Super Intelligent Humans are Coming
October 19, 2014 3:17 PM Subscribe
If you could twiddle a couple of genes and get someone radically more intelligent, that would have been far more likely to have evolved instead.
Well, the article does say it's more like a few hundred genes. Not to say that researchers will slam-dunk it on the first try...
posted by latkes at 3:34 PM on October 19, 2014 [1 favorite]
Well, the article does say it's more like a few hundred genes. Not to say that researchers will slam-dunk it on the first try...
posted by latkes at 3:34 PM on October 19, 2014 [1 favorite]
10-1 says they end up the highest-testing CPAs in Iowa because achievement has little to do with IQ.
posted by Potomac Avenue at 3:39 PM on October 19, 2014 [28 favorites]
posted by Potomac Avenue at 3:39 PM on October 19, 2014 [28 favorites]
So you're saying a 21-year-old could have the intelligence of a 210-year-old? About time.
posted by michaelh at 3:39 PM on October 19, 2014 [6 favorites]
posted by michaelh at 3:39 PM on October 19, 2014 [6 favorites]
1,000 IQ points
Utter nonsense. IQ points don't work like that.
posted by Sys Rq at 3:40 PM on October 19, 2014 [37 favorites]
Utter nonsense. IQ points don't work like that.
posted by Sys Rq at 3:40 PM on October 19, 2014 [37 favorites]
"...But which candidate would you rather have a beer with?"
posted by FreelanceBureaucrat at 3:41 PM on October 19, 2014 [5 favorites]
posted by FreelanceBureaucrat at 3:41 PM on October 19, 2014 [5 favorites]
If only a small number of genes controlled cognition, then each of the gene variants should have altered IQ by a large chunk—about 15 points of variation between two individuals. But the largest effect size 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.
IANA geneticist, but it seems to me that this conclusion is based on the premise that all of the unexplained variation is genetic - i.e. that you have completely controlled for every possible extraneous factor (cultural, nutritional, environmental etc). On the face of it, that seems fairly unlikely to me.
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.
I do not find this implication at all clear. On what basis to you assume that all of the variations are mutually additive? Especially to such a nebulous concept as g?A variant that makes you better at lateral thinking may well make you less detail oriented. An improvement in empathy may make you less analytical and so on.
Stephen Hsu is Vice-President for Research and Professor of Theoretical Physics at Michigan State University. He is also a scientific advisor to BGI (formerly, Beijing Genomics Institute) and a founder of its Cognitive Genomics Lab.
Ah, OK, so it's an ad? Carry on then.
posted by Jakey at 3:41 PM on October 19, 2014 [41 favorites]
This means that there must be at least thousands of IQ alleles to account for the actual variation seen in the general population.
IANA geneticist, but it seems to me that this conclusion is based on the premise that all of the unexplained variation is genetic - i.e. that you have completely controlled for every possible extraneous factor (cultural, nutritional, environmental etc). On the face of it, that seems fairly unlikely to me.
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.
I do not find this implication at all clear. On what basis to you assume that all of the variations are mutually additive? Especially to such a nebulous concept as g?A variant that makes you better at lateral thinking may well make you less detail oriented. An improvement in empathy may make you less analytical and so on.
Stephen Hsu is Vice-President for Research and Professor of Theoretical Physics at Michigan State University. He is also a scientific advisor to BGI (formerly, Beijing Genomics Institute) and a founder of its Cognitive Genomics Lab.
Ah, OK, so it's an ad? Carry on then.
posted by Jakey at 3:41 PM on October 19, 2014 [41 favorites]
> 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.
...or they might not be mentally functional at all because their processing potential exceeds their bandwidth capacity.
posted by ardgedee at 3:42 PM on October 19, 2014 [11 favorites]
...or they might not be mentally functional at all because their processing potential exceeds their bandwidth capacity.
posted by ardgedee at 3:42 PM on October 19, 2014 [11 favorites]
the airplane was in trouble over a mountain range, and when the president reached into the storage compartment for a parachute, he told the others "it is imperative that i survive this, because i am the leader of the free world."
the world's smartest man strapped one on too, saying that it was not to save himself, but that he held his intellect in trust for all mankind. that left the priest and the hippie.
the priest said "my faith is strong, and in a few seconds, i will be reposing in the bosom of jesus, so take the last parachute and save yourself while you can.
the hippie said "hey man, that's real cool, you being willing to die for me and all, but you don't have to, there's a parachute here for both of us."
the priest said "how could that be? i thought only three parachutes were stowed on board."
the hippie said "the world's smartest man just jumped out of the plane wearing my backpack."
posted by bruce at 3:44 PM on October 19, 2014 [101 favorites]
the world's smartest man strapped one on too, saying that it was not to save himself, but that he held his intellect in trust for all mankind. that left the priest and the hippie.
the priest said "my faith is strong, and in a few seconds, i will be reposing in the bosom of jesus, so take the last parachute and save yourself while you can.
the hippie said "hey man, that's real cool, you being willing to die for me and all, but you don't have to, there's a parachute here for both of us."
the priest said "how could that be? i thought only three parachutes were stowed on board."
the hippie said "the world's smartest man just jumped out of the plane wearing my backpack."
posted by bruce at 3:44 PM on October 19, 2014 [101 favorites]
He is also a scientific advisor to BGI (formerly, Beijing Genomics Institute) and a founder of its Cognitive Genomics Lab.
Good catch - sorry I missed that.
posted by latkes at 3:46 PM on October 19, 2014 [2 favorites]
Good catch - sorry I missed that.
posted by latkes at 3:46 PM on October 19, 2014 [2 favorites]
I'm not sure extrapolation works like that.
posted by madcaptenor at 3:47 PM on October 19, 2014 [4 favorites]
posted by madcaptenor at 3:47 PM on October 19, 2014 [4 favorites]
I chucked at the very first sentence. I dropped Soviet Physics because my courseload was too heavy, what with Ottoman Chemistry, Franco-Prussian Biology and Celtic Statistical Analysis.
posted by George_Spiggott at 3:57 PM on October 19, 2014 [13 favorites]
posted by George_Spiggott at 3:57 PM on October 19, 2014 [13 favorites]
Unfortunately, if you set all the Smarty Switches to ON, there are side effects.
posted by moonmilk at 4:04 PM on October 19, 2014 [9 favorites]
posted by moonmilk at 4:04 PM on October 19, 2014 [9 favorites]
I'm no geneticist but I know that most genes aren't linked to just one trait. If you set all of a person's 10K intelligence-linked genes to maximize cognitive ability you're also bringing along a host of genetic side effects, many of them undesirable or even fatal. I'd bet you could never achieve anything close to a genetically supercharged mind without disrupting systems throughout the body. And if you could, what would their quality of life be? I don't know of many happy, emotionally balanced super-geniuses & quite a few counter-examples.
posted by scalefree at 4:07 PM on October 19, 2014 [1 favorite]
posted by scalefree at 4:07 PM on October 19, 2014 [1 favorite]
Seems uncontroversial to me. One day, genetic engineering will make humans much smarter. I thought this was really obvious.
posted by Edgewise at 4:08 PM on October 19, 2014 [3 favorites]
posted by Edgewise at 4:08 PM on October 19, 2014 [3 favorites]
I can't get my head around the fact that people think "I'm getting good results by thinking about normal distributions in a range of +-3 SD so I'm gonna just go ahead and extrapolate that to +-100 SD." But I guess in order to get my head around that I would have to get my head around the idea that someone thinks it's makes sense to measure "intelligence" on a linear scale in the first place.
posted by escabeche at 4:09 PM on October 19, 2014 [18 favorites]
posted by escabeche at 4:09 PM on October 19, 2014 [18 favorites]
The author is dreaming if he thinks that CRISPR could or would ever be used to make hundreds of genetic changes to a single individual. One might be able to use it over 100 generations, possibly, but you might as well screen for variants the old fashioned way (i.e cross your favorite people together, like regular animal husbandry). Of course, anyone who's tried to breed animals to maintain just a couple of variants (let alone 100) knows that you have to go for inbreeding if you don't want to lose them all with the next cross, which means you'll have to deal with the numerous consequences of consanguineous pairings. Not much good having all the genes for super-intelligence if you're born without an essential organ.
And then, presuming that people commit to this 1000-year endeavor, you've bred out all the deleterious genes and perfected the art of brother-sister mating, then you might finally have somebody with all the "right" variants. However, since we don't know what the variants actually do, or how they might interact with each other, your guarantee of a more intelligent human being is far from assured. Moreover, it's quite likely that, during the execution of this multi-generation experiment, scientists will determine a way to achieve the same effect pharmacologically.
So yes, there is a conceivable pathway by which the sort of human genetic engineering and screening the author is writing about could become a reality, but it requires time, money and the continuous overstepping of some serious ethical boundaries.
posted by kisch mokusch at 4:10 PM on October 19, 2014 [6 favorites]
And then, presuming that people commit to this 1000-year endeavor, you've bred out all the deleterious genes and perfected the art of brother-sister mating, then you might finally have somebody with all the "right" variants. However, since we don't know what the variants actually do, or how they might interact with each other, your guarantee of a more intelligent human being is far from assured. Moreover, it's quite likely that, during the execution of this multi-generation experiment, scientists will determine a way to achieve the same effect pharmacologically.
So yes, there is a conceivable pathway by which the sort of human genetic engineering and screening the author is writing about could become a reality, but it requires time, money and the continuous overstepping of some serious ethical boundaries.
posted by kisch mokusch at 4:10 PM on October 19, 2014 [6 favorites]
This all assumes that g is a thing, and that intelligence is a single quantitative trait, and of course lots of scientists, including many with degrees in biology and psychology rather than physics, think that g is bullshit. This is a very in depth explanation of the many statistical problems with the hypothesis of g.
posted by hydropsyche at 4:13 PM on October 19, 2014 [23 favorites]
posted by hydropsyche at 4:13 PM on October 19, 2014 [23 favorites]
Utter nonsense. IQ points don't work like that.
I have heard the IQ test is less useful for discriminating between people at the extremes even of human intelligence, that is, you can't draw conclusions about the abilities of two people who test 140 and 160, or 40 and 60, with the confidence that you can about people who test at 80 and 100 or 100 and 120. If that is so saying that an entity has an IQ far outside this range seems really dubious. It's basically just a way of re-asserting that you've decided to postulate a superhuman intelligence, but in a way that sounds dishonestly precise and scientific, isn't it?
posted by thelonius at 4:16 PM on October 19, 2014 [1 favorite]
I have heard the IQ test is less useful for discriminating between people at the extremes even of human intelligence, that is, you can't draw conclusions about the abilities of two people who test 140 and 160, or 40 and 60, with the confidence that you can about people who test at 80 and 100 or 100 and 120. If that is so saying that an entity has an IQ far outside this range seems really dubious. It's basically just a way of re-asserting that you've decided to postulate a superhuman intelligence, but in a way that sounds dishonestly precise and scientific, isn't it?
posted by thelonius at 4:16 PM on October 19, 2014 [1 favorite]
Utter nonsense. IQ points don't work like that.
OVER A THOUSAND IQ'S SYS RQ! IMAGINE THAT, IF YOU EVEN CAN!
posted by turbid dahlia at 4:16 PM on October 19, 2014 [25 favorites]
OVER A THOUSAND IQ'S SYS RQ! IMAGINE THAT, IF YOU EVEN CAN!
posted by turbid dahlia at 4:16 PM on October 19, 2014 [25 favorites]
a logarithmic scale for ranking theorists, from 1 to 5 … Einstein was a 0.5!
