Carol Greider, molecular biologist at Johns Hopkins University, explains why scientists must be critics, not cheerleaders, of their own hypotheses.
Read the full transcript »
The Mistakes Scientists Make in Their Work I've made a number of mistakes along the way. When you make—I feel like if there's a scientific mistake—so when I publish something and it turns out that the conclusion that was drawn was not completely correct, then it's the responsibility of the scientist to then do the experiments and publish the correct answer. But in that sense, science is very self-correcting, because if something is published and it's important, and it's wrong, then a number of other people will find out that it's wrong, although that may take a number of years, and some people may be led astray. So I think that the real importance is to not be necessarily attached to a particular idea, with the idea that telomere shortening plays some role in cell death, say. The thing that we would like to do is to test that idea. And one thing that I think creeps into people's thinking sometimes is that they want to prove a hypothesis. And I never think about proving a hypothesis, but rather think about different ways to test the hypothesis. And this was something that I think I learned very early on in Liz Blackburn's lab when we first had discovered something that looked like it was elongating telomeres. And then a period of about nine months went by where we set up to ask ourselves, is it really true that we've discovered a new enzyme? And we came up with a variety of methods to shoot down our own hypothesis. Maybe we're being fooled because it's a normal DNA polymerase that's just making this, and it looks like it's something new. Or maybe we're being fooled in a different direction. And then after the discovery withstood the test of nine months of attacks from our own standpoint about how could we shoot ourselves down, then I started to really believe that it was true. And so that really taught me that it's most important to be critical of your own hypothesis rather than to be a cheerleader for it. And I have a little bit of a fear against people just cheerleading for ideas. I think it's more important to test them because that's how you then move forward, because many ideas are going to be wrong. And if you test an idea and find out its wrong, you can go in another direction. But if you're a cheerleader for an idea, it may last for a longer period of time even if it's not correct.