0:00:00.111,0:00:02.271
- [Colin] We have a saying [br]in neuroscience sometimes:

0:00:02.271,0:00:04.253
"Don't ask the person,[br]ask the brain,"

0:00:04.253,0:00:07.606
because the brain activity[br]may be something that's less

0:00:07.606,0:00:09.256
than fully conscious.

0:00:09.678,0:00:11.795
♪ [music] ♪

0:00:18.815,0:00:22.015
My research is about behavioral [br]economics and neuroeconomics.

0:00:22.015,0:00:25.138
And behavioral economics is using [br]ideas from psychology

0:00:25.138,0:00:27.739
and other social sciences[br]to make economics

0:00:27.739,0:00:31.310
a little bit more lifelike [br]and fit human behavior.

0:00:31.310,0:00:33.761
The neuroeconomics part [br]is that we actually try to see

0:00:33.761,0:00:35.211
what's happening in the brain

0:00:35.211,0:00:37.210
when people are making [br]economic decisions.

0:00:37.210,0:00:40.310
Hypothetical bias is a term [br]for when you ask somebody

0:00:40.310,0:00:43.127
whether they're going to something,[br]but there's no actual consequences,

0:00:43.127,0:00:45.042
like a lot of surveys:[br]"Are you going to vote?"

0:00:45.042,0:00:47.061
"Will you buy [br]this new product we have?"

0:00:47.621,0:00:50.397
You tend to get a kind [br]of upward "yes" bias.

0:00:50.397,0:00:52.793
People are more likely to say,[br]"Yeah, I'd think I'd buy it,"

0:00:52.793,0:00:54.239
or, "Oh yeah, I'm planning to vote."

0:00:54.239,0:00:56.126
The hypothetical bias can [br]be pretty high,

0:00:56.131,0:00:57.881
and it can be also retrospective.

0:00:57.881,0:00:59.529
So if you ask people did they vote,

0:00:59.529,0:01:02.647
70% say yes and really [br]the answer was 45%.

0:01:03.297,0:01:05.097
One application is [br]in things like marketing.

0:01:05.097,0:01:07.913
A lot of new products fail,[br]and one reason they fail

0:01:07.913,0:01:09.631
is because when [br]they test market it,

0:01:09.631,0:01:11.414
a lot of people said, [br]"Yes, I'd buy it,"

0:01:11.414,0:01:12.630
who weren't going to buy it.

0:01:12.630,0:01:14.247
One thing that people [br]have been chasing

0:01:14.247,0:01:16.780
in different fields in economics[br]and psychology is

0:01:16.861,0:01:20.044
how can we measure the size [br]of this bias and adjust for it.

0:01:20.044,0:01:22.862
So that if 70% of the people say[br]they're going to buy a new product,

0:01:22.862,0:01:25.683
we know that [br]the real number is 45%.

0:01:29.619,0:01:32.682
We did a couple of studies using[br]brain imaging to say

0:01:32.742,0:01:34.960
is there a kind of signature [br]in the brain of when

0:01:34.960,0:01:37.026
somebody says, [br]"Yes, I would buy it,"

0:01:37.026,0:01:39.309
but when they really have [br]to choose, they say no.

0:01:39.461,0:01:41.442
So we showed them pictures[br]of different goods,

0:01:41.442,0:01:43.601
and the first part [br]of the experiment, we asked them,

0:01:43.601,0:01:46.715
"Would you pay $27 [br]for this backpack, yes or no?"

0:01:46.715,0:01:49.261
That's the hypothetical part,[br]and then we kind of surprised them

0:01:49.261,0:01:51.631
when they come out of the scanner[br]and we say to them,

0:01:51.631,0:01:54.042
"Oh by the way, now we're going [br]to actually have you decide

0:01:54.042,0:01:56.436
to spend money [br]so we're going to give you $50."

