1
00:00:00,371 --> 00:00:02,467
Steve Ramirez: My first
year of grad school,

2
00:00:02,491 --> 00:00:03,907
I found myself in my bedroom

3
00:00:03,931 --> 00:00:06,207
eating lots of Ben & Jerry's

4
00:00:06,231 --> 00:00:07,891
watching some trashy TV

5
00:00:07,915 --> 00:00:11,118
and maybe, maybe listening
to Taylor Swift.

6
00:00:11,142 --> 00:00:12,859
I had just gone through a breakup.

7
00:00:12,883 --> 00:00:14,306
(Laughter)

8
00:00:14,330 --> 00:00:16,502
So for the longest time, all I would do

9
00:00:16,526 --> 00:00:20,318
is recall the memory of this
person over and over again,

10
00:00:20,342 --> 00:00:22,811
wishing that I could get
rid of that gut-wrenching,

11
00:00:22,835 --> 00:00:25,344
visceral "blah" feeling.

12
00:00:25,368 --> 00:00:27,627
Now, as it turns out,
I'm a neuroscientist,

13
00:00:27,651 --> 00:00:30,029
so I knew that the memory of that person

14
00:00:30,053 --> 00:00:33,169
and the awful, emotional undertones
that color in that memory,

15
00:00:33,193 --> 00:00:35,803
are largely mediated
by separate brain systems.

16
00:00:35,827 --> 00:00:38,304
And so I thought, what if we could
go into the brain

17
00:00:38,328 --> 00:00:40,258
and edit out that nauseating feeling

18
00:00:40,282 --> 00:00:43,228
but while keeping the memory
of that person intact?

19
00:00:43,252 --> 00:00:45,919
Then I realized, maybe
that's a little bit lofty for now.

20
00:00:45,943 --> 00:00:48,432
So what if we could start
off by going into the brain

21
00:00:48,456 --> 00:00:51,065
and just finding a single
memory to begin with?

22
00:00:51,089 --> 00:00:53,579
Could we jump-start
that memory back to life,

23
00:00:53,603 --> 00:00:57,447
maybe even play with the contents
of that memory?

24
00:00:57,471 --> 00:00:59,676
All that said, there is one person
in the entire world right now

25
00:00:59,700 --> 00:01:01,843
that I really hope is not
watching this talk.

26
00:01:01,867 --> 00:01:05,672
(Laughter)

27
00:01:05,696 --> 00:01:08,961
So there is a catch. There is a catch.

28
00:01:08,985 --> 00:01:11,749
These ideas probably remind
you of "Total Recall,"

29
00:01:11,773 --> 00:01:13,723
"Eternal Sunshine of the Spotless Mind,"

30
00:01:13,747 --> 00:01:15,026
or of "Inception."

31
00:01:15,050 --> 00:01:16,812
But the movie stars that we work with

32
00:01:16,836 --> 00:01:18,545
are the celebrities of the lab.

33
00:01:18,569 --> 00:01:20,445
Xu Liu: Test mice.

34
00:01:20,469 --> 00:01:21,573
(Laughter)

35
00:01:21,597 --> 00:01:24,727
As neuroscientists, we work
in the lab with mice

36
00:01:24,751 --> 00:01:28,137
trying to understand how memory works.

37
00:01:28,161 --> 00:01:30,706
And today, we hope
to convince you that now

38
00:01:30,730 --> 00:01:33,922
we are actually able to activate
a memory in the brain

39
00:01:33,946 --> 00:01:36,092
at the speed of light.

40
00:01:36,116 --> 00:01:39,198
To do this, there's only two simple
steps to follow.

41
00:01:39,222 --> 00:01:42,708
First, you find and label
a memory in the brain,

42
00:01:42,732 --> 00:01:46,338
and then you activate it with a switch.

43
00:01:46,362 --> 00:01:47,783
As simple as that.

