I’ve been pondering some of the points Robert Plomin has made in Blueprint, and I find myself thinking about some things that he doesn’t say, but should have. All this in the context of a book that is generally sensible.
Imagine a certain kind of personality variation: some people are aggressive, some not. The aggressive guys tend to prevail over the peaceful, but get in trouble from fighting each other when common: so you get an equilibrium, where both have the same average fitness. Hawk-dove, a la George Price and Maynard Smith. You might like or dislike hawks or doves, but you couldn’t say that either kind of personality was a disease ( A Darwinian disease, the only reasonably definition of disease). Now if those traits were influenced by many genes, there would be continuous variation, but that can be a stable state too, an ESS. Maybe some people would go too far, be too far out on the curve of aggressiveness or way too passive – say four sigma out – they might well have lower fitness, and maybe you could call that a disease – but it’s really just the far edge of a strategy.
Another case: tertiary syphilis, which used to be a common cause of mental illness, with whole wings of mental hospitals dedicated to it. Clearly a Darwinian disease.
Again: you find that having more mutational load, more deleterious mutations, increased your chance of schizophrenia, or autism, or low IQ: that strongly suggests schizophrenia, autism , and low IQ are not the far edge of some strategy. Note: people talking about shamans and schiz: you’re probably wrong. Same for autism – not a strategy.
It is not so easy to for us to determine just how deleterious a mutation is by looking at the sequence, but natural selection gives a strong hint: if they’re bad, they’re rare. And,on the whole, if they’re rare, they’re bad.
Plomin notes that “there is an excess of rare mutations in individuals with schizophrenia, autism and intellectual disability”, and also says that “individuals of extremely high intelligence have fewer of these rare mutations”, suggesting that “rare mutations are not good for you.”
But we knew that before we looked: theory, population genetics, says that deleterious variants will be rare. Selection happens. Now maybe Plomin is simplifying things for the average reader, but I get the nagging feeling that he doesn’t know much about evolutionary game theory, mutational load, mutation-selection balance, truncation selection. I don’t think that most behavioral geneticists ever took any courses covering such things – or am I wrong? Most started as psychologists, or so I’m told.
There are interesting results that make sense if (and only if) you’re thinking in terms of mutational load: intelligence is extremely polygenic, and if you have fewer bad mutations that influence IQ (true on average for people with high IQ) – you have fewer total bad mutations. Lower mutational load, which should result in better health and increased lifespan – and apparently does.
He also says “that the abnormal is normal, meaning that there are no qualitative disorders, just quantitative dimensions. The many DNA differences that are associated with what we call a disorder affect people throughout the distribution.” I don’t think that’s the clearest way of looking at things: better to say that we all have some deleterious mutations, while some people have more than average. But there really is a qualitative difference between what works and what doesn’t.
West Hunter 
Do GWAS IQ studies assume a factor based on number of rare alleles?
Do you think that the rate of autism among people with IQs > 160 is higher than that among people with IQs between 100 and 130? If yes, what’s the evolutionary reason? Also, let’s say that we took an embryo that would normally grow up to be a male with IQ 100, but then we eliminate enough of its rare mutations so that the embryo grows up to be a man with an IQ of, say, 167. Is it likely that we have increased the chances that this embryo will turn into someone who has autism?
There’s definitely something weird about autism, and I’d also be curious about Greg’s thoughts on it.
Studies seem to consistently find a positive genetic correlation between autism and intelligence, but, that said, the ones I’ve seen calculate it using common variation (SNPs). That’s odd for a few reasons, one being that intelligence has non-positive genetic correlations with nearly every disorder. Another is that severe cases of autism are associated with reduced intelligence.
My speculation: maybe common variation controls a trade-off between intelligence and autism, while rare variants are harmful to both (reducing intelligence and increasing autism risk). Here’s a relevant study https://www.ncbi.nlm.nih.gov/pubmed/28504703 , and, unless I’m misreading, it seems consistent with what they’re saying.
Anyway, if something like that is the case, it would suggest that correcting rare variants would increase intelligence and lower autism risk, but if you were to act only on common variation (such as using a SNP-based polygenic score for embryo selection), increasing intelligence would increase autism risk.
Intelligence correlates with education which correlates with being an older parent. Older dads have a higher rate of denovo mutations, which can cause autism.
