Wednesday, June 25, 2008

Parents versus peer groups

Daniel Finkelstein has been on the Judith Rich Harris again. His latest article quite rightly takes politicians to task for assuming that parenting skills have the slightest effect on the way children turn out. As I've argued many times, they don't.

However, Finkelstein seems to believe that environment must have some effect. He casts around and comes across Rich Harris's thesis. She claims that it is our peer group that forms us. This is almost certainly rubbish too. Peers have no more effect than parents.

The key to the confusion of both Finkelstein and Rich Harris is that they fail to distinguish between innate characteristics and learnt behavior. Some examples of characteristics are intelligence, extroversion, shyness, propensity to addiction, laziness, mathematical ability and optimism. You get the idea and can probably add many more. Behaviour includes language, reading and writing (although how good you are is largely a characteristic) and mathematical knowledge.

Now it is true that you can learn from both parents and peers. It may also be true that you are more likely to learn from your peers than your parents. But Rich Harris is wrong to extend the ability of peers beyond imparting behavior to also shaping our innate characteristics. They can't. Environment (aside from diet) as almost no effect on them at all.

By assuming we can change people by changing their peer group, I fear Finkelstein will send us off on a wild goose chase as pointless as the nurture assumption.

Discuss this post at Science, History and Religion - James Hannam's Forum

Click here to read the first chapter of God's Philosophers: How the Medieval World Laid the Foundations of Modern Science absolutely free.

6 comments:

Niall said...

Hi James.

The way you dismiss peer relationships, and other environmental factors, when it comes to the outcomes of child development is pretty extraordinary, not to mention misguided.

For some relevant information on the importance of peer and sibling relationships, I'd advise examining the work of Judy Dunn.

Another interesting read would be a 1973 study by Cowen et al which examined a large number of young children. Having taken account of IQ and grade scores, they found that children who had been rejected by their peers at age 8 were far more likely to have needed help due to mental health problems.

More recently a study in New Zealand found that low peer status predicted low social status in adults, even when IQ and poor family circumstances were taken into account.

Now you could argue that peer rejection was a result of some sort of genetically determined personality type or something similar. However that misses the point. We inherit predispositions that only manifest themselves when certain environmental predispositions are presented. By manipulating the environment, we determine which predispositions manifest themselves. There have been studies that show that by intervening in situations where children are not functioning well in social relationships with their peers can have positive results.

Similarly, it has been shown that parenting style is the major factor in determining attachment style. If twin studies are anything to by, then genetics plays only a modest role. Importantly, it has been demonstrated that attachment style during infancy predicts adult outcomes. For instance, it was found that it predicted behaviour in adult romantic relationships and, importantly, adult parenting styles.

The main exception to the rule was in the case of so called earned-secures, those who were insecurely attached as kids and who had a high level of self-reflection.

I find these nature-nurture debates irritating. They often seem to miss the fact children are not just acted upon, they act on those around them. It's an intricate interaction. I'm the last person who'd deny the importance of genetic factors in helping to determine outcomes in later life. What I object to is any suggestion that we are powerless to change genetically determined outcomes. I work in an applied behaviour analysis setting with autistic children. I see the powerful effect that manipulating the environment can have on a daily basis and I shudder when I think of the adult outcomes these children would have had were it not for these interventions.

James said...

Hi Niall,

Thank you for your insightful and interesting commet. I'm sorry to say though, that I'm sticking to my guns. It seems obvious to me that the Cowen study is simply tracking characteristics that the peer group pick up earlier than adults. It is a classic case of confusing correlation with causation. I'd also say ANY study that does not explicitly place genetics to the fore of its considerations is worse than useless. That basically means any study carried out prior to about 1990.

Also, why is anyone surprised that people who are badly socialised as children have low status as adult?. Why should anyone see a cause here when it is clear that the child's characteristics are simply being carried into later life.

I admire the work you and your colleagues do with autistic spectrum disorder children. Clearly and thankfully you can help them by arranging things so that they are best able to function. That doesn't effect the argument though. You don't change the characteristics of the children, you skillfully ensure that their environment is the one in which they can best cope.

