Research: Autism as a Disorder of Prediction???

Post Reply New Topic

I came across this <click> article from the Simons Center for the Social Brain at MIT suggesting that Autism is a "disorder of prediction".

Below is an excerpt <click> from an article describing this hypothesis.

"Autism is characterized by many different symptoms: difficulty interacting with others, repetitive behaviors, and hypersensitivity to sound and other stimuli. MIT neuroscientists have put forth a new hypothesis that accounts for these behaviors and may provide a neurological foundation for many of the disparate features of the disorder.
...
The researchers suggest that autism may be rooted in an impaired ability to predict events and other people’s actions. From the perspective of the autistic child, the world appears to be a “magical” rather than an orderly place, because events seem to occur randomly and unpredictably. In this view, autism symptoms such as repetitive behavior, and an insistence on a highly structured environment, are coping strategies to help deal with this unpredictable world.

The researchers hope that this unifying theory, if validated, could offer new strategies for treating autism."

I felt that way, minus the magic, until I started speaking. Speaking enabled people contact, so the world started to make more sense. Most of my "classic" autism symptoms disappeared as a result by age 6.

Maybe they've been reading my notes. Maybe I should publish my damned book.

The world is magical, but they can't see it.

Alas, I don't believe in magic or even God. I'm science based. Still Autistic (or at least I think I am).

"magical" was probably a poor choice of words. I think "unfathomable" is more to the point.

When you're a kid, there doesn't seem to be rules. Things just happen, and since you can't affect the outcome, things just seem random.

Then you get older and have some understanding that there are rules, and you try to apply logic, and what you think is high intelligence, and...you still get it wrong.

So, you seek order in the things you can affect.

Here's an example. When I was a kid I always answered any question from my mother with "I don't know." She didn't much like that, and I'd get in trouble, but her response was predictable (and mild). If I actually answered her question, there was no telling what would happen.

When I watch my kids, particularly my daughter, I think this is why she needs scripts to handle things and why unfamiliar environments or unexpected occurrences throw her for such a loop...it pulls the rug out from under her, and she is stuck in an environment that she cannot quickly adjust to, so it is completely overwhelming. I have always noticed with her, as long as things go the way they are "supposed to," even when "supposed to" = "not in her favor," she is fine. It is when something unexpected happens that we have not talked about/planned for or she hasn't experienced before that she has problems.

Closely related to this, my hypothesis is that she has difficulty learning incidentally, which further handicaps her ability to respond appropriately to the unexpected. She needs to learn everything explicitly. Once something is explained to her in a way she understands--even social things--she is good to go. She usually just can't figure it out on her own. So when something happens unexpectedly, she cannot pull from her environment or the people around her to figure out what to do. She just feels--and looks--lost.

The authors are myopically focused on their particular hypothesis, without looking deeper for other potential root causes, such as the absence or weakness of cognitive filters, especially related to social interactions.

A lack of cognitive filters goes a long way towards explaining the experience of sensory overload. Sensory overload has multiple effects, which include making the task of predictive mental calculations more complex.

What if typical predictive mental calculations are largely subconscious, and if autistic predictive mental calculations are largely conscious due to lacking filters?

Typical human behaviour would largely be based on intuitive (subconscious) probabilistic calculations, where associated uncertainties don't even enter the stream of conscious thought, and autistic human behaviour would largely be based on conscious probabilistic calculations, where awareness of associated uncertainties compromise the speed of reaction and decision making.

Autistic anxiety, procrastination, and related repetitive thoughts could result from processing too much data rather than from poor performance at probabilistic calculations. At the short time scales related to motor control and language processing in a social context, typical and intuitive probabilistic calculations that suppress uncertainty may lead to the observed differences in behaviour, i.e. subtle motor control issues, and a lack of "fluency" in social interaction due to intensive conscious processing before responding in a conversation.

In a social context I am always aware of the ambiguities and uncertainties in communication. Here on WP I have read comments from typical individuals who explain that they do not worry much, if at all, about ambiguities and uncertainties in communication. All communication is full of ambiguity and uncertainty, but autistics are possibly the main group of people who continuously (obsessively?) worry about this state of affairs.

Intuitive behaviour and decision making has clear advantages on short time scales, but is has disadvantages over longer times scales measured in hours, days, months, years, and decades. It could be argued that autistic brains are biased towards long term planning (days, months, years, decades), and are compromised when it comes to time scales of minutes, seconds, and sub-seconds.