Have the studies controlled for parental age?
Could the difference in intelligence between the two parents be a factor? Has that ever been tested?
I kind-of wonder if this is something like mental retardation. We have natural variation on a more-or-less normal distribution, and sometimes you’re dumb as a brick because you rolled all ones on your intelligence score. But there are also rare bad things that happen genetically (Downs) or environmentally (hit on the head with a hammer), and those don’t really fit well on the normal curve.
It sure seems like there’s natural variation along something like a normal curve for social skills, ability to infer thoughts/emotional states of other people, ability to tune out distracting stimuli, etc. And then there are also extreme cases way out in the tails where you never learn to speak or something. I wonder if we get these mixed up–the folks who rolled all ones on their charisma score and the folks who have some nasty mutation that messes up the development of their mind-reading module.
I’ve suspected that low-functioning autism is almost entirely a product of mutational load, but I can’t explain high-functioning autism. People do assortatively mate for “quantitative autism traits,” which some people say are related to systematizing behaviors and nonverbal intelligence.
Autism patients with rare indels are more likely to have these indels in genes expressed during brain development or synaptogenesis (https://www.sciencedirect.com/science/article/pii/S0896627311004399). These are typically gain-of-function.
I have wondered if selection for more neural connections has rendered humans at higher risk for boiling over the pot, so to speak. This could happen with a new mutation, or by sexual selection, or by your mother taking valproate while she’s pregnant with you.
It would be interesting to know how tightly autism risk and brain size, or autism risk and IQ correlate across populations.
Is autism defined sufficiently precisely for correlations with anything having any value?
I wonder if in some cases very bright people are responding to rewards of the weird modern world, being more capable of adjusting to novelty (if that Kanazawa theory is right, I don’t know if it is), and therefore looking weird to normal people, e.g. Asperger-like behavior.
Another angle–I’ve heard of cases where people had some brain damage that lead them to having some odd changes that were useful–a sudden ability to play music, or a facility with math. Not common so don’t go hitting yourself in the head with a hammer. I forget if the cause was some kind of dis-inhibition of some other brain system or some compensation or what. Maybe mutations do that sometimes.
It’s a bit like temporal lobe epilepsy–you wouldn’t want it and in most cases it doesn’t seem to help people, but possibly Dostoevsky had it, and he made great contributions. I can’t recall if someone suggested Philip K Dick had it as well, but he fits the bill as far as I know.
There’s this fun paper, “FRONTOTEMPORAL AND DOPAMINERGIC CONTROL OF IDEA GENERATION AND CREATIVE DRIVE” that looks at disease of parts of the brain that produce different creative/intellectual behaviors, so maybe mutations affecting particular parts of the brain would work in line with the model proposed in that paper.
Here’s the link to “FRONTOTEMPORAL AND DOPAMINERGIC CONTROL OF IDEA GENERATION AND CREATIVE DRIVE”: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2571074/
Readers see also:
Features and Bugs, by JayMan
Questions questions questions. Quotes by both Robert Plomin and Kevin Mitchell in this article in Greg’s favorite newspaper exploring what causes the large variation in human intelligence.
Kevin Mitchell countered Plomin regarding the virtually insignificant differences that additive genes have in IQ by saying “we should look not for the genetic dynamics that build intelligence but those that erode it.”
Additive genes have a big role in IQ. Kevin Mitchell goes from saying things that are pretty sensible ( no evidence for significant inheritance of acquired characteristics) to others that are utterly loony (” very hard to select on highly polygenic traits”)
Dumb title “If Smart is the Norm, Stupidity Gets More Interesting.” Normal is the norm. I fear I will age twenty years and go senile before science makes any sense out of why there is such a large variation in human intelligence. I want progress, these half assed philosophers keep on blathering speculation that isn’t backed up by anything and meanwhile the years and decades just keep rolling right along.
Once we find out the genetic reasons why there is a huge variation in genetic intelligence then things will get interesting really fast. Screw the bullshit artists getting their wishful dreams of human equality messed up by scientific evidence, that will be mildly entertaining but it won’t make any real difference. One country somewhere will use that genetic information to make the next generation markedly smarter, that is when things get interesting. Too bad I’ll be dead. If I could just see the look on some of the dumbshits faces that not only is evolution real, they are next, I could die happy.
