Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Increased intelligence is a myth so far (2014)

ncbi.nlm.nih.gov

Post reply on HN
Print view View original

41–46 of 46 posts

Re: Increased intelligence is a myth so far (2014)

#41

Post by foldr »

maxander wrote: post #38
isoos wrote: post #32

Earlier quoted context omitted.

The Polgár sister's story backs you in that: "Polgár and her two older sisters, Grandmaster Susan and International Master Sofia, were part of an educational experiment carried out by their father László Polgár, in an attempt to prove that children could make exceptional achievements if trained in a specialist subject from a very early age.[11] "Geniuses are made, not born," was László's thesis. He and his wife Klára…

Which begs the question, when people are willing to commit their children to this kind of super-intensive training regimen at such an early age, why on Earth are they doing it for games or music and not something (if you will excuse a crass way of saying it) useful ? Imagine a physicist or mathematician or engineer tutored in their discipline from age five! And imagine if those were as common as the little kids one s…

Quite a lot of physics and mathematics is less useful than playing the piano, since it is equally lacking in practical application and doesn't even give other people pleasure.

Re: Increased intelligence is a myth so far (2014)

#42

Post by Moshe_Silnorin »

rtl49 wrote: post #37
Moshe_Silnorin wrote: post #17

Earlier quoted context omitted.

>The most likely approaches for increasing the intelligence of humans I agree, given the gestation time of a human, AI may well come first. Even if we perfected iterated embryo selection today, 18 years is a lot of time and perhaps AI would be developed before the super geniuses reach peak productivity. >Also I don't think there's an iq 200 set of genes, there's probably trade offs that would prevent this a brain has…

Bear in mind that that each of the "high-IQ-associated genes" (alleles) is responsible for the synthesis of a polypeptide, and that the phenotypes produced by their interactions can be extremely difficult to predict. I believe the studies you have in mind, one of which appeared on HN last week, find that high IQ is not associated with any small set of highly unusual alleles (i.e. "freak mutations"). This isn't to say…

I'm getting this from Hsu. He seems to think this will not be an issue: http://arxiv.org/abs/1408.3421

Re: Increased intelligence is a myth so far (2014)

#43

Post by mannykannot »

wodenokoto wrote: post #28

I feel like the whole argument is based on "your true IQ is the highest score you'll ever receive on a test", which is a statement that by definition would make improvements in IQ impossible.[0] I still don't understand why, if everybody in the control group only change around the error ratio and the test group on average advances beyond the error ratio, that doesn't show an improvement. According to the authors what…

> I feel like the whole argument is based on "your true IQ is the highest score you'll ever receive on a test", which is a statement that by definition would make improvements in IQ impossible.

I am no statistician, but I think there is more to the argument than that. My take is that Haier is saying that IQ is not a ratio scale, and you cannot gauge the significance of a result by treating it as if it is.

Nevertheless, I get the impression he is trying to stretch the importance of this observation. He writes 'Nonetheless, the main point is that to make the most compelling argument that intelligence increases after an intervention, a ratio scale of intelligence is required' (my emphasis), which raises the question of whether there is not a not-quite-so-compelling argument to be made without such a metric.

FWIW, I don't have a horse in this race as I am on the fence over whether IQ is as fundamental a measure as its proponents insist.

Re: Increased intelligence is a myth so far (2014)

#45

Post by gwern »

jimrandomh wrote: post #6
bordercases wrote: post #4

Earlier quoted context omitted.

Don't tell Gwern that. http://www.gwern.net/DNB%20FAQ#moody-2009-re-jaeggi-2008

Huh, good catch. The too-short time limit issue was a methodological issue I had definitely not thought to check for. (Haier doesn't mention it.)

There are actually several distinct issues with Jaeggi 2008. The speeding is a concern but it increasing looks like it's not driving the score increase. The passive control group is a big issue (although Jaeggi et al, via Au, are doing their best to fight to the grim end and deny it as long as possible). As is the subsamples using different IQ tests.

Finally, as Haier spends a lot of time explaining, pre-post test gains is fundamentally invalid as a way of proving testing increases the g-factor, because such gains are necessary but not sufficient; an increase in intelligence should show up as a gain, but many non-increases will also show up as a gain because despite the best efforts, IQ tests do not measure solely intelligence.

I'm increasingly understanding that this is as big a problem for n-back as the passive control group inflation. Many attempts to increase intelligence in the past, some where there was no dispute at all that test scores did increase, have failed as soon as a latent variable analysis was applied; I'm not aware of any which passes. All of the interventions had increased scores on only some subtests, and there were no general gains; they had not increased intelligence, only taught a specific subdomain. So our prior is highly against n-back being the exception. Worse, several studies have already tried to apply the latent variable analysis, and they all indicate that n-back flunked. So our already exceedingly skeptical prior gets reinforced by evidence of no effect on intelligence, whatever one subtest like a matrix test might say. (If Au or Jaeggi have addressed this, I have not seen it yet. They prefer to concentrate on the fact that the test scores increase and that you can cast doubt on the active/passive criticism by trying to correlate with whether a study was done in the USA or not, or whether the control group gained enough pre/post.)

You can see some links discussing the topic, including Haier's methodological argument here, in point #3 in http://www.gwern.net/DNB%20meta-analysis#analysis

Re: Increased intelligence is a myth so far (2014)

#46

Post by gwern »

angelbob wrote: post #21

The abstract mostly seems to state "increased intelligence is currently too difficult to measure, so we wouldn't know." Which is fair. But "is a myth so far" suggests disproof, not proof that it's not currently true. As they repeatedly show, it would be very difficult for them to prove that, since they'd have to usefully measure delta in intelligence.

> Which is fair. But "is a myth so far" suggests disproof, not proof that it's not currently true. As they repeatedly show, it would be very difficult for them to prove that, since they'd have to usefully measure delta in intelligence.

The burden of proof should be on anyone who claims to have an intelligence boosting method. The field is littered with hucksters who have claimed to succeeded; this was true in the '60s when Jensen first wrote on it, and it's still true now. It's the cold-fusion of psychology (something which is theoretically possible, would be immensely valuable practically, and has never panned out when examined rigorously).

Anyway, Haier's paper is not intended to be comprehensive, and we can point to some examples of investigating claimed intelligence increases and seeing that they indeed turn in the predicted null effect on intelligence: te Nijenhuis et al 2007, te Nijenhuis et al 2014, etc. See my other comment with link to citations & fulltexts.

Post reply on HN