Popular in Economics

Economics

Gender and Economics

University of California Santa Barbara Shelly Lundberg

Banks and Policy Organisations

Greening the financial sector

European Investment Bank Matteo Ferrazzi

Economic History

Gravity with Gravitas: A Solution to the Border Puzzle

Boston College James E. Anderson

Economic Inequalities

Visions of Inequality

City University of New York Branko Milanovic

Economics

School Finance Equalization Increases Intergenerational Mobility

Yale University Barbara Biasi

Economics

Olivier Blanchard on U.S. Pandemic-Era Inflation

Peterson Institute for International Economics Olivier Blanchard

Eye-image as nonverbal social cue has asymmetric gender effects in dictator taking games

View more insights in Social Sciences - Economics - Behavioral economics

View more figures from University of Sheffield

View the page of Subhasish Modak Chowdhury

Read the Study

Dictator giving games often demonstrate that nonverbal social cues, such as drawn-in eyes on display, induce pro-social behavior in the form of giving more. Notably, sometimes this effect is seen to differ between males and females. Subhasish M. Chowdhury disucsses the effects of social cues on negative behavior along with its gender dimension.

Image courtesy of interviewee. May 8, 2024

Generating Workspace

Export as PDF

This video video discusses a research study exploring human behavior in economic decision-making, particularly focusing on social preferences and gender differences. Traditional neoclassical economics assumes individuals act solely out of self-interest, whereas behavioral economics acknowledges that people often show concern for others, a concept referred to as social preference.

The research utilizes a laboratory experiment known as the "dictator game" and a variation called the "taking game." In these games, participants (dictators) decide how much money to take from another participant (recipient) who has a fixed amount. The primary objective is to observe how the presence of social cues, specifically images of eyes, influences participants' decisions to take money, thereby indicating their social preferences.

Two hypotheses are tested: first, that the presence of eyes would reduce the amount taken (to counter negative behavior), and second, that this effect would be stronger for males, reflecting existing literature that suggests males are more influenced by social cues compared to females.

The findings indicate that, while the overall amount taken did not change significantly with the presence of eyes, notable gender differences emerged. Males reduced their taking behavior when eyes were present, while females increased theirs. This video video unexpected shift suggests that social cues can have different effects based on gender.

The study concludes that understanding these dynamics is essential for implementing effective policies that rely on social cues, as they can lead to unintended consequences depending on the audience's gender distribution. The authors emphasize the importance of further research to understand the mechanisms behind these behaviors and how they can inform economic and social policies.

Language:

Export as PDF

X-TIMESTAMP-MAP:LOCAL=00:00:00.000,MPEGTS=900000

00:00.035 --> 00:04.245
When we usually talk about economics, the,

00:04.445 --> 00:08.245
neoclassical economics, we, the theory, economic theory

00:08.245 --> 00:10.445
that assume that, people,

00:10.615 --> 00:12.765
human being are homo economics,

00:13.585 --> 00:15.445
who always try to think about themselves.

00:15.445 --> 00:17.365
They try to maximize their own payoff.

00:18.385 --> 00:20.445
Uh, they don't think of other people and so on.

00:20.715 --> 00:23.285
However, when you see ourselves,

00:23.625 --> 00:25.085
we reflect ourselves on the street

00:25.085 --> 00:26.965
and so on, we see that people actually try

00:26.965 --> 00:29.645
to help each other while walking on the street.

00:29.665 --> 00:31.245
If you see a homeless person asking

00:31.305 --> 00:32.925
for food, probably will buy some food.

00:33.425 --> 00:36.925
Now, that doesn't match with the formal theory that,

00:36.985 --> 00:38.365
new classical economics has done.

00:38.745 --> 00:40.565
So here are behavioral economics comes

00:41.305 --> 00:43.685
and they say that, okay, in our preferences,

00:44.795 --> 00:46.255
people actually care for others.

00:47.475 --> 00:51.355
And that caring for other is a part,

00:51.355 --> 00:52.755
something called social preference.

00:52.895 --> 00:55.115
We have some preference for other people in the society.

00:56.015 --> 00:58.275
Now, how do you measure social preference?

00:59.125 --> 01:00.355
There are many ways to do it,

01:00.355 --> 01:02.715
but one way to do it is essentially

01:02.735 --> 01:04.315
to run laboratory experiment.

01:05.295 --> 01:08.555
In laboratory experiment you bring actual human subjects

01:09.615 --> 01:11.115
who go through some instruction

01:11.115 --> 01:12.675
and play some economic gains.

01:13.975 --> 01:16.755
And one of those economic game is called dictator game.

01:17.185 --> 01:19.235
It's a very simple decision making process

01:19.895 --> 01:23.115
in which those human beings, they come to the laboratory,

01:23.225 --> 01:26.595
they don't see anybody, they're paired with,

01:26.595 --> 01:28.595
some other people, they don't see that person.

01:28.945 --> 01:31.475
However, they know that they,

01:31.475 --> 01:33.795
they are given by the experimenter.

01:33.795 --> 01:35.755
They are given some amount of money.

01:36.415 --> 01:38.075
So it can be $10

01:38.095 --> 01:40.115
or 10 pounds depending on where you are running it.

01:40.855 --> 01:43.195
And in the dictator game, they will be asked,

01:43.495 --> 01:45.235
now this money is yours,

01:45.295 --> 01:48.235
but you have the opportunity to give some part of it

01:48.535 --> 01:51.235
to someone else, in this experiment,

01:52.135 --> 01:54.115
and the other person will not be able

01:54.135 --> 01:56.795
to say anything about how much they will get.

01:57.985 --> 02:00.725
So it's totally your decision. You can give whatever.

02:00.945 --> 02:03.805
So if you give two, then it'll be eight euros

02:03.805 --> 02:04.845
and two other persons.

02:05.105 --> 02:07.045
If you give five, it'll be five five.

02:07.045 --> 02:08.845
And so the other person will have to accept that

02:08.845 --> 02:11.125
and you'll go home with that actual money.

02:13.325 --> 02:15.685
Now, standard economic theory will

02:15.685 --> 02:16.765
say, I will not give anything.

02:16.875 --> 02:19.565
Take the whole 10 pounds, go home, be happy.

02:19.585 --> 02:20.605
The other person will be visible.

02:21.105 --> 02:22.245
But that doesn't actually happen.