I would just like to point out that if somebody scores a 0.5, then it's not a 1 to 5 scale. So, numbers are not the author's strong suit is what I'm saying.
posted by tempestuoso at 4:19 PM on October 19, 2014 [52 favorites]
I would just like to point out that if somebody scores a 0.5, then it's not a 1 to 5 scale. So, numbers are not the author's strong suit is what I'm saying.
posted by tempestuoso at 4:19 PM on October 19, 2014 [52 favorites]
Socioeconomic status modifies heritability of IQ in young children
The models suggest that in impoverished families, 60% of the variance in IQ is accounted for by the shared environment, and the contribution of genes is close to zero; in affluent families, the result is almost exactly the reverse.
posted by dephlogisticated at 4:26 PM on October 19, 2014 [16 favorites]
The models suggest that in impoverished families, 60% of the variance in IQ is accounted for by the shared environment, and the contribution of genes is close to zero; in affluent families, the result is almost exactly the reverse.
posted by dephlogisticated at 4:26 PM on October 19, 2014 [16 favorites]
Interest in cultivating objectively, genetically superior human beings has always ended well, historically. And has always involved impeccable science. I don't see any way such a project could possibly go wrong.
posted by edheil at 4:28 PM on October 19, 2014 [8 favorites]
posted by edheil at 4:28 PM on October 19, 2014 [8 favorites]
There is a long discussion of the debate about measuring "g" as a general intelligence factor, including discussion of the article hydropsyche linked above, in this previous metafilter thread.
posted by demonic winged headgear at 4:33 PM on October 19, 2014 [1 favorite]
posted by demonic winged headgear at 4:33 PM on October 19, 2014 [1 favorite]
Everyone in this thread is now dumber for having read the article.
posted by benito.strauss at 4:33 PM on October 19, 2014 [4 favorites]
posted by benito.strauss at 4:33 PM on October 19, 2014 [4 favorites]
And then there's the weird way the author begins the article (emphasis mine):
> Lev Landau, a Nobelist and one of the fathers of a great school of Soviet physics, had a logarithmic scale for ranking theorists, from 1 to 5. [...] My friends in the humanities, or other areas of science like biology, are astonished and disturbed that physicists and mathematicians might think in this essentially hierarchical way.
Somebody who uninformedly assumes that nobody but physicists are obsessed with rank (and then uses a broken ranking system as his example), and lumps biology together with the humanities (as implicitly being the refuges of those of lesser brilliance), is not really advertising himself as one of the go-to authors on biological matters. I'm not particularly interested in his opinions on literature or world affairs either, to be honest.
posted by ardgedee at 4:36 PM on October 19, 2014 [1 favorite]
> Lev Landau, a Nobelist and one of the fathers of a great school of Soviet physics, had a logarithmic scale for ranking theorists, from 1 to 5. [...] My friends in the humanities, or other areas of science like biology, are astonished and disturbed that physicists and mathematicians might think in this essentially hierarchical way.
Somebody who uninformedly assumes that nobody but physicists are obsessed with rank (and then uses a broken ranking system as his example), and lumps biology together with the humanities (as implicitly being the refuges of those of lesser brilliance), is not really advertising himself as one of the go-to authors on biological matters. I'm not particularly interested in his opinions on literature or world affairs either, to be honest.
posted by ardgedee at 4:36 PM on October 19, 2014 [1 favorite]
I'll wager human super-intelligence involves vastly more training than genetic engineering. Although, we could probably achieve massively parallel human thought with brain implants that linked together many human brains, but that still involves considerable training. We might achieve aspects of super-intelligence through non-parallelizing brain implants too, but maybe that's harder.
posted by jeffburdges at 4:47 PM on October 19, 2014 [1 favorite]
posted by jeffburdges at 4:47 PM on October 19, 2014 [1 favorite]
I can readily accept that intelligence is:
a) a useful and meaningful concept that can be tested
b) largely heritable
c) subject to genetic manipulation in the future
What I can't accept or understand is the unstated assumption that higher intelligences for a very small number children who have very wealthy parents is good for anything. Good for humanity, good for the individuals involved.
The world does not lack a surfeit of smarts, we're currently plenty smart enough collectively to do whatever we need to do. What we do lack is empathy and common sense, which I can't see genetically engineered superhumans doing anything good for. Homo superior will probably spend its time writing new code for High Frequency Trading and hedge fund risk algorithms.
posted by wilful at 4:57 PM on October 19, 2014 [23 favorites]
a) a useful and meaningful concept that can be tested
b) largely heritable
c) subject to genetic manipulation in the future
What I can't accept or understand is the unstated assumption that higher intelligences for a very small number children who have very wealthy parents is good for anything. Good for humanity, good for the individuals involved.
The world does not lack a surfeit of smarts, we're currently plenty smart enough collectively to do whatever we need to do. What we do lack is empathy and common sense, which I can't see genetically engineered superhumans doing anything good for. Homo superior will probably spend its time writing new code for High Frequency Trading and hedge fund risk algorithms.
posted by wilful at 4:57 PM on October 19, 2014 [23 favorites]
This is only the second dumbest idea I've heard about how brains work, but it was up against some stiff competition.
posted by mordax at 5:00 PM on October 19, 2014 [4 favorites]
posted by mordax at 5:00 PM on October 19, 2014 [4 favorites]
g is bullshit
I've finally found the magic phrase that causes me to insta-favorite every single time it appears.
posted by RogerB at 5:01 PM on October 19, 2014 [4 favorites]
I've finally found the magic phrase that causes me to insta-favorite every single time it appears.
posted by RogerB at 5:01 PM on October 19, 2014 [4 favorites]
The cynical part of me wonders what the suicide rate among these super-brain-people would be. Because as it is, it always seemed at least to me that the smarter someone was the more likely they were to be depressed, or just generally upset with life and the state of the world in that way.
I think this has been written about at great length already, but yea. I wonder if they'd just create the most suicidal person to ever shortly live?
This article is of course hilariously dumb for a lot of reasons, but i mean to even engage with it at all made me wonder that.
posted by emptythought at 5:02 PM on October 19, 2014 [4 favorites]
I think this has been written about at great length already, but yea. I wonder if they'd just create the most suicidal person to ever shortly live?
This article is of course hilariously dumb for a lot of reasons, but i mean to even engage with it at all made me wonder that.
posted by emptythought at 5:02 PM on October 19, 2014 [4 favorites]
This all assumes that g is a thing, and that intelligence is a single quantitative trait, and of course lots of scientists, including many with degrees in biology and psychology rather than physics, think that g is bullshit. This is a very in depth explanation of the many statistical problems with the hypothesis of g.
Assuming the g is really a single factor that can be discovered is pretty stupid - and in fact the existence of so many genes that seem to affect it would seem to argue against that conclusion. That doesn't mean that IQ-type tests aren't a useful statistical measure of some set of abilities. And that many important types of cognitive performance are (sometimes unexpectedly) correlated is pretty established I think. It's fair enough to talk about "multiple intelligences" but to assume that they are anywhere close to truly independent factors is not well supported by the science.
I don't know of many happy, emotionally balanced super-geniuses & quite a few counter-examples.
I was just thinking actually about how everything I've heard about John von Neumann - mentioned in the article, a true prodigy and polymath for the ages, whose mental calculation abilities blew other manhattan project researchers away - suggests he was a pretty well-adjusted, happy guy. I feel like a lot of the attrition against prodigies has to do with unreasonable expectations. I wonder what von Neumann's parents and teachers did right? Or maybe he just got the non-anxiety genes?
posted by atoxyl at 5:07 PM on October 19, 2014 [1 favorite]
Assuming the g is really a single factor that can be discovered is pretty stupid - and in fact the existence of so many genes that seem to affect it would seem to argue against that conclusion. That doesn't mean that IQ-type tests aren't a useful statistical measure of some set of abilities. And that many important types of cognitive performance are (sometimes unexpectedly) correlated is pretty established I think. It's fair enough to talk about "multiple intelligences" but to assume that they are anywhere close to truly independent factors is not well supported by the science.
I don't know of many happy, emotionally balanced super-geniuses & quite a few counter-examples.
I was just thinking actually about how everything I've heard about John von Neumann - mentioned in the article, a true prodigy and polymath for the ages, whose mental calculation abilities blew other manhattan project researchers away - suggests he was a pretty well-adjusted, happy guy. I feel like a lot of the attrition against prodigies has to do with unreasonable expectations. I wonder what von Neumann's parents and teachers did right? Or maybe he just got the non-anxiety genes?
posted by atoxyl at 5:07 PM on October 19, 2014 [1 favorite]
that might be the stupidest article about intelligence i've read.
posted by empath at 5:08 PM on October 19, 2014 [5 favorites]
posted by empath at 5:08 PM on October 19, 2014 [5 favorites]
The world does not lack a surfeit of smarts, we're currently plenty smart enough collectively to do whatever we need to do. What we do lack is empathy and common sense, which I can't see genetically engineered superhumans doing anything good for.
This is an important point.
posted by clockzero at 5:25 PM on October 19, 2014 [9 favorites]
This is an important point.
posted by clockzero at 5:25 PM on October 19, 2014 [9 favorites]
I... that isn't how logic works. Holy shit this person is terrible at arguing. I don't like this.
*draws a red X through it in Paint*
posted by quiet earth at 5:31 PM on October 19, 2014 [13 favorites]
*draws a red X through it in Paint*
posted by quiet earth at 5:31 PM on October 19, 2014 [13 favorites]
This used to be called eugenics. I think collectively we decided that was a bad idea.
posted by beagle at 5:46 PM on October 19, 2014 [3 favorites]
posted by beagle at 5:46 PM on October 19, 2014 [3 favorites]
This kind of research is going to be a sideshow, pretty much as AI research is a sideshow.
Science doesn't need smarter people: it progresses quite nicely by lots of scientists iterating and incrementing with only the rarest step-change insight (which is as likely as not to be made by two different people at more or less the same time, because it was just ready to be had on account of the iterating and incrementing going on). Heck, a tremendous amount of scientific progress now is being achieved by simple brute force methods -- spend enough money and time, and get your masses of data, or discover the SNP, that you need, with about the reliability of a jackhammer reducing a roadway to rubble.
(AI is a sideshow because we don't need computers to think like people, as we have no lack of people to do that, we need computers to keep thinking like computers, only better.)
posted by MattD at 5:51 PM on October 19, 2014 [4 favorites]
Science doesn't need smarter people: it progresses quite nicely by lots of scientists iterating and incrementing with only the rarest step-change insight (which is as likely as not to be made by two different people at more or less the same time, because it was just ready to be had on account of the iterating and incrementing going on). Heck, a tremendous amount of scientific progress now is being achieved by simple brute force methods -- spend enough money and time, and get your masses of data, or discover the SNP, that you need, with about the reliability of a jackhammer reducing a roadway to rubble.
