De novo and mosaic copy number variants contribute to ~6% of cases and these are often different unique to single twin even if MZ: https://pubmed.ncbi.nlm.nih.gov/37601975/
While technically genetic, to the extent that these are not shared between MZ twins, they would show up as non-shared environment in twin studies, right?
There are also instances where "genetics" is environment, if we're talking causal mechanisms. Propensity for cannabis use comes to mind for schizophrenia.
This study doesn't say what you're representing it as saying, which isn't a thing that makes sense. CNVs are much more common in people diagnosed with SZ than the general population, though this is a lot less true for SZ than it is for most neurodevelopmental things with defined disorder names (it's most markedly true for intellectual disability; its truth status for autism is intermediate between SZ and ID, and I think would be similar to SZ, or lower, if you properly screened out all the diagnostically substituted intellectual disability). CNVs would be shared between MZ twins unless something very, very strange happened, though. This study is talking about a particular form of strangeness that could happen in that circumstance, but it's ascribing it as much rarer than normal CNVs, and speculating it would probably result in an attenuated phenotype. Low-level mosaicism gets weird quickly, and it's difficult to tease out the degree to which it 'matters' vs is a chance finding, and if things can be chance findings even if they look like they should 'matter' (this is a big problem re. 'pathogenic' CNVs -- many of them, especially duplications, don't actually seem to do much in most people with them, such that the line between a 'pathogenic' and 'non-pathogenic' CNV is not naturally defined).
Perhaps you already have this in mind, but I think a lot of the non genetic component of why someone does or does not develop a disorder like schizophrenia is stochastic (chance) rather than whether they were exposed to certain environmental risk factors.
Kevin Mitchell describes this well:
‘Whether that risk manifests as actual disease, and which symptoms emerge, is probabilistic, reflecting additional non-genetic factors. Given the lack of evidence for systematic environmental risk factors, this diversity of outcomes may reflect intrinsic stochastic variation in the trajectories of brain development. Chance thus plays a substantial role in determining which outcome from a wide possible range is actually realised by the processes of development in an individual.’
There was a post about this a few years ago - "environmental factors", when you examine them, turns out to mostly be code for "random luck", not "how good a school you went to" or whatever.
There are definitely some well-replicated environmental risk factors (heavy cannabis use in adolescence springs to mind), but I don’t think they can explain the variation in risk
I mostly agree with this, but IIRC the environmental component of schizophrenia variance is about half shared and half non-shared, suggesting there's more non-stochastic factors than usual.
My guess is these will be boring things like poor health and poverty, and not anything that will be specific to schizophrenia.
It's good to bear in mind that there are important ways in which heredity influences environment: Kids who are physically attractive and above average at motor coordination are more liked by their peers. Physically attractive kids are perceived by teachers as smarter, and are favored. Kids who really are quite smart, even if not physically attractive, are seen as smart by teachers. Simple things like this have a big influence on how many smiling faces kids see in their early years, and that number's a powerful feature of social environment.
I'm pretty sure poverty and poor health are going to be a result of having a schizophrenic parent. Consider the poverty cause of being related to a borderline schizophrenic person.
It'll be a contributor, but as Scott points out, the extreme rarity of schizophrenia means that most schizophrenics will not have close schizophrenic relatives themselves, even when looking at identical twins, even though genetic influences on the trait are very strong. So nature-as-nurture effects in that sense are likely to be minor, I think.
Others in the thread have suggested that heavy cannabis use is a significant contributor, though, and I think the literature suggests mild-to-moderate shared-environment influences on substance abuse disorders (maybe 20-30%), so maybe that indirectly accounts for a few percent of the variance.
"Random luck" as I would understand the phrase sounds a lot like something that could resaonably be described as "environmental". It's just something stochastic that you wouldn't necessarily expect to have a significant effect like "did you get chicken pox at 18 months or 20 months" rather than a more long term "environmental" effect like "poverty".
The really interesting thing is that a lot of "non-shared environmental" effect could actually be measurement error. This is probably less bad in the case of schizophrenia than many conditions as I understand it is somewhat two-peaked; but there must be some fuzziness; people who might or might not get a schizophrenia diagnosis depending on the doctor assessing them.
Perhaps I'm misunderstanding your point - "non-genetic" and "environmental" sound like they're being used above as equal-by-definition, but it sounds like you have a different notion of "environmental" in mind. What, in your view, does "environmental" mean?
I think ‘non genetic’ can be divided into ‘environment’ and ‘random variation’.
Environment encompasses everything from the uterine environment, birth complications, early life trauma, racial discrimination, drug use, head injury, etc. Even if the environment is completely accounted for, I believe there would be a proportion of the variance that is unaccounted for - which is due to stochastic processes / random variation / luck.
Is that a claim about the practical limits of our ability to enumerate particular environmental factors, or a claim about irreducible randomness in physical processes?
Leaving aside quantum mechanics for the moment, these two are the same thing, because the practical challenges in enumerating particular environmental causes of a given outcome go at least exponentially with the weakness of the causative mechanism (say the phase of the Moon or the position of α Centauri when neuron #123456789 was growing its 987-th dendrite).
You absolutely can assume that the ‘non genetic’ component is ‘environmental’ if you define environmental as "everything non genetic." I believe that this is effectively what is being argued.
> The really interesting thing is that a lot of "non-shared environmental" effect could actually be measurement error.
Yes! Schizophrenia is not something easy to define or verify objectively in the first place, unlike having a peg leg or a glass eye. And even with more 'mechanical' diseases like blindness or motor dysfunctions it is far from simple, as the horrors of Paralympic qualification manuals show.
A problem related to measurement error which ends up in the rubbish bin of "non-shared environmental effects" along with the former is the arbitrariness and messiness of indicators which researchers are forced to use from lack of anything better in available data. E.g. there is little or none measurement error in EduYears, but neither does it reflect in a simple way the abilities or qualities of characters that we are really interested in, and there is no reason to believe that its relationship to the latter remains reasonably constant with time (in fact we have good evidence that it does not).
Geneticists call everything which isn't genetically inherited 'environment'; it's baked in to the terminology.
I guess you can argue about this, but 'random' is not straightforward to disentangle anyway, it's generally 'stuff we don't have all the details for', so it's a pragmatic decision.
It's very similar to the old: a <cancer|whatever> quick-test has a 1% false positive rate. Your test is positive. What is the chance you have <cancer|whatever>? 99%?
when only a very small percentage of the population have the measured condition, statistics are very un-intuitive.
I had to explain that to an actual medical doctor in 2020. Had a positive covid test when we A) tested the entire county employee population without any symptoms, and B) prevalence in the county was extremely low. I told him it was probably a false poz, he insisted the false poz rate was very low. I tried to explain to him that if you test everyone across the board with no independent reason to believe they have the disease, the false poz rate would be higher (as a prosecutor at the time, I of course had heard of the Prosecutor's Fallacy.) He called me back a week later with the lab PCR test, and sheepishly admitted I was correct and it was a false positive and he sure was surprised because they'd been told that wouldn't happen. Meanwhile my false poz wasted a week of my life, caused an inordinate amount of stress on my marriage, and forced cancellation of a monthly circuit court docket resulting in a massive logjam of criminal defendants.
He didn't even have to be familiar with statistics to get this right, it's intuitive. Other doctors early in the pandemic were stating loudly on TV that you shouldn't give the test to people without symptoms (when saying that was useful to make Trump's staff look wasteful or foolish by testing.) But then the incentives changed to "make as many people afraid of this as possible, for their own good" so suddenly basic logical reasoning has to be ignored.
Hmm I'm all for ragging on doctors but I think it was an ok precaution at the time when the vibe was 'hey, lets shut everything down and see if we can eradicate this thing'? Testing in asymptomatic populations was assessed from BMJ and Cochrane reviews below, if you had a positive result from a lateral flow you were ~50-60% likely to have COVID. Agree that it's not great, there was a lot of hyper pre-cautionary stuff done at the time, but I wouldn't throw the bucket out with the bathwater on this one?
I guess my biggest problem is that when presented with an actual logical reason he was wrong, he just slavishly stuck to the script. There was so much derision directed at people like me who actually tried to learn things and apply reason and logic to them during covid, we were mocked for "doing our own research" rather than just listening to authorities, but in at least that one situation I was smart enough to know the authority was wrong about how statistics work and that I, the layman, was correct.
I could've respected it had they been honest and said "ok sure, you're right, the false positive rate is much higher due to very low prevalence in your area and the lack of any symptoms or contacts to supply an independent reason to suspect you have it, but the chances you're infected are still high enough that we would prefer you to be careful for 2-3 days just in case, and if you do feel these symptoms call me." Or heck, just admit that you tell the stupid people it's 99% accurate because they won't take it seriously otherwise, but when I call you on it acknowledge that. Instead we had the medical experts outright lying to my face about the statistics of asymptomatic testing, something that you did not have to be a doctor to understand, that any reasonably smart person could figure out. They did not seem to care that it was a lie, apparently trusting that anyone smart enough to discern the lie was going to play along, but nobody sent me the memo. I was particularly bothered because SSC was down during most of covid and I could not find any rational discussion of any of this, it felt like two years of my life where almost anything anybody said was designed to induce behavior rather than to correspond to reality.
Eventually tests *were* developed that had an extremely low false positive rate (in exchange for a not-so-great false negative rate). For a probability class I was teaching at the time, I looked up the error rates of the Covid tests that were required from students every week. I no longer remember the exact numbers, but the false positive rates were something like one in a few thousand. (So, if you ask all the students to take a test, you get false positives in the single digits; when there are about 50 positive results a a week, most of them are true positives.)
Of course, we need something more than simple probability when we try to interpret what this means. Does a positive result mean "contagious right now" or "contagious last week" or "about to become contagious"? This depends on the test as well.
The first argument is also often used to explain why in elite challenges (e.g. interviews to top companies) luck matters more than skill: since everyone who has a chance at all is already in the top percentile of skill anyway, the variance introduced by luck matters relatively more and is the main factor.
This becomes more true the more selective the process is. Since schizophrenia, like Harvard, is very selective, we expect the difference between people who get it and people who were strong candidates but didn't to be mostly luck (or environmental factors).
Yes, but this depends on the assumption that it's hard to distinguish (or not necessary to distinguish) skill among the top percent. I agree that's the case for soft skills--like communication, leadership, etc.--but hard skills can often be measured even at 4+ standard deviations out (e.g. consider chess or math competitions, where there are large gaps in performance even at top competitions).
I think this helps explain why highly selective software companies tend to use difficult algorithm questions in their recruiting, even though soft skills (architecture, code quality, etc.) are arguably more important in many cases. They are testing a hard skill that can provide a signal to distinguish the top 5% from the top 1% of candidates. Whereas soft programming skills aren't so easily distinguished that finely.
Skills vary by how easy they are to distinguish, but all skills are much harder to distinguish within the top 1% then at lower ranks (because skill levels tend to cluster at the top, due to normal distributions having thin tails).
For example, if the top two chess players in the world play a match, it's roughly a coin flip (actually worse than that, it's usually a draw). If you pick two random people from your middle school chess club, the better player has much better odds of winning than that (even as black).
Ad the nazi eugenic argument: I don’t think many schizophrenics would have children once the disease developed, especially back then when there weren’t any drugs ameliorating their condition. So the nazis killing them did not change the national genetic composition much.
Come to think of it, if the people they were killing off were long-term institutionalized populations, almost none of those people were going to have children ever again. Also because people had children young back then and schizophrenia has latish onset, a lot of these people had already reproduced.
Probably what's happening in the bigger picture is that there is a balance between schizophrenics having fewer children and de novo mutations occurring and increasing genetic risk. This is very pronounced for conditions like severe autism, there's evidence for it with schizophrenia as well e.g. https://pubmed.ncbi.nlm.nih.gov/34903660/
Another possible factor in both schizophrenia and autism is that some of the gene varients that can cause them may even be beneficial if they do not actually cause the disease, further reducing selection pressure. This is a bit speculative but I don't think it's unreasonable to consider these conditions as pathological extremes of normal human brain function.
I think the idea is independent of whether you consider SCZ to be on a continuum with typical brain function. The pleiotropic effects of the relevant genes would likely drive the positive selection. They could have effects (spitballing here) like slightly increasing the density of calcium channels in the basal ganglia or slightly downregulating the synthesis microglia, that don't necessarily contribute a phenotype that has recognizable continuity with schizophrenia.
"if the people they were killing off were long-term institutionalized populations, almost none of those people were going to have children ever again."
If so, does that mean "eugenics" in this case was a fancy-sounding excuse for cutting costs?
Totally. In the Nazi ideology, the people were thought of as an organism (Volkkoerper), with individual humans subservient to that greater whole. So even if one takes aside their obsession with racial purity and genetic health, killing off people who can't pull their weight is totally compatible with their twisted mindset.
I guess the reason why they did not systematically kill all their war invalids or retired people is less ideological and more practical.
Your interesting point got me reading about this. They killed people with Down's syndrome, senile elderly etc. who couldn't possibly have contributed to the gene pool. Cost cutting was clearly a factor, as well as the ideological aspect quiet_NaN describes. I would add that once you decide some people are too disgusting to survive, it's logical that will occasionally extend to your own group.
The other interesting thing I learned in my reading is that euthanizing the disabled raised more overt opposition in Germany than pretty much anything else the Nazis did. A while back some people on the internet condemned the people of Russia for not rising up to stop the invasion of Ukraine, and then I read an interesting article that argued that no government had ever been overthrown for cruelty to outsiders. You can find examples of leaders overthrown for failing at wars, but not for being cruel in them. For example, Milosevic lost power because he was beaten in Kosovo, not because he was cruel in Kosovo. The pattern holds in Germany. There were people in Germany willing to risk their necks to prevent the death of their own country's invalids, but far fewer willing to do so for Jews, Slavs, Frenchmen etc.
Schizophrenia doesn't have a latish onset, though. In fact age of maximum risk for a first psychotic episode corresponds very well with age of maximum fertility.
From what I can tell, typical onset is in your 20s. So yes, it's certainly possible to have children after you have schizophrenia but ~1940, there's a decent chance you've already had some. Scott's spreadsheet exercise assumes that that killing all of the schizophrenics prevents them from having kids, but clearly many will have already had some before the Nazis kill them.
Condemning the Nazis for not applying a gene-centric view more broadly seems unfair. If they'd known all that has been learned over the last ~75 years, they could have sterilized all relatives up to first/second cousins.
I do not think "understand what is going on before you start murdering people" is an unreasonable standard to have. (Of course, I prefer the simpler "don't murder people".)
Of course, the Nazis would claim that they never intended to wipe out genetic diseases (or at least the people with them) in that half-generation in which they ruled, they were famously thinking about a millennium, which was fortunately cut short after some 12 years.
In the world today, male schizophrenics have an average of 1 child. I don't know any studies of the timing of these births, but I also assume that they generally have children before onset. But this means that even if the Nazis reduced their fertility to zero, they would only have twice the effect that occurs every generation in the modern world.
Can we summarize the confusion as being something like base rate neglect? The prevalence of schizophrenia is so low that even knowing your twin has it doesn’t make your odds that high, because they’re so so so low to begin with. Going from 1% to 20% chance of having schizophrenia is a massive step up.
The obvious parallel is autism, which is also a neurodevelopmental phenomenon with weird complex messy influences, a huge spectrum, and a tendency for many people technically on that spectrum to evade diagnosis due to severity bias. Concordance for ASD in monozygotic twins is consistently far higher than for schizophrenia.
I think the basic answer is that the core of the schizophrenia spectrum is schizotypy, not psychosis -- that is, looking at schizophrenia and expecting it to tell you much about "what it means for someone to have this" is a bit like looking at someone with skin cancer and extrapolating it to the entire population of Queensland, Australia. (Highest skin cancer rates in the world!) The match between "the diagnostic category of autism" and "the Truth that the category is trying to get at" is still, um, messy, but there has been for many years a general understanding that people without the most 'severe' possible presentation can still be autistic.*
I also think, yeah, 'schizophrenia' itself has a lot of meaningful and significant environmental factors that don't apply to autism. A lot of this is just an age-of-onset thing. I don't think this is all of it, but it's certainly a lot of it.
I assume MZ twins of people with an SZ diagnosis tend to be pretty schizotypal, but I expect some of them aren't very schizotypal, because variance between monozygotes is weird and somehow happens all the time. (Also, because sometimes people get diagnosed with a functional psychosis when the more parsimonious etiology is 'injecting meth into your eyeballs' or similar.)
*The concept that historical (e.g. Kanner's) concepts of autism are "severe", in terms of "very disabled" as opposed to "very autistic", is not really accurate. I'm not entirely sure when the concept of autism and intellectual disability being genuinely (rather than misdiagnosedly) comorbid came in, but it wasn't Kanner's idea. Asperger actually believed it more, but he still only believed it in terms of what we'd call mild or upper moderate ID, reading between the lines (Frith translates him as discussing "severe retardation", but he describes that in ways that sound like mild, not severe, ID).
Perhaps because of the more binary aspect of schizophrenia?
If the 99th percentile is psychotic but their sibling in the 98th percentile isn't, that's discordant. However, if the 99th percentile in autism is obviously autistic, then his sibling in the 98th percentile who's borderline autistic probably gets diagnosed too.
This also gives a nice intuitive explanation for the nazi paradox. The pool of people at risk of schizophrenia is much larger than the pool of people who have it. Wiping out the pool of people who have it doesnt put a dent in the at risk pool, who can then pass their at risk genes to the next generation.
I am someone who takes great interest in scientific findings outside his own area of expertise.
I find it rather disheartening to discover that most of it is rather bunk, and that you need to be an expert in the field, writing simulations, to prove how bunk it is. I find it even more disheartening to discover that almost no-one who is an expert in the field bothers to notice or debunk the bunk.
So I find it very refreshing (heartening?) that you are doing so. This blog is a goldmine for anyone who wants to see how things are going in Psychology, EA, forecasting, and other topics where you point your mind.
I guess I might as well make this a question for the commentariat... what techniques do you find for helping discover great insights in fields where you have an interest, but no background, and are thus susceptible to falling for the bunk? Another way of putting it, if you were in charge of writing summary articles about all the cool stuff in various fields, how would you make sure you weren't being duped?
Torrey's arguments with the twins was actually a little more subtle than I have here (although some of his partisans made the less subtle one I made here). He first of all argued that the 15% twin concordance rate was the best (I haven't read the studies and can't comment), and that some infectious diseases (eg tuberculosis) have twin concordance rates around that level. I'd be interested in learning why that should be - maybe because there's a strong genetic component to how well the immune system does at fighting them off? - and if his point was just that it might still be infectious, fair. But I think he was implying something stronger than that.
Interactions between genes and environment are assumed to not exist in the standard model of heritability. That is, when you say something is "X% genetic", you are assuming no interaction between genes and environment, no GxE term.
Instead, the standard model assumes a SUM of a genetic effect and an environmental effect, like you did in your spreadsheet. A product is not allowed. This means that "AND" gates like "genetic susceptibility AND catch the right infection" are ruled out by assumption in the model.
They are from PSY 5137 - Introduction to Behavioral Genetics - Fall 2020 by Matt McGue, PhD Regents Professor, University of Minnesota, Department of Psychology
I think this is a really rare complete miss for you, for the reason that “is mostly genetic” is just a very naive way to think about any disorder. Both Torrey and Aftab’s articles are specifically trying to reach past this kind of naive conception to point at the very thing you discovered in your analysis: that while genetics are a primary risk factor, they simply are not sufficient to cause schizophrenia alone, and waving a hand at ‘polygenic background’ doesn’t do anything to explain why, even in the presence of enormous genetic risk, there are critical environmental factors which it would be wise for us to think of as causative (even if they wouldn’t be necessarily causative in the presence of a lower genetic risk profile).
