The Nurture Assumption

[Dylab]

I have recently read "The Nurture Assumption" by Judith Harris and was hoping to elicit some opinions on the arguments it makes from others who have read it. For people who haven't read the article the Wikipedia URL is here: http://en.wikipedia.org/wiki/The_Nurture_Assumption

I don't have any major problems with the arguments presented in the book. Her theory using peer groups is pretty interesting although fairly speculative. I don't think she would neccesarily disagree wtih that point though. My reading on the subject beyond the book isn't great so I'm not putting a large amount of confidence in my opininion.

[Roboramma]

Hi Dylab, I haven't read the book, (and wikipedia is blocked in china so I can't read your link). What is the argument that she's presenting?

[Meadmaker]

I have no scientific evidence to back it up, but the wikipedia article seems spot on to me. That's why my kid is in private school.

Actually, since becoming a parent, I have leaned more heavily toward the "nature" side of things in forming personality traits. I just think my kid showed certain traits very much like his parents, but before he could have learned them.

Of course, it may be that the most influential people on child development are not the parents, or the peers, or...whoever, but rather it may be whoever the kid happens to spend the most time with. In American (and most western) society, once kids start school parents are well down the list of people that kids interact with.

[Dylab]

Originally Posted by Roboramma

Hi Dylab, I haven't read the book, (and wikipedia is blocked in china so I can't read your link). What is the argument that she's presenting?

Yea I can do it. I've reread the wikipedia link and it isn't that good anyway.

The book is mainly two arguments. The first argument is against what Harris refers to as the nurture assumption. The idea that a child's enviroment basically is parenting and this is a major factor of how a child's personality develops. The second argument is Harris own theory to replace the non-genetic factor which uses peer groups.

The first argument is largely based on the work of behavioral geneticists. Behavioral geneticists measure various personality traits of siblings raised together, siblings who were adopted into different families and adopted children who were raised together as siblings. Identical and fraternal twins are tested as well but there are obviously not a large number raised apart.

As everyone expected there is a strong correlation in personality traits between natural siblings. What wasn't so expected was that this correlation wasn't very much different between siblings raised together and siblings raised apart. Furthermore the correlation between adopted children raised as siblings was small. Harris argues that the correlation value is trivially small.

The counter argument to this is that the home enviroment is different for each child. Basically refered to as the "micro-enviroment". As a response Harris refers to a comprehensive review of a large number of studies done on the issue of birth order and personality which came to the conclusion that birth order in siblings have no effect. Furthermore children are treated differently as a result of their birth order and thus have a different micro-enviromnet. Harris argues this along with other similar examples is strong evidence against intrafamily factors eliminating the expected correlation.

On to the second argument. Harris theorized that from an early age children's main social function is to become a good member of the peer group. This function results in adapthing personality traits what last well into adulthood. I am oversimplifying her argument here but it is the basic idea.

Much of the book is devoted on the study of group social dynamics in children (and somewhat on gorillas). She discusses a well known study which took two groups of 9 very similar boys in what they though was a summer camp. For two to three weeks they kept the groups apart keeping the existance of the other group secret. After the time period the "counselors" had the two groups compete. Immediately there was hostility between the groups. The important factor Harris notes is that the members of each group increased the similarities amongst themselves and increased the differences between the two gropus.

The evidence for the second part of her argument is unfortunately not as concrete as the evidence used in her first argument. It is difficult to decouple ones peer group and its effects and other factors like genetics. As a result Harris resorts to annecdotal evidence. She uses cases where no other factors could come to play. For example she notes that the children from immigrant families quickly learn their adopted country's language as well as the common personality traits even though the parents are resistant to chage. My favorite example and one that seems to be a common occurance is of a child around 5 who is convinced that women should not be doctors and men should not cook. This despite the fact his facter is a chef and his mother is a doctor.




I should edit this post but it is late and I need to sleep. I'll edit the unclear parts tomorow.

[Dancing David]

If the correlations are less than 60% they don't matter, the people who want to claim anything based on low levels of correlation are just wishing they had something.

There are so many counter claims that it boggles the mind, how many people come from families with alcoholism that don't have an alcohol problem are there.

There are no simple answers.

[Roboramma]

Originally Posted by Dylab

Yea I can do it. I've reread the wikipedia link and it isn't that good anyway.

Wow, thanks Dylab. Very clear rundown of the book. Cool.

Quote:

Identical and fraternal twins are tested as well but there are obviously not a large number raised apart.

Hmm... I didn't think that was necessary - I thought that by looking at how identical twins (raised together) are more similar to each other than fraternal twins (raise together) are to each other, we could also learn things about genetic vs. environment factors.

