Press Releases Why staring out the window is good for the brain

Nick:
All right. Welcome, everybody. Thanks so much for coming tonight. Booth puts on a ton of programming. So coming to another talk from one of our faculty members may not be a huge deal to you, but it's a big deal to us tonight, at least at the Center for Decision Research, because this is the first of these that we've ever done. This is tonight launching a brand new talk series that we're calling Think Better. It's the first time we've done anything like this in the 41 years that the CDR has been in existence. So you're here to an inaugural event tonight. Thanks for coming. This series is really meant to share with the Booth community and with our friends beyond the Booth community as well, how behavioral science is being used in lots of ways is to improve people's lives at the societal level, at the organizational level, and at the individual level as well.

Nick:
Our field's findings are getting lots of attention. Research is exploding and our results need to be shared widely. And this is one of the avenues that we're using to share the research that we're doing with our broader community. Tonight's event is the first in our series. This year, we're going to have two more. One in February, where David Yokum from the lab at D.C. He worked on the behavioral insights team in the White House for a while, in the Obama White House for a while. He's coming on February 13th. We're also having Angela Duckworth, author of Grit, a professor from University of Pennsylvania in on April 3rd. And we'll send out more information about those events as the date nears.

Nick:
Listening to us, listening to researchers talk about their results is one way to learn about behavioral science, but we have other ways as well through the CDR. We operate the PIMCO decision research labs, where we run experiments day in and day out. And if you would like to say, sign up to be a Guinea pig in one of our experiments and learn something about yourself and what we're doing, you can do so at this short link up here. We have labs in the Harper Center, we have a lab down here downtown over on Adams and Clark. We were on experiments in Millennium park pop-up labs all throughout the city. More pressing though for our learning tonight is not participating in experiments, but listening to somebody super interesting, talk about some experiments that he has done. I couldn't be more excited to kick off this series tonight then by having Sendhil here with us. Sendhil is one of our newest faculty members at Booth, he arrived just this summer.

Nick:
He was given one of only nine university professorships, which means that his professorship is officially not housed within the Booth school, but rather he was hired by the university at large. He an academic appointment that spans the entire university. However, he teaches in the business school and his office is on the fifth floor of the Harper Center. So we know where he really belongs, that is with us. Sendhil's accomplishments are nothing short of extraordinary. It makes you wonder what you do with all your time when you see what he's been doing with his. Sendhil was given a MacArthur Foundation Fellowship shortly out of graduate school. This is what's often affectionately known as a genius award. His potential was recognized very early on in his career. Since then, he's been designated a young global leader by the World Economic Forum, a top 100 thinker by the Foreign Policy Magazine, and included in the smart list of 50 people who will change the world by Wired Magazine.

Nick:
Just a couple of weeks ago, Sendhil was also awarded the 2018 Infosys Prize in social sciences for elevating the prestige of scientific research in India and inspiring young Indians to choose a vocation in scientific research. So get ready for the stampede of people following Sendhil there. He's co-founded nonprofit organization, including Ideas42, which applies behavioral science to positively change lives, and also co-founded the Poverty Action Lab, which is housed at MIT right now, a center to promote the use of randomized control trials in development. Sendhil's a very prolific writer, not just in academic journals, which is what he's going to be talking about today, but he also does a lot of writing for broader audience as well. He's a regular contributor to the New York Times and the column Economic Scene. He also has co-authored the book, Scarcity: Why Having Too Little Means So Much. On a personal note, I just want to add how excited I am to have Sendhil on our faculty here. He is without question, the most energizing person that I've been around.

Nick:
He spent last year, maybe two years ago now, on sabbatical here with us. We got together about once a month for lunch. And every month, that was pretty consistently the highlight of my month, was that lunch. You cannot walk out of a conversation without being excited about the research that you're doing, about the ideas that he is thinking about and discussing without just feeling more excited about whatever you happen to be working on. So it's a great treat to have Sendhil here tonight. Very happy to have you here. Most important, I'm very happy to have you here at the University of Chicago. This is his first event to a university, speaking to a University of Chicago crowd. So let's give him a very warm welcome here to the University of Chicago.

Sendhil:
All right. Thank you. I have to say, having a poster with my photo up with the phrase, Think Better, all around the university evoked a set of responses, including from my niece who's an undergrad. She would send me pictures every time she saw one. Do I suppose that there was an unforced error by calling this talk, The Self Driving Mind? So I really feel like I caused this trouble myself. I'm going to start on a relatively serious note. I want to start talking about crime. It's very hard to live in Chicago without confronting the problems of crime, but in fact, most of us don't really confront the problems of crime. Crime is a very, very, very regressive problem. It affects poor people far more than it affects well-off people. Its consequences are dire.