Honestly I don’t understand the reluctance to embrace the model that “maybe if hundreds of genes interact in unclear ways to create an individually highly variable risk, and it turns out identical genetic risk profiles will or will not develop schizophrenia based on environment, it’s not so smart to insist on thinking of causation as genetic.” It’s not like this explanation elides genetics, it just doesn’t throw in the towel on causation once we discover that causation isn’t purely, or even isn’t ‘mostly’ environmental.
There is a level where clinging to your mental model as “mostly this, therefore because of this and should be talked about as this” is unhelpful. Here’s a probably overloaded metaphor. If I build a building, and it collapses due to wind load, I don't just stop at saying the cause was wind. I talk about how the designers knew that wind was a primary risk for this sort of building and the design was insufficiently prepared for wind. What we are finding out is that certain buildings by their nature are more exposed to the wind, and are going to require additional reinforcement. Now we need to know what the specific scenarios are whereby winds arise and interact during the storm to collapse all buildings that collapse via wind; knowing that there is a lot of risk for particular styles of construction helps us, but it isn’t enough to say “wind did it” and put your fingers in your ears when it comes to the other factors that made wind a critical actualized risk rather than a potential risk. It is proper in this scenario to say that the cause is poor engineering, which should account for the wind and doesn’t. Wind is everywhere and we aren’t going to be eliminating it anytime soon, right? Likewise with the literally hundreds of genes with unclear interactions that create risk for psychiatric disorders.
Perhaps we need better ways to express "more nature than environment, and it happens frequently" and "more nature than environment, but it happens very rarely". Because in the latter case, a tiny change in environment could have prevented the outcome (but so could have even a tinier change in the genome).
I will reuse your metaphor with bridges: imagine that 99% of bridges are built so that no wind could possibly break them, and 1% of bridges are built so that a very strong wind can break them... but strong wind happens rarely, so 95% of the less resistant bridges are lucky enough to never break..
It makes sense to say "the bridge broke because it wasn't built well", and it makes sense to say "the bridge broke because of the wind". And it may kinda make sense to say "a bad bridge breaks in 5% of weathers, while a bad weather only breaks 1% of bridges, therefore the quality of the bridge matters 5x more than weather".
Torrey's arguments are, basically:
1) It is nonsense to say that engineering matters, because even if we have 2 identical bridges built by the same company, most of the time both remain intact, sometimes 1 of them breaks and the other does not, and only extremely rarely do both break.
2) Last year we have demolished all broken bridges, instead of fixing them. And yet, this year a few bridges broke again. If engineering matters, how is it possible that those bridges that broke this year didn't already break the last year?
And Scott responds:
1) Even the less resistant bridges break rarely. So if we have 2 identical bridges, if they are more resistant, neither breaks, but even if they are less resistant, still the most likely outcome is that neither breaks... and the second most likely outcome is that one of them breaks... and only exceptionally both break. That does not contradict that 99% of bridges are built to never break, and 1% depend on luck but mostly get lucky.
2) Even a less resistant bridge only has a 5% chance to break, which means a 95% chance that it will not break... followed by a 5% chance that it breaks the next year. The difference between 0.05 and 0.95 × 0.05 is small; it is 5% vs 4.75%.
I don't know if I"d call it a complete miss, the spreadsheet low-cost simulations were pretty enlightening. But I'd still agree on the common sense of the meaning "mostly genetic".
Even if you can show in terms of statistical definition that genetics contributes 80% of the effect, a situation where only 10% of people with the faulty genes will get the condition is still *not* intuitively seen as "mostly genetic". In the way that actually matters to the individual, it's more like "genetics and environment need to combine just right".
Regarding Torrey in particular: well, I try to say very little about the sort of person who's a first-degree relative of someone with Scary-Name Neurodevelopmental Disorder that blatantly represents the far end of population variance, and is absolutely dead insistent that Scary-Name Neurodevelopmental Disorder traits represent the face of evil itself, given we know that in all these cases first-degree relatives of people with Scary-Name Neurodevelopmental Disorder generally have a ton of Scary-Name Neurodevelopmental Disorder traits. But I can see why he (and many others for various SNNDs) is so enraptured by anti-genetic explanations for this particular SNND, yes. The fact Torrey is a disciple of Gajdusek is something I've been thinking about lately -- Gajdusek is one of the people with a claim to discovering prions, and is, uh, a 'character' (a self-identified "pedophiliac pediatrician", amongst other things). I'm not entirely sure what to make of the Gajdusek connection, given how little anyone focuses on it. Gajdusek is, by the words of anyone who interacted with him, not someone you forget easily.
In the specific case of schizophrenia, the "all first-degree relatives of people with SNNDs basically have SNNDs themselves" problem stands out somewhat. Research on SZ is really bad, consistently, because it comes from a psychosis-focused perspective and not a schizotypy-focused perspective -- there's increasing recognition of concepts like the "psychosis spectrum", but the keyword still tends to be psychosis. "At-risk mental states"/"prodromes" are some of the least-bad people have gotten, but are still terrible, because they're written under the assumption people diagnosable with those terms will *usually* develop psychosis, which is clearly untrue. Also, holy shit, have you seen a 'prodrome' test? They're awful. They read like they were written by someone trying to strawman the worst ends of psychiatry. "Do you have strong feelings or beliefs that are very important to you, about such things as religion, philosophy, or politics?" "Do you daydream a lot or find yourself preoccupied with stories, fantasies, or ideas?" "Do you usually prefer to be alone?" "Do you find that you have trouble getting motivated to do things?" Well, bad news, kid -- you might be a latent schizophrenic! Here, enroll in this trial where we give you antipsychotics.
Anyway: I think schizotypy, proper, is pretty intrinsic, though it's also modulated by other factors (e.g. developmental stage, paranoia-inducing experiences). When you're looking at a particular extreme subset of the schizotypal population, this is going to be muddled. I'm not sure if people who end up with an SZ diagnosis are particularly representative of the schizotypal population, not just in terms of "being on some extremes" but in much deeper and more fundamental senses. I have the strong and consistent impression that SZ vs bipolar diagnosis primarily loads on things other than presentation of the psychosis alone, for instance, and that's already selecting for the minority of people who have such obvious and serious psychosis that they get diagnosed with something and stick around the system for long enough other people notice.
"the strong and consistent impression that SZ vs bipolar diagnosis primarily loads on things other than presentation of the psychosis alone" - how come? I'd think that differential diagnosis is comparatively easy. If a paranoid-hallucinatory syndrome occurs with obvious manic symptoms and a history of episodic major affective disorders, which happens extremely rarely, I'd call it schizoaffective disorder.
Something I've wanted to mention here for a while:
One point about twin studies in general is that often the twins are not separated at birth, but several years afterwards. So overall this supports the lead-crime hypothesis, because otherwise, the environmental effects are too large, and lead proves too much (and I first heard this from the "Lucifer Curves" author). But it also means that MANY of our twin studies could be confounding "early childhood environment" with "genetics".
Of course, GWAS isn't susceptible to this bias, so when the twin studies agree with GWAS, we should probably accept the overall point.
Classical twin studies use twins raised together, and estimate heritability by exploiting the fact that identical and same-sex fraternal twins have the same upbringing but different degrees of genetic similarity. Separated-twin studies are interesting, but much less common.
Thanks to both for the clarification. So to summarize:
1) twin studies: same environment, identical-vs-halfshared genetics
2) adoption studies: different environment, identical-vs-halfshared genetics
I do think it's worth remembering that almost everything is horribly underpowered, because researchers assume that "medium" effect size is something that's subtle to notice. In reality, a "medium" effect size is something like "people who like eggs eat egg salad more often than those who don't" (i.e., something that we knew before we ran the experiment). See, eg http://datacolada.org/18
So it's worth remembering that lead is a HUGE effect size... and indeed we (should) have known that lead causes serious problems long before we started this line of research. We knew it from similar metals (eg mercury and the mad hatter), from lab experiments on mice, from case reports where workers exposed to lead went crazy. It would be flabbergasting if lead DIDN'T cause crime. But almost everything else is going to be smaller than the lead effect, and going to require many hundreds of observations per cell (cf link)... which very few studies have.
But with the convergent lines of evidence, and things like Hatemi's "Not by twins alone" (claiming 40k participants), I agree with the overall take about how large genetics is.
I'm not sure what I'm supposed to take from this. Are you arguing that the lead-crime link is due to a handful of people who received massive doses of lead and killed thousands? That would be really interesting if true! But I'd definitely like a citation if so.
Otherwise, it sounds like you're saying "the dose makes the poison", which is true, but in casual or charitable conversation normally one assumes the relevant background. I.e. the effect of lead is large RELATIVE to the amounts we were putting into the environment, and we can have a large societal impact by reducing the toxin.
"we (should) have known that lead causes serious problems long before we started this line of research."
The effects of tiny amounts of lead are not dramatic until you do research at huge scale. You can't just point to big effects from huge doses and assume tiny doses are the same because there's so many cases where that approach fails.
If I understand heritability correctly, 80% heritable means that a 1-SD increase in genetic susceptibility gives you a sqrt(0.8) = 0.89-SD increase in risk, and a 1-SD increase in environmental risk factors gives you a sqrt(0.2) = 0.45-SD increase in risk. 99th percentile is about 2.33 standard deviations. So, for example, if you have +2-SD genetic susceptibility, you would need to be at about the +1.22-SD = 89th percentile in terms of environmental risk factors to be at the 99th percentile for schizophrenia risk.
You would need to be at about +(2.33/0.89)SD = 99.6th percentile of genetic susceptibility to have a 50/50 chance of being at the 99th-percentile for schizophrenia risk. Assuming that the environmental risk is all non-shared, you would still only have 50% concordance for identical twins at the 99.6th percentile of genetic risk.
This is really cool, I love these simulation-based sanity checks for unintuitive things in probability theory.
Btw the company Orchid Health is apparently working on polygenic embryo screening for schizophrenia, but people are upset about it and the group whose data they're using, the Psychiatric Genomics Consortium, are trying to rescind permission for their data usage:
"PGC objects to such uses because its goal is to improve the lives of people with mental illness, not stop them from being born"
This is an interesting issue. One reason genetic screening makes a lot of people uncomfortable is it seems to imply that the people in their life who have a certain disorder should never have been born.
True. I guess the way to frame it is that the people who are alive are only a tiny fraction of the potential people who could have been born of the same parents instead, if a different sperm had met a different egg, and that "people" who are never conceived are not noticeably inconvenienced by that fact, so we can treat them in expectation as interchangeable - thus if someone with a certain disorder had never lived, and instead their putative sibling without that disorder had lived instead, we can treat that world as basically analogous with swapping out the person with the disorder for a version of the same person without the disorder. This is ... not wholly satisfactory, but I'm going to struggle to do better to square my intuitions here; I am very negative about people who would forbid genetic interventions that mean that future people don't need to suffer from serious life-limiting genetic conditions just because it would somehow say something negative about living people with those conditions. As long as we treat the actual people with those conditions with fairness and compassion, of course.
That phrasing raises the question: if a pill that could 100% prevent the development of psychosis was invented and given to at-risk babies at birth, would they consider that erasure of people with mental health problems since they would no longer be a distinct community.
Rather than merely making the lives of people who've developed psychosis better.
I'm still unsure about these issues. In the case of schizophrenia, the kid would have a 80% chance not to become psychotic and scaling such practice would reduce the spectrum of human personalities, maybe significantly. Who knows when a lot of schizotypal people will be needed to save humanity?
If there were single gene polymorphisms with large negative effect, they would get selected out of the population ... eventually. Which suggests that there can't be high-frequency mutations with large negative effect, unless there is some compensating advantage (like, e.g. giving you resistance to malaria).
Which leaves us with multiple mutations, each of which individually has a small effect, adding up to a large total effect. And mutation-selection balance, where random mutations are introducing harmful mutations at about the same rate the natural selection is removing them.
Or it's a recessive gene, which are hard to select against unless a lot of incest happens, which usually isn't the case because incest is already selected against.
The exception that proves the rule is that there are a bunch of highly penetrant, dominantly-inherited neurodegenerative diseases like Huntington Disease and various forms of familial ALS that that rarely do any real damage until after prime reproductive years, and thus have minimal effect on reproductive fitness.
I think it's interesting that your example kinda proves the point of the import of environmental influence on schizophrenia, and with that maybe shows the quite logical mistake people make when they focus on the fact that schizophrenia is 80% genetogenic. It seems that environmental factors would only make a difference in 20% of the cases when reasoned naively, but you can see that for the ones with a high enough genetic risk, environmental factors actually contribute 90% to the the development of schizophrenia! It's just that if you don't have the genetic risk it's nigh impossible to develop schizophrenia, even in a highly conducive environment, yet if you're one of the 10% of people that has a higher genetic risk the environmental factor becomes highly important.
This is true, but I think it's important that it's symmetrical - I put genetic risk first in my spreadsheet, but it's not like genetics merely predispose you and then environment deals the final blow. You could equally frame it as "only in people with environmental risk does the genetics become important".
(or at least this is true in my simulation, and what I expect to be true of reality - I think other people differ).
Isnt sickle cell trait a good example of why you can’t draw this conclusion from the numbers only? We know exactly which gene causes sickle cell disease, which is genetic in the intuitive sense. That gene also protects against malaria, so we must see a hereditary component to malaria if we purely look at the stats, but it’s not really what we think of as genetic.
Actually, I just realised, it's gnarlier than that: the "-genic" suffix is sometimes referring to the causal agent, as in iatrogenic harms, being harms caused by a doctor, and sometimes referring to the effect, as in carcinogenic substances being substances that cause cancer (as opposed to substances that are produced by cancers). So "genetogenic" could refer to that which causes things to be caused. Paging Sheherezade; we may have a deep regress on our our hands.
"While genetic effects are evident from the higher concordance in MZ twins than in DZ pairs, concordance in MZ twins does not approach 100%, even for highly heritable traits such as schizophrenia"
> I think you should expect very slightly fewer schizophrenics in the new generation, but the effect size wasn’t noticeable in this small granular simulation - nor, apparently, in Germany.
Since the population of Germany was so much larger than 2000, shouldn't statistical significance be much easier to pick up?
Yes, but you also don't have all the schizophrenics nicely highlighted in red - you have to hunt them down, diagnose them, and get that fact to the attention of whoever's doing a study. If you're comparing pre-Nazi to post-Nazi Germany, you also have to deal with a few decades' worth of evolving diagnostic criteria.
One might also expect that the Nazis generate a significantly worse environment, which would counteract the effect (if any) of the changes in the gene pool.
Also: it is in principle possible for something to be both generic and caused by a parasite (the genetic variation is improved resistance to parasitic infection).
(Looks at cat infected with T. Gondi suspiciously ...)
>A better model would have to take into account that people’s children aren’t clones of themselves, and that children’s environment is correlated with their parents’. But both of these would drive schizophrenia rates up, not down,
I believe the first would drive the rate up (because the variance in genetic schizophrenia factors would be higher), but the second would drive them down (think of the edge case where children's environment is perfectly correlated with their parents, and children are clones of their parents: in that case, schizophrenia/non-schizophrenia is perfectly passed on to children, so if you remove the parent schizophrenics, there won't be any schizophrenic children).
I know nothing about schizophrenia, but I was slightly surprised that the model you use is something like compute "genetic score" + "environment score", and if you are above threshold you develop it.
What I expected to see is something like "genetic score" + "environment score" give you your probability to develop schizophrenia (which can fairly low even for worst scores since it is not that widespread), then you roll the dice and either get it or not depending on your luck. In this model even if my twin I share environment with develops schizophrenia, it means that my baseline chance is likely significantly higher than average, but there is still a decent chance I won't develop it because it is fairly rare.
Your model kind of assumes that some people would be genetically "immune" to developing schizophrenia; do you think that's closer to how it works?
I think his model is a simplified one in which the "environment" score incorporates luck (such as with what is commonly labelled "non-shared environment").
This conversation brings to mind guitarist Peter Green's schizophrenia. We saw him perform in what looked like a bathrobe, with a giant crucifix swinging from his neck. With Danny Kirwan, Jeremy Spencer, etc. in a seedy carpet store on Market Street Bill Graham called Fillmore West.
Later in his music career, Green released ethereal solo recordings that were much calmer than his Fleetwood Mac work. But he got concerned about starving children shown on TV, and felt somewhat guilty about living a pop star's self-absorbed life. ("Couldn't we just make them a sandwich?")
Things got more complicated when he busted up his brother's collection of crystal ware, in an emotional outburst. The family got help for him, and he was diagnosed with schizophrenia. How? The initial story was he was basically kidnapped by a hipster cult of Germans and dosed with some particularly weird acid. That theory was eventually pretty much debunked. So, was Peter schizophrenic when he was madly grinning and playing his guitar? Was channeling a sort of madness into guitar solos therapy? I don't know about his family history, but Green, whose name by birth was Greenbaum, I believe, shortened to sound less semitic, did talk about insults he had suffered growing up Jewish.
But it likely wasn't notorious German LSD or antisemitism that caused his illness. It does cause me to wonder whether his music would have been so rich if he hadn't been a little bit 'crazy'.
The onset of schizophrenia in adulthood is more like massively amping up already existing weirdness rather than moving from the binary state of normal to mad.
Oh, we know the answer to this! It's called schizotypy, which is tri-dimensional (positive, negative, ???disorganized???). Positive and sometimes disorganized schizotypy read like they was written by someone making a satire of psychiatry; it amounts to the "Do people think you're strange? Do you?" of the famous Subgenius meme, but swaps it from "better than most people" to, in a traditional framework, "worse than most people". The quote from a 'prodrome' test I gave earlier, where it asks very seriously how much you daydream and if you care about philosophy and religion and politics, is...not wholly unrepresentative.
Negative schizotypy is the Asociality Grab-Bag. Everything in the DSM is correlated with negative schizotypy, because everything in the DSM is correlated with being socially reticent or anhedonic. It's trickier to talk about in coherent terms because of this. Disorganized schizotypy is even worse, because it serves as a dumping ground for questions about eccentricity and subthreshold thought disorder. All of them form the whole of "being schizotypal" in combination, and in most people who have a classically schizophreniform experience at some point in their lives (amongst other things), but the likelihood that someone in the general population who is very high on one is also very high on the other isn't too high.
The answer to "would this person be so [X interesting trait] if he wasn't crazy [i.e. schizospec]?", is no, because we know what the real long end of the schizospec is, and that psychosis is "massively amped-up weirdness" (to quote) rather than something that just happens to neurotypical people. The answer to, to subtweet a few other comments, "wouldn't screening-out of embryos with high polygenic risk scores for psychosis be a good thing?" is ha ha ha no, I cannot think of a single disease screening that would be worse, possibly including autism. (Autism is more trickily borderline, because its research problem is less about ignoring the ultramajority of the spectrum, and more about a hellish amount of diagnostic substitution.)
You write some exceptionally interesting posts. I wrote but didn't submit a response to one yesterday. Personally, I wish they were a bit more readable.
I get a little rant-y :) I'm working on a longer-form, more-edited-than-quick-comments piece on a a somewhat related subject (the complex grounds of what constitutes a *monogenic* genetic syndrome, in that many have much broader or milder phenotypes than you might think), though it's slow-going.
Schizotypy I tend to get maybe too rant-y about, because it's a frustrating subject for a few reasons:
1. It's *so obviously* the "core" or "real" thing to look at when looking at the schizo-spectrum. This is immediately obvious if you, say, compare the best prevalence estimates we have for schizotypal personality disorder with the best ones for schizophrenia. It's obvious if you study relatives of people with an SZ diagnosis. It's obvious if you study what people were like before they had a psychotic break. It's obvious if you look at...well, anything. But it's not obvious in any way that makes it into research. Autism provides a parallel here; it's a 'spectrum' (a tricky and complicated one itself), and it does face serious research problems for that fact, but people certainly research the 'high-functioning' autisms, and the fact they're in many ways more valuable to research than the 'low-functioning' ones (because the diagnosis is probably not wrong, which it often is for 'LF'A, and the clinical picture isn't so confounded) is recognized.
