Friday, May 04, 2012

Rushton on the skin color/behavior correlation

Phil Rushton and colleagues at the journal Personality and Individual Differences continue to push the the theory that genes explain the worldwide correlation between skin tone and important behaviors:
Pigmentation of the hair, skin, cuticle, feather and eye is one of the most salient and variable attributes of vertebrates. In many species, melanin-based coloration is found to be pleiotropically linked to behavior. We review animal studies that have found darker pigmented individuals average higher amounts of aggression and sexual activity than lighter pigmented individuals. We hypothesize that similar relationships between pigmentation, aggression, and sexuality occur in humans. We first review the literature on non-human animals and then review some of the correlates of melanin in people, including aggression and sexual activity. Both within human populations (e.g., siblings), and between populations (e.g., races, nations, states), studies find that darker pigmented people average higher levels of aggression and sexual activity (and also lower IQ). We conceptualize skin color as a multigenerational adaptation to differences in climate over the last 70,000 years as a result of “cold winters theory” and the “Out-of-Africa” model of human origins. We propose life history theory to explain the covariation found between human (and non-human) pigmentation and variables such as birth rate, infant mortality, longevity, rate of HIV/AIDS, and violent crime.
As Razib Khan has explained on this blog, pleiotropy as the explanation of these correlations simply does not work. So what plausible theories remain? Any ideas?

31 comments:

  1. "As Razib Khan has explained on this blog, pleitropy as the explanation of these correlations simply does not work."

    Where'd I miss that?

    ReplyDelete
  2. I think you'll find that characteristics are heritable, so they get inherited together. That is what causes the correlation; common ancestry, nothing more.

    ReplyDelete
  3. I don't find Razib's argument convincing when it comes to color and non-IQ behavioral traits.

    That said, Nettle (2010) has argued that there is a universal life history strategy, which is sensitive to environmental conditions (see: "Dying young and living fast: Variation in life history across English neighbourhoods.) Were this the case, a color-SES-IQ association would lead to a a color-life history association. We just need, then, to explain the color-SES-IQ association. Since we're IQologists, not sociologists,
    we really just need to explain the color-IQ association. SES, and with it, life history strategies follow. But this is the bugger. Now I've provided a mountain of evidence against moderate to large IQ X race differences in the UK and no one has produced any counter evidence. (Darkies in the UK -- Blacks and South Asians -- don't seem to have appreciably lower IQs.) So I'm now rather skeptical of a global hereditarian IQhypothesis. I would then suggest: (1) Global differences in IQ are largely conditioned by disease burden. Disease burden is correlated with climate and, as a result, human skin color. IQ conditions global SES differences. SES conditions Life History differences. (2) The US Black White difference is more complex as we are not dealing with globally representative samples.

    ReplyDelete
  4. Joe: I added a link.

    ReplyDelete
  5. Pigmentation of the hair, skin, cuticle, feather and eye is one of the most salient and variable attributes of vertebrates. In many species, melanin-based coloration is found to be pleiotropically linked to behavior. We review animal studies that have found darker pigmented individuals average higher amounts of aggression and sexual activity than lighter pigmented individuals. We hypothesize that similar relationships between pigmentation, aggression, and sexuality occur in humans. We first review the literature on non-human animals and then review some of the correlates of melanin in people, including aggression and sexual activity. Both within human populations (e.g., siblings), and between populations (e.g., races, nations, states), studies find that darker pigmented people average higher levels of aggression and sexual activity (and also lower IQ). We conceptualize skin color as a multigenerational adaptation to differences in climate over the last 70,000 years as a result of “cold winters theory” and the “Out-of-Africa” model of human origins. We propose life history theory to explain the covariation found between human (and non-human) pigmentation and variables such as birth rate, infant mortality, longevity, rate of HIV/AIDS, and violent crime.


    the section on siblings jumped out at me. i read the paper for that, but didn't find any results reported, nor did i see it in the citations. this would be a very powerful result, if true. we do know that there are many populations where pigmentation genes segregate within families, so if there *is* pleiotropy on non-IQ traits, then we'd probably see it there.

    jason malloy, i've emailed you the pdf for your perusal. mostly to check the citations, i don't have the time or energy to look myself. i did see if rushton had cited the enormous pile-on of genomics and pigmentation papers which date from between 2005 and 2010, and most of 'em are lacking. no surprise.