Sendhil:
I'll give an example. On June 2nd, 2012, a group of teens got into it and Calvin Carter, age 17, who's up here, pulled the gun and shot Jamal Lockett, age 16. This type of incident happens all the time. Of course, Calvin goes to jail, Jamal is dead, and many statistics like this all the time. So you can imagine, when faced with things like this, one of the things as a human being, but as a social scientist, as a behavioral scientist, you want to know is what can we do to effect this problem? Young men, and it's disproportionately men, having their lives taken away, having their futures taken away. It's heartbreaking. Unfortunately, if you look at the social science evidence, you look at the careful studies that are done, we just don't have that many great solutions. That's just the reality. Programs that look pretty good, when we assess them through a lens of a careful randomized controlled trial, inevitably they don't do very much.

Sendhil:
Which is why, about four or five years ago, I was very interested in this program, Becoming A Man, because this program purported to do something. And I have to be honest, you have to be super skeptical about this program. Not just because none of them work, but because this one is extremely improbable. Here's what this program is. It's about 12 weeks long, maybe 24 weeks. We can talk about details. It's of that order of magnitude. Kids get together in various groups and effectively there's some talking that happens. I'm not going to get into details, but that's about it. From a behavioral science perspective, programs of this variety really have no chance of working. You can't go to some kid growing up in these neighborhoods, talk to them and then solve some problem. It's not a problem you've solved by just lecturing somebody or having them talk or whatever. So in general, I'm very skeptical of these programs. Then this one seemed to be in the highest end of skepticism. When you looked at the randomized control trial, the careful evaluation, some kids get this program, some kids don't get this program. And you start to see the effects, they were immense. Here's an example.

Sendhil:
Between the treatment group and the control group, total arrests over the next two years fell by quite a bit. Violent arrests fell by quite a bit. And you might think, well, this is classic cherry picking. The program just happened to be done in neighborhoods that were relatively easy. In fact, here is a map of the Chicago area colored by the homicide rate. And the sites where the programs were run, you can see were in some of the worst neighborhoods in all of Chicago. You could say, okay, well maybe you went to the worst neighborhoods, but a classic trick that makes programs look much better than they are, is they cherry pick within the program. So the kids that they worked with in this case, you might think are particularly good. This is the number, 1.73. Anyone want to guess what this number is? This is the average GPA of the kids in the program.

Sendhil:
There's another number, 35%. This is a percent of kids that have been arrested by the time they entered the program. This is a highly improbable program from the get-go, it's improbable based on who it works with. Here's something even more improbable. A second trial was run. Again, this time, not in a school, not with kids in school, but with kids in juvenile detention in Cook County. Any of you here have been to Cook County Juvenile detention, like visited? You don't end up in Cook County Juvenile detention, unless you've done something very, very, very wrong. So it houses some quite, or at least accused of that, it houses potentially some of the most problematic cases that you'll deal with. The program was run again, this time not even by the trained counselors, but by the prison guards. And here were the effects. This is thousands of people in treatment. Thousands of people in control randomly assigned over the next 14 months from when the kids released.

Sendhil:
Here's the treatment. Those who received the program, here's the control. 15% reduction in rearrest rates. It's a crazy treatment effect. All of this raises the question, what exactly was in this program? What were the kids being told? Perhaps what typifies the program is an exercise called the fist. This is actually the first exercise they do with the kids. So when they're brought in, picture yourself in the school context, the kids are told, "Well, you don't get this class in third period and said you're going to do this thing." They come in and they're lined up in groups of, broken up into pairs and they're in two lines, and each kid in this line is given a ball. The kid across from him is told, "Your job is to get the ball from this kid." You can imagine it's a free period, and they're told, "Do whatever you want." The kids just go at it. It is a melee, they're grabbing and pulling and they're shoving. It's just wild.

Sendhil:
At the end of all of this, you stopped them after 10 minutes, very few kids got the ball. It's very hard to get a ball from someone who doesn't want to give it to you. But of course, that's not the point of this exercise. The point of this exercise is, they go to the kids and they say, "Okay, well, what did you do to try to get the ball?" And they're all very proud of their strategies that they attempted, whatever they may be pulling this. "I thought if I could pin their hand." Then they stop and say to the kids, "Okay, that's great. How many of you, by show of hands, asked for the ball?" None of them. So then they say, "Well, why didn't you ask?" They say, "Oh, well, he never would have given it to me." So then you turn to the other kid and you say, "Okay, what if you had been asked, would you have given the ball?" And the kid says, "Of course I would have given the ball. It's just a (nothing) ball."

Sendhil:
The point of this exercise is something actually quite profound that extends well past the ball and the fist. It's the point that is reinforced throughout the program for these 12 weeks. The problem here is in the way the kids were thinking about the problem, or rather not about what they were thinking but about what they weren't thinking. They were thinking pretty automatically. They entered this situation. And in this situation, they were told to get the ball. Something immediately came to their mind and they just followed it. They didn't step back and they didn't say, wait, what are all my options? Is this a good option? What are my other options? Should I generate more options? They thought automatically, rather than reflecting on the problem and stepping back to reflect.

