Quote note (#228)

Steve Hsu on the genetics of intelligence:

… the largest effect size [from a single allele] researchers have been able to detect thus far is less than a single point of IQ. Larger effect sizes would have been much easier to detect, but have not been seen. […] This means that there must be at least thousands of IQ alleles to account for the actual variation seen in the general population. A more sophisticated analysis (with large error bars) yields an estimate of perhaps 10,000 in total.*

Each genetic variant slightly increases or decreases cognitive ability. Because it is determined by many small additive effects, cognitive ability is normally distributed, following the familiar bell-shaped curve, with more people in the middle than in the tails. A person with more than the average number of positive (IQ-increasing) variants will be above average in ability. The number of positive alleles above the population average required to raise the trait value by a standard deviation — that is, 15 points — is proportional to the square root of the number of variants, or about 100. In a nutshell, 100 or so additional positive variants could raise IQ by 15 points. […] Given that there are many thousands of potential positive variants, the implication is clear: If a human being could be engineered to have the positive version of each causal variant, they might exhibit cognitive ability which is roughly 100 standard deviations above average. This corresponds to more than 1,000 IQ points. […] It is not at all clear that IQ scores have any meaning in this range. However, we can be confident that, whatever it means, ability of this kind would far exceed the maximum ability among the approximately 100 billion total individuals who have ever lived. …

* Hsu, S.D.H. On the genetic architecture of intelligence and other quantitative traits. Preprint arXiv:1408.3421 (2014).

March 9, 2016admin 41 Comments »
FILED UNDER :Discriminations

TAGGED WITH : , , , ,

41 Responses to this entry

  • Quote note (#228) | Neoreactive Says:

    […] By admin […]

    Posted on March 9th, 2016 at 3:03 pm Reply | Quote
  • Uriel Alexis Says:

    so, an orgy at Mensa could breed interesting humans?

    [Reply]

    Posted on March 9th, 2016 at 3:29 pm Reply | Quote
  • Erebus Says:

    Here Bostrom talks about how tremendous cognitive enhancement might be achieved in the near future, with existing and relatively mature technologies, and what gains we might plausibly be able to expect. Fascinating stuff. It’s only a matter of time — and not very much time, I’d reckon — before this sort of experiment is put to the test. It almost makes me wish I went into biology full-time. (Then I realize that, frustratingly, there’s no such thing as deep understanding in biology… not yet, anyhow. Maybe it’ll come along some time after we boost posthuman IQs into the stratosphere.)

    [Reply]

    Ahote Reply:

    >there’s no such thing as deep understanding in biology
    Yeah, probably even more shallow than Marxist understanding of economics, which makes transhumanism, what, the bolshevism of the 21st century?
    “We are going to seize the means of reproduction and centrally plan the genome!”
    “We are going to make a new man!”
    So I guess that “the curious task of genetics is to demonstrate to men how little they really know about what they imagine they can design.”

    [Reply]

    Erebus Reply:

    I disagree. This is because we know how to make things happen, even if we don’t have any understanding of the underlying biological processes. Lifting weights makes muscles grow, right? You might find this surprising, but nobody understands the biological process which underlies exercise-induced muscle growth. Lots of theories have been floated; a few years ago, it was thought that a variant of an exercise-induced protein called PGC-1α might be ultimately responsible, but this turned out to be false. There’s just too much going on, there are hundreds if not thousands of factors, and most of them are poorly understood. Trying to build a flowchart which describes the process would drive any man crazy — and would ultimately achieve nothing, as there are doubtless lots of myokines that haven’t even been identified yet…

    But you don’t need to understand a damn thing to lift weights and build larger muscles. And we know that you only need to knock out one gene (MSTN) to trick the body into growing Herculean muscles. (Though, again, the molecular mechanisms are incompletely understood.) Similarly, we can certainly boost human intelligence, increase lifespan, etc. without a complete understanding of the underlying biological processes. Bostrom’s plan, for instance, is completely plausible.

    Biology can be a very frustrating field for scientists. Cells are maddeningly complex, their inner workings are completely alien and utterly unlike any technology mankind has ever devised, and we understand them only on the most superficial of levels — but that doesn’t mean that we don’t know how to get things done. A transhumanist future is inevitable. I have no doubt that there will be misfires and catastrophes, but there will also be glorious achievements.

