Heritability of antisociality

[Wyatt Earp]

TonTon, this is a meta-analysis of the results on heritability of antosociality. I hate the fact that you can only read the abstract. When I'm at work I'll see if we've got a subscription to this journal, and if so, summarise.

Interesting that the effect is stronger for extreme manifestations, and for the studies with tighter methodologies.

Heritability results need interpreting with extreme care, it needs to be said. They measure the contribution of genetic variation to the variation of the trait, meaning that their meaning isn't absolute. They depend, in particular, on the range of environments associated with the measured population; for example, body weight is more heritable in Sweden than in India, because the range of environments is smaller.

(In the olden days of OTF, Justin used to dismiss results like this on the grounds that the measures used were necessarily imperfect. While that's true, I didn't buy the conclusion then and don't now. Imperfections in measurement would tend to drive the apparent heritability down, not up. We might wish it otherwise, but something relevant to antosociality is being inherited here.)

[Guest]

Heritability of antisociality

does it indicate the effect of the fluid social dynamic or is hertible anti sociality fixed.

[Wyatt Earp]

Heritability of antisociality

I don't understand the question, I'm afraid.

[Guest]

Heritability of antisociality

The fluid social dynamic- is that another name for beer?

[Tubby Isaacs]

Heritability of antisociality

Can you explain why imperfect ways of measurement tend to drive results in the direction you're saying?

[Wyatt Earp]

Heritability of antisociality

Because the higher the level of random noise, the looser any correlation will appear to be.

[Rogin the Armchair fan]

Heritability of antisociality

Random noise certainly makes me antisocial. I lean out of my windows and shout "will you just take that somewhere else, you cunts!".

Ice-cream vans tinkling around my road of a Saturday afternoon, Church bells ringing on Sunday morning, that kind of thing.

[DCI Harry Batt]

Heritability of antisociality

Thanks, WE. Off out for the day now but will take a good look at my leisure.

(As you know, I jerk my knee against this kind of thing. )

[Bruno]

Heritability of antisociality

Is 'antisociality' the same thing as antisocial personality disorder (ASPD)? Sociopaths and psychopaths?

[Anton Gramscescu]

Heritability of antisociality

12 studies is pretty small for a meta-analysis. Though I suppose getting the right number of twins together for these kind of heritability studies is difficult.

That 50% number - I seem to recall from Pinker's The Blank Slate that he attributed the same number to heritability in the variation of intelligence (though I can't remember exactly how this term was operationalized). D'you think this might be some sort of generalizable rule about cognitive and behavioural characteristics?

[Brandenburger Toro]

Heritability of antisociality

I have severe methodological problems with This Kind Of Thing, because of what we term the "graining" problem. If by "antisocial" you mean something very, very specific, then the heritability is likely to be minimal. If you mean something general - which the studies seem to - then the heritability is fairly trivial. So it's very hard to identify usable or interesting criteria. Do we mean playing your music too loud on a particular night is heritable, or do we mean mean causing annoyance to anybody in the course of your life is? Why do we imagine that the things we are grouping together as "antisocial" have anything biologically significant in common?

Plus, you know, This Kind Of Thing really encourages some people to get their Oliver Wendell Holmes on. Three generations of antisocial behaviour is, after all, enough.

[Wyatt Earp]

Heritability of antisociality

Toro, both meanings of "antisocial" you're batting around seem to correspond to colloquial, rather than technical, usages, meaning I'm not really sure what you're saying. Are you saying the term is ill-defined, graining-wise or otherwise, in the psychological literature? You've got more work to do, if so, I think.

Your second point, of course, has no bearing on whether or not the claim is true. It's also the wrong ground on which to argue against OWH. For on thing, 50% isn't 100%, or close to it. For another, OWH is guilty of judging individuals (potential ones, in this case) by group criteria. But most importantly, he's being authoritarian, anti-democratic and un-Kantian, and that would be true even if the heritability was 100%.

[Wyatt Earp]

Heritability of antisociality

Bruno wrote:

Is 'antisociality' the same thing as antisocial personality disorder (ASPD)? Sociopaths and psychopaths?

It's that, plus less extreme manifestations thought to be on the same spectrum; persistently affectless, un-empathetic behaviour.

[Brandenburger Toro]

Heritability of antisociality

Wyatt - yes, that is what I'm claiming, and no, I think it's fairly self-evident from most of the Ev Psych literature. I'm appealling to a colloquial usage in explaining my point colloquially, yes, but that's not really the point. All such terms are almost necessarily ill-defined, because it simply begs the question to assume that the behaviours we are lumping together have anything biologically in common.

