Back when I TAed physics, I used to tell the students that a huge chunk of physics was simply having the gall to believe that you could get the answer. In other words, “confidence is key!” (Of course this probably also leads to the well known problem of extralusionary intelligence)
In this spirit, here is an article in the Washington Post about gender stereotypes and scores on a math test. In the 90s a series of experiments showed that if you made students identify their sex (or race) on an exam then this would cause their scores on math tests to change, causing, for example, females scores to fall. The thinking here, of course, is that recalling your gender might also recall the negative stereotypes which are place on females in math. Well what the Washington Post article describes is what happens if you do the opposite. What happens if you ask questions before the exam which remind the students of their postive attributes. Well, the WaPo reports that a recent study found that in fact in this case the male test scores stayed the same and the female test scores increased such that they were indistinguishable from the male test scores! Having the gall to believe you could possibly be smart is, indeed, it seems very important. (I had an English teacher in middle school who used to berate the students for making fun of people who were doing well in class. “Why wouldn’t you want to get good grades?” she would ask. Thanks Mrs. Perry!)
Similarly, it looks like those “this is your brain on drugs” ads have a sort of placebo effect on ecstasy users.