WEBVTT

00:00:03.834 --> 00:00:06.467
I'll tell you a little bit
about irrational behavior.

00:00:06.501 --> 00:00:08.367
Not yours, of course -- other people's.

00:00:08.401 --> 00:00:09.601
(Laughter)

00:00:09.601 --> 00:00:13.467
So after being at MIT for a few years,

00:00:13.501 --> 00:00:18.167
I realized that writing academic papers
is not that exciting.

00:00:18.167 --> 00:00:20.534
You know, I don't know
how many of those you read,

00:00:20.567 --> 00:00:23.101
but it's not fun to read
and often not fun to write --

00:00:23.134 --> 00:00:24.467
even worse to write.

00:00:24.501 --> 00:00:27.467
So I decided to try and write
something more fun.

00:00:27.501 --> 00:00:30.934
And I came up with an idea
that I would write a cookbook.

00:00:32.301 --> 00:00:34.467
And the title for my cookbook
was going to be,

00:00:34.501 --> 00:00:37.334
"Dining Without Crumbs:
The Art of Eating Over the Sink."

00:00:37.367 --> 00:00:39.101
(Laughter)

00:00:39.134 --> 00:00:42.034
And it was going to be a look
at life through the kitchen.

00:00:42.034 --> 00:00:43.601
I was quite excited about this.

00:00:43.634 --> 00:00:45.901
I was going to talk
a little bit about research,

00:00:45.934 --> 00:00:47.401
a little bit about the kitchen.

00:00:47.434 --> 00:00:50.567
We do so much in the kitchen,
I thought this would be interesting.

00:00:50.567 --> 00:00:54.467
I wrote a couple of chapters,
and took it to MIT Press and they said,

00:00:54.501 --> 00:00:57.601
"Cute, but not for us.
Go and find somebody else."

00:00:57.601 --> 00:01:00.501
I tried other people,
and everybody said the same thing,

00:01:00.534 --> 00:01:02.601
"Cute. Not for us."

00:01:02.601 --> 00:01:05.467
Until somebody said,

00:01:05.501 --> 00:01:07.334
"Look, if you're serious about this,

00:01:07.367 --> 00:01:10.867
you have to write about your research
first; you have to publish something,

00:01:10.901 --> 00:01:13.601
then you'll get the opportunity
to write something else.

00:01:13.634 --> 00:01:15.901
If you really want to do it,
you have to do it."

00:01:15.934 --> 00:01:18.267
I said, "I don't want to write
about my research.

00:01:18.301 --> 00:01:19.501
I do it all day long,

00:01:19.534 --> 00:01:23.034
I want to write something
a bit more free, less constrained."

00:01:23.034 --> 00:01:25.467
And this person
was very forceful and said,

00:01:25.501 --> 00:01:28.067
"Look, that's the only way
you'll ever do it."

00:01:28.101 --> 00:01:30.401
So I said, "Okay, if I have to do it --"

00:01:30.434 --> 00:01:31.601
I had a sabbatical.

00:01:31.634 --> 00:01:34.567
I said, "I'll write about my research,
if there's no other way.

00:01:34.601 --> 00:01:36.401
And then I'll get to do my cookbook."

00:01:36.434 --> 00:01:39.667
So, I wrote a book on my research.

00:01:39.701 --> 00:01:42.267
And it turned out to be
quite fun in two ways.

00:01:42.267 --> 00:01:44.801
First of all, I enjoyed writing.

00:01:44.834 --> 00:01:48.567
But the more interesting thing
was that I started learning from people.

00:01:48.601 --> 00:01:50.201
It's a fantastic time to write,

00:01:50.201 --> 00:01:52.934
because there's so much feedback
you can get from people.

00:01:52.934 --> 00:01:55.467
People write to me
about their personal experience,

00:01:55.501 --> 00:01:57.834
and about their examples,
and where they disagree,

00:01:57.867 --> 00:01:59.467
and their nuances.

00:01:59.501 --> 00:02:01.801
And even being here --
I mean, the last few days,

00:02:01.801 --> 00:02:04.467
I've known heights of obsessive behavior

00:02:04.501 --> 00:02:06.467
I never thought about.