Quote:

Harris refers to a comprehensive review of a large number of studies done on the issue of birth order and personality which came to the conclusion that birth order in siblings have no effect.

That's interesting, because I've read of (haven't seen the studies myself) studies showning the opposite. The idea being that evolution would favour children who adapted their personality to their birth order, because being a second child is very different from being a first child. Parental strategies are different toward different children, and so the children's strategies should be different too.
I'd be interested to see what the truth is here.

Quote:

On to the second argument. Harris theorized that from an early age children's main social function is to become a good member of the peer group. This function results in adapthing personality traits what last well into adulthood. I am oversimplifying her argument here but it is the basic idea.

So far this makes a lot of sense to me, and while I don't know of any data that specifically supports the idea, at least it seems to account for what we do know.

Quote:

Much of the book is devoted on the study of group social dynamics in children (and somewhat on gorillas). She discusses a well known study which took two groups of 9 very similar boys in what they though was a summer camp. For two to three weeks they kept the groups apart keeping the existance of the other group secret. After the time period the "counselors" had the two groups compete. Immediately there was hostility between the groups. The important factor Harris notes is that the members of each group increased the similarities amongst themselves and increased the differences between the two gropus.

Just to clarify - did they increase the intragroup similarities before or after being introduced to the other group?
I actually sort of suspect both, which would seem to make a lot of sense. It's in each boy's interests to get along with the others in his group and to fit in with them. Even more so when they are involved in conflict with another group and their own group cohesion becomes an advantage in that conflict. But it certainly suggests that peer group dynamics can have at least a very important short term impact on personality.

On the whole it sounds like a very interesting book. Its an important topic with a lot of unanswered questions. Which is why it's so cool to see those questions slowly being answered the only way they were ever going to be - with science.

[AmateurScientist]

Originally Posted by Dancing David

If the correlations are less than 60% they don't matter, the people who want to claim anything based on low levels of correlation are just wishing they had something.

There are so many counter claims that it boggles the mind, how many people come from families with alcoholism that don't have an alcohol problem are there.

There are no simple answers.

That doesn't mean correlations are insignificant. Counterexamples in this instance mean little other than there isn't a perfect 1 to 1 correlation. No one is claiming that every child of an alcoholic parent will become an alcoholic. Where did you get 60% as a baseline? The scientific question is whether one observes a greater incidence when a controllable variable is introduced, and whether that greater incidence is better than one might expect due to chance. Your 60% seems to be arbitrary.

The observation which has been made in different studies is that having an alcoholic parent increases one's chances of becoming an alcoholic. The interesting questions then become, is there a genetic component to the increased risk, and is there a behaviorial or environmental component? The answers seem to be yes to both. That raises other questions, such as to what degree does nature vs. nurture have an effect, what are effective countermeasures to diminish the risk, etc.?

AS

[Wudang]

I'm reading Pinker's "Blank slate" at the moment. His introduction is interesting as he talks about the degree to which any discussion of the "nature" element is treated as heresy by many academics and lead to strawman attacks on extreme interpretations of claims, claims that any support of the nature hypothesis leads to social Darwinism and (Godwin rule).
Harris' book sounds like a good followup for my list.
As a father of a 16-month old girl it staggers me how many behaviours she has inhertited from her mother.

[Cuddles]

Originally Posted by Dylab

The book is mainly two arguments. The first argument is against what Harris refers to as the nurture assumption. The idea that a child's enviroment basically is parenting and this is a major factor of how a child's personality develops. The second argument is Harris own theory to replace the non-genetic factor which uses peer groups.

I haven't read the book, but this argument appears to be purely a straw man. Obviously, a child's environment is every single person they interact with. Since they are with their parents a large amount of time, especially while very young, their parents will have a large influence on them. However, as soon as they start joining peer groups and spending lots of time in these, their peers will have a large influence. I very much doubt anyone has ever claimed that only parents influence children.

The study that looked at children in a summer camp was viewing them under conditions where they only interacted with their peers, so of course their peers had a big influence. If the study had looked at children who spent two weeks with only their parents who would have had a big influence then?

[AmateurScientist]

Originally Posted by Wudang

I'm reading Pinker's "Blank slate" at the moment. His introduction is interesting as he talks about the degree to which any discussion of the "nature" element is treated as heresy by many academics and lead to strawman attacks on extreme interpretations of claims, claims that any support of the nature hypothesis leads to social Darwinism and (Godwin rule).
Harris' book sounds like a good followup for my list.
As a father of a 16-month old girl it staggers me how many behaviours she has inhertited from her mother.

That's an interesting insight into just how far progressive social politics, especially a particular brand of radical feminism, and later the political correctness wave in the 90s, has infiltrated academia and placed certain topics off limits for serious discussion. It's amazing that studying possible genetical influences on behavior equals advocating eugenics to certain persons who adhere to certain political views.

