3 ms·

Hard to believe IQ can be gene edited by 60 points simply like we can increase weight of chicken. And that without side effects. Is there any supporting evidenc

by rini17 2y ago

Hard to believe IQ can be gene edited by 60 points simply like we can increase weight of chicken. And that without side effects. Is there any supporting evidence? The fluorescent spunk made me chuckle. But I'd be wary of having such undeniable sign of gene editing on myself. It is going to be a huge issue.

ben_w 2y ago
Caveat: I'm compsci not biosci. Going from 70 to 130 seems plausible, but only that specific range as the nature of the scale is statistical not absolute, and outside 2σ the tests themselves aren't reliable. Going from 130 to 190 isn't really a meaningful score because there's almost exactly 8 people with IQ 190 or higher at any given moment, and which people they are depends on the previous day's rest and last hour's caffine and alcohol consumption. (Likewise in the other direction, single digit IQs are not meaningful because there's a lot more than 8 people at any given moment in the days between e.g. when Alzheimer's causes them to forget to drink water and when their liver and kidneys fail). As for evidence, there's two options for IQ: biology and environment. But even the impact of the environment is tied to your genes, e.g. lead is famously bad for brains, a quick search found some genes are associated with binding to heavy metals including lead to reduce toxicity: https://en.wikipedia.org/wiki/Metallothionein https://en.wikipedia.org/wiki/Metallothionein However, that everything comes down to "can your genes thrive in these conditions?", is not enough to say we can now engineer this. I *do not* make any claim that we actually know enough yet to do this. Proof of principle is much, much, easier than proof of specifics: we knew fusion power was possible in the 50s, we still don't have commercially viable fusion power.
- Ey7NFZ3P0nzAe 2y ago
  I think you are mistaking several IQ scales. IIRC the scale that uses 1 stdev per 15 points saturates around 150. The iq scale that goes to 190 is a different one.
  - ben_w 2y ago
    For modern IQ tests, the raw score is transformed to a normal distribution with mean 100 and standard deviation 15. It doesn't "saturate" at any level, it's just not economically viable to gather enough data for any specific test to be highly reliable outside 2 sigma.

by rini17 2y ago

ben_w 2y ago

Caveat: I'm compsci not biosci. Going from 70 to 130 seems plausible, but only that specific range as the nature of the scale is statistical not absolute, and outside 2σ the tests themselves aren't reliable. Going from 130 to 190 isn't really a meaningful score because there's almost exactly 8 people with IQ 190 or higher at any given moment, and which people they are depends on the previous day's rest and last hour's caffine and alcohol consumption. (Likewise in the other direction, single digit IQs are not meaningful because there's a lot more than 8 people at any given moment in the days between e.g. when Alzheimer's causes them to forget to drink water and when their liver and kidneys fail). As for evidence, there's two options for IQ: biology and environment. But even the impact of the environment is tied to your genes, e.g. lead is famously bad for brains, a quick search found some genes are associated with binding to heavy metals including lead to reduce toxicity: https://en.wikipedia.org/wiki/Metallothionein https://en.wikipedia.org/wiki/Metallothionein However, that everything comes down to "can your genes thrive in these conditions?", is not enough to say we can now engineer this. I *do not* make any claim that we actually know enough yet to do this. Proof of principle is much, much, easier than proof of specifics: we knew fusion power was possible in the 50s, we still don't have commercially viable fusion power.

Ey7NFZ3P0nzAe 2y ago
I think you are mistaking several IQ scales. IIRC the scale that uses 1 stdev per 15 points saturates around 150. The iq scale that goes to 190 is a different one.
- ben_w 2y ago
  For modern IQ tests, the raw score is transformed to a normal distribution with mean 100 and standard deviation 15. It doesn't "saturate" at any level, it's just not economically viable to gather enough data for any specific test to be highly reliable outside 2 sigma.