3 ms·

So why do you believe someone with 180 IQ is that much smarter than 100, if you agree there are no metrics to back it up? Surely there would be something that c

by dahart 2mo ago

So why do you believe someone with 180 IQ is that much smarter than 100, if you agree there are no metrics to back it up? Surely there would be something that correlates with a two orders of magnitude increase in intelligence. Doesn’t the lack of such data suggest it’s actually not likely? Human head size reaches 80% by age two. IQ tests for 7 years olds are completely different tests and even different scales than IQ tests for adults. (See WPPSI vs WISC vs WAIS) The IQ scoring formula originally divided ‘mental age’ by chronological age to adjust (this is the Q in IQ), because it’s well known that a 10 year old knows more things and has a better memory than a 7 year old. Child geniuses are a much bigger problem for your argument than mine. If IQ measured the innate biological component of intelligence, then it’d be no surprise at all that you can find young outliers. But if intelligence is trained, then child geniuses would be meaningless. If intelligence is entirely trained, that means it’s entirely acquired during life, and so then everyone can become a “genius” with effort, and IQ would only be measuring motivation and not intelligence.

Xirdus 2mo ago
> So why do you believe someone with 180 IQ is that much smarter than 100, if you agree there are no metrics to back it up? Because that's how it works for most other learnable, trainable skills that aren't dependent on physical limits of the body, and I don't see why intelligence would be any different. Yes I'm still saying intelligence isn't physical. I know IQ is (used to be?) age-adjusted. I wasn't talking about IQ there, I'm talling about raw intelligence. Child prodigies beat adults at chess, at maths, at programming, you name it - and it's happening way more commonly than head/brain size variation would allow for it. I never said entirely trained. For somebody who just spent 10 whole paragrahs across 3 comments (+/- 50%, like I said I wasn't really reading) chastising me for not reading carefully, you sure fumbled this one. Think of intelligence like playing piano. Most people can't play piano. But most people can learn to play piano, to some extent. How good they are depends mostly on how much they've been learning, but everyone has some point they just can't push past through no matter how much they try. Usually that point is pretty low, way before the general audience is willing to pay for watching. For a tiny minority, that ceiling is much higher and the growth they get from practice is much faster too - we call these people talented or naturally gifted. And then there are the piano prodigies who absolutely shred the keyboard at the age of 5. There are enough of them that you can't explain it away with abnormal growth of any relevant body organ. That's what I mean when I say head circumference variation is too small in children to explain the number of child geniuses, if intelligence is entirely biological. IQ measures whatever intelligence happens to be. It's just an assumption that the kind of intelligence that helps with IQ tests is entirely biological, and only under that assumption IQ measures the biological component only. But that assumption is likely wrong, considering you can train to score better at IQ tests (it's a fact). Either intelligence isn't biological, or IQ metric sucks at isolating the biological component - feel free to assume the latter.
- dahart 2mo ago
  Oh you misunderstood me; I wasn't making any claims about what you said, I was only pointing out the logical conclusion that measuring acquired skills is not measuring intelligence, it would only be measuring effort & interest. I'm trying to point out why the training part of IQ or intelligence is the far less interesting part. The more trainable it is, the more nobody cares. What many people want to know is what's the capacity for intelligence; what's the ceiling on training. Think about it: the single most valid and important use case for intelligence testing is deciding how to adapt the educational curriculum for children. BTW I'm not the one repeatedly bringing up the 20% thing, I'm trying to drop it. You are arguing that intelligence is primarily trained and not primarily a physical trait, no? You did make a claim about a two orders of magnitude intelligence difference between IQs of 180 and 100, right? You seem to agree that "intelligence" is a combination of genetics and training, is that correct? And you believe the variance in IQ scores is mostly due to training and not genetics, or what? This is somewhat going in circles now. I've agreed multiple times that IQ scores are somewhat trainable, and you're ignoring the facts that 1) it's only a little bit trainable not a lot, and 2) that people generally tend to have a stable IQ measurement over their lifetime (adults moreso than kids), and 3) that the goal of IQ tests is to try to factor out training to some degree, which is why it used to divide by age and now measures only rank. Cultural bias was a huge a valid criticism of earlier IQ tests, and they've taken steps to avoid relying on knowledge. I'd certainly love to hear a more interesting argument or any evidence at all for you two orders of magnitude claim, but I can accept you're telling me your opinion and not something defensible. What we have agreed on is that IQ doesn't measure intelligence, it measures rank. One way or another, that fact could fully explain the outcome of the article we're commenting on. > Child prodigies beat adults at chess, at maths, at programming, you name it. FWIW, that's not really true in any meaningful way, and to some degree it contradicts what you're saying about training intelligence. The child outliers are not beating the adult outliers. Some children can outperform the average adult, but what has that got to do with anything? Some adults can also outperform the average adult. If children are outperforming adults, and they've had less training time, then how do you explain the superior performance? What is a naturally gifted child prodigy, if not evidence of physical traits?
  - 2mo ago

