[RD] Discussion on IQ (split from effect of white people on America)

[red_elk]

There should be some correlation. At least, it's reasonable to expect that people who made significant achievements in science, have above average IQ

 

[Hrothbern]

Is there any correlation between ambition and IQ? I knew someone with a high IQ who died of alcoholism at a relatively young age. People had great expectations for them and they felt pressured to achieve and reacted by basically shutting down, or lacked the ambition to move forward.

ed: doh

Read the short stories "The Young Archimedes" of Aldous Huxley and the classic fairytale of Andersen "The Ugly Duckling"

Describing some of the issues.

There better be a match between your surrounding and your talents
Not everybody is that lucky, or it takes some time before you can spread your wings

 

[Senethro]

People really seem to think that IQ is your intelligence stat like its an RPG and not just some high pattern recognition capability.

 

[Berzerker]

IQ is not a measurement of actual, workable intelligence.

I know your context is IQ /= achievement which I agree with but does IQ "measure" or estimate an ability to think in abstract terms?

edit: I dont know if I agree IQ /= achievement

actually I'd expect there to be a connection but that wouldn't mean every smart person 'achieves' what is expected of them. Everyone has their own priorities, but inventors are generally smart people achieving progress for the rest of us. Thank you

 

[Hrothbern]

People really seem to think that IQ is your intelligence stat like its an RPG and not just some high pattern recognition capability.

Pattern recognition is imo hugely underrated in practical value.
Perfect for fast effortless decisions..... and at the same time a big hand if you need to sweat some logic, produce more defendable and also more reliable decisions.

EDIT
I wonder how tests would function where you would not measure serial logic supported by other nonspecific features,
but measure it directly
And yeah... that would be difficult to keep abstract

 

[bernie14]

Read the short stories "The Young Archimedes" of Aldous Huxley and the classic fairytale of Andersen "The Ugly Duckling"

Describing some of the issues.

There better be a match between your surrounding and your talents
Not everybody is that lucky, or it takes some time before you can spread your wings

nah, just watch gattaca!

 

[Hrothbern]

nah, just watch gattaca!

I had to read the summary...
I think that movie describes how determination can be the ruling factor.... and achieving something the measure stick to measure.
The determined imperfect "better" thant the undetermined perfect.
You can also see that as a perfect who did not found his match.

 

[Lexicus]

We watched Gattaca in my high school bioethics class, I credit it with giving me a very dim view of any form of biological or genetic determinism.

 

[Ryika]

Hm, what exactly have you achieved? Eg have you presented any important paper, run some experiment, proven something in science or math etc? Cause (and this isn't something i am saying to oppose you; i could just as easily have quoted some others in the thread, but i go with you because you are a mod so you aren't that likely to feel opposed) it does seem a bit funny to claim that intelligence isn't much or that (even faulty examination of it via an iq test) means little to nothing, but at the same breath write down that you supposedly have some genius-territory intelligence.
So i think it is a good question (again, not just aimed at you; you don't even have to answer if you don't feel like it) to put: what actual manifestation of a genius-territory intelligence can you point to? Published papers of note, proof of theorem, scientific experiment, patents etc. Einstein himself is said to have had an iq less than 170.

I also recall something noted by user Perfection once, in this forum: "We are all in the top 1% here".

But isn't that exactly the point? I'm a bit surprised that you would question her high IQ, but maybe that's just because you're unfamiliar with her posts - because they really read like "Hello, high-IQ person here." to me. If anything, I'm surprised to hear that she has so many downsides because of it, because she always seems "highly functional" to me.

But in any case, I don't see why you would expect some exceptional achievements purely based on her IQ. The reality is, exceptional achievements are always a combination of a multitude of factors. While IQ is one of these factors, to achieve something "outstanding", you need a ton of other factors to also be in place. In general, you can expect most people to "underperform" if all you do is to take a single factor and them compare them to the people who are similar to them in that one factor and ignore anything else. That's ... I'm not sure what it's called. "Visibility bias"? You only concentrate on the outstanding achiever, while you are "blind" to all the people who share that characteristic but have not achieved anything outstanding.

