16 point IQ difference between identical twins raised apart | Hacker News Reader

16 point IQ difference between identical twins raised apart

psypost.org

29 pointsTedDoesntTalk51 comments

First, this apparently covers one single set of twins. Tiny sample size.

Second, I can't access the actual paper, but a commenter at Marginal Revolution wrote the following:

"Everyone in the comments here arguing about the implications of this study for the heritability of intelligence (vs. other factors like parenting) seems to be missing an important piece of information buried in the study: the twin raised in the US suffered three concussions as an adult, including one in 2018 that resulted in "light sensitivity and concentration difficulty" and led her to feel like a "different person, with increased anger and anxiety." It is hardly surprising that someone who has suffered multiple major concussions would score lower on tests of cognitive ability than someone who has not."

Wow. Massive, massive omission. If true, the person who publicized this either tossed it off without doing much homework or had a serious ideological axe to grind.

Again, I'm just re-posting that comment. Would be nice if someone who could read the actual article confirmed that the comment is accurate.

Unfortunately it's not yet in the Library of Alexandra …

Once I started reading that they were trying to assess the political beliefs of the twins and measuring their feelings about things like collectivism, I started to smell that this was turning into political propaganda. Especially when they claimed one grew up in a more religious household. How does one measure that? Especially when there's a good chance there were different religions in both families.

Did you purchase the study? Can you post the bit about religion, and also the bit about concussions please?

> Especially when they claimed one grew up in a more religious household. How does one measure that?

How is that supposed to be hard to measure?

Not only this doesn't sounds especially hard to do roughly, but it is in fact a pretty well studied topic in social science so there's plenty of literature about how to do such a measurement.

I think the IQ test has been mythologized in a way that people feel it is a perfect representative of intelligence and is somehow unable to be improved upon and therefor a true measure of god-given intelligence.

This is patently rediculous. If you have ever taken an IQ test you would know the types of question are easily practiced and improved upon. The modern IQ test is used to identify potential learning disabilities by psychologists and beyond that is not really considered.

I have personally scored more than 15 points differently at different times in my life.

The weight some people give to IQ is absurd. Psychometrics as a whole suffers a perpetual replication crisis.

+/- 7 on a scale from 50 to 150+ using different tests at different periods is very stable. It’s kind of shocking how little stuff like lack of caffeine alters the results while people really do score widely differently.

Not that I think the score actually means much, but consistency isn’t the problem.

It would perhaps be a more meaningful test of intelligence if a person took it 100 times over a few years. This, however, is not very realistic.

With 100 tests, the results would likely be skewed by the training …

IQ is the most solid and well-replicated part of psychometrics and psychology in general. It doesn't belong in the same category as growth mindset and power pose and other pop psychology.

It is, it just isn’t what most people think it is, nor was it ever meant to be.

You might be right, but I'm not quite sure what most people think it is, so I don't know.

It is not intended as a measure of giftedness. It is intended to measure proper cognitive functioning. In pop culture it is often talked about as a measure of genius.

I think it is a legitimate measure of giftedness at the top end. Genius, not so much.

Having an IQ of, say, 160 means you're extraordinarily capable of a broad range of cognitive tasks. It's more than merely proper cognitive functioning. But it doesn't mean you will be a world-altering genius. There are many stories of people with staggering IQs who ended up doing nothing notable, or producing a lot of interesting and odd ideas that ultimately had no impact on the world.

IQ is great on a population level, when administered by people who know what they are doing. On an individual level, especially when administered by a non-expert (or worse, online), it just doesn't offer as much information as living life does. As in, if I know I'm not an idiot, it's more likely my IQ test is wrong than that I'm wrong.

I also feel that there would be a likely Time-of-Day difference in test results.

For example (although only indirectly related), I can do Sudoku puzzles really well after I've been awake for a while. But for the first few hours of the day, I make lots of mistakes that I simply don't make once I'm fully awake and mobilized for the day.

Since it is such a bad test, what test do you recommend to replace it?

I think it is an excellent tool for what psychologists use it for.

So it can identify disabilities, but not abilities?

So a psychologist who wrote a book condemning the original twins IQ study now publishes a paper where she purportedly refutes that study's most famous findings based on the analysis of a single set of separated twins?

Isn't this the ultimate in p-hacking? Is there any reason to draw any conclusions from a sample size of one?

Exactly, this is ridiculous...

An unasked question is how many of the identical twins in the original study were separated at birth and raised in very different environments? The answer is not many. A large number of the separated twins were separated at an older age, or were adopted by a close family member of a similar class background.

I don't see this study as any more muddled than the original study.

Other twin studies didn't look solely at twins, but compared them with non-twins. As poster civilized said https://news.ycombinator.com/item?id=31304805:

monozygotic twins reared apart show much more similar IQs than unrelated individuals raised together, indicating that genetics are the dominant driver of IQ. https://www.gwern.net/docs/iq/1996-bouchard-2.pdf

To be clear, one of these twins suffered several concussions while the other did not

Increasing the sample size is useful to get nice statistical properties out of noisy data. That doesn't mean you cannot learn anything from tiny samples, especially when the data isn't noisy (which is the case here, since they have the same set of genes, a lot of the noise is gone).

This is not an example of p-hacking. P-hacking is when you test so many different hypotheses that some of them randomly happen to have low p-values.

This is an interesting claim, because "being lost by your family at age 2 and essentially kidnapped and shipped to another country" ranks pretty high on my life impacting events list.

