4 ms·

Well, you made the arguably far less substantiated claim that intelligence is a physical attribute and thus bounded to +/-20% range for most of the population..

by Xirdus 2mo ago

Well, you made the arguably far less substantiated claim that intelligence is a physical attribute and thus bounded to +/-20% range for most of the population... And that it can't be trained, despite all the evidence to contrary. If you're no longer interested defending that position, that's okay.

dahart 2mo ago
Speaking of unsubstantiated claims… what evidence is there that intelligence can be trained? You’re confusing intelligence with IQ tests. I made no such claim that intelligence is bounded to 20%, you seem to have misunderstood. I pointed out the fact that IQ tests are by design not tied to nor proportional to actual intellectual capacity. If you disagree, maybe you can provide some relevant information? So far your objections have been either irrelevant or factually incorrect.
- Xirdus 2mo ago
  Okay, let's step back a little. When you said "there are good reasons to assume IQ works the same way [as the hypothetical HQ]" - what exactly did you mean by that? Did you mean to imply the ratio between intelligence of someone with IQ 100 and IQ 180 is smaller than a factor of 1.8, or did you mean something else? The evidence that intelligence can be trained is that every metric of intelligence ever invented is consistently going up when the subjects spend time to prepare themselves before being measured. IQ tests, school grades, chess performance, math olympiads, you name it. Sure, you can always claim that all of those are skills separate from intelligence and that's why becoming better is possible - but then how do you explain that being better at one of these automatically makes you better at all the others, if the overall intelligence stays the same? Also, I said the range is +/-20% already in my very first reply. Why didn't you correct me back then? Why did you wait until now?
  - dahart 2mo ago
    > what exactly did you mean by that? Great question. My primary point was that the IQ scale and absolute intelligence aren’t the same things, and people often assume they are (this article’s premise, for example), or even worse many people assume that absolute intelligence goes up exponentially along with the rarity of the IQ ranking. Height was simply an example of how there’s an obvious discrepancy between standard deviations and the underlying absolute metric; the rarity of one person’s height says literally nothing about how tall they are, and yet people assume without any justification whatsoever that the rarity of someone’s IQ score does say something about how smart they are. All human physical characteristics have a relatively narrow range, and they don’t increase exponentially. Running is another example - Usain Bolt can run faster than 99.9999% of people, and yet he’s only 10% faster than the average competitive sprinter and maybe ~50% faster than the average person not trained in running. You’ll see similar distributions with weight lifting. The number of brain cells doesn’t grow without bound - we all have similar amounts. The construction of the brain cells and the structure of the brain isn’t wildly different from person to person. Most skulls are approximately the same size. Most eyeballs and ears and arms and legs all vary by about the same amount height and head size varies. So, there really is good reason to start with the hypothesis that raw biological cognitive capacity and speed don’t vary by a lot more than height does, and - importantly - that they all top out at the upper end of the distribution. > Did you mean to imply the ratio between intelligence of someone with IQ of 100 and IQ 180 is smaller than a factor of 1.8, or did you mean something else? I meant that from the start, assuming the factor is 1.8 is a very bad assumption. That’s simply not how standard deviations work. Yes, please consider the possibility that the cognitive capacity of IQ 180 is less than 1.8 times more than that of IQ 100. If intelligence has a similar distribution to other physical characteristics, then difference between the mean and the first standard deviations is much larger than the difference between the first and second standard deviations. Translation: the jump from 130 to 160 is probably much smaller than the jump from 100 to 130. And going from 160 to 190 IQ, there is likely a negligible, unmeasurable difference in absolute intelligence. This is the opposite of the popular conceptions of IQ, right? > The evidence that intelligence can be trained is that every metric of intelligence ever invented is consistently going up Again, you are confusing raw