1
00:00:00,474 --> 00:00:03,283
Other people. Everyone is interested in other people.

2
00:00:03,283 --> 00:00:05,406
Everyone has relationships with other people,

3
00:00:05,406 --> 00:00:06,998
and they're interested in these relationships

4
00:00:06,998 --> 00:00:08,853
for a variety of reasons.

5
00:00:08,853 --> 00:00:10,865
Good relationships, bad relationships,

6
00:00:10,865 --> 00:00:14,011
annoying relationships, agnostic relationships,

7
00:00:14,011 --> 00:00:17,435
and what I'm going to do is focus on the central piece

8
00:00:17,435 --> 00:00:20,738
of an interaction that goes on in a relationship.

9
00:00:20,738 --> 00:00:23,074
So I'm going to take as inspiration the fact that we're all

10
00:00:23,074 --> 00:00:25,499
interested in interacting with other people,

11
00:00:25,499 --> 00:00:29,331
I'm going to completely strip it of all its complicating features,

12
00:00:29,331 --> 00:00:33,225
and I'm going to turn that object, that simplified object,

13
00:00:33,225 --> 00:00:37,375
into a scientific probe, and provide the early stages,

14
00:00:37,375 --> 00:00:39,824
embryonic stages of new insights into what happens

15
00:00:39,824 --> 00:00:43,474
in two brains while they simultaneously interact.

16
00:00:43,474 --> 00:00:45,767
But before I do that, let me tell you a couple of things

17
00:00:45,767 --> 00:00:47,466
that made this possible.

18
00:00:47,466 --> 00:00:50,247
The first is we can now eavesdrop safely

19
00:00:50,247 --> 00:00:52,958
on healthy brain activity.

20
00:00:52,958 --> 00:00:55,535
Without needles and radioactivity,

21
00:00:55,535 --> 00:00:58,398
without any kind of clinical reason, we can go down the street

22
00:00:58,398 --> 00:01:01,525
and record from your friends' and neighbors' brains

23
00:01:01,525 --> 00:01:04,063
while they do a variety of cognitive tasks, and we use

24
00:01:04,063 --> 00:01:07,797
a method called functional magnetic resonance imaging.

25
00:01:07,797 --> 00:01:10,122
You've probably all read about it or heard about in some

26
00:01:10,122 --> 00:01:14,500
incarnation. Let me give you a two-sentence version of it.

27
00:01:14,500 --> 00:01:17,984
So we've all heard of MRIs. MRIs use magnetic fields

28
00:01:17,984 --> 00:01:20,013
and radio waves and they take snapshots of your brain

29
00:01:20,013 --> 00:01:22,374
or your knee or your stomach,

30
00:01:22,374 --> 00:01:24,419
grayscale images that are frozen in time.

31
00:01:24,419 --> 00:01:26,740
In the 1990s, it was discovered you could use

32
00:01:26,740 --> 00:01:29,399
the same machines in a different mode,

33
00:01:29,399 --> 00:01:31,745
and in that mode, you could make microscopic blood flow

34
00:01:31,745 --> 00:01:35,045
movies from hundreds of thousands of sites independently in the brain.

35
00:01:35,045 --> 00:01:38,245
Okay, so what? In fact, the so what is, in the brain,

36
00:01:38,245 --> 00:01:42,077
changes in neural activity, the things that make your brain work,

37
00:01:42,077 --> 00:01:44,087
the things that make your software work in your brain,

38
00:01:44,087 --> 00:01:46,576
are tightly correlated with changes in blood flow.

39
00:01:46,576 --> 00:01:48,549
You make a blood flow movie, you have an independent

40
00:01:48,549 --> 00:01:50,888
proxy of brain activity.

41
00:01:50,888 --> 00:01:53,922
This has literally revolutionized cognitive science.

42
00:01:53,922 --> 00:01:55,913
Take any cognitive domain you want, memory,

43
00:01:55,913 --> 00:01:58,054
motor planning, thinking about your mother-in-law,

44
00:01:58,054 --> 00:02:01,769
getting angry at people, emotional response, it goes on and on,

45
00:02:01,769 --> 00:02:04,858
put people into functional MRI devices, and

46
00:02:04,858 --> 00:02:08,241
image how these kinds of variables map onto brain activity.

