Temperament, parenting, and the so-called “non-shared environment” PART I


In the prestigious journal, Science,* researchers informed us in 1990 that  “On multiple measures of personality and temperament, occupational and leisure-time interests, and social attitudes, monozygotic twins reared apart are about as similar as are monozygotic twins reared together.”  This is old news, of course, in science, and perhaps old news to many readers.  But stop a moment and think about what this really implies. First of all, research on both siblings and twins has established strongly that 40-50 percent of the variability in personality/temperament measures is purely genetic.  This leaves 50% or more to non-genetic, or environmental influences.  Monozygotic twins by definition, have essentially identical genes, with some minor caveats.  But-just when developmental psychologists were getting used to the idea that the environment–thus nurture–controlled about half of our eventual personality traits, interests, etc. –the non-shared environment burst upon the scene.

Think again about the fact that taking the monozygotic twin pair and putting each one in a different adoptive family simply did not affect their resemblance to one another in the measured behaviors.  Hmm.  Statistics for separating genes and environment are moderately complex but the bottom line is simple. One hundred percent resemblance minus the roughly 50 % resemblance due to genetics leaves you the other fifty percent.  If a pair of twins raised together in a single family do not resemble each other more than a pair raised separately there does not seem to be an environmental effect that they shared from family.  Since the  environmental effect is not shared, it is classified as non-shared.  The general belief, at present is that no more than 5% of environmental effects are shared effects–and perhaps not that much. It might help to think of non-shared effects as effects that are unique to each twin.

There are undoubtedly what we might call “background” effects that are present in both homes for the pair.  Neither twin is from an environment that failed to provide for basic needs, failed to feed, clothe, educate and generally take care of them.  If we were comparing a home with abuse, physical battering, etc. for one twin, and not for the other, we would see an effect of these underlying basic provisions.  That is not the point here, however.  Homes were assessed both physically for tools playthings, books etc.; families for amounts of education and income.  There clearly were differences on these factors, but not beyond acceptability, and such differences as there were, had very little effect on twin resemblance.

And there is more.  When you look at fraternal twins and non-twin siblings there is additional interesting evidence for the non-shared environment.  Fraternal twins share the same gene similarly as do other siblings, having only the addition of a shared uterine environment.  Early reports were that they correlated on personality measures at about  .25.  This leaves 75% of their differences as a result of non-shared genes and the total shared and unshared environment. Turning to non-twin siblings with the same shared 25% of genetic material, the first reports for  correlations on personality measures reached an estimate of only .12%  This literally suggests that siblings within the same household are less alike than their shared genes alone would predict.  Illogical as it sounds, it is as if their shared and unshared environment made them less alike.

These are truly astounding findings.  It is worth repeating this finding in the language used over a number of  years, to see the durability of this idea.  In an early study (1981) the authors noted that “whatever environmental factors influence personality–and they are substantial, accounting for half of the variance of personality–they are not shared by siblings….these results suggest that nearly all of the environmental variance in personality operates in such a way as to make siblings in a family different from one another.”**

In an article titled “Why are children in the same family so different? Nonshared environment a decade later”,  The authors emphasized “Despite the difficulties encountered in identifying specific sources of nonshared environment, the fact remains that most environmental variance affecting the development of psychological dimensions and psychiatric disorders is not shared by children growing up in the same family.”***

Finally, a 2011 paper, once again asking why children in the same family are so different, did report that a study of depression in adolescents found that negativity on the part of the mother (corporal punishment, physical abuse, parental criticism and parent-child conflict) did have a non-shared effect on the child’s depressive symptoms.  However there are two caveats here.  First, the effect was estimated to be less than 5% of the total non-shared environment, and second, it was non-shared in the sense that it was specific to the parent-child pair and not other members of the family.  The authors conclude that  “…the basic finding of the 1987 paper remains unchallenged: children growing up in the same family are very different.  It is rare in a field as complex as the behavioral sciences to discover such clear and consistent evidence for a finding that radically alters the way we think about an issue as basic as how the environment influences development.  It was reasonable to assume that the key influences on children’s development are those that are shared by children growing up in the same family:  their parents’ personality and family experiences, the quality of their parents’ marital relationship, their parents’ educational background and socioeconomic status, the neighborhood in which they are raised and their parents’ attitude to school or to discipline.”) I emphasize their conclusion here: “Yet to the extent that these influences are shared environmentally, they cannot account for individual differences in children’s development because the salient environmental influences are non-shared….the key question largely remains unanswered: why are children in the same family so different?”

This really is a rather stunning finding.  It is widely accepted by behavioral genetics scientists, accepted  less completely by developmental psychologists, and largely ignored by the public. Consider the implications.

If nearly all effects are non-shared within families, then that all those parenting factors that have been considered, tested, shown to be significant for child development in literally thousands of studies in child development research, just aren’t moving the dial when we look at these carefully crafted twin studies.  Permissive parenting? Bed sharing? Authoritative discipline?  Having high expectations?  Having low expectations?  Being especially warm and responsive?  Being somewhat cool and distant?  Providing good schools and good neighborhoods for your children.? Living, as best you can in marginal neighborhoods? Being (a bad thing) helicopter parents?  Being careful not to be helicopter parents?  The list is endless and now this research suggests it really doesn’t matter much!