intelligence with psychometric tests. Yes you can train for an IQ test. No, you cannot train intelligence. ‘Intelligence’ in this context, the kind we are discussing, is defined as the biological component of cognitive capacity that cannot be trained. Intelligence and skill are two different things, go ahead and google that question. The fact that IQ tests are not currently able to differentiate between intelligence and skill is a well known shortcoming of IQ tests, and that relates directly to my initial point: IQ is not a great measure and assuming that IQ score and happiness would have a correlation anywhere near 1 is a very bad assumption. > Sure, you can always claim that all of those are skills separate from intelligence Yes, yes I can, since it’s true. Therefore, given that fact, the question you asked is one for you to ponder and answer. What is the source of the claim that being better at one “automatically makes you better at all the others”? Does taking IQ tests give you more brain cells or increase your cognitive speed? You’re hand-waving this training argument and ignoring the fact that people who train for IQ tests still top out and hit a wall regardless of the amount of practice, they don’t increase their scores indefinitely. If intelligence could be trained, why isn’t everyone a so-called high IQ genius? > Why didn’t you correct me back then? Hehe I don’t need to argue every wrong thing you say, and doing so doesn’t make for good conversation. Your assumptions only reflect on you, not me. I assumed you’d eventually re-read the comment you replied to and figure out that using 20% in your argument was straw man given that I actually gave an example of a 50% variance in height in that same comment, and that I only used height as an analogy, not a claim about intelligence. When people seem to misunderstand my comments, whether it’s a mistake or intentional, I prefer to try to clarify my broader intent and not nitpick every word. Rebutting every single thing tends to escalate arguments and send the impression that you’re reflexively arguing rather than thinking, and seeking disagreement rather than agreement. I try to follow the HN guidelines, and I’m by no means always perfect, but they are pretty good guidelines, right?
    - Xirdus 2mo ago
      > Hehe I don’t need to argue every wrong thing you say No you don't. But if it's a central point of my argument, and ESPECIALLY when you throw it in my face later on that I based my argument around it - it feels like I wasted a lot of time and effort for nothing, and makes me wonder how much else of what I'm saying is a similar waste of time because you refrained from telling me I'm wrong about something else similarly important. Makes me hesitant to put too much effort into this conversation. > I meant that from the start, assuming the factor is 1.8 is a very bad assumption. I agree with the literal words - exactly 1.8 is almost certainly the wrong answer. But where we differ is that you believe it's less than 1.8, and I believe it's more than 100. A person with IQ 180 is likely two orders of magnitude more intelligent in absolute terms than someone with IQ 100. On the other hand, someone with IQ 60 is only marginally less intelligent than someone with IQ 100. It's because intelligence is not a physical attribute, but closer to a skill. Usain Bolt is only a little faster than a typical adult because speed is capped by the physical limits of the human body. Intelligence is not, because it's not a physical attribute. And if you want to continue insisting it is a physical attribute, I'd like to ask you for some evidence that it's actually the case, rather than assuming it a priori. As far as I know, neither brain size nor the number of neurons nor synapses is correlated with IQ.
      - dahart 2mo ago
        > it feels like I wasted a lot of time […] Makes me hesitant to put too much effort into this conversation Hahaha, buddy, this is very funny. Lack of effort is what caused your problem in the first place. I’m super surprised you’re digging in your heels on this. If it was me, I’d drop it, since it’s not a good look, but suit yourself. You got something wrong. So what? Own it. Fix it. It isn’t my responsibility to correct you, it’s yours. You did indeed waste a tiny little bit of time by arguing and jumping to conclusions before reading carefully. ;) And that’s okay, which is why I ignored it. I’ve wasted time making mistakes too. There’s nothing wrong with making mistakes. How you deal with mistakes is revealing though, and blaming others for your mistakes isn’t typically very productive here, even if some politicians seem to get away with it. :P I do honestly appreciate that your previous