Blonde children exhibit more fear response? A new paper reports:
...Hair pigmentation was found to be significantly associated with behavioral inhibition in the sense that blond children exhibited higher fear scores. As in American samples, blue-eyed children had a higher fear score than did other children, but this difference was not statistically significant.
Jerome Kagan has reported these sort of findings before. Coloration is a funny trait, for example, there is now evidence that Europeans are highly constrained on one locus which affects complexion, while being high polymorphic on another (MC1R, on which East Asians seem to show strong recent positive selection). The finding that blue eyed or blonde youth tend to be more withdrawn or shy probably isn't that surprising, and it makes one wonder how such a bias might have shaped the higher order "characters" of cultures in the north of Europe. If I had to guess I would point the finger at pleiotropy in this case. Here is the older research by Kagan:
Two independent investigations of the association between the temperamental dimensions of inhibition and lack of inhibition to the unfamiliar, on the one hand, and the degree of pigmentation of the iris, on the other, revealed a statistically significant relation in Caucasian children between behavioral inhibition to the unfamiliar and blue irises and uninhibited behavior and brown irises....
Please note that MC1R is uncorrelated with eye color, though it does have a relationship to skin tone. Update: A few clarifications and caveats. In terms of this not being surprising, I'm making the rather weak observation that the blondest nations in Europe stereotypically are characterized by societal introversion. This is not to say that I am arguing for clear and deep national characters, the ancient Romans were perceived by contemporaries (at least during the republican era) to be stoic in persona, while modern Italians are rather not typified by this stereotype. German efficiency would have been surprising news to ancient peoples. Nevertheless, I suspect there are some basal predispositions at work that might vary between populations.This paper on variation on the DRD4 receptor across populations is in the direction I suspect more data is going to be unearthed. Second, in terms of "higher order" characters, this is what I am getting at. Imagine two populations, for simplicity assume they are arbitrarily large. Assume an extroversion metric, 0 to 10, and assume that there is an underlying genetic component to this. All things being equal, assume that population A consists of 60% of individuals with extroversion level 8, and 40% extroversion level 2. Assumine that in population B the numbers are reversed, so 60% have extroversion level 2 and 40% have extroversion level 8. So, expecation of A = 8*.6 + 2*.4 = 5.6 Expectation of B = 8*.4 + 2*.6 = 4.4 But, humans do not develop independently in a vacuum, they interact with other individuals. If you assume random interactions, then in population A the individuals who are less extroverted will more likely be interacting with more extroverted individuals and those in population B who are more extroverted will interact more often with less extroverted individuals. In their book Not By Genes Alone, the anthropologists Peter Richerson and Robert Boyd review extensive psychological literature which suggests that humans seem to have a "do what most people do" cognitive bias, so the bias of who you interact with might have powerful developmental effects. Granted, the model above is highly simplified (one assumes that interactions are not random within a population), nevertheless, the basic gist is that in populations A & B would be deviated away from the expected mean toward the mode because of social factors. Ultimately the same logic applies to purely culturally heritable traits as well, perhaps explaining the "flips" in state from the ancient Romans to modern Italians (I find arguments that exogenous admixture caused these changes unpersuasive, though I am open to the possibility of endogenous change in gene frequencies). Update II: Here is the table you want:
Here is part of the discussion:
..a common precursor for the melanocyte-stimulating hormone and beta-endorphins, with the latter mediating a higher threshold of physiological arousal. Propiomelanocortin is co-produced with corticotropin-releasing hormone (CRH) in human hair skin (Kono et al., 2001) which might be of potential relevance given reports that inhibited children tend to show higher cortisol levels (Kagan et al., 1988, 1987; Fox et al., 2005). The finding of a elevated baseline cortisol in inhibited children led to the assumption higher cortisol levels may induce changes in the amygdala, thereby exacerbating or even precipitating inhibited behavior (Schmidt et al., 1997). This potential biological mechanism might contribute to an explanatory model for the association of hair pigmentation and fearfulness, as amelanocyctic hair builds seem to produce and significantly higher levels of CRH (Ito et al., 2004) and this production has been described as substantial enough to manifest itself in different cortisol serum levels (Ito et al., 2005)....
References: Moehler E, Kagan J, Brunner R, Wiebel A, Kaufmann C, Resch F, Association of behavioral inhibition with hair pigmentation in a European sample, Biol Psychol. 2006 Jan 13. Rosenberg A, Kagan J, Iris pigmentation and behavioral inhibition, 1987 Jul;20(4):377-92.
