User Comment Replies

Late-talking kids and "Einstein syndrome"

Has anyone found good data or have an informed guesstimate on what fraction of kids who are non-verbal or minimally verbal at age X, will become verbal by age Y?

I found this old (1987) study by D. Bishop, and A. Edmundson which says:

"87 language-impaired children were assessed at the ages of 4, 4 1/2, and 5 1/2 years on a battery of language measures. In 37% of children, who were termed the "good outcome group," the language disorder had resolved by the age of 5 1/2 years so that children were indistinguishable from a control group. "

There's probably... (read more)

2Steven Byrnes1y

The Camarata book has various related facts & figures (especially at the beginning of chapter 2) although I don’t recall him saying the exact thing you’re looking for. If you have a specific child in mind, then you have much much more information about them than “they are non-verbal or minimally-verbal at age X”. You (hopefully) know whether or not they’re hearing-impaired, and you hopefully know whether or not they have Down’s syndrome, and you hopefully know whether or not they have other other indications of severe autism, and you hopefully know something about their receptive language and social pragmatics, etc. So I’m not sure that the population statistics would be all that useful. (Unless the study population is a decent match to what’s known about the kid, I guess.)

The dangers of reproducing while old

garymm1y10

I believe down syndrome is caused by a chromosome disorder, which is covered in the blog post. Risk does indeed rise sharply with maternal age past ~35.

How to have Polygenically Screened Children

garymm1y30

I just spoke to someone at Orchid who says their basic pricing is $2,500 per embyro for their most thorough testing, but they do have some sort of volume discount. They didn't tell me when the discount kicks in.

2GeneSmith1y

Yikes. That's even higher than I had heard.

The dangers of reproducing while old

garymm1y10

Not sure if you saw the full post at the link, but some absolute risks, such as for miscarriage, are much higher. And for me personally a 1% risk of having a child with a serious mental disability is really scary. Perhaps not for you.

AXRP Episode 22 - Shard Theory with Quintin Pope

garymm2y30

Specifically, after self-supervised pretraining, an LLM outputs exactly the thing that it expects to see. (After RLHF, that is no longer strictly true, but RLHF is just a fine-tuning step, most of the behavioral inclinations are coming from pretraining IMO.)

Qualitatively the differences between a purely predictively-trained LLM and one after RLHF seems quite large (e.g., see the comparison between GPT-3 and InstructGPT examples from OpenAI).

2Steven Byrnes2y

I was thinking: In pretraining they use 400,000,000,000,000 bits of information (or whatever) to sculpt the model from “every possible token string is equally likely” (or similar) to “typical internet text”. And then in RLHF they use 10,000 bits of information (or something) to sculpt the model from “typical internet text” to “annoyingly-chipper-bland-corporate-speak chatbot”. So when I say “most of the behavioral inclinations”, I guess I can quantify that as 99.99999999%? Or a different perspective is: I kinda feel like 10,000 bits (or 100,000, I don’t know what the number is) is kinda too small to build anything interesting from scratch, as opposed to tweaking the relative prominence of things that are already there. This isn’t rigorous or anything; I’m open to discussion. (I absolutely agree that RLHF has very obvious effects on output.)

How to have Polygenically Screened Children

garymm2y30

I have a friend who's a geneticist and is much more pessimistic about the expected impact of embryo selection on a given desired trait. I chatted with him and read some papers he suggested and I now have a lower estimate of the expected value of embryo selection than before.

Very curious to hear from other experts in this field who disagree, especially those who don't have a financial conflict of interest!

It seems it's quite unclear how to even measure how heritable a given trait is, with different methods giving wildly different answers. And even if ... (read more)

4GeneSmith2y

Yes, I hear criticisms like those made by your friend all the time. They aren't particularly discouraging to me. The variation is not that much lower. The standard deviation of any continuous trait is about 70% that of the general population. That's still plenty for selection. The impact from rare genetic variants is mostly NOT captured in today's genetic predictors, and that's a place where the field could improve more in the future. But you don't NEED rare genetic variants for embryo selection to work. You just need a predictor that correlates strongly enough with the actual trait for selection. And for many traits we already have that. The paper you linked showing heritability is lower using relatedness disequilibrium regression was interesting. They don't include intelligence or most of the disease predictors on which embryos are currently selected in the table, so it's hard to give a good estimate of how strongly we should expect it to affect future increases in the efficacy of embryo selection. If I just kind of blindly extrapolate based on the heritability of height in their study and what I know about the heritability of height vs intelligence in other studies, I conclude that the RDR method would generate an estimate of the heritability of intelligence of about 35-40%, which sounds pretty low, but not absurd. It seems pretty likely we could still get a predictor that explained 30% of variance in the trait just using SNP arrays and large sample sizes, so my estimate of the efficacy of embryo selection for intelligence in the future doesn't change much based on that. This is true, but from everything I've read this seems to not matter that much for one simple reason: rare alleles with small effects don't account for that much of the variance in trait values. Like if the RDR method's estimate of heritability is accurate for height, we should expect rare variants to raise the variance explained from 45% to 55%, which would result in a ~10% increase in exp

