I was thinking that an interesting definition of intelligence might be “your ability to solve problems that a computer can not”. If so, then we might expect that the harder a problem is for AI to solve, the more g loaded the problem is. For example the least g loaded problems on the Wechsler are the processing speed subtests because these require simple rapid decision making, immediate memory & psycho-motor speed. I have no doubt they could program a robot that could score far better on these subtests than any human.
On the other hand it would be very hard to program a robot to score high on the more g loaded verbal comprehension tasks .
Pumpkin Person 
“… The more g-loaded verbal comprehension tasks”? I doubt that a robot wouldn’t be able to score highly on vocabulary tasks, namely giving synonyms or antonyms of a given word. I can also see a robot acing sentence completions which is the verbal form of the gestalt test, so highly g-loaded. With Reading comprehension it would be hard to program the level of nuance to determine things like “what did the author meant by this?”, but I would argue it can be achieved; although it would be much harder, of course.
Reading comp requires understanding irony, sarcasm, and metaphor which would be very tough for a robot.
And an autistic person.
Just if the computer learn about denotative and conotative language styles.
What about arithmetic and digit span? https://pumpkinperson.files.wordpress.com/2018/12/g-loadings.png
A robot would score perfect on digit span but would flunk the more g loaded Arithmetic because the math is presented verbally.
6 years without understanding what intelligence is, and counting… poor PP.
It is no by accident that you think Oprah is a genius.
So what’s PP’s response to Bird’s analysis showing that there is no evidence for “natural selection” in blacks and whites using within family GWAS showing that the “trait” evolved neutrally? He showed that 4-8 points out of the alleged 30 point gap can be “genetic” as a lower bound. Just a bunch if handwaving from “hereditarians” on Twitter—so PP et al, what are your thoughts on this negative, for hereditarians, result?
https://onlinelibrary.wiley.com/doi/abs/10.1002/ajpa.24216
https://magazine.scienceforthepeople.org/vol23-3-bio-politics/genetic-basis-genome-wide-association-studies-risk/
I think it’s something we need to be concerned about. I don’t know enough about genetics to judge the quality of his research but if it was published in a peer reviewed journal, Cochran & Piffer need to respond.
I dunno. I put a self-proclaimed “radical” like Bird claiming genes have little influence on intelligence/group differences in the same category as a coal lobbyist claiming global warming is a hoax.
The whole “we shouldn’t platform non-mainstream beliefs!!!1!” idea is super gay, so I hope Cochran and Piffer respond.
His work should be judged on the merits not his politics.
I agree, that’s why I think the pro-polygenic score camp should respond.
It says more about your politics and not Kevin’s if your bringing up Kevin’s political views.
Two favorite and stupid name callings, losers like austin love to say
Autist
And
Gay
Interestingly Bird is just behaving like a typical WHITE PSEUDO INTELLECTUAL (~50% or more of all “intellectuals” specially white), mostly hetero and pure blood…
And to believe in bullshit seems you need to be more psychotic than autist..
And austin, he believes he is supremely wise or lucid because he follows, like a sheep, sociopaths from fake hbD…).
So many layers of wrongness there. It’s not surprise when many of these mostly White sheep end up believing in all non-progressist bullshit even the most dangerously obvious…
sailer and other associated retards still don’t understand why his account on twitter was excluded… this whitetard shabbo goy has spread fake news about everything he can during 2020, specially about this pandemics… how many stupid men should misbehave and killed an elderly or inmune compromissed people???
On very avg, white people are extremely sloppy in what matter most
True rationality
Most of WHITE MAGIC is because white’s good lookingness…
White race is the best example of how wrong a population can evolve even with supposed true advancements. because when the basis (true philosophy) is rotten no matter how beautiful is your face or how gigantly technologically advanced is “your” civilization, you are already doom because nothing what we do really matter in the end and exactly because this we should try our best. tip: destroy magnificent earth natural environments by profits is not one of them.
You “were thinking” – as a human you can think (the main aspect of IQ tests). Robots, which lack minds, can’t think. So this is useless and g isn’t real.
