hit (Lovaas, 1987; McEachin et al., 1993) and in fact hit "hard" (Leaf & McEachin, 2008). While the authors mention the one small-N RCT (where the intended comparison between randomized groups was actually carried out) in the 47-year history of research into ABA-based autism interventions, they don't mention its findings (and fail to cite it correctly).

For something more empirical, see Turner and Stone (2007) re the stability of early diagnosis and whether this is related to any kind or amount of intervention. See Magiati et al. (2007) for an ABA controlled trial conducted by researchers who have no relevant conflicts of interest.

See, again, my responses Mr Doherty's boilerplate, which are here and here. And in my view, autistics deserve the kind of ethical consideration that automatically protects and benefits nonautistics; in this direction, see information about published reports of ABA success stories here.

RAJ said...


Mind reading again? I have followed Deborah Fein's research for a number of years.

If you have a spare hour you might want to click on her lecture at this page:
Behvioral interventions, speech and language therapy, even physical and occupational promote brain reorganization. 'Autism' is a disorder of early brain development. Or on't you think 'autism' involves structural abnormalities in brain development?

Michelle Dawson said...

In response to RAJ, there is a lot of evidence that autistic brains are allocated atypically and function atypically. We've shown that there are measurable behavioural differences related to atypical perception in autism starting very early in development (Mottron et al., 2007). This finding has been replicated by another group in an even younger sample (Ozonoff et al., 2008).

I've read Dr Fein's peer-reviewed work and have seen her speak at IMFAR.

KeithABA said...

"likewise, the ABA literature in the area of functional communication training includes the use of strong aversive procedures such as basket holds (e.g., Hagopian et al., 1998; this study and others similar are published in a journal Mr Molino recommends). Basket holds have been fatal."

So, when it comes to external vailidity and generalization, you give zero credit to procedures researched by within subject design. However, when you find one article in which punishment was used in conjunction with FCT, you overgeneralize and make the statement that FCT includes punishment, like basket holds!

Before making the accusation that FCT always includes punishment, you should conduct a meta-analysis of all the JABA articles and give a ratio of the number of articles which document the use of FCT with and without punishment. Or for that matter, the number of articles published in JABA which are reinforcement based vs punishment based.

You seem to always leave out the work of Murray Sidman, Glen Latham, BF Skinner, and many other behavior analysts who devoted their entire life to informing people to avoid punishment procedures.

KeithABA said...

Correction to above:

you overgeneralize and make the statement that ALL FCT includes punishment, like basket holds!

Michelle Dawson said...

In response to Keith ABA, here is what I wrote:

"Likewise, the ABA literature in the area of functional communication training includes the use of strong aversive procedures such as basket holds (e.g., Hagopian et al., 1998; this study and others similar are published in a journal Mr Molino recommends). Basket holds have been fatal."

Here is Keith ABA's remarkably false interpretation: "you overgeneralize and make the statement that ALL FCT includes punishment, like basket holds."

The fact that I cited one article as an example does not mean it is the only one. I also generally disagree with Keith ABA's priorities on this issue.

KeithABA said...

I don't know how to hyperlink, but taken from here

Michelle: ...there's those later studies showing that punishment was necessary to make "this procedure"(in reference to FCT) work.

Conduct a meta analysis to determine how many studies utilizing FCT invovled a punishment contingecy and how many did not.

Conduct a meta analysis to determine how many articles utilized positive reinforcement based procedures vs punishment contingencies in JABA.

Michelle Dawson said...

In response to Keith ABA, that is not the purpose of meta-analysis. Maybe he means systematic review.

I forgot to add, above, that Hagopian et al. (1998) is highlighted as an example of good practice in the unrefereed MADSEC report.

KeithABA said...

Yes you are right, I meant systematic review, since it is not reviewing the hypothesis. However, a meta analysis could be conducted to determine if FCT is effective without punishment.

Perhaps you could discuss in another topic your views on what is acceptable as punishment and what is not. It is a seperate topic from effectiveness, and an argument that you have pointed out against ABA in the past.

So far I think it is obvious that you are opposed to a basket hold, which I agree with 100%. In fact I will not advise any physical intervention (laying hands on someone else) unless the person has caused injury to themselves or other's in the past. Additionally PCM is what is used, which excludes the use of a basket hold.

Punishment exists within society. Every day we are subject to punishment contingencies when we drive, where we work, through the judicial system, and through our social relationships. Punishment is the top two startegies used by parents (1. Time out 2. Spanking).

Do you beleive that any form of punishment is o.k. for anyone?

Michelle Dawson said...

Outside of autism research and outside of the area of ABA, there is a consensus that meta-analysis should not be used with poor quality data. This is because poor quality studies have been found to be a major source of bias in meta-analysis.

