How culture affects genes and vice versa

March 3, 2010 • 8:10 am

A piece in yesterday’s science section of The New York Times drew my attention to a new article by Laland et al. in Nature Reviews Genetics summarizing the intriguing reciprocal interactions between human culture and human genes. It’s an interesting article, well worth reading if you’d like to see the recently-acquired evidence for natural selection in our species, and how that selection might have been affected by our culture (and vice versa).

In principle, any species that has some kind of culture (which the authors define as “information that is capable of affecting individuals’ behaviour, which they acquire from other individuals through teaching, imitation, and other forms of social learning”) can show these mutual interactions between DNA and culture: the so-called “gene-culture co-evolution.” And any species that can change the environment through its own activities, like moles digging burrows, can be viewed through the lens of “niche construction,” in which an organism actually affects the ways that natural selection impinges on it by their own behavior, evolved or otherwise. Once moles “decided” to go underground (and certainly the initial stages of that “decision” may have been based partly on genes), they then subjected themselves to all kinds of novel selection.  One is selection to lose or reduce their eyes, which aren’t needed in a dark tunnel and, indeed, can be a source of infection, or use metabolic energy that could be more profitably diverted to other traits.

These phenomena may be especially important in humans since we have a rich culture, much of which does not rest on genetic adaptations, and that culture can affect our susceptibility to natural selection. We have “invented” doctors and antibiotics, for example, so infections that would have felled our ancestors (and selected for resistance to bacteria) is no longer so strong.  Laland et al. discuss several examples like this.

The article gives examples of each phenomenon. Perhaps the best example of gene-culture coevolution is one that I discuss in WEIT: the evolution of lactose tolerance in those “pastoral” human populations that keep cows, goats, and sheep for milk.  This was a cultural change, but one that seems to have profoundly influenced the evolution of at least one gene.

Whether or not you can digest lactose (milk sugar) depends on whether you have one of two forms of the gene producing the enzyme lactase.  Ancestral human populations were lactose intolerant as adults: while babies obviously needed to drink milk, adults did not, and so the gene producing lactase (the enzyme breaking down lactose) became inactivated during development, as it still does in many humans. This is the form of the gene that produces lactose-intolerance in its carriers.

(Analysis of DNA from ancient human bones showed that, as predicted, they had the “intolerant” form of the gene: a cool result.)

But, starting about 9,000 years ago, some human populations became pastoral, and those populations now show a high frequency of lactase persistence: the gene breaking down lactose is not turned off, so that adults can digest and get nutrients from milk sugar.  So they have the “new” derived form of the lactase gene, the “tolerance” allele.  Groups that historically were not pastoral, like many Asian and African populations, retain the ancestral intolerance allele—the one that gets switched off in adults. The difference between the “on” and “off” versions of the lactase gene is based on a simple DNA change in its regulatory region.

What is nice is that we can, through population-genetic analysis, get an idea of when the new “switched-on” form of the gene arose by mutation.  It was between 3,000 and 8,000 years ago, making its appearance and rise coincident with the rise of pastoralism—and the huge energy advantage of drinking milk.   One can draw a pretty strong conclusion that individuals able to use this rich new source of food—a source deriving from the cultural adoption of pastoral behavior—had a selective advantage over non-tolerant individuals.  You can even calculate this advantage from population-genetic analysis of how fast the “tolerant” gene increased in frequency.  Apparently, tolerant individuals in pastoral cultures would have left 4-10% more offspring than non-tolerant individuals, probably because they were better fed.  This is pretty strong selection, and would have promoted rapid genetic change.

The evolution of lactose tolerance is a splendid example of gene-culture coevolution.  The authors give others.  One that I didn’t know about was the correlation between yam cultivation in Africa and the frequency of the sickle-cell hemoglobin allele, which in heterozygous form confers some protection against deadly malaria.  Apparently those human populations that cultivate yams, which involves cutting down forest, have a higher frequency of this allele than those who don’t grow yams.  This genetic difference is ascribed to one of the byproducts of yam culture: the creation of pools of water that serve as breeding sites for mosquitoes.

The authors also suggest a hypothesis that I’ve long found appealing (see WEIT): many of the superficial physical differences between human populations—those differences affecting hair texture, facial features like eye and nose shape, and the like—could reflect sexual selection, but sexual selection based on cultural preferences.  I like this idea because, first, differences in appearance between human ethnic groups must have arisen very quickly, since they presumably weren’t present when our ancestors migrated out of Africa less than 100,000 years ago; and sexual selection motivated by cultural preference can be incredibly strong.  Second, genetic differences between human groups are much larger for genes affecting external physical traits than for genes in general, which are pretty homogeneous among groups.  To me, this again suggests sexual selection.