(AI is a sideshow because we don't need computers to think like people, as we have no lack of people to do that, we need computers to keep thinking like computers, only better.)
posted by MattD at 5:51 PM on October 19, 2014 [4 favorites]
I was so hoping the Flynn Effect would turn out to be named after Joe Flynn.
posted by lagomorphius at 5:55 PM on October 19, 2014
posted by lagomorphius at 5:55 PM on October 19, 2014
beagle:That got me to thinking, so I did a naïve search and what do you know, Stephen Hsu is in the orbit of the Dark Enlightenment people.This used to be called eugenics. I think collectively we decided that was a bad idea.
posted by ob1quixote at 6:00 PM on October 19, 2014 [9 favorites]
And then, presuming that people commit to this 1000-year endeavor, you've bred out all the deleterious genes and perfected the art of brother-sister mating, then you might finally have somebody with all the "right" variants.
Kull wahad!
posted by ROU_Xenophobe at 6:02 PM on October 19, 2014 [8 favorites]
Kull wahad!
posted by ROU_Xenophobe at 6:02 PM on October 19, 2014 [8 favorites]
I do wish it hadn't started with a quote from Brave New World, as there is no genetic engineering or even selective breeding in BNW (or at least I couldn't find it). Everyone comes from the same ovaries until they're all used up. In Brave New World, there are only highly educated normal people (alphas), somewhat less educated normal people (betas), and people who have been harmed in pseudo-utero to reduce their mental capacity (gammas on down).
Which makes it that much creepier. To me, anyhow.
posted by ROU_Xenophobe at 6:07 PM on October 19, 2014
Which makes it that much creepier. To me, anyhow.
posted by ROU_Xenophobe at 6:07 PM on October 19, 2014
Just what the world needs, a smarter banker
posted by Fupped Duck at 6:12 PM on October 19, 2014 [3 favorites]
posted by Fupped Duck at 6:12 PM on October 19, 2014 [3 favorites]
Science doesn't need smarter people: it progresses quite nicely by lots of scientists iterating and incrementing with only the rarest step-change insight (which is as likely as not to be made by two different people at more or less the same time, because it was just ready to be had on account of the iterating and incrementing going on).
Indeed, if anything I think what science "needs" is even more people who are socially intelligent and good at collaboration, because this allows (to abuse an analogy) more distributed problem solving.
posted by en forme de poire at 6:39 PM on October 19, 2014
Indeed, if anything I think what science "needs" is even more people who are socially intelligent and good at collaboration, because this allows (to abuse an analogy) more distributed problem solving.
posted by en forme de poire at 6:39 PM on October 19, 2014
Mostly, though, what science needs is funding.
posted by en forme de poire at 6:41 PM on October 19, 2014 [14 favorites]
posted by en forme de poire at 6:41 PM on October 19, 2014 [14 favorites]
My friends in the humanities, or other areas of science like biology, are astonished and disturbed that physicists and mathematicians (substitute the polymathic von Neumann for Einstein) might think in this essentially hierarchical way. Apparently, differences in ability are not manifested so clearly in those fields.Really? All writing, all art is of equal value in the humanities?
Try doing an impact assessment on the whole of human history. Who has had the greatest impact on the whole of human culture? Hint: it probably wasn't a physicist.
posted by No Robots at 6:42 PM on October 19, 2014 [1 favorite]
If all of the amazing MIT graduates could be found fulfilling jobs in their fields instead of being ushered into consulting gigs that are actually hurting society at large, none of this (notional junk science) mucking around would be necessary.
posted by 1adam12 at 6:44 PM on October 19, 2014 [8 favorites]
posted by 1adam12 at 6:44 PM on October 19, 2014 [8 favorites]
Height is a good analogy to intelligence. It's also controlled by thousands of genes, each of which causes only a small increase or decrease. And being tall, to a moderate extent, also has some advantages in life.
If a person existed who was 100 standard deviations above the mean height of an American male, he would be 14 feet tall. In reality, the tallest man (who was only 9 feet tall -- 13 standard deviations) had serious physical problems because of it, and wasn't an amazing basketball player or ceiling painter or anything. (He died at age 22 of complications.)
posted by miyabo at 6:56 PM on October 19, 2014 [18 favorites]
If a person existed who was 100 standard deviations above the mean height of an American male, he would be 14 feet tall. In reality, the tallest man (who was only 9 feet tall -- 13 standard deviations) had serious physical problems because of it, and wasn't an amazing basketball player or ceiling painter or anything. (He died at age 22 of complications.)
posted by miyabo at 6:56 PM on October 19, 2014 [18 favorites]
Not a single reference to The Best and the Brightest?
posted by IndigoJones at 7:00 PM on October 19, 2014
posted by IndigoJones at 7:00 PM on October 19, 2014
Tell me of the waters of your homeworld, Usul.
posted by scalefree at 7:16 PM on October 19, 2014 [2 favorites]
posted by scalefree at 7:16 PM on October 19, 2014 [2 favorites]
If intelligence/"g" is, roughly speaking, a measure of adaptation to a modern technological world, then by some estimates Abraham Lincoln would score only an 80 or so on our modern tests. Lots of people have covered the problems with this "1,000 IQ" claim before me, but it's interesting to think about what a human would look like who had a 1,000 IQ compared to our average 100 IQ person today, the same way we can compare our modern test-takers to Lincoln.
I would just like to point out that if somebody scores a 0.5, then it's not a 1 to 5 scale. So, numbers are not the author's strong suit is what I'm saying.
Why can't a 1 to 5 scale have intermediate values between integers? The Richter scale is a logarithmic scale from 1 to 9+ and it has them. And in any case, this isn't a comment by the author of the article, but a reference by the author to a comment made by Lev Landau, a Soviet physicist. And the author, for his part, is a theoretical physicist. Do you really think that between you, Lev, and Stephen, that they're the innumerate ones? I mean, I enjoy a good joke at the expense of a science journalist as much as the next guy, but I think you botched this one...
posted by mrbigmuscles at 7:28 PM on October 19, 2014
I would just like to point out that if somebody scores a 0.5, then it's not a 1 to 5 scale. So, numbers are not the author's strong suit is what I'm saying.
Why can't a 1 to 5 scale have intermediate values between integers? The Richter scale is a logarithmic scale from 1 to 9+ and it has them. And in any case, this isn't a comment by the author of the article, but a reference by the author to a comment made by Lev Landau, a Soviet physicist. And the author, for his part, is a theoretical physicist. Do you really think that between you, Lev, and Stephen, that they're the innumerate ones? I mean, I enjoy a good joke at the expense of a science journalist as much as the next guy, but I think you botched this one...
posted by mrbigmuscles at 7:28 PM on October 19, 2014
In Bruce Sterling's Schismatrix (and a few other places), he suggests that genetic engineering for super-intelligence results in super-intelligent people who are also batshit insane. I don't think there's any evidence to suggest that this might happen in real-life - but for some reason it wouldn't surprise me at all if it turned out to be true.
posted by doctor tough love at 7:29 PM on October 19, 2014
posted by doctor tough love at 7:29 PM on October 19, 2014
If a person existed who was 100 standard deviations above the mean height of an American male, he would be 14 feet tall.
Yeah, but what about the person who's 100 standard deviations below the mean height?
And... the limitations of modeling by normal distribution suddenly become clear.
posted by escabeche at 7:29 PM on October 19, 2014 [7 favorites]
Yeah, but what about the person who's 100 standard deviations below the mean height?
And... the limitations of modeling by normal distribution suddenly become clear.
posted by escabeche at 7:29 PM on October 19, 2014 [7 favorites]
Why can't a 1 to 5 scale have intermediate values between integers?
It can. But 0.5 isn't between 1 and 5.
Not that that's really the problem. The problem is thinking of a numerical scale like this as a good idea in the first place. Maybe this is something that physicists actually do, but it's not something I've ever seen mathematicians do.
posted by escabeche at 7:31 PM on October 19, 2014 [1 favorite]
It can. But 0.5 isn't between 1 and 5.
Not that that's really the problem. The problem is thinking of a numerical scale like this as a good idea in the first place. Maybe this is something that physicists actually do, but it's not something I've ever seen mathematicians do.
posted by escabeche at 7:31 PM on October 19, 2014 [1 favorite]
Would they also be 100 standard deviations haughtier than your average MENSA member? Because that prospect runs chills down my spine.
posted by Perko at 7:37 PM on October 19, 2014 [5 favorites]
posted by Perko at 7:37 PM on October 19, 2014 [5 favorites]
It can. But 0.5 isn't between 1 and 5.
Shit, haha well now we all know where I am on the bell curve
posted by mrbigmuscles at 7:39 PM on October 19, 2014
Shit, haha well now we all know where I am on the bell curve
posted by mrbigmuscles at 7:39 PM on October 19, 2014
Super Intelligent Humans are Coming
And some of us are already hear.
*preens*
posted by Atom Eyes at 7:46 PM on October 19, 2014 [1 favorite]
And some of us are already hear.
*preens*
posted by Atom Eyes at 7:46 PM on October 19, 2014 [1 favorite]
This used to be called eugenics. I think collectively we decided that was a bad idea.
Yeah, we did. But it's coming back, and it's more subtle now.
Want to screen your unborn kid to see if they're going to have an abnormality that would doom them to a short life of suffering? You can do that, now, for a lot of things, and abort if you find that that's the case. I think a lot of people would. I would.
Want to screen your unborn kid to see if they're likely to be excruciatingly stupid? That's not a solved problem, at all, but we're making baby steps. It'll probably happen. Would you do it? I don't know that I would, in 2014, but times are a-changin'.
Want to select an egg for IVF that has the most desirable genetic properties possible given your own genes? Or, hey, for that matter, given all the genomes available in the donor bank? Well, fine, that's science fiction right now, but I could imagine ways that that day could come. As long as you're doing something invasive like IVF, why not optimize a bit?
This conversation isn't over; it's just beginning.
posted by gurple at 7:47 PM on October 19, 2014 [1 favorite]
Yeah, we did. But it's coming back, and it's more subtle now.
Want to screen your unborn kid to see if they're going to have an abnormality that would doom them to a short life of suffering? You can do that, now, for a lot of things, and abort if you find that that's the case. I think a lot of people would. I would.
Want to screen your unborn kid to see if they're likely to be excruciatingly stupid? That's not a solved problem, at all, but we're making baby steps. It'll probably happen. Would you do it? I don't know that I would, in 2014, but times are a-changin'.
Want to select an egg for IVF that has the most desirable genetic properties possible given your own genes? Or, hey, for that matter, given all the genomes available in the donor bank? Well, fine, that's science fiction right now, but I could imagine ways that that day could come. As long as you're doing something invasive like IVF, why not optimize a bit?
This conversation isn't over; it's just beginning.
posted by gurple at 7:47 PM on October 19, 2014 [1 favorite]
And some of us are already hear.
*preens*
Not sure if you meant to include a typo for humorous effect, but, preen away, I say!
posted by Perko at 7:54 PM on October 19, 2014 [7 favorites]
*preens*
Not sure if you meant to include a typo for humorous effect, but, preen away, I say!
posted by Perko at 7:54 PM on October 19, 2014 [7 favorites]
Science doesn't need smarter people: it progresses quite nicely by lots of scientists iterating and incrementing with only the rarest step-change insight
I don't know that I've ever seen a terser conflation of the naturalistic and moralistic fallacies.
posted by 7segment at 8:09 PM on October 19, 2014 [1 favorite]
I don't know that I've ever seen a terser conflation of the naturalistic and moralistic fallacies.
posted by 7segment at 8:09 PM on October 19, 2014 [1 favorite]
The problem is thinking of a numerical scale like this as a good idea in the first place. Maybe this is something that physicists actually do, but it's not something I've ever seen mathematicians do.
