Nature VS Nurture and Twin Studies


First, sorry for the first post, it’s gone. I’ve just learned something about ‘Link’ posts.


This isn’t the most comprehensive article about all the separated twin studies that have been done, but it’s new, and typical enough. Here’s an excerpt, a link to the article and the author’s data, then my comments:


“To explain this close bond, psychologists frequently look to environmental factors: Identical twins, growing up in lock step, share most of the same experiences (they are often even treated the same by others) and therefore develop their unusually strong connection. But my research on reunited twins challenges this theory. I have found, for example, that reunited identical twins report feeling greater closeness to each other than do reunited fraternal twins. And I have found that a reunited twin generally reports feeling closer to the twin he only recently met than to a genetically unrelated sibling with whom he was raised.

Given that reunited twins were not reared in the same environment, genetic factors are surely relevant to these bonds. But how do shared genes result in an immediate sense of connection?”


Nancy L. Segal, a professor of developmental psychology at California State University, Fullerton, is the author of “Born Together — Reared Apart: The Landmark Minnesota Twin Study.”


How different are the environments really? Don’t we all raise our kids the same way, with the same methods? And even in the case of these ladies, isn’t American culture and therefore American child-rearing simply a continiuation of the British version? Aren’t the apparent differences from household to household superficial? For this, I offer the relative ubiquity of parenting books and discussions where the basic premises are nearly universal while only the details of how and when parental control is exercised differ.

Which makes me wonder that we are not looking at the correlation of genes in interaction with different environments, but genes in correlation to environments that are actually also very similar. Again, within a given culture, how different are the environments provided for children? This idea would have to extend to the sorts of personality tests that the twins do so similarly at: are the described personality traits and types not also visible only in contrast with a fairly uniform background? If so, the power of our genes, and so of the Nature over Nurture argument, is going to win every time, considering that there really be very little variation in the environment children are raised in anyway.

18 thoughts on “Nature VS Nurture and Twin Studies

  1. Scarlet May 26, 2014 / 12:54 am

    I don’t imagine that the cultures are too different, there are all kinds of variables though, what are the chances of two average upbringings? An average upbringing sounds ideal to me.


      • Scarlet May 26, 2014 / 3:57 pm

        They should have sent one to Sudan 😀


        • neighsayer May 26, 2014 / 6:27 pm

          Exactly what I would need to see before I give up on my Nurture over Nature stance.


          • Scarlet May 26, 2014 / 6:35 pm

            Now all we need is someone to fund the experiment – Bayer? 😀


    • neighsayer May 26, 2014 / 7:03 am

      I figure I’m probably wrong about the larger argument, but I just wonder if that part of it has been a poor assumption, that all families’ differences, in terms of child-rearing are meaningful ones. Seems to me that the adults in my world put up a single, pretty uniform front.


    • neighsayer May 26, 2014 / 6:53 pm

      Yeah, I just can’t swallow Nature over Nurture no matter what the evidence! I just think it warrants extreme suspicion for so many reasons. It’s a denial of our responsibility for choices we make, and it always brings in the spectre of eugenics . . . and as I said, when you get to where I am on the subject of parenting and the scourge of punishing, all parenting environments look the same. Plus, what Scarlet said above: there must be twins separated and raised in very different environments, such as extreme punishing ones (fundamentalist communities?) and non-punishing ones (Buddhist communities, or some aboriginal communities?) as well, of course as other, large cultural differences. Urban VS hunter/gatherers . . . like that. Then, if they still grow up to both be car salesmen, then maybe I’ll believe.



  2. Ad Nausica May 29, 2014 / 7:03 pm

    (Following up from twitter) I see a couple of misconceptions here. First, the similarity of environment is irrelevant to the conclusions. Let’s look at a simple version of what twin studies do.

    Start with four categories: (1) identical (monozygotic, MZ) twins raised together in the same household, (2) MZ twins raised seperately, (3) unrelated individuals of the same age, reared together from babies (adopted siblings), (4) unrelated individuals of the same age, reared apart (strangers). Note that each category involves 2 people (siblings, strangers), call them A and B, so there are 800 people total: 2 people per sample x 100 samples per category x 4 categories. For simplicity let’s also make all pairs the same sex as each other (a must for MZ twins, of course).

