Today, Explained - The Stanford Prison Correction (replay)

Episode Date: December 27, 2018

In 1971, a professor locked a bunch of young men in a basement to understand evil. The results were explosive. This summer, it all came crashing down. Vox’s Brian Resnick explains what’s going on with the Stanford Prison Experiment in this holiday rebroadcast. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Brian Resnick, science reporter here at Vox. I feel like I heard a bunch about the Stanford Prison Experiment in college, and then I pretty much forgot about it. And then suddenly this summer, it's been everywhere. I see it in the news everywhere. What is this thing, Brian? What's going on? The Stanford Prison Experiment is a part of this group of studies that are trying to explain evil. It's 1971, and we're at Stanford University.

And Philip Zimbardo is a researcher there. He's wondering, what is it that brings out bad behavior in people. And the working hypothesis going into the Stanford prison study is that it's circumstance that some of us are put into positions of power and some of us are put in positions of weakness. And in those circumstances, a natural thing happens where the people in power abuse the people who are in less power. So this Stanford Prison Experiment kind of really went for this. They constructed a mock prison in the basement of Stanford's psychology building, and they recruited this group of young men. They got paid around $15 a day,

and the participants were randomly sorted into different groups. Half of them became the guards of this prison, responsible for keeping the prisoners in line, and half of them were prisoners. So it started off with the prisoners getting arrested, like a mock arrest. It was a real cop car, it was a real policeman, and there were real neighbors in the street

who didn't know that I was, that this was an experiment. They were brought to this prison, which had no windows. They had no way to keep time throughout the day. It's just like a room with a door with bars on it. You could tell it wasn't a real jail. The power differences were set up from the beginning. So the prisoners were stripped. They were deloused.

They had to wear stockings on their head to simulate their heads being shaven. This is one of the prisoners' uniforms, Prisoner 819. You can see it's really a dress. Here are the chains that the guards made the prisoners wear to remind them of their status. The guards, on the other hand, were given these mirrored sunglasses. Military uniforms the guards wore, their billy club, symbols of power and authority. So this experiment starts. And the key of this experiment was supposedly that no one was explicitly told what to do in this situation.

And the first day, nothing much happened. The guards and prisoners got along. It was pretty peaceful. And then the first day, nothing much happened. The guards and prisoners got along. It was pretty peaceful. And then the second day, like, chaos. The prisoners rebelled. They, like, took off their numbers. They wanted out. They were really rebelling against the guards. And then the guards, from that moment on, really stepped up in their cruelty and their meanness and their severity. So there are reports from this experiment that some of the guards stripped the prisoners naked. They make them do push-ups. They simulated acts of sodomy on them.

It got really intense. One of the prisoners reportedly had this mental breakdown. He began to play the role of the crazy person, but soon the role became too real. Yeah, the exact quote is like, Jesus Christ, I'm burning up inside. Don't you know? I can't say that. I'm fucked up. I don't know how to explain it. I'm fucked up inside. I want out! I want out now! What happened was that these prisoners, they became more dehumanized. They became more agitated. And then the guards, on the other hand, became more sadistic. They lived up to the situation.

819 did a bad thing. Prisoner 819 did a bad thing. Prisoner 819 did a bad thing. Prisoner 819 did a bad thing. Prisoner 819 did a bad thing. How long did this study last? We were going to run it for two weeks, but we had to end it after only six days. So what happens after that? So what happened is that this became like a blockbuster moment in psychology.

The first reports of this experiment were published in New York Times Magazine. It was like a big moment in psychology. And like I said, you know, coming out of the context of like, where does evil come from? Like this is, you know, one generation after World War II. And like, you know, getting like a kind of satisfying answer in that, well, evil comes from circumstance. And this was the conclusion of this study that has been talked about for years, that if you put people in power and you strip other people of it, you get these inequalities and you get evil. This happens naturally. It's just a cruel

component of humanity that we fall into these roles. This conclusion has been really influential. It has worked its way into congressional testimony. It has been used to explain abuses in places like Abu Ghraib in Iraq. There's been documentaries, books, and this is like in every introductory psychology textbook. If you take it, Psych 101 College, you're going to learn about the Stanford Prison Experiment. So you're saying it became pretty much the best-known study in psychology. Yeah, it's the most famous study in psychology. It transcends psychology.

