Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[martinuzz]

Intelligence does most definitly have a genetic base.

[hops]

Intelligence does most definitly have a genetic base.

What I meant is that unless you have a mental disability, having more "brainpower" isn't that important.

[iceball3]

Also, IQ isn't everything.
I say without pride that my IQ seems to be stupidly high, because it hasn't really stopped me from being a useless potato with 0 ambitions. People with much lower IQ but higher ambitions could get way more done and contribute much more to the world than I have, at this point.

[wierd]

Indeed. Mine is also supidly high, and I too have low ambition.

My big ambitious goal? Save enough money to escape the hamster wheel of american living, and live in a shack in some God forsaken wilderness far away from other people and their incessant inane bullshit.

That's it. Its all I want.

[Reelya]

And the other problem with IQ tests is that they are calibrated based on an ideology, so it's impossible to discern whether any relationships found within IQ data are real or a result of the calibration.

The ideology is well-meaning: that men and women have identical mean IQ of 100, and that the SD of IQ is 15. Basically, they periodically do tests, see which questions men and women tend to do better on, then adjust the amount and weighting of questions such that the average returns to 100 for both groups. Why men and women do better on different types of questions is largely irrelevant here. The fact is, that they shape the data through heavy-handed statistical measures to maintain 100 Mean and 15 SD based on a gender-based categorization. The choice of gender as the baseline variable to equalize is pretty damn arbitrary, if you think about it.

Can you imagine any physical science which could get away with such obvious data-manipulation to fit a preconceived theory as "psychometry" (iq testings field name) does?

Basically, for any arbitrary large-enough grouping you seem to be able to find an abritrary set of questions in which one group will do better than another group, or vice-versa. The ability to tune it so that things are equal, is also the ability to tune it such that men always score higher or women always score higher. The choice of how we calibrate IQ tests is purely a human choice made for sociopolitical reasons, not empirical justifications.

[Rolepgeek]

...IQ tests do have measurable and real predictive value, mostly when you measure for general intelligence factor. The brain is not moderated by a 'few' genes, however. It's more than likely that there are hundreds if not thousands of genes that affect intelligence. And when most traits and effects are like 50% genetic and 50% environmental, there's very little reason to believe this should be different for intelligence.

More, from what I remember they try to get that calibration for the largest number of people possible, and keep records for previous calibration standards, so that we know about things like the Flynn effect (IQ scores have been going up slowly over time despite genetic contributions decreasing slowly over time, believed to be the result of environment).
Biological sex really isn't that arbitrary of a dividing line, given the sexual dimorphism, hormones, etc.

Moreover, it's more useful for finding what environmental variables we can adjust in order to improve outcomes. Like how stopping the use of leader gasoline improved outcomes in cities considerably.
There's a means to acknowledge the ' unsavory' fact of how significant genetics are in development and behavior and characteristics and so on, without having to go full nazi. It's called "everyone has the same worth/value anyway". Finding reasons to discount real science because it might be used to justify horrible things is just...the wrong way to go about it. Evopsych isn't false just because some people use it to claim that being an asshole to women is okay. Social Darwinism doesn't prove survival of the fittest to be a fallacious model. Genetic or even racial differences in intelligence doesn't make racism okay; there is far too much noise at the individual level for it to be at all useful as a predictor there, and at the statistical level, it helps explain differences in outcomes and give us ways to help compensate. It doesn't make someone less of a person if they're not as intelligent.

Similarly, outcomes can be predicted on the large-scale statistical level from IQ/g factor but individual outcomes are incredibly variable.

Though as for ambition my life goal is to try and 'cure death' via generous application of bionanotechnology and genetic engineering of artificial symbiotic organisms. So :/

[Reelya]

And when most traits and effects are like 50% genetic and 50% environmental, there's very little reason to believe this should be different for intelligence.

That actually varies wildly based on the trait: the more critical for survival of a specified trait, the less genetic variation there is (in terms of how it affects the phenotype. genes which don't affect the phenotype as strongly can drift more). Basically, strong selection pressures correlate to traits with low genetic variability. Ones where it's 50/50 tend towards "cosmetic" differences (i.e. they could be functional, but they don't affect your reproduction chances). So it can be argued that the genes for e.g. hair color are going to have experienced genetic drift far more than the genes for intelligence.

Biological sex really isn't that arbitrary of a dividing line, given the sexual dimorphism

What I meant was that calibrating things so there's equality is arbitrary. It's an ideological statement that there's no dimorphism in a specified trait, which is not based on scientific principles, but what we wish to be true. If you think about it, "geekiness" could probably be calibrated, then the population divided 50/50, and we come up with brain-tests that calibrate so that more or less geeky people are equally "intelligent". e.g. there are definitely skills that less-geeky people can master much easier than the average "geek". Why do we automatically assume "geeks" are the smart ones, when there are a wide range of skills the average geek is terrible at compared to non-geeks.