02:22.915 --> 02:25.645
Just like, what I said, people have social preference.

02:26.025 --> 02:27.685
Uh, people feel bad not to give anything.

02:28.305 --> 02:30.195
And the amount that is given

02:30.695 --> 02:31.915
and how many times it's given,

02:32.285 --> 02:34.555
those are different measures of social preference.

02:35.955 --> 02:39.775
So dictator game is used in one way to try

02:39.775 --> 02:41.815
to find out people's social preference in this case,

02:41.955 --> 02:44.375
how nice people are, how much they're giving.

02:44.845 --> 02:48.255
Yeah. So, this particular study that came out in,

02:48.365 --> 02:50.255
general Behavioral and Experimental Economics,

02:50.255 --> 02:51.455
that's actually a joint work

02:51.455 --> 02:53.975
with Julian John from University of Reading

02:54.275 --> 02:56.575
and from Ham University.

02:57.475 --> 02:58.535
Uh, that

02:58.545 --> 03:02.845
Particular study, we use a version of the Dictator game.

03:03.125 --> 03:06.405
I just told you the most, popular version

03:06.405 --> 03:09.445
of the Ator game, which I call giving game in which I have

03:09.445 --> 03:10.885
10 pounds as a subject.

03:11.085 --> 03:13.725
I can give something. What we use is a different version.

03:13.745 --> 03:14.885
We call it taking game.

03:15.985 --> 03:19.955
In that case, still, I am making the decision

03:20.375 --> 03:22.115
and that's, that's why I'm the dictator is

03:22.115 --> 03:23.435
for a dictator again, because, you know,

03:23.435 --> 03:25.595
other person cannot say anything about my decision.

03:25.785 --> 03:27.275
They have to just take the money and leave.

03:27.855 --> 03:31.635
So in this case, we change the way, the game is framed.

03:32.215 --> 03:34.155
We say that, you know, the other person,

03:34.495 --> 03:37.395
we call them recipient, other person has 10 pounds,

03:39.005 --> 03:40.985
and you will have to make a decision

03:41.165 --> 03:42.585
how much you want to take from them.

03:43.855 --> 03:44.955
The other person cannot do anything.

03:45.235 --> 03:47.195
Whatever you want to take, take, and then you can go home.

03:47.665 --> 03:49.835
Yeah. Now see, it remains exactly the same.

03:50.375 --> 03:52.395
If I was dictator in a giving game

03:52.415 --> 03:54.635
and I had 10 pounds, I give you two pounds,

03:56.065 --> 03:59.085
then you'll have two, I'll have eight in a taking.

03:59.085 --> 04:00.965
Again, you have 10 pounds, I take eight.

04:01.555 --> 04:02.565
Then in the end, still

04:02.565 --> 04:03.845
it'll be, you'll have two, I'll have eight.

04:04.025 --> 04:05.605
It is important to use this frame

04:05.605 --> 04:08.845
because in this case, the property right

04:08.845 --> 04:10.445
of the 10 pounds is not with me.

04:10.445 --> 04:12.725
It's with you. So it's put in a negative frame

04:12.725 --> 04:14.405
that I'm actually taking something from you,

04:14.415 --> 04:15.685
which might be bad thing.

04:15.865 --> 04:19.245
We are trying to find out how people behave in this kind

04:19.245 --> 04:20.765
of situation when they are making

04:20.765 --> 04:21.925
this kind of negative behavior.

04:22.505 --> 04:23.845
And we focus on two things.

04:24.645 --> 04:28.775
First, there are different ways to nudge people

04:29.395 --> 04:31.335
to stop doing negative behavior

04:31.595 --> 04:33.015
or to data negative behavior,

04:33.395 --> 04:35.175
or to make people do good behavior.

04:35.625 --> 04:36.735
There are many different ways.

04:37.035 --> 04:39.975
One of the easiest way to do that is,

04:40.265 --> 04:42.015
shown in psychology, behavioral economics

04:42.015 --> 04:44.615
and so on, is to give something called social cue.

04:45.665 --> 04:49.565
Social cues are different cues that is either verbal,

04:49.715 --> 04:51.085
nonverbal and so on.

04:51.665 --> 04:54.205
Uh, that guide our social interaction.

04:55.525 --> 04:59.465
So it might be that when you are, talking to someone,

04:59.495 --> 05:02.225
someone talks you in a very nice sooth way

05:02.735 --> 05:03.785
that makes you feel better.

05:03.785 --> 05:05.695
That's a social cue. Okay?

05:05.695 --> 05:06.855
It can guide your conversation

05:07.715 --> 05:12.165
or if you are, going through a particular area,

05:12.545 --> 05:13.925
and you'll see there is a big notice.

05:14.235 --> 05:17.045
It's written that neighborhood wa neighborhood

05:17.045 --> 05:18.285
watch is in place.

05:19.065 --> 05:21.325
So be careful. So you can see that

05:21.355 --> 05:23.165
that makes you some error in this area.

05:24.545 --> 05:28.565
So one of the way social queue is used in experimental,

05:29.045 --> 05:31.365
research is to put in the when

05:31.365 --> 05:33.405
people are making decision in their computer screen

05:33.465 --> 05:37.685
or in their pen and paper to put a picture of a pair of eyes

05:38.855 --> 05:40.955
in that area, in the screen or in the paper.

05:42.055 --> 05:43.995
Now this picture of eyes is

05:43.995 --> 05:45.395
as if somebody is looking at you.

05:45.705 --> 05:47.635
That makes people more c conscious.

05:48.855 --> 05:52.675
It is shown in game often, in the giving game.

05:52.905 --> 05:55.355
When you put those pair of eyes, people give more

05:55.355 --> 05:58.005
to the other person because it made them more conscious.

05:58.005 --> 06:00.885
It made them feel like I should behave better and so on.

06:01.465 --> 06:05.365
So similarly, it is possible to stop people doing bad things

06:05.705 --> 06:07.325
by putting those pair of eyes.

06:07.865 --> 06:10.005
Uh, people have done it. Other researchers have done

06:10.005 --> 06:14.645
that in actual field, in which, the, in one university

06:15.155 --> 06:18.605
they have put this picture of ice with, saying

06:18.605 --> 06:20.005
that or cycle things.