44
00:01:47,807 --> 00:01:49,605
(Laughter)

45
00:01:49,629 --> 00:01:51,450
SR: Are you convinced?

46
00:01:51,474 --> 00:01:55,171
So, turns out finding a memory
in the brain isn't all that easy.

47
00:01:55,195 --> 00:01:57,974
XL: Indeed. This is way more
difficult than, let's say,

48
00:01:57,998 --> 00:02:00,378
finding a needle in a haystack,

49
00:02:00,402 --> 00:02:03,117
because at least, you know,
the needle is still something

50
00:02:03,141 --> 00:02:05,467
you can physically put your fingers on.

51
00:02:05,491 --> 00:02:07,444
But memory is not.

52
00:02:07,468 --> 00:02:10,482
And also, there's way
more cells in your brain

53
00:02:10,506 --> 00:02:15,548
than the number of straws
in a typical haystack.

54
00:02:15,572 --> 00:02:18,427
So yeah, this task does
seem to be daunting.

55
00:02:18,451 --> 00:02:22,106
But luckily, we got help
from the brain itself.

56
00:02:22,130 --> 00:02:24,561
It turned out that all we need
to do is basically

57
00:02:24,585 --> 00:02:26,554
to let the brain form a memory,

58
00:02:26,578 --> 00:02:30,384
and then the brain will tell
us which cells are involved

59
00:02:30,408 --> 00:02:32,150
in that particular memory.

60
00:02:32,174 --> 00:02:34,507
SR: So what was going on in my brain

61
00:02:34,531 --> 00:02:36,601
while I was recalling the memory of an ex?

62
00:02:36,625 --> 00:02:39,003
If you were to just completely
ignore human ethics for a second

63
00:02:39,027 --> 00:02:40,671
and slice up my brain right now,

64
00:02:40,695 --> 00:02:42,886
you would see that there
was an amazing number

65
00:02:42,910 --> 00:02:45,867
of brain regions that were active
while recalling that memory.

66
00:02:45,891 --> 00:02:48,779
Now one brain region
that would be robustly active

67
00:02:48,803 --> 00:02:50,786
in particular is called the hippocampus,

68
00:02:50,810 --> 00:02:53,241
which for decades has
been implicated in processing

69
00:02:53,265 --> 00:02:55,657
the kinds of memories
that we hold near and dear,

70
00:02:55,681 --> 00:02:58,231
which also makes it
an ideal target to go into

71
00:02:58,255 --> 00:03:01,016
and to try and find and maybe
reactivate a memory.

72
00:03:01,040 --> 00:03:03,410
XL: When you zoom in into the hippocampus,

73
00:03:03,434 --> 00:03:05,758
of course you will see lots of cells,

74
00:03:05,782 --> 00:03:08,789
but we are able to find
which cells are involved

75
00:03:08,813 --> 00:03:10,265
in a particular memory,

76
00:03:10,289 --> 00:03:12,883
because whenever a cell is active,

77
00:03:12,907 --> 00:03:14,431
like when it's forming a memory,

78
00:03:14,455 --> 00:03:18,104
it will also leave a footprint
that will later allow us to know

79
00:03:18,128 --> 00:03:20,806
these cells are recently active.

80
00:03:20,830 --> 00:03:23,164
SR: So the same way
that building lights at night

81
00:03:23,188 --> 00:03:26,617
let you know that somebody's probably
working there at any given moment,

82
00:03:26,641 --> 00:03:29,202
in a very real sense, there
are biological sensors

83
00:03:29,226 --> 00:03:31,156
within a cell that are turned on

84
00:03:31,180 --> 00:03:33,291
only when that cell was just working.

85
00:03:33,315 --> 00:03:35,601
They're sort of biological
windows that light up

86
00:03:35,625 --> 00:03:37,816
to let us know that that cell
was just active.