    [Reply]

    Ahote Reply:

    >Though, again, the molecular mechanisms are incompletely understood.

    Exactly my point. There’s The Law Of Unintended Consequences, and then there’s the past experiences, the history of trying to meddle with things that we do not understand, also of trying to better the human condition, to right wrongs, to better the world, to make a new man, etc. etc. and always has it led to nothing but disaster and ruin. Why would you believe that it will be any different this time? Also why the need to meddle, why not simply leave things to catallaxy and evolution?

    Erebus Reply:

    Ah, but you missed my point entirely. We simply don’t need to understand the molecular mechanisms involved.

    Quinine has saved countless lives, but nobody can tell you how it works on a molecular level. Metformin works extremely well, although its mechanism of action is vague and only partially understood. Anabolic steroids clearly work, even if nobody has resolved what happens downstream of the androgen receptor. (Interestingly, whether or not there’s a second androgen receptor, and what it might be responsible for, is still open to debate.) The use of these medicinal agents might be associated with unintended consequences, but not solely with disaster and ruin, and they have undoubtedly bettered the human condition. If we don’t need a complete understanding of biology to treat disease, what makes you think that we need a complete understanding of biology to enhance human capabilities? That was a rhetorical question. The answer is: We don’t require that understanding.

    Bostrom, in that video I linked to above, describes a method one might use to manufacture humans with enhanced intelligence. There’s no need to venture down to finer levels of detail or comprehension. What he proposes can be done with existing technologies. That sort of thing therefore should be done, and it will be done.

    Evolution works both ways, besides. H.G. Wells wrote an interesting article about this, titled “Zoological Retrogression,” all the way back in 1891. And it looks like we’re plunging headlong into a dysgenic abyss — are retrogressing as a species. If we want to advance, we will likely need to take matters into our own hands. Besides, transhumanism on a smaller scale is utterly inevitable, as there will always be smart and driven people who want to give themselves and their offspring every advantage.

    Ahote Reply:

    >We don’t require that understanding.

    Ah, but treating patients is not the same thing as creating incredible hulks. It’s very different thing altogether. They are different even sociologically, medicine was never considered revolutionary. Transhumanism, on the other hand, is called The Final Revolution. Isn’t it strange to you, even a little bit, that transhumanists are all leftists (with a few misguided libertarians here and there)?

    Erebus Reply:

    >”Ah, but treating patients is not the same thing as creating incredible hulks. It’s very different thing altogether.”

    Did you get a chance to look at that video? The method Bostrom proposed was basically a sped-up variation of PGD — it’s no more than a modified version of an old technology. Similarly, CRISPR can be used to treat disease — and can also be used to knock-out genes like MSTN. Even small molecule drugs can have profound effects on human capabilities and performance, which is why “doping” is always so much of an issue in sports.
    ….It can’t be a different thing altogether when the tools are the same.

    I do agree that it is different sociologically. There are lots of ethical questions surrounding these things. There are even regulatory hurdles, as the FDA and its European equivalent can only review disease treatments, and therapies that make people better-than-baseline don’t exactly fall into that category. The NIH won’t fund ethically-questionable Transhumanist research, either. (This may have interesting effects on where research is performed and how it is funded, but it won’t stop anything. This is where unintended consequences will really come into play, for if research is driven completely underground, absolutely no ethical restrictions will apply.)

    Transhumanism has nothing to do with the left. It is explicitly anti-egalitarian and elitist — and anybody who doesn’t see that is delusional, painfully stupid, or a sort of entryist.

    Ahote Reply:

    >Transhumanism has nothing to do with the left.

    O RLY? Just look at this horrid thing for example.

    Erebus Reply:

    Yes, that’s a terrible article.

    I’ll just say this: Don’t conflate the “Transhumanist movement” (which largely consists of that idiot Istvan and the total imbeciles who wrote and host that article) with Transhumanism as a concept. The concept is noble — in fact, it could not possibly be more noble — and its realization is inevitable. The movement is indeed horrid and worthless.