06:20.025 --> 06:23.725
We are looking at you and that just next

06:23.725 --> 06:26.205
to their bike stands, you know, many of the universities,

06:26.505 --> 06:28.925
people steal parts of bikes

06:28.925 --> 06:31.765
or the whole, bike bicycle.

06:32.745 --> 06:34.485
So they put those posters

06:34.825 --> 06:38.485
and that reduced the, that kind of bad, incidences.

06:39.145 --> 06:42.175
So yes, it might be possible to,

06:42.205 --> 06:43.215
have this kind of effect.

06:44.155 --> 06:45.415
So we were trying, we,

06:45.475 --> 06:49.095
but you know, if it's done in the campus, you,

06:49.155 --> 06:51.655
you don't see exactly what is happening, who is doing what.

06:52.075 --> 06:53.535
So you don't have much control.

06:53.535 --> 06:56.375
You don't know exactly through which it's happening, right?

06:56.995 --> 07:00.375
So we want to take that, we want to check that.

07:00.835 --> 07:03.855
And what we did, we ran this taking game.

07:05.495 --> 07:08.315
So, there is 10 pounds, which somebody else,

07:08.455 --> 07:10.195
you can just take that money,

07:11.195 --> 07:12.635
whatever proportion you want from zero

07:12.655 --> 07:14.395
to 10, and then you can go home.

07:15.815 --> 07:17.755
So what we do with a set of people,

07:18.335 --> 07:19.955
we ran it just like that.

07:20.065 --> 07:23.155
Another set of people ran it when they see a

07:23.155 --> 07:24.315
pair of eyes looking at them.

07:27.625 --> 07:30.005
And we also know from the literature

07:30.235 --> 07:33.525
that this kind social queue has effect on

07:33.735 --> 07:35.045
males more than females.

07:35.355 --> 07:37.085
When we are looking at in how,

07:37.585 --> 07:39.645
how people are doing good things, you know.

07:40.745 --> 07:44.525
So, we try to look at gender difference.

07:44.665 --> 07:46.165
So we had this kind of dictators.

07:46.495 --> 07:48.245
50% of dictators were male,

07:48.735 --> 07:52.965
50% dictators were female in both of the settings without,

07:53.075 --> 07:54.525
with eyes and without eyes.

07:55.585 --> 07:57.645
And we try to look what is happening.

07:57.645 --> 08:01.325
There actually, fits in three STRs of literature,

08:01.345 --> 08:05.405
of course in the social preference, literature,

08:05.505 --> 08:08.125
why people are being nice to each other and so on.

08:08.125 --> 08:09.445
There are literature on pure

08:09.625 --> 08:14.125
and impure altruism in which pure altruism means people do

08:14.125 --> 08:17.725
nice to other people just because that gives them happiness.

08:18.425 --> 08:22.045
Impure altruism is people do nice things to other people

08:22.155 --> 08:24.805
because that make them feel good that I'm a nice person

08:25.385 --> 08:27.005
and that's why they do nice things.

08:27.425 --> 08:28.685
It can be reciprocity.

08:29.265 --> 08:30.965
So somebody did good things to me,

08:30.965 --> 08:32.925
that's why I'm doing good things back to them

08:33.385 --> 08:34.845
or somebody did good things to me.

08:35.065 --> 08:37.365
So that's why I'm doing good things to somebody else.

08:38.865 --> 08:40.445
So social queue is one

08:40.445 --> 08:42.565
of those areas in, social preference.

08:42.825 --> 08:45.845
So we are contributing to that in this area.

08:47.405 --> 08:50.975
Another area that we're contributing to is

08:51.715 --> 08:53.295
gender effects in dictator again.

08:54.605 --> 08:58.325
So it's quite well known that in dictator game,

08:58.505 --> 09:00.885
in a giving game, not in a taking game, in a giving game,

09:01.435 --> 09:02.925
females give more than males.

09:04.155 --> 09:06.245
What, what happens in taking game?

09:07.135 --> 09:11.545
Well, people looked at whether giving

09:11.565 --> 09:12.625
and taking game are similar.

09:13.085 --> 09:14.825
People give similar kind of amount.

09:15.285 --> 09:19.025
Uh, other literature has shown that, yes, overall,

09:19.125 --> 09:22.265
in giving game, if on an average out

09:22.265 --> 09:24.345
of a hundred people people are giving two pounds out

09:24.345 --> 09:27.255
of weight, then in a taking game they are taking eight

09:27.255 --> 09:29.375
pounds out, and leaving two.

09:30.225 --> 09:31.955
However, this has also shown

09:32.185 --> 09:34.715
that this is not symmetric across gender.

09:36.825 --> 09:41.245
So males take more and

09:41.305 --> 09:42.405
but males give less.

09:43.585 --> 09:44.595
Females are the opposite.

09:45.175 --> 09:47.635
As a result, overall distribution three and stem,

09:47.905 --> 09:49.715
however, there's a gender difference in between.

09:50.335 --> 09:52.275
So we are trying to contribute on that

09:53.505 --> 09:55.055
while looking at the taking in

09:55.835 --> 09:58.255
and putting those social skills in our

09:58.255 --> 10:00.215
experimental, detail.

10:00.355 --> 10:02.335
We, do the following.

10:03.695 --> 10:06.715
Um, we employ a pure taking in

10:07.975 --> 10:09.345
with two different treatments.

10:09.345 --> 10:11.385
You know, treatments means two different settings

10:11.385 --> 10:13.825
that are implemented in the laboratory.

10:15.365 --> 10:19.735
So pure taking error means in this case

10:20.335 --> 10:23.305
a subject who will be a dictator will know

10:23.655 --> 10:27.065
that the other subject who's a recipient has 10 pounds.

10:27.925 --> 10:30.865
And only thing that I can do is take anything from zero

10:30.865 --> 10:31.865
to 10 from that person.

10:33.115 --> 10:34.335
So it's a pure taking error.

10:34.335 --> 10:35.615
It's not that I can give and take

10:35.615 --> 10:38.975
and both, no, no, it's just a pure taking decision.

10:39.645 --> 10:42.935
Okay? Um, we have done this to not

10:42.935 --> 10:45.815
to have any other confo that is sh shown in the literature.

10:46.315 --> 10:50.135
Uh, we just wanted to find clear effect of the social cue.

10:50.135 --> 10:51.455
That's why you used, pure taking.