The authors give a big table of genes (Table 2) in our species that may have experienced natural (or sexual) selection deriving from human culture.  There are lots of them, and of course the evidence is stronger for some than for others. Population-genetic analysis that can pinpoint “the footprints of selection” on DNA sequences is not perfect, and there can be false positives.  I, for one, am dubious about the MYH16 gene.  The authors note that “this gene is expressed primarily in the hominid mandible, and its loss is thought to result in a massive reduction in jaw muscle, with a timing that may coincide with the appearance of cooking.”  This may be true, but showing that its loss really did have that phenotypic effect may be hard (it’s impossible to do those experiments in humans!), and showing that its evolution may reflect relaxed natural selection for chewing ability even harder.  Such cases may always remain speculative.

Nevertheless, the idea of gene-culture coevolution is a good one, and certainly must explain some recent cases of selection in our species.  Likewise, the idea of “niche construction” bears thinking about, though I think its ubiquity may have been overstated by some of its proponents.  Yes, some species can affect their own evolution through their behavior, but in other cases species must certainly be passive responders to the environment.  Does the hoof of the chamois really have an effect on the tensile properties of Swiss granite?  Do those Lithops plants that resemble stones have any effect on the appearance of the stones themselves?  Does the polar bear’s white coat have any effect on the reflective properties of ice and snow? (I suppose one could counter that the bear subjected itself to coat-color selection by “choosing” to live in cold climates—if it did!).

At any rate, if you’re not up on the latest evidence for selection in our species, and how that selection may have been driven by our culture, the Laland et al. paper is a good place to start.


Laland, K. N., J. Odling-Smee, and S. Myles.  2010.  How culture shaped the human genome: bringing genetics and the human sciences together. Nature Reviews Genetics 11, 137-148 (doi:10.1038/nrg2734).

19 thoughts on “How culture affects genes and vice versa

  1. I’ve always suspected that there’s a good chance that many ethnic differences/national characteristics are likely due to the leader of a tribe’s having mated with most of the females in the tribe back in time, no different than many animal groups. Cycle that a few generations with successive leaders and you should skew characteristics quite a bit.

  2. I think that the invention of cooking had a huge effect on human evolution in that it allowed our neotenous form and its attendant psychological effects.

    We choose ‘cuteness’ a childlike face in preference to one with a strong enough jaw. Before cooking that would be selected against.

    Most adult animals are not affectionate except to the young, it’s a juvenile trait that allows bonding to parents and siblings and that disappears like lactose tolerance as animal matures.

    If you have the opportunity to successfully choose a mate with a childlike face then you will also be choosing a childlike brain-remember the Siberian foxes-so you will be reinforcing the kind of behaviour that defines modern human society.

    We care about each other, how else could that have happened?

        1. And don’t forget – sometimes they go after us. They like to start up a relationship, and then take out life insurance policies on us.

  3. My response to this supposedly startling discovery is “well, duh.” Didn’t Darwin say 150 years ago that the vast differences of domesticated animals and plants from their wild forebears were the result of human cultural selection factors? The shaping of our own evolution because of the same factors can easily be inferred – and observed.

    1. That was my reaction too when I saw the headline in the NY Times yesterday — so much so that I didn’t even bother reading the article. Reading Jerry’s description of the actual paper makes me think twice about it.

      In my Human Anatomy and Physiology class, I’m covering the digestive system this week and I discussed an example similar to the lactase one, but with salivary amylase and amount of starch in the diet: Perry, et al. 2007. Nature Genetics 39(10):1256-1260.

      The examples are useful to show students that evolutionary change happens all the time.

      1. I’m a physicist and understand the scientific method of proof, as well as the ‘accepted’ method of scientific proof which states that ” a theory is accepted as fact as long as there exists no countervailing theory”.(For example,Einstein’s theory of relativity).

        With that in mind, neither evolution (which has a countervailing theory called creationism), nor creationisn (which which has a countervailing theory called evolution) can be categorized as a science.

        Both evolution and creationism ought to be taught in the philosophy dept.

        Then the student could choose between the theories.

        Just a thought.

    2. Well, it’s one thing to see this as a possibility; it’s another to actually identify those genes that could have changed as a result of culture in our own species (not in domesticated species). The lactase story, for instance, really is new and interesting information.

      1. Yes, the amylase story is pretty compelling too. The authors show increased copy number of the gene (Amy-1), that amylase production is correlated with copy number, and that populations with high starch diets have, on average, higher diploid copy numbers than those with relatively lower starch in the diet.

        There’s also an angle in there to refute Meyer’s lame “gene-duplication-does-not-increase-information” argument…

  4. I’m honestly curious about your response (meaning Jerry) and Richard Dawkins’ to this article. You both seem to view this in a fairly favorable light, or see the evidence as promising in some fashion. Now, am I mistaken to read this article as nearly directly correlated to what David S. Wilson has been pointing out for many years with his “multi-level / group” selection?

    1. No, this article says nothing about multilevel selection, except that some populations can experience selective pressures different from others. That says nothing about the possibility that alleles deleterious in populations, but good for the group as a whole, can evolve.

