Sunday, September 02, 2007

Modern women are excellent gatherers

 
Here's an article form this week's New Scientist [Modern women are excellent gatherers]. I'd be curious to know what our adaptationist friends think about it. Is this a good example of how to do research?
Men hunted, women gathered. That is how the division of labour between the sexes is supposed to have been in the distant past. According to a new study, an echo of these abilities can still be found today.

Max Krasnow and colleagues at the University of California, Santa Barbara, have discovered that modern women are better than men at remembering the location of food such as fruit and veg in a market.

The researchers led 86 adults to certain stalls in Santa Barbara's large Saturday farmer's market, then back to a location in the centre of the market from where the stalls could not be seen. They were then asked to point to each stall's location. This requires dead reckoning - a skill that men may once have used to return from hunting, and one that men today still usually perform better than women in experiments. Despite this, the ...
The news report refers to a paper that will soon be published in Proc. Roy. Soc. B [New et al. 2007]. Here's the abstract.
We present evidence for an evolved sexually dimorphic adaptation that activates spatial memory and navigation skills in response to fruits, vegetables and other traditionally gatherable sessile food resources. In spite of extensive evidence for a male advantage on a wide variety of navigational tasks, we demonstrate that a simple but ecologically important shift in content can reverse this sex difference. This effect is predicted by and consistent with the theory that a sexual division in ancestral foraging labour selected for gathering-specific spatial mechanisms, some of which are sexually differentiated. The hypothesis that gathering-specific spatial adaptations exist in the human mind is further supported by our finding that spatial memory is preferentially engaged for resources with higher nutritional quality (e.g. caloric density). This result strongly suggests that the underlying mechanisms evolved in part as adaptations for efficient foraging. Together, these results demonstrate that human spatial cognition is content sensitive, domain specific and designed by natural selection to mesh with important regularities of the ancestral world.
As indicated in the news report, 86 adults (41 women and 45 men) were tested for their ability to remember the location of food stalls in a farmers market. The women were 9% better at this than the men.

The result confirms the authors' hypothesis that women are genetically superior at this task because of adaptation during our hunter-gatherer past.
Silverman & Eals (1992) argue that the female advantage on pencil-and-paper and desktop measures of object location memory reflects a selective pressure on ancestral women for plant-foraging efficiency. But their measures did not involve foods, tested spatial memory on a very small scale, and included no measure of vectoring; as a result, a female advantage on their measures is open to many alternative interpretations. For this reason, we deemed it important to examine whether a female advantage could be demonstrated on a task that closely resembles foraging for plant foods. From this theory, we predicted that women should remember the locations where they have previously encountered immobile resources (e.g. plants, honey) more accurately than do men.
The authors don't explain exactly how this adaptation might have happened. Presumably it went something like this ...

At some time in the ancient past all humans had a single allele for the (unknown) gathering gene. A mutation in this gene arose producing an allele where the ability to gather food was improved. Since women were the principle food gatherers, this mutant allele conferred a selective advantage on women who carried it: presumably because they didn't share their food with their friends who carried the old allele. Over time, the new allele became fixed (or very frequent) in women but men did not benefit.

The first three authors are in Departments of Psychology and the senior author is in a Department of Anthropology.


New, J., Krasnow, M.M, Truxaw, D. and Gaulin, S.J.C. (2007) Spatial adaptations for plant foraging: women excel and calories count. Proc. Roy. Soc. B. DOI 10.1098/rspb.2007.0826.

[Photo Credit: The drawing is "An artist’s impression of early Hunter-Gatherers" from Manx National Heritage]

31 comments :

  1. "I'd be curious to know what our adaptationist friends think about it. Is this a good example of how to do research?"
    You talkin to me?
    I'll bite.
    I can only read the fragment posted (and certainly New Scientist is not a good example of how to publish research), but I see nothing wrong with an approach that runs like this:
    It is thought that for the vast majority of human (pre)history, there was a division of labor between sexes. We hypothesized that correlated neural/behavioral sexual dimorphism evolved as a result. Since no paleohumans are extant, we tested a sample of modern humans...(I can't comment on the experimental design, statistics, etc., but assuming a convincing and robust sexual difference was in fact found:) Our results are consistent with our hypothesis. Alternative explanations for our results might include...?
    What's wrong with that?