0:01:56.436,0:01:59.550
If you want to buy the backpack[br]for $27, we're going to take it

0:01:59.550,0:02:02.453
out of your 50, so that now[br]they have to make real decisions,

0:02:02.560,0:02:04.694
and then we study [br]in the brain imaging,

0:02:04.694,0:02:07.776
could we tell what areas [br]were saying yes,

0:02:07.776,0:02:11.011
but actually would later say no[br]compared to the areas that said,

0:02:11.011,0:02:14.327
"Yes, I think I'll buy it,"[br]and, yes, they really did buy it.

0:02:14.452,0:02:17.658
And we found both more activity[br]in certain regions associated

0:02:17.658,0:02:20.375
with valuation and then activity [br]in different regions,

0:02:20.375,0:02:23.507
which were somewhat predictive[br]of when a yes was going to turn

0:02:23.507,0:02:25.290
into, "Well, not really."

0:02:28.667,0:02:30.748
In some other studies [br]on hypothetical bias,

0:02:30.748,0:02:33.068
we used eye tracking, [br]which is a computerized way

0:02:33.068,0:02:35.401
of seeing what you're looking at[br]for how long.

0:02:35.467,0:02:37.249
It also measures pupil dilation.

0:02:37.249,0:02:39.499
When you're aroused [br]by something that you like,

0:02:39.499,0:02:42.382
or possibly you're scared,[br]the pupil dilates a little bit.

0:02:42.701,0:02:44.871
So we used that method[br]and then another method

0:02:44.871,0:02:46.337
using mouse movements.

0:02:46.337,0:02:48.005
And so we found [br]that the mouse tracking

0:02:48.005,0:02:50.337
and the eye tracking[br]actually could give us an idea

0:02:50.337,0:02:52.974
of when people would say, "Yes, [br]I'm going to buy this product,"

0:02:52.974,0:02:55.282
but then they really didn't[br]when they had skin in the game

0:02:55.282,0:02:56.409
and had to buy it.

0:02:56.498,0:03:00.283
The quicker a person moves a mouse[br]to a box to click on something,

0:03:00.283,0:03:02.040
the more they like it.

0:03:02.424,0:03:04.924
It's like a fast trajectory,[br]"I really like this."

0:03:05.190,0:03:08.090
A slow meandering trajectory is, [br]"Well, I don't know. I'm not sure."

0:03:08.090,0:03:12.457
The motor activity in the mouse is[br]actually an index in economic value

0:03:12.457,0:03:15.273
as well as some other things,[br]like indecision.

0:03:19.701,0:03:21.624
We live in a kind of golden age[br]of social science

0:03:21.650,0:03:24.483
in which we can measure things[br]in lots and lots of different ways.

0:03:24.483,0:03:25.819
So one thing we've just begun

0:03:25.819,0:03:28.150
that I think it's going [br]to be really, really fun

0:03:28.181,0:03:32.448
is in order to study habit[br]and a bunch of other things.

0:03:32.448,0:03:35.315
We bought a smart vending machine,[br]and a smart vending machine

0:03:35.315,0:03:37.380
is basically a vending machine [br]in the back

0:03:37.380,0:03:39.129
and a giant iPad in the front.

0:03:39.217,0:03:42.285
So you could program the iPad[br]to show whatever you want.

0:03:42.285,0:03:44.850
For example, if somebody buys[br]habitually and you raise the price

0:03:44.850,0:03:47.434
by a few percent,[br]do they just ignore that

0:03:47.434,0:03:49.299
because they're not [br]even looking at the price?

0:03:49.381,0:03:52.560
Eventually, we'd like also[br]to be able to use webcam cameras

0:03:52.628,0:03:55.226
to record so when I say[br]somebody is not looking at the price,

0:03:55.226,0:03:57.537
I really mean they [br]are not looking at the price

0:03:57.537,0:03:59.903
because the camera looked[br]at where their eyes were looking.

0:04:00.065,0:04:02.398
- [Narrator] Want to see [br]more economists in the wild?

0:04:02.398,0:04:03.764
Check out our playlist.

0:04:03.764,0:04:05.148
Are you a teacher?

0:04:05.148,0:04:07.628
Here's some related material [br]for your classroom.

0:04:07.868,0:04:09.641
♪ [music] ♪