Lec 2 | MIT 9.00SC Introduction to Psychology, Spring 2011 30:45 Can we test that experimentally? So here's the experiment. And the idea is, again, that, when people observe that others violate a social norm or legitimate rule, are they more likely to violate it? If you see people around you breaking rules, are you more likely to do it? And you could say, it seems like it is. But let's do the science. So here's an independent measure. I'll show you the environment. The dependent measure is the number of people who perform a violation. Now, you can't go getting people murdering each other and stuff for an experiment. But you can do this. You can have two conditions. So on the top condition, do you ever go back to your car or bike, usually it's cars, I think, and they put leaflets in your car? And now what are you going to do? You're running away. They'll leave the leaflet on your windshield wiper. So they put it out. And they either had a wall like this, which was clean. There's a sign that says no graffiti. And they put on those white pieces of paper onto the bikes. Or they put a lot of graffiti on the wall. And people came out from their classes. They parked their bike. They ran into class. They came out. And they got all these white leaflets. And what they did is watched. And they said, how many people put the leaflet away? And how many people crumpled it up and threw it right on the ground? So we don't know what they're going to do with the leaflet after they put it away. But we could hope they're going to throw it out. But the one thing we know for sure, if you crumple it up and throw it on the ground, it's littering and not a good social norm. And sure enough, 1/3 of the people littered if the wall had no graffiti. But it more than doubled if the wall had graffiti. So that's a small thing. That's not a broken down, complicated neighborhood. But just at that moment, as you roar out of class, you get that piece of paper on your bike. And you see like, wow, this is a pretty crummy place. People are breaking the rules all the time, twice as likely to throw a piece of paper on the ground. So that shows that people very easily are influenced by very mild perceptions of whether people are following rules or not and will be their better selves or their worse selves depending on that sense of are all rules on or are all rules off. And there's a pretty good bet they're not thinking about it, right? They're not sitting there thinking, OK, here's littering, not as bad as killing somebody. Here's the wall. Well, there's 30% graffiti. I mean nobody does that. They have an emotion quick response. But the environment makes a difference in that regard. So an important thing about studies like this, so let me tell you about this one. Here's a study to look at a drug that was supposed to lower mortality from coronary artery disease. So those who took the drug, 80% or more, because people follow whether you take the drug, if you don't take the drug, of course, the drug won't work, had a 15% mortality rate within five years. Those who took the drug less than 80% had a 25% mortality. So you'd say, well, 10% better chance of living for five years, that's worth it. The nice thing is they had a placebo. And there was no difference whatsoever between those who took the active drug and those who took the placebo. The difference was those who followed the instructions to take the pill and those who didn't. So what does that mean? How do you interpret that? It's not the drug. Yeah? AUDIENCE: It could be the stress of not following the directions, meaning it's not working. PROFESSOR: It could be that. It could be the stress. It could be that. But let me make a guess. There's no scientific answer to this. People who tend not to follow instructions, are they mostly stressed by not following instructions? The intuition is they're blowing it off. They're like, yeah, that's the instructions. But I got things to do and places to go. But you could be right. So anyway, the idea is, what you pick up is, are people who tried to take care of themselves. They have a health condition. And the people who are following the pill instruction are probably also doing the best they can in terms of exercise and diet, all the things they could do to maximize their health, right? And they're taking their placebo. Or they're taking the drug. They don't know. But they're doing everything they can. And the people who blowing off stuff, they're the ones who are doing all the things that aren't healthy, including not even taking your pills. I mean, if you don't even take your pills, right-- Exercise is work. Diet is work. Popping a pill that's been handed to you that you're told might help you is probably not that much work. It's what's called self-selection bias, that you stand out if you tend to take care of yourself or not. So when we run experiments at universities that involve people, a huge issue for us is the idea is this, that we randomly recruit people. And we'll come back to this in a moment. Then we randomly assign them to one condition or another, for example, the placebo condition or the drug condition. We randomly assign them to the messy wall or the clean wall. We discover something. And then we try to generalize it as far as we can to people as a whole. We've discovered something