    Now suppose you have test to measure some aspect of these people. It could be how many fingers they have, how tall they are, personality features, behaviours, habits, or really anything you’d like. Let’s take narcissism as an example. Suppose the test ranks people’s narcissism on a scale of 1 to 10. So you have all 800 people take the narcissism test.

    Now you look at the MZ twins raised together and compare Twin A’s results with Twin B (their sibling), and you do this for all 100 sets of twins in Category 1. You notice that whatever Twin A scores, Twin B gets the same thing exactly. If Twin A scores a 7, so does Twin B. If A scores 2, so does B. Wow, highly correlated and predictive.

    Then you look at Category 2, MZ twins raised apart. They have the same result: perfect correlation. In Category 3 (adopted siblings) you find no correlation at all. If Sibling A gets a 7, Sibling B might have a 2 or a 10, or anything. There is no pattern. In Category 4 you find the same thing, no correlation.

    What can you conclude? The data is very clear: having identical genes with somebody is a perfect predictor for the amount of narcissism they have, regardless of being reared together or apart, and being reared together or apart has no effect on the amount of correlation. This would be a perfect example of a purely (100%) genetic feature.

    Now suppose the results had been different; suppose you had the same results for Category 1 (MZ twins reared together) with highly correlated results. But Category 2 (MZ twins apart) were perfectly uncorrelated, Category 3 (adoptees) were highly correlated, and Category 4 (strangers) were uncorrelated. Now genes show no predictive power, but being raised together is highly predictive. This indicates it is something about being raised together that causes whatever score they get.

    Note that the similarity or difference of the environments any of the samples are raised in has no effect on the conclusion in either scenario. They could all be from very similar suburban neighbourhoods, or “reared aparts” could be from the jungle versus wealthy mansions. The conclusions aren’t based on assumptions of the differences of environmental style, only the correlations of the 4 categories.

    The way baseline environment would affect the outcome is the same way that baseline genes affect the outcome. If you were to measure, say, number of fingers each had, and all 800 people had 10 each, there’s no variation to correlate; they all have the same baseline genes (that make them human). It’s gene variation across the population of humans that is of interest, not the 98% of genes that all humans have in common that make us human. If all 800 are from suburban environments, that chances nothing above. But, you might do more tests where some are from urban environments and some from jungle tribes, rather than just “reared apart”. Harder to do, but now you have another variation to correlate. But it answers a different question.

    If you live in a suburban environment and want to raise your children to not be narcissistic, the results of the first scenario (100% genetic) show that the range of parenting styles used in suburban environments make no difference, you are wasting your time worrying about which method to use. (You might look at therapy, pharmaceuticals, or other treatments instead, for instance. Yes, genes don’t mean what you seem to think.) If we got the results of the second scenario, then we might look at what was different in the households of the low versus high scorers and could recommend doing the parenting things that give lower scores.

    Twin studies get a little more complicated too. They include fraternal (dizygotic, DZ) twins as well, and sometimes general siblings. DZ twins are valuable because they have 50% identical genes on average. That doesn’t mean 50% are completely different, since humans have ~98% same genes that make us human. What it means is both your father and mother each have two alleles of each gene (23 pairs in total), and each egg or sperm gets one or the other from each, so for a given gene there are 4 possible sources: father’s A vs B, mother’s C vs D, So you could be AC, AD, BC, or BD at each of ~24,000 genes. MZ twins will have the same combination at each gene, DZ twins will have half the same (50%) on average at each gene. (That’s 4 to the power of 24,000 different possible children for a couple can have, which is astronomically high.) The 50% means of all the variations available to humans to inherit, DZ twins and sequential siblings have half the same on average. So you’d expect to see about half the correlation in the scenarios described above. This is also true of sequential siblings, but they are different ages so DZ twins make a nice way to remove effects of differences in time, experience of parents of one child vs the other, age of parents when raising them, etc. DZ twins also adds two more categories (reared together vs apart, 5 and 6). They strengthen the conclusions.