It becomes something you know just from being a human. Yeah, but the story Philip Zimbardo has been telling about this experiment for decades isn't the whole truth. Remember when I said that the key to this experiment was that no one was given instructions of what to do? This was the natural thing that happens to humans when some people are put in power and some people are put in not? Well, some tapes from the behind the scenes of the experiment really, really, really undercut that critical element. In the recording, you can hear the warden, who is a research assistant,

somebody working with Dr. Zimbardo, really chastising one of the guards for not being assertive enough. And if you listen to this tape, you know, it's not just like a one-off comment, like, oh yeah, you really weren't participating. It was, I believe the quote from the warden was... The guards have to know that every guard is going to be what we call a tough guard.

And the warden really pressed this. And then he went on to say... It's really important for the workings of the experiment. Because whether or not we can make this thing seem like a prison, which is the aim of the thing, depends largely on the guard's behavior. What was really happening here? If one guard was coerced into being meaner, more aggressive,

was this line that like, this whole experiment was naturalistic? You know, is this something that we should believe? Or is it something more like the experimenters created this like social pressure on the guards to push them to be meaner, to be more cruel? And then there's some other inconsistencies too. Like, there was this great expose Ben Bloom wrote about. He talked to one of the guards, and the guard said, I took the whole thing as an improv activity. Like, he was, like, acting the whole time.

He also talked to the prisoner that had that mental breakdown that I told you about, like, I had fire in my head. That prisoner has said that he was faking it, that he wanted to study for the GRE, and he was, like, kind of sick of the experiment and just wanted to get out. Overall, the main conclusion that we can take from these revelations is that it's not like these events didn't happen. It's not like there wasn't a little prison uprising. It's not like these guards didn't show acts of cruelty. It's that this assumption that they did it because

of a naturalistic human tendency that when there's an imbalance of power, you become cruel. Like that conclusion, you cannot draw that conclusion from this study. How has Zimbardo reacted to all this? Has he admitted that he was just telling a great story? I doubt it. So I called him up. Brian, Brian. I really pressed him on whether it's okay as a researcher for people to reevaluate your work. And he said, yes, of course it is. But then he kept calling them attacks. I think the ones against me, certainly by Ben Bloom, are malice. I mean, you don't say a study is a lie. But then he kept calling them attacks. I certainly see they qualify as attacks on my honesty, on my authenticity, and on the validity of the study.

I'm telling you, every fucking thing that Ben Bloom said is a lie. It's false. So this one audio clip of the guard being coached, he said it was just one time, just one guard. And Jaffe picks on this guy because he is doing nothing. He's sitting on the sideline, you know, doing nothing, watching. He's got to earn his keep as a guard. But even that is, you know, kind of evidence against his conclusion that, you know, this was naturalistic. If you have to like berate one guard for not doing things, like the message can be spread across to the other guards that it's important to the experiment that you you know

behave like in the way the experimenter wants you to okay let's say like regardless of whether guards are coached or not regardless of whether people okay you keep that wait ryan i'm gonna stop you no no no you're not gonna finish when you say guard it implies all guards and it was one guard and i don't if you i'm telling you if, if you say that again, I don't want to talk to you ever again. Because, you know, you're being thick-headed about this. He was really resistant to the academic criticism. Like, I asked him, like, you know, so what's the scientific value of the Stanford Prison Experiment? for prison experiment. It's a very powerful demonstration of a psychological phenomenon that has had relevance not only for Abu Ghraib, but for many other situations.

If you want to say, is it a scientifically valid conclusion? I'd say it doesn't have to be scientifically valid. It means it's a conclusion drawn from this powerful, unique demonstration. And that strikes me as, you know, it can be argued as a scientist, I think you need to be open to the reinterpretation of your conclusions. And it's hard, you know, because we put ourselves into our work and it's hard to separate yourself from your work sometimes. And that's important because psychology is going through a little painful period of introspection right now where a lot of these classic studies are being reinterpreted.