00:02:06.501 --> 00:02:08.134
(Laughter)

00:02:08.167 --> 00:02:10.034
Which I think is just fascinating.

00:02:10.067 --> 00:02:12.667
I will tell you a little bit
about irrational behavior,

00:02:12.701 --> 00:02:15.867
and I want to start by giving you
some examples of visual illusion

00:02:15.901 --> 00:02:17.834
as a metaphor for rationality.

00:02:17.867 --> 00:02:19.667
So think about these two tables.

00:02:19.701 --> 00:02:21.467
And you must have seen this illusion.

00:02:21.501 --> 00:02:25.334
If I asked you what's longer, the vertical
line on the table on the left,

00:02:25.334 --> 00:02:28.134
or the horizontal line
on the table on the right,

00:02:28.167 --> 00:02:30.467
which one seems longer?

00:02:30.501 --> 00:02:33.467
Can anybody see anything
but the left one being longer?

00:02:33.501 --> 00:02:35.334
No, right? It's impossible.

00:02:35.367 --> 00:02:39.267
But the nice thing about visual illusion
is we can easily demonstrate mistakes.

00:02:39.301 --> 00:02:41.734
So I can put some lines
on; it doesn't help.

00:02:41.767 --> 00:02:43.467
I can animate the lines.

00:02:43.501 --> 00:02:46.167
And to the extent you believe
I didn't shrink the lines,

00:02:46.201 --> 00:02:50.967
which I didn't, I've proven to you
that your eyes were deceiving you.

00:02:51.001 --> 00:02:54.467
Now, the interesting thing about this
is when I take the lines away,

00:02:54.501 --> 00:02:57.301
it's as if you haven't learned
anything in the last minute.

00:02:57.334 --> 00:02:59.567
(Laughter)

00:02:59.567 --> 00:03:03.267
You can't look at this and say,
"Now I see reality as it is."

00:03:03.301 --> 00:03:07.334
Right? It's impossible to overcome
this sense that this is indeed longer.

00:03:07.367 --> 00:03:09.001
Our intuition is really fooling us

00:03:09.001 --> 00:03:11.167
in a repeatable,
predictable, consistent way.

00:03:11.167 --> 00:03:13.467
and there is almost nothing
we can do about it,

00:03:13.501 --> 00:03:16.467
aside from taking a ruler
and starting to measure it.

00:03:17.101 --> 00:03:19.667
Here's another one.
It's one of my favorite illusions.

00:03:19.701 --> 00:03:22.401
What color is the top arrow pointing to?

00:03:23.467 --> 00:03:25.634
Audience: Brown.
Dan Ariely: Brown. Thank you.

00:03:25.667 --> 00:03:27.634
The bottom one? Yellow.

00:03:27.634 --> 00:03:29.001
Turns out they're identical.

00:03:29.001 --> 00:03:30.934
Can anybody see them as identical?

00:03:30.934 --> 00:03:32.334
Very, very hard.

00:03:32.334 --> 00:03:34.467
I can cover the rest of the cube up.

00:03:34.501 --> 00:03:37.901
If I cover the rest of the cube,
you can see that they are identical.

00:03:37.934 --> 00:03:40.367
If you don't believe me,
you can get the slide later

00:03:40.401 --> 00:03:43.201
and do some arts and crafts
and see that they're identical.

00:03:43.234 --> 00:03:46.567
But again, it's the same story,
that if we take the background away,

00:03:46.601 --> 00:03:48.801
the illusion comes back.

00:03:48.834 --> 00:03:51.801
There is no way for us not
to see this illusion.

00:03:51.834 --> 00:03:55.134
I guess maybe if you're colorblind,
I don't think you can see that.

00:03:55.167 --> 00:03:57.467
I want you to think
about illusion as a metaphor.

00:03:57.501 --> 00:03:59.634
Vision is one of the best things we do.

00:03:59.667 --> 00:04:02.267
We have a huge part of our brain
dedicated to vision --

00:04:02.267 --> 00:04:04.167
bigger than dedicated to anything else.

00:04:04.201 --> 00:04:08.034
We use our vision more hours
of the day than anything else.

00:04:08.067 --> 00:04:10.134
We're evolutionarily
designed to use vision.