2. The criteria for schizotypy are...at least arguably something that shouldn't be called a 'disorder' or a 'bad thing'. "Something that distresses some people with them", sure. "Something that distresses some relatives of those people" -- oh, yes. But (again, like many of the autisms), the idea they're something that should be treated through an inherently problematic lens rapidly raises tons of issues. (There is a foreseeable future, somewhere down another path, where arguments about the schizospectrum take on some of the qualities we see today in arguments about the autism spectrum -- the kind that get simplified into "high-functioning self-advocates" and "relatives of low-functioning people", neither of which are quite right or form natural categories.)
The second is very, very difficult, because people (for many reasons, many good) want to justify their intuition of schizophrenia as a wholly bad thing, and something that e.g. it would be beneficial without consequences or horrible backfires to "screen out". Because many elements of schizotypy clearly contradict this, you end up in a bind.
The former is annoying for a different reason, which is that it kneecaps research. "Causes of" schizophrenia are not a coherent concept in the sense that most of them haven't been properly cross-referenced for "is this true of schizotypal people, or just some people who develop psychosis?". (I wonder if "polygenic risk scores for schizophrenia" and "polygenic risk scores for schizotypy" are non-identical.)
To throw fuel on the bonfire, doesn't this also steer *away* from all those embryonic genetic screening business? After all, if you can have 80% schizophrenia genes but not develop it because of environmental factors, how many of the other "eek! genetic risk!" alarms are likely to occur if you let the embryo develop into a pregnancy, be born, and grow up?
I haven't seen the kind of diagnostics these specific companies do, but if they're anything like pre-FDA-intervention 23andMe you'll get the "predicted risk of $DISEASE: X% instead of population average Y%", rather than "alarms".
It just makes sense to give the raw stats instead of substituting the judgment of the company to that of the parents, especially since out of a bunch of IVF embryos no single one will be better on all measures than the others, and any choice involves a trade-off between different risks.
I agree that high PGS score/high genetic risk for condition X doesn't mean you are going to develop condition X with any particularly high degree of probability compared to someone who wasn't selected. However these sort of screening businesses don't normally screen for just one thing. they screen for dozens up to hundreds of different disorders with a genetic component. And because how averages work if you choose a "low risk" embryo with a score saying it will have on average 2 out of 100 conditions (lets assume them to all be equally bad) then this embryo will with very high probability (close to 100%) grow up to have fewer bad conditions than an unselected embryo whose score suggests it will have on average 8 out of 100 conditions.
Even though for each individual condition X out of our 100 both the selected and unselected embryos have very close probabilities of developing the condition, when you aggregate over all 100 these differences in expectation add up and when you combine this with the averaging effect of combining multiple small individual contributors then with very high probability the realisation of the screened embryo random variable will be healthier than the realisation of the unscreened embryo random variable.
As other comments elsewhere have summarized it - imagine you have several bridges. ~90% of them are structurally sound, while ~10% are not structurally sound, and are vulnerable to collapsing in high winds. But high winds don't happen very often, so only about ~10% of the vulnerable bridges (~1% overall) collapse. So, you have one group that is immune to high winds, and another group that has to depend on luck - but most of them get lucky, so only a small number actually collapse.
Regarding embryo selection, the question becomes: would you rather have a child who is immune to XYZ disorder, or would you rather have a child who needs to be lucky?
I would like to point out that this isn't a one-off disagreement about the genetics of schizophrenia (or any other disorder). As a devout Catholic, IVF/embryo selection deeply offends you, with years of commenting history attesting to that fact.
Every time this topic comes up, you grasp at any shred of evidence that embryo selection will somehow not be effective after all. You are absolutely giddy at the prospect of being able to say "see? I told you so!" to any parents who are trying to have healthy children through "immoral" means such as embryo selection.
"You are absolutely giddy at the prospect of being able to say "see? I told you so!" to any parents who are trying to have healthy children through "immoral" means such as embryo selection."
Jeepers, and I'd have gotten away with it too, if it hadn't been for you meddling kids!
Yes, that is absolutely what I was trying to do.
Or maybe, juuuuuuust maybe, I was asking an honest question? If there's a screening company that comes back with "1% chance of Obscure Syndrome You Have To Look Up Because You Never Heard Of It", what is the real risk? What is the weighting of genetic versus environmental? If there's "You carry genes for breast cancer" is it worth getting a preventative double mastectomy? Some people think the risk warrants such an intervention, others don't.
If twin studies show that even with the same genetics and brought up in a similar environment, one may be schizophrenic and the other won't be, then I think we know less than we think we know about how illness works and the genuine risks.
But sure, put it down to me wanting to pwn the libs or whatever the cant phrase of the day is.
This sounds more like a science communication (or science understanding) issue.
I asked a convenience sample at breakfast what "80% genetic" meant to them. Generally the reply was "80% of the cases have it entirely due to genetic reasons with environment playing no part, and 20% with environmental reasons and genetics playing no part". I asked what about ones where environment played some part and genetics played some part. They updated their answer to be 80% genes with no environment, an unknown X% environment with no genes, and 20%-X% mix.
I then laid out the argument and spreadsheets in the article, and asked how they would choose to describe the reality of the spreadsheets, in their own words. The consensus reply was "It's 100% genetic and 100% environmental, but X% of the population is immune regardless of their environment, and Y% of the population is immune regardless of their genetics, and those populations are non-exclusive, and you could construct a Venn diagram to easily and accurately communicate it to the general public."
Then one person pointed out something interesting: 20% environmental is somewhat circumstantial. Since there's no absolute way to define the environment, it's a number that's derived from the types of environment available for the study population, and theoretically a more powerfully negative (or positive) environment could exist or develop in the future that would change the ratio from 80/20. And therefore the 80/20 mix isn't something set in stone but instead set in the social milieu of the study and there should be an expectation that it changes over time or study population.
Thanks for taking the extra step to ask some questions! Now that I think about it, that does sound like a common interpretation. I’m less certain that the general public would easily understand if given a Venn diagram. The “X% and Y% immune” seems even worse for general understanding. But clearly, just saying “80% genetic” doesn’t communicate reality. What would a single-sentence, or ideally sub-ten-word explanation look like?
That's a good question, and one that I think will have a different answer for each individual gene and trait discussed. Take for example horse size, something I read about very thoroughly a while back for a simulation project. Suppose there are two known alleles for the LCORL gene that have these effects on height: small/small -> x1.0, small/large -> x1.03, large/large: x1.06 and two alleles of HMGA2 with these effects: small/small -> x0.81, small/large -> x0.94, large/large -> x1.0 (these numbers are chosen such that whichever allele I believe to be the wildtype scales by 1.0). And suppose there are maybe 2-4 other unknown genes with similarly sized effects, plus a bunch of less noticeable ones. Suppose also that stallions are on average 2% taller than mares. (Edit: castration also has an effect, though I forget how much and in which direction) And we know nutrition also has a large impact on size, picking numbers at a wild guess maybe a malnourished horse would be 10-20% shorter. And then some amount of size variation from randomness, let's say plus or minus 5% height.
At the end of all that, how do you go about describing how much of horse height is genetic and how much is environmental? How do you communicate that % genetic and % environmental/non-heritable is going to vary depending on whether you look at all horses, only purebred Thoroughbreds, only well-fed horses, or only clones of one particular horse?
Any way you look at it, 100% of horses have genetic factors affecting their height, and 100% of horses have environmental factors affecting their height. So I guess the only hope of being able to communicate anything about this in a short and clear sentence relies on the listener understanding that everything comes from multiple causes.
There isn't a sub-ten-word explanation for most lay people with an average level of biological literacy. To address what I think are most people's concerns in asking a question like "is schizophrenia genetic?" I would say something like "We cannot predict who will get schizophrenia. Most of the known risk is genetic, which means that people with close family members with schizophrenia are at higher risk. But most people with a family history of schizophrenia will not develop it." And follow up with quantification if desired.
> The “X% and Y% immune” seems even worse for general understanding.
I disagree. I think the "immune" framing is a really great way to describe it, but more as "X% immune" instead of "X% and Y% immune."
There is another comment chain about bridges. Imagine you have a group of bridges where ~90% are structurally sound, and ~10% are vulnerable to collapse in high winds. But high winds don't happen very often, so only ~10% of the vulnerable bridges (~1% overall) end up collapsing.
You can frame this as "~90% of the bridges are immune to high winds, and ~10% have to rely on luck - but most of them get lucky, so only a small number collapse."
I've also heard it described as "genetics loads the gun, but environment pulls the trigger."
> Then one person pointed out something interesting: 20% environmental is somewhat circumstantial. Since there's no absolute way to define the environment, it's a number that's derived from the types of environment available for the study population, and theoretically a more powerfully negative (or positive) environment could exist or develop in the future that would change the ratio from 80/20. And therefore the 80/20 mix isn't something set in stone but instead set in the social milieu of the study and there should be an expectation that it changes over time or study population.
Yes, I always considered this to be a problem with interpreting these analysis-of-variance numbers.
An example: imagine a society where some people have huge amounts of lead exposure, others have very little. Then studies will show IQ is mainly environmental, with genetics not being important.
Next, imagine a society where every child is fully nourished and has an incredibly enriching education. Then, because everyone is achieving near their genetic maximum potential, studies will show IQ is mainly genetic, with environment not being important.
Despite the opposite conclusions, in these two examples, the biological realities are identical.
What's the alternative to analyzing variance? Ideally there could be a model developed, saying something like IQ = (genetic potential) - (lead poisoning) - (lack of education) - (poor nutrition) - (luck factor). This would provide far more useful and actionable information than just factoring the variance into genetic/environment.
Does the uniform distribution make sense for the genetic component (or the environmental component for that matter)? Seems like a power or normal distribution would make more sense, especially given the low appearance rate in the population. Using those distributions changes the simple model considerably. Another consideration is the deviation in environment vs genetics - that has a big impact on how the model shapes out - if the outliers are on the environmental or genetics side drives the analysis.
Even some of the weirdness around high heritability but low variance explained by the SNPs we’ve found may be explained by rare variants (low base-rates) that aren’t found in ordinary GWAS setups. And previous missteps in psychiatric genetics can be explained by not considering the base-rate of how many (underpowered) studies were looking for candidate genes.
Nice. I ran the same simulation and got similar results. Then I ran the simulation out to 5 generations of Nazi genetic experiments. For the budding Nazis out there, after 5 generations, I did start to see an effect! G2=19, G3=15, G4=12, G5=9.
So after only five generations of murderous crimes against humanity (~100 years), Eugenics reduced schizophrenia by half! Of course, there's the pesky side effect that many of these genes probably have some positive effect when they don't manifest as disease...
But maybe a return to the atrocities committed in service to the Eugenics movement isn't necessary - even in a world where a disease is 80% genetically driven. I took Scott's same model again (I generated my own threshold based on my top 20 cutoff), and tried two different things. First, I reduced environmental contributions by 20%, and second I created an entirely new population with a baseline 20% reduction in environmental contributions. Both returned 2 schizophrenic people in the overall population of 2000 - a 90% reduction!
My model suggested a 50% schizophrenia decrease with just a 10% reduction in environmental factors, and a 100% decrease with a 30% reduction in environmental factors. In other words, you don't even have to be perfect (or particularly good!) at mitigating environmental factors to have a huge impact. So, even if genetic factors are responsible for the majority of the effect, if environmental factors are easier to control they might deserve significant focus for even heavily genetic-driven diseases.
CAVEAT: Of course this is based on Scott's oversimplified model.
I agree that environmental factors aren't easy - especially the part where we have to identify causal relationships - I'm just saying that they're still an option, even with conditions that are heavily genetically determined. For example, say that 25% of the "environmental" variance isn't environmental at all, but copy number variants (i.e. still genetics). That still leaves us with the other 75% to play with, and if we only make a 5-10% reduction in environmental contributions - because it's hard to do - we should still see more than a 5-10% reduction in overall incidence, which is encouraging!
I try to steer clear of solutions that require us to approach the asymptote of 100% implementation before we start seeing an effect. I prefer solutions that start showing efficacy below 20% implementation.
You don't have to kill anyone to do coercive eugenics; you can just sterilize them :-P
Though I am now intrigued by what would happen in the hypothetical world where a version of Project Provention (https://en.wikipedia.org/wiki/Project_Prevention) had been set up to pay people to get sterilized not on the basis of them being drug addicts, as in our timeline, but on the basis of them being schizophrenic. There is plausibly no population of candidates for that sort of intervention that would be more resistant to it.
I think the implication that Torrey is lying is uncharitable. Maybe he just doesn't know these decidedly unintuitive mathematical results, and Psychiatry Research will issue a correction (or even if this too central to the paper, a retraction).
The math is wrong. For the causal paths, you need the square root of the variance, so you need sqrt(.80) = 0.89, and sqrt(.20) = 0.45. With heritability 80% and environment 20%, you get relative strength of 2 to 1, not 4 to 1. Variances are deceptive.
I don't think this will change your simulation results that much though.
Given the nonlinearity of the normal distribution, does it make sense to define relative importance in terms of ratios of correlations? Assume that everything is normally distributed and additive, and that you get statsophrenia when you're at 99th percentile (2.33 SD) additive risk. With mean environmental risk factors, you need to be at +2.6-SD genetic risk, or +5.2-SD environmental risk with mean genetic risk. That's 99.5th percentile vs. 99.99999th percentile.
I'm not sure how to reason about this. Does it make sense to say that genetics is only twice as important when we can make statements like "There are 50,000 times as many people who get statsophrenia due to genetics despite average environmental risk factors as there are who get it due to environment despite average genetic risk factors?"
There's several decades of debate about how to measure relative importance. This debate is mostly about regression analysis, but these ACE-model results are equivalent to regression with uncorrelated predictors (no GE-correlation).
In the I/O psychology literature, though, they are generally concerned with ratio of correlations because strength of the correlation has a linear relationship with the expected return of using it for selection (hiring) purposes.
>"because strength of the correlation has a linear relationship with the expected return of using it for selection (hiring) purposes."<
Can you expand upon this a bit, if you get a chance? It seems to me that ought not be the case — but perhaps I misunderstand "expected return" in this context (in terms of winnowing candidates? in terms of quality of work?), or else something else in my mental model is dumb and wrong.
(e.g.: taking Brandon's example, but pretending "statsophrenia" is actually very desirable, is knowing a candidate's genetic risk factor only twice as valuable as knowing their environmental risk...?)
See section "Computation of Utility". The equation for computing the gains of utility are multiplied by the correlation of job performance with the predictor (IQ or something else). Thus, they scale linearly. You could also show this with a simulation of hiring with predictors of different validities and how this affects the difference in job performance between the hired and unhired groups.
Would you mind pointing to some material that explains why? Or if you feel like it, mapping out your first sentence a bit more? I have zero instinct for this type of stats, but love to get better at it...
Good post, but quick note: 80% heritability means genes are *twice* as important as environments, not four times as important. When predicting, you want to use unstandardized terms and sqrt(0.8) = 0.8944, sqrt(0.2) = 0.4472.
I just thought of a thought experiment for Aftab's point: can someone who knows genetics better than me comment?
Consider drowning. Right now I would predict that drowning is largely environmental (obviously genetic predispositions towards risk-taking that could include risky behaviors around water will affect it).
However, what if there were a mutation that gave people gills? If this mutation became widespread, it would significantly increase the genetic component to drowning risk across the population. To what degree is it correct to say that, in this hypothetical world, an individual drowning is largely caused by their lack of the "gill" gene(s)? How does that differ to the actual world we live in when someone drowns?
This is a limitation of the whole concept of dividing variance into "genetic" and "environmental" factors. It's a *description* of the current statistical distribution; it is *not a model* of the underlying mechanisms. In particular, at different times and within different populations, you'll have a potentially different genetic vs environmental split.
In your example, gills currently account for 0 variance of drowning, because nobody has gills. If in the future, some people develop gills, then the genetics for gills will explain some of the variance in drownings. And so there will be a corresponding increase in the genetic effect on drownings.
In general, heredity estimates are only applicable to the populations from which they are sampled, and to some extent to other populations which do not differ much in relevant ways. Heredity samples in different populations, at different times, or under different technological conditions can be very different. They should not be taken as timeless, universal truths.
For example, in a wealthy country where almost all people, even poor people, are well-nourished enough to attain their genetic potential height, we will find that height is very strongly heritable. However, if we look at a global sample, where a large percentage of the population does not get the nutrition necessary for this, we will find that height is significantly less heritable.
In the United States, Mandarin verbal ability is largely environmental, because the vast majority of speakers learned it by being raised either in a Mandarin-speaking country or in a Mandarin-speaking household in the US. But in China, it's largely genetic, especially if we confine our study to regions where most people speak Mandarin at home.
Another way to reframe the situation intuitively is that the odds a randomly selected person has schizophrenia are 1%. If having a twin with schizophrenia increases your odds to between 15-50%, it seems like having a schizophrenia gene increases your odds 15-50x of developing schizophrenia vs. the base rate. That's a very strong genetic connection!
This is a great explanation of the subject, very clear and concise.
> E. Fuller Torrey recently published a journal article trying to cast doubt on the commonly-accepted claim that schizophrenia is mostly genetic.
> People really hate the finding that most diseases are substantially (often primarily) genetic. There’s a whole toolbox that people in denial about this use to sow doubt..
I feel like these two lines are very bravery-debate-ish. "The evil anti-science people are using clever tactics to spread misinformation. But we're the brave defenders of science who can see through their ways!"
Also schizophrenia is a disease which is constantly getting filtered out by selective pressure and replenished with mutational load. Killing off people who have it will reduce the amount of schizophrenia genes in the next generation a lot less than the numbers of people who have it because those people would likely not have had many children to begin with, and there's always more mutations in the next generation upping the overall amount.
Truly the worst post I've ever seen from you, Scott. Uncreative self-soothing that misunderstands the existing literature on the genetics of schizophrenia—and, as others have pointed out, bad math!
Except you have just proven the point of the paper: even amongst those with highest genetic risk, a tiny minority develops schizophrenia. Hence the "80% genetic" angle is not just factually-true-but-kinda-unhelpful, it is actively misleading and used by people to discourage and/or disregard research into environmental (i.e. social, political, toxicological) causative factors.
This is emblematic of a wider problem in the field: heritability has no intuitive correlate when it comes to the individual patient on a biological patient. The missing heritability problem taught us it does not even have any correlate on the level of DNA, at least at current levels of data availability, and I would risk predicting that this will not meaningfully change with the whole genome sequencing effort people are getting hyped up about today (i.e. the 1 million genomes the EU is trying to scrape together).
Likely, we will have another round of articles lamenting the fact that the expenses have not been worth it. This is an obvious grift by the genetics community to try and stay relevant, when the other main promise they made couldn't be fulfilled: genetics based drug discovery has been a disaster and so the industry funding is drying up, I guess.
I do not understand why people have such enthusiasm for genetic causes of chronic disease - it means we are factually powerless to prevent it. Is this just residual belief in eugenics? If you are actively seeking out a Gattaca-style future of polgenic scores and genetically engineered humans, please be aware of the enormity of the dice you are rolling: more likely than not, you will be amongst the discarded. It takes a lot of delusion to believe to come out on top, kinda like a modern, more depressing gods-chosen-people belief.
I can't speak for Scott, but a lot of this seems uncharitable.
