I spent a large part of my summer in Middle-Of-Nowhere, WY, at a paleontological dig. I was lucky enough to be able to stay at a decent trailer in town, near the dinosaur museum. The only major problem with it was a past history of mice. So, along with a cat, there were numerous mouse traps set up all over the house. They'd been baited with peanut butter, which seems perfectly natural. While I was discussing this with another volunteer, however, we both mentioned the idea of baiting the trap with cheese instead. After all, everyone knows mice like cheese, right?
In steps my critical thinking skills. It's "common knowledge" that mice are attracted to cheese. But, in the wild, there's no substance anything like cheese. Wild mice like seeds, nuts, and berries. Baby mice, like other mammals, would drink milk, but adult mice wouldn't have access to it. So, why would house mice eat cheese?
As it turns out, they don't. A study back in 2006, by Dr. David Holmes at Manchester Metropolitan University in England, showed that mice prefer foods with high carbohydrates, such as pasta, foods with high sugars, like chocolate, and grains and fruits that fit their natural diet. Cheese is too rich a food, tailored for human taste. Mice won't eat it unless given no other option.
Anecdotes confirm this. The volunteer I discussed this with had been baiting mouse traps for years. He'd often caught mice with peanut butter: the sugars and nutty taste are an ideal attractant to mice. He'd never once caught a mouse in a trap baited with cheese.
So, it turns out this "common knowledge" is yet another urban myth. It's been around for a very long time, but was certainly popularized by the cartoon Tom and Jerry. I'd thought nothing of the idea for years, but a bit of skepticism showed to me, once again, that you can't believe everything you see on TV.
Dr. Holmes discussed the study in a Scientific American podcast; you can listen to it or read the transcript here.