87
00:03:37,840 --> 00:03:39,974
XL: So we clipped part of this sensor,

88
00:03:39,998 --> 00:03:43,121
and attached that to a switch
to control the cells,

89
00:03:43,145 --> 00:03:47,021
and we packed this switch
into an engineered virus

90
00:03:47,045 --> 00:03:49,609
and injected that into the brain
of the mice.

91
00:03:49,633 --> 00:03:52,243
So whenever a memory is being formed,

92
00:03:52,267 --> 00:03:54,591
any active cells for that memory

93
00:03:54,615 --> 00:03:57,333
will also have this switch installed.

94
00:03:57,357 --> 00:03:59,548
SR: So here is what the hippocampus
looks like

95
00:03:59,572 --> 00:04:01,798
after forming a fear memory, for example.

96
00:04:01,822 --> 00:04:03,938
The sea of blue that you see here

97
00:04:03,962 --> 00:04:05,890
are densely packed brain cells,

98
00:04:05,914 --> 00:04:07,435
but the green brain cells,

99
00:04:07,459 --> 00:04:10,031
the green brain cells
are the ones that are holding on

100
00:04:10,055 --> 00:04:11,374
to a specific fear memory.

101
00:04:11,398 --> 00:04:13,353
So you are looking at the crystallization

102
00:04:13,377 --> 00:04:15,740
of the fleeting formation of fear.

103
00:04:15,764 --> 00:04:19,237
You're actually looking
at the cross-section of a memory right now.

104
00:04:19,261 --> 00:04:21,690
XL: Now, for the switch
we have been talking about,

105
00:04:21,714 --> 00:04:24,615
ideally, the switch has
to act really fast.

106
00:04:24,639 --> 00:04:27,194
It shouldn't take minutes
or hours to work.

107
00:04:27,218 --> 00:04:31,458
It should act at the speed
of the brain, in milliseconds.

108
00:04:31,482 --> 00:04:32,888
SR: So what do you think, Xu?

109
00:04:32,912 --> 00:04:35,490
Could we use, let's say,
pharmacological drugs

110
00:04:35,514 --> 00:04:37,328
to activate or inactivate brain cells?

111
00:04:37,352 --> 00:04:41,391
XL: Nah. Drugs are pretty messy.
They spread everywhere.

112
00:04:41,415 --> 00:04:44,399
And also it takes them
forever to act on cells.

113
00:04:44,423 --> 00:04:48,048
So it will not allow us
to control a memory in real time.

114
00:04:48,072 --> 00:04:52,342
So Steve, how about let's zap
the brain with electricity?

115
00:04:52,366 --> 00:04:54,647
SR: So electricity is pretty fast,

116
00:04:54,671 --> 00:04:56,386
but we probably wouldn't
be able to target it

117
00:04:56,410 --> 00:04:58,839
to just the specific cells
that hold onto a memory,

118
00:04:58,863 --> 00:05:00,618
and we'd probably fry the brain.

119
00:05:00,642 --> 00:05:03,813
XL: Oh. That's true.
So it looks like, hmm,

120
00:05:03,837 --> 00:05:06,424
indeed we need to find a better way

121
00:05:06,448 --> 00:05:09,719
to impact the brain at the speed of light.

122
00:05:09,743 --> 00:05:14,805
SR: So it just so happens that light
travels at the speed of light.

123
00:05:14,829 --> 00:05:18,288
So maybe we could activate
or inactive memories

124
00:05:18,312 --> 00:05:19,785
by just using light --

125
00:05:19,809 --> 00:05:21,140
XL: That's pretty fast.

126
00:05:21,164 --> 00:05:23,025
SR: -- and because normally brain cells

127
00:05:23,049 --> 00:05:24,621
don't respond to pulses of light,

128
00:05:24,645 --> 00:05:26,579
so those that would respond
to pulses of light

129
00:05:26,603 --> 00:05:29,035
are those that contain
a light-sensitive switch.