47
00:02:08,241 --> 00:02:11,090
It's in its early stages, and it's crude by some measures,

48
00:02:11,090 --> 00:02:13,658
but in fact, 20 years ago, we were at nothing.

49
00:02:13,658 --> 00:02:16,017
You couldn't do people like this. You couldn't do healthy people.

50
00:02:16,017 --> 00:02:18,505
That's caused a literal revolution, and it's opened us up

51
00:02:18,505 --> 00:02:21,323
to a new experimental preparation. Neurobiologists,

52
00:02:21,323 --> 00:02:25,083
as you well know, have lots of experimental preps,

53
00:02:25,083 --> 00:02:28,224
worms and rodents and fruit flies and things like this.

54
00:02:28,224 --> 00:02:31,621
And now, we have a new experimental prep: human beings.

55
00:02:31,621 --> 00:02:35,382
We can now use human beings to study and model

56
00:02:35,382 --> 00:02:38,332
the software in human beings, and we have a few

57
00:02:38,332 --> 00:02:41,167
burgeoning biological measures.

58
00:02:41,167 --> 00:02:45,054
Okay, let me give you one example of the kinds of experiments that people do,

59
00:02:45,054 --> 00:02:47,731
and it's in the area of what you'd call valuation.

60
00:02:47,731 --> 00:02:49,866
Valuation is just what you think it is, you know?

61
00:02:49,866 --> 00:02:52,670
If you went and you were valuing two companies against

62
00:02:52,670 --> 00:02:55,406
one another, you'd want to know which was more valuable.

63
00:02:55,406 --> 00:02:59,285
Cultures discovered the key feature of valuation thousands of years ago.

64
00:02:59,285 --> 00:03:01,975
If you want to compare oranges to windshields, what do you do?

65
00:03:01,975 --> 00:03:04,331
Well, you can't compare oranges to windshields.

66
00:03:04,331 --> 00:03:06,586
They're immiscible. They don't mix with one another.

67
00:03:06,586 --> 00:03:08,937
So instead, you convert them to a common currency scale,

68
00:03:08,937 --> 00:03:11,643
put them on that scale, and value them accordingly.

69
00:03:11,643 --> 00:03:15,079
Well, your brain has to do something just like that as well,

70
00:03:15,079 --> 00:03:17,567
and we're now beginning to understand and identify

71
00:03:17,567 --> 00:03:19,704
brain systems involved in valuation,

72
00:03:19,704 --> 00:03:22,336
and one of them includes a neurotransmitter system

73
00:03:22,336 --> 00:03:24,968
whose cells are located in your brainstem

74
00:03:24,968 --> 00:03:28,143
and deliver the chemical dopamine to the rest of your brain.

75
00:03:28,143 --> 00:03:30,585
I won't go through the details of it, but that's an important

76
00:03:30,585 --> 00:03:32,742
discovery, and we know a good bit about that now,

77
00:03:32,742 --> 00:03:34,972
and it's just a small piece of it, but it's important because

78
00:03:34,972 --> 00:03:38,247
those are the neurons that you would lose if you had Parkinson's disease,

79
00:03:38,247 --> 00:03:40,263
and they're also the neurons that are hijacked by literally

80
00:03:40,263 --> 00:03:42,495
every drug of abuse, and that makes sense.

81
00:03:42,495 --> 00:03:44,831
Drugs of abuse would come in, and they would change

82
00:03:44,831 --> 00:03:46,620
the way you value the world. They change the way

83
00:03:46,620 --> 00:03:49,819
you value the symbols associated with your drug of choice,

84
00:03:49,819 --> 00:03:52,333
and they make you value that over everything else.

85
00:03:52,333 --> 00:03:55,354
Here's the key feature though. These neurons are also

86
00:03:55,354 --> 00:03:58,855
involved in the way you can assign value to literally abstract ideas,

87
00:03:58,855 --> 00:04:00,896
and I put some symbols up here that we assign value to

88
00:04:00,896 --> 00:04:03,616
for various reasons.

89
00:04:03,616 --> 00:04:06,305
We have a behavioral superpower in our brain,

90
00:04:06,305 --> 00:04:08,058
and it at least in part involves dopamine.

