[robotcookies]

I don't think humans evolve to be inaccurate - I think a better description is that they evolve to not all be the same. If you look at two species, the one that is very uniform is more likely to die out over the long term than the one that has more variation. That variation can be physical, or in the mind and how it makes choices. But this variation makes the species more adaptable when conditions change drastically.

If you look at the long term, in the world that the article describes (slow, gradual with rare cataclysmic changes), most wrong decision makers will still die at a higher chance than correct decision makers. It's just in those rare situations that they survive.

[bduerst]

That's a stretch, because it implies intent for diversity. Mass extinction through a lack of robustness is definitely something that natural selection selects for, but the inverse is true where too much genetic variability leads to less reproduction, and a drift towards uniformity over generations.

The bigger assumption in this article is in what natural selection selects for with humans. For sole individuals in any species, you typically need only to live long enough to reproduce viable offspring. With humans, we've evolved intelligence that has lead to a tribal culture.

That means that natural selection doesn't apply to the individual, but the majority of the species. The best individual of the entire human species is enough to hold natural selection back for everyone else (i.e. vaccines, engineering feats, etc.). That doesn't mean we've evolved to making bad decisions, it just means that the collective knowledge of our species is now being subjected to natural selection instead.

[robotcookies]

"but the inverse is true where too much genetic variability leads to less reproduction"

Two ant colonies A and B. Sugar is abundant in the area and all ants in colony B prefer sugar. 95% of the ants in colony A prefer sugar while 5% prefer peanut butter. The ants that like peanut butter have a higher risk of getting killed because peanut butter is scarce and they must travel further. They also use more energy in getting food. One day a truck drops sugar near the colonies that is poisoned. Colony B is wiped out. Colony A survives on because of the 5% of ants that prefer peanut butter. The queen may die but the surviving ants reproduce.

We all know someone who hates a particular food that most people love. We all know someone who loves a food we think is disgusting. Why don't we all like the same healthiest foods?

I'm not an expert on genes but it's possible that in one species, taste or some other variable is determined by 1 or 2 genes. In another species it may be determined by 8 or 9 genes. This complexity in taste determination may cause more variation in how it manifests. Maybe that complexity causes odd variations to occur over generations. Even as other factors select for sugar in one species, individuals keep popping up that like peanut butter.

[bduerst]

The problem with evolutionary behavior hypotheticals is that it's too easy to craft them to fit a narrative - even with yours the inverse is still true.

The next generation of the species is all PB-seeking high risk ants, which expend more energy & die off faster, reproducing less. Diversity shrinks over generations as natural selection picks off most of the high risk individuals, leading to a uniformity.

This still doesn't apply to humans, because even food preferences weren't really subject to natural selection (outside food neophobia in children), because pre-agricultural humans ate whatever they could get their hands on.

[oldmanjay]

I'm not sure which you are using to form your hypothesis, but I find the notion that intelligence is unnatural, or that the application of intelligence is unnatural, or that the products of intelligence are unnatural to be rather unsatisfying. I take it as a capitulation to marketing forces more than any sort of scientific perspective.

[bduerst]

I'm not implying that intelligence is unnatural. "Natural Selection" is the mechanism by which evolution works.

[ThrustVectoring]

Evolution does not understand and does not optimize for the survival and flourishing of a species. There are three reasons why members of a species vary - either the variance is the result of a beneficial gene, or it's the result of random changes to irrelevant genes, or the species is partway through fixing a new beneficial mutation through the entire population.

So the more complicated explanation as to why humans vary is that humans have a specific set of adaptations for learning and filling their place in the social environment. This is the entire point of childhood - trying things, seeing what works well, doing more of it, and winding up a person who does the sorts of things that work well for them. This means that if you happen to be gifted with bad eyesight and good verbal processing, you'll get early successes at storytelling that lead you to that sort of role as an adult.

That, in a nutshell, is why humans have such variance. It's the result of childhood, which is a specific algorithm that genes use to find and exploit the things the hosts happen to be better at.

And this built-in ability to vary oneself is helpful both for the strong and the weak, the pretty and the ugly, the clever and the dull, and so forth. It means instead of over-fitting for behaviors that work for the strong, the trait enacts a strategy that does strong-person behavior in strong people and weak-person behavior in weak people.