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

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

    ReplyDelete
  8. Chuck5:47 PM

    Razib,

    You offered two arguments:

    (a) About six genes condition pig differences in humans; human behavioral traits are conditioned by hundreds to thousands of genes; therefore pig genes can't have a large impact.

    (b) "if you positive that trait X is strongly correlated with pigmentation due to pleiotropy than the elucidation of the genetic architecture of the trait should b easy. if it isn't"

    As for (a), is the genetic architecture of pig. in the non-human animals which show pleiotropy awesomely more complex than it is is humans? As for (b),
    Rushton cites research mentioning the melanocortin system. Has no association turned up in humans between this and behaviors? I ask because Ducrest seemed to be pretty ambivalent, saying "Our hypothesis may thus not apply to human because there are many mutations..."

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

    ReplyDelete
  10. No, the claim that darker siblings have lower IQ, etc, is not supported anywhere in this paper.

    The Add Health data set has separate questions for race, ethnicity, and skin color (5 grades). This was several years ago, but IIRC, I did not find skin color-IQ differences for whites (of course there are major confounds like tanning, but pleiotropy is not my claim to support). I also didn't find aggression/impulsivity differences for darker hair and eye pigmentation in the NLSY.

    ReplyDelete
  11. As for (a), is the genetic architecture of pig. in the non-human animals which show pleiotropy awesomely more complex than it is is humans?

    no. as in humans it seems that most of the between individual variation in the aggregate is due to a handful of genes. lots of the human pig. genes arose in animal models (e.g., mice) and elucidation of coat patterns.

    As for (b),
    Rushton cites research mentioning the melanocortin system. Has no association turned up in humans between this and behaviors? I ask because Ducrest seemed to be pretty ambivalent, saying "Our hypothesis may thus not apply to human because there are many mutations..."


    i stopped paying attention to stuff on non-IQ behavioral studies because i just couldn't figure out how people were measuring these. though it is obvious how you can quantitative time preference.

    one thing to remember is that there is evidence of a connection between melanin production and androgen hormones. e.g., testosterone. peter frost has talked about this a lot, so i won't rehash it. most of the variation across populations is to the genes which i listed, SLC24A5, SLC45A2, KITLG, OCA2, etc. but there is a residual on the order of 10 percent which is explained by other genes and also explains sex difference within races.

    No, the claim that darker siblings have lower IQ, etc, is not supported anywhere in this paper.

    the claim, if true, would be an awesome one, as scientific findings go. if it's in the abstract why didn't a peer reviewer ask why it was in the abstract if it's not in the paper itself? (or, in the citations?) priors updated, again....

    ReplyDelete
  12. I think the Standard Model for Racial Realism goes something like this. Man emerged from Africa about 70,000 years ago. Some went north, others south and of course some stayed at home in Africa. Man has a naked skin and the sun causes damage but also creates Vitamin D. So those who went north would become lighter than those who lived near the equater. In the north the cold winters selected for inteligence. So today we find a correlation between skin color and IQ.

    The introduction of animal coloration data suggests another possiblity. Dogs don't need to be dark to live in the tropics as nearly naked people do. Yet if black dogs are indeed more aggressive this suggests the coorelation isn't though a historical association but through some existing biosynthetic pathway.

    That's a really far out thought. Melanin makes you mean?

    I can't really believe that it's true, but I certainly wish it were. It is a testable hypothesis. Lighten up a prison population and see if they become better behaved. I don't really think it works this way but it would be easy enough to start with some animal experiments.

    Statin drugs interfere with a biosynthetic pathway. Something similar that broke the pathway between melanin systhesis and testosterone (or something else) might allow us to create docile black rats and dogs. That is of course if there is such a pathway.

    Don't bother telling me I'm wrong. I don't actually believe in this. Just do the experiment. Get that Nobel Prize.