Nor biologists, particularly when talking about variable traits. That's because there's always a trade-off. Any trait that is universally advantageous will be selected for (and will no longer be variable). Variable traits are often variable for a reason, in that their advantage is context-specific. To delineate a single trait on "intelligence" genes ignores the potential detrimental effects of these variants.
posted by kisch mokusch at 8:19 PM on October 19, 2014 [2 favorites]
Nor biologists, particularly when talking about variable traits. That's because there's always a trade-off. Any trait that is universally advantageous will be selected for (and will no longer be variable). Variable traits are often variable for a reason, in that their advantage is context-specific. To delineate a single trait on "intelligence" genes ignores the potential detrimental effects of these variants.
posted by kisch mokusch at 8:19 PM on October 19, 2014 [2 favorites]
Try doing an impact assessment on the whole of human history. Who has had the greatest impact on the whole of human culture? Hint: it probably wasn't a physicist.
Well, this is provocative. But I agree. It was probably a mathematician. (Gauss, I'd conjecture.)
posted by Mental Wimp at 8:26 PM on October 19, 2014
Well, this is provocative. But I agree. It was probably a mathematician. (Gauss, I'd conjecture.)
posted by Mental Wimp at 8:26 PM on October 19, 2014
It was probably a mathematician. (Gauss, I'd conjecture.)
Google hits for "Gauss": about 18 million.
Google hits for "Dickens": about 48.5 million.
Sorry. Try again.
posted by No Robots at 8:35 PM on October 19, 2014
Google hits for "Gauss": about 18 million.
Google hits for "Dickens": about 48.5 million.
Sorry. Try again.
posted by No Robots at 8:35 PM on October 19, 2014
And then, presuming that people commit to this 1000-year endeavor, you've bred out all the deleterious genes and perfected the art of brother-sister mating, then you might finally have somebody with all the "right" variants.
Sorry, Reverend Mother, but your Kwisatz Haderach is in another castle.
posted by Steely-eyed Missile Man at 8:36 PM on October 19, 2014 [5 favorites]
Sorry, Reverend Mother, but your Kwisatz Haderach is in another castle.
posted by Steely-eyed Missile Man at 8:36 PM on October 19, 2014 [5 favorites]
I feel like I hit 1000 IQs at that precise moment between my 5th and 6th customary Friday night beers, just as the sun is lighting up the clouds from below as it sets outside my window.
I guess 'feel like' are probably the operative words here.
posted by stavrosthewonderchicken at 8:39 PM on October 19, 2014 [6 favorites]
I guess 'feel like' are probably the operative words here.
posted by stavrosthewonderchicken at 8:39 PM on October 19, 2014 [6 favorites]
The author is dreaming if he thinks that CRISPR could or would ever be used to make hundreds of genetic changes to a single individual.
Why do you think that? Drug companies are already using CRISPR systems to disable the genes that make the T-cell receptors that HIV binds to, with some considerable success so far in inducing remission in HIV+ patients — i.e. patients who have their T-cells successfully reprogrammed have undetectable viral loads and don't take antiretroviral medicines. That's just the tip of the iceberg, as far as current applications go.
posted by a lungful of dragon at 8:40 PM on October 19, 2014 [1 favorite]
Why do you think that? Drug companies are already using CRISPR systems to disable the genes that make the T-cell receptors that HIV binds to, with some considerable success so far in inducing remission in HIV+ patients — i.e. patients who have their T-cells successfully reprogrammed have undetectable viral loads and don't take antiretroviral medicines. That's just the tip of the iceberg, as far as current applications go.
posted by a lungful of dragon at 8:40 PM on October 19, 2014 [1 favorite]
It says quite it that the comments quickly divert into ranting about the Alcuberre and EM Drives. And how salt is the perfect entropic chemical.
Honestly, there really should be a law that prohibits physicists and engineers from talking about matters outside their specialties- especially when it comes to biology..
posted by happyroach at 8:40 PM on October 19, 2014 [1 favorite]
Honestly, there really should be a law that prohibits physicists and engineers from talking about matters outside their specialties- especially when it comes to biology..
posted by happyroach at 8:40 PM on October 19, 2014 [1 favorite]
To delineate a single trait on "intelligence" genes ignores the potential detrimental effects of these variants.
That's only a problem as long as society has sufficient qualms about creating people who are off-the-charts in some specific kind of mental capacity but unable to care for themselves in more banal ways.
I'm not advocating for this, you understand, but I do think it's a possible part of the future. Already, in 2014, we have companies that are quite forthright about their desire to hire high-functioning autism-spectrum employees because they can be phenomenally productive workers. What if companies could design such a workforce rather than relying on a natural supply? The obstacles are both scientific and ethical, but I'd say the bigger obstacles are ethical.
posted by gurple at 8:44 PM on October 19, 2014
That's only a problem as long as society has sufficient qualms about creating people who are off-the-charts in some specific kind of mental capacity but unable to care for themselves in more banal ways.
I'm not advocating for this, you understand, but I do think it's a possible part of the future. Already, in 2014, we have companies that are quite forthright about their desire to hire high-functioning autism-spectrum employees because they can be phenomenally productive workers. What if companies could design such a workforce rather than relying on a natural supply? The obstacles are both scientific and ethical, but I'd say the bigger obstacles are ethical.
posted by gurple at 8:44 PM on October 19, 2014
Why do you think that? Drug companies are already using CRISPR systems to disable the genes that make the T-cell receptors that HIV binds to, with some considerable success so far in inducing remission in HIV+ patients — i.e. patients who have their T-cells successfully reprogrammed have undetectable viral loads and don't take antiretroviral medicines. That's just the tip of the iceberg, as far as current applications go.
Well, knocking things out with CRISPR is easier than inserting/changing genes, and you're still doing the genes one at a time. Try to do two at once and you've got a much lower chance of success just based on probability alone not to mention the chances of chromosomal translocations and other effects of creating multiple double-stranded breaks. lt is also much easier to do with a large pool of samples (e.g. T cells) that you can then select easily for (e.g. antibody for surface CCR5 expression). I suppose one could potentially do more than one change if they could maintain the embryonic stem cells in totipotency, but I still can't see how the screening would be done without letting them grow into an organism.
I completely agree on the tip-of-the-iceberg statement. The possibilities of CRISPR are remarkable. My criticism is in it's application in human engineering.
posted by kisch mokusch at 8:57 PM on October 19, 2014 [4 favorites]
Well, knocking things out with CRISPR is easier than inserting/changing genes, and you're still doing the genes one at a time. Try to do two at once and you've got a much lower chance of success just based on probability alone not to mention the chances of chromosomal translocations and other effects of creating multiple double-stranded breaks. lt is also much easier to do with a large pool of samples (e.g. T cells) that you can then select easily for (e.g. antibody for surface CCR5 expression). I suppose one could potentially do more than one change if they could maintain the embryonic stem cells in totipotency, but I still can't see how the screening would be done without letting them grow into an organism.
I completely agree on the tip-of-the-iceberg statement. The possibilities of CRISPR are remarkable. My criticism is in it's application in human engineering.
posted by kisch mokusch at 8:57 PM on October 19, 2014 [4 favorites]
I know some people with very high IQs. They're good at crossword puzzles, but don't seem to be happier or more successful than people with high-normal IQs. Also, there's reason to think that extreme IQs may be associated with genetic disorders. There must be some reason why the world isn't run by mega-Einsteins; if g really were a good predictor for success then there would almost necessarily be a compensating disadvantage that explains why people with mediocre intelligence do so well.
posted by Joe in Australia at 8:59 PM on October 19, 2014
posted by Joe in Australia at 8:59 PM on October 19, 2014
The obstacles are both scientific and ethical, but I'd say the bigger obstacles are ethical.
The ethical obstacles are clearer. The scientific obstacles are completely unknown. It's possible that certain combinations of variants are incompatible with life. That would turn out to be a bigger obstacle, i think.
posted by kisch mokusch at 9:00 PM on October 19, 2014
The ethical obstacles are clearer. The scientific obstacles are completely unknown. It's possible that certain combinations of variants are incompatible with life. That would turn out to be a bigger obstacle, i think.
posted by kisch mokusch at 9:00 PM on October 19, 2014
I feel like I hit 1000 IQs at that precise moment between my 5th and 6th customary Friday night beers, just as the sun is lighting up the clouds from below as it sets outside my window.
It's extra funny because I can't tell if stavros accidentally pluralizes "quotient" or if he subjectively experiences multiple >1000 IQ scores
posted by clockzero at 9:00 PM on October 19, 2014
It's extra funny because I can't tell if stavros accidentally pluralizes "quotient" or if he subjectively experiences multiple >1000 IQ scores
posted by clockzero at 9:00 PM on October 19, 2014
Well, fine, that's science fiction right now, but I could imagine ways that that day could come. As long as you're doing something invasive like IVF, why not optimize a bit?
Even with the controversy of potential "custom babies", people can only screen for variants that they provide. Two blond people, for example, cannot select for a dark-haired baby. And they can forget about screening for smarter babies since the known genetic variants don't provide a strong enough effect to be meaningful. In other words, we're all playing with the same Lego set. You might think that it's possible to put the pieces together better than Mother Nature, but chances are you can't.
Putting this flaw in the premise aside, the article is kind of paradoxical in that it simultaneously imagines an impractical, unlikely future world of super-intelligence while limiting the application of human genetic engineering to the use of known variants. Which is pretty unimaginative for an article that mentions CRISPR.
posted by kisch mokusch at 9:02 PM on October 19, 2014 [1 favorite]
Even with the controversy of potential "custom babies", people can only screen for variants that they provide. Two blond people, for example, cannot select for a dark-haired baby. And they can forget about screening for smarter babies since the known genetic variants don't provide a strong enough effect to be meaningful. In other words, we're all playing with the same Lego set. You might think that it's possible to put the pieces together better than Mother Nature, but chances are you can't.
Putting this flaw in the premise aside, the article is kind of paradoxical in that it simultaneously imagines an impractical, unlikely future world of super-intelligence while limiting the application of human genetic engineering to the use of known variants. Which is pretty unimaginative for an article that mentions CRISPR.
posted by kisch mokusch at 9:02 PM on October 19, 2014 [1 favorite]
Already, in 2014, we have companies that are quite forthright about their desire to hire high-functioning autism-spectrum employees because they can be phenomenally productive workers. What if companies could design such a workforce rather than relying on a natural supply?
Damn, Gibson coming true is bad enough. Please no Vernor Vinge.
posted by Steely-eyed Missile Man at 9:03 PM on October 19, 2014 [1 favorite]
Damn, Gibson coming true is bad enough. Please no Vernor Vinge.
posted by Steely-eyed Missile Man at 9:03 PM on October 19, 2014 [1 favorite]
I didn't even want to go there.
Making people autistic so they will be better, more focused workers is a specific plot point in A Deepness in the Sky, though.
posted by Steely-eyed Missile Man at 9:34 PM on October 19, 2014 [4 favorites]
Making people autistic so they will be better, more focused workers is a specific plot point in A Deepness in the Sky, though.
posted by Steely-eyed Missile Man at 9:34 PM on October 19, 2014 [4 favorites]
doctor tough love> In Bruce Sterling's Schismatrix (and a few other places), he suggests that genetic engineering for super-intelligence results in super-intelligent people who are also batshit insane.