91
00:04:08,058 --> 00:04:12,247
We can deny every instinct we have for survival for an idea,

92
00:04:12,247 --> 00:04:16,252
for a mere idea. No other species can do that.

93
00:04:16,252 --> 00:04:19,858
In 1997, the cult Heaven's Gate committed mass suicide

94
00:04:19,858 --> 00:04:22,073
predicated on the idea that there was a spaceship

95
00:04:22,073 --> 00:04:25,858
hiding in the tail of the then-visible comet Hale-Bopp

96
00:04:25,858 --> 00:04:30,130
waiting to take them to the next level. It was an incredibly tragic event.

97
00:04:30,130 --> 00:04:33,615
More than two thirds of them had college degrees.

98
00:04:33,615 --> 00:04:37,338
But the point here is they were able to deny their instincts for survival

99
00:04:37,338 --> 00:04:40,204
using exactly the same systems that were put there

100
00:04:40,204 --> 00:04:44,246
to make them survive. That's a lot of control, okay?

101
00:04:44,246 --> 00:04:46,335
One thing that I've left out of this narrative

102
00:04:46,335 --> 00:04:48,569
is the obvious thing, which is the focus of the rest of my

103
00:04:48,569 --> 00:04:50,728
little talk, and that is other people.

104
00:04:50,728 --> 00:04:53,724
These same valuation systems are redeployed

105
00:04:53,724 --> 00:04:56,216
when we're valuing interactions with other people.

106
00:04:56,216 --> 00:04:59,487
So this same dopamine system that gets addicted to drugs,

107
00:04:59,487 --> 00:05:02,011
that makes you freeze when you get Parkinson's disease,

108
00:05:02,011 --> 00:05:05,088
that contributes to various forms of psychosis,

109
00:05:05,088 --> 00:05:09,008
is also redeployed to value interactions with other people

110
00:05:09,008 --> 00:05:11,904
and to assign value to gestures that you do

111
00:05:11,904 --> 00:05:14,478
when you're interacting with somebody else.

112
00:05:14,478 --> 00:05:17,055
Let me give you an example of this.

113
00:05:17,055 --> 00:05:20,022
You bring to the table such enormous processing power

114
00:05:20,022 --> 00:05:22,646
in this domain that you hardly even notice it.

115
00:05:22,646 --> 00:05:24,113
Let me just give you a few examples. So here's a baby.

116
00:05:24,113 --> 00:05:27,843
She's three months old. She still poops in her diapers and she can't do calculus.

117
00:05:27,843 --> 00:05:31,196
She's related to me. Somebody will be very glad that she's up here on the screen.

118
00:05:31,196 --> 00:05:33,572
You can cover up one of her eyes, and you can still read

119
00:05:33,572 --> 00:05:36,327
something in the other eye, and I see sort of curiosity

120
00:05:36,327 --> 00:05:39,924
in one eye, I see maybe a little bit of surprise in the other.

121
00:05:39,924 --> 00:05:43,103
Here's a couple. They're sharing a moment together,

122
00:05:43,103 --> 00:05:44,421
and we've even done an experiment where you can cut out

123
00:05:44,421 --> 00:05:47,428
different pieces of this frame and you can still see

124
00:05:47,428 --> 00:05:49,932
that they're sharing it. They're sharing it sort of in parallel.

125
00:05:49,932 --> 00:05:52,395
Now, the elements of the scene also communicate this

126
00:05:52,395 --> 00:05:54,630
to us, but you can read it straight off their faces,

127
00:05:54,630 --> 00:05:58,133
and if you compare their faces to normal faces, it would be a very subtle cue.

128
00:05:58,133 --> 00:06:01,480
Here's another couple. He's projecting out at us,

129
00:06:01,480 --> 00:06:04,368
and she's clearly projecting, you know,

130
00:06:04,368 --> 00:06:06,631
love and admiration at him.

131
00:06:06,631 --> 00:06:10,266
Here's another couple. (Laughter)

132
00:06:10,266 --> 00:06:15,416
And I'm thinking I'm not seeing love and admiration on the left. (Laughter)

133
00:06:15,416 --> 00:06:17,976
In fact, I know this is his sister, and you can just see

134
00:06:17,976 --> 00:06:20,489
him saying, "Okay, we're doing this for the camera,

135
00:06:20,489 --> 00:06:26,191
and then afterwards you steal my candy and you punch me in the face." (Laughter)

136
00:06:26,191 --> 00:06:28,297
He'll kill me for showing that.

