Heritability of Intelligence

[!diom]

I've always wondered exactly how this works. My mom scored a 700 (total) on the SAT in 1980 and, based on what I've seen on the internet, that would probably put her IQ somewhere between 87 and 100 -- which jives with my experience of her. Needless to say, I scored much, much higher than that and my IQ may be over 3 standard deviations higher than hers.

What's up with this?

I haven't done extensive research on the subject and I have no understanding of genetics or statistics (I also don't know my father), but here are the potential explanations I've thought up so far:

(1) My mom suffered brain damage before she took the SAT.
(2) My intelligence may be somewhere in between my mom's and dad's, with my dad being a super-genius. This is unlikely, simply because super-geniuses are unlikely.
(3) My intelligence was mostly inherited from my dad, with normal variation.
(4) I'm a mutant.
(5) Intelligence isn't all that heritable.
(6) ???

Someone help me here.

[Hustler]

If nature really counts for 50% of the variation in intelligence, and if 15 represents the standard deviation in IQ among the population, the maybe this can sort of be figured out. For a sigma of 15, the variance is 225. If genetics accounts for half of it, it means the standard deviation for offspring compared to parents is about 10.6 (the square root of half of 225). Let's say your mom has about a 95 IQ and yours is around 135. You have 40 points on her, or about 3.77 standard deviations. That's rare, but not impossible; one out of every 12,195 people will have that wide a discrepancy from his parents. If your dad has a higher IQ than your mom, and if the genetic component of your IQ is inherited from both parents, then this discrepancy goes down. If, for instance, your dad has a 115 IQ, then their average becomes 105, and you're now only 2.8 genetic standard deviations away from them. This will be the case for one out of every 391 persons.

It's possible your mother was brain damaged or has a learning disability that affected her ability to take the SAT. Anxiety disorders, dyslexia, and so on could easily diminish her score. If she was depressed or bipolar at the time, that could also make a big difference. It's also possible that genetics aren't as big a factor in determining IQ as studies thus far appear to indicate; the jury is certainly still out on pretty much everything regarding the brain.

The other possibility is that your father fucked the milkman.

[!diom]

It's possible your mother was brain damaged or has a learning disability that affected her ability to take the SAT. Anxiety disorders, dyslexia, and so on could easily diminish her score. If she was depressed or bipolar at the time, that could also make a big difference.


Most of that is likely, but it's the same for me (anxiety, depression, potential bipolar II, etcetera, lots of overlap). :(



The other possibility is that your father fucked the milkman.

:confused:

[!diom]

If nature really counts for 50% of the variation in intelligence, and if 15 represents the standard deviation in IQ among the population, the maybe this can sort of be figured out. For a sigma of 15, the variance is 225. If genetics accounts for half of it, it means the standard deviation for offspring compared to parents is about 10.6 (the square root of half of 225). Let's say your mom has about a 95 IQ and yours is around 135. You have 40 points on her, or about 3.77 standard deviations. That's rare, but not impossible; one out of every 12,195 people will have that wide a discrepancy from his parents. If your dad has a higher IQ than your mom, and if the genetic component of your IQ is inherited from both parents, then this discrepancy goes down. If, for instance, your dad has a 115 IQ, then their average becomes 105, and you're now only 2.8 genetic standard deviations away from them. This will be the case for one out of every 391 persons.


I have a question about this. If you continue to calculate the rarity of discrepancy in this way, then the odds would be one in two if my dad's IQ were 175. However, that seems completely unrealistic because the rarity of a 175 IQ is about one in 3 and a half million. Would it be safe to say there's a most probable IQ for him? Based on a few calculations, it seems that would be around 125. This seems weird though.

[Hustler]

Most of that is likely, but it's the same for me (anxiety, depression, potential bipolar II, etcetera, lots of overlap). :(

This varies a lot depending on what state you're in when you take the test. Take bipolar, for instance. Depending on whether you're up or down, it could make a difference of ~15 points in your IQ score. Also, different types of anxieties have different implications. Test anxiety (a type of performance anxiety), for instance, would be quite damaging to someone's IQ score.


I have a question about this. If you continue to calculate the rarity of discrepancy in this way, then the odds would be one in two if my dad's IQ were 175. However, that seems completely unrealistic because the rarity of a 175 IQ is about one in 3 and a half million. Would it be safe to say there's a most probable IQ for him? Based on a few calculations, it seems that would be around 125. This seems weird though.

I think you are using conditional probabilities backwards. First, you would have to calculate the probability of him having a 175 IQ, and then the probability, given that, that you would have a 135 (given what we know about your mother). You could then do that for 125. I think you'll find that 125 is a lot more likely for him than 175 if you do it that way.

The point of this analysis was just to serve as an estimate and to see if it's even plausible that you're not adopted. It turns out that it is quite plausible, as about one in 400 people will be in your exact same position if his parents have an average IQ of around 105. This estimate makes a bunch of assumptions like, for instance, your mother and father contribute equally to your genetic IQ (which I've read before is not the case - some research suggests it's all from your mother) and can just be averaged, or that the genetic IQ and non-genetic IQ both add about equally to both expected IQ and variance in IQ, and so on.

[Resonance]

It's also possible that your grandma and everyone before her didn't get enough vitamin B12 or something during the pregnancy. Brain damage doesn't have to be traumatic.

You were also most likely exposed to a lot more stimulation in infanthood than she was.

Hustler's statistical calculations work for the general population, but IQ is also normalized by age group. Younger people perform better, so the bar is higher. Your raw score is going to be oodles higher than hers, even if her normalized IQ is the same as yours.

[!diom]

You were also most likely exposed to a lot more stimulation in infanthood than she was.


Early-childhood stimulation probably has a lot to do with the development of intelligence, but I have no reason to believe that she was any less stimulated than me.



Hustler's statistical calculations work for the general population, but IQ is also normalized by age group. Younger people perform better, so the bar is higher. Your raw score is going to be oodles higher than hers, even if her normalized IQ is the same as yours.

No matter how you look at it, she's nowhere near me. The 1980 SAT more closely correlated with intelligence, and the average score for that year was 994 re-centered with her at roughly 800 re-centered, so she was below average for her own age group.

[Resonance]

oh, ok. Yeah that is a pretty bad score, especially if she wasn't like high or something.

But yes stimulation is way easier these days. Video games, toys that talk, picture books etc. are all way cheaper now than they were when she was a kid, if they existed at all.

Why do you think african americans' average IQ has gone up in strong correlation with their average socioeconomic class?