Comment by superb-owl | Hacker News Reader

Comment by superb-owl | Hacker News Reader

I'm increasingly intrigued by the idea of autism as one end of a spectrum of cognition strategies, with schizophrenia at the other end. Scott Alexander has a lot to say about this:

https://slatestarcodex.com/2018/12/11/diametrical-model-of-a...

https://astralcodexten.substack.com/p/blindness-schizophreni...

https://slatestarcodex.com/2016/09/12/its-bayes-all-the-way-...

tl;dr:

Autistic folks have really tight confidence intervals on their priors. In these circumstances, sensory data that deviates slightly from expectation becomes very grating (noise and touch sensitivity) and sensory data that conforms to expectation is exceptionally rewarding (hence stimming and deep topical science-y interests).

Schizophrenic folks have really broad confidence intervals on their priors. In these circumstances, sensory data that deviates from expectation is interpreted as meaningful rather than noise (hence hallucinations and psychotic jumps to conclusions), but they'll tend to do better on tasks that require relaxed priors (e.g. the concave/convex face)

Came here to link SSC as well. I find the predictive coding theory to be a pretty compelling model of cognitive function/dysfunction.

There seems to be another related but separate dimension to the spectrum. I remember an analogy of borderline personality disorder being the female version of autism mentioned somewhere on SSC as well, which is interesting. The disorders seem to share some sort of failure in integration between top-down priors and bottom up sensory information, but with a different focus. Borderline personality disorder seems like a difficulty in integrating more extrinsic, social concepts like self-identity and models of other people. It makes me think there's something fundamentally different between introverted and extraverted cognitive strategies.

Personally I see this as indicating it is dominated by learning method.

The autism end is highly skilled individual learning, either intrinsically or because of social deprivation during development. This does not mean they can’t learn socially, but that social learning is a net loss for them and they become a commons (in the tragedy of the commons sense)

The schizophrenia end is predominantly social learning, where either individual learning skill is not as high, leading to bad outcomes and seeking confirmation, or early social opportunity is atypically high, leading to a bias for social interaction.

This can be either based on skills/ability or more simply on differing intensity of affect as experienced in different settings.

It might have some explaining power but I can think of counter examples from my own personal life. Schizophrenia just like autism a very defuse category that covers many non overlapping conditions.

I think this silly joke by jreg about Schizotism/Autizmophrenia explains it best:

https://www.youtube.com/watch?v=IHuFSnhKG9I

...I think this video is too steeped in Layer 3 irony for me to parse.

This will clear that up for you: https://www.youtube.com/watch?v=ZjSE1h_Ad9E

The existence of autistic schizophrenic people seems like a problem for the hypothesis that they're opposite ends of a continuum though?

Interesting! Turns out autistic people are up to 3.55 times more likely to also have a schizophrenia diagnosis [1].

I've actually wondered something similar about myself: I exhibit some minor autistic and schizophrenic tendencies, but in different contexts. I like to think my position on the scale is more fluid than it is for others. Maybe some genes code for higher variance along the spectrum, while other genes push in one direction or the other.

[1] https://www.healthline.com/health/autism-vs-schizophrenia

Just don't ask too many questions about what else he believes about the range of human cognition, or what we should be doing about it!

Tell me about it.

He's a smart dude and I can't help but appreciate his work. But his defense of racialism is getting harder and harder for me to ignore.

> his defense of racialism

??

I'd suggest reading the NY Times article on him [1], as well as his rebuttal [2]

I don't think the Times was entirely fair in their coverage. But as a long time SSC/ACX reader I've often felt a subtext of racialist beliefs lurking in the corner. He never comes right out and says it, but e.g. praising Charles Murray is more than a dog whistle.

To be clear - I'm distinguishing racialism (which believes there are large, measurable, genetic differences in traits like intelligence and aggression between races) and racism (which carries an extra connotation of emotional hatred). I think of racialism as racism masquerading as objective science.

[1] https://www.nytimes.com/2021/02/13/technology/slate-star-cod...

[2] https://astralcodexten.substack.com/p/statement-on-new-york-...

Do you strongly distinguish between race and culture?

I'm not entirely sure I understand the question...

But yes - in my set of definitions, race is genetic/hardware, and culture is environmental/software.

Obviously the two correlate pretty heavily - i.e. racial groups tend to form subcultures.

Does culture not vastly complicate the racism question, perhaps often to a point beyond the human mind's ability to conceptualize it accurately, especially considering how flawed we know it is at that task?

I mean, sure, it's complicated. Are you saying complicated things aren't worth studying or discussing?

Quantum mechanics is also "beyond the human mind's ability to conceptualize" but we're still able to make progress. And discussions of race are vastly more important, at least when measuring in terms of human suffering.

I'm more so saying that your mind presents to you a representation of reality, and if you take this representation at face value, you may have an inaccurate model.

I don't know you personally, but "most" people I encounter seem to draw no distinction between their model of reality and reality itself.

Consider:

> praising Charles Murray is more than a dog whistle.

What is the meaning of "is" (to be) in this context, in a literal/architectural sense? Does your "is" align with other people's? And where this a discrepancy, how skeptical of your take on it are you?

Possibly related:

https://i.redd.it/nbqixw27k5vy.jpg

Was that the one that doxxed him?

[Edit:] Yes, yes it was the smear that doxed him and misrepresented his writings in an egregious manner. I had forgotten about that incident, but F the NYT and anyone repeating Cade Metz' lies.

In early 2020, I learned the New York Times wanted to write an article about me. They had discovered my real name and wanted to reveal it to the world.

I'm not really familar with SSC's writing, and the NYT article is paywalled. For that matter i had not heard of Chris Murray either.

But it seems really unfair to infer that someone believes X because he praised someone who believes X, especially without further context.

Reply on news.ycombinator.com