Living on Autopilot

-Welcome to a scientific revolution -- one that's not transforming how we understand distant stars or tiny microbes, but our minds.

-We like to think of ourselves as these really rational agents, but most of the time, we're relying on lots of stuff that's just on autopilot.

-We'll take you inside the autopilot system which enables you to make decisions so fast you're not even consciously aware of them.

And we'll discover the problems with autopilot decision-making, which are at the heart of stories by Michael Lewis, like "Moneyball" and "The Big Short."

-We're hardwired for certain kinds of really serious mistakes.

-And we'll find out why even Lewis was stunned to learn about the scientific revolution that explains why.

-I went, "Oh, my God!"

-We'll explore the autopilot mistakes that lead to much of the bias in our society.

-Even when there's no evildoer, we can still have bias, and we can still be affected by bias.

-We'll explore how marketers and corporations hack our autopilot system, and we'll go inside the political campaigns that brought corporate-style hacking to politics.

Best of all, we'll find out why this same scientific revolution offers tremendous hope.

-When you learn more about how the mind really works, I think it should give you an incredibly optimistic view about the future.

-This is "Hacking Your Mind."

And if you want to understand how to keep others from getting control of your mind and take control of it yourself, come around the world with us right now.

♪♪ ♪♪ ♪♪ -"Hacking Your Mind" was made possible by a major grant from the National Science Foundation, "Where discoveries begin."

♪♪ -Hello.

I'm Jake Ward.

We live in an amazing age.

We can communicate with almost anyone we want dozens of times a day via social media.

We can shop online and have what we buy delivered to our door.

We can get access to most of the information that anyone, anywhere in the world, is creating.

But the same technology that makes all of that possible has given rise to the era of big data, in which marketers and politicians and governments now have access to thousands of details about the decisions each of us makes and how we behave, and they're using that knowledge to literally hack our minds.

American marketing geniuses today have more influence over what we decide to buy than anyone in history.

The Russian government is using carefully designed behavioral science techniques to try to influence how specific groups of people in other countries vote.

And the Chinese government is on the verge of creating a brave new world where every citizen's decisions must meet the approval of the group.

But, at the same time, a new scientific field has arisen -- decision science -- that's discovered why our minds are so vulnerable to being hacked.

Its discoveries offer us a kind of owner's manual for our brains, one that comes complete with antivirus software so you can take defensive action when others attempt to hack your mind.

Just like with any new manual, the first chapter of this one explores the remarkable and surprising way that the systems that power our decisions are operated.

Two of the men who helped figure that out were Daniel Kahneman and Amos Tversky.

In 2002, Kahneman won the Nobel Prize in Economics, which he dedicated to Tversky, who passed away in 1996.

-We discovered very early on, Amos and I, that we enjoyed working together, and that's what we did.

We basically did everything together.

We walked a lot in Jerusalem, which is a very beautiful city, and we would share just about everything, so, in fact, we had access to each other's mind to a very exceptional degree.

-Best-selling author Michael Lewis became fascinated by the two men and titled his book about them "The Undoing Project."

-They were busy undoing a false view of the human mind.

What they were undoing, in particular, was a sense that, um... fallibility is sinful or shameful.

They were replacing it with an idea that we are inherently fallible.

We're hardwired for certain kinds of really serious mistakes, and the spirit in which one should approach this fallibility isn't shame.

It's, "Let's try to understand it and do our best to correct for it."

-When they first began collaborating, the world's most influential economists had an entirely different point of view.

Their theories were based on the assumption that people make decisions the same way a famous character from "Star Trek" does.

-Logical.

Flawlessly logical.

♪♪ -Kahneman and Tversky helped start a scientific revolution by demonstrating that, far from always being rational and analytical, like Mr. Spock, we make most of our decisions by relying on what Kahneman calls fast thinking.

Decisions made when we're thinking fast are nonconscious, intuitive, effortless, shaped by our gut feelings.

It's as if we're operating on autopilot.

-If I say, "Two plus two," a number comes to your mind.

And it comes to your mind -- you don't have to decide to compute it.

In fact, you can't help it.

It's an association that came to your mind.

If I say, "Your mother," something happens to you, which is emotional.

And, you know, so you have images of your mother, all sorts of good or bad thoughts or whatever.

You certainly are having an emotional reaction.

This is...

This occurs immediately.

It occurs spontaneously.

And so there is a whole system that generates those very rapid thoughts in response to anything that happens.

♪♪ -The world's great economists scoffed at the idea that we humans make most of our decisions on autopilot.

But it made sense to those who, like Laurie Santos of Yale, study our primate relatives.

Just off the coast of Puerto Rico is a unique research facility -- Cayo Santiago.

The locals call it Monkey Island, since its only residents are 1,200 rhesus monkeys.

Santos is here to study the ways in which our autopilot system is like the one these monkeys use.

-What do we have in common with monkeys?

-So, both humans and monkeys have this really fast system for making snap judgments.