    ReplyDelete
  2. I need to point out that I wrote the above comment when only the first boxed quote (from New Scientist) had been posted, or at least was visible to me.
    I still don't have a big problem with it. Apparently it's a real, robust difference in behavioral performance between sexes. Apparently it's vegetable-specific. There is no doubt that hard-wired behavioral differences between sexes exist; this one could plausibly be based (proximately) on something as simple as expression of estrogen-receptor-specific transcription factors in part of a developing brain. It's not hard (for me, anyway, with my trusty hammer) to envision meaningful variation in, say, milk quality resulting from variation in ability to remember where food has been found before. Seems entirely plausible to me. The wording in the abstract seems a little stronger than I might like, but then I haven't read the article.
    Assuming the effect is real, what's your pluralistic alternative explanation?

    ReplyDelete
  3. This gives me a feeling of unease, but I don't think I can do a good job of explaining why. I think it's because the connection between the observation and the conclusion seems tenuous to me.

    To give just one observation - if women had done most of the shopping previously, and if the location of various items within the market was reasonably standardized, then it would hardly be surprising for women to score better.

    However (to beat the drum again), I'd still be interested in an opinion from a Real Live Scientist to my comment in the previous item ...

    Serious question from a lay reader - I'd had the impression over the last few years that over-emphasis on adaptation was dying away. Is it making a comeback, or was my impression wrong?

    ReplyDelete
  4. The blog entry says: "Presumably it went something like this ...

    At some time in the ancient past all humans had a single allele for the (unknown) gathering gene. [snip] Over time, the new allele became fixed (or very frequent) in women but men did not benefit.
    "

    Why do you attribute this single allele view to the authors? Is there something in their article that is inconsistent with the sex difference* in question being the result of the interaction between several genes and the environment?

    ---------
    * I have not read the article, but assume for the sake of the discussion that the effect is real.

    ReplyDelete
  5. scott: "if women had done most of the shopping previously [...] "

    Now I've read the article and the authors of the study address this issue. In Sec 3(b) they comment that "Experience at the market did not predict performance, either as zero-order effect or in the HLM model (table 2, between-subjects effects). The female advantage is significant even after controlling for experience at the market.". In the conclusion, they discuss the fact that females are often the primary shopper for household goods and why they think that this does not affect their conclusions.

    ReplyDelete
  6. I know from experience that my wife and I differ in this area. My wife is very good at reading maps and once she has been somewhere she can remember exactly how to get there, but she seems to have little sense of directionality.

    I can often sense what direction something should be in and just start turning down streets going that way (sometimes this doesn't really work out that great).

    ReplyDelete
  7. I would say the most probable explanation is that women generally have spent more time in supermarkets in our society, looking for stuff, and having to remember it. Proposing an evolutionary hypothesis and a what? gatherer gene seems to me rather an astonishing marvel of speculation, of hoping while hypothesizing.

    But I am no biologist.

    ReplyDelete
  8. "The female advantage is significant even after controlling for experience at the market."

    Oops, so they accounted for this, that sounds a little better. I suggest another hypothesis, stores do tend to organize the layout to suit the women, note the women's apparel near the entrance at Walmart.

    "It is necessary to say plainly that all this ignorance is simply covered by impudence. Statements are made so plainly and positively that men have hardly the moral courage to pause upon them and find that they are without support. The other day a scientific summary of the state of a prehistoric tribe began confidently with the words 'They wore no clothes!' Not one reader in a hundred probably stopped to ask himself how we should come to know whether clothes had once been worn by people of whom everything has perished except a few chips of bone and stone. It was doubtless hoped that we should find a stone hat as well as a stone hatchet." (G.K. Chesterton, The Everlasting Man)

    Again, I claim no biology speciality, but this strikes me as hopeful hypothesizing.

    ReplyDelete
  9. LOL, talk about your long bows.

    1. One simple survey (repeated on separate days) at a farmer's market does not equal natural hunter/gatherer behaviour.

    2. What happens when you use the exact same methodology in another environment? Supermarket? A large garden with native foods?

    3. What happens when you normalize for socioeconomic background?

    4. What happens when you exclusively use men who go grocery shopping? Is this a behaviourally learnt trait, not a genetic trait?