10:53.235 --> 10:56.825
And now what we did,

10:57.325 --> 11:00.585
in one treatment that is without any social cue.

11:01.555 --> 11:05.445
So we had 140 subjects

11:06.195 --> 11:10.505
coming to the laboratory out of them,

11:10.855 --> 11:14.425
70 would be, dictator and seven would be recipient.

11:15.775 --> 11:17.955
So we mashed one dictator with one recipient

11:19.305 --> 11:21.285
and they had to make only one decision.

11:21.565 --> 11:22.565
Recipient cannot make any decision.

11:22.585 --> 11:25.485
Of course, the dictator will have to make only one decision

11:25.915 --> 11:27.645
that, you know, the recipient has 10

11:27.645 --> 11:28.685
power, how much you want to take.

11:28.795 --> 11:30.165
They just write how much they want to take,

11:30.265 --> 11:31.405
and that's it done.

11:32.935 --> 11:37.485
So essentially we have 70 dictators playing this game

11:37.905 --> 11:39.125
and there is no picture

11:39.225 --> 11:42.845
of anything except we put a blood black box in

11:42.845 --> 11:43.885
their decision making screen.

11:46.245 --> 11:48.105
And remember, out of those 70 dictators,

11:48.205 --> 11:50.905
we have 35 male dictators and 35 female dictators.

11:53.515 --> 11:54.535
In the other treatment

11:54.535 --> 11:57.175
or the other setting, it's exactly same.

11:57.475 --> 12:01.575
140 subjects came 70 dictators, 35 male dictators,

12:01.575 --> 12:02.615
35 female dictators.

12:03.155 --> 12:05.415
But in the screen, instead of the blood black place,

12:05.845 --> 12:08.575
it's a rectangle with picture of eyes that looking at them.

12:09.275 --> 12:12.185
So that's essentially the setting.

12:12.565 --> 12:16.585
So we had two hypothesis that, stemmed from,

12:16.655 --> 12:18.065
what we know in the literature.

12:19.115 --> 12:21.815
So the first hypothesis was about the

12:22.015 --> 12:23.135
reduction in negative behavior.

12:23.795 --> 12:28.615
As I said. Um, there are, experiments run in the field

12:28.885 --> 12:31.215
that if you put this kind of eyes of

12:31.845 --> 12:34.695
like people looking at you, then people do bad things less

12:35.285 --> 12:36.455
like stealing from bike

12:37.405 --> 12:40.265
or there's also one, experiment in which they showed

12:40.265 --> 12:42.465
that people do less littering, you know,

12:42.465 --> 12:45.105
in a busy place instead of littering a lot.

12:45.285 --> 12:47.305
If you put ice, then people do less littering.

12:47.305 --> 12:49.185
They actually put it in that, trash pit.

12:51.245 --> 12:53.505
So following that, we expected that

12:53.505 --> 12:55.545
after putting those eye pictures,

12:56.375 --> 12:58.265
that total taking will go down

12:59.855 --> 13:02.065
because taking is kind of in a negative frame

13:02.065 --> 13:03.545
with somebody else's money I'm taking.

13:03.805 --> 13:06.185
So we expected that we putting eyes, it should be go down.

13:06.185 --> 13:07.305
So that's the first hypothesis.

13:08.165 --> 13:12.165
And second hypothesis,

13:13.355 --> 13:16.735
is as I said, people have found that this kind of effect

13:16.735 --> 13:19.935
of social cue is stronger on males compared to females.

13:20.555 --> 13:22.215
Of course it was done in the giving game.

13:22.435 --> 13:25.135
So if you put, a picture of eyes,

13:25.445 --> 13:29.755
then males give more compared to females like the increment.

13:29.795 --> 13:31.155
I that's not, that's not right.

13:31.935 --> 13:35.275
So with eyes, male gives more female give more.

13:35.785 --> 13:38.395
However, the extra amount

13:38.395 --> 13:40.595
that male is giving is more significant than females.

13:42.815 --> 13:46.235
Now we expect similar in this case we'll say, okay,

13:46.405 --> 13:47.635
males will take this

13:48.295 --> 13:51.235
and this, that amount, they're taking less.

13:52.065 --> 13:53.875
That would be more significant than females.

13:53.875 --> 13:55.355
That's what our expectation was.

13:55.995 --> 13:59.635
Actually the findings, was something that

14:00.055 --> 14:01.995
partly we expected and partly we did not.

14:02.055 --> 14:03.835
And this is what makes it exciting. I think.

14:04.635 --> 14:07.855
So as I said that there are 10 pounds that,

14:08.355 --> 14:11.095
and we ran it at University of, in uk.

14:11.115 --> 14:12.655
So that's why I'm talking about British bound.

14:13.435 --> 14:14.575
Uh, what

14:15.795 --> 14:17.775
and remember, I'm just reminding you,

14:17.985 --> 14:19.575
there are four cases, okay?

14:20.275 --> 14:22.335
Uh, with eyes, without eyes

14:22.835 --> 14:24.535
and male dictator, female dictator.

14:25.195 --> 14:28.695
So we'll have four cases with eyes, male dictator

14:28.805 --> 14:32.135
with the female dictator without eyes,

14:32.135 --> 14:34.295
male dictator without a female dictator, we are able

14:34.295 --> 14:35.975
to check, those four cases.

14:37.885 --> 14:42.335
What we found is, overall, when we don't split

14:42.405 --> 14:46.325
between male and female dictator in a baseline,

14:46.325 --> 14:47.805
when there is no picture of ice,

14:48.645 --> 14:51.765
around 7.9 pounds

14:53.565 --> 14:55.105
are taken, okay?

14:55.845 --> 15:00.025
Around 7.99 pounds are taken from, the recipient

15:00.085 --> 15:03.855
by dictators and it goes to

15:04.035 --> 15:06.935
around 7.7 pounds with eyes.

15:08.575 --> 15:13.515
So 7.9 to 7.7, like that.

15:14.215 --> 15:18.475
So there's a like very little decline with ice

15:20.065 --> 15:21.765
and when you run statistical test,

15:21.765 --> 15:22.885
it's not significant at all.

15:23.305 --> 15:27.165
So essentially I did not have any effect on overall

15:27.225 --> 15:28.565
taking of the dictators.

15:30.255 --> 15:33.195
So our overall first hypothesis turned out to be rejected.