With that said though, I do not share LM's sentiments about IQ entirely, although there is certainly some truth to them. A high IQ stops being being useful for most things in everyday life after a certain break point, and can even become a negative thing, but that's true for really high IQs only; and it seems like her opinion on IQ as a whole is a bit "tainted" because of her own experiences on the issue.

In general I would argue that the biggest benefit of a high IQ is that you can actually analyze your own life, your relationships and everything else - that's such a great utility that really makes you appreciate the situation you're in. Having worked with people of sub 60 IQ during my Freiwilliges Soziales Jahr (in a sheltered workshop for people with mental and psychological disabilities), I can say that these people are basically on auto pilot. No ability to reflect on what they do. No ability to shape who they are. And while they're able to construct emotional connections with other people, they do not have the ability to really enter meaningful relationships that mean anything in the grand scale of things. I'm sure they don't feel like that, but from an outside perspective it seemed like they're just people floating in an ocean that is filled with noise but nothing of real value.

Go closer to normal IQ, and you see that people begin to think about their lives more and more. Relationships are based on a shared connection, a back-and-forth of ideas and emotions, instead of it just "being there", and as you go up further in IQ that only improves - because while a high IQ is correlated with being less social, the social connects that you do build, feel so much stronger and meaningful. That's again a bit of a questionable value, because it is not something that low-IQ people "would like to do but can't", but rather something that only becomes important to a person only because they have a rather high IQ. So maybe my opinion is also "tainted" because of my position in life (by just barely hitting the top 1% with an IQ right below 140), but I find that having that sort of analytic intelligence really adds value to everything. Well, at least if you make use of that potential I guess.

 

[Bootstoots]

@Bootstoots, I have never seen a satisfactory rebuttal to the point that the group differences are basically going to be an artifact of how you construct the groups.

It is possible to cluster human populations into a few large clusters based on how genetically similar they are to each other, and the way we define "race" ended up overlapping fairly well (although I doubt it's 99.86%) with the actual distribution of genetic variation. I suspect that, in particular, it's the "white/European" and "East Asian" groups that correspond especially closely with patterns of genetic similarity. Based on geographic isolation from other large groups, "Native American" and "Aboriginal Australian" would also correspond to genetically distinct clusters of people.

OTOH, the "black" grouping is a mess, as there's considerably more genetic variation within the African continent than outside it. So, for instance, some people might consider the Khoisan to be "black" based on their darker skin (albeit light relative to their Bantu neighbors) and African origin, but that wouldn't make any genetic sense because the Khoisan are more genetically divergent from all other human populations than any other group. They represent the most distant split between living human groups, having diverged at least 100,000 years ago from the line leading to all other humans. Ignoring them, there's still more difference between, say, most West African populations and East African populations compared to the difference between, say, whites and Native Americans.

That's not even accounting for the way "black" is defined in the US; as you know, US blacks have an average of 20-25% European ancestry and some are majority European. In Brazil they'd mostly be pardo, and the lightest-skinned ones might even be white. So obviously there's an extra layer of complication there.

I suspect that a high-resolution survey of many different ethnic groups in Africa would, if malnutrition and other known IQ-lowering effects were somehow accounted for, reveal differences in average IQ and other heritable personality traits from group to group.

I also wish I understood more about epigenetics because my current understanding of that is that the existence of the phenomenon means there are environmental factors determining how genes are expressed, which genes are triggered, and so on. To me this blows the idea that genetic = predetermined and unchangeable completely out of the water. The whole idea that genes=nature and social=nurture is wrong. And so the terms under which this discussion is generally carried on are also wrong.
When I see bits like this:

It just flabbergasts me that anyone who claims to have any understanding of modern genetics actually can believe that genes could be a fixed blueprint that determine life outcomes before a person is even born.

