Study of fluid intelligence in autism spectrum disorders

Post Reply New Topic

While reading up on the Raven's Progressive Matrices, a measure of fluid general intelligence (abstract reasoning ability), i came across two papers that used this test to identify this ability in children with autism spectrum disorders. The results? children with asperger's, while having the same full scale IQ but lower performance IQ's on the WISC, scores signfiicantly higher on the raven's compared to the controls with matched IQ's. And for children diagnosed with classic autism, the result was that they also scored well on the Raven's, and sometimes to a very large degree better than their IQ scores would have suggested (up to 70 percentile points between their IQ and Raven's scores ie. if they scored in the 2nd percentile IQ-wise (mentall ret*d), they might have scored in the 70+th percentile on the raven's test (people with an IQ slightly above the mean fall in this percentile). So what do these results mean? that in asperger's, general intelligence is overal conserved/enhanced and general fluid intelligence is generally enhanced, while in more severe autism general intelligence is generally decreased/conserved and general fluid intelligence is conserved or enhanced(relative to the mean).
You can find the papers under these names : Superior fluid intelligence in children with Asperger’s disorder and The Level and Nature of Autistic Intelligence (only abstract available without a subscription)
What do you guys think of these results? btw i also saw another paper a few weeks ago stating the same thing, people on the spectrum do better on these types of intelligence tests (which interestingly are suggested to measure general intelligence the most accurately) than their IQ scores would suggest.

Has anyone here taken the Raven's Progressive Matrices? and what were your results?

What does that mean then? What is 'fluid' intelligence? I don't know if it's relevant, but one interesting idea that I stumbled across recently is that autism might be a product of a brain that is hyper plastic.

look for the article "Fluid and crystallized intelligence" on wikipedia (not having posted before i can't post links yet)

I did poorly on the one online test I found. About 25 points lower than my formally tested composite IQ. I think I may have a co-morbid learning disability that effects symbolic processing and short term working memory. It's something I plan on bringing up with my psychologist.

Interesting.. btw i'm doing some more reading on tests that measure G, it seems there are some recent theories out there (the General fluid and crystallized theory dates from around WW2, like many still used cognitive abilities concepts..) the newest theory is "Cattell-Horn-Carroll theory" (again wikipedia article title as well), modern standardized tests seem to include the most recent model (WAIS-IV), i think WAIS-III was still largely based on solely the crystalized-fluid disinction ? can i ask what your percentiles were for the IQ test?

My last official test was 99th percentile combined score. The Raven Matrices test I took online was 78th percentile. I suspect if I could take the Raven test enough times I would develop a strategy that would allow me to get them. It is relatively common for me to have to do this. I get stuck in anything that requires too much switching between too many possibilities and my brain sort of locks up. But if I can develop a strategy that keeps things clear, then I will have a big increase in performance.

Ah, i also read that while in these studies they did better on the raven's, for asperger's the performance IQ part of WAIS tends to be lower percentile than their verbal IQ in general, was this true for you?

I don't recall as it was at least 5 years ago. But I suspect there was a measurable gap. I have always had trouble with the performance side of things. I remember linear algebra in college as sheer torture because I could grasp the theorems and concepts but had virtually a 100% failure rate when actually multiplying anything more than a 2x2 matrix. I probably would have failed the class if it were not for the fact that the tests were graded for partial correctness (if they could tell you got part way through before you messed up and were using the appropriate technique you got some credit). The class so shattered my confidence that I never fully recovered it and ended up dropping out the following year. This was long before I knew of Asperger's and was convinced my brain was hopelessly defective. I must have made my professors really scratch their heads. I told a math professor during a number theory class that one of the fundamental theorems in number theory was vacuous. Strangely, he partially agreed. But then other times I couldn't perform the simplest of computations or manipulations. I alternated between laser like intensity and clarity and utter confusion.

sounds like you're a very "crystallized intelligence" guy and not a "fluid intelligence" guy? btw i think i might be somewhat of a combination of both, with slighhtly better at fluid than crystallized (i can make up arguments about things on the spot but will forget what arguments exist already in the field, for example) it's interesting that asperger's seem to do better at the crystallized part of the WAIS and auties do better at the fluid part (might be related to hyperdeveloped hyperactive local frontal neural networks)

Are you a cog-sci professional or just very interested? I was aiming for a cog-sci grad school before the wheels came off. So I have more than a passing interest.

That's interesting. It reads to me like the difference between something like raw cognitive ability, and the ability to draw intelligently from processed and stored information, AKA wisdom. Please correct me if I'm wrong.

Wisdom and wavefreak don't often collide in the same sentence.

This is really how it feels to me. I have tremendous difficulty getting certain types of information in, but once it's in I am really good at drawing conclusions from disparate sources. It's VERY aggravating at times.

Not wisdom, fluid intelligence could be called the ability to "imagine cause and effect".

Wisdom would be the ability equally measure outcome against cost - as in Solomon and splitting the baby.

Both types of intelligence deal with cause and effect. But fluid intelligence is not dependent on experience, where Wisdom draws heavily on lifetime of exposure to different cause and effect situations. Fluid intelligence is about deducing an outcome given a set of inputs, it is a creative intelligence. Wisdom is about prioritizing inputs to achieve an outcome, it is a logical intelligence. Many researchers would combine the two types of intelligence, but they are independent although not mutually exclusive.


Even after centuries of research, the entire realm of intelligence theory is poorly structured in psychological circles.

I want to point out that finding that IQ is no lower than normal in Asperger's and can be very low in autism is not surprising, or should not be surprising, because it's one of the criteria used to differentiate the groups.

This just in: women have two X chromosomes. Men have only one.

I was under the impression that IQ was not a diagnostic criteria, rather that IQ distribution skews higher for Asperger's.