"I do not understand why people have such enthusiasm for genetic causes of chronic disease - it means we are factually powerless to prevent it"
Firstly, if we actually *are* powerless to prevent it, that's still information worth having; it means we don't need to throw resources away seeking a cure that cannot exist. But secondly, it is unlikely to actually mean that we powerless to prevent it in any case. Scott himself made the point well in this old post: https://slatestarcodex.com/2014/09/10/society-is-fixed-biology-is-mutable/ - just because a problem is biological / genetic, doesn't mean we can't do anything about it. It just means that the best treatment is likely to be a medical rather than a social intervention. And you can try to triangulate his true position from this Socratic dialogue, in case you haven't already read it: https://www.astralcodexten.com/p/galton-ehrlich-buck
"Is this just residual belief in eugenics?"
I think you may need to be clearer what you mean by "belief in eugenics". Belief as in "empirically, a eugenics program could be expected to be effective at its stated aim" is different from "morally, implementing a eugenics program would be the right thing to do" ... and that's before we even get started on the vast range of interventions, from coercive to wholly voluntary, implemented by governments or left up to individual prospective parents, that get rolled together under the label of eugenics.
I get the impression, from what I've read of Scott over the years, that he is about as opposed to Nazi-style mass-murder-based eugenics and mid twentieth century style coercive-sterilization-based eugenics as anyone else, but is optimistic about the prospects of non-coercive reprogenetics which leaves the decision whether to opt for, e.g. embryo selection with polygenic screening, in the hands of individual prospective parents. He has also in the past expressed some favorable comments about Project Prevention, a charity which offers to pay drug addicts a financial premium to get sterilized, with a view to minimizing the number of children who have to grow up with drug-addicted parents, though I'm not sure if he still stands by that today. I'm not sure to what degree Project Prevention even counts as eugenics - sure, to the degree that drug addition is genetic, and to the degree that the charity is effective at its stated aims, it will have eugenic effects, but from what I've read, they're not so much concerned about trying to reduce the frequency of genes for addictive behavior as they are concerned about trying to minimize the suffering of children of drug-addicted parents by preventing those children from coming into existence.
"more likely than not, you will be amongst the discarded"
Again, this presupposes that Scott favors coercive measures, or that if we allow voluntary measures, this somehow sets us on an inevitable course towards coercive measures. This is a non-zero risk (though in the modern era I am highly skeptical that there is much of a constituency for reinstituting the old-fashioned coercive measures), but it has to be weighed against the suffering caused by genetic conditions that we could have used voluntary biotech to prevent, but didn't because we were scared of the slippery slope. We don't get to weigh only one side of that trade-off.
Again, even the large heritability estimates do not mean that this is a "genetic disease", but rather that a subgroup of the entire population is disproportionately susceptible to the eventual, and more proximally causal, environmental factors. This is to me the main point the paper made, and it still stands, regardless of anything Scott wrote unfortunately (looking forward to the follow-up, because so far this was disheartening).
Re this is important information: we have had industry and public research speding for medical interventions against schizophrenia out the backside and nothing to show for it, except for a few drugs that are nor disease-modifying in any way, poorly tolerated and rarely lead to long term success (PMID: 35653111). So exactly how is this helping? And while in principle I would be sympathetic to the not-wasting-money angle, this is not a monogenetic disease like cystic fibrosis where you can shut down alternative etiology research once an for all. Rather, there is now a substantial bias towards environmental defeatism based on a misunderstanding of the consequences of "high heritability" and hence this line of research is likely not adequately funded.
Re eugenics/compulsion:
I do not suggest Scott is favouring compulsory eugenics. If I recall correctly, in Gattaca, there was no governmental coercion either, rather genetic discrimination was made illegal and it was mostly companies enforcing the dystopia via genetic esting of employees. People do not exist in a vacuum, and what becomes de rigeur in a society becomes to some extend compulsory by the uneven playing field it creates for those who refuse to engage. For instance, I'd rather not wear expensive clothing, but society is forcing me to buy suits and dress shirts by not allowing me to do my job in the 10-year old sweatpants and t-shirts I normally wear. This is a sort of coercion by circumstance, and the get-paid-to-sterilize "charity" is dangerously skirting the line in this direction. Economic pressures exists and can result in compulsion worse than what the most opressive governements would dream of doing (case in point: prostitution). The minute genetic optimisation becomes available to the general public, it will seriously tilt the playing field for the next generation, resulting in a new upper-class of "selected" kids and a new underclass of "their-parents-weren't-able-to-afford-it-s". How is this desireable, again?
How is that any different from any other advantage not everyone in the world can afford? Should we ban good food, clean water, medicine, education until everyone can get it?
We're going to have an upper-class either way, I wouldn't mind a saner, smarter one.
Obesity is also 80% genetic, yet we know from epidemiology that obesity increased substantially for environmental reasons.
This suggests that there are strong gene-environment interactions. But the "X% genetic" model tacitly assumes ZERO gene-environment interactions. Maybe it's not a great model?
(A similar story to obesity happened with the Flynn effect: IQ is like 80% genetic, but IQ scores increased substantially over the second half of the 20th century for what are surely environmental reasons.)
Good discussion. It illustrates how people get very confused when they don't work through the numbers but simply make judgements based on the qualitative sound of words like "most." They forget that there are multiple variables at play, and "most" may apply only to some of them.
Worth noting that exactly analogous considerations apply to IQ and some personality traits.
I'm surprised this statistical fallacy required a post to debunk. You can swap out schizophrenia for any other rare condition; imagine someone saying "IQ is heritable? Then why is it that when one twin is >145IQ, the other twin is >145IQ "only" 33% of the time?"
Smart analysis. However, when people say heritability accounts for 80% of the variation in a binary trait, they mean actual cases, not 80% of the variance of a latent variable which causes that trait above a theshold, right? How do they figure this? Through some kind of log odds ratio on the variable corresponding to identical genome?
It's important to remember that heritability studies are only as powerful as the range of environments available to researchers--which are always very narrow, relative to the range of conceivable environments. The fact that WWII-era German culture was wildly different from post-war Germany is more than enough (for me) to explain the lack of reduced rates of schizophrenia, even with high inter-culture heritability scores. Which people become schizophrenic might change drastically in a deeply religious society vs a far-future technological society vs an isolationist commune--even if _within_ each of those societies schizophrenia is 80% heritable.
My best summary of my intuition here is that there's no such thing as genetic variance that's pure and independent of environmental variance (and vice versa). Heritability measures are _always_ conditioned on a given environmental range.
"Is it wrong to give twins random environmental scores? Don’t twin pairs grow up in very similar environments? Yes"
Is it wrong? The book Innate by Kevin Mitchell changed the way I think about individual development. It looks like most of your ‘environment’ is how your brain happens to wire itself up during fetal development. That’s why identical twins can be so different, not to mention siblings in general.
You can model these questions using a liability threshold model.* For example, see the methods section of this article: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8510582/. This model is very similar to the one you've used, but it allows you to define covariation between family members. For example, you can take into account that a child and parent share ~50% of their genome. You can also take into account shared environments.
Using this model to address the first question:
- The twin of a diagnosed schizophrenic has an ~36.5% risk of developing the disease.
- The twins can share up to ~43% of their environmental factors before that risk surpasses 50%.
To answer the second question:
- A child of two parents without schizophrenia has ~0.85% chance of developing the disease.
- This number drops to ~0.76% if you assume that the child shares 50% of their environmental factors with each of their parents.
> Making everyone mate is beyond the scope of this discount-rate simulation
If you want to do it in the future:
1. Create a new column filled with random numbers.
2. Sort by that column.
3. Take the top half of the population and put them into a new column, probably on a separate sheet.
4. Take the bottom half and put them in their own column next to the top half on the new sheet.
You now have n/2 random marriages. You can take the average value of each couple, regress it to the mean, and have a set of n/2 predictions of child genetic risk.
Actual children would vary around that predicted mean. You can add a normally distributed term† to make that happen. (For example, if you want to generate multiple children from one couple!) For random mating, I believe the variance of this term should be equal to population variance in the trait (?). But there's a good chance I'm confusing the questions of "how much regression to the mean do we do based on our knowledge of the child's ancestry?" -- ultimately a question about the mean of the children's distribution -- and "how much residual variation is left among the children of two known parents?", a question about its variance.
† I get the sense from your post that your generated numbers are centiles rather than deviations. This will interfere with inheritance models.
I didn't spot this comment until posting my own comment separately, but I ran such a simulation (just using 1000 genes rather than calculating what all the necessary covariances are). Another way to do this would be to give each parent 2 normally distributed haplotypes and give the child one from each parent, and set each person's overall genetic value to the sum of their haplotypes (recombination can be ignored since it's only one generation).
This whole post seems like a strong argument that Awais Aftab is correct that "mostly genetic", or even "80% genetic", is a misleading phrase. Even you got slightly confused about exactly what it meant when running your simulation!
I frequently see people on the right say "since we know from twin studies that X is highly heretitable, that means environmental interventions are necessarily useless". Which is the exact same mistake you're rebutting here; as you just demonstrated, it's entirely possible for something to be "80% genetic" in this sense and also 100% dependent on environmental factors. (Though not necessarily ones we can change, they could be totally random.)
In colloquial terms, schizophrenia in your toy model is almost 100% dependent on genes *and* 100% dependent on environment; it's nearly impossible to get it without both. Framing it as one or the other, and mostly one, *is* misleading. (Something you've written on before: https://slatestarcodex.com/2013/06/25/nature-is-not-a-slate-its-a-series-of-levers/)
That's not to say we shouldn't do it, I'm not sure there's a better way to describe the situation in precise statistical terms. Just, maybe we need big flashing disclaimers every time reminding people of the above? Or maybe we need a new jargon word than "hereditability" that doesn't give people this impression?
> as you just demonstrated, it's entirely possible for something to be "80% genetic" in this sense and also 100% dependent on environmental factors.
No, that isn't possible and therefore can't have been demonstrated.
The problem is that you're shifting between claims about particular fixed people (and different ones in different claims!) and claims about the pool of all people.
For the very large majority of people in Scott's simulation, it is theoretically impossible for them to develop schizophrenia no matter how terrible their environment is. (Because you need a composite risk score of 0.929, and the environment can contribute at most 0.200 of that, more than 90% of people -- 0.729 / 0.8 = 0.91125 -- have zero risk of becoming schizophrenic.)
I hope you'll agree that, for that 91% of people, whether they develop schizophrenia is not 100% dependent on environmental factors?
Looking at it from the other direction, in Scott's model 1 in 50 people is, by definition, born with a genetic risk of 0.98 or worse. About one in four of these people will develop schizophrenia. Did they do that because of their environment? In a sense; the model doesn't allow for schizophrenia to occur without an environmental contribution. But whenever you look into one of these people, it is overwhelmingly likely that the environment that pushed them over the edge will have been a perfectly normal somewhat crappy environment, the kind that most people with alarmingly high genetic schizophrenia risk shrug off. What conclusion would you draw?
I do think a real criticism of the high end risk estimates for psychiatric disorders is that heritability is not just what proportion of the disease is caused by genes, but has an essential clause of how much it causes disease in a given environment.
Twin studies typically are not twins who were relocated to dire poverty or traumatic circumstances: most ended up in fairly middle to upper class circumstances as twin adoptees. When environment and nutrients are right and more standardized, then it makes sense that genes will play a relatively larger role.
The concordance rate (in Figure 2) from the 1970 paper seems higher than 33% and closer maybe to 38% or just somewhat under 40% and I don’t think it accounts for shared environmental effects.
Is concordance the probability that both twins have a phenotype given that either one of them does or is it the probability that both twins have a phenotype given that a specific twin does? I think it is the latter (or at least the 1970 paper you linked uses it that way), but the former makes a lot more sense.
Also, if we assume that schizophrenia genetic and environmental risk are normal and additive, we can derive the following formula for the concordance of twin pairs (using the paper’s definition).
Where $p$ is the population prevalence and $\varphi$, $\Phi$, and $\Phi^{-1}$ denote the standard normal PDF, CDF, and inverse CDF respectively. It is not very pretty.
Using a heritability of 0.8 and a population prevalence of 1%, we get a concordance rate of about 38% which is what the paper says (and I checked this in an R simulation and it is true or I have made at least two mistakes).
Also, for the probability that both have it given that either have it, or what I think makes more sense for concordance, we can derive the formula
Which gives a probability of about 23% for the heritability and prevalence specified in the post.
You can look at either formula at e.g. the website quicklatex or in whatever latex processor you have.
Also, the second argument is bad because the fact that Schizophrenia has low incidence means it’s immediately clear from e.g. the breeder’s equation, that culling Schizophrenics before reproduction would not change the frequency of Schizophrenia very quickly regardless of the heritability and I do not understand how the author did not see this.
Children born of older fathers are more likely to be schizophrenic (presumably because the fathers' sperm has accumulated more mutations as time has passed and something like 50% of genes are expressed in the brain). Do you know how much more likely? If the answer is a lot, i.e., if a large proportion of schizophrenics were sired by older fathers, then there is an additional reason why the Nazis' policy had little effect on the next generation, or at least one can tell a plausible story about that. The proportion of such offspring was little changed before and after WW2.
As less than two thirds of the german male cohorts born 1920-25 survived WW2, postwar children in germany indeed were sired by elder fathers on average.
The fact that people have more than one parent does make the case significantly stronger against the Germany thing. I ran a simulation in which a population of 5000 with 1000 genes each were tested for some 100% genetic trait, and the top 1% for the trait were eliminated, then had them produce children. 28 of the children (so about 0.6%) were still above the original threshold. Repeating the experiment a few times gave similar results (32 and 30 schizophrenic children the next two times) (I didn't bother working out exactly what the threshold should be on average for the top percentile, and just used the same threshold that caught 1% of the first sample every time, but that doesn't seem to have been a huge issue). If I eliminate the top 1%, run 2 generation and test the grandchildren, the rate returns even closer to the original rate, around 0.9%. If this effect is combined with the fact that real schizophrenia is only 80% heritable not 100%, it seems quite likely that the effect would be unmeasurably small.
Maybe "genetic" should mean two concurrently related things: first, that it is heritable. Second, that it is probabilistic. The first meaning that its presence in the gene pool is a certitude. The second meaning that its expression is conditional. Conditional on what? On a number of intricate factors and processes which are still genetically mediated. We know nothing about these intricate processes and factors hence our confusion. And it makes sense that a phenomenon produced by intricate processes and factors we don't understand would confound our understanding of its phenotypical expression.
I could be wrong here, but schizophrenia starts in the late teens/early 20s for men (20s for women usually), and there is a prodrome phase of depression and isolation, etc. before that, too.
So ofc the incidence of schizophrenia isn't going to change much by elimination of schizophrenics in a population - they aren't the ones doing much reproducing of those genes in the first place.
The gene naming objection does not survive even cursory thinking. You can tell that Huntington's disease is inherited from the biological parents without having any concept of what genes or DNA are. All you need is correlations. And these days, nobody disputes that height is largely genetic, despite us having very little understanding of the genes involves or their role.
Scott, what happens to this data if you expand "schizophrenia" into "thought disorders and bipolar processes?"
It seems to me that inter-rater reliability is part of the problem. In my forensic work, I have seen many of the same patients over and over, and who have also been seen by other forensic psychologist (its a rural state so the 5 of us who actually do this see the same repeat offenders a lot). Often times our Dx line up perfectly, but frequently one of us will Dx the patient with one thing, and the other six months later with something else. And both Dx are "close enough" to capture the disordered thought process that results in their being diverted away from the justice system and into the state hospital system. In fact, we work together so well, that we often consult with each other to reconcile the Dx variance before it goes to court.
What is the probability of say, for example an adult Dx with Major Depressive Disorder, recurrent, severe with psychotic features having a child who is later Dx with Schizophrenia?
I don't know whether I am reiterating a point made already on this blog or in the comments found here, but gene variants that increase your risk of being diagnosed with schizophrenia often also increase your risk of being unusually creative. Increased creativity means that relatives of people diagnosed with schizophrenia are more likely to reproduce than the norm, thus balancing out the greatly reduced reproductive rate seen in people diagnosed with schizophrenia. This likely largely explains why genes that increase the risk of being diagnosed with schizophrenia are nevertheless found in all human populations. A world without the genes that can together lead to the development of schizophrenia or increased creativity would be a boring world. And yes, I say this as a person with schizoaffective disorder and a professional background as both an award-winning advertising creative and a bioinformatician. Hope this comment is helpful.
Don’t forget about the role of mosaic variants in schizophrenia
What is the role of mosaic variants in schizophrenia?
De novo and mosaic copy number variants contribute to ~6% of cases and these are often different unique to single twin even if MZ: https://pubmed.ncbi.nlm.nih.gov/37601975/
While technically genetic, to the extent that these are not shared between MZ twins, they would show up as non-shared environment in twin studies, right?
Correct. Another example where “environment” is genetic
There are also instances where "genetics" is environment, if we're talking causal mechanisms. Propensity for cannabis use comes to mind for schizophrenia.
But that would show up with polygenic load. Unless you mean polygenic cannabis load in family/friends
Do we know which way the causality runs here? Is it possible that, e.g., people with prodromal schizophrenia are more inclined to smoke marijuana?
This study doesn't say what you're representing it as saying, which isn't a thing that makes sense. CNVs are much more common in people diagnosed with SZ than the general population, though this is a lot less true for SZ than it is for most neurodevelopmental things with defined disorder names (it's most markedly true for intellectual disability; its truth status for autism is intermediate between SZ and ID, and I think would be similar to SZ, or lower, if you properly screened out all the diagnostically substituted intellectual disability). CNVs would be shared between MZ twins unless something very, very strange happened, though. This study is talking about a particular form of strangeness that could happen in that circumstance, but it's ascribing it as much rarer than normal CNVs, and speculating it would probably result in an attenuated phenotype. Low-level mosaicism gets weird quickly, and it's difficult to tease out the degree to which it 'matters' vs is a chance finding, and if things can be chance findings even if they look like they should 'matter' (this is a big problem re. 'pathogenic' CNVs -- many of them, especially duplications, don't actually seem to do much in most people with them, such that the line between a 'pathogenic' and 'non-pathogenic' CNV is not naturally defined).
Perhaps you already have this in mind, but I think a lot of the non genetic component of why someone does or does not develop a disorder like schizophrenia is stochastic (chance) rather than whether they were exposed to certain environmental risk factors.
Kevin Mitchell describes this well:
‘Whether that risk manifests as actual disease, and which symptoms emerge, is probabilistic, reflecting additional non-genetic factors. Given the lack of evidence for systematic environmental risk factors, this diversity of outcomes may reflect intrinsic stochastic variation in the trajectories of brain development. Chance thus plays a substantial role in determining which outcome from a wide possible range is actually realised by the processes of development in an individual.’
http://www.wiringthebrain.com/2022/03/what-have-we-learned-from-psychiatric.html
There was a post about this a few years ago - "environmental factors", when you examine them, turns out to mostly be code for "random luck", not "how good a school you went to" or whatever.
There are definitely some well-replicated environmental risk factors (heavy cannabis use in adolescence springs to mind), but I don’t think they can explain the variation in risk
Perhaps "non-heritable factors" would be a clearer name.
I mostly agree with this, but IIRC the environmental component of schizophrenia variance is about half shared and half non-shared, suggesting there's more non-stochastic factors than usual.
My guess is these will be boring things like poor health and poverty, and not anything that will be specific to schizophrenia.
It's good to bear in mind that there are important ways in which heredity influences environment: Kids who are physically attractive and above average at motor coordination are more liked by their peers. Physically attractive kids are perceived by teachers as smarter, and are favored. Kids who really are quite smart, even if not physically attractive, are seen as smart by teachers. Simple things like this have a big influence on how many smiling faces kids see in their early years, and that number's a powerful feature of social environment.
I'm pretty sure poverty and poor health are going to be a result of having a schizophrenic parent. Consider the poverty cause of being related to a borderline schizophrenic person.
It'll be a contributor, but as Scott points out, the extreme rarity of schizophrenia means that most schizophrenics will not have close schizophrenic relatives themselves, even when looking at identical twins, even though genetic influences on the trait are very strong. So nature-as-nurture effects in that sense are likely to be minor, I think.
