Your Genes Determine Your Personality

I just sent off a saliva sample to 23andme to learn more about my genetic profile, and while I wait for the results, I've been continuing to look into the fascinating new discoveries about the influence of genes on personality traits. Over at Razib Khan's invaluable Gene Expression we find this table of heritability of particular personality traits (0.0 = no genetic influence; 1.0 = 100% determined by genes):

For comparison, your height is about 90% determined by your genetic makeup. By age 18, your genes determnine about 80% of your intelligence (it's hard to accurately test young children for intelligence, so the correlation increases as the test becomes more reliable and 'noise' decreases). Your genes also play a strong, if not determinative, role in your political orientation and personality structure. Of course, the headline of the post is misleading, I just gingered it up a bit to attract the attention of blank-slaters. As Khan aptly puts it: 'Biology may not be destiny, but it is definitely probability'.

One long-term trend we're seeing is the gradually increasing acceptance of findings such as these. There are still a few holdouts who, like global-warming skeptics, reject the findings out of hand, but their numbers are dwindling. I, for one, find it utterly fascinating. When I get my test results, I'll post anything interesting about them.

Comments

[Frederik]

But, as I pointed out in the original post and encounter practically every day, there is an fanatical rear-guard resistance by people who simply refuse to accept that genes have any influence on higher-order personality characteristics. This is particularly widespread in the humanities, where any mention of heritable characteristics is likely to immediately earn you a strident denunciation as a 'eugenicist' or worse. If you don't believe me, read the first few chapters of the 'Blank Slate'.

I can only speak for me, but one of the reasons that we in the humanities or the social science react quite defensive, when someone mentions "heritable characteristics" in a discussion is, that there are A LOT of racists/sexists who are adopting the language of pure and neutral science to justify a lot of bullshit. One moment you admit, that "Why yes, men tend to have more testosterone then women" and in the next moment you have to counter the idea that women did evolve to stay in the kitchen. I don´t even want to start with this fucking study about the lower IQ in Africa.

Another reason why I act defensive is, that many people who use this kind of arguments have only a very limited understanding of the scientific findings, accepting without doubt everything they like, while dismissing anything that might counter their made up world view.*

There is also the reason that a lot of this fair and neutral science is still very shaky and not scientific consensus. IQs are cultural biased and evolutionary psychology has a big problem with empiric evidence. Pinker for example is someone who´s writing I do enjoy, but whose 'evidence' is hotly debated. Take this review of "The blank slate" for example. (http://www.phil.cam.ac.uk/~swb24/reviews/Pinker.htm) Its quite long but a good read and it shows a lot of the problems of Pinkers empirical grounds. I can also recommend the book "Sex at dawn" (http://en.wikipedia.org/wiki/Sex_at_Dawn), which has its own empirical shortcomings and troubling conclusions, but which does a good job in showing the many weaknesses in the arguments of evolutionary psychology in general in Pinkers in particular.

So yes. I tend to overreact, when someone argues about the hereditary of characteristics, but, I think, with good reason. I´m also trying to be open to these theories, but would also like "the other side" to be open to the "not totally blank slate-side" of the argument.

*To be fair. This is something I very much do myself. But I like to tell myself that I did adopt my more or less "blank slate" worldview for good reasons. I think it´s easier to refuse any form of discrimination, if you don´t accept a genetic make up that determines your character. (Not that someone who does accept this will discriminate more, but I think there is one more reason for doing so.) That said if there is irrefutable evidence that im wrong I will accept it (hopefully). But for the moment, Im not seeing it.

Ps:
Im quite sorry about all the grammatical and other errors I made. Im out of time and english is not my first language.

[Andrew]

@ney: Of course, environment and genes interact in complex ways that are just beginning to be unraveled, so if one child of tall parents is systematically starved while the other is fed a normal amount, one of the children will not reach the full height of which he would likely otherwise be capable. But really, isn't this point so obvious as to be banal? The statistics in the chart clearly show that environmental factors are also at play, since genes don't come close to explaining the entire variance.

But, as I pointed out in the original post and encounter practically every day, there is an fanatical rear-guard resistance by people who simply refuse to accept that genes have any influence on higher-order personality characteristics. This is particularly widespread in the humanities, where any mention of heritable characteristics is likely to immediately earn you a strident denunciation as a 'eugenicist' or worse. If you don't believe me, read the first few chapters of the 'Blank Slate'.

[Damon]

@razibkhan: regarding point 2, a more accurate description would be (variance *attributable* to genetics)/(phenotypic variance). The attribution is key, and it is the reason I referred to the Devlin paper. The assumption of most twins studies is that similarities exhibited by identical twins separated at birth are due to genetics. However, the authors find, based on comparisons to fraternal twins and siblings born at different times, that the environment of the womb may play a significant role in these similarities and that previous studies consequently overestimate the genetic contribution to intelligence.

The point is not that genes have no influence on intelligence (they probably do) but that small differences in what researchers consider to be important can have a substantial impact on how much influence those genes are thought to have. As pointed out by Ney, it has become apparent in recent years that the interaction between genes and environment is extremely complex and therefore deserves a much more complex discussion than it is currently given in the public sphere. I usually get that level of discussion from this blog, and I therefore felt a bit disappointed by the level of discussion regarding behavioral genetics.