The more I've found out about this, the more I've realised that the fact that nurture and peers are irrelevant to characteristics is probably the most important scientific discovery of the last fifty years. No less so that so few people can yet accept it.

Niall said...

Just a couple of clarifications James.

Genetics should be to fore of any research in these areas, and over the past couple of decades it has. I think that it's important to note.

A brief review of research into the genetic basis of attachment is found here:
http://www.child-encyclopedia.com/documents/van_IJzendoornANGxp.pdf

"Four twin studies on child-mother attachment security using behavioural genetic modelling have been published to date. Three of he four studies document a minor role
for genetic influences on differences in attachment security and a rather substantial role for shared environment. The fourth study, the Louisville Twin Study, investigated the quality of attachment in twin pairs with an adapted separation-reunion procedure originally designed to assess temperament. The large role shared environmental factors play in attachment (about 50% in the Bokhorst et al. study)11 is remarkable. Differences in attachment relationships are mainly caused by nurture rather than nature, although the bias to become attached is inborn.

Is sensitive parenting the core ingredient of the shared environment? Twenty-one correlational studies have replicated a significant but modest association between parental sensitivity and infant attachment (r = .24, N = 1099). But only experimental interventions can definitely prove Ainsworth’s original hypothesis. In 24 randomized intervention studies (n = 1280), both maternal sensitivity and children’s attachment security were assessed as outcome measures. In general, attachment insecurity appeared more difficult to change than maternal insensitivity. When interventions were more effective in enhancing parental sensitivity, they were also more effective in enhancing attachment security, which experimentally supports the notion of a causal role of sensitivity in shaping attachment."

Of course, while attachment style is important, it is only one factor. I use it as an example because it is something that has been shown to have important effects on behaviour through the lifespan, because the research indicates it is determined by parenting style and not genetics and most importantly, because it has been shown that by it can be altered.

Second, I think you have some misconceptions regarding what we do with children. Yes, the environment is manipulated, but it is manipulated with the intent of helping children acquire new skills which are then generalised to new environments. We are constantly developing children's abilities to the point where many are able to attend mainstream schools. Research indicates that the ABA (applied behaviour analysis) approach leads to better outcomes than other approaches to education provision. If autistic children's outcomes were determined by nature and not nurture, then the outcomes from all education provisions would probably be the same, no?

Similarly, in the treatment of Depression, different therapies result in different outcomes. Cognitive behavioural therapy has been shown to be the most effective. It is more effective than other therapies, and in some cases, if my memory serves me correctly, it has been found to be more effective than the most effective drugs on the market. Best outcomes result when CBT is used in conjunction with medication. If nurture was not important, then surely all therapies would be shown to be equally useless?

These are the problems that jumped to mind when I read your argument. Am I missing something?

James said...

Hi Niall,

With your permission, I'd like to transfer our discussion to the board:
http://jameshannam.proboards83.com/index.cgi

I expect most people will be against me on this, so don't worry about being out numbered!

Best wishes

James

Charles Freeman said...

I don't often contribute to this blog- it is one of several I note from time to time.
Having brought up four children and having been a teacher for fifteen years, I have noted
1) At sixth form level, how some families/parents have instilled academic skills and work ethics and others have not. Those children who are well organised develop self-confidence and become higher achievers, the others have to struggle. One doesn't need surveys and studies, a teacher picks up very quickly in the classroom those students who have a stable and supportive background and those who don't. It is one of the biggest determinants of success and can certainly make up for a weak genetic background ( say in intelligence). A hunger for learning can certainly be instilled.
2) My youngest daughter's school ( a comprehensive) developed peer groups as a means of improving standards. Students helped each other. Five of her group got five 'As at "A" level and,in my daughter's case ,that was considerably above what she had originally achieved. Again she has blossomed in confidence, has just gone round the world , got a better university place and this all becomes mutually reinforcing. I hope our parenting skills had something to do with it- we certainly worked hard with our children- but the school was fantastic!
I wonder if James will still hold his views when he has actually lived a bit in the world of children!

Niall said...

Fine by me James. Go for it.