00:04:10.167 --> 00:04:13.134
And if we have these predictable
repeatable mistakes in vision,

00:04:13.167 --> 00:04:14.667
which we're so good at,

00:04:14.701 --> 00:04:17.267
what are the chances we won't make
even more mistakes

00:04:17.267 --> 00:04:20.834
in something we're not as good at,
for example, financial decision-making.

00:04:20.867 --> 00:04:22.467
(Laughter)

00:04:22.501 --> 00:04:25.134
Something we don't have
an evolutionary reason to do,

00:04:25.167 --> 00:04:27.534
we don't have a specialized
part of the brain for,

00:04:27.534 --> 00:04:29.567
and we don't do that many
hours of the day.

00:04:29.601 --> 00:04:31.567
The argument is in those cases,

00:04:31.567 --> 00:04:35.467
it might be that we actually
make many more mistakes.

00:04:35.501 --> 00:04:38.267
And worse -- not having
an easy way to see them,

00:04:38.301 --> 00:04:41.534
because in visual illusions, we can
easily demonstrate the mistakes;

00:04:41.567 --> 00:04:43.634
in cognitive illusion
it's much, much harder

00:04:43.667 --> 00:04:45.467
to demonstrate the mistakes to people.

00:04:45.501 --> 00:04:48.467
So I want to show you
some cognitive illusions,

00:04:48.501 --> 00:04:51.567
or decision-making illusions,
in the same way.

00:04:51.601 --> 00:04:55.067
And this is one of my favorite
plots in social sciences.

00:04:55.101 --> 00:04:58.934
It's from a paper
by Johnson and Goldstein.

00:04:58.967 --> 00:05:02.667
It basically shows the percentage
of people who indicated

00:05:02.701 --> 00:05:06.434
they would be interested
in donating their organs.

00:05:06.434 --> 00:05:08.367
These are different countries in Europe.

00:05:08.401 --> 00:05:10.367
You basically see two types of countries:

00:05:10.367 --> 00:05:12.967
countries on the right,
that seem to be giving a lot;

00:05:13.001 --> 00:05:15.967
and countries on the left
that seem to giving very little,

00:05:16.001 --> 00:05:17.601
or much less.

00:05:17.601 --> 00:05:18.767
The question is, why?

00:05:18.801 --> 00:05:22.101
Why do some countries give a lot
and some countries give a little?

00:05:22.134 --> 00:05:23.767
When you ask people this question,

00:05:23.801 --> 00:05:26.334
they usually think that it has
to be about culture.

00:05:26.334 --> 00:05:27.967
How much do you care about people?

00:05:28.001 --> 00:05:29.467
Giving organs to somebody else

00:05:29.501 --> 00:05:32.834
is probably about how much you care
about society, how linked you are.

00:05:32.867 --> 00:05:34.467
Or maybe it's about religion.

00:05:34.501 --> 00:05:36.467
But if you look at this plot,

00:05:36.467 --> 00:05:39.901
you can see that countries
that we think about as very similar,

00:05:39.934 --> 00:05:42.067
actually exhibit very different behavior.

00:05:42.701 --> 00:05:44.967
For example, Sweden
is all the way on the right,

00:05:45.001 --> 00:05:47.667
and Denmark, which we think
is culturally very similar,

00:05:47.667 --> 00:05:49.467
is all the way on the left.

00:05:49.501 --> 00:05:53.467
Germany is on the left,
and Austria is on the right.

00:05:53.501 --> 00:05:56.801
The Netherlands is on the left,
and Belgium is on the right.

00:05:56.834 --> 00:05:59.467
And finally, depending
on your particular version

00:05:59.501 --> 00:06:01.467
of European similarity,

00:06:01.501 --> 00:06:06.667
you can think about the U.K. and France
as either similar culturally or not,

00:06:06.701 --> 00:06:10.501
but it turns out that with organ
donation, they are very different.

00:06:10.534 --> 00:06:12.967
By the way, the Netherlands
is an interesting story.

00:06:13.001 --> 00:06:16.634
You see, the Netherlands is kind
of the biggest of the small group.