fundamental about how humans work, generalizable principles of human mind or behavior. So what really happens in the world? And you're in between sort of the best possible world and the world we actually live in. So is it really random who participates in research studies in psychology departments where a lot of the research gets done. Is it really random? No. So who volunteers? Have any of you volunteered to be a subject in an experiment anywhere? Yeah. So it's not random. Some people, they either need to make a few bucks. Or they're very curious about psychology. Or they have a friend who's in psychology. It's not random. And in fact, population-wide, people have said that psychology could be described as WEIRD research for the following reason, that the vast majority of the findings you'll read about everywhere, including in your textbook, but everywhere are from nations that are westernized, educated, industrialized, and relatively rich democracies. The vast majority are between 17 and 25. They're undergraduates or graduate students who participate in research. It's been estimated that, if you're a US college student, you're 4,000 times more likely to show up in a scientifically reported experiment than if you're a person anywhere else on the planet, 4,000 times. Because why? Because it's changing a little bit but for many, many years the huge amount of research happened in the US at universities, pretty much at better universities, pretty much at highly selected students. And then we never publish that. We never say MIT students did the following thing. Because people will say, yeah, MIT students. Because then we have a different report from Stanford or from UMass or from Michigan or Texas. We say, no, humans did an experiment. And humans are like this. And it's just something to think about as you see studies and stuff like that. And people are more and more sensitized to that. As more and more of the world gets somewhat industrializes, there's more and more research around the world. But still the vast majority of findings are you guys. Now, that's just to start with who's in the subject pool or the participant pool. How normal are you when you do an experiment? Are you like yourself? Or are you doing weird stuff? And people use this idea that you're really weird when you do an experiment in another sense. Because you don't know what's going on. You're following instructions. So they do experiments to show how ridiculously people behave when they're doing an experiment like they would never do it. So they ask them to come in. And they say, will you fill out 200 sheets of paper filled with random digits, add them pairwise? And I'll come back in a few hours. Now, if a friend asked you to do that, add random digits for a couple hours, you would go no, thank you. I've got about eight million better things to do than that, right? I could multiply random digits just to start with. But people will do it for hours. Because they're in this weird like that's the instruction. And that's what my mission is. Or even more ridiculous, they pick up a card after each page. They've added all these numbers for no good reason. And the instruction tells them, when you've done the page, tear the page up into 32 pieces. And here's the next page and the next page. It's obviously fruitless and pointless, ridiculously so. But because you're in an experiment and somebody told you do that, you just keep barreling forward. How much are we who we are when we do experiments versus people following weird instructions? And we'll come back to that in some serious context. That's the subjects. How about experimenters? Experimenters are never objective. We try to be as scientists as objective as we can. We're never, we believe certain hypotheses. We have certain goals. We recognize results that would be super exciting. We recognize results that would be boring. We're humans, too. We want to do stuff that matters. We want to do stuff that people reward us for. I mean it's just human. So here's a lowest case example of how, once you're an experimenter, you produce your own results. And believe me, it's much more when your result matters for things like your career advancement or your status in the field. So students were told that rats were either maze-bright or maze-dull. They were told that these rats were bred to be smart learners or not so smart. The rats were all just random rats. It's a fake story to undergraduates. Then they had them test on a maze. And they said for your laboratory exercise write down how quickly the rat runs the maze. And amazingly, the smart rats did way better than the dull rats. Now, these are not students trying to become famous or have a career. They're handing in an assignment. And how did that happen? Well, they went back. And it wasn't that anyone faked data or anything like that. But when they looked at films or discussed stuff with kids, students in the course, every little thing that the students did that moved the data in tiny, tiny ways always favored finding the maze-bright rats being brilliant maze runners. So for example, they would put the rat-- if you've ever run a rat in a maze, has anyone run a rat in a maze? Yeah? Is it easy? Have you picked them up and put them in the maze? It's more automated now. Is it physically easy? Are rats