Epigenetics is certainly real and interesting, but it's generally only a secondary contributor to various phenotypic traits. It is known to matter for some traits like tendency to anxiety, depression, and addiction; while studies on epigenetic effects on IQ seem inconclusive so far, it's certainly possible that some effect may be found. But it's not likely to be especially large.

Epigenetic modifications, like DNA methylation patterns, are passed on from parents to child. However, the effects fade over the course of someone's lifetime, so that the epigenomes of identical twins raised in different environments slowly diverge from each other as they get older. Because of that, epigenetic effects from, say, a famine that caused a period of malnourishment in the parents' childhoods usually fade away to undetectability within three or four generations, and they even fade within a given individual over time.

The misconception where genetics are thought to be a deterministic blueprint is definitely a problem. Just about all traits are affected both by genetics and environment (including epigenetic effects), so that even the environmental 30% contribution towards a 70% heritable trait is still huge and leaves room for all sorts of possible outcomes wrt that trait. But, at the same time, the underlying genetics still do contribute a large fraction (in this case a majority) of the total variability, such that two groups of people who have somewhat different underlying genetics, who are subjected to a similar spectrum of environments, will end up with trait differences on average..

Here's another problem: "black people," "white people," and "intelligent" are all not defined specifically enough that you can make scientifically rigorous claims about them.

I covered the problems with "black people" above. As for "white people", that's one of those groupings that is pretty well-defined at the center but fuzzy around the edges. People from the European continent (or rather, whose ancestors were European 500 years ago) are pretty closely related to each other. But where to draw the line is unclear. Should Turks be considered "white", or Iranians, or Arabs, or northern Indians? Many Slavs have a bit of Asian admixture from the Mongol invasions, many Balkan peoples have a fair amount of ancestry from everywhere else in the old Ottoman Empire, and so on. Different ways of defining it will give different average results on whatever trait you measure.

If you choose to define it fairly narrowly as "the vast majority of ancestors 500 years ago lived somewhere in the European continent", which is pretty close to how Americans define "white", and decide to measure skin melanin levels, then this population will have a very low average. If you define it more broadly, then there will be somewhat more melanin on average. Both, though, will have much lower values than any population considered "black", even though the group called "black" is extremely heterogeneous. The same can be true for traits that aren't as obvious.

"Intelligence" here is defined really narrowly as "abstract reasoning", or basically puzzle-solving skill, as measured on tests of that. Those results do correlate strongly with each other and have a fairly high correlation with life outcome, possibly because abstract reasoning is demanded by modern economies. I really wish that IQ had been called something else, because the way we use the word "intelligence" in everyday speech is far broader than it's used for psychometrics. So the statement that IQ differences exist between races (as currently defined) and might have a genetic component gets turned into "black people are generally dumber than white and Asian people", which is wrong. And of course, as LM said, there are a lot of stupid people with high IQs out there.

So far, epigenetics research has only identified certain effects that can have intergenerational effects. These have mostly do with brain's reward system (things such as fear, anxiety, depression and addiction). It's not at all clear that epigenetics has anything to do with intelligence, or if it does, the effect is most likely small (twin studies give fairly high numbers for heritability, and in these, epigenetic changes would be included in the environmental category). Epigenetics also cannot express or suppress gene variants that aren't in the genome. In any case, it's a fascinating phenomena, and one that requires more study (especially studies spanning multiple generations). Also, the IQ gaps haven't really been changing, but if they ever find a way to close the IQ gaps, then I'll gladly admit I was wrong.

Second, as far as your groupings go, not really. We can group people based on genetic similarity. These self-identified races match best-fit genetic clusters with 99,86% accuracy.

Shouldn't some of the epigenetic effect show up as inherited rather than environmental? Epigenetic effects do fade over an individual's life for environmental reasons, but they are inherited at first, and identical twins' epigenomes would remain quite a bit more similar to each other even in adulthood than to an unrelated person.