  5. Looks like a theme for your next book, Jerry Coyne:

    Why Gene-Culture Co-Evolution Is True,
    The Sleuth from Chicago Investigates

  6. There are some genes having an effect on human appearance that likely evolved due to natural, rather than sexual, selection. Loss of skin piemtation in northern European populations is one example.
    But is there a way to “confirm” that some change had to do with sexual selection? For example, Stephen Oppenheimer, author of “the Real Eve” says in his book that no evolutionary advantage to having a skin fold over the eyelid in Asian populations has been found. So likely the force that made it was sexual selection. But is there a way to know for sure.

  7. Hmmm! It’s too late for an old fart to develop cogent arguments, but I am still aware that written language lets me know the past more immediately than my genes

  8. I am copying here a comment I posted elsewhere on this fascinating lactose finding. It speaks to some of the posts above where people are saying things like “wasn’t this predicted by Darwin?” In one sense, no.

    The thing I find interesting here is the fact that there are 4 different genetic variants in 4 different human populations in which lactose tolerance has evolved. This suggests that ancestral populations did not have any variation for this trait (if they did, then different descendant populations would likely have the same variants; they don’t, which suggests de novo mutations in each of the four populations, three of which are in Africa).

    So the mutations arose de novo and then were selected for in the four different populations. But this is the thing that I find to be the most interesting possibility: what if the selection pressures of the pastoral society somehow directed or induced the mutations to occur in each of these 4 populations?

    Now, most biologists would pounce all over that suggestion. It’s Lamarckian- there’s no good evidence for that!

    Well, there is starting to be some evidence that Lamarck may have been partially right – in a limited sense (not in the sense of the giraffe reaching for the leaves and passing it on to it’s kids, though).

    But this idea of de novo mutation is actually more like something codified later, and mostly forgotten by biologists. It’s called the Baldwin Effect. Baldwin proposed that experience actually increases the likelihood of certain variants arising.

    [and look! there’s the Lactose example in the Baldwin Effect wiki 🙂 ]

    In the early 50’s, Conrad Waddington kind of went in the same direction with his genetic assimilation concept, but appeared to back away from the harder-core Baldwin Effect-like ideas in later publications. Who knows? Maybe from pressure from his skeptical colleagues?

    Anyway, it seems to me that this is testable. If the de novo mutations arose totally by chance in these 4 populations, then such lactose-tolerance variants should be present in other populations as well (it’s just that they haven’t been selected for; but they should be there, cryptically, presumably at very low frequency).

    BUT…if the de novo mutations arose by a Baldwin Effect-like mechanism, then the variants would not be found – even cryptically – in other populations, except by migration from these 4 populations in question.

    Will be cool to see how that turns out. Baldwin’s vindication day may yet come! 🙂

  9. RE: Professor Coyne’s statement on moles:

    One is selection to lose or reduce their eyes, which aren’t needed in a dark tunnel and, indeed, can be a source of infection, or use metabolic energy that could be more profitably diverted to other traits.”

    Thank you for explaining this! I’ve never really been crystal clear on why animals lose their eyes in dark habitats.

    I’ve heard the “metabolic energy diversion” argument before, and it makes sense, but it seems like it would be a comparatively weak effect (are eyes truly that “expensive” to operate?) that would require a much longer time for significant selection to occur. In other words, in any one generation, even if some of an early progenitor mole’s offspring were born with reduced vision compared to their siblings, would that really confer a “noticeable” selective advantage on them? Is it possible to give a few examples on how this “reserve” of metabolic energy might make some moles fitter than their better-sighted competitors?

    On the other hand, your statement on infection makes perfect sense, even to a neophyte like me, because obviously such a malady might directly cause premature death, or an inability to find mates — which seems like it would be rather strong selection pressure, in a single generation, as it were.

    And infections could arise nearly anywhere on a creature’s body, so I can now see how this could speed along natural selection in many other instances besides moles and dark tunnels.

  10. It might not be politically correct to say this but this would also apply to the genetic makeup for our brains.

    Simply put if your personality is too “weird” for the society you live in you’re less likely to reproduce. Overtime the cultural norms will have an impact on the genetics of the mind.

    The “racial eugenicists” were right that the genetics of different cultures were different(as an average, a generality, there will always be a small part of any population that deviates far from the norm). What they were wrong about was that theirs was the best. There is no “best”.

    A few factors will determine just how concentrated a society is at a single cognitive type or how diffuse and diverse it is. I believe more cognitive diversity is a strength for a society, allowing more collaboration between different people.

    One factor is just a general attitude of tolerance. The more tolerance is encouraged, the more people are taught to get to know someone before rejecting them the broader the cognitive genetics. This attitude would be seen the most in egalitarian, communal tribes which explains why these cultures often don’t experience mental illness, they accept everybody.

    Another is more acceptance of casual sex. The more casual sex is accepted the greater the chance of the more “weird” people passing on their genes.

    I’ve always felt different, and the more I learn about Latin America the more their culture appeals to me so I think maybe my cognitive genetics match theirs more than America.

Leave a Reply