15:33.815 --> 15:36.575
So, but this is not uncommon

15:36.575 --> 15:41.215
because there has been other, experiments as well by,

15:41.435 --> 15:43.815
both econom, economists,

15:43.835 --> 15:45.855
and ethologists and biologists.

15:46.605 --> 15:49.415
They showed that this effect of eye

15:50.835 --> 15:52.815
can be depend on many different things.

15:53.975 --> 15:56.785
This, if it's done only one, one time

15:56.845 --> 16:00.345
and a particular, job is done only one time, instead of,

16:01.525 --> 16:04.625
if the I was not,

16:05.085 --> 16:07.105
put in a very social environment enough,

16:07.445 --> 16:08.985
you might not have those kind of effect.

16:09.125 --> 16:11.505
So overall probably that might be the case we don't know

16:11.505 --> 16:15.025
yet, but we were interested in the gender difference.

16:16.965 --> 16:21.725
So now we looked at without I, how much male took

16:21.905 --> 16:23.325
and how much female took out.

16:23.595 --> 16:25.845
Females took out of this 10 pounds.

16:27.045 --> 16:31.035
Now in baseline, no eyes, males are very greedy.

16:31.625 --> 16:35.115
They took almost on an average nine pounds out of 10

16:36.205 --> 16:39.825
when there is no eye, as we know from the literature,

16:39.825 --> 16:43.305
females are nicer, they took around 6.7 pounds.

16:45.865 --> 16:48.445
So males take nine, females take 6.7.

16:48.515 --> 16:51.805
When there is no eye, this difference is significant

16:51.905 --> 16:53.045
and this matches with

16:53.075 --> 16:55.205
what we know from the literature, the females and nicer.

16:57.165 --> 17:00.535
Now we want to check what happens when we have eyes.

17:00.675 --> 17:03.415
And remember these are not the same people, these are,

17:03.415 --> 17:05.335
these are all different people facing

17:05.335 --> 17:06.935
different situation, okay?

17:06.995 --> 17:08.455
We are looking at on an average what happens.

17:09.275 --> 17:13.755
So with eyes male drop,

17:13.945 --> 17:17.515
they are taking male drop their taking behavior

17:18.105 --> 17:22.035
from nine pounds to 7.4 pounds, okay?

17:22.305 --> 17:25.745
From nine to 7.4, so around

17:26.455 --> 17:27.905
16% drop.

17:30.825 --> 17:33.285
And for female it was 6.7.

17:34.065 --> 17:38.625
And this is the shock, it goes up to 8.1, okay?

17:39.845 --> 17:41.985
So that's around 14% increase.

17:44.815 --> 17:47.755
Now, when we look at

17:49.975 --> 17:54.205
What is the difference between male

17:54.965 --> 17:56.205
baseline versus with I,

17:57.005 --> 17:58.775
that difference is statistically significant.

18:00.525 --> 18:03.815
What is the difference for female baseline versus I?

18:04.085 --> 18:05.655
That difference is statistically significant.

18:06.985 --> 18:09.525
So essentially with the social cue,

18:10.655 --> 18:13.285
males are taking less and females are taking more.

18:15.275 --> 18:17.845
Overall, what we see is there is no difference

18:17.845 --> 18:19.645
between we eyes without eyes,

18:19.865 --> 18:22.165
but the difference is happening between the genders.

18:22.665 --> 18:25.045
Of course, we ran many different tests, regression

18:25.045 --> 18:26.405
and so on, and those turned out to be,

18:26.715 --> 18:28.645
this results turned out to be correct and robust.

18:29.465 --> 18:33.005
So what we tried to find out, we tried to find out whether

18:33.745 --> 18:35.245
is is, is it happening over

18:35.245 --> 18:37.605
or not, is the particular type of people they are,

18:37.785 --> 18:39.005
behaving differently?

18:39.505 --> 18:43.085
So we divided those dictators into three types.

18:43.985 --> 18:45.565
So we know from this literature,

18:45.565 --> 18:49.005
and there is also true in our data, there are two,

18:49.085 --> 18:50.405
it's a bimodal distribution.

18:51.325 --> 18:55.285
A lot of people they just go for 50 50 split, but it's 10.

18:55.425 --> 18:57.525
You, you, you are five pounds mine, five pound done.

18:57.905 --> 19:00.685
And there are a lot of people who go for greedily nine

19:00.685 --> 19:02.085
or 10 pounds, they want to just take,

19:02.425 --> 19:04.125
and there are very few people in between they,

19:04.665 --> 19:08.325
so we divided them into egalitarian who go for 50 50,

19:09.095 --> 19:11.225
selfish, who go for 10 or nine like that,

19:11.925 --> 19:13.025
and others in between.

19:14.305 --> 19:18.575
So what interestingly turned out that

19:19.435 --> 19:22.495
for females without eyes, there are a lot of egalitarian,

19:23.235 --> 19:28.005
but with eyes, a lot of people become selfish

19:28.785 --> 19:32.665
and exactly opposite for males, lot of selfish at first,

19:32.685 --> 19:34.425
but with eyes become a lot of utilitarian.

19:34.885 --> 19:36.745
So it's not that they're distributing in between,

19:36.745 --> 19:38.105
they're just jumping from one type

19:38.105 --> 19:39.385
to another type with eyes.

19:40.485 --> 19:44.305
And from this data, it's not clearly possible for us

19:44.305 --> 19:48.585
to say exactly what make it through it, what mechanism,

19:48.685 --> 19:49.745
we can explain this,

19:50.615 --> 19:54.935
but with this change in type, we can guess,

19:55.235 --> 19:56.415
two possible mechanisms.

19:57.625 --> 19:59.965
The first one is,

20:02.135 --> 20:03.655
experimented demand effect.

20:04.075 --> 20:06.015
Uh, there's something called experimented demand effect

20:06.285 --> 20:08.855
that is, subject often think

20:08.855 --> 20:11.535
that experimenter is asking them to do something.

20:12.445 --> 20:13.825
That's why they try to do that.

20:14.245 --> 20:15.345
So in a dictator game,

20:15.575 --> 20:17.985
they feel like the experimenter is asking me to give money,

20:17.985 --> 20:18.985
that's why I'm giving money.