like, please, show me the starting line? I'm ready to go. No. What are they like? Help me out here for you guys who have done this. Are they kind of really-- AUDIENCE: [INAUDIBLE]. PROFESSOR: And they don't like to be handled that much really. And they're fighting. And these are undergraduates who've hardly ever touched a rat. They're like, please, don't attack me or something. So they're holding up. And the rat's fighting back. And because it's really uncomfortable, the rat's really unhappy. And the student's like really-- So what happens is-- and then they plunk the rat down. And they go, oh, my gosh, it's facing the wrong direction. And it's a smart one. Wait, stop the clock. Let's get him-- But if it's a dull one facing the wrong direction, it's OK. It's a dull one. Let the maze-- so lots of little things. Are you sure you wrote this down. Was it 12 seconds or 11? I guess it was 11. Oh, because it's a smart one. So completely innocently, completely innocently as far as we understand, they made the result to match up with what the expectation was. And it's not just that. And if you know this example, cognitive psychologists have studied a phenomenon like this. So try this at your desk. Each card is a letter on one side and a number on the other side. Your job is to figure out what's the smallest number of cards you have to pick up to discover if this rule is right. If a card has a vowel on one side, then it has an even number on the other side. So think about this for a moment. If a card has a vowel on one side, then it has to have an even number on the other side. What's the smallest number of cards you can pick up to test that hypothesis? So here's what people have found. The correct answer is E and seven. And about 10% of people on average come up with that. Many, many more people pick E alone or E and six. Why? Well, everybody gets that you have to look at E. Because if it's an even number, so you flipped E over. And if that one turns out to have, it's a vowel, if it has an odd number, the hypothesis is wrong, right? Everybody gets that you should check E. Pretty much everybody gets that J is irrelevant. Because it's a consonant. Now comes the interesting six and seven. Because the sentence is talking about a vowel and an even number, sometimes people feel like I got to check the even number. But it doesn't say anything about even numbers couldn't go with consonants. It doesn't tell you anything. That's not part of the hypothesis. But if you flipped over number seven and you discovered a vowel, you'd violate the rule. So what happens is they're tempted to go for things that feel like they confirm the question. And it feels like E confirms it. And it feels like six confirms it. It's what is called confirmatory bias that when you look at stuff, if you see what you think confirms it, then you tend to go for that, even when there's evidence in front if you that could contradict your expectation. So now I'm going to give you a practical suggestion for how to win somebody over if you have a brewing romantic interest in that person. So finally we get practical, right? Everything was like, I look at you, I'm not running rats. And I think it's very interesting. Because when we think about how people make decisions in their lives, we tend to think intuitively about big decisions. What's my career going to be? Where do I go to college? We think those are the big decisions. But a huge amount of life is little decisions, moment to moment. If somebody asks you for something, did you say yes or no? Somebody walks up to you at a party. They want to talk with you. Do you feel like talking or not feel like talking? Maybe you talk with them and it's a lifetime friend or romantic partner or whatever. So I think we make constantly little decisions that turn out to big effects in our lives. So here's some evidence about things like that that influence that. And this is worked on at Yale. So we describe people as warm. Or we can define them as cold or aloof, right? And we could think those words are kind of arbitrary. And most of us, if you met somebody at a party, would you rather hang out with somebody who's warm or cold? Warm, OK, just checking. Now sometimes there's an idea of embodied cognition. And here's what it is. We can't really know how warm or cold that person is or even that's an abstract idea. But we know we're comfortable when we're kind of warm. And we know we're not comfortable physically inside us when we're kind of cold. So there's an idea that when we think about other people's feelings, other people's thoughts or abstract things in the world, we base them on physical experiences that we have inside, that there's a more direct link between what we know inside us. And we use that to interpret the world out there. Embodied cognition, your body is the vehicle for understanding things out there. So here's the experiments. It's really clever. There's two temperature conditions. So what happens is you go into a building. And the examiner, an undergraduate usually or a graduate student, meets a subject in the lobby. And she's carrying or he's carrying a cup of coffee, sometimes it's hot coffee, sometimes it's cold coffee, a clipboard, two textbooks. So you can imagine. It's a set-up, right? The person's walking. They're juggling with the coffee, the