And how people react to those reinterpretations is really important. There are two ways to respond. And one is to move forward and to incorporate new evidence and to be collaborative. And the other way is to be, well, a little more stubborn. I will stand by that conclusion for the rest of my life, no matter what anyone says, regardless of what any critics say. That conclusion is a generic conclusion about this whole body of research that human behavior is more under the influence of social, situational, environmental variables

than we have thought of before, than we have considered before. How can you, Brian, not accept that. I know you guys only like one podcast today explained, but there's another one out there I want to tell you about. It's called Murderville from The Intercept. It's hosted by Liliana Segura and Jordan Smith. Together, these two got more than 30 years of experience investigating wrongful convictions, and they focus on what they think is one. A brutal murder that rocks a small town in Georgia that shocks everyone.

The cops pin it on the new guy in town. They throw him away in prison. But then another murder happens and another. In the end, you've got four bodies and two convictions. One guy in jail for a crime he likely did not commit. Murderville. Find it wherever you find your podcasts. So is the Stanford Prison Experiment the only study that's been thrown into question

lately, or is it part of a series? So, you might have heard of something called the replication crisis. Nope. No. Where have you been? This is like big news in social science. There's this other thing going on. Read Vox. We write about this a lot. For the past few years, psychologists and social scientists have realized that a lot of their

standard practices in collecting data and running experiments are actually recipes to yield false positives. So there's a lot of different stories in this replication crisis, but basically there's a movement now to go back to these classic studies and to find out like, okay, are they true or not? Do their conclusions still hold up? And in science, replication is a cornerstone. Like, if you do an experiment right here between the two of us and we find, like, I'm drinking a cup of coffee now and you're not and, you know. I got water. You got water. And the experiment finds that, like, my heart rate is going to be elevated because I have a cup of coffee. That should be replicable. That, like, every time I drink a cup

of coffee or you drink a cup of coffee, like your heart rate should elevate, and the control person, you know, their heart rate shouldn't be elevated. Okay. You have to repeat an observation multiple times to believe it. Got it. A common problem in the past has been that studies have been done on 50 college students like Harvard.

You know, that's not very representative. And also that small sample size, it turns out is actually a really good recipe to land on a false positive. Can you give me an example? Have you heard of ego depletion? Tell me more. In psychology, we use a term to describe how people don't always think through their decision making in a rational and linear way when placed under situations of stress. We call it ego or cognitive depletion. So ego depletion is this idea that willpower is finite. Like President Obama famously didn't choose his own suits because he was worried that he would like deplete his willpower store for the day to make decisions. Who chose his suits? I think he just had a few

suits and didn't make it complicated. Got it. So this is like one of those classic psychological studies that really have invaded pop culture and all that, you know, like our pop understanding of psychology. And it turns out like once you bring like very many labs to do a test on ego depletion, they find out they can't find it when you like have like really big numbers of participants, when you have these tests being done at multiple sites around the world, that if you find ego depletion in one study, you should find it in another and another and another if this is like a true thing about human psychology. And it turns out that's not looking great. So that's one idea. So we've got the Stanford Prison Experiment and ego depletion. What else is being called into question lately? Any other big ones? Something called the marshmallow test.

It's basically this test of delaying gratification. You put a marshmallow out in front of a little kid. You say, like, if you can wait 15 minutes, you'll get two. And if the kid takes the marshmallow early, it's a sign that they're not so good at sitting still. And what we found in many studies is that kids who are able to self-regulate to delay gratification by the time they're four or five or six years old have a much better chance of doing well at school, have a much better chance of thriving as adolescents and moving on in life. But recently there was a reevaluation of that much larger sample,

a much more diverse data set that they can control better for things like family background and cognitive ability. You find out that once you start controlling for things, that this effect kind of disappears. Interesting. The idea has been the story of the marshmallow test has been if you could just teach kids patience, they will become better adults. And these new results suggest like, no, you actually probably need to act further upstream.

