The Unz Review: An Alternative Media Selection
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 TeasersGene Expression Blog
The Fourth Law of Behavior Genetics
🔊 Listen RSS
Email This Page to Someone

Remember My Information



=>
BookmarkToggle AllToCAdd to LibraryRemove from Library • Show CommentNext New CommentNext New ReplyRead More
Search Text     
    Citation: The Fourth Law of Behavior Genetics, Chabris et al.

    Figure citation: Marigorta UM, Navarro A (2013) High Trans-ethnic Replicability of GWAS Results Implies Common Causal Variants. PLoS Genet 9(6): e1003566. doi:10.1371/journal.pgen.1003566

    The above is a figure from The Fourth Law of Behavior Genetics (ungated), accepted for publication in Current Directions in Psychological Science. It’s an excellent overview of the intersection of behavior genetics and genomics over the past 10 years or so. The full story is outlined in the paper, fleshed out by the copious and informative citations which litter the text. The point of the above figure is to show how robust many inferences from small effect genome-wide associations are. In particular, there is the standard caveat that a variant which is correlated with disease X in population 1 may not be correlated with disease X in population 2. It actually turns out that in most cases they are correlated across populations. Above you see the correlation in effects (odds ratios) between variants and traits between East Asians and Europeans (common variants are also predictive within families).

    The Fourth Law of Behavior Genetics makes no sense unless you know the first three laws outlined by Eric Turkheimer:

    * First Law. All human behavioral traits are heritable.

    * Second Law. The effect of being raised in the same family is smaller than the effect of genes.

    * Third Law. A substantial portion of the variation in complex human behavioral traits is not accounted for by the effects of genes or families.

    By heritable you simply mean that some of the variation of the trait in the population is explained by variation in genes in the population (you see it in the standard parent-offspring regression). The second law refers to the fact that on many behavior genetic traits the influence of shared family environment in explaining variation can be surprisingly small. The final law points to the reality that a lot of the variation we see in people in outcomes seems pretty much random. It’s labeled non-shared environment, but we should think of it more as a noise factor. These “laws” are robust regularities which you need to take into account when considering the likelihood of any given result. What is the fourth law? It’s pretty straightforward: “A typical human behavioral trait is associated with very many genetic variants, each of which accounts for a very small percentage of the behavioral variability.” To give a concrete example, it looks like the largest effect common variants for behavioral traits explain about 10% the variance as the largest effect variant for complex disease or morphological traits, on the order of 0.01% as opposed to 0.1%.

    That’s a mighty small effect. To make sense of the heritabilities estimated using classical methods that means that genetic variation is partitioned across many many genes, on the order of thousands. This is why methods to ascertain loci of effect utilizing small sample sizes (e.g., candidate gene studies) were bound to fail, because they didn’t have the power to detect true results. Rather, many of the hits were simply noise which got published because of low stringency of statistical significance.

    One objection then might be that the missing heritability consists of very low frequency (i.e., far less than the 1% threshold used to define common variants in terms of minor allele frequency) mutations which have a larger effect. The authors claim that the research currently does not support that finding. That implies that high coverage whole genome sequencing at reasonable sample sizes won’t make a big difference. Second, there are the results out of the GCTA framework. I won’t go into the details, though check out the paper in AJHG. It’s a powerful way to explore heritabilities using genomic data across unrelated individuals that’s rapidly converging on the heritabilities estimated from classical behavior genetic methods.

    Finally, in a similar vein, Dominance Genetic Variation Contributes Little to the Missing Heritability for Human Complex Traits:

    For human complex traits, non-additive genetic variation has been invoked to explain “missing heritability,” but its discovery is often neglected in genome-wide association studies. Here we propose a method of using SNP data to partition and estimate the proportion of phenotypic variance attributed to additive and dominance genetic variation at all SNPs (Math Eq and Math Eq) in unrelated individuals based on an orthogonal model where the estimate of Math Eq is independent of that of Math Eq. With this method, we analyzed 79 quantitative traits in 6,715 unrelated European Americans. The estimate of Math Eq averaged across all the 79 quantitative traits was 0.03, approximately a fifth of that for additive variation (average Math Eq = 0.15). There were a few traits that showed substantial estimates of Math Eq, none of which were replicated in a larger sample of 11,965 individuals. We further performed genome-wide association analyses of the 79 quantitative traits and detected SNPs with genome-wide significant dominance effects only at the ABO locus for factor VIII and von Willebrand factor. All these results suggest that dominance variation at common SNPs explains only a small fraction of phenotypic variation for human complex traits and contributes little to the missing narrow-sense heritability problem.

