You can’t decide what’s normal and what’s abnormal until you understand the ordinary function of any trait—whether it’s vomiting or cough or fever or nausea. You start with its normal function and in what situation it gives selected advantages. But there are a lot of places where natural selection has shaped mechanisms that express these defenses when they’re not needed, and very often that emotional response is painful and unnecessary in that instance. Then there’s a category of emotions that make us feel bad but benefit our genes. A lot of sexual longings [extramarital affairs or unrequited love], for instance, don’t do us any good at all, but they might potentially benefit our genes in the long run.

So it’s not saying that these emotions are useful all the time. It’s the capacity for these emotions that is useful. And the regulation systems [that control emotion] were shaped by natural selection—so sometimes they’re useful for us, sometimes they’re useful for our genes, sometimes it’s false alarms in the system and sometimes the brain is just broken. We shouldn’t try to make any global generalizations, we should examine every patient individually and try to understand what’s going on.