On Chris Langan, et. al., and "IQ"

from a post reply thread on the Seeds Of Science page

Prompted as a reply to this mention of the mental abilities of Chris Langan, individual extraordinaire, found in a comment for a Substack story on “the IQ discourse”:

https://www.theseedsofscience.pub/p/iq-discourse-is-increasingly-unhinged/comment/93033579

You don't assign a numerical score to an intellect like Chris Langan has. It's too paltry a container. Which is enough to lead one to wonder about the utility of assigning a linear score of intelligence, per se. To anyone.

I think people who reify their own "IQ"s are really selling themselves short. Submerging themselves in a mass, however "elite."

I'm also not sure whether (or to what extent) the rare human individuals who demonstrate protean intellectual ability or polymath mastery of multiple disciplines, skills, and crafts are evidence that a handful of outlier off-the-charts genetically superior minds walk among the rest of us, or whether their example is a rebuke to the rest of us, for not activating as much potential as we could have, with a similar level of ability.

Intelligence is funny that way. It's possible to do a treadmill test and learn the limits of your aerobic capacity and various other markers of cardiac fitness. But an IQ test just doesn't operate with the same urgency, so there's no sure way to know if it's pushing people to the limit of their mental acuity. Motivation has a surprising amount to do with it.

Don't get me wrong, I do think there are bell curves of ability. Plural. And yes, some people are sharper than others, just in general. But if you really want to watch people step up their game, put them in a wheelhouse where they have a natural affinity. A knack. Brian Wilson in his “all business” music zone. As someone once put it, "that part of a person that isn't nature or nurture, but what is it?"

Whatever it is, it isn't nothing.

I've read enough IQ references online to get the unsettling suspicion that the term is enjoying "meme popularity" of the type that it doesn't deserve. It isn't a shorthand abbreviation for the scope of human intelligence, and never will be. The verbal tests are a quite accurate measure of the academic skills required for superior school performance. If you haven't learned to read by the time you're in third grade, you won't do well at them. (Come to think of it, I've never seen or heard of a reference to an IQ test designed in any language other than English or French. I'll have to do a keyword search to see what I can find out about that.) The nonverbal tests--Raven's Progressive Matrices--are only aimed at one subset of skills: "abstract reasoning", which in this case is more like the ability to measure the next step in a linear progression by looking at visual diagrams. One interesting feature of RPM is that verbal autistics with impaired measures on verbal tests like the Weschler (WISC) often do quite well on RPM. Two other interesting features of Raven's: it's pitched as a nonverbal assessment, which it isn't; some baseline level of verbality is required in order to understand exactly what the test wants as a correct answer. That isn't automatically apparent. The other interesting feature is that because it's "nonverbal"--which is mostly the case--it no longer has "culture bias". But in my opinion RPM does have culture bias--not in the test content, but in the conclusions. The subtext of viewing RPM as equivalent to or superior to its main rival, the Wechler tests is that the skill set it measures with its linear time-space

sequencing prediction constitutes a probative measure of "abstract reasoning ability", which in turhas long been held as definitive for "Spearman's 'g'--or general intelligence. About which more here adwjeditor.substack.com…

Long story, short, even Spearman demurred about what 'g' measured. I'm going to differ with him even more: I don't think the substrate of intelligence is all about the ability to perform specific abstract reasoning tricks. And the reason Abstract Reasoning gets so much positive emphasis is that it's what makes good coders. I'm not saying that possessing superior ability to do that means that someone only has a narrow bandwidth of intelligence, much less implying that superior performance is an indication of autism. That's reading the results backward, a logic error of the first order. There are obviously people with superior abilities in the skill set associated with RPM that who also demonstrate their intelligence abilites in other ways--even, rarely, in a great many other ways.

What I am saying is that superior performance at abstract reasoning alone is a very narrow ability set. Furthermore, the fact that it's often highly rewarded monetarily tends to narrow people who focus their attention on abstract reasoning professionally in correlation with the monetary reward they receive. And that's how the political economics of the Silicon Economy relate to the circles of American culture, and the greater world adjacent to that

business. So RPM is held to be the new gold standard of IQ testing: the veracity of all "IQ" tests are ultimately reliant on correlative measures, numbers based on selected verbal criteria chosen for the purpose of interpretation. And the criteria shown are factors like advanced schooling, and income, and occupation. Is it any surprise that a society going through a time that materially rewards the narrow skill set measured by RPM is going to view those results as a probative measure of general intelligence? Also, the intelligence testing industry, currently challenged in explaining the disparity--studies showing improvement of up to 2 standard deviations improvement with RPM over the WISC, by verbal autistic test takers--for test supposedly measuring the same quality, 'g', general intelligence. Especially given that some recent studies are also showing the obverse: some high performers on the Weschler see their scores decline with RPM. So cultural bias factors are still in the mix on this subject.

The inclination to view IQ score numbers as proving something permanently real about innate intelligence potential--"because numbers, science"--is also about Culture. "IQ" is a "text meme" of the worst sort: brief, catchy, utterly misleading.