     
    • Category: Science • Tags: Behavior Genetics, Genomics 
    Hide 22 CommentsLeave a Comment
    Commenters to Ignore...to FollowEndorsed Only
    Trim Comments?
    1. Helga Vierich says:Show Comment

      Excellent. Finally we are getting somewhere. Yes, so this now makes sense of the elusive and finely sliced gradations in human individual personality. That is only one aspect of human “behaviour”, however.

      Turkheimer’s laws do, indeed, explain at least a little of the remarkable individual variation in musical talent, flare for artistic expression or imagination that we can find even in a single set of siblings… but they say almost nothing about beliefs, table manners, and individual abilities like riding horses, algebra, or tracking game.

      At some point as the generation of phenotypes progresses over time, the slow sweeping tsunami of environmental influences, causing epigenetic tweaking and culturally transmitted influences on behaviour, must – and can – explain the rest.

      Read More
      • Replies: @JayMan
      Turkheimer’s laws ... say almost nothing about beliefs, table manners, and individual abilities like riding horses, algebra, or tracking game. 
      As Razib said, that's actually BS. It shouldn't take long on my blog to find a long list of behaviors and personality quirks that are quite heritable. Or read Nancy Segal.
      At some point as the generation of phenotypes progresses over time, the slow sweeping tsunami of environmental influences, causing epigenetic tweaking and culturally transmitted influences on behaviour, must – and can – explain the rest. 
      Don't bet on it.
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    2. Razib Khan says:Show Comment

      but they say almost nothing about beliefs, table manners, and individual abilities like riding horses, algebra, or tracking game.

      that’s mostly false. a lot of that is at least moderately heritable through personality or intelligence. even horse-back riding probably has a heritable element in terms of kinesiology.

      the slow sweeping tsunami of environmental influences, causing epigenetic tweaking and culturally transmitted influences on behaviour, must – and can – explain the rest.

      i also would not characterize environment as slow. at least in modern environments. one of the reasons that people underestimate heritability is the rapidly fluctuating ‘norm of reaction’ issue.

      Read More
      • Replies: @Helga Vierich
      You are surely not actually saying that whether I eat with chopsticks, my fingers, or knife and fork has a genetic component? Whether I ride horses, or not, has a genetic component?

      Or are you are using the fact that there IS a genetic component to how well I use my body to learn physical skills, to say that my point is “mostly false”?

      Okay, let me clarify this: it is not mostly false. How well I ride a horse may depend on my balance, which might have some genetic component. Granted, there are human beings who are, for possibly genetic reasons, more klutzy than the norm. (I may even be one of them.) However this will conceivably only influence how WELL these people ride a horse, not WHETHER they do. And an individual’s ability to ride a horse or not is going to be a function of their culture, not of their genes. In some cultures, there are domesticated riding horses. In others, there are not. In some cultures people ride bicycles or drive cars; in others, they ride cattle or donkeys or camels. In some, indeed, they do not ride any animal or anything else. This is not genetic. It is cultural. A person who moves from a non-horse culture to a horse culture is not genetically unable to learn to ride a horse. Or ride a bicycle, or do algebra, or learn to use chopsticks... or are you suggesting that these are evidence of my underestimating the heritability in the rapidity of “norms of reaction”? :)

      You perhaps misunderstood my point, and for that reason, I assume that my clarification will bring our views into concordance.

      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    3. Helga Vierich [AKA "Helga Vierch"] says:Show Comment
      @Razib Khan
      but they say almost nothing about beliefs, table manners, and individual abilities like riding horses, algebra, or tracking game.

      that's mostly false. a lot of that is at least moderately heritable through personality or intelligence. even horse-back riding probably has a heritable element in terms of kinesiology.

      i also would not characterize environment as slow. at least in modern environments. one of the reasons that people underestimate heritability is the rapidly fluctuating 'norm of reaction' issue.

      You are surely not actually saying that whether I eat with chopsticks, my fingers, or knife and fork has a genetic component? Whether I ride horses, or not, has a genetic component?

      Or are you are using the fact that there IS a genetic component to how well I use my body to learn physical skills, to say that my point is “mostly false”?

      Okay, let me clarify this: it is not mostly false. How well I ride a horse may depend on my balance, which might have some genetic component. Granted, there are human beings who are, for possibly genetic reasons, more klutzy than the norm. (I may even be one of them.) However this will conceivably only influence how WELL these people ride a horse, not WHETHER they do. And an individual’s ability to ride a horse or not is going to be a function of their culture, not of their genes. In some cultures, there are domesticated riding horses. In others, there are not. In some cultures people ride bicycles or drive cars; in others, they ride cattle or donkeys or camels. In some, indeed, they do not ride any animal or anything else. This is not genetic. It is cultural. A person who moves from a non-horse culture to a horse culture is not genetically unable to learn to ride a horse. Or ride a bicycle, or do algebra, or learn to use chopsticks… or are you suggesting that these are evidence of my underestimating the heritability in the rapidity of “norms of reaction”? 🙂

      You perhaps misunderstood my point, and for that reason, I assume that my clarification will bring our views into concordance.