"All we want is a quiet place to work while God eats our brains."
posted by UrineSoakedRube at 10:53 PM on October 19, 2014 [5 favorites]
"All we want is a quiet place to work while God eats our brains."
posted by UrineSoakedRube at 10:53 PM on October 19, 2014 [5 favorites]
My IQ WAS 1000 but then I discovered alcohol.
Beer: The cause of, and solution to, all life's problems.
posted by Eyebrows McGee at 11:03 PM on October 19, 2014 [5 favorites]
Beer: The cause of, and solution to, all life's problems.
posted by Eyebrows McGee at 11:03 PM on October 19, 2014 [5 favorites]
If you could twiddle a couple of genes and get someone radically more intelligent, that would have been far more likely to have evolved instead.
This doesn't make sense to me. There are, for example, dogs. They're smart, but they're not as smart as me. If nature is so efficient and effective at making brains, why aren't dogs as smart as me? It seems possible that evolution has a wide variety of results, some of them better than others. So why do we assume that what we have now must be the best it can be?
We don't have the best eyes we can have. We don't have the best ears we could have. We don't have the best noses we could have. We don't have the best skin, heart, lungs, whatever we could have. Somewhere there's an animal that has a better version. Even within the human population, there are clearly some people whose brains are better than others. Why do you assume we have the best brains we could have? At the very least, we could all have brains as good as the best brains out there now.
posted by obiwanwasabi at 11:39 PM on October 19, 2014 [1 favorite]
This doesn't make sense to me. There are, for example, dogs. They're smart, but they're not as smart as me. If nature is so efficient and effective at making brains, why aren't dogs as smart as me? It seems possible that evolution has a wide variety of results, some of them better than others. So why do we assume that what we have now must be the best it can be?
We don't have the best eyes we can have. We don't have the best ears we could have. We don't have the best noses we could have. We don't have the best skin, heart, lungs, whatever we could have. Somewhere there's an animal that has a better version. Even within the human population, there are clearly some people whose brains are better than others. Why do you assume we have the best brains we could have? At the very least, we could all have brains as good as the best brains out there now.
posted by obiwanwasabi at 11:39 PM on October 19, 2014 [1 favorite]
Wow, I really want to come in here and shout "not all physicists." So that's where that impulse comes from.
posted by nat at 11:50 PM on October 19, 2014 [4 favorites]
posted by nat at 11:50 PM on October 19, 2014 [4 favorites]
Yeah, but what about the person who's 100 standard deviations below the mean height?
And... the limitations of modeling by normal distribution suddenly become clear.
There was an entertaining article in Significance (a magazine produced by the Royal Statistical Society aimed at younger statisticians) which calculated the probability of someone being born with negative height. It concluded that this was so improbable that we would have to wait far longer than the heat death of the universe to observe it. Oh well!
...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.
...the implication is clear: If an article writer doesn't understand statistics, they will think that extrapolating a model of human intelligence will have meaningful results.
Other's have picked this apart, but its worth mentioning that with any statistical model that you can create bizzare results simply by creating unlikely individuals that were never observed in the data.
posted by Cannon Fodder at 12:23 AM on October 20, 2014 [1 favorite]
This is not going to turn out well. Why? Because selection will first be among those with means, and it will be performed by people who want to "be first"; there are going to be lots of sociopaths among that crowd - people who don't think long term. They will be the ones who venture foryth before there is settled law in this arena; thay will want *advantage*. They are going to spread *their* genes, at 100 standard deviations above average intelligence - super-intelligent sociopaths, probably at a higher general percentage (about 1%) that are in the population at present. They will be first; they will multiply first; they will take advantage and ownership, first - of this "brave new world" of super intelligence. This is not going to turn out well.
posted by Vibrissae at 1:19 AM on October 20, 2014 [2 favorites]
posted by Vibrissae at 1:19 AM on October 20, 2014 [2 favorites]
I can't tell if stavros accidentally pluralizes "quotient"
Nah, just riffing on a joke upthread.
Also, I am large, I contain multitudes.
posted by stavrosthewonderchicken at 1:52 AM on October 20, 2014
Nah, just riffing on a joke upthread.
Also, I am large, I contain multitudes.
posted by stavrosthewonderchicken at 1:52 AM on October 20, 2014
Apparently cat scientists have found a way to combine all the many different genes for blackness into a single strand of cat DNA. This will produce a cat 1000 times blacker than the blackest thing there has ever been, so black it will suck in all the light from the entire solar system.
Won't that be something?
posted by Segundus at 2:50 AM on October 20, 2014 [9 favorites]
Won't that be something?
posted by Segundus at 2:50 AM on October 20, 2014 [9 favorites]
Won't that be something?
There's a reason "cat scientists" have many more Google hits than "Dickens" and "Gauss".
posted by effbot at 3:12 AM on October 20, 2014 [8 favorites]
There's a reason "cat scientists" have many more Google hits than "Dickens" and "Gauss".
posted by effbot at 3:12 AM on October 20, 2014 [8 favorites]
Oh we do impact analysis by counting Google hits now? And whose contributions more likely brought us teh googles...
posted by aydeejones at 3:12 AM on October 20, 2014
posted by aydeejones at 3:12 AM on October 20, 2014
Not an immediate response to the comment above, heh. That comment helps actually and is probably bettah.
posted by aydeejones at 3:13 AM on October 20, 2014
posted by aydeejones at 3:13 AM on October 20, 2014
Einstein is discussed a lot in regards to genius, and he comes up in the article right away, as many folks have commented on. But I think Einstein's real amazing ability had to do with his astounding creativity and imagination, the fact that he was able to conceive of explanations for physical reality so outside of the Newtonian physics that had come before. This is really outside of any linear measurement that could be encapsulated by a value like IQ points.
And while of course I don't dispute his incredible intelligence, and he certainly must have had a significant amount of mathematical skill compared to the average person, it's important to remember that often brilliant mathematicians like Emmy Noether and Ernst G. Straus helped him work out the maths. If anything this says that the kind of valorization of "extreme genius"--represented in this article and popping up constantly in Western culture--should probably take a back seat to the idea of the power of talented, motivated humans working together in teams, in cooperation: that's where our strengths lie. I'd like to see more emphasis on that as our path forward.
posted by dubitable at 4:24 AM on October 20, 2014 [11 favorites]
And while of course I don't dispute his incredible intelligence, and he certainly must have had a significant amount of mathematical skill compared to the average person, it's important to remember that often brilliant mathematicians like Emmy Noether and Ernst G. Straus helped him work out the maths. If anything this says that the kind of valorization of "extreme genius"--represented in this article and popping up constantly in Western culture--should probably take a back seat to the idea of the power of talented, motivated humans working together in teams, in cooperation: that's where our strengths lie. I'd like to see more emphasis on that as our path forward.
posted by dubitable at 4:24 AM on October 20, 2014 [11 favorites]
Why don't we all just come out and admit that we hope this guy is wrong? Unless you think that you're cute enough for the ubermenschen to make you a pet.
posted by Halloween Jack at 4:48 AM on October 20, 2014
posted by Halloween Jack at 4:48 AM on October 20, 2014
Anybody who knows anything about statistics, intelligence and genetics knows this guy is wrong. It's not remotely whistling past the graveyard. He leaps to absurd conclusions based on the flim<siest evidence.
posted by empath at 5:22 AM on October 20, 2014
posted by empath at 5:22 AM on October 20, 2014
Oh we do impact analysis by counting Google hits now?
I thought that would appeal to all the super-intelligent physico-empiricist internet warriors. Care to suggest another measure that would include all cultural agents?
And whose contributions more likely brought us teh googles
Dickens gave us Christmas. Look it up. On Google.
posted by No Robots at 5:24 AM on October 20, 2014
I thought that would appeal to all the super-intelligent physico-empiricist internet warriors. Care to suggest another measure that would include all cultural agents?
And whose contributions more likely brought us teh googles
Dickens gave us Christmas. Look it up. On Google.
posted by No Robots at 5:24 AM on October 20, 2014
Well, I for one contributed to creation of human super-intelligence the old fashioned way: by having sex, and then waiting 9 months to meet the little genius.
CRISPR might be faster but sex is more fun.
posted by caution live frogs at 5:47 AM on October 20, 2014
CRISPR might be faster but sex is more fun.
posted by caution live frogs at 5:47 AM on October 20, 2014
This all assumes that g is a thing, and that intelligence is a single quantitative trait
Let me get this straight—you're saying that it's nothing but a "g" thing?
posted by octobersurprise at 5:57 AM on October 20, 2014 [3 favorites]
Let me get this straight—you're saying that it's nothing but a "g" thing?
posted by octobersurprise at 5:57 AM on October 20, 2014 [3 favorites]
1000 IQ points is totally possibly. I got 1000 IQ points. It only took me 10 tests!
posted by srboisvert at 6:14 AM on October 20, 2014 [2 favorites]
posted by srboisvert at 6:14 AM on October 20, 2014 [2 favorites]
Let me guess: we are only 30 years from this incredible breakthrough. Yeah, I thought so.
It's always 30 years.
posted by clvrmnky at 6:16 AM on October 20, 2014
It's always 30 years.
posted by clvrmnky at 6:16 AM on October 20, 2014
It's theoretically possible to have a concrete definition of 1000 IQ. The scale is constructed by imposing a normal distribution on the scores of test-takers. You can do a thought experiment where you imagine giving the test to quadrillions and quintillions of people, and then constructing new tests that distinguish amongst those who max out the existing tests, using the new tests to establish a new normalisation... and then repeating this process with increasing numbers of people until you reach a test that measures IQs 60SD above the mean. By the very definition of the normal distribution, this would require about 10^784 people. Incidentally, there's a good chance that at least one of those people would perfectly answer a 1300 item multiple choice test purely by chance, assuming four alternatives per question, so the tests would have to be very long.
What this means is that -practically - to make an IQ of 1000 meaningful, it'll be necessary first to have excellent knowledge of how genetics and environment predict IQ - which is a bit circular, to be honest. I think we'd have more luck identifying a g-factor in chimpanzees and constructing an inter-species IQ test than in successfully norming an IQ test up to 60SD above the human mean.
posted by topynate at 6:40 AM on October 20, 2014 [3 favorites]
What this means is that -practically - to make an IQ of 1000 meaningful, it'll be necessary first to have excellent knowledge of how genetics and environment predict IQ - which is a bit circular, to be honest. I think we'd have more luck identifying a g-factor in chimpanzees and constructing an inter-species IQ test than in successfully norming an IQ test up to 60SD above the human mean.
posted by topynate at 6:40 AM on October 20, 2014 [3 favorites]
Interesting article. I actually do study genetics so maybe I can address 2 points. I think a lot of the criticism in this thread is unfounded. Arguments about selection - we may not have directional selection for intelligence any more, so there may be a large number of alleles that can be artificially selected for, which is what the article proposes.
To the question of whether or not intelligence can be measured, or is genetic, you may not know how the data are collected. The twin studies indicate in a well-controlled way that you are measuring something intrinsic, as you can look for differences in test scores between identical and fraternal twins. This controls for all of the basic criticisms (which are legitimate) of cultural bias etc and simply says, if you're 100% related you'll have a lot more similar scores than if you're 50% related. If IQ or SAT scores are all socioeconomic, the result would be that genetic and fraternal twins score the same. Of course there is likely to be a massive socioeconomic component, but this controls for that and says there is also a large generic component.