Several papers in JABA (the flagship ABA journal) report the use of basket holds, without reporting their serious dangers. In my view, this is irresponsible, as was the MADSEC report citing a study using basket holds as an example of good practice. I have never seen anyone else raise concerns about this.

My scientific and ethical concerns with ABA-based autism interventions are not in any way limited to the use of punishment procedures (including strong aversives). For example, I've written about reported consequences of the amount of extrinsic positive reinforcement required in supposedly "all-positive" ABA programs.

Also, not everyone sees the entire world in behaviour analytic terms. Some of us (regardless that we may be informed about behaviour analytic terminology, including the behaviour analytic meaning of "punishment") consider that there are things like cognitive processes and so on.

By the way, I've read about an autistic boy who started in ABA at a late age and who thought it was great because ABA meant winning ABA prizes all the time (there was an ABA prize cabinet, full of ABA prizes). He was sure he was never, ever punished for anything--all he did was win ABA prizes when he did the right thing, and who could argue with that? Who wouldn't want to win prizes all the time? But he was in a strict token economy with steep response costs--and response cost is punishment.

KeithABA said...

I think there is a misconception about cognitive processes. Radical behaviorism seperated from other forms because of 2 primary factors. 1. Radical behaviorists included the behavior of the scientist being subject to contingencies of reinforcement.
2. The acceptance of private events as behavior. (cognition)

ABA does not teach that cognition doesn't matter, or isn't there, but rather it makes a poor dependent variable since you can't gain interobserver agreement on what someone is thinking or feeling. This is why you will see recent articles in JABA studying, "Increasing Indices of Happiness." Things like positive comments, smiling, laughing.

I know that the use of rewards can be contriversal, but I alos think that many people disregard things like praise, attention, tickles, hugs, or singing songs together as a reinforcer. ABA programs do this stuff as well. Sometimes contingently as a reinforcer, sometimes non-contingently just to play for fun. It astonishes me when I hear comments from speech therapists or occupation therapists such as, "I didn't know ABA could be like that!"

Response cost generally is meant to be a punishment procedure, but it is something that does generalize into real life. If I don't come in to work, I have to use Paid Time Off, which I could have traded at the end of the year for money. I received a speeding ticket, which took away some of my money because I wasn't paying attention to how fast I was going.

I really would like to hear your thoughts on Murray Sidman's book "Coercion and it's fallout." I think it is one of the most important works in behaviorist literature. That and Glenn Latham's book "The Power of Positive Parenting." Neither of these books can serve as an argument in the realm of effectiveness, however, when it comes to ethics I think these books become very relevant.

Michelle Dawson said...

Murray Sidman is recognized as having been, up through to the current decade, a major influence on the NECC. See this and this and this.

See a also a recent media report about practices used with autistic children at the NECC, here. I'm not sure "coercion" is a useful word, in many situations. It's better to look at what actually happens.

Also, I have (for years now) provided an important but forgotten example of how ethics can enter into decisions about ABA-based interventions. See this. Unfortunately, this kind of ethical consideration has not in any way been extended to autistics, and leading behaviour analysts and autism advocates have deemed as reprehensible any suggestion that it should be.

By the way, if what Keith ABA writes about cognitive processes were true (that they can't be measured), then there would be no such thing as cognitive science, much less highly replicated results that go back decades. For a short, informal autism-related primer, see this.

For some reported (in legal proceedings, and/or by autism advocates) consequences of "all-positive" ABA programs in autism, see this.

Ophidian said...

Dear Michelle,

I have been reading your blog for quite a while with great interest and really appreciate your reviews on autism as well as your sharp criticisms. Although not a researcher in autism in any way (my research belongs to a totally different field of science), I would really love to understand how substantiated are the claims by various therapies.

One of the problems for people, who are not doing medical research is devilishly hard copyrighted access to the original research articles. Some researches, like Gernsbacher, well maintain their personal websites and make their articles freely accessible for public. Others, unfortunately, do not. To remedy this situation, I thought to collect and put together all the main research articles related to therapeutic interventions in autism and and their outcomes for free web access. This way one would not have to trust other people's reviews/opinions but rather make one's own judgments. The first set is based on the recently published review article, which you probably know about, by Svein Eikeseth (Research in developmental disabilities 2009;30(1):158-78.) The review covers 25-30 outcome articles on ABA, TEACCH and Colorado Health Science Program. I wonder if you could help me complete this collection.

Again, the articles will be freely available for everyone and are currently deposited here:

If you check the site, you will see what's missing.

Michelle Dawson said...

In response to Ophidian, for a list of papers, I suggest Ospina et al. (2008; this paper is open access) as a source. Ospina et al. (2008) has problems but is much better than Eikeseth (2009) at locating studies.