textbook, oh, the clipboard. And he says, on the elevator he says, could you just hold my cup while I write down your name on my list of subjects in the experiment? It's very innocent, right, on the elevator. And so you hold the cup of coffee, which half the people get a hot cup, half the people get a cold cup. Then you read a description of a person. It's just a paragraph. And you have to rate whether you find the person described as warm or cold. And if you had the warm cup of coffee, you're more likely to describe that person as a warm kind of a person. Now, these subjects have no idea. They held a cup of coffee. They probably didn't even think whether it was warm or cold. But something in their body that got warmed up to the least degree possible, that thought is still in their head. So when they read a description of something, you go, oh, this seems like a warm person. Or if it was a cold cup of coffee, they say, oh, this seems like a cold, aloof kind of person as far as I can tell, that little bit of difference. So here's the lesson just so you got it. If you want to take somebody because you have a possible friendship or romantic interest in that person, are you better off going out for a warm cup of coffee or giving them a cold glass of beer? For that first impression, scientific research shows it's something warm will work it out. It's not that powerful. You're not going to be able to control the world. But what it says is when we are in ambiguous situations, ambiguous situations, we first meet somebody. We don't really know them. Little things can start us off on one path. Or it can start us off another path, surprisingly little things. Now, then they have to worry about a confound. Can you imagine a confound here? Think of the rats. Who knows if the cup is cold or hot besides the person holding it? The examiner, right? Now, maybe the examiner doesn't try to do this, but maybe the examiner, no matter what, says, oh, we're in the warm condition. And I'm smiling because this is going to be nice. I'm helping this person feel good. Here's the aloof condition I'm giving. We're all going to be miserable. And the examiner knows it. Because they know if they have a hot or cold cup of coffee. So maybe the person's not vibing to the coffee. They're vibing to the expression or feeling or sentiment of the examiner who understands what's going on. So they do one more experiment where they give-- have you seen these hot and icy therapeutic packs that you can make hot or cold? Now somebody else makes it hot or cold. The experimenter has no idea whether it was hot or cold. So it's what we call double blind. A person doesn't know it's relevant, the experimenter. They rate the pad. And now, this moves it a little bit, they either get, after this thing's over, they're told it's a product evaluation of the pad. But they either got a hot pad or a cold pad. And they can either choose one of two rewards, a Snapple beverage or a $1 gift certificate, whichever you like better. You had two little things you could do. But here's the big decision. Do you get it for yourself? Or we can send it for you to a friend. We can send them the Snapple or the certificate. If the people got a cold pad, 75% of the times they say, please, give it to me right now. I'll take it. If they had a warm pad, the majority of the time they actually send it to their friend. That pad difference moved quite substantially the likelihood that you take it for yourself or you give it away, quite substantially. If you're a charity trying to get people to donate or you want to get help from somebody, 25% success rate is a lot worse than a 54%, right? It's a doubling, a doubling, of whether you keep it for yourself or you're generous to your friend, just that therapeutic pad. I'm going to give you two more and then go to-- How about money? When you think about money, I don't mean traumatic thing about money, just a little bit about money, how does that influence how you relate to other people? OK, here's the experiment. The idea is that money, on average, when you just think a little bit about it-- and I'll show you how ridiculously little they think about it-- their hypothesis is that money makes you think about self-reliance, you earn your money, your bank account, compared to situations where you're not thinking about money. So here's what they did. They have people come in in three conditions. And they had to unscramble sentences. You have no idea what's going on. You got a sentence like, cold it desk outside is, which you would make into a good sentence, or high a salary desk paying. So you just make those, you scramble those words around. But the money condition has a word like salary in it. Or you get the neutral sentences that have no money reference. And you see Monopoly money in the corner of the room. So we're not talking like people just sent you your tuition bill or whatever. Or you just found out that something terrible happened financially or you won the lottery. It's Monopoly money. It's just like the idea of money. And then they give you a task to do. And the experimenter, as he or she leaves the room, they say, it's a pretty hard task. You may want some help. So just come and get me. I'm glad to help you, not a problem. So what happens? Here's the probability of getting help. And it's