"This study builds on our previous findings and should help educators capitalize on the intellectual abilities of autistics," says senior researcher Laurent Mottron, the new Marcel & Rolande Gosselin Research Chair in Autism Cognitive Neuroscience of the Université de Montréal and psychiatry professor. "The limits of autistics should constantly be pushed and their educational materials should never be simplified."
I emailed Dr. Soulie'res, asking her for a .pdf copy of the study and she was nice enough to promptly reply with a copy of the study. I have done my best to read the study though I did not understand a lot of it. I am curious as to how this study which shows a subset of persons with autism to be significantly faster at processing questions on the Raven's matrices than typical controls can be utilized as a real world application which would allow persons to use the results to help autistics learn or function better in society, which it seems to me what Dr. Mottron is implying. He and his sidekick Michelle Dawson also coauthored the article, though Soulieres was the lead author.
On perusal of both the statement to the media, the abstract of the study and in fact the study itself there seem to be problems with this interpretation. First we see the initial sentence of both the abstract and the actual study itself which may serve as a potential red flag:
Recent behavioral investigations have revealed that autistics perform more proficiently on Raven's Standard Progressive Matrices (RSPM) than would be predicted by their Wechsler intelligence scores.
The study cites Dawson(2007) as evidence for this assertion in which huge differences were found between the scores of autistics on the Wechsler IQ test as opposed to much higher scores on the Raven's matrices. After the study was published Dawson seemed to think her findings should influence all public policy towards autism with the statement:
I hope this paper will have the practical effect of making it less likely that autisics will be written off, as autistics are so routinely written off by autism advocates in Canada and elsewhere.
Souleries also cites a study done by Hayashi showing that persons with Asperger's who were matched with typical controls on Wechsler IQ outscored them on the Raven's. The problem with this interpretation was that Souleries only included persons with speech delay and problems in her paper who would have an autism diagnosis and excluded those with Asperger's so Hayashi would most likely not apply.
Another problem is that a study Boelte(2009) has largely failed to replicate Dawson(2007) so the question is still open as to the superior of Raven's scores versus Wechsler in autistics. Though Boelte found that some autistics scored higher on the Raven's than the Wechsler, the effect was much less pronounced than the one found by Dawson. It was also limited to lower functioning autistics (meaning those with an IQ of less than 85). The subject pool used by Boelte was also different than those by Dawson-they were a lower functioning group. Had Boelte used a group more similar to Dawson's the effect of Wechsler's versus Raven's would likely be negligible.
In contrast to Dawson's pollyannish statement Boelte makes a more guarded statement:
in conclusion, the claim that intelligence has been underestimated in autism seems somewhat premature.”
Kim Bodner who works in Nancy Minshew's group has also studied this question with high functioning autistics. She has found no significant difference in the scores of HFAs in the Raven's vs. Wechsler. I think her work has only been confined to a limited group of autistics, only high functioning ones. Also, I don't think her work has yet been published in a peer reviewed journal, but so far has only been presented as an IMFAR poster. I think her study will be published either at the end of 2009 or the beginning of 2010.
While there is a subset of persons on the autistic spectrum who score substantially higher on the Raven's than on the Wechsler it would appear to be mostly confined to lower functioning autistics who would not necessarily even be in the majority of all autistics. Further studies by the Mottron group, Boelte, Minshew's group and others may shed more light on this question.
In the latest study there is about a 6:1 ratio of male versus female autistics as opposed to the 4:1 ratio found in the general population. Souleries does not acknolwedge any limitations in this work. Of note, the autistics were also of normal intelligence with IQs of about 100, so whether or not these findings are applicable to lower functioning autistics is in question. Since they were matched with the typical controls they were different than at least some of the subjects in Dawson (1987) who had lower Wechsler IQs overall than typically matched controls. It is possible that the lower functioning subjects in Dawson (2007) would have to be excluded from the Souleries study due to the fact they would make too many head movements, creating artifacts on the fMRI scans.