00:06:17.834 --> 00:06:20.467
It turns out that they got to 28 percent

00:06:20.501 --> 00:06:23.767
after mailing every household
in the country a letter,

00:06:23.801 --> 00:06:26.467
begging people to join
this organ donation program.

00:06:27.001 --> 00:06:29.967
You know the expression,
"Begging only gets you so far."

00:06:30.001 --> 00:06:32.467
It's 28 percent in organ donation.

00:06:32.501 --> 00:06:34.367
(Laughter)

00:06:34.767 --> 00:06:37.134
But whatever the countries
on the right are doing,

00:06:37.134 --> 00:06:39.267
they're doing a much
better job than begging.

00:06:39.267 --> 00:06:40.601
So what are they doing?

00:06:40.601 --> 00:06:44.067
Turns out the secret has to do
with a form at the DMV.

00:06:44.101 --> 00:06:45.467
And here is the story.

00:06:45.501 --> 00:06:48.001
The countries on the left
have a form at the DMV

00:06:48.034 --> 00:06:49.734
that looks something like this.

00:06:49.734 --> 00:06:54.034
"Check the box below if you want to
participate in the organ donor program."

00:06:54.067 --> 00:06:55.467
And what happens?

00:06:55.501 --> 00:06:57.801
People don't check, and they don't join.

00:06:58.501 --> 00:07:01.001
The countries on the right,
the ones that give a lot,

00:07:01.034 --> 00:07:02.634
have a slightly different form.

00:07:02.667 --> 00:07:06.167
It says, "Check the box below
if you don't want to participate ..."

00:07:06.167 --> 00:07:08.367
Interestingly enough,
when people get this,

00:07:08.401 --> 00:07:10.467
they again don't check, but now they join.

00:07:10.501 --> 00:07:13.467
(Laughter)

00:07:13.501 --> 00:07:16.401
Now, think about what this means.

00:07:16.901 --> 00:07:20.467
You know, we wake up in the morning
and we feel we make decisions.

00:07:20.501 --> 00:07:22.801
We wake up in the morning
and we open the closet;

00:07:22.834 --> 00:07:24.634
we feel that we decide what to wear.

00:07:24.634 --> 00:07:27.701
we open the refrigerator and we feel
that we decide what to eat.

00:07:27.734 --> 00:07:29.234
What this is actually saying,

00:07:29.267 --> 00:07:32.067
is that many of these decisions
are not residing within us.

00:07:32.101 --> 00:07:34.867
They are residing in the person
who is designing that form.

00:07:35.234 --> 00:07:37.167
When you walk into the DMV,

00:07:37.201 --> 00:07:40.201
the person who designed the form
will have a huge influence

00:07:40.201 --> 00:07:42.034
on what you'll end up doing.

00:07:42.034 --> 00:07:44.834
Now, it's also very hard
to intuit these results.

00:07:44.867 --> 00:07:46.201
Think about it for yourself.

00:07:46.201 --> 00:07:47.467
How many of you believe

00:07:47.501 --> 00:07:49.801
that if you went to renew
your license tomorrow,

00:07:49.801 --> 00:07:51.467
and you went to the DMV,

00:07:51.501 --> 00:07:53.767
and you encountered one of these forms,

00:07:53.801 --> 00:07:56.267
that it would actually
change your own behavior?

00:07:56.267 --> 00:07:58.567
Very hard to think
that it would influence us.

00:07:58.601 --> 00:08:02.134
We can say, "Oh, these funny Europeans,
of course it would influence them."

00:08:02.167 --> 00:08:03.934
But when it comes to us,

00:08:03.967 --> 00:08:06.601
we have such a feeling
that we're in the driver's seat,

00:08:06.634 --> 00:08:09.901
such a feeling that we're in control
and we are making the decision,

00:08:09.934 --> 00:08:12.534
that it's very hard
to even accept the idea

00:08:12.567 --> 00:08:15.234
that we actually have an illusion
of making a decision,

00:08:15.267 --> 00:08:17.334
rather than an actual decision.

00:08:17.334 --> 00:08:19.801
Now, you might say,

00:08:19.834 --> 00:08:22.267
"These are decisions we don't care about."

00:08:22.267 --> 00:08:24.534
In fact, by definition,
these are decisions

00:08:24.567 --> 00:08:27.167
about something that will happen
to us after we die.