incredibly more for the control group that did the sentences without money and way less not only for the people who unscrambled the sentence with the word salary but also way less if they saw Monopoly money in the corner. The mere idea of money, not money given to you, not financial trauma or financial gain, the mere idea of money makes you less willing to get up and get help. So the positive it makes you more self-reliant. The negative is people were less willing to help somebody. So now at the end as the examiner comes and says, I'm an undergraduate like you. I've got huge pressure to get this paper in. Can you help me code? Can you help me score up a few sheets? People were twice as much willing to help if they just did the neutral things than if they unscrambled the money sentences or saw the Monopoly money in the corner of the room. Because just that thought about money goes like, man, self-reliance, we're all in it for ourselves. I've got my money. You've got your money. Good luck. Self-reliance can be a good thing. But it doesn't actually go with helping people very much also. Yeah? AUDIENCE: [INAUDIBLE] because the Monopoly money and he's suspicious of why this money's over there, that's why you're not going to help. PROFESSOR: It could be. Yeah, the Monopoly money, I agree with you. Maybe that's why they also had the one where you just get the scrambled sentences. I agree with you. You could think there's many, many ways to think about this. The last example of this kind that I want show you is dating behavior and the way, again, an experiment can challenge various ideas. And this is specifically about dating between men and women. The story, to the extent it's been studied, would be a little bit different if you're dating within a sex. So if we're dating between men and women, who is more selective in their dating choices do you think? Who's more willing to date widely or a lot or less? What's your thought if you have to generalize about men and women on average in your-- AUDIENCE: Men. PROFESSOR: Thank you. And evolutionary psychologists, and I love evolutionary psychology. But it drives me crazy, too. Evolutionary psychologists will go, well, because by evolutionary theory men have to go and make as many offspring as they can to keep their genes going. Women have to be more selective. It takes nine months to bear a baby. It takes nine seconds to do the male part, right, estimated. So this might or might not be true. It's kind of fun. We all laugh about it. All of us find it kind of amusing to sort of think about. But how would we actually know if this is vaguely true on average? How would we actually know if men for dating behavior? How would we even, is that true? Is that a bad myth, bad rap, or whatever? Yeah? AUDIENCE: Dating websites? PROFESSOR: Yeah, thank you very much. So there's an entire research enterprise out there now that uses online dating sites, which, of course, it's only in the last 20 years or something that they've existed, I think. So it wasn't before. And the other one that's been a treasure trove of data is speed dating environments where you can observe things and measure things. And those studies have shown, for example, that online dating on average men are 1.5 times more likely to send email offerings dates to a woman than the other way around compared to women. And in speed dating, too, men are more likely to indicate they would be happy to date a person they met than women. So there's objective evidence that this is true on average in this society at this time. And so everybody, oh, yeah, this shows the evolutionary psychology. Evolutionary psychology drives you crazy. Because, on the one hand, it's certainly the case that our minds and brains evolved through evolution. So there's something about the history we went through as a species that matters for who we are now. That's got to be the case. On the other hand, it was a different world before internet dating and speed dating, right? It's hard to know how those things line up exactly. So here's the fun thing. It turns out that on average in speed dating a typical speed dating event will have four minutes, right? So you typically have women sitting at desks. Almost always they set it up that way. And the men are sitting in chairs in front of the desk. And they meet. And they talk for four minutes. The bell rings. The men jump up. And they shift over one station. And then what happens is, for each time you meet a person, you fill out a card that says yes or no I'd be willing to have a date with the person. And if you have two yeses from the man and the woman, the speed dating organizer exchanges the information so they can make direct contact. Does that sound OK? And men make more yes responses. And it turns out, it doesn't have to be this way, but almost, from my reading, that almost every speed dating thing has the men rotating around and the women sitting there waiting for the men to rotate around. So let's just ask this question. What happens if the women rotate, which almost never happens? But they did this experimentally. They said for half of the speed dating sessions the men will be sitting at the table. And the women will stand up and move a chair over every four minutes. The bell rings. So the women are moving and rotating. And here's the