    5. What happens when you use real hunter gatherers?

    Basic questions, left unanswered by the paper in Proc RSB. Pathetic. Honestly you can publish anything about ancient humans if it's crazy enough. Just so stories are systematic of the processualist anthropologists in the States.

    ReplyDelete
  10. I seem to be at least a little more adaptationist than you are, Larry, but in this case I don't believe a word of it. Not based on just one small study. Why not? Simple: the skill involved (constructing and recalling a mental map of one's surroundings) is one that can be improved with repeated use, and which atrophies if used less. For skills like that, I'm quite sure that the effect of use or non-use rapidly overwhelms any small genetic influence on the skill's extent.

    If they had data from several thousand subjects from different walks of life and the effect still held, I might believe it. As-is, though, it looks like a candidate for the Journal of Irreproducible Results.

    ReplyDelete
  11. Erik 12345 asks,

    Why do you attribute this single allele view to the authors? Is there something in their article that is inconsistent with the sex difference* in question being the result of the interaction between several genes and the environment?

    No. It could be a polygenic trait. That would make it even more difficult to fix in the female population.

    ReplyDelete
  12. Socio-biological systems have an open-ended number of variables – it is impossible to isolate them; for example, in this particular case the cultural place of women has to be factored out as far as possible. Perhaps even the political bias, career stake and reputations of the researchers involved also needs to be taken into account. However, unless we are to throw our hands up in despair and declare the whole project to be hopeless, it seems that we have to accept that given the complexity of this subject differences in the knowledge, abilities, experience and social groupings of the researches will impinge upon the conclusions drawn and bumptiousness in one’s position is uncalled for. However, it is admitted that pluralists do have the advantage of not being committed to one kind of evolutionary explanation (that is, they have a hammer, a screwdriver, a file, etc.)

    After Gould it seems to me quite possible that at certain junctures in the great configuration space of possible living forms (and their associated environments) there may be routes through that space that are analogous to the level valleys skirting mountains; genetic drift is capable of taking the phenotype along these level paths, but movement along the these “equipotential” contours doesn’t constitute adaptive advantage. It’s a bit like languages drifting apart. Changes in a language may not confer an actual advance but rather signify a mere drift along a ‘contour’. However, one thing one can say about these contours is that at each point they must represent an object that can function and survive successfully.


    The vast ‘design’ configuration space of organic forms (and their environments) seems almost too vast for us to explore. Who knows what it may hide: there may be sudden ‘drops in level’ that allow for fast diffusion into new forms – hence periods of very rapid evolution. Alternatively, there may be regions where organisms become trapped for a while - hence Gould’s punctuated equilibria. They may even be isolated regions that are unconnected by incremental change – hence the ID contention. My own feeling is that we are dealing with a system that is so complicated that our attitudes need to be rather tentative, even when confronted with ID (I actually favor the Gouldian view)


    I’m no evolutionary specialist but the above is my understanding of Gould on these kinds of issues. I read this blog in order to try and hone my understanding of evolution.


    One question: why do so many “Dawkobots” (as somebody called them) read this blog? Moreover, why is Gould under Dawkins and nearer Ken Ham in the survey table? – Does that signal a subliminal understanding that Gould is the thin end of the creationist wedge?

    ReplyDelete
  13. modern women are better than men at remembering the location of food such as fruit and veg in a market.

    this is a joke, right? All the fruit and veg are in the produce section. Which, BTW, is a great place to meet modern women.

    ReplyDelete
  14. Moreover, why is Gould under Dawkins and nearer Ken Ham in the survey table?

    I think it's because fewer people voted for Gould than for Dawkins.

    ReplyDelete
  15. You just wait until a professor of English Literature on her way up in Women's Studies shows how this proven spatial trait is correlated with the Jungian animus and will lead to the Goddess taking over the foul dominator model of the patriarchy any day now.

    ReplyDelete
  16. Larry Moran answers: "No. It could be a polygenic trait. That would make it even more difficult to fix in the female population."

    I'm afraid I still don't see what the problem is supposed to be. How is the evolution of the (inferred) sex difference in spatial ability under discussion here different from the evolution of other polygenic traits/organs like body height or the vertebrate eye? If there is no need to ridicule eye evolution or evolution of body height by attributing the single allele assumption, then why is the need greater in this case?