130
00:05:29,059 --> 00:05:30,981
Now to do that, first we need
to trick brain cells

131
00:05:31,005 --> 00:05:32,443
to respond to laser beams.

132
00:05:32,467 --> 00:05:33,513
XL: Yep. You heard it right.

133
00:05:33,537 --> 00:05:35,680
We are trying to shoot
lasers into the brain.

134
00:05:35,704 --> 00:05:37,390
(Laughter)

135
00:05:37,414 --> 00:05:40,714
SR: And the technique that lets
us do that is optogenetics.

136
00:05:40,738 --> 00:05:43,996
Optogenetics gave us this
light switch that we can use

137
00:05:44,020 --> 00:05:45,504
to turn brain cells on or off,

138
00:05:45,528 --> 00:05:48,021
and the name of that switch
is channelrhodopsin,

139
00:05:48,045 --> 00:05:50,558
seen here as these green dots
attached to this brain cell.

140
00:05:50,582 --> 00:05:53,868
You can think of channelrhodopsin
as a sort of light-sensitive switch

141
00:05:53,892 --> 00:05:56,484
that can be artificially
installed in brain cells

142
00:05:56,508 --> 00:05:58,398
so that now we can use that switch

143
00:05:58,422 --> 00:06:01,422
to activate or inactivate the brain
cell simply by clicking it,

144
00:06:01,446 --> 00:06:03,970
and in this case we click
it on with pulses of light.

145
00:06:03,994 --> 00:06:07,665
XL: So we attach this light-sensitive
switch of channelrhodopsin

146
00:06:07,689 --> 00:06:09,873
to the sensor we've been talking about

147
00:06:09,897 --> 00:06:12,328
and inject this into the brain.

148
00:06:12,352 --> 00:06:15,539
So whenever a memory is being formed,

149
00:06:15,563 --> 00:06:17,766
any active cell for that particular memory

150
00:06:17,790 --> 00:06:21,250
will also have this light-sensitive
switch installed in it

151
00:06:21,274 --> 00:06:23,651
so that we can control these cells

152
00:06:23,675 --> 00:06:27,915
by the flipping of a laser
just like this one you see.

153
00:06:27,939 --> 00:06:30,779
SR: So let's put all of this
to the test now.

154
00:06:30,803 --> 00:06:32,914
What we can do is we can take our mice

155
00:06:32,938 --> 00:06:35,842
and then we can put them in a box
that looks exactly like this box here,

156
00:06:35,866 --> 00:06:38,182
and then we can give them
a very mild foot shock

157
00:06:38,206 --> 00:06:40,254
so that they form a fear
memory of this box.

158
00:06:40,278 --> 00:06:42,337
They learn that something
bad happened here.

159
00:06:42,361 --> 00:06:44,679
Now with our system,
the cells that are active

160
00:06:44,703 --> 00:06:47,469
in the hippocampus
in the making of this memory,

161
00:06:47,493 --> 00:06:50,352
only those cells will now
contain channelrhodopsin.

162
00:06:50,376 --> 00:06:53,369
XL: When you are as small as a mouse,

163
00:06:53,393 --> 00:06:56,964
it feels as if the whole
world is trying to get you.

164
00:06:56,988 --> 00:06:58,712
So your best response of defense

165
00:06:58,736 --> 00:07:01,194
is trying to be undetected.

166
00:07:01,218 --> 00:07:03,227
Whenever a mouse is in fear,

167
00:07:03,251 --> 00:07:05,109
it will show this very typical behavior

168
00:07:05,133 --> 00:07:06,878
by staying at one corner of the box,

169
00:07:06,902 --> 00:07:09,640
trying to not move any part of its body,

170
00:07:09,664 --> 00:07:12,935
and this posture is called freezing.