Actually, the last century or so has been a huge exercise in making smart the norm in environmental stuff–essentially nobody in a first-world country has had their brain development stunted by malnutrition, we do our best to vaccinate for childhood stuff that might mess up brain developments, we add iodine and other supplements to food so we don’t have kids with screwed-up brain development because of malnutrition, we stopped using lead paint and leaded gasoline, etc.
We just don’t think of this as being anything weird or special, because we’re used to it.
Do you really think people with an IQ of 100 are smart? I don’t. Was a small percentage of the population prevented from having sub normal IQ’s because of modern medicine and diet? Yes, but it was a very small percentage of the overall population. Naysayers who wanted badly to believe that IQ isn’t highly genetic loved to point to the Flynn Effect. https://en.wikipedia.org/wiki/Flynn_effect. The Flynn effect was in effect when the education level of the general population moved considerably from illiterate up to a high school level. Of course this would happen, the IQ test is a written test, if you can’t read well you are going to do awful on it. But now that everybody has a high school education the Flynn effect has stopped and even reversed itself. If smart was the new normal than folks with an IQ of !00 with a good work ethic would go to great schools and move in to the best professions, but they very rarely do.
The average person isn’t very smart. Half of people are dumber than that.
Average world IQ is about 90.
Most likely people in the US who are referred to by others as “smart” probably average above 110 in IQ. Likely quite a bit above 110.
I think in a modern first-world country, most people are probably getting close to the intellectual potential of their genes. But in a modern third-world country like Haiti or Bangladesh, or most countries a couple hundred years ago, I suspect a lot of people are/were not getting very close to their potential. We know malnutrition, disease, and specific dietary deficiencies like iodine depress IQ, and we know a fair number of kids in the world don’t get enough food for normal development, spend their whole lives with a bunch of parasitic infections, have several serious bouts of malaria in their childhood, etc.
How smart is an IQ of 100? Not smart compared to me or the people I work with, but probably very smart compared to a Russian serf in 1600 or a slave on a sugar plantation in 1700.
Let me give you less theory and more what is going to happen. We live on this big stage, fools come and go, beliefs mutate because they are inherently flawed, they are after all wishes. Time rolls along and slowly, ever so slowly the greater truth gets spit out of a big chewing machine called science.
A new sense has been found, Intelligence. It just came about in the last 50,000 years. Last one was the eyeball and that one was way back. Oh those first eyeballed critters kicked ass, the trilobites and their kind dominated just like we are. But they were blind too, like we are, to the big picture, We are going places with this new sense organ and life won’t ever be the same. If it isn’t already. What are we on the fourth revolution? First modern like human ingenuity just starting 50.000 years ago, then the agricultural revolution 12,000 years ago. then the industrial revolution kick starting in 1850, and next engineered greater human intelligence.
Hah! Don’t believe in it, do believe in it. Think it’s good, think it’s bad. It’s gonna happen, and you don’t matter.
Why can’t it be an interaction between # deleterious variants and evolutionary strategy? Where one is on the autism spectrum, provided one isn’t too far out, does seem like a strategic tradeoff. But maybe you only get full-blown autism if you are near the edge of that continuum and you have too many deleterious variants. Meanwhile, maybe being on the other end of the spectrum reduces the impact of those variants, but makes you susceptible to some other variants (causing e.g. schizophrenia). All of these things are still basically rare, so one can imagine evolution allowing these general strategic tradeoffs along a continuum, with the penalty of sometimes having too many bad variants causing disease being more than offset by the flexibility of the general strategy.
“But there really is a qualitative difference between what works and what doesn’t.”
Is this not still based to some extent on quantitative factors, since reproductive fitness is (almost) always relative? In a population of 100 individuals, if the 20 healthiest males (on average carrying fewer deleterious mutations than the other 30) get killed by a pack of lions while out hunting, the shaman’s chance of having a couple of baby shamans has just gone up. Nothing has changed about him qualitatively, but since the distribution of deleterious mutations has been skewed, the definition of ‘what works’ has changed. Maybe I’m just splitting hairs. Interesting post, looking forward to the podcast.