The second point isn't about whether it's correct or not, and it's not about OWH, either. There're some scientific questions it's dickish to ask, whatever the answer turns out to be, and this gets along to that territory. As Tom Lehrer put it; "'Once the rockets go up, who cares where they come down? That's not my department!' says Wernher von Braun..."

[Wyatt Earp]

Heritability of antisociality

Also Sprach Zaratoro wrote:

Wyatt - yes, that is what I'm claiming, and no, I think it's fairly self-evident from most of the Ev Psych literature. I'm appealling to a colloquial usage in explaining my point colloquially, yes, but that's not really the point. All such terms are almost necessarily ill-defined, because it simply begs the question to assume that the behaviours we are lumping together have anything biologically in common.

This is behavioural genetics, not Ev Psych. And "antisociality" comes from mainstream psychiatry. I can't see an Ev Psych angle here, really (which isn't to say there isn't one).

Also Sprach Zaratoro wrote:

There're some scientific questions it's dickish to ask, whatever the answer turns out to be, and this gets along to that territory.

Hmm. I'm not convinced at all that questions about heritability of personality are inherently "dickish". Certainly, the fact that people might wrongly interpret the answers doesn't establish this.

This arose in the context of asking whether parents who smack their kids are or are not inflicting long-term psychological harm on them. Should we not ask that question? Or should we ask it, but ignore evidence that comes from lines of inquiry that Oliver Wendell Holmes might misuse? The thread was in response to a question from TonTon; should he not have asked it?

[Brandenburger Toro]

Heritability of antisociality

I should use lower case for ev psych, I suppose. But while the term is broadly useful in mainstream psychiatry, because what is of interest is primarily the symptoms, it's not very useful once you start trying to assess their causes. This is a bit of a methodological problem with psychiatry generally.

Hmm. I'm not convinced at all that questions about heritability of personality are inherently "dickish".

Nor did I come close to saying it was.

[Wyatt Earp]

Heritability of antisociality

Then I don't understand you.

Heritability is about correlation, not cause, by the way, though of course correlation can pertain to cause.

[Brandenburger Toro]

Heritability of antisociality

I'm aware of that. I'm saying firstly that identifying correlations is meaningless if we can't clearly state what we are correlating, and secondly that this is exactly the sort of study we should be immensely cautions about clarity of meaning if we're going to do it, precisely because of the misconceptions that are likely to arise.

[Wyatt Earp]

Heritability of antisociality

Well, I certainly agree with the latter. The term "heritability" is particularly unfortunate in that respect, I think, because it doesn't at all mean "inheritedness". Some clearly inherited traits, such as the ability to recognise faces, have a heritability close to zero.

But it's pertinent in the context of the other thread, because it goes to the question of how important variation in upbringing is in the variation of antisociality, as opposed to (a) variation in heredity and (b) variation in the social environment outside the home. I don't think that's a question it's dickish to ask.

[DCI Harry Batt]

Heritability of antisociality

So what does "heritability" mean, then?

[Wyatt Earp]

Heritability of antisociality

It's a measure of the contribution to the variation in a trait that's made by variation in heredity.

[Bruno]

Heritability of antisociality

Okay, now pretend you're on t.v. and I'm Regis Philbin. What's heritability?

[Wyatt Earp]

Heritability of antisociality

Who?

[Wyatt Earp]

Heritability of antisociality

OK, think about body weight.

1. Body weight varies.
2. Part of this variation is environmental; starve one twin and make ample food available to another and the latter will be heavier.
3. Part of this variation is hereditary; make ample food available to two unrelated people and they may very well not end up weighing the same, or close to the same (unlike twins).
4. It turns out that in the case of variation, it makes mathematical sense to say that x% comes from one factor and y% from another and z% from a third, etc.
5. This only works with variation; you can't say anything along those lines about the trait itself; 4 feet of my height come from my Mum and Dad and 2 feet from eating food, or something. Doesn't make sense.
6. So in principle, we can partition the variation in height into hereditary and environmental bits, plus some random noise we can't assign.
7. These can be estimated in practice, too, through things like twin and adoption studies.
8. The same applies to things like scores on psychometric tests.
9. Such scores may, and probably do, reflect what they purport to measure only imperfectly, and indeed what they purport to measure may be problematic.
10. However, high scores of heritability can only be ignored or discounted if we assert that the scores are completely meaningless and unrelated to anything, which I simply don't buy.
11. And in any case, any noise in the test scores will, as I say, tend to drive estimates of heritability down, not up.

[Bruno]

Heritability of antisociality

Thank you. Very interesting; sounds (to me) like it could get phenomenally complicated very quickly. But I'm reassured there are people working on such things, so that I don't have to. I suppose I would be most interested in how heritability applies to musical talent, but then that's probably not very mathematically measurable is it (other than, perhaps, something like say absolute pitch).