It's a system that researchers have called System One, this kind of first evolved system, and it's super good when you just have to come up with a decision fast.

You don't have time to think about it.

You just have to make a really quick snap judgment.

-Mm-hmm.

[ Monkeys snorting, hissing ] [ Monkeys squabbling ] -Right behind us, we have a monkey who just had this interaction.

She's scared.

She has to decide whether to stand there and defend her turf or to take off.

-Right.

-Afterwards, when she's kind of upset, her friend might have to decide whether or not to groom her or not.

All of these are calculus that lead monkeys to make better friendships or not.

-Right.

-They're kind of constantly making these subtle decisions that impact them, their life, and their well-being, even though they're really fast judgments.

-Right, right.

-And this is the amazing thing about our human System One, is that we're making all these decisions all the time.

We quickly and unconsciously decide whether the person who's around us is a man or a woman, or...

So, kind of fast judgments that we don't even notice we're making, but they are choices.

They are judgments.

They are decisions.

We just don't notice that we're making them.

-I think about, you know, how I act in an emergency, right?

When I'm about to drop something or I see my kid about to fall, [ Snaps fingers ] I just act.

I don't think about it.

I don't think how it would reflect on me.

I just do it.

Is it sort of like that for monkeys?

-I think that's right.

I think all of their decisions feel to them like it feels to us when we're making those quick, snap judgments in an emergency.

[ Monkey chattering ] -Decisions like that happen so fast, we're not even aware of making them, and so there's no reasoning or conscious thought involved.

-Most of the time, even though we like to think we rely on deliberative thought, we're relying on lots of other stuff, lots of stuff that's just on autopilot.

We rely on the habits we have that we formed up over learning over time.

We rely on quick intuitions that come sometimes based on our experience, sometimes based on our evolutionary history.

And oftentimes, we're relying on emotion.

We're thinking in terms of the way we feel.

Are we happy right now?

Are we hungry right now?

Are we sad?

These are the kinds of things that are affecting us way more than we think, and so we have to understand how those fast processes work and how they interact with our deliberative thought if we're to make sense of how the human mind works.

♪♪ -It's not just primates, like us and monkeys, who use an autopilot system.

So do other mammals, and reptiles, as well.

In humans, the wiring that makes that possible is in the most ancient region of our brains.

It evolved so long ago that researchers have discovered that even people who, so far as they're consciously aware, are blind, due to brain damage, can still, without even being conscious of it, react to what's going on in the visual world around them.

One of neuroscientist Bea de Gelder's patients has what's called blindsight.

What are we seeing here?

-Here you see a very, very impromptu piece of scientific evidence.

We had been testing all day, and there was no evidence of any normal visual function.

-But then, de Gelder tried setting up an obstacle course in the hallway of her office.

-We said, "Look, I mean, all we ask you, without any further instructions or any further detail, is that you walk straight ahead of you."

-Hm.

-And that's literally all we said.

-The blind man proceeded to step around every object in his way.

-We were all watching, totally flabbergasted.

How is this possible he can do it?

-Does he remember having made a left and a right?

-No, he doesn't even believe when we tell him that he avoided the obstacles.

-Get out of here.

Really?

-Yeah, he doesn't believe that.

-How is it possible that he is doing this?

-The eye is absolutely healthy.

-Mm.

-The part of that's not intact is that piece in the cortex, the cortical shield, in the back of the brain that receives the signals.

-So in these people, the connection has been broken between the eye, which is intact, and the part of the brain that consciously processes what we see.

-Yes.

-How does that image get to the part of the brain that then can respond?

-Because it travels from the eyes through alternative visual routes that do not involve consciousness.

There are those other routes.

They have been documented in animals.

They exist in humans, but they are not associated with knowing what you do, with being conscious of what you see.

That's what ancient history has equipped us with to survive.

I really believe that a whole lot of our processing is taking place all the time, sort of bypassing our conscious... -Wow.

-...our conscious awareness and also our conscious decision-making.

-Of course, we also make decisions the way we've always imagined we do -- rationally.

Kahneman calls that our slow-thinking system.

♪♪ Decisions made when thinking slowly are reasoned, conscious, time-consuming, energy-draining.

-So if I ask you, "17 times 24 -- you know, what is it?"

no number comes to your mind.

Now, you can produce that number, but it's going to be mental work.

It's going to take time.

It's going to occupy you so that you cannot hear or really do other things while you're performing that computation.

It's going to have physiological manifestations.

Your pupils will dilate.

There will be changes in your pulse, changes in your blood pressure.

So, the whole body changes when mental effort is being done.

[ Beeping ] -As we go through life, our autopilot system is constantly proposing potential decisions.

But if our slow-thinking system doesn't like what it's proposing, it can tap the brakes, turn off the autopilot system, and make a different decision.

-You want to choose based on your preference.

-That might sound easy.

-How hungry am I right now?

I'm moderately hungry.

-But as scientists at Caltech are discovering, it all happens so fast... -Wow.

Cake.

-...that it can be hard to do.