The likely answer to that conundrum, is that there's no single "g factor" that makes sense. From what I've read, some researchers have managed to split cognition ability into ~3 separate skills, in a way that each one is completely statistically independent of the other two. If the "g factor" posited by IQ proponents really existed, such a split should be impossible. And there are definitely brain-related abilities that we're not measuring with "intelligence" tests at all, because IQ tests assume that cognitive symbol manipulation of some sort is the only sort of brain-ability that matters. What about people who have amazing reflexes and neuro-physical abilities? Aren't they smart too? Their brain might be optimized to process more sensory and motor skill data, but we're not measuring that: we're not taking into account differences in how each brain is optimizing it's finite amount of data processing.

What is much more promising is looking at the brain as making trade-offs between different abilities, and that some of these trade-offs are correlated with gender via itermediary systems (pre-natal testosterone is the prime suspect) so that is what allows IQ testers to calibrate a test that equalizes "g" between the genders.

Pre-natal testosterone research is really the promising lead here. It can explain the "expected" gender differences, but also why some people buck the trend. e.g a boyish girl could just be born that way, and isn't just a result of some fluke of environment. It's certainly a better fit for the data than saying you were just a certain way "because" you had a Y chromosome, or "because" you were socialized as a boy. Neither of those older theories can explain outliers very well: i.e if either "X/Y chromosome" or socialization is so completely overwhelming, why do some people turn out completely differently given the same inputs?

[Rolepgeek]

Not genetic variability within a population; how much variation is explained by genetics versus environment. This is found with mostly really complex things (like intelligence) and, as you said, iirc, 'cosmetic' things like what someone finds attractive. Obviously simple traits/traits where only a few genes are involved are going to be more affected by genetics than anything else.

And what I was saying for the dividing line is that because they're different they have to be measured on their own scales, which is why they both get set at 100. You're setting the mean for each population to be 100 because that's what's been decided will be used as the baseline, rather like temperature.

As for g factor; no, that's not how it works. First, here, second, a summary of what that says, iirc: basically, all those cognitive abilities tend to correlate with each other, and they almost by definition correlate with g factor. Furthermore, the higher the g factor, the more they correlate with each other. (not as in that's what the g factor measures but as in they've found that when people are smarter they tend to be smarter all around, basically, rather than savants).

Statistical data being able to be fooled around with isn't really that impressive, though. I don't say that with the intent of dismissing it, I'm just saying that when g factor has been a reliable metric for a long time, saying it can't possibly exist as a useful construct/measurement (because obviously it's not a physical thing, it's an abstracted measurement based on other measurements) because a few researchers managed to separate cognitive testing into multiple clusters that appear in their analysis to have no relation...I would need more evidence, essentially.

[Reelya]

Not genetic variability within a population; how much variation is explained by genetics versus environment

But if a gene doesn't vary within the population, then 100% of the variation in the trait it controls is explained by environment. Variation only makes sense at a population level. Individuals don't have "variation" in phenotype. And traits which are important for survival don't tend to have high variability at the genetic level: the genetic variation is reduced because having the trait is important. Therefore the differences due to genetic are reduced.

And if you're talking about the effect of "genetic variation" on one individual, then it's clear that enough change in the gene could set a trait from anywhere from 0 to 100% of the current trait. So how many base-pair flips exactly are considered to equal 50% of the influence of environment? If you're not talking about mapping real genetic variation in the population to changes in the phenotype, then you can cite any amount of arbitrary base-pair flips as being the "right" amount that equates to 50% variation in the trait.

As for G, some more recent studies have put to rest that such a factor even correlates at all with the sub-abilities, as long as you break up the sub-tasks correctly:
http://www.independent.co.uk/news/science/iq-tests-are-fundamentally-flawed-and-using-them-alone-to-measure-intelligence-is-a-fallacy-study-8425911.html

The researchers took a representative sample of 46,000 people and analysed how they performed. They found there were three distinct components to cognitive ability: short-term memory, reasoning and a verbal component.
...
“The results disprove once and for all the idea that a single measure of intelligence, such as IQ, is enough to capture all of the differences in cognitive ability that we see between people,” said Roger Highfield, director of external affairs at the Science Museum in London.

“Instead, several different circuits contribute to intelligence, each with its own unique capacity. A person may well be good in one of these areas, but they are just as likely to be bad in the other two,” said Dr Highfield, a co-author of the study published in the journal Neuron.

The scientists found that no single component, or IQ, could explain all the variations revealed by the tests. The researcher then analysed the brain circuitry of 16 participants with a hospital MRI scanner and found that the three separate components corresponded to three distinct patterns of neural activity in the brain.

So what they found is that there's literally no correlation between being good or bad at each of these three important mental abilities. Other tests might correlate with each other, hence you can call the overlapping correlation amount "g". But that's just because those tests were badly designed in the first place and didn't separate out individual factors properly.

[Rolepgeek]

...What? I was saying that the variability of the genes within a population is not what is being discussed. The variation of a trait/outcome caused by genes versus environment is what's used.