20:19.755 --> 20:24.155
Okay? So if they take it as,

20:24.655 --> 20:27.195
so females take it as experimental demand effect,

20:27.195 --> 20:29.315
and the experimental demand effect becomes quite

20:29.525 --> 20:30.635
acute with eyes.

20:31.385 --> 20:34.515
They feel like in this game I'm supposed to take

20:35.095 --> 20:36.995
the experiment is asking me to take and he's

20:36.995 --> 20:38.075
or she's looking at me.

20:38.295 --> 20:39.515
That's why they're taking more.

20:40.895 --> 20:42.875
So this can be one possible explanation.

20:43.585 --> 20:46.525
The other can be the females Were already,

20:46.945 --> 20:49.725
As I told you, with those distributions,

20:49.915 --> 20:51.685
females are already,

20:51.835 --> 20:55.565
sharing it 50 50 quite a lot with eyes.

20:56.305 --> 20:58.445
If they have to share even more,

20:58.835 --> 21:01.525
that means they will have lower percentage

21:01.625 --> 21:03.645
or lower share than their recipient.

21:04.725 --> 21:07.065
You know, previous, even without ice

21:07.495 --> 21:08.745
they are splitting it five pounds.

21:08.745 --> 21:11.625
Five pounds. And with ice, if the ice telling them,

21:11.625 --> 21:13.945
share even more, then it would be two for them

21:13.945 --> 21:14.985
and eight for the other person.

21:15.645 --> 21:17.385
So this kind of,

21:17.945 --> 21:19.705
situation might actually reverse their preference.

21:20.165 --> 21:22.915
So instead of becoming,

21:23.015 --> 21:25.395
in a situation in which they are in a really poor,

21:25.505 --> 21:28.795
poor place and the recipient is rich, they'll say, okay, in

21:28.795 --> 21:32.595
that case probably we'll share less instead of 50 50.

21:33.135 --> 21:34.635
So these two are two possible

21:34.635 --> 21:36.035
mechanisms through which we can tell.

21:36.305 --> 21:39.965
However, you cannot come to any

21:39.965 --> 21:41.565
of the conclusion from our data.

21:42.185 --> 21:45.525
So probably it might be useful to run more experiments

21:45.905 --> 21:49.365
and especially, collaborating with neuroscientists

21:49.785 --> 21:52.245
and run some FMRI to check which part

21:52.265 --> 21:54.685
of our brain is giving signal with

21:54.705 --> 21:56.605
and without eyes when they're making decision.

21:57.005 --> 21:58.365
Probably that will tell us something more.

21:58.585 --> 22:00.245
So yes, we don't know the mechanism,

22:00.755 --> 22:05.165
however we know that, in a negative frame

22:06.085 --> 22:08.015
when we see that there's a null result.

22:08.475 --> 22:10.975
So, you know, economists or psychologists

22:10.995 --> 22:13.135
or others, especially policy people,

22:13.165 --> 22:15.095
they don't like null result overall.

22:15.435 --> 22:17.855
If they see that this policy is not having any effect,

22:17.925 --> 22:21.055
they say, we don't not really go for this policy, this,

22:21.685 --> 22:24.455
this paper actually tell us learn that even

22:24.455 --> 22:25.655
with re result is better.

22:26.155 --> 22:27.255
It good to dig deeper

22:27.395 --> 22:29.935
and see what is happening in, in between.

22:30.725 --> 22:32.935
Overall, we have re result with the effect of eyes.

22:32.965 --> 22:36.095
However, we see that if we dig deep,

22:36.485 --> 22:38.055
then males are behaving one way,

22:38.055 --> 22:39.135
females are behaving another way

22:39.335 --> 22:41.855
and they're, like canceling each other.

22:43.175 --> 22:47.275
So it might be good that if you know a particular

22:48.115 --> 22:50.955
behavior, which is that behavior which is driven by male,

22:51.725 --> 22:55.275
using the social cue might actually help you just like,

22:55.495 --> 22:56.635
the bike keeps.

22:58.175 --> 23:01.185
However, if you know that the bike cubes are mostly females

23:01.285 --> 23:03.905
and you put those eyes,

23:04.325 --> 23:07.305
you might have unintended consequence of your policy.

23:08.865 --> 23:12.845
So this is why it's better to know exactly even within

23:13.365 --> 23:16.125
nowadays and what is going on and how

23:16.125 --> 23:16.845
to implement those.

00:00 - When we usually talk about economics, the, neoclassical economics, we, the theory, economic theory that assume that, people, human being are homo economics, who always try to think about themselves. They try to maximize their own payoff.

00:18 - they don't think of other people and so on. However, when you see ourselves, we reflect ourselves on the street and so on, we see that people actually try to help each other while walking on the street. If you see a homeless person asking for food, probably will buy some food. Now, that doesn't match with the formal theory that, new classical economics has done. So here are behavioral economics comes

00:41 - and they say that, okay, in our preferences, people actually care for others. And that caring for other is a part, something called social preference. We have some preference for other people in the society. Now, how do you measure social preference? There are many ways to do it, but one way to do it is essentially to run laboratory experiment. In laboratory experiment you bring actual human subjects

01:09 - who go through some instruction and play some economic gains. And one of those economic game is called dictator game. It's a very simple decision making process in which those human beings, they come to the laboratory, they don't see anybody, they're paired with, some other people, they don't see that person. However, they know that they, they are given by the experimenter. They are given some amount of money.

01:36 - So it can be $10 or 10 pounds depending on where you are running it. And in the dictator game, they will be asked, now this money is yours, but you have the opportunity to give some part of it to someone else, in this experiment, and the other person will not be able to say anything about how much they will get. So it's totally your decision. You can give whatever. So if you give two, then it'll be eight euros

02:03 - and two other persons. If you give five, it'll be five five. And so the other person will have to accept that and you'll go home with that actual money. Now, standard economic theory will say, I will not give anything. Take the whole 10 pounds, go home, be happy. The other person will be visible. But that doesn't actually happen. Just like, what I said, people have social preference.

