October 25, 2012

A short, stylized dialogue on epigenetics

Me: If you analyze a host of real world outcomes using adoption studies, fraternal v. identical twin studies, twins-raised-apart studies, the history of early childhood intervention research, naturally-occurring experiments, differences between societies, changes over history, and so forth, you tend to come up with nature and nurture as being about equally important: maybe fifty-fifty. The glass is roughly half-full and half-empty.

The Latest Conventional Wisdom: You are so out of date! You see, the new science of epigenetics has proven that genes are even more powerful than you think. It's really three-fourths genes. But, the study of epigenetics also proves that one-third of the power of genes is under control of the environment!

Me: Okay, but that's what I just said: fifty-fifty. Two-thirds of three-quarters is one-half. I mean, whatever the precise mechanism under the hood, it's got to translate into what we see where the rubber hits the road, which is about fifty-fifty nature and nurture.

The Latest Conventional Wisdom: Whatever. Your fuzzy math can hardly shake my faith in the TED talk I saw about epigenetics or that article in the NYT Magazine. Where are your Powerpoints, huh? Where is your heartwarming example of a poor child winning a chess tournament due to epigenetics? Where is your galvanizing anecdote about how epigenetics caused Bob Dylan to write Like a Rolling Stone?

33 comments:

Anonymous said...

Americans are optimistic by nature. Nobody wants to hear a grim story about how half the population is condemned at birth by their inferior genetics. No matter how firm the scientific evidence is, people don't want to hear it. Not only is the truth politically incorrect, it is also depressing. So bring on the fairy tales.

Assistant Village Idiot said...

Onion Network News did a great sendup of TED talks, BTW.

Harlion said...

What about ncRNAs like microRNAs? They're heritable but not protein coding genes....

Anonymous said...

The addiction to maudlin heartwarming stories will be the downfall of European civilization.

TGGP said...

My latest post is about an academic who argues against gender essentialism. She actually agrees with Judith Harris on "nurture-via-nature", but uses that to argue that we can act to avoid magnifying the (relatively) small initial differences. It seems really tough to evaluate the causality on that one.

Anonymous said...

As soon as there is a real, medically effective solution to genetic disparities (which is nowhere yet on the horizon), the Left will shrilly demand we pay for the poor and "disadvantaged" to be upgraded.

Anon.

jody said...

the general idea in research psychology, at least in the 90s when i was studying the prevailing hypotheses, was that human behavior is definitely and measurably affected by both genes and environment, but that genes were weighed more heavily. the disagreement in the field about the degree to which it was more important. there were only a few serious researchers espousing a 50-50 hypothesis 15 or 20 years ago.

in the last 15 years though i'd say the consensus is shifting towards genes being more important than environment by a good amount. maybe 75-25 would be a rough guess. there's no way i can sign on to the idea that year 2012 accumulated research on the topic supports the idea that it's about 50-50. i certainly will read papers and research articles making the case, but i don't expect to be persuaded.

maybe if we're measuring behavior with a low performance ceiling the 50-50 hypothesis could be true, but when you start to get into competitive performance, it's almost like there are genetic limits imposed by genes on some behaviors. no amount of environment is going to turn some people into far above average performers in certain activities.

jody said...

aside from simply looking at humans at the highest levels of performance in various endeavors, what i found astonishing in the psychology research of the last 15 years is the repeated finding that even personality is highly heritable and many personality traits and quirks are directly passed on from parents to children, and that almost no amount of difference in environment will mitigate this. general personality traits and temperament regularly persist regardless of how the people grew up. there must be biological explanations for certain behavior patterns then.

anger management, for example, must be in the brain, and people with a problem controlling it, tend to make children who grow up to be adults who also can have trouble controlling it. this will happen whether they are a poor third worlder living in a hut, or the son of a bond trader living in a NYC apartment. those genes are in there, directing the proteins to develop in recurring patterns, instructing the brain to grow in a certain way, with the end result being a brain that flares with anger easily.

there is definitely a lot of room left for environment to have a big effect on people as they grow up, with respect to low performance, day to day behaviors which the typical psychological researcher would be interested in, such as how likely is the person to get an undergraduate degree, ever be arrested, get married and stay married, have children out of wedlock, and so forth. i would never deny that. that's easily established in basic research on life outcomes with large sample sizes. certainly it matters on average whether you grow up in a safe suburb versus a violent ghetto for instance. what happens in your life before age 21 or 22 or so, definitely has an effect on what kind of person you will be.

but even here, in the low performance world, there are many scenarios where the genes will put up roadblocks which are nearly insurmountable for the "environment can fix everything" crowd. no amount of environment is going to protect some people from falling into career and life derailing pathologies which they are genetically susceptible to. whether it's autism, alcoholism, dyslexia, diabetes, schizophrenia, there is no magic fix where you can take them out of one place and stick them in another place and now they're safe for good and you've permanently steered them around their trouble. those genes are in there, waiting to come out.