As for the groupings, could you post a link to that study? Although I'd expect a very strong correlation between genetic clusters and self-identified race, that number still seems really high, especially in edge cases like e.g. central Asians.

 

[Lexicus]

It is possible to cluster human populations into a few large clusters based on how genetically similar they are to each other, and the way we define "race" ended up overlapping fairly well (although I doubt it's 99.86%) with the actual distribution of genetic variation. I suspect that, in particular, it's the "white/European" and "East Asian" groups that correspond especially closely with patterns of genetic similarity. Based on geographic isolation from other large groups, "Native American" and "Aboriginal Australian" would also correspond to genetically distinct clusters of people.

I don't dispute this. But IIRC Lewontin showed that it is possible to separate the populations into clusters corresponding with Mayan, Maori, and Icelandic ancestry and use those as your 'racial' categories just as effectively: you can describe all human populations in terms of an admixture of those three archetypes. I don't see why "black" "White" and "Asian" should be taken as somehow more significant or 'fundamental' groupings.

Epigenetics is certainly real and interesting, but it's generally only a secondary contributor to various phenotypic traits. It is known to matter for some traits like tendency to anxiety, depression, and addiction; while studies on epigenetic effects on IQ seem inconclusive so far, it's certainly possible that some effect may be found. But it's not likely to be especially large.

It seems to me that anxiety, depression, and addiction are all likely to influence "intelligence".

 

[Bootstoots]

I don't dispute this. But IIRC Lewontin showed that it is possible to separate the populations into clusters corresponding with Mayan, Maori, and Icelandic ancestry and use those as your 'racial' categories just as effectively: you can describe all human populations in terms of an admixture of those three archetypes. I don't see why "black" "White" and "Asian" should be taken as somehow more significant or 'fundamental' groupings.

It's true that, in applying cluster analysis on human genetic diversity, you have to make choices involving how many clusters you wish to split humanity into, and that the input data affect the groupings you get back. If you used only Icelanders, Mayans, and Maori to "train" your cluster model, and then input genetic data from a large variety of humans, what it would spit out would be that everyone is best represented by x% Icelandic, y% Mayan, z% Maori. That would obviously be a bad choice - the model would do a poor job of covering the full space of human genetic variation, compared to if you had trained it on a large number of different ethnic groups from all over the world. That doesn't invalidate it as a tool, and it's not clear to me why anyone would think it does. It just indicates that garbage in, garbage out applies to this like anything else.

Here is a paper that does a fairly good job of doing a human genetic cluster analysis with a wider variety of populations, analyzing most of humanity and then repeating the analysis for Eurasian groups with a larger data set. For the worldwide group you'll see that they still only sampled 296 individuals from 13 populations; there would have been some differences if they'd included more populations or different ones, and even if they repeated it with different individuals from the same population, the results would have been slightly different. But the overall patterns match quite closely to those of other analyses.

They show results from different numbers of clusters (K); K=4 resulted in one cluster that was almost entirely African, one almost entirely Native American, and also a group corresponding to Europeans and another to Asians; Eurasian populations are different mixtures of the two. At K=12, quite a bit more population structure pops out within continents. The principal component analysis (PCA) shows that the sampled African populations are pretty distinct from all other populations, and that the other groups vary from each other based on their geographic distance. This is consistent with the other analyses that have been done.

For the Eurasian set, they show the results for K=7; here it's obvious that different racial or ethnic groups have genetics distributed as clines rather than distinct groups, except where substantial barriers separate them, but yet different groups of people are usually more similar to their neighbors than to people distant from them. Of course we already sort of knew that, broadly, but this both demonstrates that the method produces plausible results and gives us information about migration patterns.

What I'd conclude is that, while our concept of "race" is socially constructed, it still mostly corresponds to real genetic distance between large groups of people. How many such groups you want to define is arbitrary, but the "white", "Asian", "Native American", and even "black" groups that pop out of cluster analyses with K=4 correspond fairly well to the way we classify them socially. Dialing up the resolution then allows finer distinctions such as the relations between and migration histories of different ethnic groups.