Do hereditarians claim that g is virtually the same across all groups of people, or just that all mental abilities positively correlate with each other wherever they’re measured?
The latter. They claim that ‘g’ is different between groups.
Can’t speak for all of them, but Jensen seemed too. Indeed his main argument for the validity of the BW IQ gap was that it correlated with g. Rushton also compared Blacks, Whites and Indians in South Africa and iirc argued that the differences were g loaded, regardless of which group from which the g loadings are calculated. Gould mocked jensen for implying the intelligence gap between humans and advanced aliens would be g loaded.
Rushton coined the term Jensen effect, meaning the size of an IQ gap between groups correlates with the g loading of the test and noted that there was no Jensen effect for IQ gains caused by adoption, implying the IQ increase was spurious. Inbreeding depression is also a Jensen effect as is heritability. So it seems to be measuring something genetic but it’s possible to have biological IQ gaps that are not Jensen effects (i.e. fetal alcohol syndrome)
“Hereditarians” have proved that racial differences in intelligence are real by classical science empiricism but they has been fooled by overly abstractizing induced by Kevin Bird types…
At priori you don’t need to prove how happens by molecular levels the atmospheric process that results in the rain to know or identify a rain…
You do not need to restrictly prove what it is already evidenced or it is easily noticeable on a wide scale and/or intercontinentally…
computer is the hardware.
a software simulation of the brain would solve all problems a human could.
given it was raised properly and treated right.
The brain regions most associated with (g) are the frontal lobe connected to the parietal lobe. It is top-down from these regions. It would be very easy to have a computer simulate this. I find no reason not to.
Hey Pumpkin. Do you think that the SAT math section could possibly correlate better with verbal IQ due to the fact that the questions are presented as word problems?
Yes, in addition to g, they both likely load on a verbal factor to some degree.
This gets into what I was saying about intention and inquisition being aspects of intelligence. Artificial intelligence is inhuman because it is more or less input—>output. It lacks internal ability/desire and metacognition to incept, self-improve, and identify information and problems interesting to itself. It doesn’t conceive.
I view intelligence as just the brain’s problem solving computer and emotions (intention, curiosity) as just the problems we solve.
I assume a robot could be programmed to solve the same problems even if it lacked the subjective feelings of desire & inquisitiveness but I could be totally wrong.
At a certain point, you have to say a 130-IQ person who cures cancer is effectively a better problem solver than a 150-IQ meek, unambitious librarian. A lot of times, high-IQ burnouts and losers just weren’t ever psychologically cut out for real-world problem solving and thus had less ability. I get you want to operationalize intelligence as just IQ test aptitude, but we have to take actual productions into account.
At a certain point, you have to say a 130-IQ person who cures cancer is effectively a better problem solver than a 150-IQ meek, unambitious librarian. A lot of times, high-IQ burnouts and losers just weren’t ever psychologically cut out for real-world problem solving and thus had less ability. I get you want to operationalize intelligence as just IQ test aptitude, but we have to take actual productions into account.
We all know you can be smart and lazy, or smart and evil, or smart and unambitious. Sure we could redefine intelligence to include hard work and nobel ambition, and that would make it a better predictor of real world problem solving. Similarly we could redfine height to include speed and coordination thus making it a better measure of real work basketball performance, but what would be gained? One purpose of science is to isolate variables and in psychology they make a clear distinction between abilities (voluntary reactions) and emotions (involuntary reactions). So a high IQ should predict the ability to cure cancer, but if you don’t have the involuntary interest or temperment for scientific research, you wont VOLUNTEER to use said ability for that.
“a 130-IQ person who cures cancer” is more likely than not an illogical verbal absurdity on par with a “round square” or “dry water”
You yourself describe intelligence as “the mental ability to adapt.” Personality stems from the mind, and it unequivocally affects adaptability. Why not just abandon that chemistry teacher’s description of intelligence and just say intelligence = IQ?
