Thursday, March 27, 2014

Resiliency and Fragility

The HBD sites have had discussion recently about unimportance of environment in human development.  Some take the extreme form, that heritability accounts for so much of the variance, and factors that are clearly environmental factors so little, that we should for the time being not pay much attention to the latter.  While this sounds counterintuitive to us, they actually have some good numbers for it.

Disease seems to follow a different pattern: heritability is clearly important, but does not account for even half of the variance.  "Other factors," which are neither environment nor DNA, and are often a giant question mark, are the big ticket.

Cautionary note:  disease researchers say that genetics is only 5-10%, but they have usually narrowed their scope before saying that.  Cystic Fibrosis and Down Syndrome are clearly genetic ("Oh, that. But those are exceptions, and what we are talking about is...." Want a long long list of those exceptions, dude?"), while gunshot wounds and head injuries are pretty clearly environmental - unless one wants to take the (not unreasonable) view that these occur in definable genetic groups - males, sensation-seekers, risk-takers - more often. That is sorta kinda genetics, I suppose. What the researchers really mean is that genetics seems to provably account for only 5-10% of the variance in the diseases we are studying right now.

IQ scores and life outcomes, beloved by HBD-ers for their measurability* can convince one that environment matters little, at least over large populations. Today I ran into a great example of genius rising despite less-than-fortuitous circumstances, in the biography of Walter Pitts. (That link spends more time discussing the anthropologist Bateson.  You might like the Wikipedia spread better. Other sources here, here, and here, starting about p 139.). Mathematical prodicgy who ran away from home at 15 to hang around mathematicians and tell them where their arguments were weak. No education, no milk of human kindness as a teenager, impoverished with no prospects, but attracted the patronage of the super-brilliant by his own brilliance.  Resiliency.  You can't keep a good man down, eh?

Well yeah, you can, apparently.  He completely collapsed emotionally when Norbert Wiener cut off ties without explanation.  That is certainly hard to have one's mentor and patron drift away, but thousands of human beings bounce back from that every day.

I would say that it is easier to break things than make them.  There are lots of ways of becoming shorter, but none for becoming taller.  There are many ways to become less intelligent, but no ways to become more intelligent.

*which is why they also like East African and West African performance in track's distance and sprint events.

2 comments:

hugh lygon said...

so few comments for a fellow bgi-er?

as far as i can tell there is one huge subtlety (a contradiction in terms?) missed even by those one would expect to be more sophisticated.

geneticists i've heard get it, but hbd-ers don't. http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CCkQFjAA&url=http%3A%2F%2Fwww.bbc.co.uk%2Fprogrammes%2Fp005461w&ei=YJI3U-2pI-GyyAHv24GwDg&usg=AFQjCNFYjOcwr_hi6cZbciEsrjMBw7HuxA&sig2=kzEZpDjltXbNToIiimJ3_Q&bvm=bv.63808443,d.aWM

the statistic "heritability" is local. that is, all can agree that genes and environment together determine one's rank/status/position/etc. but the GxE plane is much larger than late 20th c minneapolis or any one place and time.

the model is iq or conscientiousness or whatever = sqrt(a)G + sqrt(1-a)E, where a = heritability.

but this is just the tangent plane or total derivative of the function which takes G and E to a certain trait.

the hereditist contention is that those genes which make for success or failure in one time and place make for the same in all times and places.

or that the plot of the GxE to trait function looks like a mountain range parallel to the E plane.

but a cactus rots in the forest and a tree sears in the desert. why should all people be trees or cacti. mayn't some be cacti and others trees?

an alien observing humans would notice:

1. human behavior is enormously varied. humans live in the sahara and in greenland. etc.

2. humans are at the genitic level homogeneous. http://www.ox.ac.uk/media/news_stories/2012/120302.html

so what the various heritabilities are really saying may be merely that some are more fit than others for the particular society they've been born into.

i have no doubt that there is a residual/hard core to iq, etc. but it's likely like health---it's so small that it can be overwhelmed by environment.

hugh lygon said...

an example of how heritability is local. http://pss.sagepub.com/content/14/6/623.abstract