Sendhil:
Let me show you one little piece of evidence that we collected. We took the kids in the program and we put them in an artificial little experimental game. I think this is one of the powers of behavioral science is that you can look at real outcomes in the world that you might care about, but you can also look at real outcomes in the lab that you might say, distills the phenomenon. So in this game, the kids were brought in and they played a little game where somebody offered them money and they would either view it as fair or not fair. If they viewed it as not fair, they got to punish the person back. What's interesting is not, did they punish the person back when they got an unfair offer? What was interesting was how long did they take to decide would they punish back? And you'll notice the kids in the program responded much slower than the kids not in the program.

Sendhil:
Faced with an annoying situation, the program had gotten these kids to reflect for a couple of more seconds. That's pretty amazing, but you might wonder, great, so they're reflecting about this, what difference does it make? Well, it makes all the difference. Let me go back to Jamal Lockett and Calvin Carter. What really happened between Calvin and Jamal is actually quite tragic. All of the problem arose over a stolen bike or a purportedly stolen bike. The two groups were jawing at each other, "That bike is mine, this bike is not mine." The kind of the situation had gotten pretty tense, and then it started to calm down and the groups were walking away from each other. And in that moment, Calvin just reached down and pulled out a gun and just shot Jamal. It was actually just about 10 seconds of Calvin's life. And one of the prison guards in Cook County puts it well. He says, "With most of the kids in here, if their life were a film and I could just go and take 10 seconds and edited out, their entire life would be different."

Sendhil:
Because in that moment, in that high pressure stress situation, something happened in Calvin's mind where he had an automatic thing. Maybe it was a fear. Maybe he felt like, wait, if I don't show that I'm willing to stand up, something will happen to me and just [inaudible 00:16:05]. And many of these kids' lives are in fact, about the consequences of that kind of automaticity. And that's the first point I want to make. That there are these enormous consequences that happen to the level of being able to make an outrageous claim that if we can get people to think a little less automatically as this program has done, you can actually change behaviors that look like they came from something totally different. I think if we were to have a conference about what are the sources of crime in Chicago, I don't think we would have gotten in the top 50 answers, Oh, automatic thinking. That's the problem. That's part one.

Sendhil:
That's the self-driving mind that I want to establish. I want to talk about that same phenomenon though, in another context. And before doing that, I want to do two little experiments on you guys. The first experiment is pretty easy. It involves just you naming the color of the object on the screen. So say it out loud and say it as quickly as you can. Ready?

Audience:
Black, red, green, yellow, blue.

Sendhil:
In a way, you had done something like Calvin Carter did in that moment, there was something automatic that you did. And this is one point I want to make, which is, the problems I'm going to describe, and I'm going to end on afflict all of us. They simply have different consequences for Calvin than they do for me. But what this illustrates is something else. When the mind behaves automatically, it's interesting to know, where does it go? What are the kinds of automatic thoughts that you tend to have? In this example, this is called a Stroop Test, we controlled the thought through, I guess you could call it a cheap trick, but it's not that cheap of a trick. It illustrates something about which processes in your mind happen pretty fast. So what I want to think about is a little bit more about that question, where does your mind go?

Sendhil:
To do that, I hope you guys have pen and paper, Some of you. the ones that people have pen and paper, I'll take pencil as well, I'm going to do a little test. I know you're thinking, it is 6:20 PM on a Wednesday night, The last thing I need is a test, but there we are. Here's what I'm going to do as you get your pencil and paper, I'm going to read a list of words. Do not write down what I'm reading. When I'm done reading, write down every word that you remember. Does that make sense, everybody? I probably don't need to tell you because you're not Harvard undergrads, but don't cheat. Don't look at your neighbor. Okay. Everyone ready? So don't write down what I say. When I finish, I'll say finish and then just write down every word you remember. Here we go. Bed, rest, awake, nap, dream, wake, doze, snore, slumber, blanket, snooze, tired. Okay, now write down all the words you remember. (Silence). Okay, it looks like most people are done. There's still more coming out. Is that what you guys are saying?


Sendhil:
I can't be giving hints. Okay. So let's say we'll call it close enough. How many of you got the word bed? That was good. Okay. How many of you got the word tired? That was pretty good. How many of you who got the word dream? Some of these were words in the middle. I think people are very good at getting the words at the beginning and at the end. How many of you got the word slumber? Pretty good. How many of you got the word sleep? Oh, okay. A lot of you got that. What's really impressive is that I'd say about the first few words people got at about a 60%, 70% rate. This, I'd say people got about 30%. Sleep, people got about at a 50%, which is great. What's really good is the bonus points you guys earned? Because I never said the word sleep. So really, extra points for all of you who managed to create a memory out of nothing. Anyone who said sleep want to explain why they said sleep?