    The results are also never as perfect as here. They usually result in things like IQ is 70% genetic, 0% household, and 30% uncorrelated. Notice the third category. Uncorrelated means there is a measured variation in score that doesn’t correspond to any categorization. That 30% could be random events in everyone’s lives, such as the peer group you fell into at school, random neuron development in utero, and so forth. Note that some of those would be environmental and some would be biological. The 0% household means that being reared together vs apart has no effect on IQ, for instance, so parenting style or home life can’t make your kid smarter or dumber.

    As far as what “genetic” means, I’ll start a different comment.


    • neighsayer May 29, 2014 / 8:48 pm

      I’m not trying to argue, but I do want to take advantage, as long as you’re still willing to explain this to me. Having said that, I need my same point clarified: what is “reared apart” presupposed to indicate then? If we’re talking about twins, is the point to compare twins who have each other around with twins who don’t? Or is it, as I assumed, that “reared apart” is assumed to mean something like “in a different environment?” Surely there is supposed to be some meaning to the comparison, “reared together” or not. I mean, if “MZ twin” VS other sorts of people means a genetic difference, what does the “reared together” or not mean? I’m not arguing with the data, I’m just worried about the premise of the studies.
      I think you know that’s where I’m stuck, but while your response is good and informative, it doesn’t really answer that point for me.


      • Ad Nausica June 3, 2014 / 7:10 am

        Great question. You’ve more or less described “reared together” correctly, as “having each other around”. “Reared apart” means that the twins were separated at birth and raised by/in separate families/parents and weren’t in contact with each other (or knew about each other). Reared together means they were brought up by the same parents in the same household. In principle there could be another category where twins were raised in separate households/parents but knew each other and interacted, but I’m not aware of any study that includes them, probably because it would be very rare.

        I think the general problem people have with the results, or the principles discussed in general, is assumptions about what the results mean. It’s far to simple to just think of things in terms of “nature” and “nurture”, or genetics vs environment. For example, if a particular characteristic is measured as 40% genetic, 5% from being raised together, and 55% not correlated, this means that it is unlikely variations in parenting style affect that characteristic. One of many possible explanations of this scenario is that to have this characteristic, one needs a particular gene variation (allele) which makes that characteristic possible, but it is only “activated” by being part of a peer group that embraces that sort of characteristic.

        Let’s take a preponderance for vandalism, for instance, and suppose it had the above results. If you didn’t have a gene for it, if you joined up with a peer group that pressured each other to vandalize property, you’d probably feel guilty and hate doing it, and resist or eventually leave the group (or avoid joining it in the first place). You’d tend not to be a vandal. If you did have a gene variation related to vandalism but were never part of a peer group that endorsed it or were part of one that opposed it, you’d likely never do it. Of course there’d be no “gene for vandalism”, but it might look something like a gene, or combination of genes, that removes inhibitions, perhaps gives a dopamine pleasure from being rebellious and breaking rules, or perhaps a pleasure from doing whatever your “in-group” does.

        This last case is an interesting one. Humans in general have in-group/out-group tribal tendencies (“us” vs “them”) for fairly clear evolutionary reasons regarding maximizing survival and reproduction with limited resources available in a region. (You can see it easily in chimpanzees.) If somebody has a genetic tendency that overemphasizes this factor (do what your peer group says) at their own expense, you’d expect to find lots of results like those above: the genetic basis for strong in-group compliance and the behaviour tending towards whatever peer group you happen to fall into, which might be a function of the first kid you sat beside in grade school. That is, more or less random. As far as these twin studies, that random effect would show up in the uncorrelated results (55% in above example).

        Contrary to this, some genetic tendencies might lead certain individuals towards certain peer groups, such as the “brainiacs”, athletic jocks, “pretty girls” cliques, etc. And none of these have anything to do with parenting, which may have little to no effect on any of it. (Of course you can try to guide kids into or out of certain peer groups, but that can certainly be quite difficult, particularly if parents don’t have the “inside scoop” on a particular clique and/or strong bonds are formed prior to attempted intervention.)