-Toblerone.

-Take what happens when our gut feelings tell us to eat something delicious, but reason tells us we should eat something healthy.

-Hm...Mrs. Fields.

Uh...I guess a Kit Kat.

Pie.

Cookie.

Goodbar.

-Researchers here can measure how fast different regions of the brain process a decision.

They can also detect when a final decision gets made.

And frequently, we decide to do what our autopilot system wants before the slower-thinking regions of our brain have even finished weighing in.

-Pringles.

-The areas of the brain that are assigning value... -Right.

-...to the foods when you're making the choice start reflecting taste information about 200 milliseconds earlier than they start reflecting health information.

-So "yummy" happens faster than health?

-A very small amount of time, but that's enough to have a big impact, given how quickly these decisions get made.

-Because our autopilot system is mostly really good at what it does, plus, fast and easy to use, we tend to rely on it whenever we can.

-Let's say that you're driving a familiar street in your hometown.

The slow-thinking part of your brain is psyched because basically that's free time, time off.

The route has become so automatic that the fast-thinking part of your brain does all the work.

This is why we sometimes get lost in thought when we're driving.

The slow-thinking part of your brain is free to go on and on about dating and your friends and why does that one friend keep dating such stupid people.

While your slow-thinking brain is going, "Noonie-noonie-noonie-noo," the fast-thinking part of your brain operates the car.

♪♪ But when you're doing something that is not automatic, that's when the slow-thinking part of your brain has to take charge.

For instance, right now, I have to drive on the wrong side of the road.

I'm in England, and something that should be totally automatic for me requires all kinds of conscious thought.

My slow brain needs to get in here and deal with every turn signal on these weird roundabouts and trying not to take anybody's door off.

And before long, I start thinking, "My brain is fried," because my slow-thinking system is running out of energy.

♪♪ But despite that, I cannot let my autopilot system take over, because it would make a terrible mistake, like turning straight into oncoming traffic, because, guess what -- I'm not in my hometown.

I'm in England, where they drive on the wrong side of the road.

♪♪ But here's the thing.

It's rarely that obvious that letting our autopilot system make the call will lead to a mistake, and slow thinking takes so much energy.

And so, when facing other decisions our autopilot system isn't any better at handling than driving on the wrong side of the road, we nevertheless stay on autopilot.

And the result is that we make serious, but predictable, mistakes.

♪♪ To prove to their skeptical colleagues that we human beings make those predictable mistakes, Kahneman and Tversky began doing all sorts of experiments with average people.

♪♪ -We're about to put on a game show -- a bizarre game show, based on an experiment that Kahneman and Tversky used to do on real people.

But first, I have to trick our contestants.

I'm gonna rig this wheel so that it only comes up either 10 or 65.

Good evening, and welcome to "The Anchoring Game."

Here is our evening's first contestant -- Cheryl.

Cheryl, where are you from?

-Lake Oswego.

-Lake Oswego.

And do you have a view there?

-Just a slice.

-Well, let's see if we can win a little something more.

Now we're gonna begin by spinning the wheel.

Vicki, please.

Alright, alright.

Let's see where it ends up.

-Come on!

Up, up!

♪♪ -Alright.

The number 10.

Cheryl, do me a favor.

Write that number down on this piece of paper.

-Okay.

-Here we go.

Here we go.

Now let me ask you a complicated question.

Is the percentage of African nations among UN members larger or smaller than the number you just wrote?

What do you think?

-Um... Higher!

Higher, higher.

-A higher number.

Alright.

So, question number two -- What is your best guess of the percentage of African nations in the UN?

-Uh...20%!

-20%, she says.

20%.

♪♪ So, here's what's really bizarre.

When the wheel lands on 10, people tend to say that the percentage of African nations in the UN is close to 10.

-30%.

-30%.

Alright.

-And when the wheel lands on 65, people tend to say that it's close to 65.

-50% percent.

-50%, he says.

-But, of course, what number this wheel landed on has absolutely nothing to do with the percentage of African nations in the UN.

So what in the name of Vanna White is going on?

According to Kahneman, when our slow-thinking system doesn't have enough information to answer a question involving numbers, we simply stay on autopilot.

And our autopilot system takes what might be called a shortcut and anchors its answer to the last number that crossed its radar, even when that number is completely irrelevant to the question at hand.

-[ Slo-mo ] Lower!

-And that leads us to reach an absurd conclusion.

♪♪ I know it seems bizarre that anyone would do that, and surely you and I, reasonable people, would never do that in our real lives.

Well, you do it all the time.

Let's talk, for instance, about what happens when you buy a used car at a lot like this.

♪♪ See the number on the windshield?

That price is not actually the price the dealer expects you to pay.

It's called an anchor.

That number anchors itself in your mind, and you begin to negotiate around it.

Now, you don't actually have enough information to know what this vehicle is worth, and yet when the dealer accepts a lower price than that, you think to yourself, "What a shrewd negotiator I am.

I got a great deal."

