Autism as a Disorder of Prediction

[dragonwolf]

http://www.pnas.org/content/early/2014/10/02/1416797111.full.pdf

This is an interesting hypothesis that I kind of look forward to seeing more research about. Basically, the idea is that one of the underlying causes of several of the Autistic symptoms stem from an impaired ability to predict unknowns based on current knowns, so things that NTs would expect, an Autistic may find surprising.

It goes into pretty great detail about how various symptoms relate to this hypothesis. The most obvious example is what the paper calls "Insistence of sameness" -- the need for routines, things to be in a certain place, etc. Things are predictable that way (go figure). Interestingly, it seems NTs will also start to show ritualistic behaviors when stress from an unpredictable situation is high enough.

This unpredictability even shows up in hypersensitivities. The paper even mentions that unpredictability is what makes torture so effective (and...well...torture). The victim can't predict when the next drop will come, which raises anxiety levels (as mentioned before, unpredictability breeds anxiety even in NT people), and leads to hypersensitivity.

It goes further and includes how it relates to things like Theory of Mind issues and "islands of proficiency."

I found it rather fascinating to read, because one of the major times I have issues is in situations where I can't predict a person's reaction, or when the reaction runs counter to what I was expecting. Such situations are what serve as the most stark reminders of my differences for me, when I can otherwise "pass," particularly while relaxed.

 

[King_Oni]

Well.. it does make me wonder where I am on the spectrum, hah.

In general I can't stand routine and I kinda thrive on chaos. Clearly within some "reason" but sameness and routine are my bane... I enjoy randomness quite a lot more. Though perhaps I should see it as impulsive and erratic at times. I'm really not as organized offline, lol.

 

[Peace]

"The Only Thing We Fear, Is The Unknown"
If you wander through the woods during the day, you fear nothing and enjoy the beauty and serenity. But go into that same woods at night, and we are afraid even though we know that there could be nothing more out there than what can be during the day. What is the difference? The unknown. If something (or someone) is there during the day, we can see them. But at night we cannot. We do not know if anything is there, The Unknown.

 

[paloftoon]

Well.. it does make me wonder where I am on the spectrum, hah.

In general I can't stand routine and I kinda thrive on chaos. Clearly within some "reason" but sameness and routine are my bane... I enjoy randomness quite a lot more. Though perhaps I should see it as impulsive and erratic at times. I'm really not as organized offline, lol.

You and I are total opposite here King! I can deal with others' chaos as long as it doesn't affect me. Only problem is, usually their chaos does affect me and I always end up being a 2nd class friend, so I respond the same way and end up with a bunch of crappy friends who aren't able to walk all over me 100%, but are totally undependable. I still have to put up with it because otherwise I would have nothing at all, and nobody is perfect, and people seem more selfish now than ever overall intentionally or not (including NTs and non-NTs alike). . .

 

[Randomperson]

"The Only Thing We Fear, Is The Unknown"
If you wander through the woods during the day, you fear nothing and enjoy the beauty and serenity. But go into that same woods at night, and we are afraid even though we know that there could be nothing more out there than what can be during the day. What is the difference? The unknown. If something (or someone) is there during the day, we can see them. But at night we cannot. We do not know if anything is there, The Unknown.

Most people do seem to have the irrational fear of the dark, but you have 4 other senses to rely on, sight is not everything. I myself prefer the night, sunlight burns me too much to enjoy the daylight.

 

[Daniel]

John Elder Robison is not a fan of this theory, as expressed here: Look Me In The Eye: Is Autism a Failure of Prediction?

This hypothesis proposes that what is different about autistic people is that we cannot estimate the conditional probabilities of going to state B given state A and temporal duration. The consequence: "This impairment renders an otherwise orderly world to be experienced as a capriciously 'magical' one."

Basically, they're saying that people on the spectrum have to do repetitive behaviours and stimming because they can't tell what's going to happen next given what's already happened.

So do any of you live in a "magical" world where there's no connection between actions and consequences? Does this sound like anyone's experience here?

If it came down to simple predictability, wouldn't we love small talk? Wouldn't we looooove having the same content-devoid conversations every time we see someone?

 

[dragonwolf]

John Elder Robison is not a fan of this theory, as expressed here: Look Me In The Eye: Is Autism a Failure of Prediction?

This hypothesis proposes that what is different about autistic people is that we cannot estimate the conditional probabilities of going to state B given state A and temporal duration. The consequence: "This impairment renders an otherwise orderly world to be experienced as a capriciously 'magical' one."