results. Basically, what's showing on the left are when the men rotate, the standard condition. First, it's romantic desire. Men feel more romantic desire, on average, for the partner they sit with. They feel that more chemistry is happening with the person they sit with. And on the bottom left corner, they also are more likely to give a yes response, I would like a date with this person. But reverse the roles, have the women rotating, and look at these bars. There's no difference at all. You completely wipe out the male female difference and self-reported romantic desire per person, self-reported chemistry they think is happening, and the yes no responses that I would have a date. You completely eliminate it when the women are rotating. It's flat even. So that's the result. What does this suggest? What happened to the evolutionary psychology? It disappeared just by having the women rotate around the room. It suggests, and we don't know this with certainty. But people are following this up who are interested in this. It suggests that there's something about getting up and approaching that person, your body going here I come, I'm not sitting in my chair hanging out, it's like here I come, that makes a person energized, feel more romantic, feel more chemistry is happening, and more likely to ask on a date. And it doesn't matter whether you're a man or a woman at all. It's the getting up and that physical approach. You're more invested. You're more out there. Who knows what it is, completely eliminates what everybody had said is an ancient, inevitable, evolutionary difference between men and women. And this isn't to say there might not be some of those streaks in us. Who knows. But speed dating is not showing that. All you got to do is have the women rotate. And you completely eliminate what's thought to be some hard-wired biological difference on this dimension. Now, talking about that, when we do experiments like this that look at sort of fun things like dating around a thing or age difference between a five-year-old and a 20-year-old or a 20-year-old and a 80-year-old, cultures, there's lots of experiments that compare say Western cultures verses Eastern Asian cultures, we'll talk about that later in the course in terms of behaviors, genes, whether you have one gene or another, all those things, are those independent variables? No. We can't assign you randomly to be a man or a woman, young or old, to come from the United States or Japan or China, or to have this gene or that gene. It's not that we couldn't do stuff in some monstrous world but not ethically. And it's never going to happen, right? So whenever you have science that talks about differences between men and women or young and old or cultures or genes, it's not really a completely independent manipulation. We can't assign you randomly. And so it means everything in the world about what it means to be a man or a woman, a 20-year-old or an 80-year old, come from this continent or that continent, have this gene or that gene, those are not things we can freely manipulate. And we have to be extra cautious in interpreting those kinds of studies. Because we can't randomly assign you to conditions. We just can't. And I didn't talk about the fact that, up until this moment, we've made every conclusion a complete generality about human nature. And people are different one from the other, individual differences. We'll talk about that later in the course, too. So for the last couple minutes I want to talk about this, folk psychology. Everybody has ideas about psychology. So if you have this piece of paper, if you would zoom through it, ideas about people in psychology and just where there is some scientific evidence, if you just zoom down and mark yes or no-- On average, do opposites attract? On average, does familiarity breed contempt? On average, more people present in an emergency, are you more likely to get some help? On average, are there such things as visual learners and verbal learners, and if they got the right kind of instruction that would help them in school? Is hypnosis baloney? Does subliminal advertising work? When things are presented so you don't see them below consciousness, does it help you buy something? Does playing Mozart to an infant boost their intelligence, Mozart effect? Is old age, on average, associated with dissatisfaction? If you're unsure of your answer when taking a test, I'm going to tell you the science about this. It's best to stick with your first hunch. Practically every website tells you that. Have you heard that many times? I'm going to tell you what the actual evidence is about that and how they get that in a moment. Ulcers are caused primarily by stress. A positive attitude can stave off cancer. Raising children similarly leads to similarities in their adult personalities. Low self-esteem is a major cause of psychological problems. People's responses on the ink blot test tells us a lot about their personalities. Interviews help identify those most likely to succeed in medical school. OK? I'll give you a minute. A bunch of these we'll come back to in the course. But the answer for all these is this. And there's evidence to it all. So I'm going to tell you about four of them for 10 minutes. One of the ones you might want to know most about is the changing your initial