Fair enough. But that’s not what happened in this case — the scientists were surprised by the results. They weren’t confirming what they already knew. Which suggests that either they’ve never had pets themselves or were extraordinarily unobservant.

I also think there are some cases where, when science apparently reports something contrary to common belief, the rational thing to do is reject the science. The so-called “Fundamental Attribution Error” research seems to me like a good example. There’s a limit to how far science can correct common sense, because at some point such “corrections” undermine the very intelligibility of other people to the point where if we bought in, we’d no longer have reason to trust any human endeavour, science included.

It’s unnerving to contemplate, but sometimes intimate familiarity with a topic just isn’t good enough. As a human living in human society, you’d expect to know all about what people are like–and yet people disagree all time about what people are like! To take just one example, there are those who say, “Men are like (this) and women are like (that) and this is innate from birth, as is obvious to anyone who’s raised children.” Others, including many parents, disagree vehemently. To find out what’s actually the case, we’re going to need something better than familiarity with human behavior—we need to design clever experiments at great expense, and do all sorts of complicated statistics, and have smart people study the problem for years—and even then, we still might be wrong.

Now, of course not all knowledge is scientific. Dog owners are perfectly justified in forming beliefs based on their experiences. But I’m also glad that some people take the time to investigate the obvious at length—just in case it turns out not to be true.

Human feelings of obviousness are just not that reliable. Better to do a rigorous study to confirm what’s right in front of our noses, than to get hosed when it turns out not to be.