Thursday, June 14, 2018 ... Deutsch/Español/Related posts from blogosphere

Young folks born in 1991 are 5 IQ points dumber than older ones born in 1975

Norwegian researchers blame the retrograde Flynn effect on "adaptation to the environment"

We often complain about the declining intellectual skills, deteriorating education systems, dropping focus on hard sciences and knowledge with beef, decreasing curiosity, and on the contrary, increase of the percentage of bogus knowledge and indoctrination, virtue signaling, safe spaces, and political correctness. Many of us have surely thought about the causes and fixes for many particular manifestations of these trends.

A simple title summarizing a paper, e.g. RT's

Dumbing down? New study suggests young people's IQs are in decline
has persuaded me to think differently, more simply about phenomena such as a low number of very young string theorists. And the conclusion is that it may be futile to try to solve any similar problem in isolation. It may be futile to fight against lousy "science journalists" and similar people because they're not really the isolated "cause" of these trends – instead, they're just symptoms, among millions of others.

What's going on? Two Norwegian IQ experts, Bernt Bratsberg and Ole Rogeberg (and yes, Scandinavia was a hotbed of lots of eugenics a century ago), have published a new paper in PNAS,
Flynn effect and its reversal are both environmentally caused,
that has looked at the mandatory IQ tests in Norway that have been performed on 18- or 19-year-old men (or big boys) for decades in order to map the army reserves. First, let's ask: Why wasn't the Idiocracy film shot before 2006? It's because people actually observed the Flynn effect, named after a psychologist from New Zealand.

The effect simply says that a "fixed method to measure the IQ" produces the IQ scores (of British kids) that are about 15 points higher half a century after the Second World War than they were right after the Second World War. Prosperity after the war and better nutrition were among the factors that were often credited for the rising IQ. Note that at every moment, the IQ axis should be nonlinearly reparameterized so that the existing world population has the IQ distributed as the normal distribution centered at 100 with the standard deviation of 15. So if you update your IQ definition in this way, you won't see any sharpening or dumbing down.

But the effect was there – a whopping 15 points in half a century.

The Norwegian folks have looked at the IQ tests, found that the kids born in 1991 were 5 IQ points dumber than those born in 1975, and checked that this rather substantial change isn't caused by mutations or demographics change because one may reproduce the same changes by averaging the trends "inside the families". So it's really the "fathers" born around 1975 that are 5 IQ points smarter than the "sons" born around 1991. To be sure, in this way, they also claim to have excluded the explanation (cited in Idiocracy) that "dumb Norwegian families have reproduced more quickly than the smart ones".

Thank God that they did it and said it – otherwise we would have to immediately ask whether the effect isn't simply due to some mass immigration to Norway. Incidentally, it's easy to see that the minorities of Norway just can't possibly be large enough for such an effect, anyway. The population of Norway is some 5.3 million and some 6% are non-Western folks. If you reduce the IQ by 6% of the "West-non-West difference", let say 6% of "100-80=20", you only reduce it by 6% of 20 which is 1.2 points. It would be much less than the observed 5 points even if the whole exotic population appeared suddenly, which it didn't.

OK, so the decrease of the IQ is "environmental" or "social" in this sense. People don't need to be intelligent anymore so the organisms adapt and suppress the intelligent behavior needed in IQ tests – and lots of analogous activities. When I give you such a particular description of the cause, it may already be too suggestive. The only thing that was really established is that the effect cannot be attributed to changes of the DNA of the sample. So it's "more social than biological". But "what kind of social" cannot be really determined from their analysis, I think.

However, these 5 points (one-third of the standard deviation) have far-reaching consequences.

Look at this simplified table of the normal distribution which will be enough for our rough purposes. In the third column, "or 1 in \(p\)", you see which fraction of the people is outside the \(n\)-sigma band going to both directions. If you multiply the column by two, you get the percentage of the people whose IQ is higher than the mean by more than \(n\) sigma. For example, 1 person in 6.3 is smarter than IQ=115 (by more than 1 standard deviation above the mean), OK?

Let's say that a certain occupation requiring a certain type of occupation needs the IQ above 115. One got 1/6 of the people over there. But when your population dumbs down by 5 points, you will only find roughly 1/10 (one-third of the way from 1/6 to 1/44, in some nonlinear sense) of the people who can do it. Already with the selection that only requires one standard deviation, the drop by 5 IQ points basically halves the number of people who get there.

The reduction of the pool becomes much more extreme if you go to a higher level of selection.

