Hacker News Reader

Comment by wycs | Hacker News Reader

>Have you considered the possibility that confounding variables exist? I have no doubt that coming from a family where your parents have received a good education makes it much more likely for you to personally also go on to get the same education for example.

Yes.

If you read the citations you will see this is based on twin adoption studies, which control for confounding variables almost perfectly.

I assure you, ever single objection you can think of off the top of your head has been raised and overcome.

The heritability of IQ is not a conclusion psychologists wanted to affirm. It is fact the field was forced to come to from the data, despite the ideological drifts of the last 50 years yearning (or in the case of Stephen Jay Gould outright falsifying data) for the opposite conclusion.

View on HN

Psychology is a field notorious for being derided against as a soft science particularly because of how little data exists with poor sample sizes and ideology / purposeful tampering coming into the way like you've said.

Any meaningful evidence for a conclusion requires a much better understanding of the brain both biologically and also as a function of general cognition. This will require many more breakthroughs within the fields of biology and computer science. We're definitely getting closer and closer everyday on that front, but as of right now, the tools that we have access to are far too crude in my honest opinion.

Psychology in that respect is akin to the alchemy that was a precursor to chemistry. I'm not saying no real science was done by the alchemists of course, just that it was far and away from what the field of chemistry would ultimately become.

I sense this is a lost cause, but we can already predict ~10 percent of the variance in educational attainment with the genome alone.

Once we get data sets with millions of genomes tagged with their donors IQ, this number will rise. If we can predict, say, 60% of the variance in IQ (based on the genome alone) will you change you mind?

That is, what sort of data would change your mind?

https://www.nature.com/articles/s41588-018-0147-3

Again, I'm not trying to dispute the fact that genes are the blueprints to our bodies.

The point I'm making is in regards to the OP's comment about how people coming from legacy will be much higher due to the fact that smart people reproducing with other smart people greatly increases the chances of producing a resultant smart baby. Random variation has much greater effect in determining the end result of that baby's genome is what I'm saying. You could have paired the person with almost any other human being, and the resultant IQ outcome would be no less likely to occur.

Only through concerted effort to discover and understand how the brain is constructed genetically in addition to developing methodologies for testing changes to those key markers will it be possible to meaningfully alter the statistics of intelligence. Anything else will be lost in the noise primarily because of, again, what I stated above. Intelligence of humans is likely already very close to some local maxima and it is also a highly conserved trait.

I used to believe everything you’ve stated until I dove into the twin studies, and more recently the GWAS data from Europe. Read the literature yourself with an open mind and you may be surprised. Ironically, as society has become more environmentally equal, ie the decline of lead, pollution, and clinical malnourishment, Nature has pulled even further ahead of Nurture, because it’s much easier to remove IQ points than to add them.

Reply on news.ycombinator.com