171
00:07:12,959 --> 00:07:17,229
So if a mouse remembers that something
bad happened in this box,

172
00:07:17,253 --> 00:07:19,852
and when we put them
back into the same box,

173
00:07:19,876 --> 00:07:21,656
it will basically show freezing

174
00:07:21,680 --> 00:07:23,941
because it doesn't want to be detected

175
00:07:23,965 --> 00:07:26,636
by any potential threats in this box.

176
00:07:26,660 --> 00:07:27,991
SR: So you can think of freezing as,

177
00:07:28,015 --> 00:07:30,206
you're walking down the street
minding your own business,

178
00:07:30,230 --> 00:07:32,278
and then out of nowhere
you almost run into

179
00:07:32,302 --> 00:07:34,123
an ex-girlfriend or ex-boyfriend,

180
00:07:34,147 --> 00:07:36,257
and now those terrifying two seconds

181
00:07:36,281 --> 00:07:38,133
where you start thinking, "What do I do?
Do I say hi?

182
00:07:38,157 --> 00:07:39,501
Do I shake their hand? Do
I turn around and run away?

183
00:07:39,525 --> 00:07:41,530
Do I sit here and pretend
like I don't exist?"

184
00:07:41,554 --> 00:07:44,714
Those kinds of fleeting thoughts
that physically incapacitate you,

185
00:07:44,738 --> 00:07:47,460
that temporarily give you
that deer-in-headlights look.

186
00:07:47,484 --> 00:07:50,751
XL: However, if you put the mouse
in a completely different

187
00:07:50,775 --> 00:07:53,912
new box, like the next one,

188
00:07:53,936 --> 00:07:56,059
it will not be afraid of this box

189
00:07:56,083 --> 00:08:00,788
because there's no reason that it
will be afraid of this new environment.

190
00:08:00,812 --> 00:08:03,968
But what if we put
the mouse in this new box

191
00:08:03,992 --> 00:08:07,579
but at the same time,
we activate the fear memory

192
00:08:07,603 --> 00:08:10,258
using lasers just like we did before?

193
00:08:10,282 --> 00:08:13,112
Are we going to bring back the fear memory

194
00:08:13,136 --> 00:08:17,109
for the first box into this
completely new environment?

195
00:08:17,133 --> 00:08:19,844
SR: All right,
and here's the million-dollar experiment.

196
00:08:19,868 --> 00:08:22,757
Now to bring back to life
the memory of that day,

197
00:08:22,781 --> 00:08:24,940
I remember that the Red Sox had just won,

198
00:08:24,964 --> 00:08:26,849
it was a green spring day,

199
00:08:26,873 --> 00:08:28,731
perfect for going up and down the river

200
00:08:28,755 --> 00:08:31,000
and then maybe going to the North End

201
00:08:31,024 --> 00:08:33,159
to get some cannolis, #justsaying.

202
00:08:33,183 --> 00:08:36,261
Now Xu and I, on the other hand,

203
00:08:36,285 --> 00:08:39,091
were in a completely windowless black room

204
00:08:39,115 --> 00:08:42,751
not making any ocular movement
that even remotely resembles an eye blink

205
00:08:42,775 --> 00:08:45,218
because our eyes were fixed
onto a computer screen.

206
00:08:45,242 --> 00:08:48,195
We were looking at this mouse
here trying to activate a memory

207
00:08:48,219 --> 00:08:50,078
for the first time using our technique.

208
00:08:50,102 --> 00:08:52,345
XL: And this is what we saw.

209
00:08:52,369 --> 00:08:54,547
When we first put the mouse into this box,

210
00:08:54,571 --> 00:08:57,660
it's exploring, sniffing
around, walking around,

211
00:08:57,684 --> 00:08:59,349
minding its own business,

212
00:08:59,373 --> 00:09:01,050
because actually by nature,

213
00:09:01,074 --> 00:09:03,029
mice are pretty curious animals.

214
00:09:03,053 --> 00:09:05,651
They want to know, what's going
on in this new box?

215
00:09:05,675 --> 00:09:07,182
It's interesting.

