Welcome back. So, in the, the first module, I tried to give you some characteristics of general cognitive ability, sometimes called general intelligence. And, in this module, we're going to look at twin and adoption research on general cognitive ability. Most of these studies are based on IQ tests, but not all of them that we'll talk about here. There would be something that's very similar to, to an IQ test. And there are hundreds of twin and adoption studies of general cognitive ability. In fact, this is actually kind of a fun module for me. Because it, it lets me go back to what was my first empirical publication when I was a graduate student. So oh, it's going to sound like ancient history to you and I suppose it is to some degree. But let me go back to a paper I published in 1981 with my adviser Tom Bouchard. Now, I wouldn't bother you with this old paper, except the data we published in 1981 are still being used today. And it's probably as good a data as we have on twin family and adoption similarity for general cognitive ability that, that, that's, that surrounds, so why not look at it. What we did back in 1981 is we, we summarize all existing behaviour with genetic research, on general cognitive ability, in terms of familial correlations, for general cognitive ability, for various classes of relatives. And at that time, we summarized was 100, I think 101 studies, over 500 correlations, familial or twin correlations being reported on samples over 100,000 individuals. So, back in 1981 we already had a very large database. Recall, I said, there's a lot of behavioural genetic research on general cognitive ability. There's been a lot of research since 1981. But I think in terms of the basic things I'm going to describe here today, really nothing has changed. So nobody has gone back and updated this summary that we published in 1981. So, what I'm reporting here is the average correlation and we call a correlation will vary between 0, no similarity to 1 perfect similarity. So the average correlations, for a quantitative measure of genocognibility, in different class of genetically related relatives. Monozygotic twins, dizygotic twins, and so on and so forth, and. And, see that it has this characteristic pattern that as we go from more genetically-related individuals to less genetically-related individuals, the average correlation declines. In this case, that we could call maybe the monozygotic twins, zero degree relatives, zero degree relatives actually correlated, extraordinarily high for their in, in their general cognitive ability. Almost 0.9 are more correlated than first degree relatives in turn, more correlated than second degree relatives who, in turn, are more correlated than third degree relatives. This pattern here, that how similar you are in genetic to a relative in genetic, in general cognitive ability. Is pro, is proportional to how genetically related you are to that individual. Implicates that there are genetic influences. But it'd be nice if there were another way of documenting this result as well. And there are other ways. One other way is to look at reared-apart twin studies. In this next table, illustrates reared-apart twin studies. Correlation for general cognitive ability in reared-apart monozygotic identical twins. So there's been five such studies published in the literature. The number of pairs is not extradordinarily large because these are hard to identify. This is the study we talked about a little bit earlier, and you've read a paper about this is the study called MISTRA that was done here at the University of Minnesota, it's the most recent of these studies. Unfortunately, there's a western bias to these studies, but what's remarkable is that the studies are very consistent in indicating that even when reared separately, genetically identical monozygotic twins have at least moderately dis, to maybe highly similar levels of general cognitive ability. The correlations there are about 0.7. And actually quite, quite comparable across the multiple studies. So, these two lines of evidence both implicate genetic influences on genetic general cognitive ability. That the degree of similarity, and these are all and reared together relatives, is proportional to their degree of genetic similarity. The degree of similarity in general cognitive ability proportional to genetic overlap. And secondly, still a rather strong correlation in monozygotic twins who had been reared apart. The study we did in 1981 also, though, implicates the importance of environmental factors. The first way it implicates the importance of environmental factors, I'll go back to this, is that the monozygotic twins here aren't, aren't perfectly similar. Right, and they're genetically identical, but their correlation is not one. And because they're genetically identical, that the difference, the deviation of one here, has to be due to a, an aspect of their environment, what we would call, right, the non sharing arm. But the, the IQ, the general cognitive ability correlations also implicate the shared environment. And we know that for two reasons. One is that reared together relatives for general cognitive ability are more similar than reared apart relatives. So here's the average general cognitive ability correlation for reared together monozygotic twins. This is the average when they're reared apart. There's a difference there. Full siblings when reared together correlate almost 0.5. When reared apart, they're less similar. The difference between these two is growing up in the same home here, but not there. In both cases, they're full genetic siblings. Similarly for parent offspring, the difference between the blue bars