AS

[Dylab]

Originally Posted by Roboramma

Wow, thanks Dylab. Very clear rundown of the book. Cool.

Hmm... I didn't think that was necessary - I thought that by looking at how identical twins (raised together) are more similar to each other than fraternal twins (raise together) are to each other, we could also learn things about genetic vs. environment factors.

Yea of course. That is done as well. I think the idea is to have as many data points as possible.

Originally Posted by Roboramma

That's interesting, because I've read of (haven't seen the studies myself) studies showning the opposite. The idea being that evolution would favour children who adapted their personality to their birth order, because being a second child is very different from being a first child. Parental strategies are different toward different children, and so the children's strategies should be different too.
I'd be interested to see what the truth is here.

I should stress that the review is somewhat contentious. Ed Sulloway in his book on birth order effects reanalyzed the study and came to the conclusion that birth order does have an effect. Harris includes an essay attacking his conclusion. In this case it is basically messy data anylisis. I'm not too familiar with the more recent stuff though. At her website there are four essays on the subject including the essay against Sulloway. http://home.att.net/~xchar/tna/birth-order/index.htm

I should also mention that she argues for context-specific behaviour. Basically people use different programming depending on the social context. Harris does believe that parents can affect the home social-context behaviour. Children act differently at home depending on the strictness of their parents for example. However she differs from many researchers in that she doesn't believe children take these effects outside the home.

For what it is worth, from an evolutionary stand point I would imagine this makes the most sense. Obviously it would be very auspicious for one to adapt their behaviour according to the home enviroment. However it wouldn't make much sense to make use of the adaptions in a completely different enviroment.

Originally Posted by Roboramma

Just to clarify - did they increase the intragroup similarities before or after being introduced to the other group?
I actually sort of suspect both, which would seem to make a lot of sense. It's in each boy's interests to get along with the others in his group and to fit in with them. Even more so when they are involved in conflict with another group and their own group cohesion becomes an advantage in that conflict. But it certainly suggests that peer group dynamics can have at least a very important short term impact on personality.

Your suspicion is correct. In the example I used one of the groups consciously decided to stop swearing completly because that the other group swore a lot. In another example she discussed the study of a group of girls playing dodgeball. When they were by themselves they were fairly aggressive and physical however when they played against a group of boys they were much less aggressive... more "girl-like". In this case the change was unconcious.

I think she would disagree that it is because of the increase need for group cohesion in time of conflict. Rather she would argue for children's desire to increase "groupiness". One way to do this is by increasing the lines between other groups.

[ponderingturtle]

Originally Posted by Meadmaker

I have no scientific evidence to back it up, but the wikipedia article seems spot on to me. That's why my kid is in private school.

Actually, since becoming a parent, I have leaned more heavily toward the "nature" side of things in forming personality traits. I just think my kid showed certain traits very much like his parents, but before he could have learned them.

Of course, it may be that the most influential people on child development are not the parents, or the peers, or...whoever, but rather it may be whoever the kid happens to spend the most time with. In American (and most western) society, once kids start school parents are well down the list of people that kids interact with.

Well to make this issue more complicated there have been interesting studdies in monkeys that showed that nature is much more than DNA. They showed that some behavior was hereditary, but not genetic or learned. They did this by swaping the invetro fertilizations between different mothers and seeing how that effected things. It stayed dependant on the birth mother and not the genetic mother.

[Dylab]

Originally Posted by Cuddles

I haven't read the book, but this argument appears to be purely a straw man. Obviously, a child's environment is every single person they interact with. Since they are with their parents a large amount of time, especially while very young, their parents will have a large influence on them. However, as soon as they start joining peer groups and spending lots of time in these, their peers will have a large influence. I very much doubt anyone has ever claimed that only parents influence children.

The study that looked at children in a summer camp was viewing them under conditions where they only interacted with their peers, so of course their peers had a big influence. If the study had looked at children who spent two weeks with only their parents who would have had a big influence then?

Maybe I overstated her attack. I don't think she ever stated that people believed parenting was the only part of the enviroment only that parenting was very significant aspect of the enviroment. I'm sure no one discounts the effect of peer groups totally but many people do have a tendency to blame parents if their children screw up. In general she argues that a large amount of psychology tries to link (in her opinion unsuccesfully) behaviour of offspring with the parenting styles used based on the idea of the Nurture Assumption. Having the tendency to ignore the effects of heredity and outside social enviroment.

The study I mentioned was used to only look at group dynamics and behaviour. I don't think she intended to use the study to argue that children have a long term personality influence.