Where are my smart genes? Searching for intelligence in our DNA

Reproductive technologies such as in-vitro fertilization (IVF) and preimplantation genetic diagnosis (PGD) mean it is currently possible for parents to create a range of embryos and make decisions about which to implant on the basis of their genetic makeup.  One interesting possibility is that we will soon be able to use such technologies to influence the intelligence of our future children. It is known that intelligence has at least a moderately important genetic component. Identical twins are significantly more similar in intelligence than dizygotic twins, who are in turn significantly more similar than adopted siblings raised together. In fact, a range of studies indicate that the heritability of intelligence is approximately 0.7, which is only slightly lower than the heritability of height. This means that 70% of the variation we observe in intelligence is due to genetic factors. Once we can identify the genes which explain this variation it will be relatively simple to test embryos for them, meaning it will be technically possible for parents to select embryos on the basis of their likely intelligence.

However scientists are finding it surprisingly difficult to locate the specific genes which affect intelligence. A recent study concluded that most reported genetic associations with general intelligence are false positives. This means that the genetic basis for intelligence is still largely a mystery.

An interesting hypothesis concerning our failure to find intelligence genes (proposed by Dr Kevin Mitchell here and discussed in the New York Times here) is that we are thinking about the issue in the wrong way.  Rather than thinking about ‘the genetics of intelligence’, we should be thinking about ‘the genetics of stupidity’. Current research methods are best suited at identifying specific ‘smart genes’ which occur in some but not all of the population. However it’s possible that what explains differences in intelligence is not the presence or absence of smart genes per se, but rather the accumulation of random mutations in the genome. This would mean that the genetic variants which influence intelligence will be different for different people. The mutations that affect Jack’s intelligence would be different to the mutations that affect Jill’s.

This hypothesis builds on the reasonable assumption that most of the genes which contribute to humans being more intelligent than other animals (those that affect brain development for instance) would have been fixed in the genome early in our evolutionary history.  Given the vast majority of mutations are deleterious it is likely that any mutations in these specific genes would be selected against. This would mean that we would expect little difference between individuals in relation to the genes that build intelligence: we all share the same basic blueprint for an intelligent brain. Where we differ, and what explains observed differences in intelligence, is how well our bodies can follow this blueprint. The most intelligent individuals will be the ones whose bodies most accurately follow and reproduce their genetic code. Innate differences in intelligence are not explained by differences in design but by differences in construction.

If this view of intelligence is correct, then as Dr Mitchell observes, we need to start looking at different traits to explain the heritability of intelligence. One trait that we inherit and which could affect intelligence in the above manner is developmental stability. Developmental stability describes the accuracy with which our genetic code is read. The idea that intelligence is influenced by developmental stability is supported by observed correlations between intelligence and a diverse range of physical traits including symmetry, fecundity, and longevity.

If developmental stability is a major determinate of intelligence this would have interesting implications for the debate regarding the screening of embryos on the basis of their expected intelligence. Assuming we can identify genetic variants which influence developmental stability, this would be one manner in which we could select for intelligence using IVF and PGD. Selecting for intelligence in this way may be less controversial than selecting for intelligence directly. It is sometimes claimed that it is permissible to use reproductive technologies like PGD in order to avoid disease, but not to select for positive traits like intelligence. However if what explains differences in intelligence is differences in developmental stability – the ability to read and replicate DNA – then genes which influence intelligence may be more likely to be looked at in the same way as disease-related genes.

Further, genes which affect developmental stability would also affect a range of other important traits. Increasing developmental stability would not only make prospective children more intelligent but also healthier, longer lived and more attractive.  Selecting all of these traits in embryos simultaneously will likely be seen as more desirable, and sometimes more morally permissible, than selecting embryos solely on the basis of intelligence. A future where parents undergoing IVF can select embryos that are likely to be the most intelligent could therefore be less controversial than is often thought.