A recent series of studies to be published in Psychological Science (Boothby, Cooney, Sandstrom & Clark) shows that following conversations with new people, participants believe that they like the other person more than the other person likes them. Logically this cannot be true. The bias, which we call the liking gap, contrasts with many other established biases which are self-serving. Why does this occur? We believe that, when meeting new people, our participants we were self-protective and a bit too self-focused as a result. They likely criticized their own performance (thinking, for instance, I should have said something else) but were less critical of their partners. Their self-focus also seems to have prevent them from picking up on signs that the partner did like them – signs that objective observers viewing the conversation did pick up.