hunch answer. How do you know that? So the people have gone back over test scores and looked where people crossed out an answer or erased an answer. That's the best source of evidence they have in large samples. And people are twice as likely to give a better, the correct answer, than to move to a wrong answer. That's the empirical evidence. So it's a fantastically interesting disconnect that you see so many places. In teachers' advice, you go with your first hunch when you're unsure. The empirical evidence is you're much more likely to be correct when you go with your best final answer. I mean, of course, you're not sure. So you're going to miss a fair bit. But you're twice as likely to come out with the correct answer when you go with your best possible answer. So that people who crossed out their answers or erased their answers, they were twice as likely to move to a good answer than leave behind a good answer. Yeah? AUDIENCE: Do you know that maybe it's they picked an answer and then they switched their answer and then picking their best and going back later-- PROFESSOR: Yeah, that's an excellent question. We don't know. They've kind of gone like five times back and forth. The assumption has been, on average, that people will go once back and forth. But you're right. But ask yourself the other question, too. What's the evidence ever that going with your hunches? And nobody's been able to show that. So where there's evidence, it's all in favor of best answer. And nobody's shown evidence that hunch is the best way to go empirically. That doesn't mean that sometimes it doesn't happen. But on average, that's the way the ends go. But your point's excellent. How about Rorschach tests? Every psychology program, clinical psychologist in the country, is required to learn how to administer and evaluate a Rorschach. You know this? So this obviously shows, obviously it shows, some sort of horrible sex act between animals. You know these kinds of tests. You're supposed to read it. And then the examiner listens to you. And they figure out something, right? There's been about 10,000 publications about this. There's no evidence that there's anything predictive or correct about this test, none, zero, none. People argue it. People give it. People have to learn how to give this test. So this is a disconnect that happens a lot between clinical stuff and experimental work. I mean researchers can't grasp why people still give this. Medical school interviews, those of you who will apply to medical school will get interviews. You'll get interviews for jobs. You'll get interviews for all kinds of things in your life. What's the evidence that interviews are any good for anything? How could you know whether interviews are good for anything? A company interviews you. A university interviews you. How would you even decide whether? So I can tell you this. Social psychologists have shown repeatedly that interviews are cesspools for discrimination. Because people tend to like people who are like them. Oh, you went to MIT? Yeah, course nine. Then you go, oh, man, this is going to be good, right? I went to CalTech. We didn't like MIT people. We thought they were a little bit snobby or something. And you go that's not going to be so good. Social psychologists have shown that, if you just vary information in practice interviews, people tend to like people who are like them, by background, by appearance, and so on. On average, they try not to. They try to be fair. So we know that. So here's what they did. Yale compared students who were accepted or rejected on the basis of an interview. So they all had scores that got them to the interview. And then they looked at them where they went in pairs or larger than pairs to other medical schools. So let's say you got accepted by Yale. And you were accepted or rejected on the interview. But you ended up as two students at Penn or two students at Michigan. You're sitting next to each other. One of you was accepted by interview at Yale. One of you was not. Who does better at that other place? They're dead even. The interview added nothing for predicting grade point average or completion of medical school. Here's one more example, University of Texas enlarged in the midst of their, at Houston, in the midst of their interviews, it enlarged from 150 to 200 students. They had interviewed 800. The 150 who came were among their top ones. They interviewed them. They liked them. They came. And then all of a sudden because the state legislation said, by law, mandating that you take 50 more, they took 50 more who didn't do so well in the interview. Those 50 more performed identically to the ones who got in with the successful interview. They couldn't find a difference between them. Now, there's at least one thing you might think the interview might have done. But we don't know. What do you think it might have done? So the outcome was your grades in medical school and your likelihood of completing medical school. What else could an interview have been relevant for? Maybe, maybe, maybe something about bedside manner or how you relate to people, maybe, maybe, maybe, but there's no evidence that that's true. Now you're starting to grasp at straws. So there's no evidence that interview processes make any difference anywhere. But everybody does interviews