by dahart 2mo ago

Xirdus 2mo ago

> So why do you believe someone with 180 IQ is that much smarter than 100, if you agree there are no metrics to back it up? Because that's how it works for most other learnable, trainable skills that aren't dependent on physical limits of the body, and I don't see why intelligence would be any different. Yes I'm still saying intelligence isn't physical. I know IQ is (used to be?) age-adjusted. I wasn't talking about IQ there, I'm talling about raw intelligence. Child prodigies beat adults at chess, at maths, at programming, you name it - and it's happening way more commonly than head/brain size variation would allow for it. I never said entirely trained. For somebody who just spent 10 whole paragrahs across 3 comments (+/- 50%, like I said I wasn't really reading) chastising me for not reading carefully, you sure fumbled this one. Think of intelligence like playing piano. Most people can't play piano. But most people can learn to play piano, to some extent. How good they are depends mostly on how much they've been learning, but everyone has some point they just can't push past through no matter how much they try. Usually that point is pretty low, way before the general audience is willing to pay for watching. For a tiny minority, that ceiling is much higher and the growth they get from practice is much faster too - we call these people talented or naturally gifted. And then there are the piano prodigies who absolutely shred the keyboard at the age of 5. There are enough of them that you can't explain it away with abnormal growth of any relevant body organ. That's what I mean when I say head circumference variation is too small in children to explain the number of child geniuses, if intelligence is entirely biological. IQ measures whatever intelligence happens to be. It's just an assumption that the kind of intelligence that helps with IQ tests is entirely biological, and only under that assumption IQ measures the biological component only. But that assumption is likely wrong, considering you can train to score better at IQ tests (it's a fact). Either intelligence isn't biological, or IQ metric sucks at isolating the biological component - feel free to assume the latter.

dahart 2mo ago
Oh you misunderstood me; I wasn't making any claims about what you said, I was only pointing out the logical conclusion that measuring acquired skills is not measuring intelligence, it would only be measuring effort & interest. I'm trying to point out why the training part of IQ or intelligence is the far less interesting part. The more trainable it is, the more nobody cares. What many people want to know is what's the capacity for intelligence; what's the ceiling on training. Think about it: the single most valid and important use case for intelligence testing is deciding how to adapt the educational curriculum for children. BTW I'm not the one repeatedly bringing up the 20% thing, I'm trying to drop it. You are arguing that intelligence is primarily trained and not primarily a physical trait, no? You did make a claim about a two orders of magnitude intelligence difference between IQs of 180 and 100, right? You seem to agree that "intelligence" is a combination of genetics and training, is that correct? And you believe the variance in IQ scores is mostly due to training and not genetics, or what? This is somewhat going in circles now. I've agreed multiple times that IQ scores are somewhat trainable, and you're ignoring the facts that 1) it's only a little bit trainable not a lot, and 2) that people generally tend to have a stable IQ measurement over their lifetime (adults moreso than kids), and 3) that the goal of IQ tests is to try to factor out training to some degree, which is why it used to divide by age and now measures only rank. Cultural bias was a huge a valid criticism of earlier IQ tests, and they've taken steps to avoid relying on knowledge. I'd certainly love to hear a more interesting argument or any evidence at all for you two orders of magnitude claim, but I can accept you're telling me your opinion and not something defensible. What we have agreed on is that IQ doesn't measure intelligence, it measures rank. One way or another, that fact could fully explain the outcome of the article we're commenting on. > Child prodigies beat adults at chess, at maths, at programming, you name it. FWIW, that's not really true in any meaningful way, and to some degree it contradicts what you're saying about training intelligence. The child outliers are not beating the adult outliers. Some children can outperform the average adult, but what has that got to do with anything? Some adults can also outperform the average adult. If children are outperforming adults, and they've had less training time, then how do you explain the superior performance? What is a naturally gifted child prodigy, if not evidence of physical traits?
- 2mo ago