Tag: iq

Urban Revitalization Factors

the system compared 1.6M pairs of photos taken 7 years apart to test several hypotheses about the causes of urban revitalization. Density of highly educated residents, proximity to central business districts and other physically attractive neighborhoods, and the initial safety score assigned by the system all correlate strongly with improvements in physical condition. Raw income levels, housing prices or neighborhoods’ ethnic makeup do not to predict change.

IQ 1000

Given that there are many 1000s of potential positive variants, the implication is clear: If a human being could be engineered to have the positive version of each causal variant, they might exhibit cognitive ability which is 100 standard deviations above average. This corresponds to more than 1000 IQ points.

or perhaps merely 550:

Most humans have 1000 rare (-) alleles for intelligence and height, and someone who is 1 standard deviation above average has 30 fewer (-) variants. A human with none of the negative alleles might be 30 SD above average! Such a person has yet to exist in human history. When current IQ tests were developed, the median raw score of the norming sample is defined as IQ 100 and scores each standard deviation (SD) up or down are defined as 15 IQ points greater or less, although this was not always so historically. By this definition, 95% of the population scores an IQ between 70 and 130, which is within 2 standard deviations of the mean. 30 SD above average would be an IQ of 550.

here is an urgent rationale to increase everyone’s IQ by 30 points: survival of the species depends on it. our problems are getting harder faster than our ability to adapt, and you need a level of intelligence to set aside your short term aims in favor of a global view. my guess would be that’s around 130. the linked article has proposals how to bring in the rear, which is a great start. and there’s also more modest proposals:

KL-VS did not curb decline, but it did boost cognitive faculties regardless of a person’s age by the equivalent of 6 IQ points. KL-VS will be the most important genetic agent of non-pathological variation in intelligence yet discovered.

another overview:

individual differences in human intelligence can 50%–80% be explained by genetic influences making intelligence one of the most heritable traits. However, present GWAS studies can capture 22% of this heritability. Each gene has a small effect on intelligence. 95% of the genetic variants are located in intronic and intergenic regions and might have a gene regulatory function. 1.4% of associated SNPs are located in DNA fragments that are translated into protein. By 2025, between 100M and 2B human genomes could be sequenced.

and it goes beyond IQ too:

There are people with rare but highly beneficial genes. Adult whole-body gene therapy could make what is currently good and rare into something that is common or universal.


2023-12-15: A sketch how IQ augmentation in adults might be possible

  1. Determine if it is possible to perform a large number of edits in cell culture with reasonable editing efficiency and low rates of off-target edits.
  2. Run trials in mice. Try out different delivery vectors. See if you can get any of them to work on an actual animal.
  3. Run trials in cows. We have good polygenic scores for cow traits like milk production, so we can test whether or not polygenic modifications in adult animals can change trait expression.
  4. (Possibly in parallel with cows) run trials on chimpanzees
    The goal of such trials would be to test our hypotheses about mosaicism, cancer, and the relative effect of genes in adulthood vs childhood.
  5. Run trials in humans on a polygenic brain disease. See if we can make a treatment for depression, Alzheimer’s, or another serious brain condition.
  6. If the above passes a basic safety test (i.e. no one in the treatment group is dying or getting seriously ill), begin trials for intelligence enhancement.

AIs beat IQ tests

Our model can reach the intelligence level between the people with bachelor degrees and those with master degrees

and

it’s taken 60 years of AI research to build a machine in 2012 that can come anywhere close to matching the common sense reasoning of a 4-year old. But the nature of exponential improvements raises the prospect that the next 6 years might produce similarly dramatic improvements. So a question that we ought to be considering with urgency is: what kind of AI machine might we be grappling with in 2018?

Vocabulary Test

Ghent University in Belgium has created an online, almost arcade-game-like test of word knowledge which is almost BS-proof. Know the word? Press J. Don’t? Press F. But don’t lie! You will be punished.

You said yes to 76% of the existing words. You said yes to 0% of the nonwords. This gives you a corrected score of 76% – 0% = 76%. This is a high level for a native speaker.
On a different test, I’m in the 75% for native speakers who took the test, but allegedly 3x the vocab of ESL speakers. which seems ridiculous.

Cognitive Inequality

i wonder how the cognitive surplus from the internet is distributed. it could be as high as 20 IQ points (my guess) over the last 30 years. i suspect though that most of the gains went to people already on the right hand side of the distribution, perhaps with a cutoff point around 120 IQ points. this would support the idea expressed by Steve Jurvetson that perhaps the increasing returns to intelligence went primarily to the already intelligent, and the gap has increased? it is true that the flynn effect holds true for the entire spectrum of intelligence but perhaps this is a higher level effect than what is measured?

this is old but still very true.

the internet makes dumb people dumber and smart people smarter. If you don’t know how to use it, or don’t have the background to ask the right questions, you’ll end up with a head full of nonsense. But if you do know how to use it, it’s an endless wealth of information. The internet is now a major driver of the growth of cognitive inequality.

and the effect seems to be increasing:

In a large US-representative adolescent sample, a Flynn Effect was found for IQs ≥ 130, and a negative effect for IQs ≤ 70. IQ changes also differed substantially by age group. A negative Flynn Effect for those with low intellectual ability suggests widening disparities in cognitive ability

Ethics arbitrage

In China, a research project aims to find the roots of intelligence in our DNA; searching for the supersmart

it would appear that the biggest sequencing lab in the world has no qualms about research that would not pass ethics approval in the west. of course, the results of that research will not be ignored by the west, how convenient.