But the thing is, the dealer intentionally placed this anchor into the transaction, and, as a result, you probably paid more than you should've.

How can you avoid being hacked?

The best defense is to train your slow-thinking system to be on the alert for anchors, and as soon as you spot one, tap the brake and turn off your autopilot system before its emotional investment in having that car right now leads to a real mistake.

♪♪ Understanding that making decisions on autopilot can lead to mistakes is even more important when it comes to people.

That's because every one of us regularly uses our autopilot system to make decisions about other people that it's simply not equipped to make.

For instance, colleagues of Kahneman and Tversky's demonstrated that we frequently rely on an autopilot shortcut -- a quick glance at a person's face -- to decide whether they are smart or dumb, mean or nice.

-For centuries, human beings have believed that the face, and the eyes especially, are windows to the soul, that we can read a person's character from their face.

And, of course, as modern humans, we know that that is not true.

We teach children this.

We tell them, "You don't judge a book by its cover.

You don't look at people and make decisions about them."

But what we should know, of course, is that humans use the wrong facial cues to make all sorts of decisions about people's character.

-First, let's look at this face.

Now let's look at this face.

Which of these people is nice?

-Simple things like the distance between a person's eyes or the distance between the nose and the upper lip or whether the mouth is turned up or down are all signals to us in some way about a person's underlying properties, like their character, their goodness, their trustworthiness.

-Okay.

Let's look at this face.

Now let's look at this face.

Which of these people is mean?

-Our question was a very simple one -- Is this learned over the course of our development as humans?

Or is this feature so deep in us that even the youngest children we can test will show it?

To our great surprise, we discovered that even the youngest children we were able to test -- 3-year-olds -- performed very similarly to adults.

They're using exactly the same cues to make the same decisions about trust, about who's likely to be dominant versus submissive, who's smarter or less smart.

And by the time they're 5, they're exactly like adults.

And so what this shows us is that these shortcuts, or these quite incorrect ways of judging people, are fundamental to our nature -- so fundamental that even a 3-year-old will show it.

♪♪ -Making decisions that way can lead us to not just think that somebody very smart is dumb.

It can lead us to make mistakes in some of our most important decisions.

♪♪ It turns out that, if you show people photographs like this, photos that go by so fast, you can't even really consciously register the faces, and then you ask people to rate how competent and likable these people are, the fast-thinking part of your brain will deliver a decision.

But all that part of your brain can really detect is whether the person in the photo has a confident smile and a strong chin.

That is it.

Here's why that's important.

These are the faces of people who are running for office in the United States, and it turns out that the candidate whose photo got a higher rating of competence and likability, based on nothing more than a quarter-second glance, well, they won their elections 70% of the time.

That is a way higher percentage than can be explained by chance.

And so the research suggests that it's not my slow-thinking, rational, thoughtful brain that's making these decisions -- the part of my brain that I would hope would be in charge.

Instead, it's the fast-thinking, automatic part of my brain that is casting my vote.

It turns out that even experts make autopilot decisions based on their emotions and intuitions.

And when that expert is a surgeon, it can be a matter of life and death.

In a study that Tversky conducted at one of the most prestigious hospitals in the country, he found that, when doctors were told that a surgery for lung cancer had a 10% mortality rate, a surprising number of them said, "Well, I can't possibly recommend that to my patients."

But when the doctors were told that the surgery had a 90% survival rate, far more or them chose to recommend it.

The facts were the same.

10% mortality, 90% survival -- it's the same thing.

But whether it was framed as a loss -- death -- or as a gain -- survival -- determined how they felt about the surgery, and that influenced their decision.

When the results were revealed, the doctors were shocked that they could make a mistake like that.

And studies exposing the autopilot mistakes made by experts in other fields have also left those participants stunned.

-And it turns out, I think, that this is the key to the appeal or the effect of the theory, is that people discover you're not exactly what you thought you were, because a very important part of your thinking is done outside of your awareness and outside of your control.

♪♪ -Of course, we also have that other way of making decisions that only we, of all the species on Earth, have -- the slow-thinking, logical way.

And that way of thinking has enabled us to build one civilization after another and transform the world.

So if slow thinking can be so useful and our autopilot system sometimes makes mistakes, why didn't we humans evolve beyond it, just like we evolved beyond other species in all sorts of ways?

The answer can be found not where most of us live today, but in the environment our hunter-gatherer ancestors lived in thousands of years ago, the one our brains evolved to cope with.

What was life like in that environment?

We are on the way right now to find out.

We are about 10 hours into the interior of Tanzania, in a very remote part of the country, and we're looking for a unique set of people -- the Hadza.

The Hadza are a nomadic tribe.

That means they go where the food is available, and when that food is gone, they move on.

It also means that there's no e-mailing them to arrange a spot to meet.

They don't have cellphones.

And so we have no idea where they are, and we have to go find them.

After searching for hours, we thought we might never find them.

And then my guide, Deus Haraja, spotted one of the Hadza children.