In addition, though I have not seen it discussed, there are equally serious implications for all sorts of public policy issues.  Providing free access to preschools, reducing  teacher-pupil ratios,  using busing to promote integration and diversity, anti-poverty programs to change young children’s lives.  These are all community-based rather than family based, but you would think that the same principles would apply.

Then what is nonshared environment/.  How does it do its magic?

Unfortunately, researchers would like to find a good answer to this as much as we would.  There are quite a number of candidates that have been suggested over the years, but all have fallen short in some degree–mostly in explaining only a very small part of the supposed effects of this.  Recall  that, for identical twins (and much more so for less closely related children) genes account for about half of the trait resemblance between them.  This leaves total environment responsible for the rest, and researchers assure us that nearly all of this is non-shared environment.  So, we have to look for forces, events, etc. that are not shared between a pair of such twins and therefore act to make them different from one another.

Some obvious events would be a debilitating illness or a traumatic accident that affected the course of development for one twin only.  That could certainly be life changing.  However, it has been noted of all the various events studied that “the effects of these variables are generally limited, either because only a small number of people encounter the variable or because its effects are weak.”**

Another source of unique events is actually related to genetics.  That is a fairly well established fact that children tend to seek out environmental experiences differently, depending on their own genetically based traits.  The physically active and moderately aggressive child may look for and join all sorts of activities and sports that fit these traits, while a quiet and perhaps introverted sibling may actively avoid them.  In this way, they help to shape the environment that in turn shapes them.

The idea has also been entertained that relationships to one or more siblings form  a unique source of difference for each sibling.  This seems logical but scattered research does not seem to suggest this as a major non-shared factor.  Taking it further it has been suggested that perception might be more important than fact.  If brother x perceived brother y as terrifying and oppressive, this might be more important than brother Y’s actual qualities.  Interesting, but no strong support for this as yet.

And so, the author of the paper Commentary:  Why are children in the same family so different?  Non-shared environment three decades later concluded that “…my main conclusion has to be that the key question remains largely unanswered:  why are children in the same family so different?”


*Bouchard, T., Lykken, David et al. (1990).  Sources of Human Psychological Differences:  The Minnesota Study of Twins Reared Apart. Science, 250 (4978) 223-228.

**Rowe, D. & Plomin, R. (1981).  The Importance of Nonshared (E¹) Environmental Influences in Behavioral Development.  Developmental Psychology 17(5) 517-531.

***Plomin, R., Asbury, K., and Dunn, J. (2001). Why are Children in the Same Family So Different?  Nonshared Environment a Decade Later.  Canadian Journal of Psychiatry, 46(3) 225-233.

*****Plomin, R. (2011) .  Commentary:  Why are Children in the Same Family so Different?  Nonshared environment three decades later.  International Journal of Epidemiology, 40, 582-592.



This entry was posted in And the Family, Anything/Everything-Temperament and tagged , , , . Bookmark the permalink.

One Response to Temperament, parenting, and the so-called “non-shared environment” PART I

  1. Mickey says:

    I do enjoy reading your blog and the semmtioes insightful comments, but at the end of the day, I could never afford to quit my law firm job, not because I’m scared to explore, but mostly because I’m scared financially to be unemployed. And as you have recognized in previous postings, you know how lucky you are to have a husband who can financially support you through your journey. I’ve been beaten and battered by this career I worked for the state government where I was overworked and underpaid and the cases were far too personal and emotional. I worked for biglaw and was disgusted by what my life had become a servant to the partners and arrogant clients. I worked for a small litigation boutique that laid me off after personally making the partners close to $500,000 on just one case during a 6-month period. So in lieu of quitting, I decided to give law another go just one more change for a fulfilling career in law (to Jack above why only chase fulfillment over career? You can have a fulfilling career.). I was able to find a satisfying law firm job that does not chain me to my desk or my blackberry. They do exist, even in NYC. They are hard to find and even harder now, in this economy. For those who are questioning their careers in law, there are other options you just need to be able to step down from the big $250K salary and so-called prestige of working in biglaw (which is really just prestigious among lawyers . regular every day people do not know the difference between Cravath and Sullivan Cromwell, I promise you that). I make a solid 6-figure salary (one which pays the bills, allows me to travel to foreign places a couple of times a year, buy nice clothes, eat at nice restaurants, and even save a little bit for a rainy day). I work on interesting cases for interesting corporate and individual clients. No, the cases are not as high profile as biglaw, but that’s okay because my name is on every single piece of paper that goes out the door, the clients know who I am, the only red light going off on my blackberry on most weekends is calendar entry reminders for my own personal events, and I get to go home on average before 8 pm. I might not become partner, but I don’t know if I even want to become partner or I would be content being of counsel, especially when I decide to have a family. I have a life and I have a career.

Leave a Reply

Your email address will not be published. Required fields are marked *