Dysgenic Fertility

Dysgenic fertility means that there is a negative correlation between intelligence and number of children. Its presence during the last century has been demonstrated in several countries. We show here that there is dysgenic fertility in the world population quantified by a correlation of − 0.73 between IQ and fertility across nations. It is estimated that the effect of this has been a decline in the world’s genotypic IQ of 0.86 IQ points for the years 1950–2000. A further decline of 1.28 IQ points in the world’s genotypic IQ is projected for the years 2000–2050. In the period 1950–2000 this decline has been compensated for by a rise in phenotypic intelligence known as the Flynn Effect, but recent studies in 4 economically developed countries have found that this has now ceased or gone into reverse. It seems probable that this “negative Flynn Effect” will spread to economically developing countries and the whole world will move into a period of declining genotypic and phenotypic intelligence.

the singularity can’t come soon enough 🙂

Against NYT

The Times’ list is completely fictional. Made up. Divorced from reality. The stated goal of the list is to find (and promote) books that Times editors want people to read, not books that are actually selling a lot. So, they make up ‘rules’ to appear consistent. When Harry Potter was selling like crazy, they invented a new list so that they could take JK Rowling’s books off the real list. When diet and other books started selling a lot, they made up a new ghetto (miscellaneous) for those books. When books started selling in places like Walmart (thus driving the snootiness factor down) the Times penalized sales in chain outlets. And books like the Bible are banished because they’re not current enough.

Carefully curated so as not to have dreck like harry potter or diet books on it.
2013-02-14: Aww, how inconvenient. NYT makes up story, gets owned by pervasive data logging in the car.

After a negative experience several years ago with Top Gear, a popular automotive show, where they pretended that our car ran out of energy and had to be pushed back to the garage, we always carefully data log media drives. While the vast majority of journalists are honest, some believe the facts shouldn’t get in the way of a salacious story. In the case of Top Gear, they had literally written the script before they even received the car (we happened to find a copy of the script on a table while the car was being “tested”). Our car never even had a chance. The logs show again that our Model S never had a chance with John Broder. In the case with Top Gear, their legal defense was that they never actually said it broke down, they just implied that it could and then filmed themselves pushing what viewers did not realize was a perfectly functional car. In Mr. Broder’s case, he simply did not accurately capture what happened and worked very hard to force our car to stop running.

2013-09-12: NYT got trolled by Putin. Syria op-ed by Putin, publication date 9/11. Masterful pr.

RECENT events surrounding Syria have prompted me to speak directly to the American people and their political leaders. It is important to do so at a time of insufficient communication between our societies.

2014-03-19: NYT fails at economics. The rag of record apparently slept through economics class.

most states have limits on direct sales by auto manufacturers. These rules are meant to ensure competition, so that buyers can shop around for discounts from independent dealers, and to protect car dealers from being undercut by automakers.

2016-03-21: NYT middlebrows Terrrrists

To summarize, if you see something on someone’s computer screen that fits the description below, the person with the computer could be an ISIS terrorist!
It looks like “a line of gibberish across the screen.”
It’s “a bunch of lines, like lines of code.”
There’s “no image.”
There’s “no Internet.”

2020-06-26: nyt delenda est

This morning, like many others, I woke up to the terrible news that Scott Alexander—the man I call “the greatest Scott A. of the Internet”—has deleted SlateStarCodex in its entirety. The reason, Scott explains, is that the New York Times was planning to run an article about SSC. Even though the article was going to be positive, NYT decided that by policy, it would need to include Scott’s real surname (Alexander is his middle name). Scott felt that revealing his name to the world would endanger himself and his psychiatry patients. Taking down his entire blog was the only recourse that he saw

2021-02-21: NYT is middlebrow. The reason the NYT is so smug, and yet also consistently wrong about everything, is because it is middlebrow.

The NYT has 7.5M subscribers, mostly progressives in the 90-99% range. These people feel very smart, and they are in fact smarter than 90% of the population. So there’s no point bemoaning the fact that the NYT is not about to tell it’s readers that, “Actually, we provide middlebrow news analysis, and if you want brilliant inspired analysis you need to read blogs like SlateStarCodex.”

Yes, the NYT story is awful in all the ways that are currently being discussed by its critics, but the fundamental problem is inescapable. Any time a powerful middlebrow entity (which wrongly thinks it’s highbrow) evaluates an actual highbrow entity, you will end up with a mixture of resentment and incomprehension. This case is no different. It’s just how things work.

2022-10-24: NYT pretends to do corrections, but only on things that do not matter.

if you’re willing to correct the spelling of 1 vowel in somebody’s middle name or the location of a statue of a rambunctious horse, you should be willing to correct the erroneous statement, “Researchers find that female-named hurricanes kill 2x as many people as similar male-named hurricanes because some people underestimate them,” or various erroneous economic and education statistics.

For Brooks or Kristof to admit to a non-trivial error, as you suggest they do, would be for their admissions to immediately become high-profile fodder for their critics. And not just then, but forever more — a link to the admission, a quote of it, will be repeated at any occasion when a club to use against their credibility is wanted. But more to the point, such a club will work. It will work because people strongly recall that someone was proven wrong about something and especially they recall when someone admitted to it and, finally, they weight that information very heavily when evaluating credibility.