216
00:09:07,206 --> 00:09:10,633
But the moment we turned
on the laser, like you see now,

217
00:09:10,657 --> 00:09:13,665
all of a sudden the mouse
entered this freezing mode.

218
00:09:13,689 --> 00:09:18,096
It stayed here and tried not
to move any part of its body.

219
00:09:18,120 --> 00:09:19,724
Clearly it's freezing.

220
00:09:19,748 --> 00:09:22,307
So indeed, it looks
like we are able to bring back

221
00:09:22,331 --> 00:09:24,371
the fear memory for the first box

222
00:09:24,395 --> 00:09:27,738
in this completely new environment.

223
00:09:27,762 --> 00:09:29,850
While watching this, Steve and I

224
00:09:29,874 --> 00:09:31,983
are as shocked as the mouse itself.

225
00:09:32,007 --> 00:09:33,245
(Laughter)

226
00:09:33,269 --> 00:09:36,552
So after the experiment,
the two of us just left the room

227
00:09:36,576 --> 00:09:38,305
without saying anything.

228
00:09:38,329 --> 00:09:41,701
After a kind of long,
awkward period of time,

229
00:09:41,725 --> 00:09:43,913
Steve broke the silence.

230
00:09:43,937 --> 00:09:46,254
SR: "Did that just work?"

231
00:09:46,278 --> 00:09:49,228
XL: "Yes," I said. "Indeed it worked!"

232
00:09:49,252 --> 00:09:51,345
We're really excited about this.

233
00:09:51,369 --> 00:09:53,969
And then we published our findings

234
00:09:53,993 --> 00:09:55,665
in the journal Nature.

235
00:09:55,689 --> 00:09:58,136
Ever since the publication of our work,

236
00:09:58,160 --> 00:10:00,551
we've been receiving numerous comments

237
00:10:00,575 --> 00:10:02,676
from all over the Internet.

238
00:10:02,700 --> 00:10:06,426
Maybe we can take a look at some of those.

239
00:10:06,450 --> 00:10:08,883
["OMGGGGG FINALLY... so much more to come, virtual reality, neural manipulation, visual dream emulation...
neural coding, 'writing and re-writing of memories', mental illnesses. Ahhh the future is awesome"]

240
00:10:08,907 --> 00:10:10,883
SR: So the first thing
that you'll notice is that people

241
00:10:10,907 --> 00:10:13,786
have really strong opinions
about this kind of work.

242
00:10:13,810 --> 00:10:16,340
Now I happen to completely
agree with the optimism

243
00:10:16,364 --> 00:10:17,156
of this first quote,

244
00:10:17,180 --> 00:10:19,964
because on a scale
of zero to Morgan Freeman's voice,

245
00:10:19,988 --> 00:10:22,465
it happens to be
one of the most evocative accolades

246
00:10:22,489 --> 00:10:24,015
that I've heard come our way.

247
00:10:24,039 --> 00:10:25,833
(Laughter)

248
00:10:25,857 --> 00:10:27,784
But as you'll see, it's not
the only opinion that's out there.

249
00:10:27,808 --> 00:10:29,348
["This scares the hell out of me... What if they could do that easily
in humans in a couple of years?! OH MY GOD WE'RE DOOMED"]

250
00:10:29,372 --> 00:10:31,658
XL: Indeed, if we take
a look at the second one,

251
00:10:31,682 --> 00:10:33,765
I think we can all agree that it's, meh,

252
00:10:33,789 --> 00:10:35,748
probably not as positive.

253
00:10:35,772 --> 00:10:37,933
But this also reminds us that,

254
00:10:37,957 --> 00:10:40,119
although we are still working with mice,

255
00:10:40,143 --> 00:10:43,636
it's probably a good idea
to start thinking and discussing

256
00:10:43,660 --> 00:10:46,627
about the possible ethical ramifications

257
00:10:46,651 --> 00:10:48,575
of memory control.