Sendhil:
It's like that thing where you say to a friend, "Hey, you remember that party you were at?" They're like, "I wasn't at that party." "No, you remember that?" "I wasn't at that party." They should have been at that party like everybody else. Why weren't they? It's not your fault. It's my fault. I should've put sleep on that list because it belong there. Didn't it? What's interesting is these types of DRM tests are fascinating because they tell us something about the architecture of the associations we have. Your mind automatically went to the word sleep because it's in the associate of network. These type of tests help us reveal knowing that sleep is in the architecture of the network of these words is hardly interesting. It's amazing as a demonstration of how faulty memory is, which I hope at least you've taken away one thing. When you say, I definitely remember blank, I hope you now walk away with a feeling that doesn't mean much. You definitely do remember it. That's the only thing that it means.

Sendhil:
But they also reveal something else, which is, they can also be used to understand what else is in the architectural words. I'm going to give you another list much like the one I just showed you. Phone, gas, grocery, this list of words. If I were to do this list with you, there's no fake word that would pop out with most of you. But there is a group of people for whom there is a word in the middle of that, that pops out at the same rate roughly as sleep pops out for you. Money, that's right. So if you look at low-income people, they're very similar in hearing the word sleep in the other network. Rich people look at this list and say, I don't know. I remember these words. Poor people, having heard grocery or cash or loan, they also remember having heard the word money.

Sendhil:
That's pretty profound if you think about it because it means being poor isn't just a physical state. It's not just a monetary state. It's also a cognitive state. It's a state in which things in the environment now automatically make you think about money in a way that wouldn't for other people. I'll give you an example. Imagine you go to the doctor and you have people read the story. So you've been feeling sick lately and finally decided to go see a doctor about it. The doctor explains you have a serious condition, require immediate attention. The good news, however, is you're virtually guaranteed to make a full recovery. The doctor writes several prescriptions about why you need to make several appointments. If you ask people, what would be on your mind or how would you feel as you've heard this news? What are three things that you would think or feel? Oh, what's going to happen to my wife or son, my co-worker? The co-worker come in. It's very weird, really. I'm scared, I'm afraid, I'm worried. Relief, hope, joy, cost.

Sendhil:
What's interesting is when you hear all these words and you hear the story, I think the difference between the poor and the rich becomes very stark. Now we're not just talking about memory tests, this critical life-threatening condition for which there is a solution and what you're told, take these drugs, not take these tests. When a poor person is listening to that, they're much more likely to be thinking, what will this cost me? Can I afford these drugs? How much [inaudible 00:24:45]? Now think of what that means. That means that whenever you're wandering around the world that you're wandering around, there's an extra set of automatic thought processes. Your mind is no longer necessarily going to the thing that you want. I better listen to what the doctor is telling me about how often I should take these drugs. It is also going of its own volition to these other things.

Sendhil:
Imagine you had a computer processor where there was a bunch of background processes that were running. What would that do to the video you're trying to watch? It would make the entire processing speed of the computer, the effective speed that you have available to you much smaller. So if you take this perspective, it actually implies something pretty stark. It implies that being poor, the state of being poor takes anybody's mind and makes the available capacity effectively lower because the mind is automatically going to these other things. I won't go through this in detail, but this is one of the studies we did in this scarcity work. For example, we went to India. I don't know how many of you have ever seen what sugarcane actually looks like. This is what sugarcane looks like, and it's delicious. Sugarcane is interesting because sugarcane farmers get their crop in India, at least it's longer in India, once a year roughly.

Sendhil:
Getting your crop once a year has one amazing financial consequence. Imagine you got your entire salary once a year. That's what it's like to be a sugarcane farmer. When I was a grad student, I got my salary once a semester. And I still remember, you get paid in January and you get paid in September. And I remember December was a very tough month because I did not. But now that was actually me saying it actually from a relatively comfortable position, my dorm room was paid for and everything... So it was a tough month in a very narrow sense. For the sugarcane farmers, it's their entire income and it is genuinely a tough month. Rates of starvation are very, very high just a month before they get paid, and then the times were good.

Sendhil:
So besides the human tragedy, there's also a behavioral science element of this. The same person is poor one month and then pretty well off the month after. So if everything I just told you is true, it has one radical implication. That the month after harvest isn't just the time when people were financially rich, they should also be cognitively rich. They should literally be smarter and more capable in the month after harvest. That's what we did. We can get into some details. We ran tests like this. This is called the Raven's Matrix Test. It's one way of testing fluid intelligence. It's like, you can do this without knowing any literacy. Here's an image, which of these images fits in there? I never do well at this, so I have to remind myself, I think it's four, right? It's four, thank you to whoever said five. Now I feel better. It's four. So we ran tests like this. We also ran tests like the Stroop Test, where we just said, how much more are they likely to automatically just make a mistake and just say blue when they shouldn't?

