Theory of Mind

Post Reply New Topic

Couldn't agree with you more, Ettina.

I do feel very irritated about the damage (in my view) that Baron-Cohen has done, and the misconceptions he has disseminated, in the cause of his own self-interest.

Here is an introduction to some of the flawed methodology Baron-Cohen used. Yes, his findings were based on the flaws, and they SUCK!

The Sally-Anne test, where 61 children (20 autistic, 14 Down's Syndrome and 27 neuro-typical) were shown two dolls, is an example of bad "science." Sally has a basket in front of her, while Anne has a box. The Sally doll, presumably made to move by an adult, which further complicates the test, puts a marble into her basket and leaves the room. While she is gone, Anne takes the marble from Sally's basket and places it in the box. When Sally returns, the child is asked, "Where will Sally look for the marble?" Only 20 percent of the autistic children were able to correctly answer the question -- Sally will look in her basket.

From the test results Simon Baron-Cohen concluded "that the core problem in autism is the inability to think about other people or one's own thoughts" according to the blog, holah.co.uk. Except that his test did not take into consideration the challenges many autistic children have in sequencing, language problems, misunderstandings of prepositions, the level of anxiety or stress levels of the autistic participants at the time of testing. Nor did it take into account literal thinking, something many autists have, all of which made the test and the questions asked that much more challenging.

For any research findings to be valid, the researcher must identify and take into account all "confounding variables". This is elementary, in the social sciences, students learn this in introductory courses.

Here is a made up example, of ignoring confounding variables, just for the purposes of illustration of bad research methodology.

Suppose you were a qualified psychologist and you decided to explore whether ASD children were less competitive than others. So you designed a study based on running races for same age children. And as the ASD children mostly came in last, you claim your theory is proven correct!! !! The evidence "shows" that your theory is correct!! ! Only it isn't, because there is no consideration of the confounding variables: a proportion of ASD children have poor muscle tone, dyspraxia, problems with loose joints, enormous stress when pressured to do something for which their bodies are not designed.

Your "scientific research" in the example I made up here, would only show or prove what a very poor researcher you were, and this is precisely the problem, in a nutshell, with Baron-Cohen's work.

It happens all the time in social science, which is why you should take all "research proves/shows" claims with a grain of salt, or at least considerable suspicion, unless you can see the whole methodology, data, and and on what the researcher actually measured (and what he or she did not measure). Career building academics know this, and know that the general public is not savvy to critical analysis of their claims. Hope this helps clarify my earlier comments for you.

Interesting. While the methodology he uses to prove his theories may have it's flaws, I find his theories themselves have some merit in understanding autism.

For eg, as an aspie, for most of my life, I have been a "non-thinking" person.
Also, his theory that the autistic mind is an extremely systemizing one is an useful insight. Does it have it's flaws? Yes, but frankly, I haven't come across a better hypothesis for the workings of the autistic mind. If there is a better explanation or theory, I would like to know about it.

I failed the sally-anne test as an adult who has no trouble processing the simple language in the test or pretending that a doll is a person, and I failed in the same way that the children failed, by pointing to the real location of the marble instead of where sally would look for it. It was just my fast automatic response and definitely didn't involve thinking about what someone else was thinking. It was my spontaneous behavior to not think about what other people are thinking. I have read some people try to account for the sally-anne test failure by 80% of autistic children back in the 1980s when autism was mostly classic autism phenotype, and none of the accounts by people who passed the test made more sense than my interpretation that I didn't think about what other person thought, but answered only what was important to me, which was the location of the marble.