As soon as the Hadza children saw Deus and me, they rushed to greet us and acted exactly like my own children.

[ Children giggling ] People are people wherever you go.

Yet life in this environment could not be more different from my life.

Every day, the Hadza eat what they've hunted and gathered that day.

When there's no more food to be found near this location, the Hadza will move on.

And while they can trade for a few things, like clothes, with the people in the villages they pass, if they can't find more food, they'll starve.

And if they get hurt or sick, there are no doctors or hospitals to go to.

-[ Singing in foreign language ] ♪♪ ♪♪ -The Hadza are able to not only cope with life in this dangerous and unforgiving landscape, but thrive, by relying on the same combination of autopilot decision-making and slow thinking that every other human on Earth relies on.

But in this environment, the one the human brain evolved in, it's much easier to see why making decisions on autopilot can be not only useful, but essential.

For a hunter like Dofu, it can be the difference between life and death.

Think for a second about the world in which Dofu lives.

I mean, it is a world that punishes hesitation.

He's got to act incredibly quickly in order to not just feed his family and himself, but also to avoid being killed by the elements and whatever else.

I mean, you know, thinking -- that's for later.

You got to act, right now.

♪♪ ♪♪ -Dofu is a badass hunter.

-[ Laughs ] -He's one of the best hunters at the camp, and he killed a leopard.

-He killed a leopard?

-He did, yes.

-Wow!

-By bow and arrow.

Then, he killed a giraffe.

♪♪ [ Bow snaps ] -Dofu has managed to do something truly remarkable -- spot an animal called a hyrax 40 feet away and almost instantly shoot it through the eye with an arrow.

-[ Speaking foreign language ] -[ Blowing ] -The hyrax's closest relative is the elephant, but it's only as big as a rabbit.

And so the Hadza will soon be back out on the savanna, hunting and gathering, and very often making autopilot decisions in response to what happens around them.

Now, I understand that it's difficult for those of us who live in modern society with our cellphones and our cars and our houses to understand that we make decisions in this way, very suddenly.

But if you spend time with the last of the hunter-gatherers, you begin to understand just how well it served our ancestors to make decisions this way.

And if it hadn't, we wouldn't be here.

♪♪ But here's where it gets tricky -- not for the Hadza, but for you and me.

Our autopilot system does not work as well in our world as theirs works here.

Why?

Because the intuitions and gut feelings we humans experience, and which our autopilot system relies on to make fast decisions, evolved to improve the chances of survival in these conditions, not the ones we live in.

So, for instance, whenever the Hadza get a hold of fat, salt, or sugar in the form of honey, their gut feelings tell them to eat as much of it as they possibly can.

Scientists call that a bias.

The Hadza's autopilot decisions about what to eat are biased towards sugar, fat, and salt.

[ Harp trilling ] And, of course, our autopilot system is biased toward eating sugar, fat, and salt, too, but we don't live in that world.

We live in this one.

-A very good example of why the biases that paid off in the old world may not be useful today is sugar and fat.

Our ancestors would not have survived had they not evolved to store every little smidgen of sugar and fat.

That's what allowed them to survive.

That very same thing that allowed them to survive is killing us off today because of an abundance of sugar and fat in our environment.

[ Whistle blows ] -And just like the kinds of food we have access to has changed, so have countless other things.

We no longer live in small groups of less than 50, made up of only people who look like us.

Our weapons do vastly more damage than spears and bows and arrows did.

And the decisions we make can affect the lives of everyone else on the planet.

But despite all those changes, we still frequently make decisions based on autopilot biases that made sense long ago, but don't make sense in this world.

And that can lead us to make serious mistakes.

-The very thing that may have paid off in the past, today, could kill us off, unless we do things like go against our natural selves.

But this is not at all an easy thing to do.

-It's especially hard to overcome our autopilot biases because, much of the time, we're not even aware we're experiencing them.

For instance, here's an autopilot bias I can almost guarantee you're not aware of -- being biased in favor of one person over many people.

One of Kahneman and Tversky's closest colleagues studies how that bias distorts the decisions we make.

-Take a look at this image of the little boy from Syria, Alan Kurdi, washed up on a beach in Turkey when he and his family were trying to flee the civil war there.

-I have to tell you, this is the longest I've ever been forced to look at this photo 'cause I click away from it.

I just can't...

I have a 3-year-old daughter.

I can't even look at this photo, it's so emotional.

-Most people saw that slide.

It went around the world, and it created a very strong emotional reaction.

The day this photograph appeared, the donations to the Red Cross increased by 100 times what they were... -Wow.

-...the day before.

There had been, you know, the statistics, the death toll in Syria and refugees dying trying to escape Syria for years prior to this picture.

And those statistics were periodically reprinted in, you know, major newspapers, and the world basically was asleep.

Then this photograph came out, and people woke up.

I think that, through evolution, our feelings system, you know, was designed to react to individuals.

You know, we didn't have to worry about large numbers of people thousands of miles away in distress, and so -- -It's just whoever's with you around the fire or in your cave.