Basically, they're saying that people on the spectrum have to do repetitive behaviours and stimming because they can't tell what's going to happen next given what's already happened.

So do any of you live in a "magical" world where there's no connection between actions and consequences? Does this sound like anyone's experience here?

If it came down to simple predictability, wouldn't we love small talk? Wouldn't we looooove having the same content-devoid conversations every time we see someone?

I think you and John are missing/overlooking a couple of things:

1. It's a hypothesis, not a study (only once did John say it was a hypothesis, the rest of the time he called it a study). A hypothesis is pretty much by definition not entirely right, but something to test against. In the scientific method, you start with a hypothesis, then test it, then form a conclusion and refine the hypothesis and/or form a new one. It's not perfect, by any stretch, but it's a starting point (and frankly, it's far better than the idea that we're mentally handicapped).

2. The paper never said it was the underlying cause, but that it's possibly an underlying cause. Autism is complex, we all know that. So, just like not all of us have the same symptoms, not all of us may have the same underlying cause(s). That whole meeting one person with Autism thing.

3. Impairment doesn't mean total lack. A hearing impaired person may not actually be completely deaf (in fact, non-total impairment is more common than total impairment). That person can hear certain pitches and certain volumes, but not enough to fully function in society without help (hearing aids) or adaptations (sign language, writing). Most of us can certainly predict certain things -- particularly things with only a handful of potential outcomes and the cause-effect relationship is fairly constant (ie - fire generates heat; anything exposed to it will either heat up or burn entirely) -- but how many times have you seen Autistics complain that NTs are unpredictable, or that the world is overwhelmingly chaotic? Most of us run into that to one degree or another, I think. Some of us have been able to mitigate issues by learning to consciously run through what amounts to a prediction model, but it's still a concerted, conscious effort.

4. Impairments aren't always so obvious. Again, for some people, predictive impairments may be more subtle. As I mentioned in my first post, I have the most trouble when I'm in a situation where I can't predict a person's reaction, or if their reaction is the opposite of what I predicted (because it then throws my entire predictive process into question and I no longer have "cause-effect data" to draw from). I can predict things. In fact, it's pretty much a coping mechanism for me -- predicting what others will say, so that I can prepare a response. However, I get most anxious when that prediction mechanism breaks down. This isn't just with unexpected reactions, but also with things like schedule changes that I didn't initiate (especially last-minute changes), or when someone goes back and forth on something (my husband's constant back and forth about whether we have enough money in general is a good example). These things break the prediction and question the process.

Regarding small talk, I think our general dislike for it may stem from something else. Not everything has only one cause, especially when it comes to the brain. Additionally, it could also be the same underlying cause, but a different effect. A similar phenomenon comes to mind with the gene that gives a person resistance to Malaria. That same gene causes Sickle-Cell Anemia. Same cause, different effects. This is far from uncommon.

I agree that the paper isn't perfect in its ideas. I also agree that it suffers from the same "outside looking in" perspective. However, the authors of that paper did, in fact, look to the first-person accounts to support their hypothesis. John's writings weren't the only ones they consulted. Not perfect, but a start.

 

[Daniel]

I think you and John are missing/overlooking a couple of things...

I don't disagree with any of your points, and I do understand that it's a hypothesis. I included Robison's post because I agreed that the hypothesis feels to me like an outsider-looking-in perspective.

As I mentioned in my first post, I have the most trouble when I'm in a situation where I can't predict a person's reaction, or if their reaction is the opposite of what I predicted (because it then throws my entire predictive process into question and I no longer have "cause-effect data" to draw from). I can predict things. In fact, it's pretty much a coping mechanism for me -- predicting what others will say, so that I can prepare a response. However, I get most anxious when that prediction mechanism breaks down. This isn't just with unexpected reactions, but also with things like schedule changes that I didn't initiate (especially last-minute changes), or when someone goes back and forth on something (my husband's constant back and forth about whether we have enough money in general is a good example). These things break the prediction and question the process.

I think you've touched on it here. We like things to be rule-based, consistent, logical--yes, predictable. I feel like what's different is that we use prediction in social situations in the first place, as opposed to just "going with the flow".

 

[Heissonear]

I call this research "rent seekers", filling in grant money.

In science, something is not true until it can be observed and tested as true.

How is this research aspect to be tested?

If it came in the mail I would call it junk mail and toss it. But in the new view of "Autism is more common than we think" there lies funding for researchers. In academia it is a game to live by.

We may need to watch for "research in Autism " that follows the grant seeking course.