02:26 - people feel bad not to give anything. And the amount that is given and how many times it's given, those are different measures of social preference. So dictator game is used in one way to try to find out people's social preference in this case, how nice people are, how much they're giving. Yeah. So, this particular study that came out in, general Behavioral and Experimental Economics, that's actually a joint work

02:51 - with Julian John from University of Reading and from Ham University. that Particular study, we use a version of the Dictator game. I just told you the most, popular version of the Ator game, which I call giving game in which I have 10 pounds as a subject. I can give something. What we use is a different version. We call it taking game. In that case, still, I am making the decision

03:20 - and that's, that's why I'm the dictator is for a dictator again, because, you know, other person cannot say anything about my decision. They have to just take the money and leave. So in this case, we change the way, the game is framed. We say that, you know, the other person, we call them recipient, other person has 10 pounds, and you will have to make a decision how much you want to take from them. The other person cannot do anything.

03:45 - Whatever you want to take, take, and then you can go home. Yeah. Now see, it remains exactly the same. If I was dictator in a giving game and I had 10 pounds, I give you two pounds, then you'll have two, I'll have eight in a taking. Again, you have 10 pounds, I take eight. Then in the end, still it'll be, you'll have two, I'll have eight. It is important to use this frame because in this case, the property right

04:08 - of the 10 pounds is not with me. It's with you. So it's put in a negative frame that I'm actually taking something from you, which might be bad thing. We are trying to find out how people behave in this kind of situation when they are making this kind of negative behavior. And we focus on two things. First, there are different ways to nudge people to stop doing negative behavior

04:31 - or to data negative behavior, or to make people do good behavior. There are many different ways. One of the easiest way to do that is, shown in psychology, behavioral economics and so on, is to give something called social cue. Social cues are different cues that is either verbal, nonverbal and so on. that guide our social interaction. So it might be that when you are, talking to someone,

04:59 - someone talks you in a very nice sooth way that makes you feel better. That's a social cue. Okay? It can guide your conversation or if you are, going through a particular area, and you'll see there is a big notice. It's written that neighborhood wa neighborhood watch is in place. So be careful. So you can see that that makes you some error in this area.

05:24 - So one of the way social queue is used in experimental, research is to put in the when people are making decision in their computer screen or in their pen and paper to put a picture of a pair of eyes in that area, in the screen or in the paper. Now this picture of eyes is as if somebody is looking at you. That makes people more c conscious. It is shown in game often, in the giving game. When you put those pair of eyes, people give more

05:55 - to the other person because it made them more conscious. It made them feel like I should behave better and so on. So similarly, it is possible to stop people doing bad things by putting those pair of eyes. people have done it. Other researchers have done that in actual field, in which, the, in one university they have put this picture of ice with, saying that or cycle things. We are looking at you and that just next to their bike stands, you know, many of the universities,

06:26 - people steal parts of bikes or the whole, bike bicycle. So they put those posters and that reduced the, that kind of bad, incidences. So yes, it might be possible to, have this kind of effect. So we were trying, we, but you know, if it's done in the campus, you, you don't see exactly what is happening, who is doing what. So you don't have much control.

06:53 - You don't know exactly through which it's happening, right? So we want to take that, we want to check that. And what we did, we ran this taking game. So, there is 10 pounds, which somebody else, you can just take that money, whatever proportion you want from zero to 10, and then you can go home. So what we do with a set of people, we ran it just like that. Another set of people ran it when they see a

07:23 - pair of eyes looking at them. And we also know from the literature that this kind social queue has effect on males more than females. When we are looking at in how, how people are doing good things, you know. So, we try to look at gender difference. So we had this kind of dictators. 50% of dictators were male, 50% dictators were female in both of the settings without,

07:53 - with eyes and without eyes. And we try to look what is happening. There actually, fits in three STRs of literature, of course in the social preference, literature, why people are being nice to each other and so on. There are literature on pure and impure altruism in which pure altruism means people do nice to other people just because that gives them happiness. Impure altruism is people do nice things to other people because that make them feel good that I'm a nice person

08:25 - and that's why they do nice things. It can be reciprocity. So somebody did good things to me, that's why I'm doing good things back to them or somebody did good things to me. So that's why I'm doing good things to somebody else. So social queue is one of those areas in, social preference. So we are contributing to that in this area. Another area that we're contributing to is

08:51 - gender effects in dictator again. So it's quite well known that in dictator game, in a giving game, not in a taking game, in a giving game, females give more than males. What, what happens in taking game? Well, people looked at whether giving and taking game are similar. People give similar kind of amount. other literature has shown that, yes, overall, in giving game, if on an average out

09:22 - of a hundred people people are giving two pounds out of weight, then in a taking game they are taking eight pounds out, and leaving two. However, this has also shown that this is not symmetric across gender. So males take more and but males give less. Females are the opposite. As a result, overall distribution three and stem, however, there's a gender difference in between.

09:50 - So we are trying to contribute on that while looking at the taking in and putting those social skills in our experimental, detail. We, do the following. we employ a pure taking in with two different treatments. You know, treatments means two different settings that are implemented in the laboratory. So pure taking error means in this case

10:20 - a subject who will be a dictator will know that the other subject who's a recipient has 10 pounds. And only thing that I can do is take anything from zero to 10 from that person. So it's a pure taking error. It's not that I can give and take and both, no, no, it's just a pure taking decision. Okay? we have done this to not to have any other confo that is sh shown in the literature. we just wanted to find clear effect of the social cue.

10:50 - That's why you used, pure taking. And now what we did, in one treatment that is without any social cue. So we had 140 subjects coming to the laboratory out of them, 70 would be, dictator and seven would be recipient. So we mashed one dictator with one recipient and they had to make only one decision. Recipient cannot make any decision. Of course, the dictator will have to make only one decision

11:25 - that, you know, the recipient has 10 power, how much you want to take. They just write how much they want to take, and that's it done. So essentially we have 70 dictators playing this game and there is no picture of anything except we put a blood black box in their decision making screen. And remember, out of those 70 dictators, we have 35 male dictators and 35 female dictators.

11:53 - In the other treatment or the other setting, it's exactly same. 140 subjects came 70 dictators, 35 male dictators, 35 female dictators. But in the screen, instead of the blood black place, it's a rectangle with picture of eyes that looking at them. So that's essentially the setting. So we had two hypothesis that, stemmed from, what we know in the literature. So the first hypothesis was about the

12:22 - reduction in negative behavior. As I said. there are, experiments run in the field that if you put this kind of eyes of like people looking at you, then people do bad things less like stealing from bike or there's also one, experiment in which they showed that people do less littering, you know, in a busy place instead of littering a lot. If you put ice, then people do less littering. They actually put it in that, trash pit.

