Anyone who is daft enough to question the most cherished notions of a culture had better be prepared for the flak that follows. I am the author of two controversial books, The Nurture Assumption and No Two Alike , both of which question the notion that parents play a central role in shaping their child’s personality. In an essay in Prospect magazine published earlier this year, I revealed my heretical views and was widely attacked.

I’m used to being attacked for my beliefs. What I can’t get used to is being attacked for things I never said and don’t believe. It is true that I say that the important environment for children is the world outside the home. I also maintain that it is peers, rather than parents, who play a central role in the formation of the child’s personality. But I don’t claim it’s “peer pressure” that does it. Nor do I base my views on my personal experiences as a mother and grandmother. My theory of personality development is firmly based on evidence from hundreds of research studies.

To someone involved in the day-to-day job of rearing children it may seem irrelevant to speak of research evidence. Why should we have to do research on this topic? Aren’t our intuitions enough? Shouldn’t we, therefore, reject out of hand any counterintuitive results? That was the position of some of the commentators who wrote about my Prospect essay. Well, if intuition were always correct, there would be no need to do research. And, if it were always correct, the results of research would never surprise us, and sometimes they do surprise us.

Some of the research described in my books produced results that surprised even the scientists who carried it out.

These scientists belong to a field called behavioural genetics, but the results that surprised them were not about genetics. They were not surprised that children turn out the way they do partly because of genes inherited from their parents. What shocked the researchers was what they discovered about the environment. It turned out that virtually all resemblances between parents and children — the tendency of parents who are good readers to have children who are good readers, the tendency of aggressive parents to have aggressive children — are due to genetic influences on these traits. To heredity. The evidence indicated that the environment provided by the parents had no important or lasting effects on such traits.

Notice that I said “lasting”. There is no question that parents have short-term effects on their children. They certainly have effects on how their children behave at home. There are also short-term effects on the children’s vocabulary and other measures of intelligence. After all, young children cannot learn words they never hear! But the child who never hears a three-syllable word at home will eventually hear (or read) it outside the home. Research on adopted children, whose genes were provided by one set of parents and whose environment is provided by a different set, shows that the effects of the rearing home gradually fade. By late adolescence, adopted children reared in three-syllable homes are no smarter, on average, than those reared monosyllabically.

I spent a year reading research results like those. By the end of that year, I had lost faith in the effectiveness of the home. I turned my attention to the environment outside the parental home, where children are destined to spend their adult lives. If childhood is preparation for adult hood, shouldn’t the child be alert to what’s going on in the wider world? Isn’t getting along with one’s parents — frankly — a short-term goal? Not exactly. Parents were given short shrift in The Nurture Assumption but they fare somewhat better in No Two Alike . In the eight years between the two books, my theory — shall we say — matured.

In No Two Alike , I propose that children have three social goals to accomplish in childhood. The first is to form and maintain good relationships with the significant people in their lives. The second is to fit in — to adapt to their culture. The third is to work out a strategy for competing successfully with their rivals — to strive for status. Specialised departments of the brain, called “modules” or “systems”, provide the motivations (and other cognitive equipment) to accomplish these goals.

I do not deny that parents are important people in a child’s life. If a good relationship with a parent is forged in childhood, it is likely to be maintained. Parents don’t stop being important merely because one has reached adulthood. But relationships with parents leave no permanent marks on the personality. Research has shown, counterintuitively, that the way someone behaves with his father does not predict how he will behave with his boss. Likewise, the way someone behaves with her siblings does not predict how she will behave with her friends. Children who fight with their siblings like cat and dog are fully capable of having serene relationships with friends. Though parents and siblings play a major role in the relationship system, it is a limited one.

For the socialisation system — the one that motivates children to fit in — the important people are peers. But it’s not a matter of peer pressure. Children don’t have to be pushed to conform; they want to.

They learn to behave in a way that’s appropriate to their culture by observing their peers. They can’t just imitate their parents because parents are adults, and children aren’t expected to behave like adults. A child who behaved like an adult would appear abnormal. So the way children become socialised, according to my theory, is by figuring out what kind of people they are — child or adult, male or female — and then observing how that kind of person behaves. A little girl gets an idea of appropriate little-girl behaviour by observing other little girls. This kind of learning goes on largely at an underground level; it never reaches the conscious mind. That’s why people’s intuitions about how they became socialised are likely to be wrong.

Peers are important in a different way for the status system. Children are not only motivated to conform to their peers: they are also motivated to be better than their peers. But better in what way? To figure out where their efforts are most likely to yield success, children have to learn about their own strengths and weaknesses. They obtain this self-knowledge by comparing themselves, or by being compared, with their peers. A child doesn’t have to know more than his father to consider himself smart, but he has to know more than the other kids his age.

All this goes on in the world outside the parental home. But parents have some control over that world, too. By deciding where they will live and where their children will go to school, parents determine who their children’s peers will be. If they move to Canada, their children’s peers will be Canadian. Such decisions have lasting effects. Though the parents will retain their British accent, their children will soon acquire the accent of the peers they go to school with. Accents are an informative indicator of socialisation, because (unlike most behaviours) they aren’t influenced by genes. When children resemble their parents in liking to read or behaving aggressively or being religious, the strands of genetic, parental and cultural influences are hard to separate. With accents we can separate them. Accents are part of a culture (or subculture) and children get their culture from their peers.

Peers are important all through childhood. But the term “peer group” is usually applied to teenagers. In adolescence, young people sort themselves out into different groups within a school or neighbourhood on the basis of their interests, abilities, propensities or ethnic groups. Some groups are composed mainly of kids who couldn’t find another group to accept them. Adolescent peer groups clearly have a short-term influence on the behaviour of their members. Whether they have a long-term one is difficult to judge because most of the characteristics that distinguish the members of these groups were present well before the groups were formed.

Do parents have any power to determine which peer group within a school their teenager will join? It’s theoretically possible — music lessons, tennis lessons or clothing might make a difference — but I haven’t seen any convincing evidence. Most studies done on adolescent peer groups are of little value because the research method provides no way of distinguishing environmental effects from the effects of genes. My guess is that parents generally have little or no influence. Even teenagers who would like to follow their parents’ advice may be powerless to do so because in many cases it isn’t the kid who selects the peer group; it’s the peer group that selects the kid.

Judith Rich Harris is the author of The Nurture Assumption (Bloomsbury, 1998) and No Two Alike: Human Nature and Human Individuality (Norton, 2006)

Times recommends

• The pink mile: women on death row
• The top 10 richest streets in the world
• Japan's new professional seducers