If it is, why do we get regression to the mean in intelligence?
-
-
-
How does regression to the mean operate as an argument against the heritability of intelligence? Regression to the mean simply means that the underlying correlation is not perfect (not +/-1).
-
No, b/c in that case it would not be a consistent trend, but 'intelligence' would fluctuate randomly about a function of the mean intelligence of the parents. In other words it would be a function of the parents' intelligence plus an error term - and as far as I know it's not.
-
Someone with a high score has, on average, both better than average genes and better-than-average environmental luck. On average the next generation has the same genes but average luck.
-
So if you picked very tall parents, average of next generation would be less tall, although still considerably taller than average.
-
But if people in the 2nd generation intermarry, the 3rd generation has the same average height as the 2nd. You've created a taller population.
-
This is assuming the 2d generation is taller than the third. But I am not sure what any of this means. Nutrition is a far important determinant of height I would have thought, but significance testing is given more emphasis than estimation...as is true of all cognitive science
-
Given these assumptions, and constant environments, the 2nd generation is taller than average, but not as tall as the 1st generation. The 3rd generation will have the same average height as the 2nd. Selection. works other traits in the same way as it does with height.
End of conversation
New conversation -
-
-
I suspect there are already wealthy parents using in vitro fertilization to select embryos which carry the most genes for high IQ. The next generation of the 1% may well carry a significant intellectual advantage until gene therapy allows the rest of us to “fix it in post”.
-
Remember that the 1% is not a stable set of people. It is highly conpetitive and the people that are the 1% change fast. A fortune seldom stays in a family for longer than 3 generations.
-
I suspect regression to the mean accounts for most of that... and how will that change now that they can ensure the next generation actually accelerates away from the mean? Maybe there is a sweet spot around IQ 130-140, and push to 180 will backfire?
- 1 more reply
New conversation -
-
-
A sincere thanks, sir, for your speeches and writings. They've been a great resource, and they've provided much fodder for discussions in my own circle about a world allegedly "going to Hell in a handbasket."
-
Thomas Smith did an
@seriouspod episode about this during the week. Well worth a listen.
End of conversation
New conversation -
-
-
If intelligence is a fixed essence why do people score higher the more often they do IQ tests?
-
This, a big reason why I'm sceptical about the value of the IQ test: You can learn to output what it wants you to be. It's just like conditioning a dog to do nonsensical tricks in exchange of a treat, yet it doesn't mean that it could apply it in any other situations/contexts.
-
Can you explain how you can solve ravens progressive matrices without understanding what you do?
End of conversation
New conversation -
-
-
Still no agreed-upon definition of what 'intelligence' is, good luck 'finding the genes'.
-
It's in the abstract: 'the ability to learn, reason and solve problems'
- End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.