Regarding point 3: you correctly point out that the Devlin paper is a bit old, but as far as I understand, much of the data used in the field (e.g. the identical twins studies typically used to calculate heritabilities) is also quite old. :)

[Ney]

"For comparison, your height is about 90% determined by your genetic makeup." Which is nonsense. Without my genetic makup, I would not exist. Without an environment which supplied me with food, I would never have grown. So what do they want to say? Maybe that for people who grew up in roughly the same environmental conditions, 90% of the variability in height is accounted for by genes? Or something like that? Meanwhile, some of those scientists who really understand genetics have started to demonstrate that a little bit more complex reasoning is needed than is found in the mostly crude public discussions about genes and environment. See for example
http://mindhacks.com/2012/06/12/gene-environment-interaction-of-your-neighbourhood/

[twitter.com/razibkhan]

1) just want to be clear that heritability is more accurately a population level statistic, though it has individual level ramifications

2) concisest way to define heritability is (genetic variance)/(phenotypic variance). i.e., proportion of variation due to variation in genes

3) your nature reference is a little old. people have spent decades arguing about heitability estimates using relative designs. therefore, i think these genomic methods with unrelated individuals is the way to go in the future:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3182557/?tool=pubmed

[Michael]

It tests for 241 diseases and conditions. I am a hypochondriac and I can't resist this test.

[Damon]

@Andrew: Likewise, you don't need to be a genetic determinist to believe that in-born differences between humans exist. There are other potential explanations (epigenetics, stochastic variations in in utero development, etc.) for such differences that don't require genetics.

Your point that IQs are predictive of success (at least of a certain kind of success) is true. However, it doesn't have any bearing on whether intelligence is either genetically determined or culturally defined. Smart people may be smart their entire lives and may be more likely to be successful than their stupid peers (who have been stupid their entire lives), but that doesn't mean that they acquired this smartness from the genes inherited from their parents. As for cultural definitions of success, people with lower IQs tend to have more children and, at least by one measure--and a fairly important measure in biology--are more evolutionarily successful than their counterparts with higher IQs. Just a thought.

Although I'm sure that you probably understand the pitfalls and caveats of heritability, I wanted simply to note these caveats for your readers. It is easy for most people, even very intelligent people, to confuse heritability with inheritability. Heritability includes both genetic AND environmental (read: everything else) influences on a trait of interest. Moreover, heritability isn't a true measure of whether a particular trait is genetically inherited because it doesn't take into account whether the traits of the parents resemble the traits of the offspring.

I'm perfectly willing to agree that the blank slate hypothesis is, strictly defined, almost certainly wrong. I would, however, assert that the principle is useful for a free and fair society, but that is a question for another post, I suppose...

[Andrew]

@Damon: All of your caveats are well-taken, but I don't think they do much to weaken the point of the paper (which, of course, is only one of hundreds showing comparable results). Further, I don't agree that IQ (or better, general intelligence) is a culturally-determined phenomenon. It's just been too widely-studied for that objection to hold much water. In particular, the overwhelming predictive power of 'g' is hard to argue with: people with higher intelligence, ceteris paribus, achieve much better life outcomes than people with low intelligence, everywhere across the world, period.

The main reason I highlight these studies, aside from the fact that they're fascinating, is that there are still vast armies of blank-slaters out there who literally reject any genetic influence on behavior, personality, socialization, etc. They may not move in your more scientific circles, but they certainly do in mine.

It should also go without saying that none of these studies compel, or even suggest, normative conclusions. You don't have to be a racist reactionary to be interested in genetic variation -- in fact, the more sensible people begin acknowledging these realities, the better.

[Damon]

I thought your previous post on genetics and intelligence was supposed to be tongue-in-cheek so I didn't say anything at the time. However, now I'm not so sure that you are just joking.

Thus, although I realize that you haven't come out and said what you think about the genetics of personality traits, I feel that it is my duty as a geneticist to point out that the term "heritability" is somewhat misleading--perhaps due to the fact that it sounds similar to "heredity". First, heritability is actually influenced by both genes and environment (see the Wikipedia entry for the definition), and although the scientists who study the heritability of certain traits try to control for the effect of environment, it is virtually impossible to eliminate its contribution from their calculation. For example, one publication (Nature 388:468) found that the apparent contribution of genetics to the heritability of intelligence (which can be more easily quantitated than, say, extroversion) was much less when they controlled for the shared environment of the womb. Much depends on whether scientists control for these factors and what semi-arbitrary numbers they use to control for them.

Second, heritability is not a correlation between a trait (e.g. intelligence) and genetic relatedness. Rather, it is a correlation between the variance of a trait and genetic relatedness.

Finally, I've always felt that it should be noted that personality traits are often culturally defined (e.g. IQ) and/or difficult to quantitate (e.g. aggression or extroversion), which would make the measurement of such traits rather more spurious.

That's not to say that genes don't influence personality (they probably do). However, reported heritabilities ought to be taken with a rather large grain of salt.