Now I will get to one major problem I have with this article. All of these alleles that underlie intelligence have to have an additive effect for the argument to be true - having 10 intelligence alleles is 10x better than having 1. The twin studies cited in this paper do NOT test or support that assumption, as no/very few people in the studies will have such a large number of alleles. If you imagine a biochemical "stupidity" pathway, it's possible there are 10 genes that can break the pathway and make you smarter - but having a mutant allele in 1 gene would give you the exact same benefit as all 10. Furthermore, it also possible that these genes can interfere with each other, such that a "smart" allele in one person's genome becomes a "stupid" allele in another. In general, he's taking a small linear pattern and extrapolating it will continue to be linear over orders of magnitude beyond the original data - that's definitely not a valid approach. In his defense, though, genes for other things like height *are* additive, so he may be right.
So I'd say you guys are being too harsh - some of the data are quite good, and it's possible this is true. However, there are some critical assumptions that might also make it not true - just gotta keep studying it and see whether the genes are additive or not. Probably it'll be a mix of both, and probably in the next few decades it really will be possible to select one of 10 embryos and improve your child's expected intelligence. Whether that's useful or moral is a separate story, but maybe let's not criticize the actual science because of it's moral implications, okay?
posted by Buckt at 6:59 AM on October 20, 2014 [2 favorites]
To the question of whether or not intelligence can be measured, or is genetic, you may not know how the data are collected. The twin studies indicate in a well-controlled way that you are measuring something intrinsic, as you can look for differences in test scores between identical and fraternal twins. This controls for all of the basic criticisms (which are legitimate) of cultural bias etc and simply says, if you're 100% related you'll have a lot more similar scores than if you're 50% related. If IQ or SAT scores are all socioeconomic, the result would be that genetic and fraternal twins score the same. Of course there is likely to be a massive socioeconomic component, but this controls for that and says there is also a large generic component.
Now I will get to one major problem I have with this article. All of these alleles that underlie intelligence have to have an additive effect for the argument to be true - having 10 intelligence alleles is 10x better than having 1. The twin studies cited in this paper do NOT test or support that assumption, as no/very few people in the studies will have such a large number of alleles. If you imagine a biochemical "stupidity" pathway, it's possible there are 10 genes that can break the pathway and make you smarter - but having a mutant allele in 1 gene would give you the exact same benefit as all 10. Furthermore, it also possible that these genes can interfere with each other, such that a "smart" allele in one person's genome becomes a "stupid" allele in another. In general, he's taking a small linear pattern and extrapolating it will continue to be linear over orders of magnitude beyond the original data - that's definitely not a valid approach. In his defense, though, genes for other things like height *are* additive, so he may be right.
So I'd say you guys are being too harsh - some of the data are quite good, and it's possible this is true. However, there are some critical assumptions that might also make it not true - just gotta keep studying it and see whether the genes are additive or not. Probably it'll be a mix of both, and probably in the next few decades it really will be possible to select one of 10 embryos and improve your child's expected intelligence. Whether that's useful or moral is a separate story, but maybe let's not criticize the actual science because of it's moral implications, okay?
posted by Buckt at 6:59 AM on October 20, 2014 [2 favorites]
Oh and to the IQ standard deviation thing, that's true, but who gives a shit? He's talking about having people who are so smart they break the scale! That's cool to think about, and of course you have to do it with the scale we currently have. Maybe in the future we'll have to pick a new scale, but that doesn't mean the people won't be smart! It's true though that a 1000 IQ is not meaningful, so this article isn't actually about people with 1000 IQs, just really really smart people.
posted by Buckt at 7:08 AM on October 20, 2014 [1 favorite]
posted by Buckt at 7:08 AM on October 20, 2014 [1 favorite]
i> Probably it'll be a mix of both, and probably in the next few decades it really will be possible to select one of 10 embryos and improve your child's expected intelligence.
I have no doubt that intelligence can be improved via genetic engineering. Simply selecting out Down Syndrome embryos for example. My extreme skepticism about the article is based on whether intelligence can be improved by many multiples. I suspect that the human brain is already working near maximum efficiency given the space and energy requirements. There's simply not much headroom for improving intelligence by huge leaps and bounds. At best, one could create a 'new average', where everyone is concentrated around what is now the 140 level, rather than the 100 level, with less variation.
posted by empath at 7:14 AM on October 20, 2014
I have no doubt that intelligence can be improved via genetic engineering. Simply selecting out Down Syndrome embryos for example. My extreme skepticism about the article is based on whether intelligence can be improved by many multiples. I suspect that the human brain is already working near maximum efficiency given the space and energy requirements. There's simply not much headroom for improving intelligence by huge leaps and bounds. At best, one could create a 'new average', where everyone is concentrated around what is now the 140 level, rather than the 100 level, with less variation.
posted by empath at 7:14 AM on October 20, 2014
Unless of course, some of those genes select for gigantic craniums and more blood flow to the brain, and larger hips in females, etc.
posted by empath at 7:15 AM on October 20, 2014 [1 favorite]
posted by empath at 7:15 AM on October 20, 2014 [1 favorite]
Oddly, a story about how difficult genetically enhancing intelligence turns out to be is getting spun as though it's about how easy it is.
posted by grobstein at 7:30 AM on October 20, 2014
posted by grobstein at 7:30 AM on October 20, 2014
Actually consider a different take on the same data -- I imagine that animals also have some variation in intelligence, also controlled by thousands of different alleles. It should be possible, then, to uplift certain mammals to human intelligence and beyond.
posted by empath at 7:36 AM on October 20, 2014
posted by empath at 7:36 AM on October 20, 2014
Yeah I think it's entirely possible you're right - in the framework I was talking about it would manifest with competition or context dependence of the alleles. On the other hand, crazy stuff can happen in genetics - you can double lifespan, for example, in lots of species. I'm completely undecided, but I think it's unfair to dismiss it.
posted by Buckt at 7:55 AM on October 20, 2014
posted by Buckt at 7:55 AM on October 20, 2014
I think we can argue this we all like, but it is already going on in China and will be a reality in that country very soon. They may not know exactly which genes increases intelligence but they will stack the deck and hope for the best, so to speak. Like someone said in another metafilter thread, each country is running their own experiments wrt their cultural values, and the results are coming in... capitalist vs. socialist; democratic vs. state control; universal health care vs. fend for yourselves you sorry pleebs; and now it will be embryonic selection vs let nature take its course. Place your bets now, me-fites!
Vice article about BGI Shenzhen. They are analyzing the genetics of the smartest people they can find, recruiting through word of mouth (conferences etc) and of course evidence of smarts.
Any given couple could potentially have several eggs fertilized in the lab with the dad’s sperm and the mom’s eggs. Then you can test multiple embryos and analyze which one’s going to be the smartest. That kid would belong to that couple as if they had it naturally, but it would be the smartest a couple would be able to produce if they had 100 kids. It’s not genetic engineering or adding new genes, it’s the genes that couples already have.
How does Western research in genetics compare to China’s?
We’re pretty far behind. We have the same technical capabilities, the same statistical capabilities to analyze the data, but they’re collecting the data on a much larger scale and seem to be capable of transforming the scientific findings into government policy and consumer genetic testing much more easily than we are. Technically and scientifically we could be doing this, but we’re not.
Why not?
We have ideological biases that say, “Well, this could be troubling, we shouldn’t be meddling with nature, we shouldn’t be meddling with God.” I just attended a debate in New York a few weeks ago about whether or not we should outlaw genetic engineering in babies and the audience was pretty split. In China, 95 percent of an audience would say, “Obviously you should make babies genetically healthier, happier, and brighter!” There’s a big cultural difference.
posted by St. Peepsburg at 8:00 AM on October 20, 2014
Vice article about BGI Shenzhen. They are analyzing the genetics of the smartest people they can find, recruiting through word of mouth (conferences etc) and of course evidence of smarts.
Any given couple could potentially have several eggs fertilized in the lab with the dad’s sperm and the mom’s eggs. Then you can test multiple embryos and analyze which one’s going to be the smartest. That kid would belong to that couple as if they had it naturally, but it would be the smartest a couple would be able to produce if they had 100 kids. It’s not genetic engineering or adding new genes, it’s the genes that couples already have.
How does Western research in genetics compare to China’s?
We’re pretty far behind. We have the same technical capabilities, the same statistical capabilities to analyze the data, but they’re collecting the data on a much larger scale and seem to be capable of transforming the scientific findings into government policy and consumer genetic testing much more easily than we are. Technically and scientifically we could be doing this, but we’re not.
Why not?
We have ideological biases that say, “Well, this could be troubling, we shouldn’t be meddling with nature, we shouldn’t be meddling with God.” I just attended a debate in New York a few weeks ago about whether or not we should outlaw genetic engineering in babies and the audience was pretty split. In China, 95 percent of an audience would say, “Obviously you should make babies genetically healthier, happier, and brighter!” There’s a big cultural difference.
posted by St. Peepsburg at 8:00 AM on October 20, 2014
In any field you're in, there is always someone who cuts in shrieking that thanks to their (nearly always half informed and limited) knowledge, they have the information that will blow it and all human society apart and remake it in some glorious image they have decided is the right one. I'm in Latin literature and even we encounter this on a regular basis. It's just that in some areas we give these people a very public platform, especially if they call themselves CEO of anything.
posted by lesbiassparrow at 8:01 AM on October 20, 2014 [1 favorite]
posted by lesbiassparrow at 8:01 AM on October 20, 2014 [1 favorite]
Oh and to the IQ standard deviation thing, that's true, but who gives a shit?
I do, because I'm a mathematician, and I get a bug up my ass when people use bad math to make speculation look like certainty.
posted by escabeche at 8:17 AM on October 20, 2014 [7 favorites]
I do, because I'm a mathematician, and I get a bug up my ass when people use bad math to make speculation look like certainty.
posted by escabeche at 8:17 AM on October 20, 2014 [7 favorites]
That BGI Shenzhen article. There are so many bad ideas in such a short article...
Also, this picture. Yes, that is exactly what DNA manipulation looks like. You wouldn't believe how much of my day I spend holding things up next to each other and staring at them vaguely out of focus. If I had a dime for every time I held up the red RNA(?) to the blue DNA to match them, man, I wouldn't have to work this lousy 9 to 5, I tell you that!
(Sorry, but that image is up there with the news clip where a siRNA molecule turned into Pacman and started eating up DNA strands.)
posted by maryr at 9:17 AM on October 20, 2014 [2 favorites]
Also, this picture. Yes, that is exactly what DNA manipulation looks like. You wouldn't believe how much of my day I spend holding things up next to each other and staring at them vaguely out of focus. If I had a dime for every time I held up the red RNA(?) to the blue DNA to match them, man, I wouldn't have to work this lousy 9 to 5, I tell you that!
(Sorry, but that image is up there with the news clip where a siRNA molecule turned into Pacman and started eating up DNA strands.)
posted by maryr at 9:17 AM on October 20, 2014 [2 favorites]
kisch mokusch: "And then, presuming that people commit to this 1000-year endeavor, you've bred out all the deleterious genes and perfected the art of brother-sister mating, then you might finally have somebody with all the "right" variants."
The Kwisatz Haderach.
posted by dendrochronologizer at 9:47 AM on October 20, 2014
The Kwisatz Haderach.
posted by dendrochronologizer at 9:47 AM on October 20, 2014
As a livestock geneticist this is a fascinating, and entertaining, discussion to read. Particularly because the quality of the science done at BGI is very uneven (as an example, consider the rather innovative, and awful, approach they took to the goat genome). I'm not a human geneticist, so maybe they really are way ahead of us there, but they're not much of a player in the livestock world. If they're not producing high-quality research using large mammals that are (relatively) free of ethical considerations (relative to humans) I have no reason to conclude that they're doing better work on the human side.
