“
In short, distinctively human cognitive mechanisms are tracking targets that move too fast for genetic evolution. In a stable phase, “as similative alleles”—genes that reduce the experiencedependence of a cognitive gadget’s development—may increase in frequency. But when the environment shifts, there will be selection against assimilative alleles because their bearers will be slower to adjust to the new conditions (Chater et al., 2009). Once again, let’s take imitation as an example. As long as gestural markers of group membership, bonding rituals, and technologies remain constant, alleles that privilege and accelerate learning of particular matching vertical associations could be targets of positive selection. For example, people who more readily associate matching trunk movements (for example, you lean forward, I lean forward) than complementary trunk movements (you lean forward, I lean back), might have higher reproductive fitness than people who learn matching and complementary trunk movements at the same rate. But when conventions or technologies change, those assimilative alleles would hamper the development of imitation mechanisms with a now more effective repertoire of matching vertical associations. The people who had once been such effective social operators would now be losing social capital by leaning in when they should be leaning back. This kind of problem could be avoided if mutation produced a universal imitation mechanism, like the cognitive instinct postulated by Meltzof and Moore (1997), which could copy the topography of any body movement. However, this would be standard genetic evolution, not genetic assimilation, and, given that no one has worked out how such a mechanism could operate (Chapter 6), it is plausible that—like wheels (Dennett, 1984)—it lies outside the range of available genetic variation.
”
”