Whenever something says 'you just have to break up the 'X' correctly', I become suspicious, because of how easy it is to do p-hacking with things like that, and furthermore, this has been one study. Especially given the motivations inherent in certain ideologies to prove anything that implies differences between race, gender, and class. (Class is largely gonna be an environmental thing, honestly) The newest study, or the biggest, is not the 'best', that disproves all the others instantly. If there are metastudies along those lines, I would love to see them, because metastudies are often the only way to figure out anything at all about fields like these, where things are so complex and difficult to measure that single studies become incredibly noisy. Especially when you consider that MRIs are actually fairly unreliable for use in neuroscience, people're finding. See?

Like, what do they have that proves their tests are the ones that are properly designed, and the original ones weren't?


And about that motivation thing: literally this can have massive effects, which are really hard to prevent. The story/study I would relate except he does it better. Like, there's this established thing which is reliable in determining outcomes and useful for that purpose and has been found to be affected by environment in meaningful(if small) ways, and consistently so for many years, and a single study is going to disprove that? It opens the way for further study into that area, certainly, but science is slow and it needs more evidence than that to overturn scientific consensus.

[Reelya]

..What? I was saying that the variability of the genes within a population is not what is being discussed. The variation of a trait/outcome caused by genes versus environment is what's used.

But this is a nonsensical statement then.

If there's 0 variability in a gene within a population, then how can you attribute any of the measured variation to the gene? It's not a comprehensible claim.

Sure, you could talk about theoretical gene variation and the effect it would have on an individual, but then you have no way of calibrating what is an "average" amount of variation of that gene, which you can attribute to causing 50% of the variability in the outcome, since you're basing "average variation" on a purely theoretical basis, not population data.

[Rolepgeek]

Reelya. The variations of specific genes are not the concern here. Obviously it is greater than zero. But that is not what it is being measured. They are looking at how well genetic similarities versus environmental similarities correlate to similar outcomes, usually if not always using twin studies. How did you get '0 variability in gene' from 'not what is being discussed'?

I'm saying you're looking at the wrong part of this to look at what is actually used in these studies. They don't examine specific genes, they examine the overall genetic component of it, aka how heritable it is, controlling for factors like parenting.

[wierd]

You could isolate that using an animal model that is a close genetic relative, and examine the different nonhuman gene.

EG, one could explore the differences in some HOX genes between humans and chimpanzees, and say that "gene FOO is responsible for increased thickness of cortex in humans compared, which contributes to human intelligence (vs chimpanzee)", etc.

[Reelya]

Twin studies probably overestimate their effectiveness:
https://www.madinamerica.com/2013/03/the-trouble-with-twin-studies/

However, this “twins create their own environment” argument is a circular one, because twin researchers’ conclusion that identical pairs behave more similarly because they are more similar genetically is based on the assumption that identical pairs behave more similarly because they are more similar genetically. This means that twin researchers’ position that genetic factors explain the greater behavioral resemblance of identical twin pairs is, illogically, both a conclusion and a premise of the twin method. In defending the validity of the twin method, modern twin researchers refer to the premise in support of the conclusion, and then refer back to the conclusion in support of the premise, in a continuously circular loop of faulty reasoning.

Additionally, a research paper by the same author, where they showed that similarity in measurable environmental factors were in fact much higher for identical twins vs fraternal twins, undermining a core assumption of twin studies:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4411885/

Using results from prior twin studies, we tested if intraclass correlations for the following five categories of child social adversity are larger in identical than fraternal twins: bullying, sexual abuse, physical maltreatment, emotional neglect and abuse, and general trauma. Eleven relevant studies that encompassed 9119 twin pairs provided 24 comparisons of intraclass correlations, which we grouped into the five social exposure categories. Fisher’s z-test revealed significantly higher correlations in identical than fraternal pairs for each exposure category (z ≥ 3.53, p < 0.001).

... which explains why they have a "twins shape their own environment" theory in the first place: it hand-waves away the fact that the central assumption behind twin studies is provably false. The key assumption that the environments of identical and fraternal twins is the same is not consistent with the twin researcher's own data. "Twin Studies" therefore have the same huge methodological flaws as psychometry in general.

[Rolepgeek]

We're looking at/for what are essentially minute differences in intelligence between humans, wierd; an animal model is useful for finding what makes us intelligent compared to animals, say, but for the subtleties involved in the brain operating at peak efficiency, it's not going to cut it.

Reelya, again, is there more than one guy who puts forward those views? Looking at the website, his style of talking about it, and his responses to comments, gives me the impression that's he's not exactly unbiased. Again, it bears looking into more, but one guy vs. scientific establishments is how we got anti-vaxxers, at the extreme, and when he publishes it on a site that specifically mentions social justice in it's tagline (and social justice has a vested interest in disproving genetics playing a role in just about anything). Again, p-hacking is a thing, which is why metastudies are important. I see that he used multiple studies in his analysis, and so I'll ask this; does he at any point propose any other way to test for genetic factors in complex mental issues, or is his core proposal that there is no genetic factor and no way to test for one effectively?