I guess it comes down to how we interpret my chemistry teacher’s definition. I see personality as part of the situation you’re adapting to while you see it as part of the adaptability. I see it as part of the problem you’re trying to solve while you see it as part of the ability to problem solve. The definition is ambiguous & of limited scientific value but I find it poetic.
IQ test measure your ability to adapt to a wide range of situations (i.e. WAIS subtests) under circumstances where personality is relatively controlled because almost all personalities find the WAIS fun and are motivated to try their best, and it’s the psychologists job to make sure you do.
And yes, Edwin G Boring infamously defined intelligence as what intelligence tests measure which I don’t much disagree with.
G: Many consider MJ the GOAT hooper, and he’s like 6′ 5″ barefoot. The statistically expected IQ of someone who makes such a breakthrough is probably around 4 sigma, however, so I hear you.
I to an extent agree with you and Boring in that I think a lot of cognitive tests have high face and construct validity. It baffles me that an anti-IQist could take an arithmetic or matrix test and claim it has nothing to do with intelligence. I just hesitate to say it’s the same thing. You could say I see IQ as height, personality/emotional regulation as coordination and speed, and intelligence as basketball ability – or something like that. Intelligence is too complicated to encapsulate with some number on a particular test, but IQ definitely has some value per se and more or less represents one’s ability to understand.
PS: I’ve seen you use “i.e.” a couple times when I think you mean “e.g.”
Not trying to be pedantic, but I thought you’d care 🙂
Thanks for the correction. Information was tied for my lowest score on the WISC-R so I did not know i.e. was wrong. 🙂
Solving scientific problems loads heavily on IQ. Not all challenges in life do, and life is the real IQ test. A numerical indicator like an IQ score is only meaningful insofar as it predicts (correlates with) life outcomes, and just because it “predicts life outcomes” doesn’t mean that it predicts ALL life outcomes equally!
There is a great reason why more intelligence isnt better mainly due to the comorbidity of having a high intelligence corresponding to mental illnesses. If youre a 150 with a high amount of ADHD you will have scattered thoughts. If you have schiz you will not be able to contain the imaginative side of your intelligence. If youre autistic you will have a tough time investing the emotional side of things.
This is why people like Anime though I note his IQ isnt exceptional are always struggling to come up with clear constructive ideas that actually progress forward with time through follow-ups and build-ups of new and different varieties.
More is not always better is the point here. And I would say the sweet spot for intelligence can be as low as 130 and as high as 170. The range is just attributable to which person has the most talent aside from IQ. It generally comes with more costs than actual benefits so that is why a lot of very highly intelligent people do not become characteristically successful at least not to the level they should.
ADHD seems plausible.
Why is Santo allowed to comment and im not? He sounds like a raging monkey!
”He sounds like a raging monkey!”
Exactly for that
I’ll teach you!1
I’ve thought the same, but it might not last for long.
Did you know that OpenAI gave the Raven’s Progressive Matrices to one of their neural nets? It did pretty well at the simple “continue the visual pattern” items, but was incapacitated to answer problems which required abstract reasoning. https://openai.com/blog/dall-e/
It says:
DALL·E is often able to solve matrices that involve continuing simple patterns or basic geometric reasoning, such as those in sets B and C. It is sometimes able to solve matrices that involve recognizing permutations and applying boolean operations, such as those in set D. The instances in set E tend to be the most difficult, and DALL·E gets almost none of them correct.
The different sets are not showing on my computer. Damn.
I took screenshots of each set and uploaded them to an imgur album. Note that only the element in the bottom-right of each picture is part of the robot’s answer, with the remainder of the image having been fed into it as the prompt.
You’re so resourceful! Thanks!
When doing something as simple as putting screenshots into an imgur album and sending you the link demonstrates remarkable resourcefulness, it’s time to seriously think about why there are so many morons in your comment section
(Not an attack at you or the content you write; I really want to know why)
Ganzir despite my somewhat negative reputation on this blog you can vouch for me in saying that you think im a competent and capable person right?
See link that explains these thoughts:
https://www.unz.com/jthompson/james-flynn-rip/#comment-4341095
Please read comments 36, 53, 54, 63, 66, and 67 for the theme and its development.