It seems to me that anxiety, depression, and addiction are all likely to influence "intelligence".

Sure; IQ results are affected to some extent by each of these conditions. I don't know much about the relation between IQ and psychological conditions, TBH. But, thus far, very little (that I'm aware of) has been found on epigenetic effects wrt IQ specifically, although the data are more solid for some psychological conditions. It would be very interesting if epigenetic effects on IQ are found.

 

[Bootstoots]

Frankly, Boots, I feel like you're wildly off-base.

Let's say there is unequivocal proof released that black people are less intelligent than white people. What happens next?

What will be done towards an ethical and compassionate response that is different than what is already being done/advocated for?

Why is this information valuable?

You plainly recognize that this will be used as fuel for white nationalism. So you must believe that the benefits outweigh the costs. What benefits do you see coming from this "real science"?

Why? Because white supremacists release cherry-picked data based on flawed methodologies saying it's true?

I don't think that suppression or denial of scientifically established facts is likely to work in the long run. I mean, this stuff is already getting out pretty quickly to white nationalists and their ilk, and they're spreading it fairly effectively. When people on one side of an argument really do have pretty good data on their side, and the other side can't hold up on the evidence and attempts to shout them down instead, it tends to make relatively neutral people more likely to believe the more data-driven side. The way the current dynamic is, people who see the evidence and end up "switching sides" have this tendency to overinterpret it, think that it means a lot more than it really does, and turn into full-fledged racists/white nationalists.

What I'm definitely not saying is that people should try to spread this information widely. But there are some countermeasures that might be helpful. One I can think of is to make sure people know what IQ is - puzzle-solving ability, not the sum total of human cognition. I'd like to develop more, because I think that this topic will keep coming up and being used as ammunition for the far right whether we do anything or not, especially as information about what genes impact IQ and how their alleles are distributed among racial groups becomes known.

 

[Synobun]

I don't understand how you can consider their data to be scientifically established fact, or that it's "pretty good", unless you inherently believe the message they're peddling.

 

[Bootstoots]

I don't understand how you can consider their data to be scientifically established fact, or that it's "pretty good", unless you inherently believe the message they're peddling.

What I consider to be established are the existence of average IQ differences (not that there is a significant genetic component; this is plausible but not proven) and that the pattern of human genetic variation is such that what we think of as "races" do roughly end up clustering together if you do the genetic analysis.

I really didn't want to believe this at all, and still don't. The possibility of a genetic basis for this - which might be found relatively soon - is highly disturbing. I'd love if someone could actually shoot this stuff down on the facts, and I was hoping that someone would.

 

[Synobun]

What I consider to be established are the existence of average IQ differences (not that there is a significant genetic component; this is plausible but not proven) and that the pattern of human genetic variation is such that what we think of as "races" do roughly end up clustering together if you do the genetic analysis.

I really didn't want to believe this at all, and still don't. The possibility of a genetic basis for this - which might be found relatively soon - is highly disturbing. I'd love if someone could actually shoot this stuff down on the facts, and I was hoping that someone would.

That's a completely different assertion, though. You're confirming two things: IQ differences exist, and there is such a thing as race (genetically speaking). There aren't many that would disagree. Most of the "race isn't real" people mean it in a social sense, and IQ differences are easily explained (made for a certain kind of society).

But those two ideas are completely different than what's being asserted: IQ is dependent on race and (shockingly) those who are described as 'black' are significantly inferior compared to their northern 'white' counterparts. You can suggest those two ideas without meshing them together into a white supremacist's dream come true.

 

[Bootstoots]

That's a completely different assertion, though. You're confirming two things: IQ differences exist, and there is such a thing as race (genetically speaking). There aren't many that would disagree. Most of the "race isn't real" people mean it in a social sense, and IQ differences are easily explained (made for a certain kind of society).