I guess judges cannot be expected to understand that within each population genetics sets the mean IQ and thus academic abilities:
http://dailycaller.com/2018/07/22/new-mexico-graduation-ruling/
A man who inherits a null copy of FOXP3 from his mother has one extra deleterious variant— mere noise in the context of the quantitative variation in the number of deleterious variants people have genome wide— but not having regulatory T cells is a qualitative difference in immune systems. A lethal one. There are thousands of examples like this genome wide.
You use the word ‘deleterious’. I’m not sure it means what you think it means.
I think it means “causing harm or damage,” what do you think it means?
“The aggressive guys tend to prevail over the peaceful, but get in trouble from fighting each other when common”: I suppose that’s why everyman cheers one gangster kills another.
And cheer even more when we get most of our aggressive guys who want to go kick some ass into the military and later (when they’re older and less hotheaded) into the police force, so they’re mostly protecting the rest of us instead of preying on the rest of us.
Let’s see now: Ashkenazi Jews have high rates of schizophrenia according to their own studies (30 to 40% above normal populations?). As well as other mental conditions due to their self imposed victimization ideology. They also have high rates of other genetic problems including Tay Sachs. How does Plomin explain this since they have fairly high IQs? Their group strategy ….does it protect them? If not would many of them have died out? Since they have extreme assorted mating for the most part that increase these problems.
I get the feeling you won’t understand the answers to your questions.
One unfortunate answer (its not the only one) is that high IQ, like low IQ is a bit dysgenic. Both ends correspond with higher rates of mental illnesses and disorders. It seems to me that the extremes of variation usually do.
Left-handedness seems to be a strategy (it’s an advantage when few other people are left-handed), but it also correlates with things like low birth weight, like it’s an aberrant phenotype that’s more likely when things aren’t working quite as well. It wouldn’t surprise me if it also correlated with mutational load.
Lefties are more likely to have mental disorders and mood disorders.
http://journals.sagepub.com/doi/pdf/10.1177/2158244013503166
Oops – revise that. Mental disorders only. Mood disorders at about the same rate as the population.
Lefties have a big advantage in face to face fighting, which was important for wife getting.
Maybe a generation or two ago, the current incarnation of noodle-armed soy creatures is laughable.
There’s some evidence that % of population left-handed is associated with interpersonal violence. E.g. Yanomami rates of lefthandedness are a whopping 22%. https://www.nature.com/news/2004/041206/full/news041206-6.html
A little mania or depression is a good thing in many human situations – a little schizophrenia never is. Similarly, personality disorders and anxiety disorders are useful when at least somewhat under the control of the will, but destructive to the tribe and the individual when not.
I have heard arguments that a little autism – HFA or Aspie-ness – can be advantages, with engineering types identified as examples. I consider this possible, but also note that it gets very difficult to define and measure what we are talking about at that point. We may be calling anecdotes data, and fitting data to the theories.
It probably doesn’t help in trying to figure this out that the definition of “Autism” has changed and broadened greatly in the last thirty years. People defined as “mildly autistic” today would probably just been though of as being a little odd back in the 50’s and 60’s – maybe not even that. Autism certainly exists, but it has been expanded in the modern drive to pathologize everything, especially male behavior that women do not find particularly attractive (see also ADHD).
Are there any good things associated with schizophrenia–say, in family members who aren’t afflicted?
In practice, people diagnosed with schizophrenia have low fertility, less than half normal, particularly depressed in men.
Do I think that there is some compensating effect in relatives, as many silly people have suggested?
No, I don’t. I Think it’s just mutational pressure. Brains are complex, lots of genes involved, lots of ways to go wrong. the excess of rare variants in schizophrenics is a strong hint.
Be interesting to see if homosexuals had higher rates of rare variants. Although of course nobody will look.
I think I have seen that the relatives of schizophrenics have lower IQs than average.
Does higher mutational load correlate with schizophrenia? That’s what I’d expect, and it would explain relatives having depressed IQ. You’d then also expect schizophrenics to average shorter and less intelligent and less healthy than other people, even before onset of symptoms.
Some Darwinian diseases might boost your fitness on a one-time basis, just as maladaptive iterated prisoner dilemma strategies can work if you only play one game.
A compulsive gambler might actually win the lottery. An obsessive-compulsive might actually escape a plague. A schizophrenic might actually convince some people he’s the Messiah.