    Albertosaurus

    ReplyDelete
  13. Anonymous1:00 PM

    Pleiotropy is not sufficient as a strong cause between differentiated populations - you just have to look at Tamils and Pakistanis vs Englishmen to understand this.

    However, it might be sufficient as a weak cause between individuals in a population.

    There may well be a pleiotropy so darker (more melanistic) individuals are more aggressive, but there are a fuckton of other variables you can manipulate.

    Between population differences may wash out because the different populations simply readjust their levels of/responsiveness to testosterone, epinephrine and dopamine (not to mention brain structure) until they have a level of aggression and exploratory behaviour suitable for their environmentally required melanism and their other environmental pressures.

    But within populations, where other elements of the genetic matrix are not able to be "tuned" to the appropriate level for an individual's melanism, the correlation can still emerge.

    Of course, the implication here is that, if the above is true, the F2 generation of interbreeding between say, Bangladeshis and Danes (mixtures of biracial Bangladeshi-Danes) could be really diverse, with some really aggressive dark individuals and really timid light individuals who are outside the range for their parental populations.

    ReplyDelete
  14. Anonymous3:50 PM

    Chuck said "I've provided a mountain of evidence against moderate to large IQ race differences in the UK".
    Do you think you could provide, oh I don't know, maybe a link to a reference. How did you control for Selection Bias. While teaching in Africa, I met a number of very intelligent Africans who went to the UK precisely for the higher education, lived with a family of two Somalis both UK college grads for a week too. What methods did you use to control for this bias?

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

    ReplyDelete
  16. Anonymous3:52 PM

    Meant to sign that last post.
    JKL-PHD, MPH

    ReplyDelete
  17. Anonymous4:34 PM

    I too was disappointed at the quality of the Rushton, Templer "review". Any paper of this kind should discuss limitations, both the pro and the con evidence. Their understanding of epidemiology is limited where they mention ""ecological correlations" (widely used in epidemiology)." I have published such peer-review studies myself and spent several paragraphs discussing their limitations. Usually such studies, which are not that widely used, are generally acknowledge as used for hypothesis generation, rather than drawing conclusions. This is because the group measures used do not necessarily apply to the individuals of interest. Although, in the study they cite, this is not as likely to be a serious problem.
    JKL-PHD

    ReplyDelete
  18. Anonymous4:42 PM

    As a retired federal govt. epidemiologist in the USA, I think your readers would appreciate guidelines for interpreting epi evidence. Their seems to be a complete lack of the "Weight of Evidence" approach, here is an example of guidelines:
    The strength of epidemiologic evidence is enhanced by the presence of the following characteristics:

    1. Temporal relationship (plausible that exposure precedes disease).

    2. Consistency (independent studies or analyses demonstrate the same association).
    3. Strength of association (measured by estimate of relative risk).

    4. Dose-response or biologic gradient (measure of effect positively correlated with estimated dose).

    5. Specificity of the association (single clearly defined exposure associated with specific disease, note that absence of this characteristic is often uninformative).

    6. Biological plausibility (toxicology, pharma-cokinetics, genotoxicity, etc. support an effect).

    7. Collateral evidence (coherence with natural history and biology of the disease).

    None of these characteristics establishes causality. The absence of any one or even several of these aspects does not preclude a causal interpretation.

    When evaluating epidemiologic studies it is also necessary to consider extraneous factors such as chance and nonrandom error such as bias and confounding:

    1. Chance can be measured by examining the range of the 95% confidence interval.

    2. Bias is broadly defined as any trend in the collection, analysis, interpretation, or publication of data that can lead to conclusions that are systematically different from the truth. Examples include errors in measurement of exposure or disease, flaws in data collection or study design, recall bias when cases are more diligent at remembering past exposures, differential quality (accuracy) of information collected from compared groups, or differences in those who volunteer to participate in a study and those who do not.
    3. Confounding is typically a factor) that can cause or prevent the disease of interest that is also associated with the factor under investigation.

    ReplyDelete
  19. Anonymous5:29 PM

    Again apologize for not signing as JKL-PHD in the previous post on 7 types of evidence to consider and 3 types of extraneous factors. Note that pleitropy goes to biological plausibility. Its absence does not preclude a conclusion of causation.