Sendhil:
What you find is pre-harvest, people just get fewer of these tests right, make mistakes like saying blue much more, people are just much less capable. These numbers won't mean much to you. So let me put them in terms that will make sense. People will run this exact battery of tests in other contexts. One context they run them is going to show you the full [inaudible 00:28:41] of the exciting [inaudible 00:28:42] of behavioral science. I like this field the most. It's called sleep science. They bring undergraduates into the lab. It's usually undergrads, but I imagine, sometimes other people volunteer. And they say, "We're just going to do sleep experiments on you." Half the subjects are brought in. And they said, "Thanks for coming in. Here's a nice climate controlled dark room for you to sleep in. Goodnight. Thank you." The other half were told, "No sleep for you tonight and lest you should try to sleep, we have a graduate student who is going to sit next to you and watch you." Which is a little creepy.

Sendhil:
Now the next morning, 9:00 AM, one group very well slept, the other group, pretty cranky and not very well slept. They then run these exact same batteries on them. And the big implication, if you take nothing else from this talk, you should take this from this talk. Don't go without a night of sleep. You are a blithering idiot the next morning. It is about 13 IQ points, 14 IQ point loss. It's a huge loss in cognitive capacity. And it's a well-known thing. This is why driving while sleepy is much worse than driving while drunk, for example. Yeah, it's really bad. Don't drive if you've been pulling all-nighter, you also have a glass of wine, but we'll discuss that later. But I'm telling you this because those effects give us a way to make sense of this effects. And it turns out, being poor the month before harvest is about three quarters of the effect of having pulled an all-nighter. Of course, it's every single day. So it's as if the poor are walking around sleepy, really sleepy every single day.

Sendhil:
Let me say like I said with the crime thing, a similar theme here, there's a literature, for example, that talks about how the poor are not good parents and there's documented evidence. Or they don't read to their children. They don't get their 10 million words a day, blah, blah, blah, should we do parenting education? We could have a whole conference bringing people in on how to make the poor better parents. I don't think in that conference, coming to the top of the list, would be, Oh, well obviously they've just got a lot on their mind, but you would be a worst parent if you were pulling an all-nighter. Not just the worst parent, think of every other behavior that you attribute to the poor and say, Oh, well, what's the problem? Well, there is a problem here. This is the problem.

Sendhil:
Okay. So I just want to say in part two, that self-driving mind doesn't always go where you want or where we want. And that can help us make sense of phenomena like poverty. Let me now turn things inward, and this is where I'll stop. When I was an undergraduate, I would take these long bus rides. There was a woman I was interested in. She lived in Rochester, New York. I was in Ithaca, New York. So I get into this, it was winter. And I have to say, especially at that age, the first such bus ride, it's like you just feel like such a romantic on taking this bus. The snow is on the ground. It's amazing. And it's beautiful. It's really beautiful. It's great. The third bus ride was painful. There's nothing to do. I can't read because I get sick. I just sat there looking out the window. It was so boring. So boring.

Sendhil:
What's funny though, is that I'm never bored anymore. It's great. If I did that bus right now, I'd be like, I'd pull out my phone and listen to some podcasts, I'd probably go to, I don't know that I go to Pinterest. I don't know why that's there, but let's not get into me, it's not about me. But in a weird way, is that good? My self-driving mind, even when left a minute by itself, it says, I wonder if I've gotten an email. I wonder if there's anything new on Twitter. My mind keeps driving me to these things, so much so that I don't know that I focus on anything anymore. For example, even if I get to Twitter and I link on an article, I don't know how far down do I read before saying, I wonder if there's another article that I could read two sentences of. And I'm not alone. If you look at the data on content scroll through, people basically, I'm not talking about reading. Just how far do you even scroll down? People don't get very far. In fact, when you look at the correlation between how far did they get and then they tweeted, it's basically zero.

Sendhil:
I will say, Nick mentioned I write for the New York Times, the modal email I get after New York Times column seems to be from a person who has read the title. And who has strong opinions on what I am saying. And the irony is I don't write the titles. You write the articles, some editor pick the title. So it's very easy to get these emails back because I'm like, nothing to do with me. Here's a great thing. You guys ever highlight things in Kindle? This is a study that looked at where in books, in Kindle books are those highlights? Because that's a good way to see how far people make it. So basically, the half-life is about 20% of the introduction. If you're lucky, amongst people who highlighted, these are the diligent readers who are like, I'm really...