137
00:06:28,297 --> 00:06:31,094
All right, so what does this mean?

138
00:06:31,094 --> 00:06:34,444
It means we bring an enormous amount of processing power to the problem.

139
00:06:34,444 --> 00:06:38,092
It engages deep systems in our brain, in dopaminergic

140
00:06:38,092 --> 00:06:40,910
systems that are there to make you chase sex, food and salt.

141
00:06:40,910 --> 00:06:43,804
They keep you alive. It gives them the pie, it gives

142
00:06:43,804 --> 00:06:46,708
that kind of a behavioral punch which we've called a superpower.

143
00:06:46,708 --> 00:06:50,362
So how can we take that and arrange a kind of staged

144
00:06:50,362 --> 00:06:53,060
social interaction and turn that into a scientific probe?

145
00:06:53,060 --> 00:06:55,751
And the short answer is games.

146
00:06:55,751 --> 00:07:00,155
Economic games. So what we do is we go into two areas.

147
00:07:00,155 --> 00:07:03,491
One area is called experimental economics. The other area is called behavioral economics.

148
00:07:03,491 --> 00:07:07,569
And we steal their games. And we contrive them to our own purposes.

149
00:07:07,569 --> 00:07:10,536
So this shows you one particular game called an ultimatum game.

150
00:07:10,536 --> 00:07:12,381
Red person is given a hundred dollars and can offer

151
00:07:12,381 --> 00:07:16,104
a split to blue. Let's say red wants to keep 70,

152
00:07:16,104 --> 00:07:20,190
and offers blue 30. So he offers a 70-30 split with blue.

153
00:07:20,190 --> 00:07:23,041
Control passes to blue, and blue says, "I accept it,"

154
00:07:23,041 --> 00:07:24,997
in which case he'd get the money, or blue says,

155
00:07:24,997 --> 00:07:29,304
"I reject it," in which case no one gets anything. Okay?

156
00:07:29,304 --> 00:07:32,696
So a rational choice economist would say, well,

157
00:07:32,696 --> 00:07:34,752
you should take all non-zero offers.

158
00:07:34,752 --> 00:07:38,514
What do people do? People are indifferent at an 80-20 split.

159
00:07:38,514 --> 00:07:42,038
At 80-20, it's a coin flip whether you accept that or not.

160
00:07:42,038 --> 00:07:44,929
Why is that? You know, because you're pissed off.

161
00:07:44,929 --> 00:07:48,538
You're mad. That's an unfair offer, and you know what an unfair offer is.

162
00:07:48,538 --> 00:07:51,242
This is the kind of game done by my lab and many around the world.

163
00:07:51,242 --> 00:07:53,786
That just gives you an example of the kind of thing that

164
00:07:53,786 --> 00:07:57,524
these games probe. The interesting thing is, these games

165
00:07:57,524 --> 00:08:01,231
require that you have a lot of cognitive apparatus on line.

166
00:08:01,231 --> 00:08:04,159
You have to be able to come to the table with a proper model of another person.

167
00:08:04,159 --> 00:08:07,372
You have to be able to remember what you've done.

168
00:08:07,372 --> 00:08:08,792
You have to stand up in the moment to do that.

169
00:08:08,792 --> 00:08:12,142
Then you have to update your model based on the signals coming back,

170
00:08:12,142 --> 00:08:15,114
and you have to do something that is interesting,

171
00:08:15,114 --> 00:08:17,711
which is you have to do a kind of depth of thought assay.

172
00:08:17,711 --> 00:08:21,044
That is, you have to decide what that other person expects of you.

173
00:08:21,044 --> 00:08:23,998
You have to send signals to manage your image in their mind.

174
00:08:23,998 --> 00:08:26,851
Like a job interview. You sit across the desk from somebody,

175
00:08:26,851 --> 00:08:28,220
they have some prior image of you,

176
00:08:28,220 --> 00:08:30,971
you send signals across the desk to move their image

177
00:08:30,971 --> 00:08:34,891
of you from one place to a place where you want it to be.

178
00:08:34,891 --> 00:08:38,276
We're so good at this we don't really even notice it.