      Read More
      • Replies: @Razib Khan
      i thought you were talking about heritability.

      as someone whose parents are of a different culture, i am aware of inter-cultural difference. i'm actually suspecting this is one reason i was more 'blank slate' in orientation when i was a teenager, as the correlation between my parents was so much stronger than anything between their offspring and themselves. that being said, for most americans this is not their experience, as they are raised by parents of broadly similar culture.

      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    4. Karl Zimmerman says:Show Comment

      I’m beginning to wonder how long it will be before theists notice that a significant amount of personality variation isn’t apparently due to genetics or upbringing, and try to rope it in as evidence of an immortal soul (or alternatively, reincarnation). Despite being a logical fallacy, many marginally religious persons seem to be a fan of many “God of the gaps” type of arguments after all.

      Read More
      • Replies: @notanon
      Isn't that epigenetics?
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    5. @Helga Vierich
      You are surely not actually saying that whether I eat with chopsticks, my fingers, or knife and fork has a genetic component? Whether I ride horses, or not, has a genetic component?

      Or are you are using the fact that there IS a genetic component to how well I use my body to learn physical skills, to say that my point is “mostly false”?

      Okay, let me clarify this: it is not mostly false. How well I ride a horse may depend on my balance, which might have some genetic component. Granted, there are human beings who are, for possibly genetic reasons, more klutzy than the norm. (I may even be one of them.) However this will conceivably only influence how WELL these people ride a horse, not WHETHER they do. And an individual’s ability to ride a horse or not is going to be a function of their culture, not of their genes. In some cultures, there are domesticated riding horses. In others, there are not. In some cultures people ride bicycles or drive cars; in others, they ride cattle or donkeys or camels. In some, indeed, they do not ride any animal or anything else. This is not genetic. It is cultural. A person who moves from a non-horse culture to a horse culture is not genetically unable to learn to ride a horse. Or ride a bicycle, or do algebra, or learn to use chopsticks... or are you suggesting that these are evidence of my underestimating the heritability in the rapidity of “norms of reaction”? :)

      You perhaps misunderstood my point, and for that reason, I assume that my clarification will bring our views into concordance.

      i thought you were talking about heritability.

      as someone whose parents are of a different culture, i am aware of inter-cultural difference. i’m actually suspecting this is one reason i was more ‘blank slate’ in orientation when i was a teenager, as the correlation between my parents was so much stronger than anything between their offspring and themselves. that being said, for most americans this is not their experience, as they are raised by parents of broadly similar culture.

      Read More
      • Replies: @Helga Vierich
      I see. I am a bit perplexed that such things as “beliefs, table manners, and individual abilities like riding horses, algebra, or tracking game” did not conjure up the idea that hess might be examples of behaviour best explained by “cultural” variations.

      I meant to contrast this list with “genetic” explanations, not the more broad category of “heritability” explanations. Table manners, belief, and even skills like knitting and tracking game can indeed be heritable patterns of behaviour and skill sets.

      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    6. jb says:Show Comment

      I’m going to ask a question that might not make sense, but I’m going to ask it anyway.

      It seems to me that the more genes there are that influence a given trait, the less that trait should vary — at least in a thoroughly mixed population. For example, because the number of genes that influence skin color is small, even a well mixed population of European and African origin will contain light skinned and dark skinned individuals, just by luck of the draw. But if there were a thousand genes for skin color that couldn’t happen, because once the two populations were mixed, everyone would cluster around the average in terms of light and dark genes, and no one would ever approach the coloration of either original population.

      So my question is this: if human behavioral traits are “associated with very many genetic variants,” why is there still so much variation in those traits?

      I suppose one answer is that there isn’t: that there is in fact very little variation in human behavioral traits, and that what little there is only seems like a lot to us because it’s all we ever see. This would imply that there is the potential for really huge variation that we never see, because nobody every gets dealt a genetic hand with that many cards pushing them in a particular direction. E.g., it would imply that an IQ of 250 is possible, but that it’s just too unlikely to ever happen. (I’m not sure what that kind of variation with other traits, like extroversion, would even look like!) With physical variation — e.g., height — you reach a point where it just isn’t going to happen. If you got all the height genes there were, you probably wouldn’t be able to stand up. Maybe it’s the same for behavioral traits.

      Anyway, I’d like to know if you think my question makes any sense, and if it does how would you answer it.