    ReplyDelete
  17. wolfwalker wrote: "Simple: the skill involved (constructing and recalling a mental map of one's surroundings) is one that can be improved with repeated use, and which atrophies if used less."

    The authors of the study address this issue. They may be totally wrong or their argument may be unconvincing, of course, but until you've explained how their reasoning is flawed you have not significantly addressed their study.

    ReplyDelete
  18. Of course it's one study in one setting. Of course it owuld be nice to have corroboration from other populaitons and circumstances. So what? Science gets published one study at a time. In this case there is nothing a priori silly about the hypothesis, the experiment, or the conclusions from the data collected. So what's the problem? It doesn't fit your preconceived ideas? There is certainly nothing wrong with skepticism, but it ought to be based on something other than subjective incredulity.

    ReplyDelete
  19. I'll pitch in with those who are skeptic of a single epidemic study, in a standardized instead of randomized environment no less.

    On the larger question, I presume hypotheses and predictions looks fine within their strategy. It is the experiment and the initially weak support that looks questionable on the surface of it.

    I think it's because fewer people voted for Gould than for Dawkins.

    (¬_¬)

    I guess if it is really a strong concern you could introduce a random line scramble every hour. (In anticipation of many choices the first day of posting the questionnaire.)

    But then it would make reading the result graph awkward if it wasn't separated out. Anyway, another design is probably moot since it is a commercial site hosting - Blogger service the widgets and presumably they want to keep it that way.

    ReplyDelete
  20. subjective incredulity

    Yes, you could turn this around and start discussing analyzes in blogs and their value.

    I'm not sure that is entirely fair in this case. In a way every scientist who posts behind a paywall sets themselves up for snap judgments (or better delayed judgment) by posters and stray readers. We would have to start over with an open paper.

    ReplyDelete
  21. Sven DiMilo asks,

    I'm afraid I still don't see what the problem is supposed to be. How is the evolution of the (inferred) sex difference in spatial ability under discussion here different from the evolution of other polygenic traits/organs like body height or the vertebrate eye?

    Because in the case of height and eyes we have very good evidence that they have a genetic component.

    If there is no need to ridicule eye evolution or evolution of body height by attributing the single allele assumption, then why is the need greater in this case?

    Because the authors start with the unstated assumption that the ability to find fruit in a farmer's market has a genetic component. People have been questioning those kind of naive assumptions about behavior for over thirty years. So far, there's practically no evidence that individual behavioral traits like that are due to particular alleles.

    When your whole research program is based on a faulty assumption, that's silly. When you fail to even address your critics, that's bad science.

    ReplyDelete
  22. uh...for the record, it wasn't me who said that stuff.

    ReplyDelete
  23. Larry Moran: "Because in the case of height and eyes we have very good evidence that they have a genetic component."

    I agree that compared to height and eyes, the evidence for a genetic component in the sex difference in question is much weaker. The point is that instead of attributing to the authors the view that differences in spatial ability arise in much the same way as height and eyes (i.e. as the result of contributions from many genes and the environment), you first attributed the view that the spatial ability is controlled by a single allele. After I asked why, you indicated that it could also be polygenic but then it would be "even more difficult to fix in the female population". As far as I can see, that type of a priori objections could equally well be directed against the evolution of height or eyes: If height or eyes need not be single allele phenomena, then why assume that (differences in) spatial ability are? If it is not a problem that height or eyes are polygenic, then why would it be a problem that (differences in) spatial ability are polygenic?

    Larry Moran: "So far, there's practically no evidence that individual behavioral traits like that are due to particular alleles."

    So now you're back to the single allele view. Why do you attribute this view to the authors of the study? It seems much more likely (because it is common among this kind of researchers and because it simply makes much more sense than the single allele view) that the authors have in mind some sort of quantitative genetics scenario. The simplest such scenario would be that spatial ability gets additive contributions from a number of different loci and from the environment so that one can talk about the heritability of spatial ability. More complicated scenarios would relax the "additive" part and require more sophisticated quantities than heritability for quantative discussion. Qualitatively, the details do not seem to be matter that much. Then if there is a statistical dependence between reproductive success and spatial ability, the average spatial ability would change over time. Under some conditions it is probably also possible for sex differences in average spatial ability to evolve (after all males and females have different average height so its not that strange).