Others in the thread have suggested that heavy cannabis use is a significant contributor, though, and I think the literature suggests mild-to-moderate shared-environment influences on substance abuse disorders (maybe 20-30%), so maybe that indirectly accounts for a few percent of the variance.
Good article in any case, thumbs up.
"Random luck" as I would understand the phrase sounds a lot like something that could resaonably be described as "environmental". It's just something stochastic that you wouldn't necessarily expect to have a significant effect like "did you get chicken pox at 18 months or 20 months" rather than a more long term "environmental" effect like "poverty".
The really interesting thing is that a lot of "non-shared environmental" effect could actually be measurement error. This is probably less bad in the case of schizophrenia than many conditions as I understand it is somewhat two-peaked; but there must be some fuzziness; people who might or might not get a schizophrenia diagnosis depending on the doctor assessing them.
I think the point Mitchell makes is that brain development is itself a stochastic process.
So we can’t just assume the ‘non genetic’ component is ‘environmental’ unless/until we have compelling evidence for this.
Perhaps I'm misunderstanding your point - "non-genetic" and "environmental" sound like they're being used above as equal-by-definition, but it sounds like you have a different notion of "environmental" in mind. What, in your view, does "environmental" mean?
I think ‘non genetic’ can be divided into ‘environment’ and ‘random variation’.
Environment encompasses everything from the uterine environment, birth complications, early life trauma, racial discrimination, drug use, head injury, etc. Even if the environment is completely accounted for, I believe there would be a proportion of the variance that is unaccounted for - which is due to stochastic processes / random variation / luck.
Is that a claim about the practical limits of our ability to enumerate particular environmental factors, or a claim about irreducible randomness in physical processes?
I don’t think the randomness is reducible.
I’m not an expert in genetics though, it’s just what I’m taking from articles like the one I cited above
Leaving aside quantum mechanics for the moment, these two are the same thing, because the practical challenges in enumerating particular environmental causes of a given outcome go at least exponentially with the weakness of the causative mechanism (say the phase of the Moon or the position of α Centauri when neuron #123456789 was growing its 987-th dendrite).
For all I know, it might also include random effects from specific diseases and from pollution.
You absolutely can assume that the ‘non genetic’ component is ‘environmental’ if you define environmental as "everything non genetic." I believe that this is effectively what is being argued.
> The really interesting thing is that a lot of "non-shared environmental" effect could actually be measurement error.
Yes! Schizophrenia is not something easy to define or verify objectively in the first place, unlike having a peg leg or a glass eye. And even with more 'mechanical' diseases like blindness or motor dysfunctions it is far from simple, as the horrors of Paralympic qualification manuals show.
A problem related to measurement error which ends up in the rubbish bin of "non-shared environmental effects" along with the former is the arbitrariness and messiness of indicators which researchers are forced to use from lack of anything better in available data. E.g. there is little or none measurement error in EduYears, but neither does it reflect in a simple way the abilities or qualities of characters that we are really interested in, and there is no reason to believe that its relationship to the latter remains reasonably constant with time (in fact we have good evidence that it does not).
Geneticists call everything which isn't genetically inherited 'environment'; it's baked in to the terminology.
I guess you can argue about this, but 'random' is not straightforward to disentangle anyway, it's generally 'stuff we don't have all the details for', so it's a pragmatic decision.
It's very similar to the old: a <cancer|whatever> quick-test has a 1% false positive rate. Your test is positive. What is the chance you have <cancer|whatever>? 99%?
when only a very small percentage of the population have the measured condition, statistics are very un-intuitive.
I had to explain that to an actual medical doctor in 2020. Had a positive covid test when we A) tested the entire county employee population without any symptoms, and B) prevalence in the county was extremely low. I told him it was probably a false poz, he insisted the false poz rate was very low. I tried to explain to him that if you test everyone across the board with no independent reason to believe they have the disease, the false poz rate would be higher (as a prosecutor at the time, I of course had heard of the Prosecutor's Fallacy.) He called me back a week later with the lab PCR test, and sheepishly admitted I was correct and it was a false positive and he sure was surprised because they'd been told that wouldn't happen. Meanwhile my false poz wasted a week of my life, caused an inordinate amount of stress on my marriage, and forced cancellation of a monthly circuit court docket resulting in a massive logjam of criminal defendants.
He didn't even have to be familiar with statistics to get this right, it's intuitive. Other doctors early in the pandemic were stating loudly on TV that you shouldn't give the test to people without symptoms (when saying that was useful to make Trump's staff look wasteful or foolish by testing.) But then the incentives changed to "make as many people afraid of this as possible, for their own good" so suddenly basic logical reasoning has to be ignored.
Hmm I'm all for ragging on doctors but I think it was an ok precaution at the time when the vibe was 'hey, lets shut everything down and see if we can eradicate this thing'? Testing in asymptomatic populations was assessed from BMJ and Cochrane reviews below, if you had a positive result from a lateral flow you were ~50-60% likely to have COVID. Agree that it's not great, there was a lot of hyper pre-cautionary stuff done at the time, but I wouldn't throw the bucket out with the bathwater on this one?
https://www.cochrane.org/CD013705/INFECTN_how-accurate-are-rapid-antigen-tests-diagnosing-covid-19
https://www.bmj.com/content/373/bmj.n1411/rr
I guess my biggest problem is that when presented with an actual logical reason he was wrong, he just slavishly stuck to the script. There was so much derision directed at people like me who actually tried to learn things and apply reason and logic to them during covid, we were mocked for "doing our own research" rather than just listening to authorities, but in at least that one situation I was smart enough to know the authority was wrong about how statistics work and that I, the layman, was correct.
I could've respected it had they been honest and said "ok sure, you're right, the false positive rate is much higher due to very low prevalence in your area and the lack of any symptoms or contacts to supply an independent reason to suspect you have it, but the chances you're infected are still high enough that we would prefer you to be careful for 2-3 days just in case, and if you do feel these symptoms call me." Or heck, just admit that you tell the stupid people it's 99% accurate because they won't take it seriously otherwise, but when I call you on it acknowledge that. Instead we had the medical experts outright lying to my face about the statistics of asymptomatic testing, something that you did not have to be a doctor to understand, that any reasonably smart person could figure out. They did not seem to care that it was a lie, apparently trusting that anyone smart enough to discern the lie was going to play along, but nobody sent me the memo. I was particularly bothered because SSC was down during most of covid and I could not find any rational discussion of any of this, it felt like two years of my life where almost anything anybody said was designed to induce behavior rather than to correspond to reality.
Eventually tests *were* developed that had an extremely low false positive rate (in exchange for a not-so-great false negative rate). For a probability class I was teaching at the time, I looked up the error rates of the Covid tests that were required from students every week. I no longer remember the exact numbers, but the false positive rates were something like one in a few thousand. (So, if you ask all the students to take a test, you get false positives in the single digits; when there are about 50 positive results a a week, most of them are true positives.)
Of course, we need something more than simple probability when we try to interpret what this means. Does a positive result mean "contagious right now" or "contagious last week" or "about to become contagious"? This depends on the test as well.
The first argument is also often used to explain why in elite challenges (e.g. interviews to top companies) luck matters more than skill: since everyone who has a chance at all is already in the top percentile of skill anyway, the variance introduced by luck matters relatively more and is the main factor.
This becomes more true the more selective the process is. Since schizophrenia, like Harvard, is very selective, we expect the difference between people who get it and people who were strong candidates but didn't to be mostly luck (or environmental factors).
Yes, but this depends on the assumption that it's hard to distinguish (or not necessary to distinguish) skill among the top percent. I agree that's the case for soft skills--like communication, leadership, etc.--but hard skills can often be measured even at 4+ standard deviations out (e.g. consider chess or math competitions, where there are large gaps in performance even at top competitions).
I think this helps explain why highly selective software companies tend to use difficult algorithm questions in their recruiting, even though soft skills (architecture, code quality, etc.) are arguably more important in many cases. They are testing a hard skill that can provide a signal to distinguish the top 5% from the top 1% of candidates. Whereas soft programming skills aren't so easily distinguished that finely.
Skills vary by how easy they are to distinguish, but all skills are much harder to distinguish within the top 1% then at lower ranks (because skill levels tend to cluster at the top, due to normal distributions having thin tails).
For example, if the top two chess players in the world play a match, it's roughly a coin flip (actually worse than that, it's usually a draw). If you pick two random people from your middle school chess club, the better player has much better odds of winning than that (even as black).
Ad the nazi eugenic argument: I don’t think many schizophrenics would have children once the disease developed, especially back then when there weren’t any drugs ameliorating their condition. So the nazis killing them did not change the national genetic composition much.
Come to think of it, if the people they were killing off were long-term institutionalized populations, almost none of those people were going to have children ever again. Also because people had children young back then and schizophrenia has latish onset, a lot of these people had already reproduced.
Probably what's happening in the bigger picture is that there is a balance between schizophrenics having fewer children and de novo mutations occurring and increasing genetic risk. This is very pronounced for conditions like severe autism, there's evidence for it with schizophrenia as well e.g. https://pubmed.ncbi.nlm.nih.gov/34903660/
Another possible factor in both schizophrenia and autism is that some of the gene varients that can cause them may even be beneficial if they do not actually cause the disease, further reducing selection pressure. This is a bit speculative but I don't think it's unreasonable to consider these conditions as pathological extremes of normal human brain function.
I think the idea is independent of whether you consider SCZ to be on a continuum with typical brain function. The pleiotropic effects of the relevant genes would likely drive the positive selection. They could have effects (spitballing here) like slightly increasing the density of calcium channels in the basal ganglia or slightly downregulating the synthesis microglia, that don't necessarily contribute a phenotype that has recognizable continuity with schizophrenia.
Alternatively, risk variants experience positive selection pressure in the broader population.
If schizophrenia is partially caused by trauma, might post-war Germany have been a higher-trauma environment than most?
"if the people they were killing off were long-term institutionalized populations, almost none of those people were going to have children ever again."
If so, does that mean "eugenics" in this case was a fancy-sounding excuse for cutting costs?
Totally. In the Nazi ideology, the people were thought of as an organism (Volkkoerper), with individual humans subservient to that greater whole. So even if one takes aside their obsession with racial purity and genetic health, killing off people who can't pull their weight is totally compatible with their twisted mindset.
I guess the reason why they did not systematically kill all their war invalids or retired people is less ideological and more practical.
Your interesting point got me reading about this. They killed people with Down's syndrome, senile elderly etc. who couldn't possibly have contributed to the gene pool. Cost cutting was clearly a factor, as well as the ideological aspect quiet_NaN describes. I would add that once you decide some people are too disgusting to survive, it's logical that will occasionally extend to your own group.
The other interesting thing I learned in my reading is that euthanizing the disabled raised more overt opposition in Germany than pretty much anything else the Nazis did. A while back some people on the internet condemned the people of Russia for not rising up to stop the invasion of Ukraine, and then I read an interesting article that argued that no government had ever been overthrown for cruelty to outsiders. You can find examples of leaders overthrown for failing at wars, but not for being cruel in them. For example, Milosevic lost power because he was beaten in Kosovo, not because he was cruel in Kosovo. The pattern holds in Germany. There were people in Germany willing to risk their necks to prevent the death of their own country's invalids, but far fewer willing to do so for Jews, Slavs, Frenchmen etc.
Schizophrenia doesn't have a latish onset, though. In fact age of maximum risk for a first psychotic episode corresponds very well with age of maximum fertility.
From what I can tell, typical onset is in your 20s. So yes, it's certainly possible to have children after you have schizophrenia but ~1940, there's a decent chance you've already had some. Scott's spreadsheet exercise assumes that that killing all of the schizophrenics prevents them from having kids, but clearly many will have already had some before the Nazis kill them.
Condemning the Nazis for not applying a gene-centric view more broadly seems unfair. If they'd known all that has been learned over the last ~75 years, they could have sterilized all relatives up to first/second cousins.
I do not think "understand what is going on before you start murdering people" is an unreasonable standard to have. (Of course, I prefer the simpler "don't murder people".)
Of course, the Nazis would claim that they never intended to wipe out genetic diseases (or at least the people with them) in that half-generation in which they ruled, they were famously thinking about a millennium, which was fortunately cut short after some 12 years.
I believe the modern justification for the state committing atrocities on dubious evidence is the "precautionary principle."
...which funnily seems to be another german invention. https://de.wikipedia.org/wiki/Vorsorgeprinzip
In the world today, male schizophrenics have an average of 1 child. I don't know any studies of the timing of these births, but I also assume that they generally have children before onset. But this means that even if the Nazis reduced their fertility to zero, they would only have twice the effect that occurs every generation in the modern world.
Can we summarize the confusion as being something like base rate neglect? The prevalence of schizophrenia is so low that even knowing your twin has it doesn’t make your odds that high, because they’re so so so low to begin with. Going from 1% to 20% chance of having schizophrenia is a massive step up.
That sounds about right, yeah. All families have either 0% or 20% schizophrenia, 5% are in the 20% bucket.
The obvious parallel is autism, which is also a neurodevelopmental phenomenon with weird complex messy influences, a huge spectrum, and a tendency for many people technically on that spectrum to evade diagnosis due to severity bias. Concordance for ASD in monozygotic twins is consistently far higher than for schizophrenia.
Really? I hadn't realized that. Do you understand why that would happen if they both have about the same heritability?
I think the basic answer is that the core of the schizophrenia spectrum is schizotypy, not psychosis -- that is, looking at schizophrenia and expecting it to tell you much about "what it means for someone to have this" is a bit like looking at someone with skin cancer and extrapolating it to the entire population of Queensland, Australia. (Highest skin cancer rates in the world!) The match between "the diagnostic category of autism" and "the Truth that the category is trying to get at" is still, um, messy, but there has been for many years a general understanding that people without the most 'severe' possible presentation can still be autistic.*
I also think, yeah, 'schizophrenia' itself has a lot of meaningful and significant environmental factors that don't apply to autism. A lot of this is just an age-of-onset thing. I don't think this is all of it, but it's certainly a lot of it.
I assume MZ twins of people with an SZ diagnosis tend to be pretty schizotypal, but I expect some of them aren't very schizotypal, because variance between monozygotes is weird and somehow happens all the time. (Also, because sometimes people get diagnosed with a functional psychosis when the more parsimonious etiology is 'injecting meth into your eyeballs' or similar.)
*The concept that historical (e.g. Kanner's) concepts of autism are "severe", in terms of "very disabled" as opposed to "very autistic", is not really accurate. I'm not entirely sure when the concept of autism and intellectual disability being genuinely (rather than misdiagnosedly) comorbid came in, but it wasn't Kanner's idea. Asperger actually believed it more, but he still only believed it in terms of what we'd call mild or upper moderate ID, reading between the lines (Frith translates him as discussing "severe retardation", but he describes that in ways that sound like mild, not severe, ID).
Perhaps because of the more binary aspect of schizophrenia?
If the 99th percentile is psychotic but their sibling in the 98th percentile isn't, that's discordant. However, if the 99th percentile in autism is obviously autistic, then his sibling in the 98th percentile who's borderline autistic probably gets diagnosed too.
Yes. It is a very typical confusion of absolute vs relative rates
This also gives a nice intuitive explanation for the nazi paradox. The pool of people at risk of schizophrenia is much larger than the pool of people who have it. Wiping out the pool of people who have it doesnt put a dent in the at risk pool, who can then pass their at risk genes to the next generation.
I am someone who takes great interest in scientific findings outside his own area of expertise.
I find it rather disheartening to discover that most of it is rather bunk, and that you need to be an expert in the field, writing simulations, to prove how bunk it is. I find it even more disheartening to discover that almost no-one who is an expert in the field bothers to notice or debunk the bunk.
So I find it very refreshing (heartening?) that you are doing so. This blog is a goldmine for anyone who wants to see how things are going in Psychology, EA, forecasting, and other topics where you point your mind.
I guess I might as well make this a question for the commentariat... what techniques do you find for helping discover great insights in fields where you have an interest, but no background, and are thus susceptible to falling for the bunk? Another way of putting it, if you were in charge of writing summary articles about all the cool stuff in various fields, how would you make sure you weren't being duped?
Torrey's arguments with the twins was actually a little more subtle than I have here (although some of his partisans made the less subtle one I made here). He first of all argued that the 15% twin concordance rate was the best (I haven't read the studies and can't comment), and that some infectious diseases (eg tuberculosis) have twin concordance rates around that level. I'd be interested in learning why that should be - maybe because there's a strong genetic component to how well the immune system does at fighting them off? - and if his point was just that it might still be infectious, fair. But I think he was implying something stronger than that.
I hadn't realized "infectious" was in competition with "mostly genetic". I thought the argument (as per Cochran) was something like:
genetic susceptibility + catch the right infection = horrible disease
That's also possible!
Interactions between genes and environment are assumed to not exist in the standard model of heritability. That is, when you say something is "X% genetic", you are assuming no interaction between genes and environment, no GxE term.
Instead, the standard model assumes a SUM of a genetic effect and an environmental effect, like you did in your spreadsheet. A product is not allowed. This means that "AND" gates like "genetic susceptibility AND catch the right infection" are ruled out by assumption in the model.
(The model sucks, in other words.)
Here's something by Matt McGue for people interested into delving into something outside their own area of expertise:
Topic #11: Schizophrenia: Clinical Genetics
https://sites.google.com/umn.edu/behavioralgenetics-mcgue/home-topics/topic-11-schizophrenia-clinical-genetics
as well as
Topic #12: Schizophrenia: Identifying the Genetic Variants
https://sites.google.com/umn.edu/behavioralgenetics-mcgue/home-topics/topic-12-schizophrenia-identifying-the-genetic-variants
There's more too, but this should be enough for starters.
Those videos in the first link are really good. Do you know, are they part of a psychiatry/MD program or is it more genetics specific?
They are from PSY 5137 - Introduction to Behavioral Genetics - Fall 2020 by Matt McGue, PhD Regents Professor, University of Minnesota, Department of Psychology
https://sites.google.com/umn.edu/behavioralgenetics-mcgue/home-topics
Hope this helps!
PS: New to this site here. No idea why my earlier reply seems to have vanished.
This may not answer your question, but it does remind me of https://slatestarcodex.com/2019/06/03/repost-epistemic-learned-helplessness/
I think this is a really rare complete miss for you, for the reason that “is mostly genetic” is just a very naive way to think about any disorder. Both Torrey and Aftab’s articles are specifically trying to reach past this kind of naive conception to point at the very thing you discovered in your analysis: that while genetics are a primary risk factor, they simply are not sufficient to cause schizophrenia alone, and waving a hand at ‘polygenic background’ doesn’t do anything to explain why, even in the presence of enormous genetic risk, there are critical environmental factors which it would be wise for us to think of as causative (even if they wouldn’t be necessarily causative in the presence of a lower genetic risk profile).
Honestly I don’t understand the reluctance to embrace the model that “maybe if hundreds of genes interact in unclear ways to create an individually highly variable risk, and it turns out identical genetic risk profiles will or will not develop schizophrenia based on environment, it’s not so smart to insist on thinking of causation as genetic.” It’s not like this explanation elides genetics, it just doesn’t throw in the towel on causation once we discover that causation isn’t purely, or even isn’t ‘mostly’ environmental.
There is a level where clinging to your mental model as “mostly this, therefore because of this and should be talked about as this” is unhelpful. Here’s a probably overloaded metaphor. If I build a building, and it collapses due to wind load, I don't just stop at saying the cause was wind. I talk about how the designers knew that wind was a primary risk for this sort of building and the design was insufficiently prepared for wind. What we are finding out is that certain buildings by their nature are more exposed to the wind, and are going to require additional reinforcement. Now we need to know what the specific scenarios are whereby winds arise and interact during the storm to collapse all buildings that collapse via wind; knowing that there is a lot of risk for particular styles of construction helps us, but it isn’t enough to say “wind did it” and put your fingers in your ears when it comes to the other factors that made wind a critical actualized risk rather than a potential risk. It is proper in this scenario to say that the cause is poor engineering, which should account for the wind and doesn’t. Wind is everywhere and we aren’t going to be eliminating it anytime soon, right? Likewise with the literally hundreds of genes with unclear interactions that create risk for psychiatric disorders.
