5 ms·

The article mentions that much of this research is based on twin studies. I have a question about how these studies work - perhaps someone here has done this so

The article mentions that much of this research is based on twin studies. I have a question about how these studies work - perhaps someone here has done this sort of research and can answer. Here's my understanding of twin studies: you have some identical and fraternal twins. Each pair of twins are raised in the same household, and presumably experience the same environmental "nurture" effects. So, if we see stronger intellectual similarities between identical twins than fraternal twins, then we conclude that this difference is genetic. But how valid is the assumption that the environments that fraternal twins face are as similar as those of identical twins? Let's assume for instance that: 1) women and men have the same innate potential for mathematical skill 2) our education system does a better job fostering mathematical ability in men than women 3) over time, this discrepancy causes women to score lower on tests of mathematical skill than men While our identical twin pairs will all be of the same gender, the fraternal twin pairs may not. Two identical twin boys might receive the same environmental effects on their mathematical ability, but a girl and her twin brother will not, even if they have equal mathematical potential. Given these assumptions, if we conducted a twin study on mathematical ability, we would see a stronger correlation between the scores of identical twins than those of fraternal twins, due to gender difference, and we would conclude that mathematical ability has a genetic component. And our conclusion wouldn't be wrong! If gender impacts mathematical ability, and gender is genetic, then mathematical ability does arguably have a genetic component. But this gender difference comes about not because of an innate intellectual difference between women and men, but because of how a child's environment "nurture" changes depending on an arbitrary genetic trait "nature". In a world where these assumptions are true, a child's genes influence their future mathematical abilities - but in another world without gender-based educational discrepancies, the same genetic difference might not have any effect. How do twin studies account for these more nuanced situations, where society causes nature and nurture to coincide in roundabout ways? Does it make sense to simply call these "second-order" socio-genetic effects "nature"?

HALtheWise 8y ago
The best twin studies look at pairs of identical twins who were separated at birth and raised in different households. If they still end up more similar to each other than to their adopted siblings on some trait, we conclude that trait has a significant genetic component, and more sophisticated analysis can measure how significant it is. Note that the technique I described is not strictly able to determine whether the effect is from genetics or some other pre-birth shared environment, like maternal alcohol consumption. For many practical purposes, that distinction doesn't really matter.

HALtheWise 8y ago

The best twin studies look at pairs of identical twins who were separated at birth and raised in different households. If they still end up more similar to each other than to their adopted siblings on some trait, we conclude that trait has a significant genetic component, and more sophisticated analysis can measure how significant it is. Note that the technique I described is not strictly able to determine whether the effect is from genetics or some other pre-birth shared environment, like maternal alcohol consumption. For many practical purposes, that distinction doesn't really matter.