Also, there's the problem of providing Smith, Groen and Wynn (2000) while failing to provide the authors' errata (Smith, Groen and Wynn, 2000, 2001), which are pertinent to outcomes and were ignored by Eikeseth (2009).

But mostly I suggest that Ophidian get permission from authors, if Ophidian is going to repost non-open-access papers on Ophidian's own site, as opposed to linking to existing, available pdfs.

See the conditions under which (many) scientists provide free pdfs of their work here.

Even if Ophidian has no respect for copyright (which is Ophidian's own business), there is the just plain courtesy of asking researchers whether you can re-post entire (non-open-access) papers of theirs on your own site and for your own purposes.

I'm a big fan of open access, but large parts of the world don't work that way yet. Also, there are different definitions of open access, some more restrictive than others, and even the most open of open access comes with conditions (e.g., that the open access paper be available in a central location like PMC).

By the way, the html version of Lovaas (1987) used by Ophidian, who borrowed it from FEAT, has (at a glance) at least one minor error compared to the original.

Ophidian said...


Thanks for the reference to Ospina et al, 2008. The total number of references (171 in Ospina, 2008 vs. 76 in Eikeseth, 2009) suggests it is a more comprehensive review of interventions. In the subdomain of the ABA therapies it is not great, though. The review is missing important references to a number of normally cited ABA studies
(Anderson, 1987, Bibby, 2002, Eikeseth, 2007) as well as for LEAP (Harris, 1990)
and TEACCH (Lord, 1994). It seems also very weak in the critical analysis of the therapies. The take-home message of the review is almost laughable: "While this review suggests that Lovaas may improve some core symptoms of ASD compared to special
education, these findings are based on pooling of a few, methodologically weak studies with few participants and relatively
short-term follow-up. As no definitive behavioral or developmental intervention improves all symptoms for all individuals with ASD, it is recommended that clinical management be guided by individual needs and availability of resources."

Are those the only conclusions the authors could derive after analyzing 101 original studies? (It reads as "the rest of the studies are of such poor quality that they are even worth mentioning".)

You write: "Also, there's the problem of providing Smith, Groen and Wynn (2000) while failing to provide the authors' errata (Smith, Groen and Wynn, 2000, 2001)..."

It is also true for Ospina, 2008. Most of the reviews on interventions do not cite the errata. The only critical review I found, which does cite the errata, was the
review by Shea, 2004.

You write: "By the way, the html version of Lovaas (1987) used by Ophidian, who borrowed it from FEAT, has (at a glance) at least one minor error compared to the original."

Thanks for noticing it. I replaced it with the original article in PDF . What was the error, by the way?

Michelle Dawson said...

In response to Ophidian, Ospina et al. (2008) provide criteria for studies they are interested in, a list of excluded studies, and extensive information about how and when their search was conducted.

This does not mean the authors did not also err in including/excluding/classifying studies.

Ospina et al. (2008) has many weaknesses, including failing to describe and cite Smith, Groen and Wynn (2000, 2001) correctly, and treating the ABA literature as higher quality than it actually is.

It looks like Ophidian would regard many systematic reviews that are conducted properly as having ridiculous conclusions.

Harris et al. (1990) is not a LEAP study, and if Lord (1994) is a TEACCH trial, I've never heard of it.

I'll leave Ophidian to locate the error in the FEAT-generated (back in 2003 it was on more than one FEAT site) html version of Lovaas (1987). It's instantly obvious if you read it.

Memory boost in five steps said...

I have been looking for this valued information with references for years. Thank you so much Michele. I work with a lot of families who use ABA or IBI but have absolutely no idea of the basic root. I am a humanistic therapist and I cannot comprehend conditioning as a defendable approach to mental disorder or neurological disorder.

I will use those quotes and your research to educate the parents I work with. They are all loving and reasonable parents who simply implemented what was available and funded, either in Canada or Australia.

Thanks again and again

Claudie Gordon-Pomaes

Ophidian said...


"I have been looking for this valued information with references for years. Thank you so much Michele."

In fact, the list of references to the original articles related to the ABA therapy is available in many places. It does not help much because the articles themselves are not freely available. The researchers in the field (both behaviorists and their critics) made no effort whatsoever to make the articles a public domain of knowledge, which they should be by all means. The same, unfortunately, applies to the author of this blog.

Limba said...

I have no interest whatsoever in changing Sam, all I want is to be able to communicate with him. Thank you again for this under-represented and important perspective.

Dave said...

With your permission to post on your blog, Michelle, my number one all time teacher in Behavioral Ethics, I give you my latest post to enjoy:

An actually autistic man defines and explains Applied Behavior Analysis (Analy$i$)(ABA).