In one of my papers I discussed the genetic merit of an animal we might produce if we could select only the "best" haplotypes in the current population and combine them in a single animal. The answer was awesome -- it was probably more like 35 or 40 genetic standard deviations, not 100, but I also do not believe that such an animal would actually perform at that level. In dairy cattle, milk yield and fertility are negatively correlated. There are lots of reasons for this, but it basically comes down to interactions among genes and innate limitations of physiology. The kind of physiology that an animal has to have to produce 120 pounds of milk per day at peak lactation is not one that is conducive to high fertility, and I'm not sure there's a way to break that correlation, CRISPRs or TALENs or no.
posted by wintermind at 10:36 AM on October 20, 2014 [10 favorites]
In one of my papers I discussed the genetic merit of an animal we might produce if we could select only the "best" haplotypes in the current population and combine them in a single animal. The answer was awesome -- it was probably more like 35 or 40 genetic standard deviations, not 100, but I also do not believe that such an animal would actually perform at that level. In dairy cattle, milk yield and fertility are negatively correlated. There are lots of reasons for this, but it basically comes down to interactions among genes and innate limitations of physiology. The kind of physiology that an animal has to have to produce 120 pounds of milk per day at peak lactation is not one that is conducive to high fertility, and I'm not sure there's a way to break that correlation, CRISPRs or TALENs or no.
posted by wintermind at 10:36 AM on October 20, 2014 [10 favorites]
kisch mokusch: "And then, presuming that people commit to this 1000-year endeavor, you've bred out all the deleterious genes and perfected the art of brother-sister mating, then you might finally have somebody with all the "right" variants."
The Kwisatz Haderach.
posted by dendrochronologizer at 11:47 AM on October 20
[+] [!]
If only your powers of prescience had told you that joke has already been made several times in this very thread.
Admit it, you're in the Spacing Guild.
posted by Steely-eyed Missile Man at 10:51 AM on October 20, 2014 [1 favorite]
The Kwisatz Haderach.
posted by dendrochronologizer at 11:47 AM on October 20
[+] [!]
If only your powers of prescience had told you that joke has already been made several times in this very thread.
Admit it, you're in the Spacing Guild.
posted by Steely-eyed Missile Man at 10:51 AM on October 20, 2014 [1 favorite]
"1000 different variables determine how fast a car can go. Imagine if we maxed out all 1000 of them, we'd have race cars that go 10,000 miles per hour"
posted by empath at 10:56 AM on October 20, 2014 [4 favorites]
posted by empath at 10:56 AM on October 20, 2014 [4 favorites]
So humans will go from
INT: 15 WIS: 7
to
INT:30 WIS: 7
?
I'm not sure that's really what we need.
posted by freecellwizard at 10:57 AM on October 20, 2014 [3 favorites]
INT: 15 WIS: 7
to
INT:30 WIS: 7
?
I'm not sure that's really what we need.
posted by freecellwizard at 10:57 AM on October 20, 2014 [3 favorites]
BGI seems to be founded by yes men who think that buying a lot of shit and running it fast is the way to do research. At this point, my guess is that they are desperate to prove to their government masters that they were a better investment than a new plastics factory.
I agree with wintermind -- as a bioinformaticist, much of the BGI stuff I have dealt with is shoddy.
posted by benzenedream at 11:04 AM on October 20, 2014 [1 favorite]
I agree with wintermind -- as a bioinformaticist, much of the BGI stuff I have dealt with is shoddy.
posted by benzenedream at 11:04 AM on October 20, 2014 [1 favorite]
The microbiome sequencing that I've seen coming out of BGI hasn't raised any red flags for me in terms of quality control, but I also haven't looked very deeply yet.
posted by en forme de poire at 11:26 AM on October 20, 2014
posted by en forme de poire at 11:26 AM on October 20, 2014
Maybe this is something that physicists actually do, but it's not something I've ever seen mathematicians do.
This is poorly sourced, but Wikipedia says that Rao says that Berndt says that Erdos said: "Suppose that we rate mathematicians on the basis of pure talent on a scale from 0 to 100, Hardy gave himself a score of 25, Littlewood 30, Hilbert 80 and Ramanujan 100'"
posted by madcaptenor at 12:00 PM on October 20, 2014
This is poorly sourced, but Wikipedia says that Rao says that Berndt says that Erdos said: "Suppose that we rate mathematicians on the basis of pure talent on a scale from 0 to 100, Hardy gave himself a score of 25, Littlewood 30, Hilbert 80 and Ramanujan 100'"
posted by madcaptenor at 12:00 PM on October 20, 2014
If there's one thing I've learnt here it's that I really have to read Dune.
posted by kisch mokusch at 2:09 PM on October 20, 2014 [1 favorite]
posted by kisch mokusch at 2:09 PM on October 20, 2014 [1 favorite]
To the question of whether or not intelligence can be measured, or is genetic, you may not know how the data are collected. The twin studies indicate in a well-controlled way that you are measuring something intrinsic, as you can look for differences in test scores between identical and fraternal twins. This controls for all of the basic criticisms (which are legitimate) of cultural bias etc and simply says, if you're 100% related you'll have a lot more similar scores than if you're 50% related. If IQ or SAT scores are all socioeconomic, the result would be that genetic and fraternal twins score the same. Of course there is likely to be a massive socioeconomic component, but this controls for that and says there is also a large generic component.
But what is the "something intrinsic" that is being measured? And are we sure it is a single "something"? I am a biologist, but I am also a teacher. And in my experience, "intelligence" is not one single thing. I have students who are good students because they memorize well and think fast on their feet. That is the sort of thing that SAT tests test for. (I've never taken an IQ test, so I don't actually know what they're like, but I know IQ scores are correlated with SAT scores). I have students who are good students because they are able to easily make connections between concepts. I have students who are good students because they are very good at expressing themselves in writing. I have students who are good students because they understand mathematics well and can see connections between equations and graphs and the real world. I have students who are good students because they are very creative in developing hypotheses and methods to test those hypotheses. And finally, I have students who are good students because they are incredibly hard working and persistent even if some of these things don't come as easily to them.
And among scientists, I think we can all see that different scientists have been successful for different reasons, and I would argue in an awful lot of cases that some of our most innovative scientists are in that latter group, that they have come up with amazing ideas and carried them through by simply working harder and more persistently than anyone else.
So, what is intelligence? And do SAT/IQ test scores actually tell us anything useful?
posted by hydropsyche at 2:17 PM on October 20, 2014 [2 favorites]
But what is the "something intrinsic" that is being measured? And are we sure it is a single "something"? I am a biologist, but I am also a teacher. And in my experience, "intelligence" is not one single thing. I have students who are good students because they memorize well and think fast on their feet. That is the sort of thing that SAT tests test for. (I've never taken an IQ test, so I don't actually know what they're like, but I know IQ scores are correlated with SAT scores). I have students who are good students because they are able to easily make connections between concepts. I have students who are good students because they are very good at expressing themselves in writing. I have students who are good students because they understand mathematics well and can see connections between equations and graphs and the real world. I have students who are good students because they are very creative in developing hypotheses and methods to test those hypotheses. And finally, I have students who are good students because they are incredibly hard working and persistent even if some of these things don't come as easily to them.
And among scientists, I think we can all see that different scientists have been successful for different reasons, and I would argue in an awful lot of cases that some of our most innovative scientists are in that latter group, that they have come up with amazing ideas and carried them through by simply working harder and more persistently than anyone else.
So, what is intelligence? And do SAT/IQ test scores actually tell us anything useful?
posted by hydropsyche at 2:17 PM on October 20, 2014 [2 favorites]
do SAT/IQ test scores actually tell us anything useful?
They're a starting point. The measurements of cognitive function are crude, but they get the job done. You control as best you can and I think you are somewhat limited by the sorts of groups you can study (this is not my field), but given a large enough cohort, a GWAS will find genetic determinants that contribute to whatever your selecting for (in this case, success in IQ tests). In your set of examples, one might expect multiple hits across the population, depending upon which "pathway to success" was being employed. Even the ability to work hard and be persistent can be expected to have some genetic basis. It doesn't have to be a single "something". These sorts of studies can tolerate multiple "somethings". Although the better defined the phenotype the cleaner the data.
The next step is to identify what the variants do. If they're anything like GWAS disease studies, most will reside in non-coding DNA. So the first step is to work out what genes they influence and how, what cells express them and under what conditions. That sets the scene for functional studies (i.e. gene knockout and pharmacological studies). It'll take another 10 years, but we will develop a picture of what the variants do.
Maybe they'll all work in an indirect manner, as you suggest. Maybe they will control things like metabolism and energy usage, things that enable people to focus better by reducing the rate of fatigue during an IQ test. But the genetic basis will nevertheless be real.
Of course, the environmental influence is almost certainly greater. So at the end of the day, you'll probably find genetic variation influencing "intelligence" by a few percentage points while we already know that things like in utero and post-natal nutrition, early childhood learning and a whole bunch of other socio-economically-dependent factors affect cognitive ability to a much greater extent.
posted by kisch mokusch at 3:58 PM on October 20, 2014 [3 favorites]
They're a starting point. The measurements of cognitive function are crude, but they get the job done. You control as best you can and I think you are somewhat limited by the sorts of groups you can study (this is not my field), but given a large enough cohort, a GWAS will find genetic determinants that contribute to whatever your selecting for (in this case, success in IQ tests). In your set of examples, one might expect multiple hits across the population, depending upon which "pathway to success" was being employed. Even the ability to work hard and be persistent can be expected to have some genetic basis. It doesn't have to be a single "something". These sorts of studies can tolerate multiple "somethings". Although the better defined the phenotype the cleaner the data.
The next step is to identify what the variants do. If they're anything like GWAS disease studies, most will reside in non-coding DNA. So the first step is to work out what genes they influence and how, what cells express them and under what conditions. That sets the scene for functional studies (i.e. gene knockout and pharmacological studies). It'll take another 10 years, but we will develop a picture of what the variants do.
Maybe they'll all work in an indirect manner, as you suggest. Maybe they will control things like metabolism and energy usage, things that enable people to focus better by reducing the rate of fatigue during an IQ test. But the genetic basis will nevertheless be real.