and the red bars here. It implicates the importance of that shared family environment. Because growing up together, you're more similar than growing up apart. Again, just like with the genetic influence, it'd be nice if there are other aspects of behavioural genetic research that also impliculted, implicated the shared environment and indeed there are. Here, one aspect of the research is the difference between reared-together and reared-apart relatives, both being genetically related. The other line of research that implicates the importance of the shared environment is genetically unrelated individuals who grow up together. Adopted siblings and adopted parent offspring. I'll just focus on adopted siblings, here. Adopted siblings are individuals who are not genetically related to one another, who grow up in the same home. We've talked about this earlier in this course. There's no genetic reason for them to be similar on any behavioural trait. But yet, their correlation for general cognitive ability is on the order of 0.3 to 0.35. So, it's not as high as the twin correlations, but it's certainly not a trivial correlation. The only reason for that correlation, is that there must be something that they're sharing in that common home that's affecting their level of genetic, of general cognitive ability. So, looking at the correlations from that 1981 review. We can observe patterns that are consistent with their being some genetic influence. And we can also observe multiple patterns that are consistent with the importance of the environmental influences. If we recall from week three, what biometrics tries to do is actually quantify these impressions. And people have applied biometric models to these familial and twin correlations, to try to come up with estimates of the underlying genetic and environmental contributors to individual differences in general cognitive ability. Those biometric models really being kind of analogous to the Faulkner or ACE model that we talked about in week two. But here, they're a little bit more complicated because we're not just talking about reared-together twins. We have the other relatives as well. We don't need to worry about the, the statistical complexities here. What we should look at are the results of applying these models to try to estimate how important genetics and the environment appears to be in accounting for these familial resemblance and twin resemblances. There are actually three different research groups that have tried to fit biometric models. To the twin family and adoption correlations that we published back in 1981. And these are the three groups here and if you're interested here's the citations. It's actually, their results are very very consistent. First of all, genetic factors or the heritability appear to account for about 50% of the variants. Variants being a measure of individual differences for a quantitative trait. So 50% of individual differences appears to be associated with genetic factors. Again, that's an approximation, could be 40 %, sure. Could be 60%? It may also be possible. But, it appears to be that genetic factors are an important contributor here. The shared environment, that is, the impact of growing up in the same home also appears to be important, depending upon the study here contributing anywhere from 20 to 40% of the variance. Third, the non-shared environment. Recall that the non-shared environment corresponds to the effect of factors that individuals growing up in the same home don't share. And because they don't share, they can contribute to their differences on the phenotype. In all of these studies, the estimate of the non-shared environment is 14%. So that 14% of the variance appears to be associated with non-shared environmental factors. You may recall also that I mentioned in week three, that the non-shared environmental component also incorporates into it the effects of measurement error. If you measure a trait on one day and you come back a week later and give the same, for example, general cognitive ability test, the correlation between the two won't be perfect, right? People, they, they may not have had enough sleep the night before, so maybe they don't perform as well. So, if you take the same person and measure them over, let's say, a two week span. The two tests might correlate 0.9, 0.92 or something, which means 8-10% of the variance on the test is associated with measurement error, kind of instability of performance. A good chunk of this non-shared environment for general cognitive ability, is probably measurement error. So, and this is actually something that's not true of many psychological traits. For general cognitive ability, the predominant source, the behavioural genetic literature, suggest that the predominant source of environmental influence is due to the common family environment. And, that's something we'll come back to. Finally, the only difference in these three studies is really the importance accorded, the prenatal environment. In this study, they believe that the prenatal environment was very important, it counted for 20% of the, the variance. These studies actually didn't try to model that. We won't really say a lot about this. It, it's, it, it may be very important there's still not a lot of research on how important this might be, even though this paper was published a few years ago. So, twin and adoption studies have begun to tell us something about the behavioural genetics of general cognitive ability. Next time, we'll extend this by looking at what I'll call gene-environment interplay and looking at twin and adoption studies, of general cognitive ability with that in mind. [SOUND]. [BLANK_AUDIO].
