I’m interested in blue eyes. Specifically, I wonder why they’re around at all. Unlike blonde hair, there’s only one region of the world where blue eyes are extant at high frequencies, and there is a pretty regular drop off as a function of distance. It seems that variants of OCA2 are associated with blue eyes in Europeans. If you check Haplotter it looks like the region around this gene has been subject to a powerful bout of recent selection (i.e., within the last 10,000 years). Why this selection? Well, there’s no definitive explanation yet. But I don’t want to focus on hypotheses for why OCA2 has been subject to selection as much as what correlates there seem to be with the phenotype of blue eyes. Specifically, behavioral correlates.

There’s a fair amount of recent work in this area, but reading Racial Adaptations I stumbled upon some older studies, and, a model to explain the outcomes which I’m not really qualified to judge. So I’m going to report and let you decide, and hopefully, inform (especially those of you with cognitive neuroscience backgrounds).

So here I go. From page 66:

…eyes of different colors are related to differents in perception and innate behavior, as psychologists have discovered.23 Some of their tests have shown that dark-eyed persons are more aware of color and lighter-eyed ones of form; the former prefer bright hues and the latter lighter ones. Light eyes tend to envision panoramas, dark ones to concentrate on details

I don’t know anything about fashion or design, but I wonder if those of you who know this area (assman?) might be able to map some macrosocial trends back to these individual differences? After all, the vast majority of Scandinavians are blue eyed, and they’ve produced a fair amount of modern design. While Italians are well represented in the world of fashion and the arts. And the frequencies of eye colors are probably inverted in these two populations.

In any case, moving on, later on the page:

Behavioral variations are focused on the differences between “self-paced” and “reactive” responses to sudden stimuli. In the first subjects follow a well-known plan of animal behavior of pausing and deliberating before decision. In the second the subject flies into instant action (in animals), to attack or to flee. Of course, these behaviors are elaborated in man to govern many more-complex and subtler actions in speech and deed.

Light-eyed subjects are more likely to be self-paced, dark ones reactive. These differences are statistically significant and are patently genetic because they are equally represented in all age groups from kindergarten through professional life. When the iris color categories are extended from two to three, the subjects in the middle, with the mixed, green-to-hazel eyes, are found to share the benefits of the two extremes.

In one experiment ten of each of blue-eyed male, brown-eyed male, blue-eyed female, and brown-eyed female college students were wired to polygraphs and shown arousing pictures of sex and violence with appropriate sound effects. The brown-eyed subjects and the females responded more emotionally than the blue-eyed and male ones did.24

In another test, the same investigator gave Rorschach tests to forty blue-eyed and forty brown-eyed males. The blue-eyed ones fared better with form than with color and vice versa.25 In both tests only pure blue-eyed and pure brown-eyed persons were used.

That’s a lot to throw at you, but pretty much line with more recent work. The author does note that these studies were performed upon subjects of European ancestry. Whatever differences one can see between groups of blue and brown eyed Europeans, obviously it wouldn’t predict to other genetic backgrounds. East Asians tend to exhibit some of the same behavioral tendencies vis-a-vis Europeans that blue eyed Europeans exhibit vis-a-vis brown eyed ones. Obviously brown eyes can’t explain this since East Asians have brown eyes. This isn’t that strange, lots of the recent research in regards to human evolution suggests that East Asians and Europeans can converge upon the same phenotype via alternative genetic pathways. Blue eyes may simply be a byproduct of selection for another phenotype.

But on to the author’s model, which requires some knowledge of cognitive neuroscience and brain chemistry to evaluate. From page 74:

Few people other than ophthalmologists seem to have looked at retinas, nor to have considered it remarkable that the fundus is of virtually the same color as the person’s skin and for obvious reason that the underside of the retina is epidermis

At this point the author draws upon some photographs of the retinas of various racial groups, and observes the variation in color. He takes lithographs of these photos and basically measures the amount of light which can penetrate them. Here is the exposition that is relevant:

The Negro and the mulatto get 1.75 fcp; the Hindu and the American Indian 1.16 fcp; the brunet European 0.66 fcp; and the Chinese, the blond and the albino get 0.22 fcp. The Negro’s and the mulatto’s retinas let through eight times as much light as did those of the Chinese, the blond, and the albino.

OK, here’s the part where he lays out his argument for color and behavior:

Once inside the cranial cavity, neural impulses produced by the visible light that has passed through the retinal screen follow one of two paths. One lot goes to the hypothalamus…This part of the brain is the primary control tower of the central nervous system for almost all of the self-starting and self-regulating activity of the body-the sleep cycle, body temperature, the digestive process, fighting and loving.