comment tried to de-escalate a bit. Asking what I meant was a positive step, and asking that earlier would have saved you some time. Always assuming good faith and making sure you read carefully and understood everything is a good thing to learn to do on HN, that’s what the guidelines are trying to hint at. I have learned this lesson myself the hard way, multiple times, FWIW. > Intelligence is not, because it’s not a physical attribute Well okay this is our core problem. We have been talking about different things. And, I need to concede on this point that you are right. If I google “is intelligence a physical trait” the results I get say: no. In my defense, I thought I was careful to qualify that what I was talking about earlier as the biological component of intelligence, but I’m okay admitting I’ve made some mistakes here. It is worth noting that “Intelligence is one of the most heritable behavioural traits” and that “studies have consistently shown that genetic influence on individual differences in intelligence is substantial”. https://www.nature.com/articles/mp2014105 https://www.nature.com/articles/mp2014105 > A person with IQ 180 is likely two orders of magnitude more intelligent in absolute terms than someone with IQ 100. There is zero scientific basis for that claim. This is exactly what I was talking about from the top. This popular interpretation of extremely high IQ scores substantially overstates what can legitimately be inferred from the numerical scores. That’s why I call it a misconception. IQ only directly measures how unusual someone’s intelligence is, not how good it is. IQ is at best only moderately good at predicting cognitive performance on some tasks, and whether it predicts anything is highly dependent on the type of task. And after reading about it more, I’m going to correct myself and agree that IQ does not predict cognitive brain capacity, and we don’t yet have anything that does. But scientists are looking for that. Anyway, this belief you have that 180 IQ vs 100 IQ is 100x more intelligent - whatever more intelligent actually means - is extremely extreme!! Why do you believe this, and what evidence is there of your claim? You’re absolutely right that I disagree with you there. But I will read and consider any supporting links or data you can provide, and I am open to changing my mind if the data is compelling. There’s a reason that IQ is considered outdated now, that scientists do not accept that intelligence is 1-dimensional, and that cognitive scientists now talk about multiple intelligences, and use g instead. https://en.wikipedia.org/wiki/G_factor_(psychometrics) https://en.wikipedia.org/wiki/G_factor_(psychometrics) > As far as I know, neither brain size nor the number of neurons nor synapses is correlated with IQ. Human brain size does correlate with IQ somewhat. Now you know. https://en.wikipedia.org/wiki/Brain_size#Intelligence https://en.wikipedia.org/wiki/Brain_size#Intelligence https://pmc.ncbi.nlm.nih.gov/articles/PMC7440690/ https://pmc.ncbi.nlm.nih.gov/articles/PMC7440690/ https://www.scientificamerican.com/article/does-brain-size-matter1/ https://www.scientificamerican.com/article/does-brain-size-m...

by Xirdus 2mo ago

dahart 2mo ago

Speaking of unsubstantiated claims… what evidence is there that intelligence can be trained? You’re confusing intelligence with IQ tests. I made no such claim that intelligence is bounded to 20%, you seem to have misunderstood. I pointed out the fact that IQ tests are by design not tied to nor proportional to actual intellectual capacity. If you disagree, maybe you can provide some relevant information? So far your objections have been either irrelevant or factually incorrect.

Xirdus 2mo ago
Okay, let's step back a little. When you said "there are good reasons to assume IQ works the same way [as the hypothetical HQ]" - what exactly did you mean by that? Did you mean to imply the ratio between intelligence of someone with IQ 100 and IQ 180 is smaller than a factor of 1.8, or did you mean something else? The evidence that intelligence can be trained is that every metric of intelligence ever invented is consistently going up when the subjects spend time to prepare themselves before being measured. IQ tests, school grades, chess performance, math olympiads, you name it. Sure, you can always claim that all of those are skills separate from intelligence and that's why becoming better is possible - but then how do you explain that being better at one of these automatically makes you better at all the others, if the overall intelligence stays the same? Also, I said the range is +/-20% already in my very first reply. Why didn't you correct me back then? Why did you wait until now?
