Some time ago I read a paper on fruit flies, in which the researchers bred a population of flies to avoid a particular fruit. They did this by exposing the flies to two different types of food (say, e.g., grapes and bananas), one of which was poisoned and killed all the flies that ate it.
I forget the exact details, but I seem to recall that after about twenty generations they had successfully bred fruit flies that instinctively avoided the "poisoned" fruit, even when the fruit wasn't actually laced with poison. (I think they alternated poisons every couple generations to ensure that they weren't breeding a distaste for the poison rather than the fruit).
I was wondering (1) if this sort of technique would work with mosquitoes and (2) whether such a change would be likely to persist. It sort of seems like the answer to both ought to be 'yes', but if so then why has no one done it? So I figure I must be missing something. Some thoughts:
1) I forget which fruits the study used, but I remember that they were pretty distinctive. Perhaps people are too similar to other prey animals for such discernment in mosquitoes? Then again, we're bipedal and have much less hair than most mammals. Probably smell different, too. Those would seem to be pretty obvious traits.
2) On one hand, it seems like having fewer food sources would be evolutionarily disadvantageous, suggesting that the trait would not persist if released into the broader mosquito population as a whole. On the other hand, I'm guessing that a much larger proportion of mosquitoes get swatted after biting people than after biting, say, a horse or a cow. Which should make the 'avoid humans' trait evolutionarily advantageous and tend to spread throughout the entire population.
2b) Mosquitoes breed by the thousand, so clearly they can accept a large proportion of failure-to-breed from a species perspective, so swatting isn't a strong deterrent. But neither should be failure-to-breed from starving be a strong deterrent, unless people are effectively the only prey species in the area. Whether you lose 10% from swatting or 10% from starving, it seems like it makes no significant difference from a species perspective.
Anyway, I'm trying to figure out why this (apparently, or else someone would have done it already) doesn't work. I figure I'm probably missing some key detail about mosquitoes or evolution that will solve the mystery, but I'm not sure what.
Edit: If it's not clear from the above, I read the paper on fruit flies quite some time ago (probably a decade or more, and no guarantee that it was new research then). So some of my details could well be off. If there's any reason to suppose that my memory is so excessively faulty as to make such a scenario implausible or otherwise questionable, please let me know. I'm fairly sure I have the gist of the paper correct, which for the purposes of this question is that at least one type of insect can be bred over the course of a relatively small number of generations to strongly prefer one type of food over another, but not so sure that I'd stick to my guns if a subject-matter expert called it into serious question.