    Posted on March 9th, 2016 at 4:48 pm Reply | Quote
  • Carl Says:

    Steve is assuming linearity and massively extrapolating. It is likely that there are diminishing returns, even if every possible variant is artificially selected. The further you get from naturally existing people, the more uncertainty in the prediction. I’m hoping Steve is using rhetoric to build hype, and not just blindly falling into the cognitive trap of assuming linearity.

    [Reply]

    Uriel Alexis Reply:

    well, he certainly can be used for framing a case in favor of endogamy among super-IQ people. variation outside the endogamy that produces high IQs could also be imported when confirmed.

    seriously, put the mensa people to breed.

    [Reply]

    VKR Reply:

    Many of these thousands of variants are likely to have potentially negative side effects. If radically increasing intelligence was straightforward without detrimental consequences for the organism, then nature would have already done it. Evolution tends to converge towards viable compromises.

    [Reply]

    Posted on March 9th, 2016 at 5:18 pm Reply | Quote
  • grey enlightenment Says:

    An IQ above >500 would probably result in neurosis. It would be like putting a person of normal IQ in solitary confinement

    [Reply]

    Uriel Alexis Reply:

    this is something interesting. if such an intelligent person wasn’t arrogant and dominating, or an outright psychopath, she would most likely feel alone and impotent. so much to learn, so many idiots around to say bullshit.

    [Reply]

    VKR Reply:

    People with very high IQs already complain about social isolation. So things are already bad in the 150+ range, let alone 500 or more. Such people would practically be a different species, and feel like a lone human forced to live in a society of chimpanzee – only worse.

    [Reply]

    SVErshov Reply:

    still it does not imply that their life is total torture. most smart kids already know each other and thinking hard how to kill everybody else.

    check our here
    http://www.dota2.com/leaderboards/#europe

    Posted on March 9th, 2016 at 6:31 pm Reply | Quote
  • Ahote Says:

    Meh, HBD is buncha’ hogwash, as is all that IQ stuff (and hereditarianism in general, but especially HBD and IQ nonsense, it just serves to move conversation away from important topics). Humans evolved from earlier apes. Likewise, under right conditions something more intelligent than Ashkenazim and East Asians could evolve from Sub-Saharan Africans. Of course, it also means that something less intelligent than Sub-Saharan Africans could evolve from Ashkenazim and East Asians. Governance i.e. setting up the “right conditions” matters, genes… not so much (the fact proven by current road to extinction traversed upon by Germans and East Asians).

    [Reply]

    neosoliar Reply:

    And this government that matters will never (in many many years) come from populations with bad genes that apparently don’t matter.

    [Reply]

    Ahote Reply:

    What we have now is those with “good genes” running the worst government in history. In fact, high intelligence probably contributed to this badness, just think about it, universities working day and night, inventing new grievances and dysgenic policies, all that intellectual power invested into justification and propagation of Leftism. Obviously memes>genes.

    [Reply]

    Uriel Alexis Reply:

    i have a suspicion that HBD needs a little more HBE to get the full picture of social Darwinism exactly is:

    https://en.wikipedia.org/wiki/Human_behavioral_ecology

    [Reply]

    Posted on March 9th, 2016 at 7:25 pm Reply | Quote
  • slumlord Says:

    Need to read up on the work of Keith Stanovich.

    [Reply]

    Posted on March 9th, 2016 at 8:12 pm Reply | Quote
  • Liberal replacements for natural selection Says:

    […] What have we lost by replacing Darwinian improvement with “progress,” basically a political allegiance test? Among other things, as Outside In notes (quoting HBD writer Steve Hsu) we are missing out on a chance to improve our inner traits, including intelligence: […]

    Posted on March 9th, 2016 at 10:21 pm Reply | Quote
  • spigot Says:

    So I’m not familiar with the arguments in favour of IQ as a useful, meaningful measure of a person’s intelligence and as I understand it it’s a pretty shitty measure with a lot of flaws. I’m willing to be convinced otherwise and if someone could supply the arguments that would be great.

    P.S. Is this supposed to be IQ as in IQ tests or is there some other way of measuring it?

    [Reply]

    michael Reply:

    nice try

    [Reply]

    Posted on March 9th, 2016 at 10:42 pm Reply | Quote
  • SanguineEmpiricist Says:

    Is everyone’s keyboard missing their keys in this thread? Strange responses.

