Tuesday, May 15, 2007

A Gene for Children's Bad Behavior? Marital Conflict, Child Conduct, and Genes

Marital problems often make for some troubling child behavior – bullying, shoplifting, running away. A good deal of research links parents’ marital conflict to child conduct disorder, psycho-jargon for all that bad behavior, you know, the neighborhood deviant who set your mailbox on fire. No surprise, right? Kid hears all that yelling, the screaming, the door slamming, so then he or she acts out.

Not so simple according to researcher K. Paige Harden and associates reporting on their latest research in the February issue of Child Development. Their conclusion? Blame it on the genes.

The genes? Parents with an intrinsically combative personality (that is in some part genetically programmed) pass this on to their children. Genes spawn aggressive, argumentative parent. Genes spawn aggressive, argumentative child. Bad behavior from all parties ensues. Or as one MSNBC internet article spun it, some are “born to be bad”.

Now how did Harden and crew figure this out? Has a bad-behavior gene been identified on the human genome? Did Harden swab all these people for their DNA? Are we watching too much CSI? Probably not though there are many a social scientist doing swabs, brain scans, and other techniques once considered the realm of only medical types.

What we have here is an example of a classic twin study. Designed to untangle the genetic and the environmental components of a particular behavior or characteristic. Here’s how it works. Identical twins share all of their genes (although there is some recent evidence disputing this longheld belief). The fraternal twins share only a portion of genes (50%). If a trait is genetically influenced, then we expect the identical twins will be more similar on this trait then the fraternal twins who share only some genes. On the flip side, if the trait is not genetically influenced then we expect not only will the identical twins be dissimilar (given their different lives, different spouses, etc.) but so will the fraternal twins – the trait being influenced only by environment.

Our study today is a complicated version with more than 1,000 twins and their more than 2,000 children. Let’s recap. Hardin and team expected child conduct to be influenced by genes and not so much by parental marital discord. So let’s take the identical twin families. If both identical twins in the family had lots of marital conflict, their children showed lots of bad conduct. The reverse was also true, kids of peaceful marriages showed few conduct issues. But here we can’t say whether the kid behavior was the results of parent behavior or genes.

So we look to families with one twin in a bad marriage and one in a good marriage. Obviously here the parents have the same genes but different marriages, and thus different environments for their children. We can’t attribute their children’s behavior to genes because these children have their parent’s identical genes. So difference between their children should be attributed to something other than genetics, say marital conflict. Now for children of fraternal twins, who share fewer genes, we expect the difference between them to be even greater (some of the difference arising by the unshared portion of the genes). In other words, genetics did account for some of the child behavior. And this is what the researchers found. They also tested a model with all the components – I’ll spare you that heavy statistical feat – only to say that the model supported the larger genetic contribution to child conduct.

Now many of you may be wondering about a missing component in the above design – the spouses of the twins, fathers of the children. Without getting into the stats behind it, let’s just say I’m not sure including them would have changed the results much.

So what does this mean? Genes make children act out? There’s a gene that causes people to yell at their spouse? A gene to start fires? More likely genetic tendencies are expressed through personality, temperament, how we interact with other people, how we perceive our environment. Those tendencies are passed on to children who act accordingly.

Now let’s be clear about what this study did not find, in fact, did not address. First, there are likely a variety factors that influence child conduct disorder as well, ones that do not have heavy genetic influences. They just weren’t included here. Also, let’s not negate the numerous ways that children likely are affected by marital conflict - through depression, school performance, substance abuse to name just a few.

Read the abstract on the Blackwell Publishing website. You'll have to pay for the article. The research design and statistical analyses are pretty complicated but the first few intro pages and discussion at the end are an easy read.

Harden, K.P., Turkheimer, E., Emery, R.E., D'Onofrio, B.M., Slutske, W.S., Heath, A.C., Martin, N.G. (2007). Marital Conflict and Conduct Problems in Children of Twins. Child Development, 78(1), 1-18. DOI: 10.1111/j.1467-8624.2007.00982.x