00:08:27.167 --> 00:08:29.301
How could we care about something less

00:08:29.334 --> 00:08:31.601
than about something
that happens after we die?

00:08:31.634 --> 00:08:34.567
So a standard economist,
somebody who believes in rationality,

00:08:34.567 --> 00:08:35.834
would say, "You know what?

00:08:35.867 --> 00:08:39.301
The cost of lifting the pencil
and marking a "V" is higher

00:08:39.334 --> 00:08:41.467
than the possible benefit of the decision,

00:08:41.501 --> 00:08:43.301
so that's why we get this effect."

00:08:43.334 --> 00:08:44.334
(Laughter)

00:08:44.367 --> 00:08:46.867
But, in fact, it's not because it's easy.

00:08:46.867 --> 00:08:49.901
It's not because it's trivial.
It's not because we don't care.

00:08:49.934 --> 00:08:52.334
It's the opposite. It's because we care.

00:08:52.367 --> 00:08:54.301
It's difficult and it's complex.

00:08:54.334 --> 00:08:56.867
And it's so complex
that we don't know what to do.

00:08:56.901 --> 00:08:58.867
And because we have no idea what to do,

00:08:58.901 --> 00:09:02.467
we just pick whatever it was
that was chosen for us.

00:09:03.101 --> 00:09:04.667
I'll give you one more example.

00:09:04.667 --> 00:09:07.267
This is from a paper
by Redelmeier and Shafir.

00:09:07.301 --> 00:09:10.801
And they said, "Would this
effect also happens to experts?

00:09:10.801 --> 00:09:13.934
People who are well-paid,
experts in their decisions,

00:09:13.967 --> 00:09:15.334
and who make a lot of them?"

00:09:15.334 --> 00:09:17.534
And they took a group of physicians.

00:09:17.567 --> 00:09:20.001
They presented to them
a case study of a patient.

00:09:20.034 --> 00:09:23.467
They said, "Here is a patient.
He is a 67-year-old farmer.

00:09:23.501 --> 00:09:26.234
He's been suffering from
right hip pain for a while."

00:09:26.267 --> 00:09:28.101
And then, they said to the physicians,

00:09:28.101 --> 00:09:29.467
"You decided a few weeks ago

00:09:29.501 --> 00:09:31.467
that nothing is working for this patient.

00:09:31.501 --> 00:09:33.934
All these medications,
nothing seems to be working.

00:09:33.967 --> 00:09:36.734
So you refer the patient
for hip replacement therapy.

00:09:36.734 --> 00:09:38.467
Hip replacement. Okay?"

00:09:38.501 --> 00:09:41.467
So the patient is on a path
to have his hip replaced.

00:09:41.967 --> 00:09:43.934
Then they said to half of the physicians,

00:09:43.934 --> 00:09:46.301
"Yesterday, you reviewed
the patient's case,

00:09:46.301 --> 00:09:48.934
and you realized that you forgot
to try one medication.

00:09:48.934 --> 00:09:50.534
You did not try ibuprofen.

00:09:51.501 --> 00:09:54.834
What do you do? Do you pull
the patient back and try ibuprofen?

00:09:54.834 --> 00:09:57.634
Or do you let him go
and have hip replacement?"

00:09:57.634 --> 00:10:00.301
Well, the good news is
that most physicians in this case

00:10:00.334 --> 00:10:03.134
decided to pull the patient
and try ibuprofen.

00:10:03.167 --> 00:10:04.901
Very good for the physicians.

00:10:04.901 --> 00:10:06.967
To the other group
of physicians, they said,

00:10:07.001 --> 00:10:10.867
"Yesterday when you reviewed the case,
you discovered there were two medications

00:10:10.867 --> 00:10:13.401
you didn't try out yet --
ibuprofen and piroxicam."

00:10:13.434 --> 00:10:15.667
You have two medications
you didn't try out yet.

00:10:15.701 --> 00:10:18.334
What do you do? You let him go,
or you pull him back?

00:10:18.367 --> 00:10:21.801
And if you pull him back, do you try
ibuprofen or piroxicam? Which one?"

