Hi, thanks for the feedback!

"interesting metaphor ...than...underlying mechanism."

As for the mechanism part, no, this paper does not seek to investigate the neurobiology behind autism, just the outward behaviors. For example, the 'eye' has evolved multiple times in the animal kingdom, for different underlying biological reasons.

If one accepts Solomonoff induction, which is based on Kolmogorov Complexity, as a formalization of science, then it might be that evolutionary pressures would favor the development of a human phenotype that more closely approximates Kolmogorov Complexity. This paper suggests that such pressures may contribute to the emergence of the autism phenotype. https://www.lesswrong.com/posts/Kyc5dFDzBg4WccrbK/an-intuitive-explanation-of-solomonoff-induction#Formalized_Science

You said the paper "describes autists as successful, but [you] think most of them are not". This could be from multiple reasons, such as:

The nature of Kolmogorov complexity. Autistic people may study something, but not find a clean, useful solution. But when it works, you can find successful physicists.
Missing some pieces that encourage the computation of Kolmogorov Complexity. As autism is a spectrum, there could be someone who has savant-like discrete memory, but lacks mathematical background. Or someone with tunnel vision, who does not have a good memory. But when all the pieces come together, you get someone who can do a version of Solomonoff Induction really well.
Tradeoffs for having this phenotype, related to flexible thinking patterns.

Again, thanks for the feedback!

Posts

Wikitag Contributions

Comments

Posts

Wikitag Contributions

Comments