Sunday, June 21, 2009

Autistics Faster at Problem Solving? Low Functioning Autism Subjects Excluded from Soulières Mottron Dawson Study

The mainstream media is buzzing about a study by lead author Isabelle Soulières, a post-doctoral fellow at Harvard University who completed an experiment at the Université de Montréal which, according to a U de M press release, purports to find that "autistics are up to 40 percent faster at problem-solving than non-autistics".

The press release states that the study participants "were asked to complete patterns in the Raven's Standard Progressive Matrices (RSPM) – a test that measures hypothesis-testing, problem-solving and learning skills". An image from the test is provided in the U de M press release:

The U de M press release, which by definition is intended for consumption by media outlets, for dissemination to the general public, refers to "autistics" and "autism" but not once to the specific diagnoses of the participants or to their functioning level. The study report itself, which requires purchase by non-subscribers and takes more effort to locate than the press release, also refers generally to autism and autistics. The study even uses those terms to refer to footnoted articles which in their titles specify that the subject of the article is high functioning "autistics".

The study report, and the press release, insist on referring to "autistics " and "autism" as though the study referred to all persons with autism disorders. It is clear from the characteristics of the participants as identified in the study itself however that they were all High Functioning Autistics. Low Functioning Autistics were excluded from the study:

TABLE I. Participant characteristics

Based on the IQ information provided in the study report there were no persons with Low Functioning Autistic Disorder included in this study. It seems strange that a study which includes Dr. Laurent Mottron as a co-author would not specify their functioning level -Low Functioning Autism or High Functioning Autism. Dr. Mottron has published many study reports over the years in which he took great care to identify the participants as Low Functioning or High Functioning, Autistic Disorder, Aspergers and, in some cases, Autistic Savants. His studies focused almost exclusively on persons with Aspergers and High Functioning Autism as any check on Google Scholar can confirm. This paper does make it clear that persons with Aspergers were excluded from the study but offers no mention of the exclusion of persons with Low Functioning Autistic Disorder below the 100 IQ range. (The paper does not actually refer to the specific diagnosis of Autistic Disorder referring instead to "autism").

Dr. Mottron has tended to exclude low functioning autism subjects from his studies but he has in the past made it clear that his study participants were high functioning. This study by Dr. Mottron and his colleagues again focuses on high functioning autistic persons but obscures that fact for general media and public consumption.

Perhaps someday Dr. Mottron will cease excluding low functioning autistic subjects from his studies.

In the meantime it is difficult to see why a study which excludes low functioning autism subjects should be used to make broad statements about the abilities of autistic persons on all points of the autism spectrum of disorders.

In the meantime I humbly suggest that the U de M press release be modified to read:

" high functioning autistics are up to 40 percent faster at Raven's Standard Progressive Matrices (RSPM) problem-solving than non-autistics".

Bookmark and Share


Stephanie Lynn Keil said...

It already known that people with HFA are better at nonverbal IQ tests than verbal IQ tests. I did a self test once of Raven's Progressive Matrices and my score was off the chart, but this is to be expected. Jessy Park, who has HFA, who did not begin talking until she was 9 (I believe) also has a very high IQ on Raven's Progressive Matrices but her IQ is average otherwise, like mine.

I don't see why they need to keep repeating this over and over again, except to promote some sort of strange agenda. Yes, people with HFA have better nonverbal problem solving abilities than people without HFA. We know.

jonathan said...

Actually the study can be obtained by emailing Isabel Soulieres, the lead author of the study. She sent me a copy very quickly.

Actually research of late has shown the effect of higher RSPM scores as opposed to Wechsler IQ in lower functioning rather than higher functioning autistics. Michelle Dawson's study has not been replicated by the one of Sven Boelte who showed the effect confined to lower functioning. Kim Boddner in Nancy Minshew's group has also found no effect of increased Raven's over Wechsler in higher functioning autistics. Souleries' study just worked from the premise of Michelle Dawson's unreplicated study as well as Hayashi's study showing an effect with Asperger's syndrome yet excluded AS from her study and just had persons with autism.

I agree with Harold Doherty that they did not acknowledge the limitations of the population of autistics who mostly had average and above average intelligence and put a disingenous and misleading spin with the media on their study.

Stephanie Lynn Keil said...

Save Our Sophie

I know ND people would be ALL over this, which is REALLY, REALLY sad...

Roger Kulp said...

How is this all that different from this 2007 study?

This says:

From this sample,autistic children who had a known,
diagnosable genetic condition or additional neurological
condition were excluded.

I don't want to Google,so could someone remind me what they mean by "additional neurological condition"? Seizures? Nonverbal learning disabilities? Sterotypy or self mutilation? Developmental delay ?You don't have to be low functioning,you just have to be a nonaspie to be excluded from a lot of these studies.

By cherry picking your subjects like this,you can exclude and distort all you want.

In both the articles about this study,and the ones about the 2007 study,they make a point of saying that at least 40-50 % of all "autistics" are not learning disabled,and imply we should focus on the ones that don't.There is an obvious neurodiversity/aspie bias in the media.

You can see where the Kevin Leitches,and Ari Ne'emans get off calling anything that isn't Asperger's "comorbid conditions" with researchers like this calling the shots.

It does get discouraging,I know. There are serious researchers out there,who are not doing this sort of garbage,but not enough people are paying attention to them when you have a biased media,that may be filled with "self diagnosed" aspies.

Autism Reality NB said...

Roger you make a lot of very good points.

Foresam said...

Why didn't he have higher functioning NT's who could score better than 100 on the test? Since IQ range into the 200's, you'd think he could have found a few. Looks to me like he rigged the study from both directions.

Autism Reality NB said...


Mottron and Dawson were co-authors on the study. Isabelle Soulieres, a colleague, was the lead "researcher" on this "study".

I think they excluded Aspergers persons so they could refer expressly to "autism" without the difficulty that including persons with Aspergers would cause.

Yet they limited it to HFA only. While purporting to draw conclusions about all persons with "autism".

The study mirrors the Neurodiversity ideology which views autism through the eyes of persons with HFA and Aspergers.

RAJ said...

No doubt that 'Rainman' would have aced that test... That doesn't mean that he is not profoundly handicapped.