Perhaps we need better ways to express "more nature than environment, and it happens frequently" and "more nature than environment, but it happens very rarely". Because in the latter case, a tiny change in environment could have prevented the outcome (but so could have even a tinier change in the genome).
I will reuse your metaphor with bridges: imagine that 99% of bridges are built so that no wind could possibly break them, and 1% of bridges are built so that a very strong wind can break them... but strong wind happens rarely, so 95% of the less resistant bridges are lucky enough to never break..
It makes sense to say "the bridge broke because it wasn't built well", and it makes sense to say "the bridge broke because of the wind". And it may kinda make sense to say "a bad bridge breaks in 5% of weathers, while a bad weather only breaks 1% of bridges, therefore the quality of the bridge matters 5x more than weather".
Torrey's arguments are, basically:
1) It is nonsense to say that engineering matters, because even if we have 2 identical bridges built by the same company, most of the time both remain intact, sometimes 1 of them breaks and the other does not, and only extremely rarely do both break.
2) Last year we have demolished all broken bridges, instead of fixing them. And yet, this year a few bridges broke again. If engineering matters, how is it possible that those bridges that broke this year didn't already break the last year?
And Scott responds:
1) Even the less resistant bridges break rarely. So if we have 2 identical bridges, if they are more resistant, neither breaks, but even if they are less resistant, still the most likely outcome is that neither breaks... and the second most likely outcome is that one of them breaks... and only exceptionally both break. That does not contradict that 99% of bridges are built to never break, and 1% depend on luck but mostly get lucky.
2) Even a less resistant bridge only has a 5% chance to break, which means a 95% chance that it will not break... followed by a 5% chance that it breaks the next year. The difference between 0.05 and 0.95 × 0.05 is small; it is 5% vs 4.75%.
I don't know if I"d call it a complete miss, the spreadsheet low-cost simulations were pretty enlightening. But I'd still agree on the common sense of the meaning "mostly genetic".
Even if you can show in terms of statistical definition that genetics contributes 80% of the effect, a situation where only 10% of people with the faulty genes will get the condition is still *not* intuitively seen as "mostly genetic". In the way that actually matters to the individual, it's more like "genetics and environment need to combine just right".
Regarding Torrey in particular: well, I try to say very little about the sort of person who's a first-degree relative of someone with Scary-Name Neurodevelopmental Disorder that blatantly represents the far end of population variance, and is absolutely dead insistent that Scary-Name Neurodevelopmental Disorder traits represent the face of evil itself, given we know that in all these cases first-degree relatives of people with Scary-Name Neurodevelopmental Disorder generally have a ton of Scary-Name Neurodevelopmental Disorder traits. But I can see why he (and many others for various SNNDs) is so enraptured by anti-genetic explanations for this particular SNND, yes. The fact Torrey is a disciple of Gajdusek is something I've been thinking about lately -- Gajdusek is one of the people with a claim to discovering prions, and is, uh, a 'character' (a self-identified "pedophiliac pediatrician", amongst other things). I'm not entirely sure what to make of the Gajdusek connection, given how little anyone focuses on it. Gajdusek is, by the words of anyone who interacted with him, not someone you forget easily.
In the specific case of schizophrenia, the "all first-degree relatives of people with SNNDs basically have SNNDs themselves" problem stands out somewhat. Research on SZ is really bad, consistently, because it comes from a psychosis-focused perspective and not a schizotypy-focused perspective -- there's increasing recognition of concepts like the "psychosis spectrum", but the keyword still tends to be psychosis. "At-risk mental states"/"prodromes" are some of the least-bad people have gotten, but are still terrible, because they're written under the assumption people diagnosable with those terms will *usually* develop psychosis, which is clearly untrue. Also, holy shit, have you seen a 'prodrome' test? They're awful. They read like they were written by someone trying to strawman the worst ends of psychiatry. "Do you have strong feelings or beliefs that are very important to you, about such things as religion, philosophy, or politics?" "Do you daydream a lot or find yourself preoccupied with stories, fantasies, or ideas?" "Do you usually prefer to be alone?" "Do you find that you have trouble getting motivated to do things?" Well, bad news, kid -- you might be a latent schizophrenic! Here, enroll in this trial where we give you antipsychotics.
Anyway: I think schizotypy, proper, is pretty intrinsic, though it's also modulated by other factors (e.g. developmental stage, paranoia-inducing experiences). When you're looking at a particular extreme subset of the schizotypal population, this is going to be muddled. I'm not sure if people who end up with an SZ diagnosis are particularly representative of the schizotypal population, not just in terms of "being on some extremes" but in much deeper and more fundamental senses. I have the strong and consistent impression that SZ vs bipolar diagnosis primarily loads on things other than presentation of the psychosis alone, for instance, and that's already selecting for the minority of people who have such obvious and serious psychosis that they get diagnosed with something and stick around the system for long enough other people notice.
"the strong and consistent impression that SZ vs bipolar diagnosis primarily loads on things other than presentation of the psychosis alone" - how come? I'd think that differential diagnosis is comparatively easy. If a paranoid-hallucinatory syndrome occurs with obvious manic symptoms and a history of episodic major affective disorders, which happens extremely rarely, I'd call it schizoaffective disorder.
Something I've wanted to mention here for a while:
One point about twin studies in general is that often the twins are not separated at birth, but several years afterwards. So overall this supports the lead-crime hypothesis, because otherwise, the environmental effects are too large, and lead proves too much (and I first heard this from the "Lucifer Curves" author). But it also means that MANY of our twin studies could be confounding "early childhood environment" with "genetics".
Of course, GWAS isn't susceptible to this bias, so when the twin studies agree with GWAS, we should probably accept the overall point.
I think this comment is conflating twin studies and adoption studies. Classic twin studies involve normal twins who were never separated/adopted.
I agree that it's very surprising they don't find a lead effect for more traits.
Classical twin studies use twins raised together, and estimate heritability by exploiting the fact that identical and same-sex fraternal twins have the same upbringing but different degrees of genetic similarity. Separated-twin studies are interesting, but much less common.
Thanks to both for the clarification. So to summarize:
1) twin studies: same environment, identical-vs-halfshared genetics
2) adoption studies: different environment, identical-vs-halfshared genetics
I do think it's worth remembering that almost everything is horribly underpowered, because researchers assume that "medium" effect size is something that's subtle to notice. In reality, a "medium" effect size is something like "people who like eggs eat egg salad more often than those who don't" (i.e., something that we knew before we ran the experiment). See, eg http://datacolada.org/18
So it's worth remembering that lead is a HUGE effect size... and indeed we (should) have known that lead causes serious problems long before we started this line of research. We knew it from similar metals (eg mercury and the mad hatter), from lab experiments on mice, from case reports where workers exposed to lead went crazy. It would be flabbergasting if lead DIDN'T cause crime. But almost everything else is going to be smaller than the lead effect, and going to require many hundreds of observations per cell (cf link)... which very few studies have.
But with the convergent lines of evidence, and things like Hatemi's "Not by twins alone" (claiming 40k participants), I agree with the overall take about how large genetics is.
"lead is a HUGE effect size"
That entirely depends on dose.
The effects of tiny amounts of lead are not so easy to pick out.
You could equally say that drinking water has a HUGE effect size when looking at causes of death.
Because a small number of people drinking insanely huge quantities destroy their kidneys or throw their electrolytes out so much they die.
Statistical effect size and your gut feeling of how impactful something is aren't concepts that can be smoothly swapped.
I'm not sure what I'm supposed to take from this. Are you arguing that the lead-crime link is due to a handful of people who received massive doses of lead and killed thousands? That would be really interesting if true! But I'd definitely like a citation if so.
Otherwise, it sounds like you're saying "the dose makes the poison", which is true, but in casual or charitable conversation normally one assumes the relevant background. I.e. the effect of lead is large RELATIVE to the amounts we were putting into the environment, and we can have a large societal impact by reducing the toxin.
If it's something else, I'm baffled.
It's in response to this statement.
"we (should) have known that lead causes serious problems long before we started this line of research."
The effects of tiny amounts of lead are not dramatic until you do research at huge scale. You can't just point to big effects from huge doses and assume tiny doses are the same because there's so many cases where that approach fails.
If I understand heritability correctly, 80% heritable means that a 1-SD increase in genetic susceptibility gives you a sqrt(0.8) = 0.89-SD increase in risk, and a 1-SD increase in environmental risk factors gives you a sqrt(0.2) = 0.45-SD increase in risk. 99th percentile is about 2.33 standard deviations. So, for example, if you have +2-SD genetic susceptibility, you would need to be at about the +1.22-SD = 89th percentile in terms of environmental risk factors to be at the 99th percentile for schizophrenia risk.
You would need to be at about +(2.33/0.89)SD = 99.6th percentile of genetic susceptibility to have a 50/50 chance of being at the 99th-percentile for schizophrenia risk. Assuming that the environmental risk is all non-shared, you would still only have 50% concordance for identical twins at the 99.6th percentile of genetic risk.
This is really cool, I love these simulation-based sanity checks for unintuitive things in probability theory.
Btw the company Orchid Health is apparently working on polygenic embryo screening for schizophrenia, but people are upset about it and the group whose data they're using, the Psychiatric Genomics Consortium, are trying to rescind permission for their data usage:
"PGC objects to such uses because its goal is to improve the lives of people with mental illness, not stop them from being born"
https://www.science.org/content/article/genetics-group-slams-company-using-its-data-screen-embryos-genomes
This is pretty terrible if you ask me, and I hope it doesn't turn out to be a legal hurdle for Orchid.
"PGC objects to such uses because its goal is to improve the lives of people with mental illness, not stop them from being born"
One really has to wonder what their possible objection could be to the obvious rejoinder: “why not both?”.
This is an interesting issue. One reason genetic screening makes a lot of people uncomfortable is it seems to imply that the people in their life who have a certain disorder should never have been born.
True. I guess the way to frame it is that the people who are alive are only a tiny fraction of the potential people who could have been born of the same parents instead, if a different sperm had met a different egg, and that "people" who are never conceived are not noticeably inconvenienced by that fact, so we can treat them in expectation as interchangeable - thus if someone with a certain disorder had never lived, and instead their putative sibling without that disorder had lived instead, we can treat that world as basically analogous with swapping out the person with the disorder for a version of the same person without the disorder. This is ... not wholly satisfactory, but I'm going to struggle to do better to square my intuitions here; I am very negative about people who would forbid genetic interventions that mean that future people don't need to suffer from serious life-limiting genetic conditions just because it would somehow say something negative about living people with those conditions. As long as we treat the actual people with those conditions with fairness and compassion, of course.
That phrasing raises the question: if a pill that could 100% prevent the development of psychosis was invented and given to at-risk babies at birth, would they consider that erasure of people with mental health problems since they would no longer be a distinct community.
Rather than merely making the lives of people who've developed psychosis better.
I'm still unsure about these issues. In the case of schizophrenia, the kid would have a 80% chance not to become psychotic and scaling such practice would reduce the spectrum of human personalities, maybe significantly. Who knows when a lot of schizotypal people will be needed to save humanity?
If there were single gene polymorphisms with large negative effect, they would get selected out of the population ... eventually. Which suggests that there can't be high-frequency mutations with large negative effect, unless there is some compensating advantage (like, e.g. giving you resistance to malaria).
Which leaves us with multiple mutations, each of which individually has a small effect, adding up to a large total effect. And mutation-selection balance, where random mutations are introducing harmful mutations at about the same rate the natural selection is removing them.
I remember coming across a paper on exactly this topic: "Extreme Polygenicity of Complex Traits Is Explained by Negative Selection"
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6732528/
Or it's a recessive gene, which are hard to select against unless a lot of incest happens, which usually isn't the case because incest is already selected against.
The exception that proves the rule is that there are a bunch of highly penetrant, dominantly-inherited neurodegenerative diseases like Huntington Disease and various forms of familial ALS that that rarely do any real damage until after prime reproductive years, and thus have minimal effect on reproductive fitness.
I think it's interesting that your example kinda proves the point of the import of environmental influence on schizophrenia, and with that maybe shows the quite logical mistake people make when they focus on the fact that schizophrenia is 80% genetogenic. It seems that environmental factors would only make a difference in 20% of the cases when reasoned naively, but you can see that for the ones with a high enough genetic risk, environmental factors actually contribute 90% to the the development of schizophrenia! It's just that if you don't have the genetic risk it's nigh impossible to develop schizophrenia, even in a highly conducive environment, yet if you're one of the 10% of people that has a higher genetic risk the environmental factor becomes highly important.
This is true, but I think it's important that it's symmetrical - I put genetic risk first in my spreadsheet, but it's not like genetics merely predispose you and then environment deals the final blow. You could equally frame it as "only in people with environmental risk does the genetics become important".
(or at least this is true in my simulation, and what I expect to be true of reality - I think other people differ).
Agreed!
Isnt sickle cell trait a good example of why you can’t draw this conclusion from the numbers only? We know exactly which gene causes sickle cell disease, which is genetic in the intuitive sense. That gene also protects against malaria, so we must see a hereditary component to malaria if we purely look at the stats, but it’s not really what we think of as genetic.
"genetogenic"
I ... understand exactly why this *ought* to be the word for "caused by genes", but my natural reaction is still "Thanks, I hate it" :-)
I think it's pretty swell. "Having its origins in origins."
Actually, I just realised, it's gnarlier than that: the "-genic" suffix is sometimes referring to the causal agent, as in iatrogenic harms, being harms caused by a doctor, and sometimes referring to the effect, as in carcinogenic substances being substances that cause cancer (as opposed to substances that are produced by cancers). So "genetogenic" could refer to that which causes things to be caused. Paging Sheherezade; we may have a deep regress on our our hands.
Or Aristotle. We should check with the apologists, genetogene could be another of God's names (Gene for short, naturally).
I would also recommend a recent (2023) Nature Human Behavior review article on the topic of twin concordance vs. heritability: https://www.nature.com/articles/s41562-023-01609-6
"While genetic effects are evident from the higher concordance in MZ twins than in DZ pairs, concordance in MZ twins does not approach 100%, even for highly heritable traits such as schizophrenia"
> I think you should expect very slightly fewer schizophrenics in the new generation, but the effect size wasn’t noticeable in this small granular simulation - nor, apparently, in Germany.
Since the population of Germany was so much larger than 2000, shouldn't statistical significance be much easier to pick up?
Yes, but you also don't have all the schizophrenics nicely highlighted in red - you have to hunt them down, diagnose them, and get that fact to the attention of whoever's doing a study. If you're comparing pre-Nazi to post-Nazi Germany, you also have to deal with a few decades' worth of evolving diagnostic criteria.
One might also expect that the Nazis generate a significantly worse environment, which would counteract the effect (if any) of the changes in the gene pool.
I don't know that they increased cannabis use, but they did use meth.
Also: it is in principle possible for something to be both generic and caused by a parasite (the genetic variation is improved resistance to parasitic infection).
(Looks at cat infected with T. Gondi suspiciously ...)
>A better model would have to take into account that people’s children aren’t clones of themselves, and that children’s environment is correlated with their parents’. But both of these would drive schizophrenia rates up, not down,
I believe the first would drive the rate up (because the variance in genetic schizophrenia factors would be higher), but the second would drive them down (think of the edge case where children's environment is perfectly correlated with their parents, and children are clones of their parents: in that case, schizophrenia/non-schizophrenia is perfectly passed on to children, so if you remove the parent schizophrenics, there won't be any schizophrenic children).
This is the accepted model: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1685995/pdf/ajhg00354-0087.pdf
I know nothing about schizophrenia, but I was slightly surprised that the model you use is something like compute "genetic score" + "environment score", and if you are above threshold you develop it.
What I expected to see is something like "genetic score" + "environment score" give you your probability to develop schizophrenia (which can fairly low even for worst scores since it is not that widespread), then you roll the dice and either get it or not depending on your luck. In this model even if my twin I share environment with develops schizophrenia, it means that my baseline chance is likely significantly higher than average, but there is still a decent chance I won't develop it because it is fairly rare.
Your model kind of assumes that some people would be genetically "immune" to developing schizophrenia; do you think that's closer to how it works?
I think his model is a simplified one in which the "environment" score incorporates luck (such as with what is commonly labelled "non-shared environment").
Would tacking on some more randomness to the model yield any additional insight? This simple one captures the desired phenomena.
This conversation brings to mind guitarist Peter Green's schizophrenia. We saw him perform in what looked like a bathrobe, with a giant crucifix swinging from his neck. With Danny Kirwan, Jeremy Spencer, etc. in a seedy carpet store on Market Street Bill Graham called Fillmore West.
Later in his music career, Green released ethereal solo recordings that were much calmer than his Fleetwood Mac work. But he got concerned about starving children shown on TV, and felt somewhat guilty about living a pop star's self-absorbed life. ("Couldn't we just make them a sandwich?")
Things got more complicated when he busted up his brother's collection of crystal ware, in an emotional outburst. The family got help for him, and he was diagnosed with schizophrenia. How? The initial story was he was basically kidnapped by a hipster cult of Germans and dosed with some particularly weird acid. That theory was eventually pretty much debunked. So, was Peter schizophrenic when he was madly grinning and playing his guitar? Was channeling a sort of madness into guitar solos therapy? I don't know about his family history, but Green, whose name by birth was Greenbaum, I believe, shortened to sound less semitic, did talk about insults he had suffered growing up Jewish.
But it likely wasn't notorious German LSD or antisemitism that caused his illness. It does cause me to wonder whether his music would have been so rich if he hadn't been a little bit 'crazy'.
Fleetwood Mac burned through a surprisingly large number of guitarists. I wouldn't have even thought of him as the one allegedly lost to a cult!
The onset of schizophrenia in adulthood is more like massively amping up already existing weirdness rather than moving from the binary state of normal to mad.
Oh, we know the answer to this! It's called schizotypy, which is tri-dimensional (positive, negative, ???disorganized???). Positive and sometimes disorganized schizotypy read like they was written by someone making a satire of psychiatry; it amounts to the "Do people think you're strange? Do you?" of the famous Subgenius meme, but swaps it from "better than most people" to, in a traditional framework, "worse than most people". The quote from a 'prodrome' test I gave earlier, where it asks very seriously how much you daydream and if you care about philosophy and religion and politics, is...not wholly unrepresentative.
Negative schizotypy is the Asociality Grab-Bag. Everything in the DSM is correlated with negative schizotypy, because everything in the DSM is correlated with being socially reticent or anhedonic. It's trickier to talk about in coherent terms because of this. Disorganized schizotypy is even worse, because it serves as a dumping ground for questions about eccentricity and subthreshold thought disorder. All of them form the whole of "being schizotypal" in combination, and in most people who have a classically schizophreniform experience at some point in their lives (amongst other things), but the likelihood that someone in the general population who is very high on one is also very high on the other isn't too high.
The answer to "would this person be so [X interesting trait] if he wasn't crazy [i.e. schizospec]?", is no, because we know what the real long end of the schizospec is, and that psychosis is "massively amped-up weirdness" (to quote) rather than something that just happens to neurotypical people. The answer to, to subtweet a few other comments, "wouldn't screening-out of embryos with high polygenic risk scores for psychosis be a good thing?" is ha ha ha no, I cannot think of a single disease screening that would be worse, possibly including autism. (Autism is more trickily borderline, because its research problem is less about ignoring the ultramajority of the spectrum, and more about a hellish amount of diagnostic substitution.)