all the ways. Because we always think we're kind of good judges of human nature or something. So interviews, it's very hard to tell that they ever do anything. So here's my last thing. Self-esteem, you cannot watch Oprah or Dr. Phil without some discussion of self-esteem at some moment. And it sounds good to everybody. I mean nobody would wish for you as a parent or friend for you to have low self-esteem. And in fact, higher self-esteem is correlated with higher initiative and persistence, happiness and emotional resilience, and also, unfortunately, with people who are narcissistic and bullies. And how they measure it, just so you know, because this is a problem, too, how do you measure it? There's several scales. But the most widely used scale is this Rosenberg scale. So you have that, too, if you want to try it out on yourself. Self-report, and when they reviewed 15,000 studies some years ago, they could find no evidence that higher self-esteem leads to better things. It all seemed the other way around that doing better at things led to higher self-esteem. They could never find that pushing your self-esteem up doesn't. Having said that, I can tell you as a parent of two young kids or if you have friends and you care about them and you know this, we like to praise people to build up their self-esteem. We tell you you're awesome. Someday you'll go to MIT. I know you will. You're a wonderful person. And you're smart. And you're beautiful. Parents want their children and teachers, too, to grow up and thrive and build up that self-esteem. And the easiest way to do that is praise, right, just telling your kid how awesome he or she is or your younger sibling how awesome her or she is. Get their self-esteem up. So the last thing I'm going to tell you and then we'll be done in about three minutes is work from Carol Dweck that shows you how dangerous it is and counterproductive it is to have praise where you don't think through its relation to people. And in one sense we'll talk about, are you going to succeed in life by your effort or your talent? What's a bigger question than that? Wherever you get to in your life, it's going to be based on your effort or your talent or both. I'll just have a sentence about that now. So here's the experiments she did with fifth graders. She had them perform a pretty hard task. And afterwards, randomly, they were either praised for their intelligence. So a person said to them, you, fifth grader, you are so smart. I can't believe how great you did on that, fifth grader. Or they were praised for their hard work. This shows a lot of hard work, different kids. And some kids got no praise at all in this experiment. Then they get a task that's even a little bit harder. How does the different kind of praise influence things? And here was the finding. If you were praised for your intelligence, you did worse on the second test. If you were praised for your hard work, you did better. If nobody said anything, your performance didn't change. And the idea is this. You understand, praised for your intelligence, you did worse on a hard task. Praised for your effort, you did better. And here's the idea that, if we praise people for their intelligence to build up their self-esteem, they start to care more about how their performance reflects on them than the performance itself. Because you're trying to prove that you're an intelligent person. The students who get that kind of praise for their intelligence, they persist less in a difficult task. They say they don't enjoy challenges. Why? They just want to ace everything to keep proving they're smart. The kids who got praised for effort, they're kind of like, OK, I'm going to make some mistakes. But I'm just going to keep going. Because with effort I can do things sometimes. And it turns out that it makes a remarkable thing. And so Dweck, and we'll come back to this, talks about you can imbue your intelligence as a trait, I have only so much, or as a growth, the harder I work, the smarter I get. But I have to do the hard work to get there. And we live in a society that often goes by fixed traits. American Idol, the harder the work, the better you sing or who has the talent is discovered? Who has the talent is discovered, right? Because the idea is you have a certain amount of musical talent. And now the judges show. And you break into the world. And you're a famous singer, right? So ask yourself this. When you are good at what you do, is it because something in your genes and environment made you good at it or is it because you try really hard? So Malcolm Gladwell has this statement that's floating around, gets a lot of attention, that anybody can become great at anything if they work on it for 10,000 hours. So how much of your success in the past, currently, and in the future is going be based on your effort? And how much is it going to be based on your talents? And how would we know that? And we're a very talent-oriented society. And it seems like, the more we encourage effort, maybe the more people might flourish. So we talked about how psychology and science, experiments, brief history of psychology experimentation, some issues in experimentation, and some psychological topics, and some practical information for being generous and kind and winning friends rapidly. OK, thanks very much.