- dahart 2mo ago
  > what exactly did you mean by that? Great question. My primary point was that the IQ scale and absolute intelligence aren’t the same things, and people often assume they are (this article’s premise, for example), or even worse many people assume that absolute intelligence goes up exponentially along with the rarity of the IQ ranking. Height was simply an example of how there’s an obvious discrepancy between standard deviations and the underlying absolute metric; the rarity of one person’s height says literally nothing about how tall they are, and yet people assume without any justification whatsoever that the rarity of someone’s IQ score does say something about how smart they are. All human physical characteristics have a relatively narrow range, and they don’t increase exponentially. Running is another example - Usain Bolt can run faster than 99.9999% of people, and yet he’s only 10% faster than the average competitive sprinter and maybe ~50% faster than the average person not trained in running. You’ll see similar distributions with weight lifting. The number of brain cells doesn’t grow without bound - we all have similar amounts. The construction of the brain cells and the structure of the brain isn’t wildly different from person to person. Most skulls are approximately the same size. Most eyeballs and ears and arms and legs all vary by about the same amount height and head size varies. So, there really is good reason to start with the hypothesis that raw biological cognitive capacity and speed don’t vary by a lot more than height does, and - importantly - that they all top out at the upper end of the distribution. > Did you mean to imply the ratio between intelligence of someone with IQ of 100 and IQ 180 is smaller than a factor of 1.8, or did you mean something else? I meant that from the start, assuming the factor is 1.8 is a very bad assumption. That’s simply not how standard deviations work. Yes, please consider the possibility that the cognitive capacity of IQ 180 is less than 1.8 times more than that of IQ 100. If intelligence has a similar distribution to other physical characteristics, then difference between the mean and the first standard deviations is much larger than the difference between the first and second standard deviations. Translation: the jump from 130 to 160 is probably much smaller than the jump from 100 to 130. And going from 160 to 190 IQ, there is likely a negligible, unmeasurable difference in absolute intelligence. This is the opposite of the popular conceptions of IQ, right? > The evidence that intelligence can be trained is that every metric of intelligence ever invented is consistently going up Again, you are confusing raw intelligence with psychometric tests. Yes you can train for an IQ test. No, you cannot train intelligence. ‘Intelligence’ in this context, the kind we are discussing, is defined as the biological component of cognitive capacity that cannot be trained. Intelligence and skill are two different things, go ahead and google that question. The fact that IQ tests are not currently able to differentiate between intelligence and skill is a well known shortcoming of IQ tests, and that relates directly to my initial point: IQ is not a great measure and assuming that IQ score and happiness would have a correlation anywhere near 1 is a very bad assumption. > Sure, you can always claim that all of those are skills separate from intelligence Yes, yes I can, since it’s true. Therefore, given that fact, the question you asked is one for you to ponder and answer. What is the source of the claim that being better at one “automatically makes you better at all the others”? Does taking IQ tests give you more brain cells or increase your cognitive speed? You’re hand-waving this training argument and ignoring the fact that people who train for IQ tests still top out and hit a wall regardless of the amount of practice, they don’t increase their scores indefinitely. If intelligence could be trained, why isn’t everyone a so-called high IQ genius? > Why didn’t you correct me back then? Hehe I don’t need to argue every wrong thing you say, and doing so doesn’t make for good conversation. Your assumptions only reflect on you, not me. I assumed you’d eventually re-read the comment you replied to and figure out that using 20% in your argument was straw man given that I actually gave an example of a 50% variance in height in that same comment, and that I only used height as an analogy, not a claim about intelligence. When people seem to misunderstand my comments, whether it’s a mistake or intentional, I prefer to try to clarify my broader intent and not nitpick every word. Rebutting every single thing tends to escalate arguments and send the impression that you’re reflexively arguing rather than thinking, and seeking disagreement rather than agreement. I try to follow the HN guidelines, and I’m by no means always perfect, but they are pretty good guidelines, right?
  - Xirdus 2mo ago
    > Hehe I don’t need to argue every wrong thing you say No you don't. But if it's a central point of my argument, and ESPECIALLY when you throw it in my face later on that I based my argument around it - it feels like I wasted a lot of time and effort for nothing, and makes me wonder how much else of what I'm saying is a similar waste of time because you refrained from telling me I'm wrong about something else similarly important. Makes me hesitant to put too much effort into this conversation. > I meant that from the start, assuming the factor is 1.8 is a very bad assumption. I agree with the literal words - exactly 1.8 is almost certainly the wrong answer. But where we differ is that you believe it's less than 1.8, and I believe it's more than 100. A person with IQ 180 is likely two orders of magnitude more intelligent in absolute terms than someone with IQ 100. On the other hand, someone with IQ 60 is only marginally less intelligent than someone with IQ 100. It's because intelligence is not a physical attribute, but closer to a skill. Usain Bolt is only a little faster than a typical adult because speed is capped by the physical limits of the human body. Intelligence is not, because it's not a physical attribute. And if you want to continue insisting it is a physical attribute, I'd like to ask you for some evidence that it's actually the case, rather than assuming it a priori. As far as I know, neither brain size nor the number of neurons nor synapses is correlated with IQ.