This is interesting but not new. A study from 1973 found that monozygotic twins reared apart have mean absolute difference of over 6 IQ points, with maximum difference observed several times that. A 16-point difference would be less common but there would be no difficulty finding one in a population of a thousand twin pairs, especially with massive family and cultural differences. https://link.springer.com/chapter/10.1007/978-94-011-6129-9_...

Also, another commenter notes the lower-scoring twin had a life-altering concussion. Pretty enormous omission from this reporting.

Despite this single shoddily-reported example, monozygotic twins reared apart show much more similar IQs than unrelated individuals raised together, indicating that genetics are the dominant driver of IQ. https://www.gwern.net/docs/iq/1996-bouchard-2.pdf

Nurture vs. nature. What's been the effect of all "medical/biological" occurrences since conception? What's the error-range of IQ tests?

This strikes me as a test for which there are too many variables to draw a conclusion.

Assuming IQ tests are fairly good (they’re probably not, but that’s a separate issue), there is one major conclusion to be drawn from this — even when you control for nature, extrinsic factors (nurture, environment) can have an effect at least this extreme.

In this case, nurture is several concussions

You can always degrade genetically highly potential humans to a lower level with environmental factors. But you can never lift up a human to a high level when the genetic prerequisites are not met.

You can turn this around to see the massive positive impact of environmental factors... There were plenty of people around before the invention of written language who could not have even taken one of these tests. Is that an environmental degradation, or proof that we've largely succeeded in improving human intelligence through cultural inventions?

"Never" is a strong word. Perhaps we will be able to fix those missing faculties in the future somehow. We certainly learnt how to fix many other congenital conditions.

My wife was born with a heart defect that would have killed her 60 years ago. 30 years ago, it was already fixable and now she can live a pretty normal life except for "do not exercise too much". Maybe we will be able to do the same in 2040 or 2050 with people who were born with various mental defects.

Maybe there will be a magic pill that makes everybody much more intelligent. We are only using a few percent of our brain after all. Until then I assume that the evidence presented by mainstream psychology is correct which says that between 50% and 80% of IQ is inherited.

It need not be a pill. It might be genetic manipulation of an already living organism. As of today, the tools like CRISPR aren't very nimble, but who knows what tools will be used in 2040 or 2050 for that purpose. The difference between them and today's CRISPR might be the same as difference between Intel 8080 and Intel Pentium.

I would think the number of people maxing out their IQ is pretty small. There's a whole lot of work, including that done by HN readers, that's pretty rote. Not too much "deep thinking" required to clone an application, follow a tic-list, ...

Similarly, the number of people who can't be lifted to "a high level" is similarly small. I assume you don't mean "those who can't function on their own", so you're discussing the mass of people who toil along in "typical jobs for typical people". Healthy, fully-formed w/o congenital defects, thankfully, is common and usually sufficient.

Edit: these tests are culture-specific. Doubt too many of the readers of HN would look functional on one for Inuit or Japanese.

It is worth noting that one of the twins had contracted measles:

> She was later taken to a hospital that was approximately 100 miles away from her family’s residence and diagnosed with the measles

Which in some cases affects cognition due abnormal immune responses which result in brain swelling.

Just to be a contrarian it’ll still turn out to be nature, but mostly gut microbiome rather than genetics ;-)

> Just to be a contrarian it’ll still turn out to be nature, but mostly gut microbiome rather than genetics ;-)

Wouldn't gut biome still be nuture?

A healthy gut microbiome is an outcome that can be strongly influenced by environment.

> The twin who remained in South Korea was raised in a more supportive and cohesive family atmosphere. The twin who was adopted by the U.S. couple, in contrast, reported a stricter, more religiously-oriented environment that had higher levels of family conflict.

This part is unsurprising to me. Religious upbringings can, and do, cause emotional trauma which can lead to problems with mental and physical health, which affects cognitive ability. It's not a problem with the individual interpretation of a religion, but with the ideology itself, which insists that bronze age moral standards should apply in today's world—including those based around blood sacrifice.

The real tragegy is that many consider it good parenting to raise their children to be religious, even if they aren't religious themselves.

I'm the older brother of identical twins. I can tell you that they can be both startlingly similar and startlingly divergent.

For several years during their childhood, I was the only one in the family (2 adults 5 kids) who could reliably tell the twins apart. They would regularly blame stuff on the other twin. And get away with it.

One ended up as a very-liberal TV producer and the other ended up as an evangelical preacher.

It's heavily implied that the twin who was lost, diagnosed with measles, and raised in the US has the lower IQ, but never actually stated in the article.

The article states: "The twin raised in South Korea scored considerably higher on intelligence tests related to perceptual reasoning and processing speed, with an overall IQ difference of 16 points"

How did I miss that? Thank you.

Sample size = 1

This should be flagged.

And that's not very surprising. While I expect low IQ people to mostly benefit from nootropics (e.g magnesium l threonate giving a 9 IQ point increase on average..) I expect for people without existing deficits, to be able to gain ~50 IQ points by optimal training. If not then IQ tests should be improved to better reflect the ability of most humans to become rationalist virtuoses by optimal leveraging of their synaptic plasticity. BTW anyone got a link to a "proper" online free IQ test? It's been a while since I've last benchmarked my brain and BTW I never understood how IQ tests could be comparable given that there exists multiple distinct sets of recognized (by whom) IQ tests.

I would also like to see more atypical/interesting intelligence benchmarks for humans, such as e.g. the amazing https://www.kaggle.com/c/abstraction-and-reasoning-challenge

Reply on news.ycombinator.com