Heritability and mental illness (long)

<<<What you say is just the opposite of what I've been led to think.>>>

Hi Monica,

I'm a little disconcerted here ... what did you think I was saying, and what is it that you've been led to think that is so different ? Genes are hugely important in every aspect of our lives, but things that run in families are not neccessarily genetically inherited to the same degree, and even strongly suspecting something of genetically inherited doesn't mean we know the specific mechanism by which it arises. Finally, strong genetic inheritance doesn't equate to having no important influences from other sources.

All that furor a few years back about Dean Hamer's "gay gene" turned out to be over about the influence of a particular gene on roughly 2% of the variance in phenotype taken over the whole population. An interesting result but hardly a complete explanation for differences in attraction and behavior between homosexual and heterosexuals. The genetic origin of the vulnerability for schizophrenia is not that much better understood as far as I know.

I know this becomes a charged political issue, for several reasons (a few of them good reasons), so I'll list somewhat formally where my perspective is.

1. Our structural proteins and enzymes are all coded by nucleic acids, and the nucleic acid code is inherited.

2. Nearly every aspect of who and what we are is influenced to some degree by structural proteins and enzymes, and therefore directly or indirectly related to the inherited code. There is no doubt that our genes are a very central part of what we are, both as a species and as individuals, including most of our shared characteristics and at the very least a non-trivial portion of our individual variation (likely more).

3. Humans vary more genetically within a group than between groups, because they are historically a single interbreeding population.

4. The large scale structure of the human brain varies to roughly the same degree as the human face, we each have a slightly different brain but most of it follows the same basic pattern.

5. Human beings vary significantly more in behavior than they vary in the details of their genetic code or brain structure, presumably because developmental trajectories, individual experience, and cultural artifacts tend to amplify differences between individuals even while causing behavior to converge for groups. We tend to differentiate ourselves somewhat more than our genome differentiates us.

4. Most of our knowledge of the heritability of individual differences comes from behavioral genetics (which compares identical twins and other relations raised together and apart).

5. Heritability, in behavior genetics, is a measure of what extent alternative genes explain the differences between people.

6. Heritability data from behavior genetics does NOT provide a reliable measure in any sense for the importance of genes in the development of any given character. That is an interesting technical question but we have little reliable information about it yet. It also is specific to a population; the heritability ratio varies significantly from one population to the next. Thirdly, characteristics can run in families and yet have very low biological heritability, since there are important forms of inheritance that are ecological or cultural rather than genetic.

7. Very, very, very little of our knowledge of heritability comes from identifying specific genetic causal mechanisms, because of the complexity of developmental processes and the multiple interactions between different genes needed to explain most phenotypic traits, as well as the role of various structural, historical, developmental, and environmental constraints in producing a human being.

8. It has long been noted credibly that many illnesses tend to run in families, particularly things like bipolar disorder, Attention Deficit, and schizophrenia.

9. Given that most of what we know of heritability comes not from specific knowledge of genetic etiology or brain structure or chemistry but from family histories and behavior genetics, it behooves us to understand the strengths and limits of the methodologies used in behavior genetics:

A. Behavior genetic results based on twins do not provide sufficient data to distinguish genetic from ecological inheritance, and as a result they are generally likely to be inflated as estimates of genetic heritability.

B. They assume, for example, that identical twins experience equivalent amounts of their shared environment as heterozygous twins. However, this is very likely a bad assumption. Identical twins share more of the womb environment and are treated more similarly by other people than fraternal twins, and tend to construct more similar environments for themselves, as well as having the same genetics. So their remarkable similarities in behavior are not just due to their genome.

C. Also, behavior geneticists do not take epistasis (gene-gene interactions) or gene-environment interactions into account in the interpretations of their numbers, because that complicates the models impractically.

D. Finally, the assumption that adopted children are randomly placed and so have randomly different environments simply isn't true. Children "raised apart" in BG studies actually have often spent significant time together before adoption and the choice of adopting parents is frequently made with careful selection criteria.

The conclusion I draw is that heritability estimates widely quoted today mostly come from behavior genetics, and those estimates are extemely sensitive to the simplifying assumptions contained in the model being used, and can be assumed to at least sometimes be inflated if assumed to be purely genetic inheritance.

So there are other reasons for things to run in families, but I think it is reasonable to claim that the lions share of our temperament is inherited genetically, as are vulnerabilities for the most highly heritable mental and physical illnesses, and probably less but still a significant amount of our personality and academic or formal intelligence (IQ). There is still good reason to suspect that personality, intelligence, and even the most highly heritable mental illnesses are all influenced to a significant degree by social interaction during development.

One of the most interesting things I've ever read on schizophrenia and which was very influential for my view on it was John Modrow's "How to Become a Schizophrenic." It is a fascinating and revealing first-person account of Modrow's experience as he began suffering the symptoms of the illness, and then the effects of people's response to his illness. He makes the compelling case that at the very least how he was treated was a factor in the course of the disease, both in its debilitating effects and eventually his remarkable recovery.

I don't buy entirely into the Szasz perspective of this being a non-illness, or Breggin's radical view that this is "nothing but" someone becoming overwhelmed by life. However I think it's reasonable to expect that treating something as an inevitably progressive and disabling condition with only symptomatic relief will tend to alienate and dehumanize people who are diagnosed and contribute to negative outcomes.

The question is not whether schizophrenia runs in families or even the degree to which it is genetically inherited, both of which are no doubt true to a great degree, but whether we can really justify considering someone suffering from it as a lost cause to be institutionalized and sedated or failing that, given up for lost to wander the streets as happened in my area when the local state hospital closed.

kind regards,

Todd

Replies:

There are no replies to this message.

| Behavior OnLine Home Page | Disclaimer |

Copyright © 1996-2004 Behavior OnLine, Inc. All rights reserved.