Anonymous said...

What is it with the Bob Dylan worship?

I just don't get it - never have, never will.

Anonymous said...

maybe if we're measuring behavior with a low performance ceiling the 50-50 hypothesis could be true, but when you start to get into competitive performance, it's almost like there are genetic limits imposed by genes on some behaviors. no amount of environment is going to turn some people into far above average performers in certain activities.

But from the point of view of "conditional" probabilities, if you take two subjects with roughly equal genetic potential, but one of them keeps his nose clean and busts his ass to get ahead in life, whereas the other one spends his entire life wasting away again in Margaritaville, then I can guaran-damn-tee you which one is going to be a success and which one isn't.

Whiskey said...

Dont' be so sure there is no DNA changes on the horizon. link here:

---
Laboratory experiments have already shown that the science can work. In 2005, Ronald Evans, a hormone expert working at the Salk Institute of Biological Studies in La Jolla, California, showed how genetic modification can increase the athletic power of mice.

Evans produced a group of genetically modified mice with an increased amount of slow-twitch muscle fibre. This type of fibre is associated with strong cardiovascular muscles and boosts an athlete’s endurance.

Evans’s mice could run for an hour longer than normal mice, were resistant to weight gain no matter what they were fed on, and remained at peak fitness even when they took no exercise. A form of genetic modification is already being tested in medicine, in the form of gene therapy for diseases such as cystic fibrosis.

Most gene modification techniques involve placing genetically modified DNA inside a virus and injecting it into the human body. The virus then enters human cells, and its modified DNA attaches itself to the human DNA inside those cells.

peterike said...

What is it with the Bob Dylan worship? I just don't get it - never have, never will.

Maybe because he's the greatest rock songwriter that ever lived?

Anonymous said...

Maybe because he's the greatest rock songwriter that ever lived?

Are you kidding? I don't think he even scratches the top 50.

Eric said...

But from the point of view of "conditional" probabilities, if you take two subjects with roughly equal genetic potential, but one of them keeps his nose clean and busts his ass to get ahead in life, whereas the other one spends his entire life wasting away again in Margaritaville, then I can guaran-damn-tee you which one is going to be a success and which one isn't.

So why does half the political spectrum spend its time convincing that guy in Margaritaville that the deck is so stacked against him he may as well keep firing up the blender? My biggest frustration in political life.

Severn said...

As soon as there is a real, medically effective solution to genetic disparities (which is nowhere yet on the horizon), the Left will shrilly demand we pay for the poor and "disadvantaged" to be upgraded.


Not just the left - that's the logic of libertarianism as well. In fact Randall Parker has written a number of posts which trend in that direction. If we can make everyone equally smart via genetic engineering, don't we have to do it?

a Newsreader said...

Maybe because he's the greatest rock songwriter that ever lived?

That reminds me of something about the tallest building in Kansas.

JayMan said...

Here's three relevant links:

Is Greed in the Genes?, Bryan Caplan

Heritability of behavioral traits | Gene Expression

Taming the “Tiger Mom” and Tackling the Parenting Myth « JayMan's Blog

socks said...

"As soon as there is a real, medically effective solution to genetic disparities (which is nowhere yet on the horizon), the Left will shrilly demand we pay for the poor and "disadvantaged" to be upgraded. "

I don't see the left clamoring for universal access to high quality stimulants now. Adderall improves most people's performance. It helps those with lower IQ's more than those with higher IQs, and it sometimes hurts creativity, but I suspect OTC adderall would improve productivity with low (individual) risk. Instead, people make it just available enough to sooth their conscious,mostly to really average people who won't be helped enough to compete with them.

As it becomes more obvious that growth has ground to a halt, these people (in particular the verbalist left) will simply find reasons, mostly subconsciously, to restrict access to anything that helps others compete. We see severe (ongoing?) Adderall shortages now, along with bizarre pushes to both make smoking tobacco more difficult, but smoking weed easier.

Anonymous said...

It's even worse than the dialogue would suggest.

There's very little evidence that epigenetic mechanisms have any substantial contribution to phenotypic variability in traits of social importance. The best evidence to date relates to metabolism (e.g. diabetes) and stress response.

The question of development plasticity is an interesting one, but again there is very limited evidence that the traits people care about are possible to modify in the directions we want.

Nanonymous said...

Two-thirds of three-quarters is one-half.

Stop flaunting your incredible math prowess, will you? :-)

It's not really clear what's your idea of epigenetics is and what it means to you. These things aren't actually straightforward. Maybe you could consider a serious post on it at some point.

Anonymous said...

Whiskey,

Genetic engineering is in its pre-infancy. The virus approach mentioned in your link is both easier in mice (smaller, fewer cells) and risky. Do you remember Jesse Gelsinger and the happenings at Penn a few years ago? Intelligence is likely to be a trait influenced by many genes of low individual effect spread across the genome. That is not going to be an easy fix even when more advanced genetic engineering technologies become available.

socks said...

