4 ms·

“Heritability” may or may not have anything to do with genetics or biology. For example socioeconomic status and proximity to zip code are also heritable.

JumpCrisscross 4y ago
> “Heritability” may or may not have anything to do with genetics or biology. For example socioeconomic status and proximity to zip code Twin studies [1]. Different parents, socioeconomic statuses, possibly countries. Sustained significant statistical effects. That’s the genetic component of intelligence. [1] https://www.nature.com/articles/mp2014105 https://www.nature.com/articles/mp2014105 first four references
- snowwrestler 4y ago
  Twin studies provide evidence that there is a biological component to intelligence. They do not tell us how much more biologically likely it is for high-IQ parents to produce high-IQ kids—which is usually what people mean by the “heritability” of IQ. In plain English: the twin studies show us how similar twins are, but they don’t tell us how true the movie Idiocracy is.
  - JumpCrisscross 4y ago
    > do not tell us how much more biologically likely it is for high-IQ parents to produce high-IQ kids "Individual differences in intelligence tend to cluster in families" [1]. There are studies which include "parents in a twin design." They found "no significant shared environmental influence [between the adoptive and non-adoptive children]: all variance could be explained by additive genetic factors and environmental factors that are not shared by children raised in the same family" with "significant genetic transmission for intelligence...found at all ages" (§ 1.3). [1] https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=b3f170a4f2e7699a40df303846d69774bbf6b5b1 https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&d...
    - Apocryphon 4y ago
      But are the parents actually high-IQ, or do they exhibit some other set of characteristics that correlate with high-IQ children?
      - sfblah 4y ago
        Why are you looking so hard for ghosts in the machine here?
        Apocryphon 4y ago
        It's a simple question. Frederick I of Prussia, father of Frederick the Great, attempted to marry off his soldiers of great stature with tall women, to create generations to fill the ranks of his Potsdam Giants. That endeavor did not result in anything strategically fruitful. One may observe that heritable traits can manifest across generations in different ways. The previous response did not directly say if those studies suggested high-IQ parents resulted in high-IQ children. So did it?
        sfblah 4y ago
        [flagged]
        Apocryphon 4y ago
        I was asking what IQ heritability actually entailed. Thank you for answering the question. I am satisfied. The editorialization was superfluous, and composed in vain. I am not going to read it.
        sfblah 4y ago
        [flagged]
        Apocryphon 4y ago
- tptacek 4y ago
  This isn't responsive to the parent comment. There may be a strong discernible genetic component of intelligence, but it doesn't proceed logically from heritability studies; heritability is the ratio of genetically determined variance over all variance, and so can mean either maximum genetic determination, minimum determination, or anything in between. I get that you're probably trying to take this thread in a different direction from raw heritability numbers, but the parent commenter was, correctly, rebutting a flawed previous argument based solely on heritability.

JumpCrisscross 4y ago

> “Heritability” may or may not have anything to do with genetics or biology. For example socioeconomic status and proximity to zip code Twin studies [1]. Different parents, socioeconomic statuses, possibly countries. Sustained significant statistical effects. That’s the genetic component of intelligence. [1] https://www.nature.com/articles/mp2014105 https://www.nature.com/articles/mp2014105 first four references

snowwrestler 4y ago
Twin studies provide evidence that there is a biological component to intelligence. They do not tell us how much more biologically likely it is for high-IQ parents to produce high-IQ kids—which is usually what people mean by the “heritability” of IQ. In plain English: the twin studies show us how similar twins are, but they don’t tell us how true the movie Idiocracy is.
- JumpCrisscross 4y ago
  > do not tell us how much more biologically likely it is for high-IQ parents to produce high-IQ kids "Individual differences in intelligence tend to cluster in families" [1]. There are studies which include "parents in a twin design." They found "no significant shared environmental influence [between the adoptive and non-adoptive children]: all variance could be explained by additive genetic factors and environmental factors that are not shared by children raised in the same family" with "significant genetic transmission for intelligence...found at all ages" (§ 1.3). [1] https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=b3f170a4f2e7699a40df303846d69774bbf6b5b1 https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&d...
  - Apocryphon 4y ago
    But are the parents actually high-IQ, or do they exhibit some other set of characteristics that correlate with high-IQ children?
    - sfblah 4y ago
      Why are you looking so hard for ghosts in the machine here?
      - Apocryphon 4y ago
        It's a simple question. Frederick I of Prussia, father of Frederick the Great, attempted to marry off his soldiers of great stature with tall women, to create generations to fill the ranks of his Potsdam Giants. That endeavor did not result in anything strategically fruitful. One may observe that heritable traits can manifest across generations in different ways. The previous response did not directly say if those studies suggested high-IQ parents resulted in high-IQ children. So did it?
        sfblah 4y ago
        [flagged]
        Apocryphon 4y ago
        I was asking what IQ heritability actually entailed. Thank you for answering the question. I am satisfied. The editorialization was superfluous, and composed in vain. I am not going to read it.
        sfblah 4y ago
        [flagged]
        Apocryphon 4y ago
tptacek 4y ago
This isn't responsive to the parent comment. There may be a strong discernible genetic component of intelligence, but it doesn't proceed logically from heritability studies; heritability is the ratio of genetically determined variance over all variance, and so can mean either maximum genetic determination, minimum determination, or anything in between. I get that you're probably trying to take this thread in a different direction from raw heritability numbers, but the parent commenter was, correctly, rebutting a flawed previous argument based solely on heritability.