These activities are managed by the production inside the hypothalamus of regulatory hormones. In some cases directly, but more commonly indirectly, these hormones control the fabrication and release of other, more specialized hormones in the pituitary, or “master gland,” seated in its bony saddle at the base of the braincase.

Two of these pituitary hormones become the raw material for making MSH (melanosome-stimulating hormone), so named because it darkens the pigment in amphibia and other cold-blooded animals. In man as in other mammals, it has yielded its pigment-darkening role to the built-in enzymes of the melanosomes themselves. Its only retention of its earlier function is to darken melanosomes.

Meanwhile, the rest of the neural impulses flow through a complicated channel into the brain’s third ventricle. From there they continue through several different parts of the brain stem into the pineal gland.

…Among other hormones, the pineal makes melatonin, which fl
ows onto the cortext…Melatonin is an inhibitor; MSH is a stimulator, and one of its results is the secretion of a substance that switches melatonin making in the pineal on and off in countermeasure with is own rate of flow. Thus, the more light the retina lets in, the more MSH will be secreted and the less melatonin.

MHS has two divisions…one affects the peripheral nervous system only, the other may reach the brain, while melatonin bathes only the latter…all else equal, the stronger the light that penetrates the retina the more automatic are the responses to it, and the weaker the light the more the same responses fall under the control of the learning and thinking part of the brain….

To show those who may doubt that MSH and melatonin really affect the behavior of mammals…the National institute of Child Health and Human Development of Bethesda, Maryland, removed the pineal glands from some black rats (hair color, not skin color), pretended to do so with others, and left a third lot unscathed. The victims of real surgery became hyperactive and nervous, but when she injected melatonin into them they calmed down; both the other lots behaved normally throughout.29 Other researchers have given their animals shots of melatonin without operations. These injections reduced their avoidance responses,30 making them pause in the face of danger, rather than fight or flee. These are the “self-paced” and “reactive” responses found among blue-eyed and black-eyed students.

But not all our melatonin is made in the pineal gland. In daylight the choroid, which encases the retina, makes more of this hormone than the pineal does, and in darkness the pineal secretes more than the eye does. This discovery was made by experiments on chicks and rats.31 If one seeks to apply it to man, it might be well to remember that the stroma, or outer part of the iris lying above the lens, is mesoderm and an extension of the vascular choroid, also mesoderm, in contrast to the retina, which is ectoderm, like the pigmented layer of the skin. An investigation along this line might possibly explain the differences found in reactions blue blue- and brown-eyed subjects by psychologists mentioend earlier, because blood from the choroid flows direclty into the main bloodstream, some of which feeds the brain.

That was a lot to throw out there, but the main reason I’m posting this is so that cognitive neuroscience people can throw cold water on this model or not. Obviously a lot has happened in neuroscience since the 1970s, when the author thought this idea up. Myself, I tend to get confused on the various biochemicals which modulate brain chemistry and all the different pathways and modulations, so clarification would be nice too. Two points I’d also like to point out. 1) the Chinese clustered with the blond European in terms of the retina light values. 2) If the amount of light which manages to get through the retina is a major issue, that can explain variation by latitude and climate in terms of temper and personality, since obviously the amount of sunlight and radiation which reaches the surface varies a great deal.

In the interests of putting informative physical anthropological data on the web, here is something I scanned out of Racial Adaptations. The map below the fold combines both hair & eye color data.

Just talked to my mother today and she phoned relatives in Dhaka. Everything is OK, electricity is coming back online. She seemed dismissive of the idea when I asked if it was all well in the rural areas where some of my distant relatives live; it wasn’t that big of a deal. On the other hand a cousin married a man whose family is from Barisal. Her husband went to check out the situation (family has estates and properties around the Sundbarbans) which was in the path of the storm, and it wasn’t as pretty there as everything had been blasted away and tossed around. The main aim of my call was that I wanted to know if I should send money to dispossessed relatives, and my mother stated there was really no one to send money to. So to the Red Cross it goes.

Over at Edge Armand Leroi was interviewed about human genetics & variation. Most of the talking points are based on his book Mutants.

Nothing too surprising in this story about attitudes toward alcohol in Europe. From Finland:

Since the government cut tax on alcohol by one third in March 2004, deaths and diseases from alcohol have all jumped by similar amounts in hard-drinking Finland.

The cut was made due to cheap alcohol imports from neighbouring Estonia.

People respond to tax incentives!