      Read More
      • Replies: @Razib Khan
      http://en.wikipedia.org/wiki/Central_limit_theorem

      a million coin flips converges to the distribution.

      re: real traits, in reality the tails are fat, and the normal distribution approximation works best +/- 2 standard deviations. the genetic architecture might get weirder at extremes.

      , @BDoyle
      jb, you can't look at IQ as some kind of absolute measure of intelligence. The values are crammed into a normal distribution by definition, so all it means is that some kind-of-smart people decided that was the way to do it.
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    7. @Karl Zimmerman
      I'm beginning to wonder how long it will be before theists notice that a significant amount of personality variation isn't apparently due to genetics or upbringing, and try to rope it in as evidence of an immortal soul (or alternatively, reincarnation). Despite being a logical fallacy, many marginally religious persons seem to be a fan of many "God of the gaps" type of arguments after all.

      Isn’t that epigenetics?

      Read More
      • Replies: @Razib Khan
      no.
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    8. @notanon
      Isn't that epigenetics?

      no.

      Read More
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    9. @Razib Khan
      i thought you were talking about heritability.

      as someone whose parents are of a different culture, i am aware of inter-cultural difference. i'm actually suspecting this is one reason i was more 'blank slate' in orientation when i was a teenager, as the correlation between my parents was so much stronger than anything between their offspring and themselves. that being said, for most americans this is not their experience, as they are raised by parents of broadly similar culture.

      I see. I am a bit perplexed that such things as “beliefs, table manners, and individual abilities like riding horses, algebra, or tracking game” did not conjure up the idea that hess might be examples of behaviour best explained by “cultural” variations.

      I meant to contrast this list with “genetic” explanations, not the more broad category of “heritability” explanations. Table manners, belief, and even skills like knitting and tracking game can indeed be heritable patterns of behaviour and skill sets.

      Read More
      • Replies: @Razib Khan
      I meant to contrast this list with “genetic” explanations, not the more broad category of “heritability” explanations. Table manners, belief, and even skills like knitting and tracking game can indeed be heritable patterns of behaviour and skill sets.

      why are you using quotes around heritable and genetic? those are precise terms in genetics. i'm a geneticist, so i'm going to use them as a geneticist would.

      http://en.wikipedia.org/wiki/Heritability#Definition

      many genetic traits are not heritable because there is no population variance. if a behavior genetic context ironically it doesn't make much sense to talk about a trait as "genetic" since behavior genetics is concerned with variation within the population.

      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    10. Razib Khan says:Show Comment
      @Helga Vierich
      I see. I am a bit perplexed that such things as “beliefs, table manners, and individual abilities like riding horses, algebra, or tracking game” did not conjure up the idea that hess might be examples of behaviour best explained by “cultural” variations.

      I meant to contrast this list with “genetic” explanations, not the more broad category of “heritability” explanations. Table manners, belief, and even skills like knitting and tracking game can indeed be heritable patterns of behaviour and skill sets.

      I meant to contrast this list with “genetic” explanations, not the more broad category of “heritability” explanations. Table manners, belief, and even skills like knitting and tracking game can indeed be heritable patterns of behaviour and skill sets.

      why are you using quotes around heritable and genetic? those are precise terms in genetics. i’m a geneticist, so i’m going to use them as a geneticist would.

      http://en.wikipedia.org/wiki/Heritability#Definition

      many genetic traits are not heritable because there is no population variance. if a behavior genetic context ironically it doesn’t make much sense to talk about a trait as “genetic” since behavior genetics is concerned with variation within the population.

      Read More
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    11. @jb
      I'm going to ask a question that might not make sense, but I'm going to ask it anyway.

      It seems to me that the more genes there are that influence a given trait, the less that trait should vary -- at least in a thoroughly mixed population. For example, because the number of genes that influence skin color is small, even a well mixed population of European and African origin will contain light skinned and dark skinned individuals, just by luck of the draw. But if there were a thousand genes for skin color that couldn't happen, because once the two populations were mixed, everyone would cluster around the average in terms of light and dark genes, and no one would ever approach the coloration of either original population.

      So my question is this: if human behavioral traits are "associated with very many genetic variants," why is there still so much variation in those traits?

      I suppose one answer is that there isn't: that there is in fact very little variation in human behavioral traits, and that what little there is only seems like a lot to us because it's all we ever see. This would imply that there is the potential for really huge variation that we never see, because nobody every gets dealt a genetic hand with that many cards pushing them in a particular direction. E.g., it would imply that an IQ of 250 is possible, but that it's just too unlikely to ever happen. (I'm not sure what that kind of variation with other traits, like extroversion, would even look like!) With physical variation -- e.g., height -- you reach a point where it just isn't going to happen. If you got all the height genes there were, you probably wouldn't be able to stand up. Maybe it's the same for behavioral traits.