-We're very sensitive, very sensitive to individuals -- -Yeah.

-Is this someone that I would like?

Or is this child, you know, like one of my own?

-No, I can imagine putting those shoes on him.

I can imagine having chosen -Exactly.

Exactly.

-that outfit for him.

-So, once people's compassion had been triggered by that shot of the one boy on the beach, there were then photographs of thousands more people.

Did people's compassion rise with the number of people pictured?

-No, it actually declined back to where it was before the photograph first appeared.

The human mind tends to be lazy, and it's much easier just to go with our gut feelings.

And, all too often, our feelings are asleep when the numbers roll by us.

Among us researchers, one of the things we're trying to understand is, when can we trust our feelings, and when should we recognize that this is a situation where we'd better think more slowly and do the math and do the arithmetic properly in order to appreciate the seriousness of the problem?

♪♪ -We human beings experience all sorts of other autopilot biases.

For instance, you and I and everyone else is sometimes guilty of self-serving bias.

That's because when I use my feelings as a shortcut way of, say, deciding if I did the right thing or not, my feelings tend to tell me, "Hey, I was right!"

After all, this is me we're talking about.

And since, hard as it is to believe, I am not always right, that's definitely a bias.

And then there's the kind of bias that we usually mean when we say the word bias -- you know, racism, sexism, ageism, and so on.

Sometimes, that's the result of explicit prejudice that we're consciously aware of.

But far more often, it's the result of mental shortcuts that we all use without even realizing it.

Take sexism.

It sometimes leads to sexual predators feeling free to violently abuse women.

But sexism has an even wider impact experienced by virtually every woman because it can also result from the way that men tend, without even being aware of it, to categorize women.

How can we men not be aware of it?

[ Indistinct conversations ] Hey!

How you doing?

Great to see you.

Thanks for coming.

Hey, do you know these guys?

-Absolutely.

-You do, right?

Okay.

Alright.

Great.

There is a simple human truth in all of our lives.

Hey, uh, remind me I want to introduce you to somebody.

Okay.

Great.

Human beings are always meeting other human beings, in all sorts of circumstances.

Hey, congratulations on the gig.

You deserve it.

That's great.

Hi, you guys.

Thanks for coming.

Nice to see you.

Hello!

And we have to figure out how to respond to one another.

[ Doorbell rings ] You too, man.

[ Suspenseful music plays ] Say I'm having a party and a woman I haven't met before comes in the door.

Without my even being aware of it, my autopilot system instantly categorizes her.

[ Sultry hip hop plays ] ♪♪ But that's not much information, is it?

I mean, I barely know anything about her.

But that's no problem for my autopilot system, because it has a category into which it instantly puts young blonde women.

In other words, even though I'm happily married, my biases can still make me sexist.

And here's what's really nuts -- the more elaborate a story I make up about her, the more likely my fast-thinking brain is to believe that it's true.

♪♪ -When you're constructing a story, the measure of confidence that we have in the stories we tell ourselves is how coherent the story is.

Is it a good story?

And we are much more sensitive to that than we're sensitive to the quality and quantity of the evidence that was used in constructing that story.

-Occasionally, my slow-thinking system manages to find out just how wrong my bias has made me.

♪♪ So can I learn to stop categorizing women, and men, too?

Yes, if I slow down and really work at it.

But if I don't, my autopilot system will keep doing what it always does.

Got to be a college professor, teaches psychology.

Pretty good, right?

Or maybe not, because that's just a story I instantly made up based on my biases.

♪♪ Bottom line -- when we're operating on autopilot, we humans tend to automatically categorize everyone.

Some of the most damaging consequences result from categorizing people who look like we do as good and categorizing people who don't look like us as bad.

At Yale's Social Cognitive Development Lab, Yarrow Dunham is finding that young children appear to be hardwired to treat people that way, even when the differences between groups are incredibly small.

-So, Arlee, are you ready to play a game?

-Mm-hmm!

-Okay.

Well, the most important thing to know about this game is there are two groups of kids.

There's an orange group.

-Orange group.

-And... -Green group.

-Exactly.

There's an orange group and a green group.

And the kids in the orange group wear orange shirts, and the kids in the green group, they wear green shirts.

So the first thing we have to do in this game is figure out which group you're gonna be in.

How do you think we're gonna do that?

-Spinning.

-Let's do it.

[ Gasps ] -Both: Orange!

-That means you're gonna be in the orange group.

Oh, you're way ahead of me.

Ready?

You ready to put on your orange shirt?

-Uh-huh!

-Here we go!

Our goal in this experiment is to come up with some, what we hope, is a pretty trivial dimension and use that to create two groups, so we do that here through the T-shirt colors.

Which one of these children would you like to play with?

Okay.

And why'd you choose that one?

-Because she has the same shirt.

-Oh, okay, because she has the same shirt.

But here's a boy in the green group, and here's a boy in the orange group.

Which of these kids would you rather play with?

-Orange.

-Okay.

And why did you choose that color?

