The more you know about what sets a friend off, what makes him or crazy, the better the relationship with that person. That's according to a study described in this brief unsatisfying news story.
Psychological Science, but I can't find it. The two listed authors from the news story do not appear in the journal's online index of authors. Despite being labeled as "published" it's probably been accepted and is in press for later publication.
Back to what little we can tell from the short news item. People filled out a long list of behavioral "triggers" people might find annoying, such as being a perfectionist, and they answered questions about how much these trigger their friends. Not knowing what set your friend off, that can affect the quality of a friendship (measured, I don't know how, given I can't see the actual article). But this is kinda useful. For example, my son goes to college next year and I may insist he and his lucky future dorm roommate engage in just this exercise.
Okay, so you're not surprised by the findings. Still, it's a neat little approach to the study of friendship -- even if it smacks just a wee bit of researching the obvious. But to be fair, no doubt there's a lot more to the study itself than the brief news story provides, both in terms of theory and methodology, so I'm more than willing to give it the benefit of the doubt.