      Anyway, I'd like to know if you think my question makes any sense, and if it does how would you answer it.

      http://en.wikipedia.org/wiki/Central_limit_theorem

      a million coin flips converges to the distribution.

      re: real traits, in reality the tails are fat, and the normal distribution approximation works best +/- 2 standard deviations. the genetic architecture might get weirder at extremes.

      Read More
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    12. JayMan says: • WebsiteShow Comment
      @Helga Vierich
      Excellent. Finally we are getting somewhere. Yes, so this now makes sense of the elusive and finely sliced gradations in human individual personality. That is only one aspect of human “behaviour”, however.

      Turkheimer’s laws do, indeed, explain at least a little of the remarkable individual variation in musical talent, flare for artistic expression or imagination that we can find even in a single set of siblings... but they say almost nothing about beliefs, table manners, and individual abilities like riding horses, algebra, or tracking game.

      At some point as the generation of phenotypes progresses over time, the slow sweeping tsunami of environmental influences, causing epigenetic tweaking and culturally transmitted influences on behaviour, must - and can - explain the rest.

      Turkheimer’s laws … say almost nothing about beliefs, table manners, and individual abilities like riding horses, algebra, or tracking game.

      As Razib said, that’s actually BS. It shouldn’t take long on my blog to find a long list of behaviors and personality quirks that are quite heritable. Or read Nancy Segal.

      At some point as the generation of phenotypes progresses over time, the slow sweeping tsunami of environmental influences, causing epigenetic tweaking and culturally transmitted influences on behaviour, must – and can – explain the rest.

      Don’t bet on it.

      Read More
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    13. bbartlog says:Show Comment

      Do you have any idea on what basis the authors dismiss the idea that much of the variance is accounted for by rare(r) variants of large(r) effect? There seems to be a lot of room for such things, and I don’t really see an easy way to rule it out – but maybe they have some clever statistical argument to that effect?

      Read More
      • Replies: @Razib Khan
      1) for certain traits the capture of heritability has actually followed theoretical predictions based on the common variant model. height for example. basically visscher predicted years before the sample sizes got large that N sample size = capture X % of the heritability. for height everything is one schedule.

      2) they cite literature where this just isn't found when the appropriate technical tools are used to look for these variants.

      3) there are theoretical reasons why large effect variants should not be the major drivers of phenotypic variation.

      #1 and #2 are more important for me.
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    14. Razib Khan says:Show Comment
      @bbartlog
      Do you have any idea on what basis the authors dismiss the idea that much of the variance is accounted for by rare(r) variants of large(r) effect? There seems to be a lot of room for such things, and I don't really see an easy way to rule it out - but maybe they have some clever statistical argument to that effect?

      1) for certain traits the capture of heritability has actually followed theoretical predictions based on the common variant model. height for example. basically visscher predicted years before the sample sizes got large that N sample size = capture X % of the heritability. for height everything is one schedule.

      2) they cite literature where this just isn’t found when the appropriate technical tools are used to look for these variants.

      3) there are theoretical reasons why large effect variants should not be the major drivers of phenotypic variation.

      #1 and #2 are more important for me.

      Read More
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    15. Senator Brundlefly says:Show Comment

      Considering that there seems to be a popular ethos more or less geared toward pure nurturism, I’m surprised that the first law of behavior genetics isn’t YOU DO NOT TALK ABOUT BEHAVIOR GENETICS. Based on your experience, is the field of psychology as a whole open to genetic explanations? Or is it only a subset of rebels with the aforementioned authors as examples?

      Read More
      • Replies: @Robert Ford
      i majored in Beh. PSY i'd say no. they have absolutely no idea what's going on. they don't even like to acknowledge neuroscience. i graduated ten years ago but, based on what i overhear in classes now, i'd say not much has changed. thinking back, it often seems like a cult as there is a lot of bizarre rule learning and if they do acknowledge the "outside world" people might realize they have no reason to exist anymore.
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    16. Robert Ford says: • WebsiteShow Comment
      @Senator Brundlefly
      Considering that there seems to be a popular ethos more or less geared toward pure nurturism, I'm surprised that the first law of behavior genetics isn't YOU DO NOT TALK ABOUT BEHAVIOR GENETICS. Based on your experience, is the field of psychology as a whole open to genetic explanations? Or is it only a subset of rebels with the aforementioned authors as examples?

      i majored in Beh. PSY i’d say no. they have absolutely no idea what’s going on. they don’t even like to acknowledge neuroscience. i graduated ten years ago but, based on what i overhear in classes now, i’d say not much has changed. thinking back, it often seems like a cult as there is a lot of bizarre rule learning and if they do acknowledge the “outside world” people might realize they have no reason to exist anymore.