But those two ideas are completely different than what's being asserted: IQ is dependent on race and (shockingly) those who are described as 'black' are significantly inferior compared to their northern 'white' counterparts. You can suggest those two ideas without meshing them together into a white supremacist's dream come true.

I suspect that some would dispute the "average IQ varies by race" claim (I omitted the "by race" in the post above). Also, some would dispute the meaningfulness of IQ. But, fair enough.

I find it worryingly plausible that the genetic component of abstract reasoning/IQ really will turn out to be different among racial groups on average, in such a way that whites and Asians have higher average IQ than blacks. I really hope this is not the case, but we may find convincing genetic data within the next few years that would imply that.

And yeah, if that happens, everyone from Richard Spencer to Charles Murray will have a field day. White nationalist beliefs and other racism might surge. So what we need is a contingency plan in case it turns out like this, something better than denying it stridently and calling them racist.

 

[Ryika]

That's a completely different assertion, though. You're confirming two things: IQ differences exist, and there is such a thing as race (genetically speaking). There aren't many that would disagree. Most of the "race isn't real" people mean it in a social sense, and IQ differences are easily explained (made for a certain kind of society).

But those two ideas are completely different than what's being asserted: IQ is dependent on race and (shockingly) those who are described as 'black' are significantly inferior compared to their northern 'white' counterparts. You can suggest those two ideas without meshing them together into a white supremacist's dream come true.

What we know is that we can measure large differences with our IQ tests and that the base assumption that we can draw from that - people from certain localities have much lower IQs than people from other localities - is very naive and inaccurate. Of course there are other factors, and those likely contribute to a gap that seems much wider than the actual difference in the thing that we try to measure as "IQ" is, but there is currently no way to definitively come to a conclusion on whether the remaining gap is at all significant or not.

So all you're saying in your post is that it "sounds racist, therefor it's probably not true", but that makes no sense at all. You are making a logical jump here, because we are literally talking about data that we do not have. Showing that a gap is not as large as the raw data suggests is not the same as showing that the gap does not exist.

In reality, it could go either way and there's no way to know with certainty at this point in time.

 

[Lemon Merchant]

Hm, what exactly have you achieved? Eg have you presented any important paper,

Yes.

run some experiment

Yes.

it does seem a bit funny to claim that intelligence isn't much or that (even faulty examination of it via an iq test) means little to nothing, but at the same breath write down that you supposedly have some genius-territory intelligence.

I never said that intelligence isn't important. I said that an IQ test isn't a definitively good way to measure it. I also said that above a certain level, an IQ score has little correlation in the real world. It's meaningless. I never said anything about intelligence not mattering. I also listed my own IQ to make the point how meaningless it really is. I didn't put it there for bragging rights.

So i think it is a good question (again, not just aimed at you; you don't even have to answer if you don't feel like it) to put: what actual manifestation of a genius-territory intelligence can you point to? Published papers of note, proof of theorem, scientific experiment, patents etc.

Why should I have to:

A. List any published papers I have authored? Doing so would out me on the forum and put me in the dangerous situation of some deranged idiot looking me up in RL. No thank you. I had a stalker once. I don't wish to repeat the experience.
B. Have some manifestation to begin with? What difference does it make? My life is in my work. I work to improve people's lives. If I have something to share, I will tell a colleague, or send an email. Why do I have to have a body of work in addition to the work I do everyday? Why must I have this ambition? I've already defended a thesis and dissertation, why must I defend my "lack of ambition?" I don't care about the trappings of "smartness." I care about the people that I help.
C. Achieve anything to satisfy you or anyone else? Again, what difference does it make?

Einstein himself is said to have had an iq less than 170.

His score, and Stephen Hawking's score was 160, apparently. I am not as intelligent as either of those two men. I am just better at solving puzzles.

 

[Senethro]

"Sounds racist, therefore its probably not true" is a rule of thumb I use on a near daily basis sometimes.