-Because he's in my group.

-So what we're now finding is that children at least as young as age 3 seem to, quite automatically, come to prefer the groups to which they belong.

Which one do you think has bubble gum as their favorite flavor?

-This one.

That one.

-That one.

Why do you think she wants to give you candy?

-Because she has a green shirt, and some candies are green.

-One of these children would definitely share their lunch with you.

Which one do you think would share their lunch with you?

-Orange!

-So one of the ways I think about it is that putting them into this very sort of simple and really kind of, you know, this really kind of meaningless social grouping sets off a whole cascade of biases.

If you want to give stickers to this girl, you put them in this cup.

If you want to give stickers to this girl, you put them in this cup.

And you can give them out however you want.

Why don't you give a try?

Okay.

-How much time do they have to have spent with an orange shirt on or a green shirt on -for this to happen?

-This is one of the remarkable things, I think, is that, within 30 seconds, you know, within a few minutes, these biases seem to be there.

♪♪ The sort of bet is that real-world groups, including really important groups, like race, that there's a moment in a child's developmental history where that distinction was, to them, very similar to the distinction we made in the lab today.

They've had a moment where they go, "Oh, my society divides people up based on, call it skin color."

As something like that, as race is continually reinforced, it becomes a little more salient, something that we draw on more automatically, more regularly, as just sort of a lens through which to understand people.

So the speed with which bias can become manifest, can become real in the world is, I think, the real scary note that I would strike.

-Here's one more thing it's essential to get.

Though the kids were clearly biased in favor of one group over the other, they weren't aware of it, because they were doing it on autopilot.

And so each of them insisted afterward that they'd treated both groups exactly the same, and appeared to genuinely believe it.

♪♪ That's why these biases are called implicit biases.

The slow-thinking part of our brain doesn't even realize that we have them.

How warm or cold do I feel towards young people?

You know what?

I don't really like young people, to be perfectly honest with you, so somewhat cold, I would say.

This is an implicit association test.

It's measuring whether I tend to quickly and automatically associate negative words with old faces, but have to slow down and think hard before I can manage to make myself associate those negative words with young faces.

And it does the reverse with positive words.

Basically, it's trying to trick my fast-thinking brain into revealing how it feels about young and old people.

♪♪ Oh, man, that's bad.

I would've said that I have an automatic preference for old people, but according to this test, I have a slight automatic preference for young people.

That's discouraging.

Oh, well.

And, you know, I'm not the only one to be surprised by my results.

Millions of people have taken these tests online, and among the results are that 80% of people automatically associate more negative characteristics with elderly people than with young people.

And not only that -- 75% of white people who take tests like these show an implicit preference for white people over black people.

[ Indistinct shouting ] In 2015, news reports featuring cellphone video of white cops manhandling black teenagers at a pool party sparked yet another national debate about whether American cops are biased against black people.

It's a debate that triggers intense emotions on every side, but what would we discover about how bias affects the treatment of black people by police if there was a way to study thousands of actual interactions between police officers and both white and black citizens?

We're at Stanford University because we've been invited to join, in action, a groundbreaking study into the differences between how police officers treat black people and white people.

So, tell me about your project here.

-So, we've been working with the Oakland Police Department on analyzing footage from body-worn cameras.

They clip them on the front of their uniform.

Let me just show you a piece of footage, actually, that comes from a stop from one of these cameras.

-Can you turn the car off for me, sir?

-What?

-Is the engine off?

-How much footage do you have?

-Oakland has had footage for maybe nearly seven years, now, and so there's a huge amount of footage.

Well, we're interested in stepping back and actually looking at the footage as data, and to not just look at the footage for one case, but to look at the footage for all of these stops.

-What kinds of patterns do you think you could detect?

-Well, we want to look to see, for example, if African Americans are treated any differently from white Americans during these stops, and we're using these computational linguistic tools, help us to understand the language used in these stops.

-Yeah, do you have your driver's license with you, sir?

-What Eberhardt's researchers have found sheds new light on how police officers across America treat white people and black people.

-Pretty good.

The reason I pulled you over, you got a brake light out back here on your passenger side.

You aware of it?

-Oh.

If I get out just a second, I'll fix it.

-Alright.

[ Static ] -Put your hands behind your back.

-Why did y'all stop me, for one?

-Face away!

-Why did you stop me?

-So, what I'm interested in is not just what the officer is saying, but when they are saying it.

-Mm.

-And it turns out that the order in which the officers say things matters a lot.

-So what does one of these conversations typically look like?

-So, a typical conversation starts with a greeting.

It goes on to giving the reason for the stop and then getting the documents, asking for documents.

-"License and registration, please," yep.

What has your data told you about the differences between the script that a white person receives and the script that a black person receives?

-Based on our analysis of around 1,000 stops, what we see is that white motorists are given the reason for the stop earlier on in the conversations and more consistently than black motorists.

And sometimes, black motorists are not given the reason for the stop at all.

-Wow!

-Tell me what you stopped me for!