        So it can get quite complicated. “Nature vs nurture” is a poor description then. In fact, “nurture” usually implies parenting as we don’t consider peer groups as nurture even though they are a significant part of a person’s environment. “Genes vs environment” is better, but still misses the interaction. More accurately it is the combinations of both, as the results tend to show. If you don’t have genes that tend towards a certain behaviour, no environment will change that. If you do, then environment will have a big effect, but the particular component of environment may matter significantly (parenting, peer group, general culture or sub-culture, even food choices).

        To be clear, most of this comment is a set of speculated examples. Twin studies generally just result in the estimates of proportions from these categories (40%, 5%, 55%), and not how the genes do their part (combinations, lower inhibitions, raise “in-group compliance” pleasure, direct pleasure from the behaviour, etc.) or which parts of the uncorrelated component are biological or environmental, or which components of environment are important or not. There is still a lot of detail to understand, but it is clear we are not simply blank slates that nurture/parenting, culture, or promotion of ideas can write whatever we like to get the behaviours we want.


        • neighsayer June 3, 2014 / 8:36 pm

          OK, Ad, that was all great stuff regarding genetics and the studies that help us learn about genetics, but you know it’s the other side of the equation that interests me, and sorry to say, I’m still not hearing an answer to my concern. I’ll spell it out as clearly as I can, after a few comments.
          I understand that Nature VS Nurture is a debate of the previous century, that environment and genetics are in interaction;
          I think these twin studies are very useful in showing genetic commonalities, but they also necessarily show the other side of the coin as well, (as well as you say, the “other” things, the various aspects of life that intrude and belong neither to our genes or our family dynamics) the amount of commonality that can be attributed to family and environment.
          Now, again, the question, and I realize I may only be asking for an opinion:

          Do you think the variable “raised together/raised apart” is assumed to indicate “raised in different environments?” Is that not part of the background theory of these studies?

          And, in case I’ve forgotten, thank you for your patience.


        • neighsayer June 3, 2014 / 8:42 pm

          Taking another tack here – considering the interactive nature of the relationship between our genes and our environments, wouldn’t it be reasonable to say that the impressive correlations that the twins show in these studies is a strong indication of my whole thesis in this discussion? That the strong similarities with these separated twins must indicate that both the genes and the environments are largely the same?


  3. Ad Nausica May 29, 2014 / 7:36 pm

    OK, so as far as the meaning of genetic, environment, nature, nurture, random, a lot of things get misunderstood. For example, I see the comment: “I just can’t swallow Nature over Nurture no matter what the evidence! … It’s a denial of our responsibility for choices we make, and it always brings in the spectre of eugenics”

    Well, no, not at all. And the only way in which that could be interpreted as true wouldn’t change anything about holding people responsible (except maybe doing a better job of it).

    Just because something is genetic doesn’t mean we are a slave to it. Nobody questions that our craving for sweets and high fat foods it genetic (nature). Indeed it makes a lot of sense why that is genetic, having evolved in an environment where calories were scarce and/or expensive to get, and rare sweets and high fats would be a huge boost in calories. We should expect genes that gave people such cravings to spread fast as it helped the people who have them to survive and reproduce better than those who didn’t have such cravings.

    But that isn’t good in a world of abundance today. So do we not have any choice but eat sweets and high fat food? If I put a cookie in front of you does that mean you are powerless to refuse it, or plan your diet, or cheat it by using artificial sweeteners? Of course not. The tendency (urge) might be genetic, but genes give us multiple competing systems in our bodies. One can overrule another. Our cognitive brain is one of them. We can see the cost of eating the sweet, weigh the options, and say no to it.

    Likewise you can alter your behavioural tendencies. You might genetically have a short temper, but be aware of it, recognize it, learn to step away, and count to 10 to cool down. Your genes set up your biochemical processes and hormones, but you can augment that with pharmaceuticals. Pain is nature, and pain reaction is genetic. But Tylenol changes it.