    - dahart 2mo ago
      > it feels like I wasted a lot of time […] Makes me hesitant to put too much effort into this conversation Hahaha, buddy, this is very funny. Lack of effort is what caused your problem in the first place. I’m super surprised you’re digging in your heels on this. If it was me, I’d drop it, since it’s not a good look, but suit yourself. You got something wrong. So what? Own it. Fix it. It isn’t my responsibility to correct you, it’s yours. You did indeed waste a tiny little bit of time by arguing and jumping to conclusions before reading carefully. ;) And that’s okay, which is why I ignored it. I’ve wasted time making mistakes too. There’s nothing wrong with making mistakes. How you deal with mistakes is revealing though, and blaming others for your mistakes isn’t typically very productive here, even if some politicians seem to get away with it. :P I do honestly appreciate that your previous comment tried to de-escalate a bit. Asking what I meant was a positive step, and asking that earlier would have saved you some time. Always assuming good faith and making sure you read carefully and understood everything is a good thing to learn to do on HN, that’s what the guidelines are trying to hint at. I have learned this lesson myself the hard way, multiple times, FWIW. > Intelligence is not, because it’s not a physical attribute Well okay this is our core problem. We have been talking about different things. And, I need to concede on this point that you are right. If I google “is intelligence a physical trait” the results I get say: no. In my defense, I thought I was careful to qualify that what I was talking about earlier as the biological component of intelligence, but I’m okay admitting I’ve made some mistakes here. It is worth noting that “Intelligence is one of the most heritable behavioural traits” and that “studies have consistently shown that genetic influence on individual differences in intelligence is substantial”. https://www.nature.com/articles/mp2014105 https://www.nature.com/articles/mp2014105 > A person with IQ 180 is likely two orders of magnitude more intelligent in absolute terms than someone with IQ 100. There is zero scientific basis for that claim. This is exactly what I was talking about from the top. This popular interpretation of extremely high IQ scores substantially overstates what can legitimately be inferred from the numerical scores. That’s why I call it a misconception. IQ only directly measures how unusual someone’s intelligence is, not how good it is. IQ is at best only moderately good at predicting cognitive performance on some tasks, and whether it predicts anything is highly dependent on the type of task. And after reading about it more, I’m going to correct myself and agree that IQ does not predict cognitive brain capacity, and we don’t yet have anything that does. But scientists are looking for that. Anyway, this belief you have that 180 IQ vs 100 IQ is 100x more intelligent - whatever more intelligent actually means - is extremely extreme!! Why do you believe this, and what evidence is there of your claim? You’re absolutely right that I disagree with you there. But I will read and consider any supporting links or data you can provide, and I am open to changing my mind if the data is compelling. There’s a reason that IQ is considered outdated now, that scientists do not accept that intelligence is 1-dimensional, and that cognitive scientists now talk about multiple intelligences, and use g instead. https://en.wikipedia.org/wiki/G_factor_(psychometrics) https://en.wikipedia.org/wiki/G_factor_(psychometrics) > As far as I know, neither brain size nor the number of neurons nor synapses is correlated with IQ. Human brain size does correlate with IQ somewhat. Now you know. https://en.wikipedia.org/wiki/Brain_size#Intelligence https://en.wikipedia.org/wiki/Brain_size#Intelligence https://pmc.ncbi.nlm.nih.gov/articles/PMC7440690/ https://pmc.ncbi.nlm.nih.gov/articles/PMC7440690/ https://www.scientificamerican.com/article/does-brain-size-matter1/ https://www.scientificamerican.com/article/does-brain-size-m...