00:10:21.834 --> 00:10:23.001
Now, think of it:

00:10:23.001 --> 00:10:26.867
This decision makes it as easy to let
the patient continue with hip replacement,

00:10:26.901 --> 00:10:30.034
but pulling him back, all of the sudden
it becomes more complex.

00:10:30.067 --> 00:10:31.867
There is one more decision.

00:10:31.901 --> 00:10:33.234
What happens now?

00:10:33.234 --> 00:10:36.867
The majority of the physicians
now choose to let the patient go

00:10:36.901 --> 00:10:38.401
for a hip replacement.

00:10:38.434 --> 00:10:40.467
I hope this worries you, by the way --

00:10:40.501 --> 00:10:41.734
(Laughter)

00:10:41.767 --> 00:10:43.467
when you go to see your physician.

00:10:44.134 --> 00:10:46.834
The thing is that
no physician would ever say,

00:10:46.867 --> 00:10:50.667
"Piroxicam, ibuprofen, hip replacement.
Let's go for hip replacement."

00:10:50.701 --> 00:10:53.467
But the moment you set this
as the default,

00:10:53.501 --> 00:10:57.467
it has a huge power over whatever
people end up doing.

00:10:57.501 --> 00:11:00.834
I'll give you a couple of more examples
on irrational decision-making.

00:11:00.867 --> 00:11:02.667
Imagine I give you a choice:

00:11:02.701 --> 00:11:06.301
Do you want to go for a weekend to Rome,
all expenses paid --

00:11:06.334 --> 00:11:10.901
hotel, transportation, food,
a continental breakfast, everything --

00:11:10.934 --> 00:11:12.467
or a weekend in Paris?

00:11:12.501 --> 00:11:15.767
Now, weekend in Paris, weekend
in Rome -- these are different things.

00:11:15.767 --> 00:11:18.567
They have different food,
different culture, different art.

00:11:18.567 --> 00:11:21.934
Imagine I added a choice to the set
that nobody wanted.

00:11:21.967 --> 00:11:23.701
Imagine I said, "A weekend in Rome,

00:11:23.734 --> 00:11:25.101
a weekend in Paris,

00:11:25.134 --> 00:11:26.734
or having your car stolen?"

00:11:26.734 --> 00:11:29.767
(Laughter)

00:11:29.801 --> 00:11:33.034
It's a funny idea, because why
would having your car stolen,

00:11:33.067 --> 00:11:34.667
in this set, influence anything?

00:11:34.701 --> 00:11:36.801
(Laughter)

00:11:36.834 --> 00:11:41.267
But what if the option to have your car
stolen was not exactly like this?

00:11:41.301 --> 00:11:43.834
What if it was a trip to Rome,
all expenses paid,

00:11:43.867 --> 00:11:46.134
transportation, breakfast,

00:11:46.167 --> 00:11:48.467
but it doesn't include
coffee in the morning?

00:11:48.501 --> 00:11:51.867
If you want coffee, you have to pay
for it yourself, it's two euros 50.

00:11:51.901 --> 00:11:52.901
(Laughter)

00:11:52.934 --> 00:11:54.467
Now in some ways,

00:11:54.501 --> 00:11:56.801
given that you can have Rome with coffee,

00:11:56.834 --> 00:11:59.201
why would you possibly
want Rome without coffee?

00:11:59.234 --> 00:12:02.167
It's like having your car stolen.
It's an inferior option.

00:12:02.801 --> 00:12:03.967
But guess what happened?

00:12:04.001 --> 00:12:06.301
The moment you add Rome without coffee,

00:12:06.334 --> 00:12:09.934
Rome with coffee becomes more popular,
and people choose it.

00:12:09.967 --> 00:12:12.467
The fact that you have Rome without coffee

00:12:12.501 --> 00:12:14.467
makes Rome with coffee look superior,

00:12:14.501 --> 00:12:17.467
and not just to Rome without coffee --
even superior to Paris.

00:12:17.501 --> 00:12:21.567
(Laughter)

00:12:21.601 --> 00:12:24.067
Here are two examples of this principle.