Another germane issue is the lack of discrepancy between the performance IQ scores and the verbal IQ scores of the subjects, an average of about 99 on the verbal portion of the Wechsler and about 103 on the perfomance. This might indicate these autistics were less impaired than other persons with autism such as myself who have a 40 point discrepancy between the two tests. My verbal IQ has been tested in about the low 120s and my performance IQ in the low 80s. I am good in arithmetic, general knowledge, similarities, high average on vocabulary and below average in comprehension. On the performance test I score in the severely retarded range in both the block design and the object assembly test (which entails putting together puzzles) which I have great difficulty in due to my perceptual motor problems. There are probably also examples of a number of autistic persons who have the opposite profile, relatively high scores on the performance test as compared to verbal IQ. I wonder if this research could be applicable to myself or those on the spectrum who conversely have a discrepancy with a substantially higher performance than verbal IQ.
In at least one paragraph of her paper Souleries is more cautious about the interpretations of her work than coauthor Mottron:
Although theresponse time advantage for difficult RSPM problems weobserved may reflect an underlying processing advantagein reasoning mechanisms enjoyed by autistics, additionalstudies directed at this specific question will be requiredto fully explore this possibility.
In the last paragraph of the study unfortunately she makes a statement similar to Mottron's about how this work could be applied in the real world to help persons with autism. Not just the study's subjects who are clearly not an extremely representative sample of autistics, but all autistics, as she does not seem to acknowledge the limitations of this study due to the characteristics of the experimental group.
On a more positive note, in spite of problems with the interpretations the author's claim have real world applications, some of the things in the study were quite fascinating (at least to me). Dr. Souleries talked about how faster processing times may have reflected the lack of neural connections in autistics that had been cited by other studies in various brain areas, resulting in compensatory mechanisms in other areas:
Regarding possible developmental mechanisms leadingto the atypical autistic activity patterns seen in our study,clues may be found in recent studies of white mattermicrostructure [Barnea-Goraly et al., 2004; Courchesne etal., 2001; Herbert et al., 2004; Ke et al., 2008; Keller et al.,2007] and functional connectivity differences in autism[Just et al., 2004]. In autistics, Just et al., have observedreduced functional connectivity between frontal and parietalcortex in a variety of tasks, including sentence comprehension[Just et al., 2004; Kana et al., 2006], n-backworking memory tasks [Koshino et al., 2005, 2008] andresponse inhibition tasks [Kana et al., 2007]. Similarly,reduced functional connectivity between early visual areas(BA17) and inferior frontal cortex was found in autisticsduring a visuomotor coordination task [Villalobos et al.,2005], but this decrease was concomitant with increasedfunctional connectivity between the thalamus and its frontaltargets [Mizuno et al., 2006]. Given existing reports ofatypical connectivity in autism, there are several availableexplanations for our findings.One possibility, based on proposals advanced by Just etal. , is that increased use of occipital brain regions inautistics reflects compensatory activity arising from anatypical neurodevelopmental trajectory, based on significantcommunication restrictions between prefrontal andoccipital regions. In this scheme, inefficiencies in engagingprefrontal mechanisms could result in the development ofcompensatory strategies and processing mechanisms moreheavily reliant on occipital and posterior parietal corticalregions. These compensatory mechanisms would have tobe as effective in supporting reasoning as the more typicalmechanisms relying on prefrontal function.......
So even though I find Mottron's statement to the media to be offensive, this study, I must concede, may not actually be useless. It may actually give some clues to how the brain in at least a subset of persons with autism functions and processes information. The compensatory mechanisms of the occipital lobes in persons with autism as opposed to the deficient connections in the frontoparietal areas of autism that have been reported. These findings also seem to be consistent with the mirror neuron hypothesis that Marco Iacobonni and Mirella Dapretto have pushed as an etiology in autism. It may be similar to a blind person who has compensated hearing and can be prodigious in music or reports of deaf people being able to see more easily out of the corner of their eyes. Whether it could actually have any applications for even a subset of autistics developing compensatory mechanisms could be used in the real world is another question, however.
So, Dr. Mottron, I would like to ask you. What is the basis of this research, given all of its limitations, being able to help all autistics in educational and other achievements? What difference will it make in their lives? If you are going to make such sweeping statements to the media, why can't you be more specific? I really wonder what the answer to these questions are.