179
00:08:38,276 --> 00:08:42,043
These kinds of probes exploit it. Okay?

180
00:08:42,043 --> 00:08:43,850
In doing this, what we've discovered is that humans

181
00:08:43,850 --> 00:08:46,181
are literal canaries in social exchanges.

182
00:08:46,181 --> 00:08:49,578
Canaries used to be used as kind of biosensors in mines.

183
00:08:49,578 --> 00:08:53,138
When methane built up, or carbon dioxide built up,

184
00:08:53,138 --> 00:08:57,324
or oxygen was diminished, the birds would swoon

185
00:08:57,324 --> 00:08:59,650
before people would -- so it acted as an early warning system:

186
00:08:59,650 --> 00:09:02,630
Hey, get out of the mine. Things aren't going so well.

187
00:09:02,630 --> 00:09:05,584
People come to the table, and even these very blunt,

188
00:09:05,584 --> 00:09:08,574
staged social interactions, and they, and there's just

189
00:09:08,574 --> 00:09:11,590
numbers going back and forth between the people,

190
00:09:11,590 --> 00:09:13,789
and they bring enormous sensitivities to it.

191
00:09:13,789 --> 00:09:16,478
So we realized we could exploit this, and in fact,

192
00:09:16,478 --> 00:09:19,034
as we've done that, and we've done this now in

193
00:09:19,034 --> 00:09:21,728
many thousands of people, I think on the order of

194
00:09:21,728 --> 00:09:23,893
five or six thousand. We actually, to make this

195
00:09:23,893 --> 00:09:26,117
a biological probe, need bigger numbers than that,

196
00:09:26,117 --> 00:09:29,791
remarkably so. But anyway,

197
00:09:29,791 --> 00:09:31,795
patterns have emerged, and we've been able to take

198
00:09:31,795 --> 00:09:35,631
those patterns, convert them into mathematical models,

199
00:09:35,631 --> 00:09:38,320
and use those mathematical models to gain new insights

200
00:09:38,320 --> 00:09:40,451
into these exchanges. Okay, so what?

201
00:09:40,451 --> 00:09:43,764
Well, the so what is, that's a really nice behavioral measure,

202
00:09:43,764 --> 00:09:47,083
the economic games bring to us notions of optimal play.

203
00:09:47,083 --> 00:09:49,567
We can compute that during the game.

204
00:09:49,567 --> 00:09:52,520
And we can use that to sort of carve up the behavior.

205
00:09:52,520 --> 00:09:56,850
Here's the cool thing. Six or seven years ago,

206
00:09:56,850 --> 00:09:59,400
we developed a team. It was at the time in Houston, Texas.

207
00:09:59,400 --> 00:10:02,794
It's now in Virginia and London. And we built software

208
00:10:02,794 --> 00:10:06,001
that'll link functional magnetic resonance imaging devices

209
00:10:06,001 --> 00:10:10,036
up over the Internet. I guess we've done up to six machines

210
00:10:10,036 --> 00:10:12,017
at a time, but let's just focus on two.

211
00:10:12,017 --> 00:10:15,075
So it synchronizes machines anywhere in the world.

212
00:10:15,075 --> 00:10:18,244
We synchronize the machines, set them into these

213
00:10:18,244 --> 00:10:20,227
staged social interactions, and we eavesdrop on both

214
00:10:20,227 --> 00:10:21,893
of the interacting brains. So for the first time,

215
00:10:21,893 --> 00:10:25,500
we don't have to look at just averages over single individuals,

216
00:10:25,500 --> 00:10:28,397
or have individuals playing computers, or try to make

217
00:10:28,397 --> 00:10:31,160
inferences that way. We can study individual dyads.

218
00:10:31,160 --> 00:10:33,945
We can study the way that one person interacts with another person,

219
00:10:33,945 --> 00:10:36,509
turn the numbers up, and start to gain new insights

220
00:10:36,509 --> 00:10:39,024
into the boundaries of normal cognition,

221
00:10:39,024 --> 00:10:41,756
but more importantly, we can put people with

222
00:10:41,756 --> 00:10:45,093
classically defined mental illnesses, or brain damage,

223
00:10:45,093 --> 00:10:48,644
into these social interactions, and use these as probes of that.