How to have Polygenically Screened Children

garymm2y10

Thanks for this awesome post. Biggest update for me is "there might be a way to get screening for traits not advertised by Genomic Prediction", but I still have no idea of the cost or the probability of success :-) Would be great to hear from people who have more info on this.

I recently weighed the pros and cons of IVF vs old-fashioned conception and went old-fashioned because:

1. This article claims "different embryo culture media give rise to different birthweights and growth patterns in children" and "children born after ART have altered epigenetic profi... (read more)

2GeneSmith2y

I wrote the section on cost to give you a better idea of the prices involved. Hopefully that's helpful. But I take your point that what is really needed is a “calculator” of some sort where you can input relevant variables and see your expected gains and costs. I am working on something like this at the moment but it may be several months until it's finished. Apart from the randomized control trial looking at different embryo media, I find all the studies presented in this paper to be highly suspect. For example they cite showing that fresh embryo transfer is associate with preterm birth. But the study THEY cite doesn’t even control for the differences in maternal age between parents that do IVF and those that don’t! And surprise surprise, there is a major difference in PTB rates between women in their late 20s and early 30s and those in their late 30s to early 40s. Perhaps I am wrong about this, but my best guess right now is that the downsides of doing IVF are very minor and are massively outweighed by the upsides of embryo selection. The cost is still a big barrier, so I can understand if you don't do it for that reason. Yes, this is still a problem. For IQ gain in particular though the difference is not that big. I believe east and south asians, for example, have an expected IQ gain of about 75% that of Europeans (so like 3.5-4 points vs 5). Maybe that's a big enough difference for it to not be worth it, but it's not a huge reduction.

Are c-sections underrated?

garymm2y60

This article claims "different embryo culture media give rise to different birthweights and growth patterns in children" and "children born after ART have altered epigenetic profiles".

I'm not an expert but I read it and found it quite plausible that there are ways that IVF can cause worse health outcomes. Would love to read a thorough critique of it!

1GeneSmith2y

Interesting study, thanks for the link. Most of the differences listed in this analysis are basically retrospective cohort studies comparing outcomes of parents who sought IVF with those of parents who didn't. I think these are likely to result in spurious hypothesis about negative effects of IVF. We know that parents who seek IVF are disproportionately unhealthy, and as such are likely to pass on their conditions to their children. But I'd make on exception to this generality, which is the differences observed when using different culture media. In particular, the study from Kleijkers et al, which was a double-blind RCT of 836 couples. You almost never get a randomized study that large in IVF. It found a difference in birth weight of 158 grams between children born using two different culture media. That's a difference of about 4-6%. That's a relatively small effect. For comparison, women with preeclampsia who give birth early tend to have babies with birth weights about 350g below that of babies born to mothers without the condition (when matching for gestational age). So while the effect is not very large, I believe it.

Happiness and Children

garymm2y10

Posting this here since I found it useful.

The pains and pleasures of parenting: when, why, and how is parenthood associated with more or less well-being?

I think the really worth-while stuff is in the section "When Is Parenthood Associated With Well-Being?
Exploring Moderators" which starts on page 33.

Parenting and Happiness

garymm2y10

Posting this here since I found it useful.

The pains and pleasures of parenting: when, why, and how is parenthood associated with more or less well-being?

I think the really worth-while stuff is in the section "When Is Parenthood Associated With Well-Being?
Exploring Moderators" which starts on page 33.

Cryonics signup guide #1: Overview

garymm3y70

Note: Alcor is no longer accepting new associate members (according to the page linked to in OP).

For people over 18 years old, membership dues are currently $15 x (age at time of sign up).

Would be great if you could update the post so it stays up-to-date.

Thanks for writing this!

Appendices to cryonics signup sequence

garymm3y90

I just contacted Alcor about paying with a trust and they wrote back "Alcor does not accept Cryopreservation Funding Trusts at this time."
It seems like life insurance and prepayment are the only options.

2mingyuan3y

Thank you for the information! I've updated the post accordingly.

LESSWRONG
LW

All of garymm's Comments + Replies