    [Reply]

    nydwracu Reply:

    І stole all theіr “і”s.

    [Reply]

    nydwracu Reply:

    І mean — іt seems that a partіcular letter has gone mіssіng. Іsn’t that іnterestіng? І have no іdea what mіght have caused that. І had nothіng to do wіth іt and am defіnіtely not sappіng the metaphorіcal bodіly fluіds from XS replіes to power an artіfіcіal superіntellіgence desіgned to іnvade the entіre planet іn the name of neocameralіsm and crown іtself kіng. Entіrely іmpossіble.

    [Reply]

    SVErshov Reply:

    because people nowadays typing without physical keyboards on screens of their tablets, so grammar becoming context independent.

    Posted on March 9th, 2016 at 11:58 pm Reply | Quote
  • Neocolonial Says:

    Because linear extrapolation is a thing.

    [Reply]

    Posted on March 10th, 2016 at 12:01 am Reply | Quote
  • michael Says:

    while fun to imagine it seems like you could end up loading an iphone with the program that run big blue and have no room to use it. Alo reading and watching doc on idiot savants and autistic s and the capacities and feats of these people is astounding but it seems like they are finding its pretty simple wiring issues when they do the advanced mri type testing.

    [Reply]

    Posted on March 10th, 2016 at 2:53 am Reply | Quote
  • TJEL Says:

    Okay, so it’s not just me seeing the missing i’s. Has Land set up a filter to troll us?

    [Reply]

    Posted on March 10th, 2016 at 3:54 am Reply | Quote
  • John Morris Says:

    The assumptions here are daft.

    That there is just one dimension here is the big one. Most of those 10K variations almost certainly aren’t just good/bad and are more like adaptation A and B, both conserved in the gene pool because at different times and/or in combination with the others, both are valuable. If one went through and managed to plot each against the one measure of IQ as we measure it now and selected for each based on that, even assuming many weren’t mutually exclusive in their benefit, assuming you didn’t just breed up a super sperg, assuming all that there is little chance you get something desirable.

    The current natural system works because we spew out a bunch of variations and whatever the problem is somebody has a brain that can solve it. Super select too much and we end up with gaps in the skill pool.

    Until we actually understand how this stuff works we probably shouldn’t be breeding up humans by randomly poking at the genetic controls. I’m talking understand as in more than look at thousands of people and spot the wee p-value on IQ tests and more look at a gene and figure out exactly what change all of the variations makes in a brain and the behavioral difference that leads to.

    [Reply]

    Posted on March 10th, 2016 at 5:50 am Reply | Quote
  • TheDividualist Says:

    While IQ is associated with all kinds of good things up to about 150 or so, beyond that it seems further gains have little utility. The issue is that all too often the 190 IQ savants end up just entertaining themselves with riddles instead of curing cancer or taking over the world. A classic example is chess players, from Kasparov to Polgar. They score around 170-190 and they don’t do anything useful, just play. Why? I think because designing new products or conducting scientific research or scheming for power would not be challenging enough and too boring. If 130 IQ mere mortals can do it, it would be as boring for them as for for the 130 IQ guy driving a crane would be.

    You see the same thing at a Mensa meeting. A lot of really high IQ guys just entertain themselves with riddles and otherwise have unremarkable sysadmin jobs.

    Why? I think because they get very easily bored, and apparently the world does not supply them with enough real-world problems that are sufficiently difficult yet solvable. I think if you’d hire Kasparov to solve the Mises-Keynes debate and figure out what to do in a recession he’d come back saying the reasoning is not watertight in either case, so it is not solvable and the empirical evidence is shit so just use your common sense.

    So all that is achieved by making 200+ humans is that they find the world desperately boring and lock themselves into a virtual world of chess-like games and riddles.

    So I think we win nothing by breaking through the 200 IQ barrier. We would win the most by ensuring no human being is under 110.

    This is also why Yudkowsky’s Unfriendly AI threat is hogwash. Is Kasparov (IQ 194) better at grabbing power than Putin (IQ 130) ? Is Terence Tao (225) becoming world dictator any day now? I get it that they cannot replace the charisma or social communication module in them and machine can, but still.

    Not having stupid people around would matter far more than having geniuses around.