In specific domains, after a lot of practice and conscious analytical processing, people with an autistic cognitive lens may well become good at or excel in narrow domains requiring sub-second decision making. In my case, I was never any good at team sports, firstly because I don't relate to the notion of "competition", and secondly because I don't seem to have the ability to intuitively relate to the situation of a game at the scale of seconds and sub-seconds.

At the same time, with many years of practice, I have trained my sense of balance to the point where I do reasonably well on ice skates, a snowboard, or on a windsurfer. I believe this was possible because for all these activities what matters is concentration on a non-social dynamic context and on the position of my body in this context. At the same time I am no expert in any of these activities, I am simply much better than someone who has never seriously given them a try. Windsurfing is the best example, as it has the longest learning curve, and most people give up much too early. In this context autistic perseverance is a real asset.

Rather than postulating autism as a disorder of prediction, the authors might be well advised to explore autism as a lack of subconscious cognitive filters, with impacts on subconscious short-term decision making. The former thesis then becomes an explanation for observed autistic behaviour at short time scales rather than a comprehensive explanation for autism.

Can anyone relate to what I outline above?

I think most people get it wrong most of the time. People think they can affect others, affect the environment, affect their lives, but its illusory. "The more you tighten your grip, Tarkin . . ."
The human species seems driven to continually modify whatever it comes in contact with. People need to chill out and respect the natural state of things (which seems to barely exist anymore.)

jbw: That is a very interesting theory that you outline. I can see how lack of cognitive filters could be an underlying cause for poor observed prediction performance. I relate to the experiences you describe. In general, it seems to me that, when I first begin doing a certain activity, I naturally struggle with learning the activity more than other people who are also learning. But, over time, I am able to get fairly good at those activities I am interested in because of my commitment and perseverance, and possibly increased attention to certain aspects. It is as though my learning curve is shaped differently from that of many other people.

Unfortunately, life involves many novel situations where we are unable to practice before facing those situations. I think could account for poor prediction. Social situations are definitely one of the most impacted by these patterns, in my life. With people I know very well, I feel I can communicate quite well. But each new person I meet, it is like learning a new skill almost from scratch, to learn how to predict that person's responses to some degree and to understand that person.

I think like a lot of the theories it's only partially correct.

The real world is magical.
The social conceptual world is insane.
Naturally the environment would be very unpredictable for a very small child of any neurotype.
The NTs eventually become engulfed into the social conceptual world where things become mostly intuitive.
For the NTs, the real physical world often remains quite mysterious, but it doesn't really bother them so much because they have less interest in it in comparison to the social world.

The higher functioning autistic will learn to adapt and use logic to make the real world more predictable. They will try the same strategy with the social conceptual world, but that does not always work out so well no matter if they have interest in it or not.
The social world does not follow the same rules as the physical world.
And yes, I think the cognitive filters make all the difference as a real explanation of why certain things remain unpredictable.

jbw,

Actually, they acknowledge the Intense World Hypothesis as well as sensory overload. They say that what is behind both is the impairment in prediction. They explained that for sensory overload, it seems to happen in autism due to a failure to habituate. This failure of habituation comes from difficulty predicting how long it lasts, how often it will occur, etc., the researchers said.

Basically, they are taking a shot at a universal cause of autism, it seems.

I wonder if actual objective tests will emerge from this research, instead of always relying on self-reports, observation, and reports from others.

I've said this to my mom on several occasions. I hate how unpredictable people or situations are. I do feel overwhelmed when things don't go as I had thought or hoped they would. Also, order to a system that flows seamlessly on a consistent basis usually keeps me on track and calm. I'm more focused and understanding when things aren't chaotic (or normal in the case of the article). So order and predictability, I do understand this article and can agree with it.

I am glad to see that they are starting to understand it a bit better.

I hope so too, beneficii.

I am on disability service because of autism and one of the main factors people decided I was not able to be able to work was that I have a severe disability in not being able to predict people, which would leave me vulnerable to abuse, which has happened a lot.

There is some validity to the claim in that article.
I myself don't feel too disordered in my predictive abilities (that is, i don't think i'm worse than the next person), but the conclusion for the "general autistic" (even though those don't exist) follows from the mentioned effects indeed.

using "magic" as a descriptor isn't the right choice of words though, "illogical" would have sufficed