      Read More
      • Replies: @Senator Brundlefly
      Cool thanks. I was hoping it was just popular culture's conception of the field. Guess that conception derives from the majority of its practitioners.
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    17. @jb
      I'm going to ask a question that might not make sense, but I'm going to ask it anyway.

      It seems to me that the more genes there are that influence a given trait, the less that trait should vary -- at least in a thoroughly mixed population. For example, because the number of genes that influence skin color is small, even a well mixed population of European and African origin will contain light skinned and dark skinned individuals, just by luck of the draw. But if there were a thousand genes for skin color that couldn't happen, because once the two populations were mixed, everyone would cluster around the average in terms of light and dark genes, and no one would ever approach the coloration of either original population.

      So my question is this: if human behavioral traits are "associated with very many genetic variants," why is there still so much variation in those traits?

      I suppose one answer is that there isn't: that there is in fact very little variation in human behavioral traits, and that what little there is only seems like a lot to us because it's all we ever see. This would imply that there is the potential for really huge variation that we never see, because nobody every gets dealt a genetic hand with that many cards pushing them in a particular direction. E.g., it would imply that an IQ of 250 is possible, but that it's just too unlikely to ever happen. (I'm not sure what that kind of variation with other traits, like extroversion, would even look like!) With physical variation -- e.g., height -- you reach a point where it just isn't going to happen. If you got all the height genes there were, you probably wouldn't be able to stand up. Maybe it's the same for behavioral traits.

      Anyway, I'd like to know if you think my question makes any sense, and if it does how would you answer it.

      jb, you can’t look at IQ as some kind of absolute measure of intelligence. The values are crammed into a normal distribution by definition, so all it means is that some kind-of-smart people decided that was the way to do it.

      Read More
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    18. another fred says:Show Comment

      Just wanted to say thanks for this article so that Razib and the mods will know that it is appreciated by some non-geneticists. Not being trained in, and not being willing to dedicate the time to learn, the math of genetics, some of that goes over my head, but I think I get the gist of it.

      Thanks

      Read More
      ReplyAgree/Disagree/Etc. More... This Commenter Display All Comments
    19. @Robert Ford
      i majored in Beh. PSY i'd say no. they have absolutely no idea what's going on. they don't even like to acknowledge neuroscience. i graduated ten years ago but, based on what i overhear in classes now, i'd say not much has changed. thinking back, it often seems like a cult as there is a lot of bizarre rule learning and if they do acknowledge the "outside world" people might realize they have no reason to exist anymore.

      Cool thanks. I was hoping it was just popular culture’s conception of the field. Guess that conception derives from the majority of its practitioners.

      Read More
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    20. M says:Show Comment

      @JB

      I’m not sure about population level variation.

      However, a point I don’t see mentioned often is (I think, and I may be wrong) that part of why children differ from their parental mean on traits which are substantially genetic, despite getting equal contributions from both, is that the child gets a random sampling of information stretches of either of their parents’ two chromosomes at a position (to reduce the information in their parents four chromosomes down to the child’s two chromosome size).

      Random loss of some information in this “chunky” way from the parents thereby helps to create the variation in the set of children (in a way that wouldn’t hold in blending or very fine scale combination from the parents chromosomes). Sets of children thus end up with substantial variation in traits which are substantially genetic (even though the average for all the children taken tends towards the parental average); family level variation is still quite large compared to population level variation, from this effect.

      Not sure if anything like this random sampling process effect holds at the population level. (I expect the answer would be “No, as any population genetics textbook would tell you!”).

      Read More
      ReplyAgree/Disagree/Etc. More... This Commenter Display All Comments
    21. Helga Vierich says:Show Comment

      Well thank you for pointing out the problem – I did not mean to annoy with my quote marks, only to stress that I was using the terms to distinguish between obligate genetic effects, like having a digestive system, from the aspects of variation that could be at least partially attributed to environment, like frequencies of cancer of the digestive system, which appear to owe as much to diet and stress levels as to genetics, and are therefore a function of cultural environment.

      I am sorry if my reference to heritability strayed from the sense you are familiar with. But if you will recall, your post was concerned with behavioural variability, and as such, the degree to which environment might contribute to the variation seems rather large. But perhaps you might explain my error a bit further, as I am not really sure where I went astray? As far as I know, the term is frequently used to indicate that part of phenotypic variation in a given population is genetic, true, but not without some reference to the role of environment. Moreover, there is an area, concerning the distinction between obligate and facultative aspects of the phenotype, where it becomes a bit vague as “heritability estimates are always relative to the genetic and environmental factors in the population, and are not absolute measurements of the contribution of genetic and environmental factors to a phenotype”.