    As far as, say, criminal punishment, genetics don’t change the value of punishment per se. If a rapist’s libido and lack of empathy are genetic, it’s not like we have to let them run amok in society, with no choice but to let them go. We lock them up for our safety. We also punish them so their cognitive brain learns costs. Just like the cookie, having the craving or urge doesn’t mean you have to acquiesce; knowing indulging them has costs (jail, in this case) can change your actions. We also might not let them go so easily, knowing they still have the same genetic tendencies, if we’ve done nothing about it. Which is where we might even do better. Knowing they rape because of their genetic brain and biological structures (overdeveloped libido, underdeveloped empathy), we can treat those problems directly. Therapy, pharmaceuticals, gene therapy, and so forth can actually correct the problem and thereby reduce the likelihood of re-offending if/when they are released. But you can ONLY do that if you understand the cause, and in this example if that cause is genetic.

    Put another way, saying “my genes made me do it” is no less or more of a problem than saying “my environment/upbringing made me do it”. Associating it with general environment (society) or nurture (parenting) isn’t any better or worse, just different. Understanding it is not the same as excusing it, or doing nothing.

    Eugenics is also a non sequitur. Eugenics is simply a human breeding program, that same as we’ve done with dogs, farm animals, crops, or any other domesticated being throughout history. There’s nothing about genetic causality that justifies eugenics. Genetic tendencies describe an “is”, not an “ought”. Whether or not genes are significant components of behaviours has no effect on that. Nobody questions that genes affect height, but that doesn’t justify selectively breeding only tall people and sterilizing short ones so they don’t reproduce. You have to justify those sorts of actions on their own merits, and has nothing to do with it being genes or environment, nature or nurture.


    • neighsayer May 29, 2014 / 9:26 pm

      OK, first, I’m not quite s ignorant of social things as I may be about genetic studies. But I’ll pick a few points.
      1. It’s fine to talk about free will regarding eating sugar and all that, but the size of that industry and the size of North American people will show us that there’s a largish gap between free will and personal responsibility on one side and the power of a focused industry capitalizing on the genetic tendency on the other. It’s the difference between one person’s will and statistical realities.
      2. Regarding punishment, have a poke around in my blog here, you’ll see I don’t have any faith in that particular non-genetic way to change behaviour for the better. (Not the point of this discussion, I know.)
      3. This bit was good:
      ” Therapy, pharmaceuticals, gene therapy, and so forth can actually correct the problem and thereby reduce the likelihood of re-offending if/when they are released. But you can ONLY do that if you understand the cause, and in this example if that cause is genetic.”
      – Fair enough.
      4. When have any of the horrors that come out of eugenic thinking ever been supported by real science? That is not required for someone who wants to do that stuff. It is exactly phrases like “genetic causality” that these type love to use. I know, science must march on, truth above all. I wouldn’t spread any lies or suppress anything, but I think there is a tendency for people to deny environment, particularly in child-rearing. I hear the term “human nature” to explain things we do and things we cause way too much.


  4. Pamela Spiro Wagner October 6, 2014 / 6:08 pm

    This was an interesting discussion but I come from an experience that might interest you. I am an MZ twin who was raised in the same family as my twin but as far as I am concerned we were raised in different “environments” insofar as expectations and attitudes towards us were concerned. This changed the outcome tremendously…if there are “genes” or epigenetic factors that contributed to this, well and good, they did so, but all I can say is, the treatment, the nurturing or non-nurturing that we received from our parents and their messages to us verbal and non verbal, and physical as well in terms, yes, of punishment, were what changed us into what we became today, completely different people. We aren’t even on speaking terms, that is how different we are.


    • neighsayer October 6, 2014 / 7:36 pm

      wow, that’s wild. Ah, I was wrong, this isn’t the best ones on this subject. look later on, I think it’s “A Revolution in Nature VS Nurture parts #1 and #2 that I finally lay out the conclusion that got started here . . .


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s