    [Reply]

    Uriel Alexis Reply:

    here’s a fine case for a little stupidity.

    [Reply]

    SVErshov Reply:

    that is quite old question, related to what side to choose, stack to your gods and preserve your independence or let phycist to dril the hole in your brain and incert impeller in it. let see if it going to work.

    Epicritus to Meneceu

    “Our will is autonomouse and independent and to it we can atribute praise and disaproval. thus in order to keep our freedom, it would have been better to remain attached to our belief in god rather then been a slaves to the fate of phisicists: the former give us hope of winning benevolence of deities through promises and sucrifice; the later, on the contrary, bring with it invoilable necessity.”

    and still untesolved. (or otherwise, what difference it can make now)

    it is not like on one hand we can put our fate in IA (intelligence amplification) genetic, chimeric ect. and on another hand … what is on another hand? it’ve been already decided for us and in full scale production. we no need another hand anymore, one is enough. our destiny in SA (stupidity amplification). With billions spent for sputidity enhancement, I just cant see how IA can be achieved without heavy dose of idealisation.

    [Reply]

    Blogospheroid Reply:

    Having pretty much tooted the horn for AI in this thread, I think I’ll need to answer this one.

    If you think that the high IQ people and by extension, AI is not that much of a threat, please don’t think solo people, think teams.

    Sergey Brin, Larry Page are not much of a threat on their own. Combine them with Eric Schmidt and you have a world dominating force.

    Mark Zuckerberg on his own may not have been much but mix with Sean Parker and Sheryl Sandberg and you have a world dominating force.

    The AI need not be everything to everyone. It could profitably fit in a team which can try to be everything to everyone. (whatever goes on inside Amazon)

    And if this doesn’t faze you, then think of Satosh Nakamoto. $800 million in net worth and no one even knows (for sure) who they are.

    [Reply]

    Posted on March 10th, 2016 at 10:17 am Reply | Quote
  • Jack Says:

    Ameliorating the exponentially-deteriorating genetic quality of the masses should take precedence over genetically engineering a bunch of geniuses. Counter-dysgenics — successful maintenance of the genetic status-quo against dysgenic trends — comes first, experimental eugenics later. Incentivizing the intelligent to reproduce (naturally or by sperm and ovary donations) and sterilizing the inferior is called for.

    [Reply]

    Posted on March 10th, 2016 at 10:52 am Reply | Quote
  • Blogospheroid Says:

    Guys, Alpha go just beat Lee Sodol. Genetic engineering done even today will not yield results until 14 years. By that time, you’re basically looking at AI being so freaking ahead of humanity that aligning the values of AIs will probably be the only game left in town.

    [Reply]

    Posted on March 10th, 2016 at 5:14 pm Reply | Quote
  • Craken Says:

    If we identified all ~10,000 IQ alleles, and if they represent 100 SDs of potential IQ improvement–then putting 10% of them to work lifts us up to an IQ level surpassing any extant today. This means we could take some care to select only those allelic alterations least likely to produce negative consequences alongside the IQ effects. There is no reason, considering the evolutionary mechanisms operating on small effect alleles, to suppose that Shakespeare/Newton/Mozart represent the maximum achievable in human cognition. How far past them we can travel, given the physical constraints of brain engineering, is another question.

    The fast (and potentially dangerous) way to play this game is genetic engineering enabled by CRISPR-Cas9 or something like it. The slow way is genetically-informed embryo selection (although, theoretically, iterated embryo selection is also a fast route).

    Basic embryo selection could perform wonders in ensuring that those people least adapted to modernity have much more capable children–both IQ and personality are susceptible to alteration. Hsu mentions a 15 IQ point jump in a single generation by this simple method–sufficient to close the venerable white-black IQ gap in America, if only applied to blacks. More aggressive embryo selection could also be applied to the most problematic cases (low IQ, psychopathy, physical issues) by extracting more eggs and creating more embryos,–thereby generating higher selective power. All sorts of expensive social problems could be solved.

    Is there a better candidate than eugenics for being the world’s most impressive medium term upside? Given that Moore’s Law has lapsed into a coma, it may yet be that the superintelligence of the visionary future is us.

    [Reply]

    Posted on March 12th, 2016 at 6:40 am Reply | Quote

Leave a comment