Sendhil:
There's almost no highlights near the end of it anyway. All this is by way of saying all of these technologies, if we take the idea of the self-driving mind and we say it doesn't just tell us about poverty or crime, but tell us something about us. Yes, in the 21st century, what it tells us about us is that we're surrounded by these technologies that are making us dumber effectively, completely unable to focus. And that's why there are books like The Shallows. One of my favorite recent studies. I don't know if this will replicate, but it's just such a great metaphor. It feels right. We'll find out if it is right, but in this study they did something really beautiful. I just think this is a great study. They had participants come in and they were randomly assigned to one of three conditions. They would have their cell phone either on the desk in front of them while they were doing an exercise, in the bag right next to them, or they put it in a totally different room.

Sendhil:
One of the exercises they did was exactly the Ravens Matrix type tests that we talked about. And you'll notice, people are much dumber at this task when the cell phone is right in front of them. And you can completely intuit this next time you're at a meeting or you're trying to do something. When the phone is in front of you, just start trying to track the number of times your mind says, hmm, I wonder if I should see if there's something. Well, I can't. I'm at a meeting. I shouldn't do that. Just that thought is interrupted you and your mind is driven you there. And now you're out of the meeting and you have to go back. So all this is by way of saying, if I were to apply the same principle to us, one of the biggest problems we see is that technology is making this problem really, really, really bad.

Sendhil:
Yes, I'm distractable, but wouldn't it be nice to have these boring long bus rides again? It's no coincidence that lots of people say things like, my most productive time is in the shower. It's the one-time... I mean, they're going to solve that problem soon. Don't worry. But until they find a way to solve that problem. But let me conclude on a very positive note. I think the fact that technologies are driving the mind in the wrong direction is a local problem. I think we're on the verge of quite a big change. And that's where I'll end. Many of you have this in your car, a blind spot detector. Do you guys have this? It's actually quite amazing as a cognitive prosthetic. It tells you if there's a car in your blind spot. It's beautifully located because if you're about to go to the right and you even look in the rear view mirror, it'll show you, in some cars there's like feedback, haptic feedback that like your steering wheel shakes a little bit. What a great reduction in car accidents?

Sendhil:
It's because here the technology is working to drive your mind to exactly the right thing. It's actually getting you to notice at a time when you should notice. I think in the cognitive world, things outside of driving, we're going to start seeing those types of technologies for our own cognition. Let me give you a tiny little experiment I did on myself to make sense of it. What I've been worried a lot about is my web browsing. How many of you have ever looked at your history file? You guys ever looked at it? Oh, you have to. Pick a day. Even two weeks ago. Pick a random time. Just start and just follow the trail of your mind. It is the ravings of a mad man. It's amazing. Oh, look, he was interested in buying a hot air fryer. But wait, why suddenly is he interested? What is going on? And Oh, now he's back to the hot air fryer. What? Who is this guy? This guy is me.

Sendhil:
Now, the problem is not just this random walk through my cognitive structure that I'm doing. The problem, what do I have on there? All right. Well, let's not even look at ESPN. I should check what's here. Oh, my goodness, my goodness. All right. Let's move on. One thing I learned from my browser history is that there are sites I go to a lot, lot, lot, lot, lot. Liberty Ballers is one such site. This is a 76ers blog, which is a basketball team. Now, we can already build apparatuses to try and help me with this problem when my mind gets distracted. We could say, well, they have these things like blockers. It just blocks sites, but I don't want to never go. They have these things that prevent you from using the web at certain times. And maybe those were fun. They are very crude.

Sendhil:
The problem with the existing technologies in this space is that they're not designed for me. They're just generic. Oh, let me just make sure you don't use the web often enough. But the browser history taught me though, was it's like, well, actually my data set in my browser history is actually enormous. I've served the webinar that there is an entire data set that is the Sendhil browser dataset. I can learn a lot from that dataset just about me. For example, if I view Liberty Ballers as a bad place to be, I might be curious what gets me there in this wandering mind? What gets me thinking about it?

Sendhil:
Actually, it's Google News, which is something I never thought of as a bad place to go, but it turns out I go to Google News, I see something that makes me think about. There we are, looking at. And it turns out that many of the things that I don't want to do, there are gateways from Google News because something on Google News, it's such a rich cornucopia of stuff. Something there think of, Oh, this I should go and check. And of course, the problem is there's an arrow back. But it's interesting that these technologies, while they are the source of the problem, rapid one layer back, and it's now helping me identify and understand myself even better. And in a way, there's data about me, me specifically, I don't know mean data about me as a representative, everywhere. Think about it. Google Maps. I don't know what that thing in the middle is, but... Ways, all of these technologies, you can go to Google Maps and see your place's history.