You write some exceptionally interesting posts. I wrote but didn't submit a response to one yesterday. Personally, I wish they were a bit more readable.
I get a little rant-y :) I'm working on a longer-form, more-edited-than-quick-comments piece on a a somewhat related subject (the complex grounds of what constitutes a *monogenic* genetic syndrome, in that many have much broader or milder phenotypes than you might think), though it's slow-going.
Schizotypy I tend to get maybe too rant-y about, because it's a frustrating subject for a few reasons:
1. It's *so obviously* the "core" or "real" thing to look at when looking at the schizo-spectrum. This is immediately obvious if you, say, compare the best prevalence estimates we have for schizotypal personality disorder with the best ones for schizophrenia. It's obvious if you study relatives of people with an SZ diagnosis. It's obvious if you study what people were like before they had a psychotic break. It's obvious if you look at...well, anything. But it's not obvious in any way that makes it into research. Autism provides a parallel here; it's a 'spectrum' (a tricky and complicated one itself), and it does face serious research problems for that fact, but people certainly research the 'high-functioning' autisms, and the fact they're in many ways more valuable to research than the 'low-functioning' ones (because the diagnosis is probably not wrong, which it often is for 'LF'A, and the clinical picture isn't so confounded) is recognized.
2. The criteria for schizotypy are...at least arguably something that shouldn't be called a 'disorder' or a 'bad thing'. "Something that distresses some people with them", sure. "Something that distresses some relatives of those people" -- oh, yes. But (again, like many of the autisms), the idea they're something that should be treated through an inherently problematic lens rapidly raises tons of issues. (There is a foreseeable future, somewhere down another path, where arguments about the schizospectrum take on some of the qualities we see today in arguments about the autism spectrum -- the kind that get simplified into "high-functioning self-advocates" and "relatives of low-functioning people", neither of which are quite right or form natural categories.)
The second is very, very difficult, because people (for many reasons, many good) want to justify their intuition of schizophrenia as a wholly bad thing, and something that e.g. it would be beneficial without consequences or horrible backfires to "screen out". Because many elements of schizotypy clearly contradict this, you end up in a bind.
The former is annoying for a different reason, which is that it kneecaps research. "Causes of" schizophrenia are not a coherent concept in the sense that most of them haven't been properly cross-referenced for "is this true of schizotypal people, or just some people who develop psychosis?". (I wonder if "polygenic risk scores for schizophrenia" and "polygenic risk scores for schizotypy" are non-identical.)
To throw fuel on the bonfire, doesn't this also steer *away* from all those embryonic genetic screening business? After all, if you can have 80% schizophrenia genes but not develop it because of environmental factors, how many of the other "eek! genetic risk!" alarms are likely to occur if you let the embryo develop into a pregnancy, be born, and grow up?
With IVF I believe there are typically large numbers of embryos to choose from. So maybe you can go from a 10% probability of a bad outcome to %1.
I haven't seen the kind of diagnostics these specific companies do, but if they're anything like pre-FDA-intervention 23andMe you'll get the "predicted risk of $DISEASE: X% instead of population average Y%", rather than "alarms".
It just makes sense to give the raw stats instead of substituting the judgment of the company to that of the parents, especially since out of a bunch of IVF embryos no single one will be better on all measures than the others, and any choice involves a trade-off between different risks.
I agree that high PGS score/high genetic risk for condition X doesn't mean you are going to develop condition X with any particularly high degree of probability compared to someone who wasn't selected. However these sort of screening businesses don't normally screen for just one thing. they screen for dozens up to hundreds of different disorders with a genetic component. And because how averages work if you choose a "low risk" embryo with a score saying it will have on average 2 out of 100 conditions (lets assume them to all be equally bad) then this embryo will with very high probability (close to 100%) grow up to have fewer bad conditions than an unselected embryo whose score suggests it will have on average 8 out of 100 conditions.
Even though for each individual condition X out of our 100 both the selected and unselected embryos have very close probabilities of developing the condition, when you aggregate over all 100 these differences in expectation add up and when you combine this with the averaging effect of combining multiple small individual contributors then with very high probability the realisation of the screened embryo random variable will be healthier than the realisation of the unscreened embryo random variable.
As other comments elsewhere have summarized it - imagine you have several bridges. ~90% of them are structurally sound, while ~10% are not structurally sound, and are vulnerable to collapsing in high winds. But high winds don't happen very often, so only about ~10% of the vulnerable bridges (~1% overall) collapse. So, you have one group that is immune to high winds, and another group that has to depend on luck - but most of them get lucky, so only a small number actually collapse.
Regarding embryo selection, the question becomes: would you rather have a child who is immune to XYZ disorder, or would you rather have a child who needs to be lucky?
I would like to point out that this isn't a one-off disagreement about the genetics of schizophrenia (or any other disorder). As a devout Catholic, IVF/embryo selection deeply offends you, with years of commenting history attesting to that fact.
Every time this topic comes up, you grasp at any shred of evidence that embryo selection will somehow not be effective after all. You are absolutely giddy at the prospect of being able to say "see? I told you so!" to any parents who are trying to have healthy children through "immoral" means such as embryo selection.
"You are absolutely giddy at the prospect of being able to say "see? I told you so!" to any parents who are trying to have healthy children through "immoral" means such as embryo selection."
Jeepers, and I'd have gotten away with it too, if it hadn't been for you meddling kids!
Yes, that is absolutely what I was trying to do.
Or maybe, juuuuuuust maybe, I was asking an honest question? If there's a screening company that comes back with "1% chance of Obscure Syndrome You Have To Look Up Because You Never Heard Of It", what is the real risk? What is the weighting of genetic versus environmental? If there's "You carry genes for breast cancer" is it worth getting a preventative double mastectomy? Some people think the risk warrants such an intervention, others don't.
If twin studies show that even with the same genetics and brought up in a similar environment, one may be schizophrenic and the other won't be, then I think we know less than we think we know about how illness works and the genuine risks.
But sure, put it down to me wanting to pwn the libs or whatever the cant phrase of the day is.
This sounds more like a science communication (or science understanding) issue.
I asked a convenience sample at breakfast what "80% genetic" meant to them. Generally the reply was "80% of the cases have it entirely due to genetic reasons with environment playing no part, and 20% with environmental reasons and genetics playing no part". I asked what about ones where environment played some part and genetics played some part. They updated their answer to be 80% genes with no environment, an unknown X% environment with no genes, and 20%-X% mix.
I then laid out the argument and spreadsheets in the article, and asked how they would choose to describe the reality of the spreadsheets, in their own words. The consensus reply was "It's 100% genetic and 100% environmental, but X% of the population is immune regardless of their environment, and Y% of the population is immune regardless of their genetics, and those populations are non-exclusive, and you could construct a Venn diagram to easily and accurately communicate it to the general public."
Then one person pointed out something interesting: 20% environmental is somewhat circumstantial. Since there's no absolute way to define the environment, it's a number that's derived from the types of environment available for the study population, and theoretically a more powerfully negative (or positive) environment could exist or develop in the future that would change the ratio from 80/20. And therefore the 80/20 mix isn't something set in stone but instead set in the social milieu of the study and there should be an expectation that it changes over time or study population.
Thanks for taking the extra step to ask some questions! Now that I think about it, that does sound like a common interpretation. I’m less certain that the general public would easily understand if given a Venn diagram. The “X% and Y% immune” seems even worse for general understanding. But clearly, just saying “80% genetic” doesn’t communicate reality. What would a single-sentence, or ideally sub-ten-word explanation look like?
That's a good question, and one that I think will have a different answer for each individual gene and trait discussed. Take for example horse size, something I read about very thoroughly a while back for a simulation project. Suppose there are two known alleles for the LCORL gene that have these effects on height: small/small -> x1.0, small/large -> x1.03, large/large: x1.06 and two alleles of HMGA2 with these effects: small/small -> x0.81, small/large -> x0.94, large/large -> x1.0 (these numbers are chosen such that whichever allele I believe to be the wildtype scales by 1.0). And suppose there are maybe 2-4 other unknown genes with similarly sized effects, plus a bunch of less noticeable ones. Suppose also that stallions are on average 2% taller than mares. (Edit: castration also has an effect, though I forget how much and in which direction) And we know nutrition also has a large impact on size, picking numbers at a wild guess maybe a malnourished horse would be 10-20% shorter. And then some amount of size variation from randomness, let's say plus or minus 5% height.
At the end of all that, how do you go about describing how much of horse height is genetic and how much is environmental? How do you communicate that % genetic and % environmental/non-heritable is going to vary depending on whether you look at all horses, only purebred Thoroughbreds, only well-fed horses, or only clones of one particular horse?
Any way you look at it, 100% of horses have genetic factors affecting their height, and 100% of horses have environmental factors affecting their height. So I guess the only hope of being able to communicate anything about this in a short and clear sentence relies on the listener understanding that everything comes from multiple causes.
There isn't a sub-ten-word explanation for most lay people with an average level of biological literacy. To address what I think are most people's concerns in asking a question like "is schizophrenia genetic?" I would say something like "We cannot predict who will get schizophrenia. Most of the known risk is genetic, which means that people with close family members with schizophrenia are at higher risk. But most people with a family history of schizophrenia will not develop it." And follow up with quantification if desired.
> The “X% and Y% immune” seems even worse for general understanding.
I disagree. I think the "immune" framing is a really great way to describe it, but more as "X% immune" instead of "X% and Y% immune."
There is another comment chain about bridges. Imagine you have a group of bridges where ~90% are structurally sound, and ~10% are vulnerable to collapse in high winds. But high winds don't happen very often, so only ~10% of the vulnerable bridges (~1% overall) end up collapsing.
You can frame this as "~90% of the bridges are immune to high winds, and ~10% have to rely on luck - but most of them get lucky, so only a small number collapse."
I've also heard it described as "genetics loads the gun, but environment pulls the trigger."
> Then one person pointed out something interesting: 20% environmental is somewhat circumstantial. Since there's no absolute way to define the environment, it's a number that's derived from the types of environment available for the study population, and theoretically a more powerfully negative (or positive) environment could exist or develop in the future that would change the ratio from 80/20. And therefore the 80/20 mix isn't something set in stone but instead set in the social milieu of the study and there should be an expectation that it changes over time or study population.
Yes, I always considered this to be a problem with interpreting these analysis-of-variance numbers.
An example: imagine a society where some people have huge amounts of lead exposure, others have very little. Then studies will show IQ is mainly environmental, with genetics not being important.
Next, imagine a society where every child is fully nourished and has an incredibly enriching education. Then, because everyone is achieving near their genetic maximum potential, studies will show IQ is mainly genetic, with environment not being important.
Despite the opposite conclusions, in these two examples, the biological realities are identical.
What's the alternative to analyzing variance? Ideally there could be a model developed, saying something like IQ = (genetic potential) - (lead poisoning) - (lack of education) - (poor nutrition) - (luck factor). This would provide far more useful and actionable information than just factoring the variance into genetic/environment.
Does the uniform distribution make sense for the genetic component (or the environmental component for that matter)? Seems like a power or normal distribution would make more sense, especially given the low appearance rate in the population. Using those distributions changes the simple model considerably. Another consideration is the deviation in environment vs genetics - that has a big impact on how the model shapes out - if the outliers are on the environmental or genetics side drives the analysis.
Man, it’s always the damn base-rates! Always.
Even some of the weirdness around high heritability but low variance explained by the SNPs we’ve found may be explained by rare variants (low base-rates) that aren’t found in ordinary GWAS setups. And previous missteps in psychiatric genetics can be explained by not considering the base-rate of how many (underpowered) studies were looking for candidate genes.
Nice. I ran the same simulation and got similar results. Then I ran the simulation out to 5 generations of Nazi genetic experiments. For the budding Nazis out there, after 5 generations, I did start to see an effect! G2=19, G3=15, G4=12, G5=9.
So after only five generations of murderous crimes against humanity (~100 years), Eugenics reduced schizophrenia by half! Of course, there's the pesky side effect that many of these genes probably have some positive effect when they don't manifest as disease...
But maybe a return to the atrocities committed in service to the Eugenics movement isn't necessary - even in a world where a disease is 80% genetically driven. I took Scott's same model again (I generated my own threshold based on my top 20 cutoff), and tried two different things. First, I reduced environmental contributions by 20%, and second I created an entirely new population with a baseline 20% reduction in environmental contributions. Both returned 2 schizophrenic people in the overall population of 2000 - a 90% reduction!
My model suggested a 50% schizophrenia decrease with just a 10% reduction in environmental factors, and a 100% decrease with a 30% reduction in environmental factors. In other words, you don't even have to be perfect (or particularly good!) at mitigating environmental factors to have a huge impact. So, even if genetic factors are responsible for the majority of the effect, if environmental factors are easier to control they might deserve significant focus for even heavily genetic-driven diseases.
CAVEAT: Of course this is based on Scott's oversimplified model.
Too bad environmental factors are randomness and/or other factors we don't know.
I agree that environmental factors aren't easy - especially the part where we have to identify causal relationships - I'm just saying that they're still an option, even with conditions that are heavily genetically determined. For example, say that 25% of the "environmental" variance isn't environmental at all, but copy number variants (i.e. still genetics). That still leaves us with the other 75% to play with, and if we only make a 5-10% reduction in environmental contributions - because it's hard to do - we should still see more than a 5-10% reduction in overall incidence, which is encouraging!
I try to steer clear of solutions that require us to approach the asymptote of 100% implementation before we start seeing an effect. I prefer solutions that start showing efficacy below 20% implementation.
"murderous crimes against humanity"
You don't have to kill anyone to do coercive eugenics; you can just sterilize them :-P
Though I am now intrigued by what would happen in the hypothetical world where a version of Project Provention (https://en.wikipedia.org/wiki/Project_Prevention) had been set up to pay people to get sterilized not on the basis of them being drug addicts, as in our timeline, but on the basis of them being schizophrenic. There is plausibly no population of candidates for that sort of intervention that would be more resistant to it.
I think the implication that Torrey is lying is uncharitable. Maybe he just doesn't know these decidedly unintuitive mathematical results, and Psychiatry Research will issue a correction (or even if this too central to the paper, a retraction).
The math is wrong. For the causal paths, you need the square root of the variance, so you need sqrt(.80) = 0.89, and sqrt(.20) = 0.45. With heritability 80% and environment 20%, you get relative strength of 2 to 1, not 4 to 1. Variances are deceptive.
I don't think this will change your simulation results that much though.
Thanks, I was wondering about exactly this point.
Given the nonlinearity of the normal distribution, does it make sense to define relative importance in terms of ratios of correlations? Assume that everything is normally distributed and additive, and that you get statsophrenia when you're at 99th percentile (2.33 SD) additive risk. With mean environmental risk factors, you need to be at +2.6-SD genetic risk, or +5.2-SD environmental risk with mean genetic risk. That's 99.5th percentile vs. 99.99999th percentile.
I'm not sure how to reason about this. Does it make sense to say that genetics is only twice as important when we can make statements like "There are 50,000 times as many people who get statsophrenia due to genetics despite average environmental risk factors as there are who get it due to environment despite average genetic risk factors?"
There's several decades of debate about how to measure relative importance. This debate is mostly about regression analysis, but these ACE-model results are equivalent to regression with uncorrelated predictors (no GE-correlation).
https://scholar.google.dk/scholar?hl=es&as_sdt=0%2C5&q=relative+importance+regression+analysis
In the I/O psychology literature, though, they are generally concerned with ratio of correlations because strength of the correlation has a linear relationship with the expected return of using it for selection (hiring) purposes.
>"because strength of the correlation has a linear relationship with the expected return of using it for selection (hiring) purposes."<
Can you expand upon this a bit, if you get a chance? It seems to me that ought not be the case — but perhaps I misunderstand "expected return" in this context (in terms of winnowing candidates? in terms of quality of work?), or else something else in my mental model is dumb and wrong.
(e.g.: taking Brandon's example, but pretending "statsophrenia" is actually very desirable, is knowing a candidate's genetic risk factor only twice as valuable as knowing their environmental risk...?)
Hunter and Schmidt discusses it in many places. See e.g. this 1980s paper:
https://home.ubalt.edu/NTYGMITC/645/articles/Hunter%20and%20Hunter%20%201984%20Validity%20and%20Utility.pdf
See section "Computation of Utility". The equation for computing the gains of utility are multiplied by the correlation of job performance with the predictor (IQ or something else). Thus, they scale linearly. You could also show this with a simulation of hiring with predictors of different validities and how this affects the difference in job performance between the hired and unhired groups.
Would you mind pointing to some material that explains why? Or if you feel like it, mapping out your first sentence a bit more? I have zero instinct for this type of stats, but love to get better at it...
You need to use the path coefficient in SEM. Read something like this. https://jslefche.github.io/sem_book/latent-variable-modeling.html
Thanks heaps!
Good post, but quick note: 80% heritability means genes are *twice* as important as environments, not four times as important. When predicting, you want to use unstandardized terms and sqrt(0.8) = 0.8944, sqrt(0.2) = 0.4472.
I just thought of a thought experiment for Aftab's point: can someone who knows genetics better than me comment?
Consider drowning. Right now I would predict that drowning is largely environmental (obviously genetic predispositions towards risk-taking that could include risky behaviors around water will affect it).
However, what if there were a mutation that gave people gills? If this mutation became widespread, it would significantly increase the genetic component to drowning risk across the population. To what degree is it correct to say that, in this hypothetical world, an individual drowning is largely caused by their lack of the "gill" gene(s)? How does that differ to the actual world we live in when someone drowns?
This is a limitation of the whole concept of dividing variance into "genetic" and "environmental" factors. It's a *description* of the current statistical distribution; it is *not a model* of the underlying mechanisms. In particular, at different times and within different populations, you'll have a potentially different genetic vs environmental split.
In your example, gills currently account for 0 variance of drowning, because nobody has gills. If in the future, some people develop gills, then the genetics for gills will explain some of the variance in drownings. And so there will be a corresponding increase in the genetic effect on drownings.
In general, heredity estimates are only applicable to the populations from which they are sampled, and to some extent to other populations which do not differ much in relevant ways. Heredity samples in different populations, at different times, or under different technological conditions can be very different. They should not be taken as timeless, universal truths.
For example, in a wealthy country where almost all people, even poor people, are well-nourished enough to attain their genetic potential height, we will find that height is very strongly heritable. However, if we look at a global sample, where a large percentage of the population does not get the nutrition necessary for this, we will find that height is significantly less heritable.
In the United States, Mandarin verbal ability is largely environmental, because the vast majority of speakers learned it by being raised either in a Mandarin-speaking country or in a Mandarin-speaking household in the US. But in China, it's largely genetic, especially if we confine our study to regions where most people speak Mandarin at home.
Another way to reframe the situation intuitively is that the odds a randomly selected person has schizophrenia are 1%. If having a twin with schizophrenia increases your odds to between 15-50%, it seems like having a schizophrenia gene increases your odds 15-50x of developing schizophrenia vs. the base rate. That's a very strong genetic connection!
This is a great explanation of the subject, very clear and concise.
> E. Fuller Torrey recently published a journal article trying to cast doubt on the commonly-accepted claim that schizophrenia is mostly genetic.
> People really hate the finding that most diseases are substantially (often primarily) genetic. There’s a whole toolbox that people in denial about this use to sow doubt..
I feel like these two lines are very bravery-debate-ish. "The evil anti-science people are using clever tactics to spread misinformation. But we're the brave defenders of science who can see through their ways!"
Also schizophrenia is a disease which is constantly getting filtered out by selective pressure and replenished with mutational load. Killing off people who have it will reduce the amount of schizophrenia genes in the next generation a lot less than the numbers of people who have it because those people would likely not have had many children to begin with, and there's always more mutations in the next generation upping the overall amount.
