 One of the interesting ways that some of these large-scale frauds have happened is that someone who's like, say, the head of the lab gives the junior people a set of data and says, look, here's the data set. Let's analyze this and write a paper on it. What never pops into their head is that their mentor, in a way, their scientific hero, the person they wanted to come and work with, often you go to a university to work with a very specific individual to get your degree. Nowadays, it just doesn't occur to them that that person might have made up the data the weekend before and given it to me to discover, quote unquote, what their results are and so on when they already know. That's an example of what we'd call a commitment. You sort of have the assumption that the data came from a legitimate process, like someone ran the experiment somewhere and now I have the data and you don't ask enough questions and asking your boss, did you fake this last week? Not going to go over well. There's always the social and the power and the other dynamics about asking questions, which is part of why it's hard. Even just the social dynamics of being like the one guy in the sales meeting who asks hard questions of the salespeople, like that can be hard in some cultures, some cultures, some organizational cultures and so on. Being the jerk is not easy and that's part of what sort of fuels this. That's one reason why we say in the book, it's often good to try to get advice from an outsider who doesn't necessarily have the same assumptions in their head, who doesn't have the same history of familiarity and trust with the sources, who hasn't been paying attention to all the things you have and maybe has been looking at some different things. There are plenty of examples where massive frauds could have been avoided if the person who was defrauded listened to somebody else who thought it was a fraud before they got started into it. We have a great example of a con artist who was about to con some French businessman out of $3 million or something like that. The businessman's friend happened to walk in during the Skype call where this con was happening and said, when the friend went to the bathroom, he said, I think this is a con and hadn't occurred to the guy yet. That's why in science, we do try to have a culture of independent replication. We make things public, we publish the papers, other scientists should be free to try to replicate the same stuff and they have a different view. Sometimes it's unpleasant when you publish something and then someone else says, I tried it again and I didn't get exactly the same results, but we're engaged in a collective endeavor, not individual glory-seeking process in science. Although some people are in it partly for that reason or entirely for that reason, but that's not the social purpose of it, which we have to keep in mind. Our vanity and wanting to get ahead of everyone else plays into that line as well, where it's like, oh, I've got the skinny, I'm moving up, I got this, I can't share this with anybody. That's exactly what they want you to think, because then you start mentioning, and of course, people are going to tell you, that doesn't add up and that you're being scammed. And of course, you want to believe that you do, that you were the smart guy who figured out the shortcut, who asked the right questions and got the right line. So some people have the reputation and many people had this reputation in the past. Their experiments always work and people started even coming up with explanations for this. One prominent psychologist said that you have to have flair, f-l-a-i-r, like some kind of ineffable ability to make the experiment work that can't be written down. There's no recipe for it. It's almost like magic or something like that. I don't think he intended it to mean magic, but it's a quality that's so hard to bottle that you either have it or you don't, and that is not the way science works. Science doesn't work where if I assemble the rocket, it makes it to the moon, but if the next guy with the same training and so on assembles the rocket by the same instruction manual, it crashes. That's not the way it's supposed to work. And yes, incentives of course can cause greater effort and greater work and greater achievement, but they can also cause greater fraud. Because if there are two ways of getting the award, the legitimate way and the illegitimate way, you're probably going to cause more of each way to happen once you put the award out there. You might almost argue that the more rewards and honors and so on there aren't a field, the more attempts there are going to be to game them in some way.