I don’t know of any genetically-based diseases that spread themselves this way (life is prisoners dilemma with a lot of iterations, and you lose more than you win). But you don’t have to read much history to find examples of people (usually male) who enjoyed massive Darwinian success…while being completely crazy.
Maybe there’s an “insanity” niche that genes can fill (if not in humans, than in animals). What you’d need is unusual behavior that has a small chance of a big payoff. Something like the racehorse paradox.
(Contrary to what the author says, this doesn’t work in poker, which is a series of hands played with a couple of people, not a single big race)
The behavior only works if it’s rare, and if the game is “winner takes all”. There’s still a 80% chance that the unpredictable horse will lose to a given normal horse. But if you only care about the winner of the race, the odds favor the freak.
Does breeding work this way, though? It seems to me that it doesn’t.
Most people breed, and breeding is not a winner-take-all race. Quite the contrary. If breeding frequently has winners in any generation (i.e., those who reproduce at or above the average rate of the society around them) and not losers (i.e., people who can’t reproduce at the same rate as those around them), then the “unpredictable” breeder – the one who is, in your example, below average most of the time – will also most likely belong to the group who will lose out in the end.
It’s easy to imagine a “swinging for the fences” strategy paying off, and in fact, that’s what you see in males vs females–higher variance for males in most stuff, which corresponds to the much wider possible variance in male fitness.
But I can’t imagine schizophrenia being an example of this working out–it might be an example of the high-variance strategy misfiring, but for every crazy person who’s treated as a holy man, there will be hundreds who get abandoned or killed by the rest of the tribe for being madmen, or who die in one of the zillion easy ways to die that can befall a crazy/insensible hunter-gatherer or resident in a small farming/fishing village. Or who just can’t find a mate because even if the neighbors don’t kill the harmless crazy guy who lives in the woods near the village, they’re not marrying their daughter off to him.
The inconsistent race horse is “anti-fragile.”
Even most of the evolutionary anyhropologists I know who study the topic don’t know a whole lot about the quantitative theory that really underlies it all. I doubt behavioral geneticists (who are mostly psychologists) are any better. (Not that anthropologists are generally all that impressive.) There’s some irony there because I think the whole concept of heritability (narrow sense) comes from Fisher, who showed that it predicts the response to selection in simple cases. Ie the origin of heritability, which behavioral geneticists care so much about, lies in evolutionary theory. I think. (Correct me if I’m getting this wrong.)
Henry Markram of the blue brain project pushes the idea that autism is caused by learning too fast. The idea is that certain behaviours “burn in” before they should, which leads to the typical social ineptitude and motor disfunction. It also explains the high IQ variety of autism, where kids often learn to speak and read extremely early.
Somehow this fool Markram got the European Union to fund his Human Brain Project to the tune of 1.3 billion. Less than two years into it the project was recognized to be mismanaged and Markram was asked to step down. As per wiki.
What might explain speed-reading?
That seems pretty implausible.
“Now if those traits were influenced by many genes, there would be continuous variation” – nice prediction; The so called “warrior gene” is influenced by at least 3 different genes all residing in the X chromosome – and there are several continious variations; it’s extreme form in europeans is in 0.5% of the population, its less extreme -but still leading to very agressive and vengefull behaviour is spread among 20% of europopulation and its dove’s variant frequency is 80%. There are also 2 unstudied (or unreported) variants with unknown effect and frequencies. However, in some other populations the ratio of doves/hawks is nearly 50/50.
Now, about intelligence, it is trickier.
-rs2251499 (increasing IQ) is spread among only 23% of europeans
-rs13010010 (increasing IQ) has a 35% frequency
-rs9320913, rs2490272, rs11138902 (all increasing IQ) all have frequency of about 50%.
Remarkably, rs41352752 (t or t/t) has a frequency of 97% among europeance and decreases IQ; which means, that the IQ benefiting allele has a frequency of only 3% – quite rare, is not it?
About gene variants, predisposing towars asperger’s or autistic disorder – basically none of them have any verified relation to IQ influencing genes; they are just different genes, placed soomplace else in the genome. Most of them. Now there is something really interesting about asperger’s/autistic alleles – they are actually quite frequent as distribution, some of them reaching 97% in european populations; wich makes the rest – that 3% into genes, protecting from autism and asperger’s; there are a lot of these – all low frequency, 3-10%, all protecting against autism or asperger’s. Make what you wish from this.