    Often the strongest evidence for causation in epidemiology studies is dose-response and strength of association. A 10-fold increase in lung cancer was associated with smoking and there was also dose-response evidence (TH to UK's Bradford Hill who helped to develop these criteria).

    Rushton and Templer show an 11-fold increase for Blacks and crime over Whites and there is an intermediate association found for those with mixed-race parents, which is evidence of dose-response.

    I hope your main commenters will in future present both pro and con evidence, to help us better evaluate their points.
    JKL-PHD, MPH

    ReplyDelete
  20. As Razib Khan has explained on this blog, pleiotropy as the explanation of these correlations simply does not work.



    I don't understand the hang-up on pleitropy.

    What matters is the human behavior. Whether the behavior in question is the result of several different and completely unconnected genes or of pleitropy (a single gene having multiple effects) is an interesting but ultimately academic question.

    ReplyDelete
  21. JKL PHD,

    If Lynne-Rushton-Templer etc. started worrying about the ecological fallacy they would have to find a new line of work.

    It is nice to see an Epi comment on these blogs. The 11 fold increase and a "dose response" is a good argument. Too bad the well regarded CDC epidemiologist Eugenie Calle never noticed it.

    http://www.truecrimereport.com/2009/02/murder_in_midtown_atlanta_prom.php

    Ahh...irony

    ReplyDelete
  22. The introduction of animal coloration data suggests another possiblity. Dogs don't need to be dark to live in the tropics as nearly naked people do. Yet if black dogs are indeed more aggressive this suggests the coorelation isn't though a historical association but through some existing biosynthetic pathway.


    That's a really far out thought. Melanin makes you mean?



    "Black dogs" are dogs with black hair, not black skin.

    Historically speaking the "meanest" and most aggressive people on Earth have been Europeans, which would seem to throw a sizable wrench into the the melanin=aggression construct.

    ReplyDelete
  23. Anonymous6:39 PM

    Awеsomе blоg! Ιѕ your themе custom madе or did you download it
    fгom somewheге? A theme likе yours with a few ѕimple adjustеmеnts wοuld reallу maκe my blog
    јump out. Please let me κnοw whеre уou got yοur ԁesіgn.
    Bless you
    Here is my website : loans for bad credit

    ReplyDelete
  24. Anonymous4:05 AM

    Ӏ am аctually thankful to thе owner of this ѕіte who
    has ѕhагed this impressive post at at this tіme.
    Look at my web-site - loans for bad credit

    ReplyDelete
  25. Anonymous1:32 AM

    ӏ savour, lead to Ι discovered exactly
    ωhat I waѕ loοking foг. You've ended my 4 day lengthy hunt! God Bless you man. Have a nice day. Bye

    my blog post; diet plans

    ReplyDelete
  26. Anonymous7:37 AM

    Hi, i think that i saw you νisited my sitе thus i came tо
    “return the favor”.I am tгying to find things to enhance my web ѕite!
    I suppoѕe its ok to usе a fеw of youг ideas!
    !

    Alѕο ѵiѕit my web ѕite payday loans

    ReplyDelete
  27. Anonymous4:32 PM

    tramadol 50 mg tramadol codeine high - buy tramadol pay cod

    ReplyDelete
  28. Anonymous8:51 AM

    bookmarkeԁ!!, I lοve your sіte!



    Feel freе to surf to my web-site payday loans

    ReplyDelete
  29. Anonymous11:05 PM

    you're actually a excellent webmaster. The web site loading velocity is amazing. It kind of feels that you are doing any unique trick. Also, The contents are masterpiece. you have performed a magnificent activity on this topic!

    Check out my weblog - Same Day Payday Loans

    ReplyDelete
  30. Anonymous10:30 AM

    I loѵe looking through a pоst thаt can makе people thinκ.
    Αlѕo, thаnκs for allowing
    foг me to сοmment!

    Αlso νiѕіt my webpage - payday loans

    ReplyDelete

Are gun owners mentally ill?

  Some anti-gun people think owning a gun is a sign of some kind of mental abnormality. According to General Social Survey data, gun owners ...