The New York Times recently reported on a twin study into child food neophobia. This research, from my old Alma Mater, reports that 78% of the variance in food neophobia is inherited. Genetic. No-one’s fault, no-one’s choice.
“Interesting”, I thought to myself. So then I went and found the actual article, published in the American Journal of Clinical Nutrition. You can read it here if you want to (you may find the Discussion at the end quite interesting reading).
The study was based on a cohort of about five and a half thousand pairs of identical and non-identical twins, each member of whom completed a four-item measure of food neophobia (the Child Food Neophobia Scale). The correlations between CFNS scores for identical twin pairs and non-indentical pairs were then examined, with the result that these correlations were higher for the identical twin pairs, indicating a genetic influence on the neophobia trait.
Using structural equation modelling, it was then found that the model best fitting the data distinguishes between two sources of variation: Genetic, and Non-shared environmental factors/measurement error. Even if the estimate of the genetic influence at 78% is an over-estimate, it is clear that the genetic component accounts for a majority of the variation.
The remaining 22% of variance is attributed to a combination of non-shared environmental factors and measurement error, though in what proportions it isn’t clear. What is clear is that excluding shared environmental factors from the model does not result in a significant reduction of fit. From which it may be inferred that shared environmental factors have minimal influence in determining a child’s food neophobia, as they have elsewhere been shown to be less influential than non-shared environment in determining other traits.
Which leads me, as a non-expert, to idly wonder, what is the difference between shared and non-shared environmental factors (and which is why I pointed you towards the Discussion in the article, which explains exactly this). In a nutshell, it is the idea that shared environments can have non-shared effects. The same home, parents, and culture can be experienced differently by different children, and thus have different effects upon them. Whether this is because of the environment (eg parents) responding differently to their different offspring, or because of the child interpreting the environment differently is another question. I’d be amazed if it wasn’t a bit of both.
You may also be wondering how they separated out the effects of these undefined shared and non-shared environmental factors. Well, the latter component was estimated by subtracting the heritability estimate from the correlation for identical twins. If you subtract this and the heritability estimate from the total, you are left with the estimate for the non-shared component (plus error). Neat, huh?
So anyway, what we can take from this study is that while environment (including parenting) accounts for about a fifth of the variation in children’s food neophobia, it is genetic influence that explains the lion’s share. For children with food neophobia (which I think we all were), it is caused largely by our genes.
ps Of the many things that I am not an expert in, behaviour genetics is one of them. So the mechanism by which one’s genetic inheritance might give rise to food neophobia (as I have experienced it) is a mystery to me. But this study suggests to me that it might be at least partly psychosocial. To put that theory into better words, that would be to say that one’s genetic endowment leads one to have a psychosocial interaction with one’s environment, that in turn leads to the whole food neophobia thing. As it might also lead to obssessive-compulsive tendencies, for eg. Is this a post-hoc rationalisation of my experience? Or is there no such thing as free will? I don’t know. Any thoughts?