12:51 - So following that, we expected that after putting those eye pictures, that total taking will go down because taking is kind of in a negative frame with somebody else's money I'm taking. So we expected that we putting eyes, it should be go down. So that's the first hypothesis. And second hypothesis, is as I said, people have found that this kind of effect of social cue is stronger on males compared to females.

13:20 - Of course it was done in the giving game. So if you put, a picture of eyes, then males give more compared to females like the increment. I that's not, that's not right. So with eyes, male gives more female give more. However, the extra amount that male is giving is more significant than females. Now we expect similar in this case we'll say, okay, males will take this and this, that amount, they're taking less.

13:52 - That would be more significant than females. That's what our expectation was. Actually the findings, was something that partly we expected and partly we did not. And this is what makes it exciting. I think. So as I said that there are 10 pounds that, and we ran it at University of, in uk. So that's why I'm talking about British bound. what and remember, I'm just reminding you,

14:17 - there are four cases, okay? with eyes, without eyes and male dictator, female dictator. So we'll have four cases with eyes, male dictator with the female dictator without eyes, male dictator without a female dictator, we are able to check, those four cases. What we found is, overall, when we don't split between male and female dictator in a baseline, when there is no picture of ice,

14:48 - around 7.9 pounds are taken, okay? Around 7.99 pounds are taken from, the recipient by dictators and it goes to around 7.7 pounds with eyes. So 7.9 to 7.7, like that. So there's a like very little decline with ice and when you run statistical test, it's not significant at all. So essentially I did not have any effect on overall

15:27 - taking of the dictators. So our overall first hypothesis turned out to be rejected. So, but this is not uncommon because there has been other, experiments as well by, both econom, economists, and ethologists and biologists. They showed that this effect of eye can be depend on many different things. This, if it's done only one, one time and a particular, job is done only one time, instead of,

16:01 - if the I was not, put in a very social environment enough, you might not have those kind of effect. So overall probably that might be the case we don't know yet, but we were interested in the gender difference. So now we looked at without I, how much male took and how much female took out. Females took out of this 10 pounds. Now in baseline, no eyes, males are very greedy. They took almost on an average nine pounds out of 10

16:36 - when there is no eye, as we know from the literature, females are nicer, they took around 6.7 pounds. So males take nine, females take 6.7. When there is no eye, this difference is significant and this matches with what we know from the literature, the females and nicer. Now we want to check what happens when we have eyes. And remember these are not the same people, these are, these are all different people facing different situation, okay?

17:06 - We are looking at on an average what happens. So with eyes male drop, they are taking male drop their taking behavior from nine pounds to 7.4 pounds, okay? From nine to 7.4, so around 16% drop. And for female it was 6.7. And this is the shock, it goes up to 8.1, okay? So that's around 14% increase. Now, when we look at

17:49 - What is the difference between male baseline versus with I, that difference is statistically significant. What is the difference for female baseline versus I? That difference is statistically significant. So essentially with the social cue, males are taking less and females are taking more. Overall, what we see is there is no difference between we eyes without eyes, but the difference is happening between the genders.

18:22 - Of course, we ran many different tests, regression and so on, and those turned out to be, this results turned out to be correct and robust. So what we tried to find out, we tried to find out whether is is, is it happening over or not, is the particular type of people they are, behaving differently? So we divided those dictators into three types. So we know from this literature, and there is also true in our data, there are two,

18:49 - it's a bimodal distribution. A lot of people they just go for 50 50 split, but it's 10. You, you, you are five pounds mine, five pound done. And there are a lot of people who go for greedily nine or 10 pounds, they want to just take, and there are very few people in between they, so we divided them into egalitarian who go for 50 50, selfish, who go for 10 or nine like that, and others in between. So what interestingly turned out that

19:19 - for females without eyes, there are a lot of egalitarian, but with eyes, a lot of people become selfish and exactly opposite for males, lot of selfish at first, but with eyes become a lot of utilitarian. So it's not that they're distributing in between, they're just jumping from one type to another type with eyes. And from this data, it's not clearly possible for us to say exactly what make it through it, what mechanism, we can explain this,

19:50 - but with this change in type, we can guess, two possible mechanisms. The first one is, experimented demand effect. there's something called experimented demand effect that is, subject often think that experimenter is asking them to do something. That's why they try to do that. So in a dictator game, they feel like the experimenter is asking me to give money,

20:17 - that's why I'm giving money. Okay? So if they take it as, so females take it as experimental demand effect, and the experimental demand effect becomes quite acute with eyes. They feel like in this game I'm supposed to take the experiment is asking me to take and he's or she's looking at me. That's why they're taking more. So this can be one possible explanation.

20:43 - The other can be the females Were already, As I told you, with those distributions, females are already, sharing it 50 50 quite a lot with eyes. If they have to share even more, that means they will have lower percentage or lower share than their recipient. You know, previous, even without ice they are splitting it five pounds. Five pounds. And with ice, if the ice telling them,

21:11 - share even more, then it would be two for them and eight for the other person. So this kind of, situation might actually reverse their preference. So instead of becoming, in a situation in which they are in a really poor, poor place and the recipient is rich, they'll say, okay, in that case probably we'll share less instead of 50 50. So these two are two possible mechanisms through which we can tell.

21:36 - However, you cannot come to any of the conclusion from our data. So probably it might be useful to run more experiments and especially, collaborating with neuroscientists and run some FMRI to check which part of our brain is giving signal with and without eyes when they're making decision. Probably that will tell us something more. So yes, we don't know the mechanism, however we know that, in a negative frame

22:06 - when we see that there's a null result. So, you know, economists or psychologists or others, especially policy people, they don't like null result overall. If they see that this policy is not having any effect, they say, we don't not really go for this policy, this, this paper actually tell us learn that even with re result is better. It good to dig deeper and see what is happening in, in between.

22:30 - Overall, we have re result with the effect of eyes. However, we see that if we dig deep, then males are behaving one way, females are behaving another way and they're, like canceling each other. So it might be good that if you know a particular behavior, which is that behavior which is driven by male, using the social cue might actually help you just like, the bike keeps. However, if you know that the bike cubes are mostly females

23:01 - and you put those eyes, you might have unintended consequence of your policy. So this is why it's better to know exactly even within nowadays and what is going on and how to implement those.

Faculti AI can make mistakes