"Intelligence is likely to be a trait influenced by many genes of low individual effect spread across the genome."

Regardless of how accurate (and/or trendy) this may be overall, there are several genes thought to have a significant impact by themselves, even if they contribute little to overall human IQ variance. To take a few examples from the ashkenazi intelligence paper, Torsion Dystonia, CAH, and Gaucher's all have 1/3-1 SD advantages by themselves, iirc. I'm betting you could make triple-9's out of most people with half a dozen genes.

Or maybe I'm wrong. Maybe you hit a bottleneck at some point and need digestive and metabolic support genes to get more than a couple of SDs.

Anonymous said...

Are you kidding? I don't think he even scratches the top 50.

Except that I'd add a few more zeroes - maybe top 5000?

I don't know that he's even in the Top 50 of Jewish R-n-R songwriters.

Whenever I see the hardcore Dylan idolatry, I just scratch my head and think to myself, "WTF?"

The Good Lord Almighty said...

Or maybe I'm wrong. Maybe you hit a bottleneck at some point and need digestive and metabolic support genes to get more than a couple of SDs.

Trust Me, when it comes to creating a creature as complicated as Homo sapiens, you would do well to assume that you will almost always be wrong.

The Good Lord Almighty said...

Or maybe I'm wrong. Maybe you hit a bottleneck at some point and need digestive and metabolic support genes to get more than a couple of SDs.

Trust Me, when it comes to creating a creature as complicated as Homo sapiens, you would do well to assume that you will almost always be wrong.

Rob S said...

Where are your Powerpoints, huh?

Exactly, Steve. This is why none of the other Republicans bothered learning Excel. Can you make a bumper sticker with Excel? Q.E.D.

eah said...

maybe fifty-fifty

Not buyin' it.

Pat Boyle said...

Epigenetics is now and always has been a mechanism for political liberals.

Although the term itself is about seventy years old it first became in the popular lexicon with Eric Erickson as a kind of neo-Freudian psycho-sexual stage (e.g. anal fixation).

Erickson was concerned with early childhood education. The ideas was that there was some critical period in which if you didn't get educated it would effect you forever. Normal environmentalism wouldn't do. If you missed fifth grade for example because you broke your leg, you could catch up. But by the sixties it was clear that the black kids weren't catching up. Erickson and others needed a mechanism that would take an environmental deficit and make it permanent.

A decade later when DNA methylation was discovered, they seemed to have found just such a mechanism. An early trauma turned off some genes.

A favorite example of epigenetic fans is the thinness of Audrey Hepburn. She got starved in Holland at the end of WWII and stayed thin the rest of her life. This seems like a model for black underachievement. No proper early upbringing and the black kids grow up mentally stunted.

It would be a model except for two things. Audrey Hepburn's kids are not abnormally thin. And Head Start failed.

Albertosaurus

The Legendary Linda said...

Will advances in genetics eliminate the black-white IQ gap in the U.S. or will both races get smarter while the 15 point genetic gap between them remains?

And of course IQ tests will have to be made super hard so that the average IQ stays 100

The Legendary Linda said...

A favorite example of epigenetic fans is the thinness of Audrey Hepburn. She got starved in Holland at the end of WWII and stayed thin the rest of her life. This seems like a model for black underachievement. No proper early upbringing and the black kids grow up mentally stunted.

But you're assuming liberals admit blacks are mentally stunted. I think liberals believes blacks are just as intelligent as whites, but IQ tests are culturally biased. One can't argue whether the intelligence gap is nature or nuture until all agree that the IQ gap is an intelligence gap.

rob said...

Harlion said...
What about ncRNAs like microRNAs? They're heritable but not protein coding genes....

The RNA themselves aren't inherited. DNA coding for the RNA and the DNA that regulate the RNA expression is inherited. Pretty much exactly like tRNA genes or ribosomal RNA genes, though they might have different promoters, enhancers, etc.

Harlion said...

"Anonymous rob said...

Harlion said...
What about ncRNAs like microRNAs? They're heritable but not protein coding genes....

The RNA themselves aren't inherited. DNA coding for the RNA and the DNA that regulate the RNA expression is inherited."

-Ergo, microRNAs are inherited. I'm well aware DNA is the heritable material through which miRNAs are carried from one generation to the next.

"Pretty much exactly like tRNA genes or ribosomal RNA genes, though they might have different promoters, enhancers, etc."

-My first point was that they add a layer of complexity to the regulation of genes, and not only that, but that they favor a genetic explanation for traits. My second point was that there are additional levels of complexity that add to these discussions.

Anonymous said...

I can't say I've ever heard that argument taken from epigenetics, but I'm not surprised. I don't think it matters much if things like heritability of intelligence are epigenetic or not. Epigenetic effects work more or less the same as plain old genetics; only slightly faster.

Sure is interesting stuff though.