    It's fine if you think that such a quantitative genetics scenario is much more weakly supported for the spatial ability under discussion here than e.g. for height and other traits. But regardless of how well or poorly supported it is, it seems like a quite unfair to require that the authors support the single allele scenario that you have formulated, since you (I assume) don't require the same of, say, researchers studying evolution of height or eyes.

    ReplyDelete
  24. This comment has been removed by the author.

    ReplyDelete
  25. Larry Moran: "... the authors start with the unstated assumption that the ability to find fruit in a farmer's market has a genetic component. People have been questioning those kind of naive assumptions about behavior for over thirty years. So far, there's practically no evidence that individual behavioral traits like that are due to particular alleles."

    This is really rich. There is much evidence that individual behavioral traits like the ones tested here have a genetic basis, and it is bordering on denialism to claim otherwise. Analogous traits in animals clearly are. To give only one example, the particular set of tasks (hunting, herding, retrieving, etc) that one breed of dog excels at compared to other breeds.

    And the evidence is strong such traits can have a genetic component in humans. The studies of human twins on intelligence and personality traits are one such *strong* set of data. It might not be "denialism" to dismiss this evidence, but it sure looks like it in this case when it’s coupled with the charge that people who believe it are "naïve".

    Divalent

    ReplyDelete
  26. sven dimilo said...
    " There is certainly nothing wrong with skepticism, but it ought to be based on something other than subjective incredulity."

    Exactly, but.........

    This was published in Proceedings of the Royal Society B. It's a fairly prestigious journal. A fairly prestigious journal usually demands rigourous scientific methodology. There are so many questions regarding their methodology that you wonder whether the peer reviewers were familiar with the field.

    Such is the case with a number of high profile 'just so' human evolution studies. A simple Sunday afternoon experiment is conducted without considering all the factors and then a crazy sweeping generalization of a hypothesis is put forward which is not justified by the results. Now if this was any other field of science, the paper would sent back to the authors with a note saying 'Try a lesser journal'. Or at the very least 'needs more experimental data'. But because everyone loves a 'Just so' story about human evolution, these substandard pieces of work continue to be published. No journal is exempt from this, I have seen this time after time in Nature and Science (the two big science journals).

    The reasons for this are many
    1) The peer review is unfamiliar with the field of study (particularly when it involves anthropology and genetics)
    2) The authors know a guy on the editorial board
    3) The journal understands that human evolution stories = media press release = free advertising

    Many anthropologists believe reason 3 is the main driving force. But reason one is also equally valid. I had to help review a paper once for a genetics colleague with no anthropology training where the authors tried to claim on the basis of nothing that we had inherited red hair from Neanderthals. Pure conjecture, not backed up by their results. The geneticist has not picked this up, because the way it was written implied that every anthropologist already knew that Neanderthals had red hair to begin with. I pointed this out, we wrote up the review. But it got published without changes in a major journal. Que sera, some might say.

    But this has huge consequences. What it does it muddy the fields of anthropology and genetics. Anthropologists (and social scientists in general) struggle with scientific arguments, particularly when it comes to just so stories. The literature is so full of 'just so' papers that whole books have been written about the phenomena trying to understand why scholars gravitate to these positions. Using the neanderthal paper example again, I am in no doubt that we will see claims published linking red hair to neanderthals exclusively because that one paper I reviewed mentioned it. Is that science? No.

    So in summary. It is easy to write a 'just so' story and pass it off as reality (especially when the journals give you a free pass), but more often than not, your position is not defensible on the basis of your data.

    ReplyDelete
  27. Eric 12345 said

    "If it is not a problem that height or eyes are polygenic, then why would it be a problem that (differences in) spatial ability are polygenic?"

    But first you have to assume that complex behaviour is a) inherited, b) sex-linked and c) not significantly influenced by life experience, neither of which was demonstrated in this paper. Linking complex behaviour to genetics is extraordinarily difficult due to confounding effects, and as of today, few studies have actually convincingly demonstrated it in higher vertebrates.
    -----------
    "So now you're back to the single allele view."