224
00:10:48,644 --> 00:10:50,994
So we've started this effort. We've made a few hits,

225
00:10:50,994 --> 00:10:53,443
a few, I think, embryonic discoveries.

226
00:10:53,443 --> 00:10:56,255
We think there's a future to this. But it's our way

227
00:10:56,255 --> 00:10:58,815
of going in and redefining, with a new lexicon,

228
00:10:58,815 --> 00:11:02,837
a mathematical one actually, as opposed to the standard

229
00:11:02,837 --> 00:11:05,415
ways that we think about mental illness,

230
00:11:05,415 --> 00:11:07,482
characterizing these diseases, by using the people

231
00:11:07,482 --> 00:11:10,489
as birds in the exchanges. That is, we exploit the fact

232
00:11:10,489 --> 00:11:14,733
that the healthy partner, playing somebody with major depression,

233
00:11:14,733 --> 00:11:17,643
or playing somebody with autism spectrum disorder,

234
00:11:17,643 --> 00:11:21,493
or playing somebody with attention deficit hyperactivity disorder,

235
00:11:21,493 --> 00:11:24,712
we use that as a kind of biosensor, and then we use

236
00:11:24,712 --> 00:11:27,356
computer programs to model that person, and it gives us

237
00:11:27,356 --> 00:11:29,826
a kind of assay of this.

238
00:11:29,826 --> 00:11:31,957
Early days, and we're just beginning, we're setting up sites

239
00:11:31,957 --> 00:11:35,367
around the world. Here are a few of our collaborating sites.

240
00:11:35,367 --> 00:11:37,676
The hub, ironically enough,

241
00:11:37,676 --> 00:11:40,565
is centered in little Roanoke, Virginia.

242
00:11:40,565 --> 00:11:42,834
There's another hub in London, now, and the rest

243
00:11:42,834 --> 00:11:46,843
are getting set up. We hope to give the data away

244
00:11:46,843 --> 00:11:50,516
at some stage. That's a complicated issue

245
00:11:50,516 --> 00:11:53,510
about making it available to the rest of the world.

246
00:11:53,510 --> 00:11:55,357
But we're also studying just a small part

247
00:11:55,357 --> 00:11:57,624
of what makes us interesting as human beings, and so

248
00:11:57,624 --> 00:11:59,665
I would invite other people who are interested in this

249
00:11:59,665 --> 00:12:02,234
to ask us for the software, or even for guidance

250
00:12:02,234 --> 00:12:04,453
on how to move forward with that.

251
00:12:04,453 --> 00:12:06,794
Let me leave you with one thought in closing.

252
00:12:06,794 --> 00:12:08,736
The interesting thing about studying cognition

253
00:12:08,736 --> 00:12:12,468
has been that we've been limited, in a way.

254
00:12:12,468 --> 00:12:15,411
We just haven't had the tools to look at interacting brains

255
00:12:15,411 --> 00:12:16,611
simultaneously.

256
00:12:16,611 --> 00:12:19,081
The fact is, though, that even when we're alone,

257
00:12:19,081 --> 00:12:23,192
we're a profoundly social creature. We're not a solitary mind

258
00:12:23,192 --> 00:12:27,565
built out of properties that kept it alive in the world

259
00:12:27,565 --> 00:12:31,513
independent of other people. In fact, our minds

260
00:12:31,513 --> 00:12:34,383
depend on other people. They depend on other people,

261
00:12:34,383 --> 00:12:35,924
and they're expressed in other people,

262
00:12:35,924 --> 00:12:39,576
so the notion of who you are, you often don't know

263
00:12:39,576 --> 00:12:42,264
who you are until you see yourself in interaction with people

264
00:12:42,264 --> 00:12:44,670
that are close to you, people that are enemies of you,

265
00:12:44,670 --> 00:12:47,215
people that are agnostic to you.

266
00:12:47,215 --> 00:12:50,991
So this is the first sort of step into using that insight

267
00:12:50,991 --> 00:12:54,286
into what makes us human beings, turning it into a tool,

268
00:12:54,286 --> 00:12:56,264
and trying to gain new insights into mental illness.

269
00:12:56,264 --> 00:12:59,385
Thanks for having me. (Applause)

270
00:12:59,385 --> 00:13:02,474
(Applause)