4 ms·
My argument isn't about if it's advantageous or not. It's that the mechanism (DNA, a long string) may not be well-suited to expressing and adjusting very compl
by rwmj 2y ago
My argument isn't about if it's advantageous or not. It's that the mechanism (DNA, a long string) may not be well-suited to expressing and adjusting very complex emergent behaviours in the phenotype. There may either not be a way at all to express "show strong maternal instincts but reject all non-related children and surrogate children like pets", or there are too many steps to get there and natural selection would never find it in any reasonable evolutionary timescale.
- pessimizer 2y ago> My argument isn't about if it's advantageous or not. I don't know if you're being fair to yourself. It's probably advantageous to have simple heuristics for determining if you should care for something, if that something will turn out to be your offspring 98% of the time. What possible pressure would there be to push that to 100%? It would be possible for DNA to do better with some sort of signalling. Select for kids that inherit a smell, for example. But to develop that you'd have to start with a bunch of false negatives, and a bunch of deceptive smells would develop to start giving you false positives again.
- zmgsabst 2y agoMy point is that it may just be retained because it’s a positive and hence selected for, not merely because it’s a local optimum.