Historically they would have been selected against because they lack ‘social brain’ functions, but with complex societies and the division of labour, former diseases might become eugenic: for example in Silicon Valley.
Isn’t the so-called “warrior gene” a loss-of-function mutation that causes less Monoamine oxidase A enzyme? We should be able to predict its effect by whether it knocks out MAO-A or not.
https://www.snpedia.com/index.php/Rs909525(G;G); https://www.snpedia.com/index.php/Rs909525
https://www.snpedia.com/index.php/Rs6323
More like it modulates MAO A enzyme production depending on the particular alleles we’ve inherited.
It horrifies me for multiple reasons – its psychological expression is different depending on carrier’s gender; how much of our personality traits are linked to the MAO genes (far beyond warrior/none warrior traits); it is transmitted via X chromosome as are most of personality traits; it creates a loop between mother and son (or father and daughter) -mothers are hardworking and focused (thanks to the warrior versions) but also often abusive and violent towards their progeny; the divorce rates are higher than average and single mom families happen more often; which is (socially) the worst thing that could happen to a son that carries this combination because when the stressed mother becomes abusive towards her sons they respond with highly increased likelihood for antisocial and criminal behavior as adults; Perhaps a must-have gene to get in hunter-gatherer society since the carriers are multiple times more likely to engage in bloody family/group vendettas (all the way and all out) when challenged. For a contemporary PC society it would look like the sons have become victims to the wrongdoings of their fathers and their lack of social values – divorcing or getting estranged from their hardworking wifes etc. – but it is not. Not really. It is something else that often creates a typical loop of single mother/sociopathic male progeny. Daughters would often not inherit it from their mothers since they carry 2 X chromosomes and would get out of the loop. But the fathers carrying it would pass it to their daughters – and not to their sons, who would also get out of the loop. I wonder how much this mechanism of inheritance has to do with rise and fall of different royal dynasties in the past and their kingdoms respectively.
Also – could this be the “assabyah” gene?
Can you cite where you found the reference that explains the frequency of the SNP rs41352752 (T/T) in the European population
Glad to hear someone telling the truth about autism. It’s a handicap, not some gift. It is true that some lucky (more lightly-affected) autistic individuals have accomplished great things, but it’s in spite of it, not because of it.
Re “Darwinian disease,” the hawk/dove equilibrium in humans and T. pallidum are both products of natural selection, like all disease. And though the former involves only one species while syphilis involves two, host behavioral factors are such that syphilus involves related equilibria like risk-prone and risk-averse.
“, like all disease.” silicosis?
Well, Greg… silicosis probably serves some sort of fitness function. Most likely, people whose bodies mounted a robust healing response to lung damage in a time with no treatments of any kind for ferocious URIs lived longer than those whose bodies did little and therefore had more surviving offspring. Said healing response would eventually kill them, but probably after their kids were of reproductive age.
I think Pamela Anderson has that disease, although in her case science doesn’t seem to be straining itself for a cure.
I read about the French settlers in Quebec–the boldest ones who moved far out to get land married young and had big families, but boldness is also risky. This would be a general result for pioneers. Even among animals you can find differences in boldness/shyness or we could say risk tolerance. A study in Chicago found that some coyotes were much less afraid of cars and people, and these animals found more food but were also more likely to get hit by a car.
In Quebec, everybody had to marry young, girls till age 16 and boys till 20, by royal decree. French-Canadian founder population was only about 2,600, all of them bold enough to settle in that wilderness.
The Filles du Roi
http://www.cbc.ca/2017/canadathestoryofus/most-french-canadians-are-descended-from-these-800-women-1.4029699
I recall that there is a professor in the UC California system (Berkeley, I believe) who studied the whole “shamans and schiz” hypothesis. His conclusions were (1) that primitive people all over the world had no problem identifying schizophrenics as completely crazy (sorry, Thomas Szasz) (2) they definitely were not serving any kind of preeminent role like that of shaman (what moron came up with this idea?) and that (3) seemingly low rates of schizophrenia in such societies are merely a reflection of the fact that primitive people swiftly grow irritated with schizophrenics and quietly murder them in short order.
That last brutal conclusion will not surprise anyone who has dealt with unmedicated schizophrenics for any length of time.