00:12:24.067 --> 00:12:26.734
This was an ad in The Economist
a few years ago

00:12:26.767 --> 00:12:28.834
that gave us three choices:

00:12:28.867 --> 00:12:31.701
an online subscription for 59 dollars,

00:12:31.701 --> 00:12:35.201
a print subscription for 125 dollars,

00:12:35.234 --> 00:12:37.467
or you could get both for 125.

00:12:37.501 --> 00:12:39.467
(Laughter)

00:12:39.501 --> 00:12:42.101
Now I looked at this,
and I called up The Economist,

00:12:42.101 --> 00:12:44.667
and I tried to figure out
what they were thinking.

00:12:44.701 --> 00:12:47.834
And they passed me from one person
to another to another,

00:12:47.867 --> 00:12:51.601
until eventually I got to the person
who was in charge of the website,

00:12:51.601 --> 00:12:55.201
and I called them up, and they went
to check what was going on.

00:12:55.234 --> 00:12:59.067
The next thing I know,
the ad is gone, no explanation.

00:12:59.101 --> 00:13:01.001
So I decided to do the experiment

00:13:01.001 --> 00:13:03.667
that I would have loved
The Economist to do with me.

00:13:03.667 --> 00:13:06.067
I took this and I gave it
to 100 MIT students.

00:13:06.101 --> 00:13:07.667
I said, "What would you choose?"

00:13:07.701 --> 00:13:11.667
These are the market shares --
most people wanted the combo deal.

00:13:11.701 --> 00:13:13.901
Thankfully, nobody wanted
the dominant option.

00:13:13.934 --> 00:13:15.567
That means our students can read.

00:13:15.601 --> 00:13:16.967
(Laughter)

00:13:17.001 --> 00:13:19.901
But now, if you have an option
that nobody wants,

00:13:19.901 --> 00:13:21.601
you can take it off, right?

00:13:21.634 --> 00:13:23.467
So I printed another version of this,

00:13:23.501 --> 00:13:25.334
where I eliminated the middle option.

00:13:25.334 --> 00:13:28.834
I gave it to another 100 students.
Here is what happened:

00:13:28.834 --> 00:13:31.601
Now the most popular option
became the least popular,

00:13:31.601 --> 00:13:34.267
and the least popular
became the most popular.

00:13:35.134 --> 00:13:38.467
What was happening
was the option that was useless,

00:13:38.501 --> 00:13:42.367
in the middle, was useless
in the sense that nobody wanted it.

00:13:42.367 --> 00:13:45.534
But it wasn't useless in the sense
that it helped people figure out

00:13:45.567 --> 00:13:46.701
what they wanted.

00:13:46.734 --> 00:13:49.467
In fact, relative
to the option in the middle,

00:13:49.501 --> 00:13:54.001
which was get only the print for 125,

00:13:54.034 --> 00:13:57.767
the print and web for 125
looked like a fantastic deal.

00:13:57.801 --> 00:13:59.901
And as a consequence, people chose it.

00:13:59.934 --> 00:14:01.567
The general idea here, by the way,

00:14:01.567 --> 00:14:04.267
is that we actually don't know
our preferences that well.

00:14:04.301 --> 00:14:06.767
And because we don't know
our preferences that well,

00:14:06.801 --> 00:14:10.067
we're susceptible to all of these
influences from the external forces:

00:14:10.101 --> 00:14:13.767
the defaults, the particular options
that are presented to us, and so on.

00:14:14.134 --> 00:14:15.467
One more example of this.

00:14:15.501 --> 00:14:18.867
People believe that when we deal
with physical attraction,

00:14:18.867 --> 00:14:22.134
we see somebody, and we know immediately
whether we like them or not,

00:14:22.167 --> 00:14:23.467
if we're attracted or not.

00:14:23.501 --> 00:14:25.934
This is why we have
these four-minute dates.

00:14:25.967 --> 00:14:28.467
So I decided to do
this experiment with people.

00:14:28.501 --> 00:14:32.467
I'll show you images here, no real people,
but the experiment was with people.

00:14:32.501 --> 00:14:35.467
I showed some people a picture
of Tom, and a picture of Jerry.

00:14:35.501 --> 00:14:37.567
and I said, "Who do you want to date?

00:14:37.601 --> 00:14:39.067
Tom or Jerry?"

00:14:39.101 --> 00:14:42.367
But for half the people,
I added an ugly version of Jerry.

