less gendered than externalizing behaviours.
Hundreds of studies have shown that stress caused by verbal and physical violence, mental illness, and substance abuse greatly increases children’s chances of suffering from excessive aggression, alcoholism, and ADHD. Though this correlation is well-documented, it is not incontrovertible. Plenty of children come from stressful homes yet grow up to be well-adjusted adults. What benevolent force saves them from the downward spiral of anger and violence that claims so many of their peers? Is it blind luck? Is it social support? Or do they possess some innate genetic filter that keeps environmental pollutants out of their systems?
A Canary in the Coal Mine
A 2005 study led by Sara Jaffee compared the effects of a stressful home environment on identical and fraternal twins. Fraternal twins are conceived when two eggs are fertilized by two different sperm, while identical twins occur when a fertilized egg — called a zygote — splits into two separate organisms during its first few replications. Identical twins have identical genomes, which explain their great physical similarities. Fraternal twins, on the other hand, are simply two siblings who shared a womb. Genetically speaking, they are no more alike than a typical brother and sister.
In technical terms, identical and fraternal twins are distinguished by their zygosity, or the number of fertilized eggs from which they were born. Identical twins, born from a single zygote, are called monozygotic, while fraternal twins are called dizygotic.[21]
| Zygosity | Twin type | DNA in common |
| Dizygotic | Fraternal (non-identical) twins | 50 percent |
| Monozygotic | Identical twins | 100 percent |
The goal of Jaffee’s study was to show that a person’s genes were correlated to his or her sensitivity to the long-term effects of maltreatment. She developed a framework that stratified children’s risk of exhibiting externalizing behaviour based on their zygosity and the presence or absence of externalizing symptoms in their twins. This premise may seem confusing, but it is actually fairly simple. If a trait has a genetic link, then one of your relatives possessing that trait increases your chances of possessing it as well. The closer that relative is to you genetically, the more alleles you share, and the greater the odds that you both display the same version of a given trait. Dizygotic twins have roughly 50 percent of their polymorphisms in common, so the odds of them sharing a genetic trait are approximately one in two. Monozygotic (identical) twins, on the other hand, have 100 percent of their polymorphisms in common, so the odds of them not sharing a genetic trait are slim to none (although it is possible for a gene to mutate shortly after the zygote separates, and certain genetic traits can be switched on or off by the environment, a phenomenon we will address in more detail later in this book). Therefore, if you observe a genetic trait in one dizygotic twin, there is a decent chance it will be visible in the other twin as well; observe the same trait in a monozygotic twin, and its presence in the other twin is basically guaranteed.
Bearing this in mind, Jaffee developed a 4-point scale of genetic risk, its underlying principle being that a symptom in one twin predicts the presence of that same symptom in the other twin. Your sibling becomes a sort of canary in the coalmine of your genes, their behaviours a genetic portent of what you may one day face.
The lowest-risk children were those whose monozygotic (identical) twins showed no signs of externalizing behaviour. Considering these twins share all of their DNA, the absence of a genetic trait in one twin can safely predict its absence in the other. Next were children whose dizygotic twins showed no signs of externalizing behaviour. Though the trait could arise from the 50 percent of alleles the children don’t share, we can at least guarantee that one half of their DNA is safe. Riskier still were children whose dizygotic twins showed signs of externalizing behaviour. At this level of the scale, we have confirmed that the trait exists within the subject’s immediate family. Perhaps they inherited it or perhaps they did not, but it is a confirmed possibility. Lastly, children whose monozygotic (identical) twins showed signs of externalizing behaviour were the highest risk of all. If one twin has the trait, and the trait comes from a polymorphic variation, then the other twin almost certainly has the offending variation, too.
Which is not to say that he or she has the trait.
Anyone who has met a set of identical twins could tell you that the term “identical” is not, in the most technical sense, accurate. Spend enough time with them and you will be able to distinguish one from the other just by looking at them. The telltale sign may be a mole or a freckle, or a slight difference in height or weight, or a result of their varying mannerisms and inflections. Though casual acquaintances may not be capable of discerning one from the other by appearance alone, a brief conversation should suffice. The “identical” aspects of monozygotic twins rarely extend to their personalities. In fact, identicals often exaggerate the differences between each other, perhaps in order to more firmly establish unique identities. A twin girl who is slightly more extroverted than her identical sibling will make a point of acting as outgoing as possible, while her more reserved twin will draw further inward. These behaviours change the way each twin is treated by her teachers, her peers, and even her own family, which in turn reinforces the differences between them.
Living things are more than the sum of their parts. Genes are not our overlords, especially when it comes to complex behavioural traits. One’s environment is equally important, and we would expect the home life of these children to play an important role in their development (or lack thereof) of externalizing behaviours. Jaffee is, of course, aware of this, which is why her study grouped children based on environmental factors as well as genetic ones. More specifically, she considered whether or not they had been physically maltreated.
Interviewers asked the twins’ mothers a series of “probe questions” designed to elicit conversation about their children’s past maltreatment without coming across as accusatory or hostile. Questions included “Do you remember any time when your child was disciplined severely enough that he may have been hurt?” and “Did you worry that you or someone else may have harmed or hurt your child?” The interviewers chose their words carefully, avoiding any implications that the mother may have been at fault. If the mother reported any maltreatment, the interviewer probed for details, keeping careful notes on what happened, when, how often, and if the police or child services were involved. They also kept an eye out for subtler clues, observing how the mother and child acted during their conversation. A nervous glance, downcast eyes, fidgeting, and increased tension in the presence of the father, or any hint at a darker truth was noted, analyzed, and used to determine whether the child was potentially being maltreated. The evidence didn’t need to be ironclad; for inclusion in the study’s test group, children could be classified either probable or definite cases of maltreatment.
Jaffee took the maltreated and non-maltreated cohorts (containing 307 and 809 children, respectively), subdivided each of them into groups based on her 4-point risk scale, and tallied the frequency of conduct problems in each group, as reported by the children’s mothers and teachers. Conduct problems included fighting, bullying, lying, stealing, cruelty to people or animals, vandalism, and breaking rules. As expected, each degree up the maltreatment risk scale correlated to an increase in conduct disorder in both cohorts. Lowest risk children, regardless of whether or not they were maltreated, were less likely to act out than children of the same cohort in the next bracket up. However, within each risk group, the maltreated children were more likely to exhibit conduct disorders than the non-maltreated children. Moreover, the gap between risk levels was wider in maltreated than non-maltreated children. As the risk level rose, the amount of maltreated children displaying conduct disorders increased considerably more than it did in the non-maltreated cohort. In the lowest genetic risk group, maltreatment caused conduct disorders to rise 2 percent compared to non-maltreated children of the same risk level; in the highest genetic risk group, the increase was 24 percent.
Jaffee, S.R., Caspi, A., Moffitt, T.E., Dodge, K.A., Rutter, M., Taylor, A., and Tully, L.A. (2005). “Nature × Nurture: Genetic Vulnerabilities Interact with Physical Maltreatment to Promote Conduct Problems.” Development and Psychopathology, 17(1), 67–84.
Jaffee’s study is