gwern comments on Why Are Individual IQ Differences OK? - Less Wrong

39 Post author: Eliezer_Yudkowsky 26 October 2007 09:50PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (526)

Sort By: Old

You are viewing a single comment's thread. Show more comments above.

Comment author: Lumifer 14 August 2014 12:49:55AM 0 points [-]

intelligence is metabolically expensive

That's true comparing chimps to humans. I am not sure that's true comparing an IQ70 human to an IQ130 human.

Comment author: gwern 17 August 2014 12:30:59AM *  2 points [-]

If there was no metabolic difference between building an IQ70 brain and an IQ130 brain, why would there be any effects from micronutrient deficiency?

it doesn't seem to me that all the extra-smart people have unusually large heads and eat more than usual.

Remember, expensive isn't limited to adult basal metabolic rates, there are other ways to be expensive; for example, a better brain could suck up tons of iron, iodine, and protein in childhood, requiring lots of meat and fat and seafood, and if a fetus or child's metabolic needs are not met, whups, there goes some myelination (fat), some non-cretinism (iodine), some energy and lassitude (iron and protein)...

Comment author: drethelin 17 August 2014 06:27:49AM 2 points [-]

Also cranial capacity is in fact correlated to iq

Comment author: private_messaging 17 August 2014 06:51:08AM 1 point [-]

If there was no metabolic difference between building an IQ70 brain and an IQ130 brain, why would there be any effects from micronutrient deficiency?

Well, hypothetically, if we have a chip fab, and it has a "micronutrient deficiency", it can produce noisier circuits that don't consume less power, or which would even consume more power.

It would seem that there are some basic requirements which need to be met to build the brain correctly, requirements that are proportional to the brain volume, with no gains from exceeding those requirements. One could further hypothesise that those requirements are met in almost all "IQ130" brains.

Comment author: gwern 17 August 2014 04:07:09PM 1 point [-]

Well, hypothetically, if we have a chip fab, and it has a "micronutrient deficiency", it can produce noisier circuits that don't consume less power, or which would even consume more power.

Sure. Chip fabs probably even have 'micronutrient deficencies' in a very similar way - if you can't get enough of the exact right exotic element or mineral for say doping semiconductors, the engineers can probably work around it but won't get as power-efficient or fast a chip. (Now I'm imaging correlating chip fab 'brain damage' to global commodity prices...)

Comment author: private_messaging 17 August 2014 04:25:44PM 1 point [-]

I don't think deficiency in dopants can ever arise, though, as they're used in incredibly tiny amounts.

For micro-nutrient deficiencies the issue is often not so much with obtaining the micronutrient as with the lack of craving for it. We can smell iodine, but we don't crave it when deficient, so we didn't have seaweed and the like as a high value spice which everyone craves.

Comment author: gwern 17 August 2014 04:55:29PM 1 point [-]

I don't think deficiency in dopants can ever arise, though, as they're used in incredibly tiny amounts.

But they need to be extremely pure and in the right form to be used. (If just having the raw material was enough, no one would ever die of thirst drinking salt-water and plants would never lack for nitrogen.)

We can smell iodine, but we don't crave it when deficient, so we didn't have seaweed and the like as a high value spice which everyone craves.

Maybe we can smell very large quantities of iodine, but can one really smell deficiency-relevant amounts in seaweed?

Comment author: private_messaging 17 August 2014 05:19:40PM *  1 point [-]

But they need to be extremely pure and in the right form to be used.

Yeah, but so is silicon (and even more so in terms of purity), and there's million times the silicon. I think industry is sort of similar to the ancestral animal that is eating a diet where it obtains enough micronutrients alongside macronutrients. But if we were to try to build a self replicating factory on the moon... we'd probably just ship anything like this from the earth.

Maybe we can smell very large quantities of iodine, but can one really smell deficiency-relevant amounts in seaweed?

The RDA is 300 micrograms per day, 0.3mg, and if I have a 3% solution of iodine, that's 10mg of that solution. 1 drop of water is 50mg , and I think you could easily smell 1/5th of a drop (or a drop 58% the size of a regular water drop), but probably not if its mixed up in food. Still it is close enough that given an absence of such adaptation I wouldn't expect any other complex adaptations to lack of iodine. edit: plus we can detect seaweed without smelling iodine itself.

Comment author: gwern 17 August 2014 10:23:34PM 1 point [-]

The RDA is 300 micrograms per day, 0.3mg, and if I have a 3% solution of iodine, that's 10mg of that solution. 1 drop of water is 50mg , and I think you could easily smell 1/5th of a drop (or a drop 58% the size of a regular water drop), but probably not if its mixed up in food. Still it is close enough that given an absence of such adaptation I wouldn't expect any other complex adaptations to lack of iodine.

Oh, you mean that smell is one of the easiest adaptations for dealing with a lack of iodine, and since we don't have a smell adaptation, we don't have any more complex adaptations? Sure, I agree with that. Tweaking smell sensitivity seems to be pretty easy. Humans aren't dogs, but we can still smell some things at very low thresholds. For example, t-butyl mercaptan can be smelled at 0.3 parts per billion, it seems. (Although now that I think about it, what on earth was the selection pressure for that? Maybe some smell thresholds are just random.)

edit: plus we can detect seaweed without smelling iodine itself.

Seaweed has an awful lot of stuff in it; we could be smelling any of the components without smelling a particular component. You can easily smell tobacco, but can one smell important parts like nicotine?

Comment author: private_messaging 18 August 2014 10:22:31AM *  2 points [-]

Yeah and also we used to have a much better sense of smell. Smell is also used for identification (it seems, in most mammals except humans) and for that the more compounds you detect at the lesser concentrations the better. With mercaptans I think it'd be related to bacterial toxins in rotting meat, which kill at absurdly low concentrations. We can't detect poisons themselves but we can detect other stuff that goes along with it.

Seaweed has an awful lot of stuff in it

Yeah, that's the point - those who live next to the coast (within what ever range you can have a preventable iodine deficiency - right next to the coast maybe nobody ever gets it, but some distance inland...) could evolve taste for seaweed based on some other compounds or their combination, to get iodine.

Comment author: gwern 19 August 2014 12:56:28AM 1 point [-]

Yeah, that's the point - those who live next to the coast (within what ever range you can have a preventable iodine deficiency - right next to the coast maybe nobody ever gets it, but some distance inland...) could evolve taste for seaweed based on some other compounds or their combination, to get iodine.

Would that fit food transportation patterns? I've never heard of seaweed being collected and shipped in large quantities. Most communities were generally pretty self-sufficient as far as food goes. If you don't have optional access to seaweed, there's not going to be anything evolution can exploit - you'll just get entire communities of deficient people.

Which apparently did exist historically: https://en.wikipedia.org/wiki/Cretinism#History The consequences are pretty severe: starting from birth, retardation, small size, frequent infertility. And lots of variation within villages and between regions - which sounds like there would have been a lot of selection pressure for particular food preferences, within-village and between-village, but there doesn't seem to've been any kind of adaptation.