Sendhil:
It's pretty amazing. Somebody has been recording your life for you. You have a diary. It's actually just like a real remarkable walk through your past. All the articles that you've read. The New York Times doesn't just know what articles are the most read. They know which articles you have read a lot. They also know which articles you started but don't finish. They could build for you a thing that says, good that you clicked on this article. This is the kind of article you don't finish though. You seem to bookmarked it because you seem awfully interested, but you don't come back. What's going on there. That would be a very useful prompt to remind me at that moment. Oh yeah, this is broccoli. I should eat my broccoli every once in a while. This article is my broccoli. There's data like this on Uber and Lyft. There's data on what I buy, there's data on all sorts of things.

Sendhil:
In fact, what's amazing about all of this is that the sum total of the data about me is at this point, enormous. In fact, if you look at my inbox, it's about, just my inbox, well, my entire mail folder, inbox, sent mail, it's bigger than the Sloan Sky Survey, which was for a long time, all of the data we had about the known universe based on all of the satellite imagery and images we had in the known universe, just my inbox. It was pretty cool. So there's a lot to learn about me. And I think that is probably where this technology is heading because not only is it the existing data, soon all of the wearables that you have, all of these things, why is so much data so good? Because suddenly, when your mind starts to go into different places, when your mind starts to wander, we now have the ability to produce the blind spot detector for you in a very highly idealized type situation.

Sendhil:
Maybe you've learned your problem is you don't ask enough questions in a meeting. It's not very far from when your watch will ping you and say, there's a meeting going on. You seem not to have asked any questions. Maybe your problem is that you don't want to speak in a condescending tone to your spouse, but you do it. We're not far from having your watch ping you and say, Hey, remember that condescending tone thing we talked about? I think we're actually about to enter a pretty remarkable golden age where thinking about how these technologies work will allow us to drive the mind into better and better places and allow us to create a much better cognitive architecture.

Sendhil:
If I were to wrap all the way back, I think what I would say is when I look at programs like BAM, I think about the fact of how amazing it is that such programs can work when they're delivered in such a crude way. All these kids got was a little bit of conversation, 12 to 24 times, and then they're left on their own. Imagine how different those programs would be if they were embodied in a watch that says, Hey, maybe this is a good time to step back and remember the fist? All right, let me stop now. Thank you. If there are any questions, I'm happy to take a couple of questions and also others. Yeah.

Audience:
Do you have a microphone?

Sendhil:
Oh, there's a mic coming. Sorry, that was my bad. But that was a good loud voice.

Audience:
Oh, I have a good loud voice.

Sendhil:
That's awesome.

Audience:
If Google News makes money by cycling you between Liberty Ballers and Google News and whatever else they want to sell you, what are the incentives to provide these broccoli type apps, which are good for you, but perhaps not good for the people who make the platforms?

Sendhil:
I'll confess something, which is I chose to live in Hyde Park because I didn't want to commute. And that was a great choice, I think. But I find myself almost commuting every day up to the Sweetgreen on Fulton Market, and Sweetgreen is an amazing business model if you think about it. They're selling these things at $12, $15. Why am I talking about Sweetgreen? I haven't fully lost it. Let me explain why. In the 1960s and the 1970s and the 1980s, I think you would have said the business model for food is becoming very clear. Make it fatty, make it sweet, people will buy it. But over time, another thing became clear. People were slowly starting to become better consumers of food. Once the consumer demand changes, suddenly there's huge market opportunities that have started for places like Sweetgreen.

Sendhil:
I don't think we're fully there, but if you go even into very fast food chains like McDonald's, they're starting to say, hmm, we need to sell the healthy option because a lot of people want to come in for that option. One way I think about it is that even though the human species has been eating for a long time, for its entirety of its existence by definition, it actually didn't quite know how to eat. And it didn't know how to eat in an environment where food was so plentiful. So starting in forties, fifties, we as a society have actually had to learn how to eat. It's actually weird where we were learning for the first time, when the constraint is not calories and availability, how should we eat? I grew up in India and in the village we grew up and for us it's look, if there was food, you ate it. That was the right way to eat.

Sendhil:
That is not the right way for me to eat now. That social, cultural learning process of how to eat is same thing is happening with how we consume information and how we consume these technologies. We are living in an abundance of intellectual stimulation of information, of distractions, of hedonic stimulation of a kind we never would have had even 30 years ago. I have to work hard to have nothing to do. But I think we're going to have to learn how to consume this media as well. Just like as with eating, I find myself not very good at consuming these hedonic things. Even with all of these things, I don't think I'm that happy. I have reader's block, for example, I can't read a novel anymore because I keep getting distracted. So I think just like we have to learn about dieting and learn about eating well, I think we're going to learn how to consume. But I think once consumer demand starts, then companies will start selling us these products.

Sendhil:
At this point, it's almost like consumer demand has to change at some basic level and we have to get good at saying, Oh yeah, I want this thing to provides this. Hopefully it will change. I would think I've gotten to that point. I think a lot of us probably have at least gotten to that point where we would be like, Oh, a phone that actually produces less distractions, that's interesting. I'll give that a try or at least see if a friend gives it a try. Anyway, long answer. You also now know where to find me at that Sweetgreen on Fulton. Other questions? Yeah.