I have used this estimate a few times, most recently two days ago – apologies for that high frequency. But I think it's fair to estimate to say that some 10,000 people have the natural IQ to turn themselves into string theorists, and 2,000 out of them have actually done the steps to join that group.

The required edge in this selection is something close to a 5-sigma effect. By the table, 1 person in 3.5 million has the IQ above 175 (more than 5 standard deviations above the mean), which would give you 2,000 people with the abilities in the world. And 1 in one billion (more than 6 standard deviations above the mean) has the IQ above 190 (so that's some 7 people in the world, and it becomes a matter of subtleties in definitions and measurements to decide whether the number really "should be" exactly seven at each moment or whether the Gauss-like definition of the IQ still allows this number to fluctuate).

OK, so to have the abilities to seamlessly do string theory, one needs to be roughly 4.67 standard deviations above the mean – the IQ should better be above 170. We rarely hear about the people with the IQ above 170 but the basic rules of the normal distribution imply that some 7 people should be above 190, as I mentioned.

What does the decrease by 5 points, one-third of the standard deviation, does to this selection? You move by some 1/3 of the way from the selection "1 person in 3.5 million" (5 sigma) away from "1 person in 1 billion" (6 sigma), i.e. in the opposite direction. The ratio of 1 billion and 3.5 million is some 300, and the third root of it is between 6 and 7.

So if the standard deviation remains the same, the edge of 4.67 sigma becomes 5 sigma in the dumber sample and the downward shift by 5 IQ points is predicted to reduce the number of string theory-ready youth by a factor of 6 or 7. Maybe it's a bit below 6 – you may calculate the exact figure. Nevertheless, it's a dramatic decrease that you won't overlook. Sadly, it may be compatible with the observations.

Whenever you're tempted to localize the blame and say that "someone particular is doing something wrong", think again. If the grad schools really get robbed of the talent so much in one generation, they just can't beat the basic numbers. The decrease isn't quite enough to go from "studying string theory textbooks" to "studying popular books attacking beauty in physics", but it's not too much smaller, either.

So what or who is the culprit and what is the fix? The culprit is the overall societal atmosphere combined with all the pressures – and especially the non-existence of pressures – that exists around us or doesn't exist around us. A huge fraction of the people must be considered a part of the problem. Maybe it's not just the people. Maybe it's enough to live in a world in which machines really think on our behalf so we don't need to do it. Kids' organisms quickly realize it and they just can't solve the progressive matrices (and other, more meaningful things that nevertheless require some of the similar skills).

Of course, the two Norwegian researchers realize the difference between the crystallized intelligence and the fluid intelligence. I think that good IQ tests should really measure just the latter – and the latter should be minimally affected by "how the kids are being trained and educated" since their childhood. But is there some influence, anyway? Did you train things like "progressive matrices" before you were given them for the first time? And if the decrease is due to this training, why don't the rather intelligent, demanding games on tablets etc. that kids see these days improve their "fluid intelligence" more than we were helped by seemingly stupider toys in the late 1970s?

OK, forget about string theory. If that decrease continued like that, the mankind could solve more existential problems rather soon – and Idiocracy sketched them in detail. Note that 5 points per 15 years is 1 point per 3 years. You may remind yourself about the definition of idiots, imbeciles, morons, and cretins. Cretins are non-quantitative (a thyroid problem), morons have IQ between 50 and 69, imbeciles between 20 and 49, and idiots are below 20. I think that the word "imbecile" sounds the harshest but by the definitions, it's actually in between the "morons" (mild) and "idiots" (worst).

At the rate of 1 IQ point per 3 years, you need about a century for one-half of the mankind to become at least morons (below 70) and about 150 years for one-half of the mankind to become at least imbeciles (below 50). As you can see, the real world trends seem faster than those in Idiocracy by a factor of 3-5. We should better invent ways to slow down the dumbing down by a factor of 3-5 to realize at least the optimistic scenario in Idiocracy in which most people were morons and imbeciles only after the rather distant year 2500.

But the proliferation of morons and imbeciles has already sufficiently advanced so that many more people already fight global warming than the actual real threats. You know, I am intuitively repelled by eugenics as a science – and even more so by eugenicist policies. This isn't about the DNA, the study concluded, so we should fix the "environmental" factors that have caused the reversal in the first place. Is it possible at all? Isn't the dumbing down inevitable – and an explanation of the Fermi problem (why don't we see lots of extraterrestrials)?

Add to Digg this Add to reddit

snail feedback (0) :

(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','//','ga'); ga('create', 'UA-1828728-1', 'auto'); ga('send', 'pageview');