00:14:42.401 --> 00:14:47.467
I took Photoshop and I made
Jerry slightly less attractive.

00:14:47.501 --> 00:14:48.901
(Laughter)

00:14:48.934 --> 00:14:52.467
For the other people, I added
an ugly version of Tom.

00:14:52.501 --> 00:14:55.467
And the question was,
will ugly Jerry and ugly Tom

00:14:55.501 --> 00:14:59.267
help their respective,
more attractive brothers?

00:14:59.301 --> 00:15:01.534
The answer was absolutely yes.

00:15:01.534 --> 00:15:03.701
When ugly Jerry was around,
Jerry was popular.

00:15:03.701 --> 00:15:05.867
When ugly Tom was around, Tom was popular.

00:15:05.867 --> 00:15:07.467
(Laughter)

00:15:07.501 --> 00:15:10.001
This of course has two
very clear implications

00:15:10.034 --> 00:15:12.334
for life in general.

00:15:13.301 --> 00:15:16.467
If you ever go bar-hopping,
who do you want to take with you?

00:15:16.501 --> 00:15:22.234
(Laughter)

00:15:22.267 --> 00:15:26.034
You want a slightly uglier
version of yourself.

00:15:26.067 --> 00:15:27.467
(Laughter)

00:15:27.501 --> 00:15:29.734
Similar, but slightly uglier.

00:15:29.767 --> 00:15:31.201
(Laughter)

00:15:31.234 --> 00:15:35.001
The second point, or course, is that
if somebody invites you to bar hop,

00:15:35.034 --> 00:15:36.767
you know what they think about you.

00:15:36.801 --> 00:15:39.467
(Laughter)

00:15:39.501 --> 00:15:41.767
Now you get it.

00:15:41.801 --> 00:15:43.467
What is the general point?

00:15:43.501 --> 00:15:44.767
The general point is that,

00:15:44.767 --> 00:15:48.467
when we think about economics, we have
this beautiful view of human nature.

00:15:48.501 --> 00:15:51.034
"What a piece of work is a man!
How noble in reason!"

00:15:51.067 --> 00:15:53.634
We have this view of ourselves, of others.

00:15:53.667 --> 00:15:58.867
The behavioral economics perspective
is slightly less "generous" to people;

00:15:58.901 --> 00:16:01.001
in fact, in medical terms,

00:16:01.034 --> 00:16:02.201
that's our view.

00:16:02.234 --> 00:16:07.367
(Laughter)

00:16:07.401 --> 00:16:09.467
But there is a silver lining.

00:16:09.501 --> 00:16:11.367
The silver lining is, I think,

00:16:11.401 --> 00:16:15.467
kind of the reason that behavioral
economics is interesting and exciting.

00:16:15.501 --> 00:16:18.367
Are we Superman, or are we Homer Simpson?

00:16:18.801 --> 00:16:21.767
When it comes to building
the physical world,

00:16:21.801 --> 00:16:23.834
we kind of understand our limitations.

00:16:23.867 --> 00:16:25.034
We build steps.

00:16:25.067 --> 00:16:28.201
And we build these things
that not everybody can use, obviously.

00:16:28.234 --> 00:16:30.101
(Laughter)

00:16:30.134 --> 00:16:31.701
We understand our limitations,

00:16:31.701 --> 00:16:33.367
and we build around them.

00:16:33.401 --> 00:16:35.967
But for some reason, when it comes
to the mental world,

00:16:36.001 --> 00:16:39.334
when we design things like healthcare
and retirement and stock markets,

00:16:39.367 --> 00:16:41.701
we somehow forget the idea
that we are limited.

00:16:41.734 --> 00:16:44.634
I think that if we understood
our cognitive limitations

00:16:44.667 --> 00:16:47.267
in the same way we understand
our physical limitations,

00:16:47.301 --> 00:16:50.134
even though they don't stare us
in the face the same way,

00:16:50.134 --> 00:16:52.634
we could design a better world,
and that, I think,

00:16:52.634 --> 00:16:54.034
is the hope of this thing.

00:16:54.067 --> 00:16:55.467
Thank you very much.

00:16:55.501 --> 00:17:02.134
(Applause)