An idea about polygenic inheritance

    In Bruce Charlton's post, "How are highly intelligent people sometimes born to unintelligent parents (and ancestors)", he writes: 

    "This (assuming the phenomenon is real) seems hard to explain in the way that intelligence is normally considered - in terms of intelligence being a consequence of very large numbers (thousands?) of genes-for-intelligence.  With intelligence genes conceptualized as additive in effect, and in such large numberes, it is hard to understand how a very highly intelligent child could emerge by change from low intelligence parents.

But if a person's level of intelligence is also determined by the number of deleterious mutations they inherit from their parents, and these mutations are numbered in tens - then it is imaginable that, by chance, a child may be born with a very few deletrious mutations, despite his parents having a relatively heavy mutation load. 

This notion is perhaps testable, on the basis that a low mutation load should be associated with generally higher fitness - so the high intelligence child of low intelligence parents would be expected to be (on average) taller, healthier, more symmetrical, more long-lived than his low intelligence parents."

    This seems plausible.  

    Intelligence, height, and skin color are standard examples of polygenetic traits, where instead of a single gene, many genes each with a small effect contribute to the expression of the trait.  If an individual has more genes that contribute, then the expression of their trait will greater.  If the genes are inherited independently from one another, then their expression in the population follows a Bell Curve.  

    Related to this, I wonder if there is another way that two moderately intelligent parents may have an intelligent child or two moderately tall parents may have a tall child.  Suppose that they are both intelligent or tall "in two differents ways".  Here is a simple model of this: 

    Suppose that we have four genes that affect intelligence each with two alleles, one that has no effect, represented by a lowercase letter, and one that increases intelligence, represented by an uppercase letter.  

    Now, consider two people with equal intelligence but in different ways:

I: AA BB cc dd 

II: aa bb CC DD

    Now, person I and person II had a child (Person III), that child would inherit one allele of each gene from the parent, so that child's genotype would be as follows: 

III: Aa Bb Cc Dd

    The child would have equal intelligence to the parents since all three people have four intelligence-increassing alleles.  However, if Person III had a child with someone else either with a similar genotype to themselves or similar to either of their parents, they would have a chance of having a more intelligent child than themself.  

Consider person IV: AA BB cc dd

Then, there is a chance III and IV could have a child with the following genotype: 

    V: AA BB Cc Dd

    With six intelligence-increasing alleles, V would be more intellient than V's parents or grandparents.

   Or, if III had a child with someone with a similar genotype to themself: 

VI: Aa Bb Cd Dd

Then there is a chance they could have as a child: 

VII: AA BB CC DD

who with 8 intelligence-increasing alleles is significantly more intelligent than VII's parents or grandparents.  On the other hand, they could also have a child 

VIII: aa bb cc dd 

who is less intelligent than either VIII's parents or grandparents.

  Thus, according to this model is correct, after two people who are intelligent in two different ways have a child, there is a possibility for their grandchild to be more intelligent than either the parents or grandparents.  

    It is possible that different populations might have evolved different sets of genes that which code for intelligence.  The genes not used by a population might have alleles which, unlike deleterious mutations, don't decrease intelligene, they just have no effect.  For example, some inhabitants of the Solomon islands have blond hair, but the genes that code for this trait are different than the ones that cause blond hair in Euoropeans.  So, both traits express a similar phenotype, but with different underlying genes.  

    It is also not necessary for there to be no overlap at all amongst such genes.  For instance, consider the following individuals: 

1: AA BB CC DD ee ff

II: aa bb CC DD EE FF

    I and II both have intelligence increasing alleles on the "c" and "d" genes, but they don't overlap on the "a", "b", "e", and "f" genes.  In this case, we would also see a situation in the second generation where a child more intelligent than the parents or grandparents to be born.    

2 comments:

  1. I think Bruce's point is that the process you describe could only work if intelligence were determined by a relatively small number of genes. You flip a coin 10 or 20 times, and you just might luck out and get significantly more "heads" (intelligence-increasing alleles) than "tails." Flip a coin 5,000 times, though, and the odds of that happening are negligible.

    Interesting coincidence: Just last night, one of my students said to me, "My parents and grandparents all wore glasses from a very early age, but I have perfect eyesight. How did that happen?"

    ReplyDelete
    Replies
    1. Thanks for the comment. After reading it, I realize that when writing this I had not fully understood Bruce's argument.

      I see now that the same situation of non-overlapping genes would have to occur with the deleterious mutations as well; some of the mutations of each parent would have to occur on different genes.

      I think the situation I described still could work, if there are enough non-overlapping genes.

      Interesting to hear about your student; that definitely sounds like this kind of situation. I wonder also if someone who is farsighted and another person who is nearsighted have a child, could the traits "cancel out", leading to better eyesight in the child?

      Delete

The real AI agenda

    On a post  by Wm Briggs, about artificial intelligence, a commenter with the monniker "ItsAllBullshit" writes:           "...