They are from PSY 5137 – Introduction to Behavioral Genetics – Fall 2020 - by Matt McGue, PhD
Regents Professor, University of Minnesota, Department of Psychology
https://sites.google.com/umn.edu/behavioralgenetics-mcgue/home-topics
Hope this helps!
Truly the worst post I've ever seen from you, Scott. Uncreative self-soothing that misunderstands the existing literature on the genetics of schizophrenia—and, as others have pointed out, bad math!
Except you have just proven the point of the paper: even amongst those with highest genetic risk, a tiny minority develops schizophrenia. Hence the "80% genetic" angle is not just factually-true-but-kinda-unhelpful, it is actively misleading and used by people to discourage and/or disregard research into environmental (i.e. social, political, toxicological) causative factors.
This is emblematic of a wider problem in the field: heritability has no intuitive correlate when it comes to the individual patient on a biological patient. The missing heritability problem taught us it does not even have any correlate on the level of DNA, at least at current levels of data availability, and I would risk predicting that this will not meaningfully change with the whole genome sequencing effort people are getting hyped up about today (i.e. the 1 million genomes the EU is trying to scrape together).
Likely, we will have another round of articles lamenting the fact that the expenses have not been worth it. This is an obvious grift by the genetics community to try and stay relevant, when the other main promise they made couldn't be fulfilled: genetics based drug discovery has been a disaster and so the industry funding is drying up, I guess.
I do not understand why people have such enthusiasm for genetic causes of chronic disease - it means we are factually powerless to prevent it. Is this just residual belief in eugenics? If you are actively seeking out a Gattaca-style future of polgenic scores and genetically engineered humans, please be aware of the enormity of the dice you are rolling: more likely than not, you will be amongst the discarded. It takes a lot of delusion to believe to come out on top, kinda like a modern, more depressing gods-chosen-people belief.
I can't speak for Scott, but a lot of this seems uncharitable.
"I do not understand why people have such enthusiasm for genetic causes of chronic disease - it means we are factually powerless to prevent it"
Firstly, if we actually *are* powerless to prevent it, that's still information worth having; it means we don't need to throw resources away seeking a cure that cannot exist. But secondly, it is unlikely to actually mean that we powerless to prevent it in any case. Scott himself made the point well in this old post: https://slatestarcodex.com/2014/09/10/society-is-fixed-biology-is-mutable/ - just because a problem is biological / genetic, doesn't mean we can't do anything about it. It just means that the best treatment is likely to be a medical rather than a social intervention. And you can try to triangulate his true position from this Socratic dialogue, in case you haven't already read it: https://www.astralcodexten.com/p/galton-ehrlich-buck
"Is this just residual belief in eugenics?"
I think you may need to be clearer what you mean by "belief in eugenics". Belief as in "empirically, a eugenics program could be expected to be effective at its stated aim" is different from "morally, implementing a eugenics program would be the right thing to do" ... and that's before we even get started on the vast range of interventions, from coercive to wholly voluntary, implemented by governments or left up to individual prospective parents, that get rolled together under the label of eugenics.
I get the impression, from what I've read of Scott over the years, that he is about as opposed to Nazi-style mass-murder-based eugenics and mid twentieth century style coercive-sterilization-based eugenics as anyone else, but is optimistic about the prospects of non-coercive reprogenetics which leaves the decision whether to opt for, e.g. embryo selection with polygenic screening, in the hands of individual prospective parents. He has also in the past expressed some favorable comments about Project Prevention, a charity which offers to pay drug addicts a financial premium to get sterilized, with a view to minimizing the number of children who have to grow up with drug-addicted parents, though I'm not sure if he still stands by that today. I'm not sure to what degree Project Prevention even counts as eugenics - sure, to the degree that drug addition is genetic, and to the degree that the charity is effective at its stated aims, it will have eugenic effects, but from what I've read, they're not so much concerned about trying to reduce the frequency of genes for addictive behavior as they are concerned about trying to minimize the suffering of children of drug-addicted parents by preventing those children from coming into existence.
"more likely than not, you will be amongst the discarded"
Again, this presupposes that Scott favors coercive measures, or that if we allow voluntary measures, this somehow sets us on an inevitable course towards coercive measures. This is a non-zero risk (though in the modern era I am highly skeptical that there is much of a constituency for reinstituting the old-fashioned coercive measures), but it has to be weighed against the suffering caused by genetic conditions that we could have used voluntary biotech to prevent, but didn't because we were scared of the slippery slope. We don't get to weigh only one side of that trade-off.
Again, even the large heritability estimates do not mean that this is a "genetic disease", but rather that a subgroup of the entire population is disproportionately susceptible to the eventual, and more proximally causal, environmental factors. This is to me the main point the paper made, and it still stands, regardless of anything Scott wrote unfortunately (looking forward to the follow-up, because so far this was disheartening).
Re this is important information: we have had industry and public research speding for medical interventions against schizophrenia out the backside and nothing to show for it, except for a few drugs that are nor disease-modifying in any way, poorly tolerated and rarely lead to long term success (PMID: 35653111). So exactly how is this helping? And while in principle I would be sympathetic to the not-wasting-money angle, this is not a monogenetic disease like cystic fibrosis where you can shut down alternative etiology research once an for all. Rather, there is now a substantial bias towards environmental defeatism based on a misunderstanding of the consequences of "high heritability" and hence this line of research is likely not adequately funded.
Re eugenics/compulsion:
I do not suggest Scott is favouring compulsory eugenics. If I recall correctly, in Gattaca, there was no governmental coercion either, rather genetic discrimination was made illegal and it was mostly companies enforcing the dystopia via genetic esting of employees. People do not exist in a vacuum, and what becomes de rigeur in a society becomes to some extend compulsory by the uneven playing field it creates for those who refuse to engage. For instance, I'd rather not wear expensive clothing, but society is forcing me to buy suits and dress shirts by not allowing me to do my job in the 10-year old sweatpants and t-shirts I normally wear. This is a sort of coercion by circumstance, and the get-paid-to-sterilize "charity" is dangerously skirting the line in this direction. Economic pressures exists and can result in compulsion worse than what the most opressive governements would dream of doing (case in point: prostitution). The minute genetic optimisation becomes available to the general public, it will seriously tilt the playing field for the next generation, resulting in a new upper-class of "selected" kids and a new underclass of "their-parents-weren't-able-to-afford-it-s". How is this desireable, again?
"How is this desireable, again?"
How is that any different from any other advantage not everyone in the world can afford? Should we ban good food, clean water, medicine, education until everyone can get it?
We're going to have an upper-class either way, I wouldn't mind a saner, smarter one.
This sounds like Dr. Aftab's point; I'll write another post addressing it.
I guess this one is bunk then? https://www.independentsciencenews.org/health/the-great-dna-data-deficit/
It's hard to know what you can believe about science reporting these days without becoming a specialist yourself...
Obesity is also 80% genetic, yet we know from epidemiology that obesity increased substantially for environmental reasons.
This suggests that there are strong gene-environment interactions. But the "X% genetic" model tacitly assumes ZERO gene-environment interactions. Maybe it's not a great model?
(A similar story to obesity happened with the Flynn effect: IQ is like 80% genetic, but IQ scores increased substantially over the second half of the 20th century for what are surely environmental reasons.)
Good discussion. It illustrates how people get very confused when they don't work through the numbers but simply make judgements based on the qualitative sound of words like "most." They forget that there are multiple variables at play, and "most" may apply only to some of them.
Worth noting that exactly analogous considerations apply to IQ and some personality traits.
I'm surprised this statistical fallacy required a post to debunk. You can swap out schizophrenia for any other rare condition; imagine someone saying "IQ is heritable? Then why is it that when one twin is >145IQ, the other twin is >145IQ "only" 33% of the time?"
Smart analysis. However, when people say heritability accounts for 80% of the variation in a binary trait, they mean actual cases, not 80% of the variance of a latent variable which causes that trait above a theshold, right? How do they figure this? Through some kind of log odds ratio on the variable corresponding to identical genome?
Or a likelihood ratio?
You learn something new every day. Thank you, Scott!
Sapolsky's lectures are great for developing more intuition here: https://www.youtube.com/watch?v=OareDiaR0hg
It's important to remember that heritability studies are only as powerful as the range of environments available to researchers--which are always very narrow, relative to the range of conceivable environments. The fact that WWII-era German culture was wildly different from post-war Germany is more than enough (for me) to explain the lack of reduced rates of schizophrenia, even with high inter-culture heritability scores. Which people become schizophrenic might change drastically in a deeply religious society vs a far-future technological society vs an isolationist commune--even if _within_ each of those societies schizophrenia is 80% heritable.
My best summary of my intuition here is that there's no such thing as genetic variance that's pure and independent of environmental variance (and vice versa). Heritability measures are _always_ conditioned on a given environmental range.
"Is it wrong to give twins random environmental scores? Don’t twin pairs grow up in very similar environments? Yes"
Is it wrong? The book Innate by Kevin Mitchell changed the way I think about individual development. It looks like most of your ‘environment’ is how your brain happens to wire itself up during fetal development. That’s why identical twins can be so different, not to mention siblings in general.
You can model these questions using a liability threshold model.* For example, see the methods section of this article: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8510582/. This model is very similar to the one you've used, but it allows you to define covariation between family members. For example, you can take into account that a child and parent share ~50% of their genome. You can also take into account shared environments.
Using this model to address the first question:
- The twin of a diagnosed schizophrenic has an ~36.5% risk of developing the disease.
- The twins can share up to ~43% of their environmental factors before that risk surpasses 50%.
To answer the second question:
- A child of two parents without schizophrenia has ~0.85% chance of developing the disease.
- This number drops to ~0.76% if you assume that the child shares 50% of their environmental factors with each of their parents.
If you want to play around with these models yourself, I uploaded my code here: https://github.com/tgiardina/Some-Unintuitive-Properties-Of-Polygenic-Disorders
* This is a variation on the model we use at Orchid to estimate the disease risk reduction using embryo screening (https://portal.orchidhealth.com/risk-calculator).
> Making everyone mate is beyond the scope of this discount-rate simulation
If you want to do it in the future:
1. Create a new column filled with random numbers.
2. Sort by that column.
3. Take the top half of the population and put them into a new column, probably on a separate sheet.
4. Take the bottom half and put them in their own column next to the top half on the new sheet.
You now have n/2 random marriages. You can take the average value of each couple, regress it to the mean, and have a set of n/2 predictions of child genetic risk.
Actual children would vary around that predicted mean. You can add a normally distributed term† to make that happen. (For example, if you want to generate multiple children from one couple!) For random mating, I believe the variance of this term should be equal to population variance in the trait (?). But there's a good chance I'm confusing the questions of "how much regression to the mean do we do based on our knowledge of the child's ancestry?" -- ultimately a question about the mean of the children's distribution -- and "how much residual variation is left among the children of two known parents?", a question about its variance.
† I get the sense from your post that your generated numbers are centiles rather than deviations. This will interfere with inheritance models.
I didn't spot this comment until posting my own comment separately, but I ran such a simulation (just using 1000 genes rather than calculating what all the necessary covariances are). Another way to do this would be to give each parent 2 normally distributed haplotypes and give the child one from each parent, and set each person's overall genetic value to the sum of their haplotypes (recombination can be ignored since it's only one generation).
This whole post seems like a strong argument that Awais Aftab is correct that "mostly genetic", or even "80% genetic", is a misleading phrase. Even you got slightly confused about exactly what it meant when running your simulation!
I frequently see people on the right say "since we know from twin studies that X is highly heretitable, that means environmental interventions are necessarily useless". Which is the exact same mistake you're rebutting here; as you just demonstrated, it's entirely possible for something to be "80% genetic" in this sense and also 100% dependent on environmental factors. (Though not necessarily ones we can change, they could be totally random.)
In colloquial terms, schizophrenia in your toy model is almost 100% dependent on genes *and* 100% dependent on environment; it's nearly impossible to get it without both. Framing it as one or the other, and mostly one, *is* misleading. (Something you've written on before: https://slatestarcodex.com/2013/06/25/nature-is-not-a-slate-its-a-series-of-levers/)
That's not to say we shouldn't do it, I'm not sure there's a better way to describe the situation in precise statistical terms. Just, maybe we need big flashing disclaimers every time reminding people of the above? Or maybe we need a new jargon word than "hereditability" that doesn't give people this impression?
> as you just demonstrated, it's entirely possible for something to be "80% genetic" in this sense and also 100% dependent on environmental factors.
No, that isn't possible and therefore can't have been demonstrated.
The problem is that you're shifting between claims about particular fixed people (and different ones in different claims!) and claims about the pool of all people.
For the very large majority of people in Scott's simulation, it is theoretically impossible for them to develop schizophrenia no matter how terrible their environment is. (Because you need a composite risk score of 0.929, and the environment can contribute at most 0.200 of that, more than 90% of people -- 0.729 / 0.8 = 0.91125 -- have zero risk of becoming schizophrenic.)
I hope you'll agree that, for that 91% of people, whether they develop schizophrenia is not 100% dependent on environmental factors?
Looking at it from the other direction, in Scott's model 1 in 50 people is, by definition, born with a genetic risk of 0.98 or worse. About one in four of these people will develop schizophrenia. Did they do that because of their environment? In a sense; the model doesn't allow for schizophrenia to occur without an environmental contribution. But whenever you look into one of these people, it is overwhelmingly likely that the environment that pushed them over the edge will have been a perfectly normal somewhat crappy environment, the kind that most people with alarmingly high genetic schizophrenia risk shrug off. What conclusion would you draw?
I do think a real criticism of the high end risk estimates for psychiatric disorders is that heritability is not just what proportion of the disease is caused by genes, but has an essential clause of how much it causes disease in a given environment.
Twin studies typically are not twins who were relocated to dire poverty or traumatic circumstances: most ended up in fairly middle to upper class circumstances as twin adoptees. When environment and nutrients are right and more standardized, then it makes sense that genes will play a relatively larger role.
The concordance rate (in Figure 2) from the 1970 paper seems higher than 33% and closer maybe to 38% or just somewhat under 40% and I don’t think it accounts for shared environmental effects.
Is concordance the probability that both twins have a phenotype given that either one of them does or is it the probability that both twins have a phenotype given that a specific twin does? I think it is the latter (or at least the 1970 paper you linked uses it that way), but the former makes a lot more sense.
Also, if we assume that schizophrenia genetic and environmental risk are normal and additive, we can derive the following formula for the concordance of twin pairs (using the paper’s definition).
$$p^{-1}\int_{-\infty}^{\Phi^{-1}(p)} \Phi\left(\frac{\Phi^{-1}(p) - h^2x}{\sqrt{1 - (h^2)^2}}\right)\varphi(x)dx$$
Where $p$ is the population prevalence and $\varphi$, $\Phi$, and $\Phi^{-1}$ denote the standard normal PDF, CDF, and inverse CDF respectively. It is not very pretty.
Using a heritability of 0.8 and a population prevalence of 1%, we get a concordance rate of about 38% which is what the paper says (and I checked this in an R simulation and it is true or I have made at least two mistakes).
Also, for the probability that both have it given that either have it, or what I think makes more sense for concordance, we can derive the formula
$$\frac{\int_{-\infty}^{\Phi^{-1}(p)} \Phi\left(\frac{\Phi^{-1}(p) - h^2x}{\sqrt{1 - (h^2)^2}}\right)\varphi(x)dx}{2p- \int_{-\infty}^{\Phi^{-1}(p)} \Phi\left(\frac{\Phi^{-1}(p) - h^2x}{\sqrt{1 - (h^2)^2}}\right)\varphi(x)dx}$$
Which gives a probability of about 23% for the heritability and prevalence specified in the post.
You can look at either formula at e.g. the website quicklatex or in whatever latex processor you have.
Also, the second argument is bad because the fact that Schizophrenia has low incidence means it’s immediately clear from e.g. the breeder’s equation, that culling Schizophrenics before reproduction would not change the frequency of Schizophrenia very quickly regardless of the heritability and I do not understand how the author did not see this.
Children born of older fathers are more likely to be schizophrenic (presumably because the fathers' sperm has accumulated more mutations as time has passed and something like 50% of genes are expressed in the brain). Do you know how much more likely? If the answer is a lot, i.e., if a large proportion of schizophrenics were sired by older fathers, then there is an additional reason why the Nazis' policy had little effect on the next generation, or at least one can tell a plausible story about that. The proportion of such offspring was little changed before and after WW2.
As less than two thirds of the german male cohorts born 1920-25 survived WW2, postwar children in germany indeed were sired by elder fathers on average.
The fact that people have more than one parent does make the case significantly stronger against the Germany thing. I ran a simulation in which a population of 5000 with 1000 genes each were tested for some 100% genetic trait, and the top 1% for the trait were eliminated, then had them produce children. 28 of the children (so about 0.6%) were still above the original threshold. Repeating the experiment a few times gave similar results (32 and 30 schizophrenic children the next two times) (I didn't bother working out exactly what the threshold should be on average for the top percentile, and just used the same threshold that caught 1% of the first sample every time, but that doesn't seem to have been a huge issue). If I eliminate the top 1%, run 2 generation and test the grandchildren, the rate returns even closer to the original rate, around 0.9%. If this effect is combined with the fact that real schizophrenia is only 80% heritable not 100%, it seems quite likely that the effect would be unmeasurably small.
Maybe "genetic" should mean two concurrently related things: first, that it is heritable. Second, that it is probabilistic. The first meaning that its presence in the gene pool is a certitude. The second meaning that its expression is conditional. Conditional on what? On a number of intricate factors and processes which are still genetically mediated. We know nothing about these intricate processes and factors hence our confusion. And it makes sense that a phenomenon produced by intricate processes and factors we don't understand would confound our understanding of its phenotypical expression.
"P(A|B) = [P(A)*P(B|A)]/P(B), all the rest is commentary."
That's all you need to explain the raised but still low prevalence in the twins of those affected by a rare condition.
I could be wrong here, but schizophrenia starts in the late teens/early 20s for men (20s for women usually), and there is a prodrome phase of depression and isolation, etc. before that, too.
So ofc the incidence of schizophrenia isn't going to change much by elimination of schizophrenics in a population - they aren't the ones doing much reproducing of those genes in the first place.
The gene naming objection does not survive even cursory thinking. You can tell that Huntington's disease is inherited from the biological parents without having any concept of what genes or DNA are. All you need is correlations. And these days, nobody disputes that height is largely genetic, despite us having very little understanding of the genes involves or their role.
Scott, what happens to this data if you expand "schizophrenia" into "thought disorders and bipolar processes?"
It seems to me that inter-rater reliability is part of the problem. In my forensic work, I have seen many of the same patients over and over, and who have also been seen by other forensic psychologist (its a rural state so the 5 of us who actually do this see the same repeat offenders a lot). Often times our Dx line up perfectly, but frequently one of us will Dx the patient with one thing, and the other six months later with something else. And both Dx are "close enough" to capture the disordered thought process that results in their being diverted away from the justice system and into the state hospital system. In fact, we work together so well, that we often consult with each other to reconcile the Dx variance before it goes to court.
What is the probability of say, for example an adult Dx with Major Depressive Disorder, recurrent, severe with psychotic features having a child who is later Dx with Schizophrenia?
Yep, took the Bloody Code and it's brethren half a millenium
I don't know whether I am reiterating a point made already on this blog or in the comments found here, but gene variants that increase your risk of being diagnosed with schizophrenia often also increase your risk of being unusually creative. Increased creativity means that relatives of people diagnosed with schizophrenia are more likely to reproduce than the norm, thus balancing out the greatly reduced reproductive rate seen in people diagnosed with schizophrenia. This likely largely explains why genes that increase the risk of being diagnosed with schizophrenia are nevertheless found in all human populations. A world without the genes that can together lead to the development of schizophrenia or increased creativity would be a boring world. And yes, I say this as a person with schizoaffective disorder and a professional background as both an award-winning advertising creative and a bioinformatician. Hope this comment is helpful.