Of course, the environmental influence is almost certainly greater. So at the end of the day, you'll probably find genetic variation influencing "intelligence" by a few percentage points while we already know that things like in utero and post-natal nutrition, early childhood learning and a whole bunch of other socio-economically-dependent factors affect cognitive ability to a much greater extent.
posted by kisch mokusch at 3:58 PM on October 20, 2014 [3 favorites]
Yeah, as kisch mokusch said, doing gene association studies does not imply a commitment to g as a single unitary biological entity, and could actually end up undermining that idea more than supporting it (I'd guess that if they find much of anything it's going to be distributed across a variety of different pathways). I also agree that the biggest gains from this type of research would be from a basic science perspective (e.g. better understanding human brain development in general, and maybe how that development can be affected in disease states) as opposed to engineering hypersmart metahumans or wtfever.
posted by en forme de poire at 4:34 PM on October 20, 2014 [2 favorites]
posted by en forme de poire at 4:34 PM on October 20, 2014 [2 favorites]
Man, I almost never comment anymore but I saw this and came in here to make a joke but there's already Dune and Vernor Vinge and Peter Watts, and... where else on the internet are people going to make the exact same references you were going to make? You people never cease to creep me out are wonderful.
posted by Dormant Gorilla at 12:09 AM on October 21, 2014 [1 favorite]
posted by Dormant Gorilla at 12:09 AM on October 21, 2014 [1 favorite]
Even if you do get association analysis results that seem to make some kind of sense, well, validation can be quite difficult. Even in a species like humans where the annotation is of [relatively] high quality (which is definitely not the case in the cow) it's extremely difficult to untangle how sequence differences result in changes in phenotypes, doubly so for complex (quantitative) traits. Heck, it's hard even when you can clearly see from the data that a mutation is truncating a protein.
en forme de poire, I think their raw sequence data can be okay, but I wouldn't care to work with a de novo assembly from BGI.
posted by wintermind at 8:42 AM on October 21, 2014 [3 favorites]
en forme de poire, I think their raw sequence data can be okay, but I wouldn't care to work with a de novo assembly from BGI.
posted by wintermind at 8:42 AM on October 21, 2014 [3 favorites]
Google hits for "Gauss": about 18 million.
Google hits for "Dickens": about 48.5 million.
Google hits for Justin Bieber: about 209 million. I'm not sure that's a proper metric. Should we ask a mathematician?
posted by Mental Wimp at 9:31 AM on October 21, 2014 [1 favorite]
Google hits for "Dickens": about 48.5 million.
Google hits for Justin Bieber: about 209 million. I'm not sure that's a proper metric. Should we ask a mathematician?
posted by Mental Wimp at 9:31 AM on October 21, 2014 [1 favorite]
Perhaps mathematicians and physicists, and humanists for that matter, should simply face the fact that there are cultural agents that far outweigh their own favorites.
posted by No Robots at 9:51 AM on October 21, 2014
posted by No Robots at 9:51 AM on October 21, 2014
Perhaps mathematicians and physicists, and humanists for that matter, should simply face the fact that there are cultural agents that far outweigh their own favorites.
Technology, which is undergirded by mathematics, has had such a vast impact on society that it is hard to argue that literature or art is competitive. Notice I'm not saying positive impact or uplifting impact. Just impact. And when you look around you, at your desk, out your window, at your car and your home, the food you eat, the clothes you wear, the way you amuse yourself, and the way you communicate with the world, you'll notice the impact of technology more than the combined works of all the poets, writers, painters, actors, musicians, directors, composers and sculptors. Now, a few people have escaped this environment and eschew much of this technology, but for the vast, vast majority of the planet, technology has transformed the way they live like nothing before. Not saying it's a good or bad thing, just a fact.
posted by Mental Wimp at 10:28 AM on October 21, 2014
Technology, which is undergirded by mathematics, has had such a vast impact on society that it is hard to argue that literature or art is competitive. Notice I'm not saying positive impact or uplifting impact. Just impact. And when you look around you, at your desk, out your window, at your car and your home, the food you eat, the clothes you wear, the way you amuse yourself, and the way you communicate with the world, you'll notice the impact of technology more than the combined works of all the poets, writers, painters, actors, musicians, directors, composers and sculptors. Now, a few people have escaped this environment and eschew much of this technology, but for the vast, vast majority of the planet, technology has transformed the way they live like nothing before. Not saying it's a good or bad thing, just a fact.
posted by Mental Wimp at 10:28 AM on October 21, 2014
for the vast, vast majority of the planet, technology has transformed the way they live like nothing before
Do you count written language and numerical systems as technology? And if so, isn't the entire history of literature and mathmatics part of the history of technology?
Trying to quantify who or what has had the most influence on humanity, Gauss, Shakespere or Beber, seems like a total waste of time as we all interact and influence each other across specialties, cultures and time.
posted by latkes at 12:01 PM on October 21, 2014 [2 favorites]
Do you count written language and numerical systems as technology? And if so, isn't the entire history of literature and mathmatics part of the history of technology?
Trying to quantify who or what has had the most influence on humanity, Gauss, Shakespere or Beber, seems like a total waste of time as we all interact and influence each other across specialties, cultures and time.
posted by latkes at 12:01 PM on October 21, 2014 [2 favorites]
Trying to quantify who or what has had the most influence on humanity, Gauss, Shakespere or Beber, seems like a total waste of time as we all interact and influence each other across specialties, cultures and time.
Agree. I was just responding to the provocations of "Hint: it probably wasn't a physicist."
The thought experiment of imagining the counterfactual reality without, say, mathematics or art, is kind of fun, though. For each, I can imagine experience being degraded severely, but in totally different ways. The cold mechanics of technology without the leavening of art would be a dreary existence, whereas an art-filled existence without the comforts and egalitarianism that technology has provided would be a hard one. And, of course, dealing with the cross-fertilization between, say, art and technology even at the most gross and obvious level (e.g., development of materials vis-a-vis artistic tools and creations) is what makes this thought-experiment totally subjective entertainment rather than a serious endeavor.
posted by Mental Wimp at 12:38 PM on October 21, 2014 [1 favorite]
Agree. I was just responding to the provocations of "Hint: it probably wasn't a physicist."
The thought experiment of imagining the counterfactual reality without, say, mathematics or art, is kind of fun, though. For each, I can imagine experience being degraded severely, but in totally different ways. The cold mechanics of technology without the leavening of art would be a dreary existence, whereas an art-filled existence without the comforts and egalitarianism that technology has provided would be a hard one. And, of course, dealing with the cross-fertilization between, say, art and technology even at the most gross and obvious level (e.g., development of materials vis-a-vis artistic tools and creations) is what makes this thought-experiment totally subjective entertainment rather than a serious endeavor.
posted by Mental Wimp at 12:38 PM on October 21, 2014 [1 favorite]
Now, a few people have escaped this environment and eschew much of this technology, but for the vast, vast majority of the planet, technology has transformed the way they live like nothing before.
I mean, I guess so, but a long, comfortable life without art wouldn't be worth living.
posted by empath at 1:59 PM on October 21, 2014 [1 favorite]
I mean, I guess so, but a long, comfortable life without art wouldn't be worth living.
posted by empath at 1:59 PM on October 21, 2014 [1 favorite]
-Adaptive evolution and non-coding regions
-On the genetic architecture of intelligence and other quantitative traits
-Embryo Selection for Cognitive Enhancement: Curiosity or Game-changer?
posted by kliuless at 6:33 AM on November 11, 2014 [1 favorite]
-On the genetic architecture of intelligence and other quantitative traits
-Embryo Selection for Cognitive Enhancement: Curiosity or Game-changer?
Human capital is an important determinant of individual and aggregate economic outcomes, and a major input to scientific progress. It has been suggested that advances in genomics may open up new avenues to enhance human intellectual abilities genetically, complementing environmental interventions such as education and nutrition. One way to do this would be via embryo selection in the context of in vitro fertilization (IVF). In this article, we analyze the feasibility, timescale, and possible societal impacts of embryo selection for cognitive enhancement. We find that embryo selection, on its own, may have significant (but likely not drastic) impacts over the next 50 years, though large effects could accumulate over multiple generations. However, there is a complementary technology – stem cell-derived gametes – which has been making rapid progress and which could amplify the impact of embryo selection, enabling very large changes if successfully applied to humans.also btw, fwiw...
- Scientists can now delete and fabricate memories in mice. Are humans next? "Therapists already try to retrain problematic associations through some cognitive-behavioral therapy techniques and other treatments. But PTSD and some addictions can still be quite difficult to treat in many people. So it's possible that memory modification may, one day, be able to help."
- The Science Of Smart - "We'll head to Toronto to meet one of the world's leading experts on bilingualism and the brain. Then it's on to Utah where schoolchildren are learning to speak Chinese in a statewide effort to boost overall school success. Next: UCLA researchers explain how looking at motor skills affects how we learn and remember. The American RadioWorks crew wraps up in Georgia, tagging along with a medical student who is using brain research to change how he studies -- and changing his Ds into As in the process."
- The Evolution of Human Science - "We should always remember that the technologies that made metahumans possible were originally invented by humans, and they were no smarter than we."
- Paralysed man walks again after cell transplant - "A paralysed man has been able to walk again after a pioneering therapy that involved transplanting cells from his nasal cavity into his spinal cord."
- Embryonic Stem Cells Restore Vision In Preliminary Human Test - "Isabella Beukes of Santa Rosa, Calif., has been legally blind for more than 40 years. But within weeks of getting the cells, she started to see better. She could make out the cursor on her computer screen and the color of her clothes. Today, she can hike the hills near her house all by herself."
- Scientists generate first human stomach tissue in lab with stem cells - "Scientists used pluripotent stem cells to generate functional, three-dimensional human stomach tissue in a laboratory -- creating an unprecedented tool for researching the development and diseases of an organ central to several public health crises, ranging from cancer to diabetes. Scientists used human pluripotent stem cells -- which can become any cell type in the body -- to grow a miniature version of the stomach."
- Cloning Cows From Steaks (and Other Ways of Building Better Cattle) - "Can animal geneticists breed—or clone—a cow that will taste great and have a smaller environmental hoofprint too?"
- For $100,000, You Can Clone Your Dog - "Behind glass in a never-before-used operating room inside a just-built cabin at the end of a freshly paved road, Dr. Hwang Woo Suk is chasing rogue flies with an electrified bug swatter that looks like a small tennis racket. He wears baby blue scrubs branded with the logo of Sooam Biotech, his South Korea-based research company, and is making final checks of this temporary facility, erected from scratch in eight days in the Chinese city of Weihai. Here, in a few hours, he'll deliver the first cloned puppies in the country's history."
- Our Biotech Future - "Now, after three billion years, the Darwinian interlude is over. It was an interlude between two periods of horizontal gene transfer. The epoch of Darwinian evolution based on competition between species ended about ten thousand years ago, when a single species, Homo sapiens, began to dominate and reorganize the biosphere. Since that time, cultural evolution has replaced biological evolution as the main driving force of change. Cultural evolution is not Darwinian. Cultures spread by horizontal transfer of ideas more than by genetic inheritance. Cultural evolution is running a thousand times faster than Darwinian evolution, taking us into a new era of cultural interdependence which we call globalization. And now, as Homo sapiens domesticates the new biotechnology, we are reviving the ancient pre-Darwinian practice of horizontal gene transfer, moving genes easily from microbes to plants and animals, blurring the boundaries between species. We are moving rapidly into the post-Darwinian era, when species other than our own will no longer exist, and the rules of Open Source sharing will be extended from the exchange of software to the exchange of genes. Then the evolution of life will once again be communal, as it was in the good old days before separate species and intellectual property were invented."
posted by kliuless at 6:33 AM on November 11, 2014 [1 favorite]
« Older The story of the Mamas and the Papas, as "an epic... | An enthusiastic public reading journal..... Newer »
This thread has been archived and is closed to new comments
Or, to put it more simply, I refuse to believe that nature is so crap at putting brains together that you're a couple of simple genetic switches away from having a vastly more powerful brain in the same head. I think we'll get higher intelligence, but not without using something totally off the evolutionary track, like electronic processors. I think genetic engineering could result in lessar improvements, but I think it's more likely that you could do something like making everyone's intellectual potential as high as the strongest thinkers we have now, not making superminds.
posted by Mitrovarr at 3:33 PM on October 19, 2014 [21 favorites]