-We'll tell ya.

-[ Shouting indistinctly ] ♪♪ -It's the other side here, your brake light.

-Oh, my brake light?

-Uh-huh.

-Oh, I didn't know that.

-Okay.

-I'll get it fixed.

-Alright.

-So, I'm looking specifically at respect.

And, once we can identify these particular things about the officer's language, we can make an estimate of sort of their overall respectfulness at any given time.

And so a key part of that is looking at how particular kinds of words are being used more when an officer is speaking to a black person, versus a white person.

-Right.

Right.

"Apologizing.

Reassurance.

Formal titles."

-Those, for instance, are... we would traditionally think of as pretty respectful, and it turns out that they're significantly more common in stops when the officer is talking to a white community member.

-Wow.

But questions, where they're probing for more information, informal titles... -Like "dude" and "bro" and "man."

-Right, right, right.

Oh, wow.

And "Hands on the wheel."

I can tell you right now, I have never been told to put my hands on the wheel.

-Yeah, and those words are more common in stops with black community members.

-Wow.

-A surprising fact is that, if we just look at the words that were used by the officer, we can predict the race of the person that they're talking to about 2/3 of the time.

-Wow!

Even if you don't know the race of the person that was stopped, you can tell it from how that person was spoken to?

That's crazy.

-I swear.

Ow!

Oh, my gosh!

-A black person who gets pulled over is worried about how they're going to be treated.

-Put your hands behind you.

-I do have a firearm on me.

-Okay.

Don't reach for it, then.

-They see these officer-involved shootings on television.

[ Gunshots ] -Aah!

Aah!

Aah!

Aah!

-You just killed my boyfriend.

-They know other people who've been pulled over.

-[ Moans ] I can't breathe, Officer.

-They've heard stories about negative interactions.

"How is this gonna go for me?"

Right?

That's what they're thinking, "How is this gonna go for me?"

-Eberhardt's goal, and that of the Oakland Police Department, is to use what her team has found to get police officers themselves to recognize that cops sometimes treat black people very differently than they treat white people.

-It's about your brain and the culture, you know, that we're immersed in, that those don't get in the way of your decision-making, they don't get in the way of the actions you're taking.

-Eberhardt gets cops to discuss not who is to blame, but how nonconscious autopilot decisions can be shaped by implicit bias.

-There are no evil people required.

There's no bad intention.

Even without awareness, even unintentionally, even when there's no evildoer, right?

We can still have bias, and we can still be affected by bias.

-Eberhardt has found that, when that's the message, cops are far less likely to instantly become defensive, and to instead slow down and start thinking seriously about what her research reveals and what they can do to keep implicit biases from influencing how they treat people.

-Alright.

Thank you, guys.

-You can understand it the wrong way or incompletely and think that it's really an attack on someone's values and their character because, you know, emotionally, at least, you kind of go, "That's not me.

I've worked very hard not to be that.

In fact, I'm the opposite of that."

But, as you learn more about what implicit bias actually is, I think you understand that these are part of the frailties of humankind, and we all suffer from that.

We all have it, so it's not something to avoid or be defensive about, but it's something to understand and to work towards mitigating the negative effects of it.

♪♪ -And that's true not only for police officers, but for all of us, no matter what race or sex or age we are.

We all sometimes make decisions about other people that are influenced by implicit biases, not because we're all rabid racists or sexists or ageists, but because we all frequently make decisions about other people on autopilot.

And, when we do that, we aren't basing our decisions on who that person really is.

We're basing them on an imaginary version of that person shaped by our biases.

♪♪ Kahneman and Tversky described how our biases can influence our decisions this way -- we make many of our most important choices based on what can only be called cognitive illusions.

This is embarrassing for us to realize, but the evidence for it is undeniable.

♪♪ So, how do we stop making decisions based on illusions created by our biases?

Kahneman suggests that, when possible, we tap the brakes and turn off our autopilot system and let our slow-thinking system make the call.

-There are big decisions where slowing down has major advantages.

It makes you more suspicious of the stories you tell yourself.

And, on some occasions, it may cause you to actually stop yourself from making a mistake.

I think some people learn to be more self-critical about themselves and to use slow thinking instead of fast thinking, and just that metaphor itself is useful to some people.

Should I slow down?

I'll slow down.

And slowing down has large effects on the choices that people make.

[ Beeping ] -Of course, slow thinking takes a lot more time and effort, so why should you bother?

Because, when you're operating on autopilot, it is shockingly easy to hack your mind.

Everyone from marketers to politicians to governments knows how to break into the system and trigger a predictable mistake that is bad for us, but great for them.

And the last thing the hackers want is for you to understand how they do it and how to stop them.

But we will clue you in on the next episode of "Hacking Your Mind."

♪♪ ♪♪ -Learn more about "Hacking Your Mind" by visiting pbs.org.

To order this program on DVD, visit Shop PBS or call 1-800-PLAY-PBS.

This program is also available on Amazon Prime Video.

♪♪