In PLoS Genetics, there’s a report on the identification of a locus that leads to white coat color in horses. This locus is KIT, a proto-oncogene (ie. certain mutations in this gene lead to increased cell growth and sometimes cancer) important for the survival of melanoblasts early in development. It’s a nice story on its own, and the authors have an interesting historical genetics perspective:

Two thousand years ago, the Romans already knew of the phenotypic differences of depigmented horses, which they described as candidus (white) or glaucus (grey). The Roman historian Tacitus described the use of sacred white horses for auguries by German tribes. The so-called white horse of the Saxons is depicted on the flags of the German states of Lower Saxony and North Rhine-Westphalia. It is thus of considerable historic interest to trace the origins of white horses, particularly because the nature of their white color can have different causes, some of which are KIT mutations such as those described here. We do not know whether the Roman terms candidus and glaucus actually correspond to our modern coat color designations of white and grey. Archaeogenetics on historic DNA samples may help to identify the genetic constitution of these horses.

But something about that gene name bothered me: where had I heard it before? Ah yes, there was a recent report of an association with variants near KITLG (KIT ligand; ie. a binding partner of KIT) and blonde hair in humans. The genetic architecture underlying pigmentation isn’t all that different across mammals, of course, so this isn’t surprising. But still, it’s nice to see these connections– sometimes, biology makes sense.

My Genome, Myself: Seeking Clues in DNA:

For as little as $1,000 and a saliva sample, customers will be able to learn what is known so far about how the billions of bits in their biological code shape who they are. Three companies have already announced plans to market such services, one yesterday.
Offered the chance to be among the early testers, I agreed, but not without reservations. What if I learned I was likely to die young? Or that I might have passed on a rogue gene to my daughter? And more pragmatically, what if an insurance company or an employer used such information against me in the future?
But three weeks later, I was already somewhat addicted to the daily communion with my genes. (Recurring note to self: was this addiction genetic?)

Update: Hsien-Hsien Lei has much more.

Update: OK, not to be morbid, but the confirmed death tolls is now greater than 3,000, and there is a tentative projected death toll on the order of 10,000. That’s in line with my comments below. End Update
The death toll is pushing 2,000. I think I might be safe in saying that the fatalities are going to be one order of magnitude lower than the 1991 and 1970 typhoons. Chris Mooney considers whether Sidr was as bad as hurricane Mitch in 1998. Mitch caused 11,000 fatalities. Sidr might cause as many. But the combined population for Nicaragua and Honduras is about 13 million today (one assumes it was a bit smaller back then). The population of Bangladesh today is in the range of 150 million. In 1970 East Pakistan (what became Bangladesh) had less than half the population of Bangladesh today, so that puts the fatality contrast in some perspective as well. The numbers here lead me to suggest that the impact of Mitch in relative terms was more comparable to the typhoons of ’91 or ’70 which hit Bangladesh rather than Sidr. Some of this might be chalked up to better preparedness, though much of it surely has to be that Sidr made landfall in a relatively underpopulated region, the Sundarbans. It might have been a different story if the bull’s-eye had been on the mouth of the Meghna, one of the most densely populated regions of country (which combines the flow of the Ganges and Brahmaputra as well as several other local rivers).
Here’s a donation page for the International Red Cross/Crescent.

A few weeks ago I reported some research that seemed to show a relationship between a gain in IQ due to breastfeeding and a particular genetic variant. Looks like I spoke too soon. p-ter has the goods:

The fact that they have a measure of maternal IQ but don’t directly include it in the published multiple regression suggests that they tried it, but didn’t like the results. They didn’t include parental phenotype in any of their previous studies, but there, at least, there was some functional evidence linking the polymorphism and the phenotype. Here, there’s nothing. Considering the fact mentioned in a previous post that other researchers find absolutely no evidence for link between IQ and breastfeeding (the entire basis for this study), this has to be classified as highly questionable. And regardless of the veracity of any gene-environment interactions here, the huge effects of breastfeeding on IQ shown by the authors are clearly artefacts of the heritability of IQ, and it’s unfortunate that they are being propagated.

This is the classic problem that Judith Rich Harris points on in her work, many researchers just don’t control for the correlation between parents and offspring genetically.

Over at Overcoming Bias Eliezer sings the praises of simple math. Steve was the first person to bring to my attention this phenomenon. For me, a little algebra, probability and statistics goes a long way in making one’s thoughts more precise and rigorous (even one’s own internal monologue). You can attempt this through pure verbal description, but unfortunately the precision may not be perceived the same from different vantage points. In other words, verbal arguments often tend to communicate different things to different people (or different things to the same person at different times) because of subtle and implicit semantic interpretations (the mental algebra obviously works out differently if the coefficients in front the variables vary from person to person). If you speak in cautious and specific philosophical language it may only clarify the misunderstandings further.* Another angle is that of simple data analysis. Half Sigma and Inductivist do this well. The data is out there, insight will come.

* This is confusing. Clarify misunderstandings? If two people are discussing a topic and their understanding of the terms is such that they have two different conceptions of what they are discussing, further specificity of the language won’t necessarily bring them together, rather they simply clarify in their own minds their interpretation of the model.