Heritability != DNA.

Do you expect embryo selection startups to fail? Come on. I know you're smart enough to have heard about GWAS.

I'll bet you 3:1 odds embryo selection works. If you're serious about your anti-hereditarian position, take me up on my offer.

▲

tptacek 5 days ago | parent | next [-]

I think embryo selection companies, to the extent they don't actually result in people selecting for embryos with autism, are brilliant products. They're "magician's choice" setups: embryo selection promises single-to-low-double-digit improvements in metrics that aren't fully evaluable for over a decade after the product is paid for, on metrics with huge variability and low test-test reliability. The people paying for the products are generally upper-income and already predisposed to invest in educational achievement, which is the actual outcome the customers care about to begin with. It's a can't-lose proposition for the vendors.

So, no, of course I'm not going to take you up on that bet. It's like betting against Bitcoin. I think Bitcoin is a farce but I'm not dumb enough to short it.

I'm not an "anti-heriditarian". I think there's probably a lot of value, long term, in embryo selection for things like disease avoidance. I also believe there's natural variability in cognitive ability; I don't believe all people are "blank slates"; that's a caricature (or, if you like, a deliberate wrong-footing of people who reflexively reject psychometrics and genetics for ideological reasons) of the actual concern I have.

Finally, I don't know what anything you said has to do with what I said. I said, very simply, "heritability != DNA". That's an objective, positive claim. Was this bet your attempt at rebutting it?

▲

quotemstr 5 days ago | parent [-]

It's interesting: to the extent the orthodox position acknowledges that genetically mediated trait inheritance exists, it cases it in terms of "disease" and "treatment". It's morally wrong to select an embryo for height, but acceptable, even imperative, to use genetics to screen for "shortism".

I'm sure you've read Gwern's essay on polygenic trait inheritance. I'm not sure repeating the literature would be productive here. We have every reason to believe that embryo selection and genetic engineering more generally won't just "cure disease" but make us taller, smarter, more beautiful, and longer lived -- and there's nothing wrong with that.

Of course there's a lot of variability. At some point technology will improve to the point that denying the effect exists will seem ridiculous, although I'm sure plenty will try.

I will say, though, that downplaying trait inheritance and the way genetics is the mechanism for this inheritance produces models that don't predict reality nearly as well as models that incorporate hereditary via genetics, and especially when it comes to education, we're throwing public money down the toilet as long as we make policy using inaccurate models.

	▲	tptacek 5 days ago \| parent [-]
		I have no idea what the first paragraph you wrote means. I don't have a moral issue with embryo selection. Select them for eye color for all I care. I don't know what any of the rest of this has to do with what I said. I ask again: are you writing all this by way of declaring that "heritability == DNA"? That's a straightforward discussion we can have. Why avoid it?

▲

kasey_junk 5 days ago | parent | prev [-]

What is the success criteria of the bet?

	▲	tptacek 5 days ago \| parent [-]
		I'm just happy for an opportunity to rattle off my embryo-selection rant! :)