Hacker News Reader

Comment by llamaimperative | Hacker News Reader

No, it's another datapoint for IQ. General intelligence was discovered empirically, not hypothesized and then searched-for.

The basic observation was that people who score highly on one intelligence-related test will generally score highly on every other intelligence-related test. The "dimensions" of intelligence, while hypothetically separable and distinct, don't actually appear to be so in practice. Smart people are smart at most things (in general), dumb people are dumb at most things (in general).

This only gets shaky when one uses it to make moral arguments about people's value or worth, so I'll just explicitly say it: being smart is not a moral victory or very likely even something you earned at all. Smart and dumb people have equal claim to a dignified life here on earth.

View on HN

If you have the time, I am curious what you think about this post, "Three-Toed Sloth."

http://bactra.org/weblog/523.html

It is obviously just one person's blog post, but I did find it interesting and compelling. However, I must admit that some of the areas of discussion surpass my level of knowledge of statistics and psychometrics.

I am just curious what your opinions are since you seem much more versed in this topic than I am.

To be clear I’m nowhere close to an expert. But my reaction to that blog post is that I’m not quite sure what it’s rebutting? It seems to argue (successfully, I guess) that g isn’t some singular source of intellect that’s powering all other dimensions of intellect. I am not sure if other people believe this to be the case, but I certainly don’t, as that seems contrary to the structure of the brain itself. There is no “smartness” lobe that’s supporting the “smartness” of all the other things the brain is doing and smart people have a bigger/better smartness lobe than dumb people. That’s just not what the brain looks like.

What this blog post does say, which I believe to be true, is that intelligence on one dimension is highly correlated with intelligence on other dimensions. Not because they’re necessarily all driven by the “same source,” which seems like an implementation detail with, AFAICT, pretty much zero bearing on anything of practical concern?

This is just a distortion, and shows you haven't actually engaged with the scientific criticism of IQ tests:

> General intelligence was discovered empirically, not hypothesized and then searched-for.

General intelligence has been hypothesized - and assumed to exist - by almost all peoples since prehistory. The g factor is what you are referring to, and that was discovered "empirically." What was not at all empirical was the decision to call the g factor a measure of general intelligence. That is a hypothesis which seems flatly wrong, and I have yet to see an argument for its validity that isn't circular or specious. My point is that written cognitive tests don't measure intelligence, they measure something much more shallow. The fact that written cognitive tests have a correlating g factor does not mean that the g factor itself is any less shallow.

From the first line of the Wiki article: "The g factor (also known as general intelligence,...)"

I was referring to g, not to the concept of general intelligence, which very obviously has pretty much always existed. Apologies for the ambiguity.

In any case, I don't think it really matters how precisely or reliably we can actually measure g directly, which is really what your sticking point is. Every important decision we need to make around the idea of general intelligence has to do with 1) whether it exists, 2) how well it correlates to positive outcomes in life under specific social and economic regimes, and 3) how heritable it is. All of these seem pretty darn obvious from various experiments not relying on measuring g directly.

Why do you think it matters whether we can measure g? What decisions are there to be made with that information?

> The "dimensions" of intelligence, while hypothetically separable and distinct, don't actually appear to be so in practice.

Circularity. People have normally distributed IQ test results because the numbers are sent through a process that makes them normally distributed.

If this was true then the actual raw test results wouldn't come in different categories. (And there wouldn't be "autistic savants". And being a grandmaster chess player would transfer to other fields instead of coming entirely from practicing since age 5.)

1. I wasn’t making any claim to the distribution of IQ test results across the population.

2. You can have different categories and still have correlations between them

3. Presence of outliers (savants) doesn’t discredit the general correlation

4. No one is claiming skills cannot be learned

> Presence of outliers (savants) doesn’t discredit the general correlation

You're now claiming a "correlation", but a correlation is meaningless so we don't need to credit it in the first place - since, you know, correlation is not causation. It is evidence against a useful causation.

(If the causation is "low IQ causes poor skills except when it doesn't" this isn't useful since it's a tautology.)

> No one is claiming skills cannot be learned

You were claiming that smart people are better at everything before this, but it's well known that they're not better at this even though it's a purely mental skill.

Strictly speaking you were claiming that people better at some test are better at every test, but presumably you meant to generalize there, since if they just had high "test taking skills" nobody would care. Also a likely circularity lurking in the definition of "test"…

I didn’t claim anything other than a correlation, and no, I don’t see why causation is important but correlation is not. What decision hinges on this?

You’re speaking as if there’s no correlation between IQ and performance in chess. This is not true. https://www.sciencedaily.com/releases/2016/09/160913124722.h...

“Strictly speaking,” no, I didn’t say people better at some test are better at every test.

Yes, intelligence is more a multiplier of effects, rather than something useful in and of itself.

You still need knowledge, skill, training, etc. Intelligence will allow you to acquire all of that more quickly and make better use of it, but intelligence without knowledge is useless.

And just like taller people aren't better than shorter people, more intelligent people aren't better than less intelligent people. Nothing can make you a good person except your choices to do good things.

> more intelligent people aren't better than less intelligent people.

I completely agree with this, and I wish it was something that was a bit more emphasized.

However, I do think there lies the dangers of intelligence research. In seems that so much of the world is catered towards the idea that more intelligence = better person. I suppose the notion is that more intelligence people can produce greater value, thus they are more important/beneficial to society?

In society, we have "gifted" programs, "elite" universities, "knowledge" workers, etc.. I'm probably rambling and wording this all poorly, but I guess what I am trying to allude to is that society definitely ranks people's value by intelligence in overt and subtle ways.

Don't you think most parents would rather have a child with average empathy with gifted intelligence than a child that with gifted empathy and average intelligence?

(Tangent below)

I was required to take a legitimate IQ test (WAIS-IV) for the purposes of diagnosing/confirming a disability when I was in college. I'm not kidding when I say this, but I wish I was able to remain ignorant of the composite and subtest scores. Sure, it told me somethings about myself, but at the same time, I felt like it was some kind of report describing what domains my aspirations and limitations in life should be confined to.

To me, it seems like intelligence is not important to those that have a lot of it, and it's significant to those that lack it (whether they realize it or not).

Weirdly, I'd say exceptional skill/talent winds up fostering empathy rather than hindering it.

When you have exceptional talent, you quickly learn not everyone is like you and that you need to understand their perspectives. Whereas if you're within the average range, you think everyone should be in the average range, so you can wind up overly frustrated with both those below it and those above it.

People mistakenly believe that a proficiency in one area means a deficiency in others, like we're all given a set of stat points at birth we have to allocate. When the truth is some people are just generally proficient. There are people with higher intelligence and greater empathy. It's not a choice. And ironically, empathy is one of those things that is actually positively correlated with intelligence. Which kind of tracks as a component of intelligence is pattern recognition and empathy is a form of pattern recognition.

Reply on news.ycombinator.com