      Heritability estimates, therefore, as far as I know, reflect the amount of variation in genotypic and in environmental effects, as the wiki blurb I just quoted. The various contributors to this source suggest, as an example of the latter, which language a person speaks.

      Now language and certain other environmental components might also passed down from one generation to the next, and so across many generations, but due to a different kind of inheritance. Functioning through the medium of social learning and shared customs and values, such cultural environments, through things like parenting norms, dietary preferences, and activity regimes lead to behavioural continuities over time, and these are superimposed on the basic human genomic variation.

      I think it is worth considering that some things about a particular population, that might initially have been assumed were largely due to obligate genetic factors, (such as are indicated by statistics on adult height, rates of maturation, frequency of mental illness, and physical skills) but appear now to have have a significant environmental component. In many cases, cultural practices relating to health constitute a very major part of this.

      You state the fourth law thus ““A typical human behavioral trait is associated with very many genetic variants, each of which accounts for a very small percentage of the behavioral variability.”

      Where is the environment in this formulation? The first three Laws you list do not actually leave it out; in fact they seem to make it progressively more powerful as an explanatory tool.

      * First Law. All human behavioral traits are heritable.

      * Second Law. The effect of being raised in the same family is smaller than the effect of genes.

      * Third Law. A substantial portion of the variation in complex human behavioral traits is not accounted for by the effects of genes or families.

      All of these point to a third aspect; neither genes nor families, but also a heritable aspect of behavioural variation. What constitutes the biggest and most obvious aspect of human behavioural variation across the globe? It is culture… it is those various systems of economy, of social organization and association beyond the mere “family” unit, and systems of of codes and manners and paradigms of explanation, that constitute culture.

      Perhaps I have misunderstood what you mean by “a typical human behaviour trait”? Perhaps you mean universal kinds of behaviours? Perhaps you mean the use of symbolic representations in communicating through story telling, dance, art, and music? Agreed, there will be genetic variations involved in the facility of individuals in the use of the various aspects of symbolic communication. However, environment also plays a role in the very development of the brain and nervous system that make human behaviour happen. Neural tissues are affected by toxins, malnutrition, infections, and stress even during the prenatal period, and certainly after. The environment certainly must be acknowledged in explaining some part of the variation we see in human facility with symbolic communication … and every other aspect of human behaviour.

      But no, you made a statement, your fourth law, with would seem to preclude that. “A typical human behavioural trait” you suggest, “is associated with very many genetic variants” . Well yes, but where do we find epigenetic phenomena accounted for here? And where is the role of learned a shared behaviour?

      Most behaviours are influenced by particular cultural environments, which act on other manifestations of human nature, variously dialling up or down, cognitive processing and behavioural responses to things like sensitivity to unfairness, need for secure attachment to caregivers when young, and so on. These vary across the whole species, but also appear related to particular patterns of cultural difference. Where should we first seek an explanatory hypothesis, for example, for a human population which produces a high proportion of impulsive and aggressive individuals, compared to another population where they constitute any a small proportion of the variation?

      In the genome, looking for polymorphism differences that might account for the variation in that behavioural trait? That was tried with the DR4 variant repeats, and with the MAO-variants of the primary amine oxidase activity in mitochondria. Parenting and other aspects of social environment turned out to be crucial there. Variation in family life? But the second and third laws state that this would explain very little of the variation, even though, according to the first law, all human behaviour is heritable.

      Clearly there is a point at which heritability is not just about genes, it is also about the “variation in environmental effects”: epigenetic and developmental and cultural.

      To sum up, you cannot have a law about genetic human behavioural variation that stands on its own.

      Read More
      • Replies: @Razib Khan
      alright, i read your whole comment

      first, Perhaps I have misunderstood what you mean by “a typical human behaviour trait”? i think they're talking most about personality. i can ask james lee directly if you are curious, or request he leave a comment to clarify.

      second, i don't really know how to respond to a lot of your comment. i don't see what you are getting at, as much of what you are saying is tautological to what behavior genetics is interested in.

      third, epigenetics is a buzzword. it's real, but waving it around like that is only going to annoy me. some of my best friends are epigeneticists (literally), so i certainly know its validity as a science, but it's being used like magic right now. probably relevant

      www.psmag.com/health-and-behavior/epigenetics-not-revolutionizing-biology
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
    22. Razib Khan says:Show Comment
      @Helga Vierich
      Well thank you for pointing out the problem - I did not mean to annoy with my quote marks, only to stress that I was using the terms to distinguish between obligate genetic effects, like having a digestive system, from the aspects of variation that could be at least partially attributed to environment, like frequencies of cancer of the digestive system, which appear to owe as much to diet and stress levels as to genetics, and are therefore a function of cultural environment.

