what autism tells us about humans in general

Many years ago, I had the good fortune to be invited to give a big plenary talk at
a conference in New England on deep psychology and psychiatry. I talked about how
the new mathematics we are developing in neural networks fits with, and actually deepens our
understanding of serious human psychology and the subconscious mind.  As an example, I described how I developed the algorithm now called "generalized backpropagation" by translating certain ideas from Sigmund Freud into mathematics.

At that same conference, the keynote dinner talk was delivered by a woman named Temple Granit, who is very famous among people who study autism. She is famous for many reasons - such as how she made a lot of money as a successful entrepreneur even though she is autistic herself. The talk itself was also very powerful. Somewhere near the beginning, she described how she thinks in images instead of words, and discussed what that feels like. Most people at my table thought it sounded very bizarre, but there was one other engineer or mathematician at my table who also said "of course we think that way, in images too.." and I mentioned a classical study of great mathematicians where they did show that all of them think "in images," the same way (except for Von Neumann, who was 50-50 balanced -- strong image thinking but also strong verbal thinking). (Much later, at a conference of mathematicians, I mentioned this to a guy who knew Einstein; he said "No question, Uncle Albert was rather autistic himself, that's no exaggeration, let me elaborate...")

She went on to say: "Autism is basically a well-developed right hemisphere and a weak left hemisphere.  Autism is not the real mental health problem in this country. The real problem is
the mirror image of autism,  Williams' syndrome, where the right hemisphere is underdeveloped and the left hemisphere is overdeveloped. All bullshit and no reality -- and not even the honest real bullshit I made money by shoveling. Why does everyone worry about autism, when the Williams syndrome is a more serious problem? It's because the patients are running the asylum. Who do you think is running the country?" That reminded me of a technical paper I saw, based on psychology tests at different times, which concluded that the heads of US industry used to be balanced 50-50 on the whole between verbal intelligence and spatial intelligence -- but that it changed to be more verbal.

Much much later I googled on "Williams syndrome," and saw a rather different definition.
But that does not invalidate her point. In any case, psychology has a habit of shifting the definitions of words over time, and I didn't want to spend time tracking down the history of that phrase.

Most recently -- Granit has a new book on Autism, which I half read half scanned this week.
Most of it was refreshingly concrete and honest -- but there was one problem which alienated me a lot halfway through the book. She  began asking: "What is the common threasd? What is the true basis
of what autism is in general?" I would ask: "What makes you think that there IS a common thread?
And why do you folks assume there is even a Spectrum here... a one-dimensional continuum
of 'degree of autism'?" Fortunately, as a concrete thinker, she didn't go back
and edit most of her memories and experience to fit  that oversimplified picture of reality. Her book
provided plenty of concrete examples giving us a feeling for the different TYPES or DIMENSIONS of autism.

For example, it was clear from a real brain scan that her left hemisphere was physically smaller
than her right hemisphere, to an extreme degree. But it was also clear that lots of
autistic people (maybe even most) don't have that particular TYPE of autism.

The book says a lot about how autistic people (of various types) learn to cope with reality.
But in fact... a lot of the same issues apply to ALL humans. She talks about the interplay between
"Freudian effects" (which I interpret here as traumatic or euphoric memories, both of which mislead people), biological predispositions, and sheer lack of cognitive development.  In a different context, I sometimes talk about the need to balance between digesting or defusing memories, versus the need to develop new memories to broaden one's understanding of the world. She places great emphasis on the issue of overstimulation versus understimulation. She doesn't mention the tension between intolerance of cognitive dissonance and novelty seeking... but that clearly is in play in the stuff she describes.

But... this blog is probably not the place to elaborate.

Just one last thought.

In my recent papers in Neural Networks (2009 and 2012), I stressed how'
the human species is basically "half man, half ape."  Perhaps the entire human population
is autistic to some degree, relative to what humans imagine they are (which would entail sanity or integrity, the main theme of the 2012 paper).  But since "autism" is not really one variable, that is a simplification...