    Actually Larry said

    "So far, there's practically no evidence that individual behavioral traits like that are due to particular alleles."

    To summarize Larry's point for you:
    Q: Does spatial ability have a sex-linked genetic component?
    A: It's possible, but this paper does not adequately answer or address that question.

    ReplyDelete
  28. Yeah, well, if you're just looking for cabbages and beans. On the other hand, I know guys that can locate a bar with an uncanny sense, find one after another, and still find their way home.

    I do my grocery shopping at typical grocery stores, which seem to be set up according to fairly standard plans. I assume the farmers' market was set out more chaotically, such that melons may be found here and there, but the best ones at Bocciagalupe's barrow. That would make for a better test than searching in a supermarket.

    ReplyDelete
  29. Just so stories are systematic of the processualist anthropologists in the States.

    I can't help but think there is an interesting story somewhere in this.

    Such is the case with a number of high profile 'just so' human evolution studies. A simple Sunday afternoon experiment is conducted without considering all the factors and then a crazy sweeping generalization of a hypothesis is put forward which is not justified by the results.

    When you put it that way... at least the general frustration stands out in relief. I sympathize, I started out in a subfield with a similar frustrating mass of ad hoc hypotheses and little organizing verified models. [Reactive sputtering for thin films and electronics, but I digress.]

    I still think leaving open questions unanswered isn't serious if the hypothesis is testable. Nor is it a problem to make sweeping generalizations or loose ad hocs (which of course often turns out to be wrong). But if there is little connection or foreseeable justification it is a problem for that hypothesis (which may or may not warrant publication), and if it is many such studies around it may be a problem for the field.

    It is all well and nice to play the ball, but someone must eventually score. And people play the current game, not the ideal strategy. So if these are the concerns...

    ReplyDelete
  30. the archaeologist, in the particular comment (of mine) that you replied to, the issues under discussion are these:

    1. The blog entry attribute the assumption that this sex difference arose via the replacement of a single allele with another single allele.

    2. A subsequent comment says that the sex difference could also be polygenic but then it would be very difficult to fix in the population.

    These two points are a bit strange because they are inconsistent with the standard quantitative genetics scenario one would employ for e.g. height. Height is neither the result of a single allele nor is it necessary that a single allele/genotype becomes fixed in the population in order for average height (or height differences) to evolve. All that is required is that allele/genotype frequencies change systematically over time. The fact that such quantitative genetics scenarios are typical in studies like the one we're discussing here makes it fairly odd to presume the scenario outlined in the blog entry.

    You summary unfortunately leaves out 1 & 2. Could you explain the reasoning behind 1 and 2?

    ReplyDelete
  31. G'day Eric 12345,

    Thank you for your follow up question. I think I understand where your confusion is coming from.

    You are 100% correct in your assertions about quantitative genetics, but you are missing the point.

    So with the single allele stuff that Larry wrote in the original blog, I read it as a joke and most other commentators on this post obviously didn't. I understood the post to be Larry making his own just so story about allele selection as a satirical basis for pointing out how flawed their reasoning was.

    It does not matter whether the trait of spatial awareness is a single allele, or a polygenic system or whether it actually it has any genetics basis at all. The point that Larry was trying to make was that this study made a flawed assumption that such a trait could be genetics based without considering all the options or for that matter designing a more robust methodology.

    Now a few of these commentators have gotten caught up with the fact that Larry mentioned allele (single) in the original post and then he says polygenic and alleles later on. GASP!! Call Fox News. But you're missing the point of the blog post, which was to point out that the authors of the paper had no basis to argue that the results that they saw have any relationship to genetics. They merely observed a phenomena, using a dodgy methodology and failed to follow up all possible conclusions. Whether it has a basis in genetics is irrelevant. Whether it is a single, poly allele or no basis in genetics is irrelevant.

    What you have to understand is that the contention of the authors of the original paper that spatial awareness is an inherited trait is unsupported by their own methodology. Their focussed attention on finding a answer based in natural selection means that other more likely theories have been dismissed.

    The take home message is that the paper is flawed science. Any conjecture on whether or not this phenomena is a single or poly allele based system is just that, conjecture and therefore pointless for discussion.

    ReplyDelete