      I am sorry if my reference to heritability strayed from the sense you are familiar with. But if you will recall, your post was concerned with behavioural variability, and as such, the degree to which environment might contribute to the variation seems rather large. But perhaps you might explain my error a bit further, as I am not really sure where I went astray? As far as I know, the term is frequently used to indicate that part of phenotypic variation in a given population is genetic, true, but not without some reference to the role of environment. Moreover, there is an area, concerning the distinction between obligate and facultative aspects of the phenotype, where it becomes a bit vague as “heritability estimates are always relative to the genetic and environmental factors in the population, and are not absolute measurements of the contribution of genetic and environmental factors to a phenotype".

      Heritability estimates, therefore, as far as I know, reflect the amount of variation in genotypic and in environmental effects, as the wiki blurb I just quoted. The various contributors to this source suggest, as an example of the latter, which language a person speaks.

      Now language and certain other environmental components might also passed down from one generation to the next, and so across many generations, but due to a different kind of inheritance. Functioning through the medium of social learning and shared customs and values, such cultural environments, through things like parenting norms, dietary preferences, and activity regimes lead to behavioural continuities over time, and these are superimposed on the basic human genomic variation.

      I think it is worth considering that some things about a particular population, that might initially have been assumed were largely due to obligate genetic factors, (such as are indicated by statistics on adult height, rates of maturation, frequency of mental illness, and physical skills) but appear now to have have a significant environmental component. In many cases, cultural practices relating to health constitute a very major part of this.

      You state the fourth law thus ““A typical human behavioral trait is associated with very many genetic variants, each of which accounts for a very small percentage of the behavioral variability."

      Where is the environment in this formulation? The first three Laws you list do not actually leave it out; in fact they seem to make it progressively more powerful as an explanatory tool.

      .

      All of these point to a third aspect; neither genes nor families, but also a heritable aspect of behavioural variation. What constitutes the biggest and most obvious aspect of human behavioural variation across the globe? It is culture... it is those various systems of economy, of social organization and association beyond the mere “family” unit, and systems of of codes and manners and paradigms of explanation, that constitute culture.

      Perhaps I have misunderstood what you mean by “a typical human behaviour trait”? Perhaps you mean universal kinds of behaviours? Perhaps you mean the use of symbolic representations in communicating through story telling, dance, art, and music? Agreed, there will be genetic variations involved in the facility of individuals in the use of the various aspects of symbolic communication. However, environment also plays a role in the very development of the brain and nervous system that make human behaviour happen. Neural tissues are affected by toxins, malnutrition, infections, and stress even during the prenatal period, and certainly after. The environment certainly must be acknowledged in explaining some part of the variation we see in human facility with symbolic communication ... and every other aspect of human behaviour.

      But no, you made a statement, your fourth law, with would seem to preclude that. "A typical human behavioural trait” you suggest, “is associated with very many genetic variants” . Well yes, but where do we find epigenetic phenomena accounted for here? And where is the role of learned a shared behaviour?

      Most behaviours are influenced by particular cultural environments, which act on other manifestations of human nature, variously dialling up or down, cognitive processing and behavioural responses to things like sensitivity to unfairness, need for secure attachment to caregivers when young, and so on. These vary across the whole species, but also appear related to particular patterns of cultural difference. Where should we first seek an explanatory hypothesis, for example, for a human population which produces a high proportion of impulsive and aggressive individuals, compared to another population where they constitute any a small proportion of the variation?

      In the genome, looking for polymorphism differences that might account for the variation in that behavioural trait? That was tried with the DR4 variant repeats, and with the MAO-variants of the primary amine oxidase activity in mitochondria. Parenting and other aspects of social environment turned out to be crucial there. Variation in family life? But the second and third laws state that this would explain very little of the variation, even though, according to the first law, all human behaviour is heritable.

      Clearly there is a point at which heritability is not just about genes, it is also about the “variation in environmental effects": epigenetic and developmental and cultural.

      To sum up, you cannot have a law about genetic human behavioural variation that stands on its own.

      alright, i read your whole comment

      first, Perhaps I have misunderstood what you mean by “a typical human behaviour trait”? i think they’re talking most about personality. i can ask james lee directly if you are curious, or request he leave a comment to clarify.

      second, i don’t really know how to respond to a lot of your comment. i don’t see what you are getting at, as much of what you are saying is tautological to what behavior genetics is interested in.

      third, epigenetics is a buzzword. it’s real, but waving it around like that is only going to annoy me. some of my best friends are epigeneticists (literally), so i certainly know its validity as a science, but it’s being used like magic right now. probably relevant

      http://www.psmag.com/health-and-behavior/epigenetics-not-revolutionizing-biology

      Read More
      ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments

    Comments are closed.

    Subscribe to All Razib Khan Comments via RSS