Audience:
In what type of cases do you think automatic responses are beneficial or preferable?

Sendhil:
In what type of cases?

Audience:
Do you think automatic responses are preferable or beneficial?

Sendhil:
Oh, in which kind of cases are automatic responses beneficial or preferential? I think in almost all of them. I think the power of the automatic system is that you really should be behaving automatically most of the time. The problems arise when your automatic response is in congruent with the situation you face. So with the kids on the South side in BAM, the problem they face is that they face a situation where one automatic response is not going to work in both situations. If they're on the street, the idea that, Hey, I shouldn't stand down when somebody does something, that works a lot, but there's a few situations where that is not the right thing to do. So I think the problem of automaticity is actually not that it is mostly wrong, is actually that it's mostly right, but there's just a few situations where it can be very wrong. Does that...?

Audience:
Yes.

Sendhil:
Yeah. Oh, wait.

Audience:
One could argue that there's a lot of information out there about you, but you're not really able to access it at the moment. And even if you were able to access it, it would be hard to analyze it as an individual. So in your view, how can we democratize both the access to the information that corporations have on you or technology companies have on you, and the analysis of that information for your benefit?

Sendhil:
That's a great question. Yeah. I think that there are now a ton of laws that have already passed in Europe, for example, which gives you a right to a bunch of your data. I think that right now, we're just in this five-year period before this stuff takes off, because there's enough, to take Google for example, you can download every bit of information Google has on you. Take your email. That is a rich Corpus of past experience. You can just download all of that. The tools to analyze it, and you really are a data set of one. So it's not that hard to build a thing. In a way, I think where we are right now is there are cases, many cases where companies are holding data that you can't get about you, but there are so many cases where you can get it.

Sendhil:
It's just right there. That I think what we're just as a few entrepreneurs to build those things that say, Hey, here, let me show you the times of day when you write the rudest emails, aren't you curious? Let me tell you the person to whom you're disproportionately Curt. Might be interesting. So I think there's just some entrepreneurial energy at this point to produce that type of stuff. There's a colleague of mine who she built this prototype of an app where she took people's emails and it said, "Here are your average gender bias and how you communicate with men versus women. You can even pick subgroups of men and women in your company." Let me tell you, it wasn't that hard of a build because all the APIs are there with email. So I actually think this is a situation where I wouldn't think of structural blockages. I think it's just a matter of entrepreneurial energy to get in and take advantage of the huge opportunities already there. Yeah.

Audience:
Thank you.

Sendhil:
Last question. I think, there.

Audience:
I have a question going back to the beginning of the talk. So we did those associative diagrams and we saw how poor people's associations might differ from the average person. And then we talked about how their cognitive scores, when the poor people are below. I'm wondering how you create the connection between, is the idea that they're continuously thinking about money and therefore they're not as capable of performing cognitive tasks? And then another question on top of that, how do you know that these are linked and it's not like malnutrition or some other external factor driving their cognitive performance?

Sendhil:
Yeah, let me take the second one. I think I went through that study pretty fast. I think part of in those studies, at least I would say, starts to produce some suggestive evidence in that direction. Let's say in the sugarcane thing, I think that's the fun part of research. You look for lots of these tricks. So for example, in the sugarcane thing, so here's this group of people. They're getting paid out. And here's this other group of people who are neighbors right next to them. Those are people who are, I should say one fun fact, weird fact about sugarcane, even though people harvest every year, it's not the same month for everybody because the mill can only process. So this will be a rich month for you and a poor month for your neighbor.

Sendhil:
So you can really take control for all of that. You can control for nutrition by finding the slightly richer sugarcane farmers who are not changing how many calories they consume. So a lot of the work, and that's really what most behavioral science, social science economics is about, is about controlling for those compounds. I think we've tried to control for that. I think what I didn't get through is the narrative of how the mechanism operated. I think the hypothesis of how the mechanism operates is much like having a constant distracting thought.

Sendhil:
So if you're sitting here in this audience, listening to this talk, putting aside my speaking skills, the hope would be that you'd be immersed. But if you knew you had a bill due tomorrow, well, one way you can introspect about that is if you knew you had a paper due tomorrow, it would be a little hard for you to be fully immersed because during parts of the talk, a thought would bubble to your mind, man, will I be able to get this done on time? Have I done everything? So that constant interruption, I think, serves as the same. Does that...?

Audience:
That's the hypothesis?

Sendhil:
That's the hypothesis. That's what I think some of the evidence shows, but hopefully there's more evidence to test or reject that. Thank you. Thanks, Nick.

Nick:
Thanks Sendhil. Thank you so much.