So if you've been seeing just like an avalanche of stories on like, okay, ego depletion, the marshmallow effect, or something called the facial feedback hypothesis was just like, if you put a pencil in your mouth, it activates your smile muscles and you become happier. That doesn't look like it's real either. So a lot of these stories in this replication crisis, and very similar to the prison experiment, in that when you look more closely at data,

when you bring more rigorous methods to these questions, the story that we tell from the conclusion starts to look different. But to be clear here, a lot of these experiments you're talking about, the ego depletion, the marshmallow, it's sort of shoddy science, whereas the Stanford prison experiment seems a lot more like it was deception. Yes. The Stanford prison experiment is an example of what happens when you withhold information. But in some of these other examples, something similar is happening too. If you only publish The Stanford Prison Experiment is an example of what happens when you withhold information. But in some of these other examples, something similar is happening, too.

If you only publish results where, oh, we found this statistically significant result, and then you forget about all those times where it didn't happen, that is deception. And that used to be standard practice in a lot of social science, where you could just ignore results that didn't work. Or there's a lot of incentives in academia to just publish, publish, publish. And that's how you get jobs and that's how you get tenure and that's how you get grant money. And I think a lot of scientists realized they might have thought of it was like a lesser evil just to quietly push away things that didn't work and don't report them. But now there's a big movement. It's called open science as really trying to bring a lot more transparency and to make sure like

science has a stronger footing going forward. Why is all this happening right now? Why is it that science is being re-evaluated right now or psychology is being re-evaluated right now? Why is it that science is being re-evaluated right now or psychology is being re-evaluated right now? In 2015, the journal Science published this huge research effort. 270 scientists around the world tried to replicate 100 psychological experiments that were published in top journals, and only 40% of the studies held up. And how are scientists reacting to this, this upheaval? Some better than others. Like, like in my interview with Zimbardo, he called all these replication attempts attacks. Sounds like he's had some ego depletion. Yes, yes. And then there are others who basically publish like, this is the experimental plan I'm doing. And they're making their other data open

for people to evaluate. They're posting early versions of their articles so the public can weigh in on them before they get into the academic journals. So there's a lot of scientists that are really excited to be a part of this revival of psychology, to put it on a sure-fitting. Is psychology particularly at risk here because you're studying human behavior, human conditions, as opposed to like rocks, magma? You know, the problem with psychology is you're trying to understand the human mind from inside the human mind. And there's just like so many traps that get laid by that. Psychology is messy, but it's an important science. Hey, man, I didn't say it was. No, no, no. And it tends to be like a good gateway into science because as people, we're really interested in ourselves and psychology is the science of ourselves.

And I think to commend psychology and social science more broadly, they are out front on these issues. So replication is not only a problem in social science. And I think they've been particularly transparent about their efforts at reform. What about you as a journalist of science? Has this affected the way you think about scientific studies that you write about every week? Absolutely. We think about that a lot here at the science team at Vox. The newest study is not necessarily the truest. And I think what I've grown a lot more comfortable with is just when scientists don't know the answer to the question, that's fascinating. It's fascinating that we're still investigating the origins of human evil. I did a silly story a few years ago on like whether

rats have feelings and scientists have been trying to figure out if like rats have like empathetic emotions for decades and like we're getting there. But when you see like a breaking study in the news that, you know, like... The scientists now saying that... That's a good local newscaster impression. Like, if you're eating grapes, are going to lower your chances of diabetes, you know? Yeah. One study is never going to tell you a whole story. In science, like, the answer of,

I don't know, is more common than a lot of science journalists like to admit. And that's cool. Cool. Brian Resnick, thank you so much. Thank you. Brian, Brian. Resnick is a science reporter at Vox. I'm Sean Ramos from This Is Today Explained. Irene Noguchi is the executive producer of the show. Bridget McCarthy is the editor.

Noam Hasenfeld produces the show, and Luke Vanderplug produces too. Afim Shapiro is the engineer, and the replicated Breakmaster Cylinder makes music for us. You can follow along on Twitter at today underscore explained. Today Explained is produced in association with Stitcher, and we're part of the Vox Media Podcast Network. Thank you.