258
00:10:48,599 --> 00:10:50,775
SR: Now, in the spirit of the third quote,

259
00:10:50,799 --> 00:10:52,749
we want to tell you about a recent
project that we've been

260
00:10:52,773 --> 00:10:55,345
working on in lab that we've called
Project Inception.

261
00:10:55,369 --> 00:10:58,588
["They should make a movie about this. Where they plant ideas into peoples minds,
so they can control them for their own personal gain. We'll call it: Inception."]

262
00:10:58,612 --> 00:11:02,154
So we reasoned that now
that we can reactivate a memory,

263
00:11:02,178 --> 00:11:05,106
what if we do so but then begin
to tinker with that memory?

264
00:11:05,130 --> 00:11:08,139
Could we possibly even turn
it into a false memory?

265
00:11:08,163 --> 00:11:12,238
XL: So all memory
is sophisticated and dynamic,

266
00:11:12,262 --> 00:11:15,217
but if just for simplicity,
let's imagine memory

267
00:11:15,241 --> 00:11:16,619
as a movie clip.

268
00:11:16,643 --> 00:11:19,289
So far what we've told you
is basically we can control

269
00:11:19,313 --> 00:11:21,220
this "play" button of the clip

270
00:11:21,244 --> 00:11:25,805
so that we can play this video
clip any time, anywhere.

271
00:11:25,829 --> 00:11:28,336
But is there a possibility
that we can actually get

272
00:11:28,360 --> 00:11:31,196
inside the brain and edit this movie clip

273
00:11:31,220 --> 00:11:34,092
so that we can make it
different from the original?

274
00:11:34,116 --> 00:11:36,270
Yes we can.

275
00:11:36,294 --> 00:11:38,485
Turned out that all we need
to do is basically

276
00:11:38,509 --> 00:11:42,675
reactivate a memory using
lasers just like we did before,

277
00:11:42,699 --> 00:11:46,114
but at the same time,
if we present new information

278
00:11:46,138 --> 00:11:50,088
and allow this new information
to incorporate into this old memory,

279
00:11:50,112 --> 00:11:52,526
this will change the memory.

280
00:11:52,550 --> 00:11:56,189
It's sort of like making a remix tape.

281
00:11:56,213 --> 00:11:59,047
SR: So how do we do this?

282
00:11:59,071 --> 00:12:01,004
Rather than finding a fear
memory in the brain,

283
00:12:01,028 --> 00:12:02,720
we can start by taking our animals,

284
00:12:02,744 --> 00:12:05,697
and let's say we put them in a blue
box like this blue box here

285
00:12:05,721 --> 00:12:08,344
and we find the brain cells
that represent that blue box

286
00:12:08,368 --> 00:12:10,607
and we trick them to respond
to pulses of light

287
00:12:10,631 --> 00:12:12,341
exactly like we had said before.

288
00:12:12,365 --> 00:12:14,465
Now the next day, we can take
our animals and place them

289
00:12:14,489 --> 00:12:17,164
in a red box that they've never
experienced before.

290
00:12:17,188 --> 00:12:19,427
We can shoot light
into the brain to reactivate

291
00:12:19,451 --> 00:12:21,299
the memory of the blue box.

292
00:12:21,323 --> 00:12:23,043
So what would happen here
if, while the animal

293
00:12:23,067 --> 00:12:24,972
is recalling the memory of the blue box,

294
00:12:24,996 --> 00:12:27,580
we gave it a couple of mild foot shocks?

295
00:12:27,604 --> 00:12:30,273
So here we're trying to artificially
make an association

296
00:12:30,297 --> 00:12:32,188
between the memory of the blue box

297
00:12:32,212 --> 00:12:33,691
and the foot shocks themselves.

298
00:12:33,715 --> 00:12:35,477
We're just trying to connect the two.

299
00:12:35,501 --> 00:12:37,013
So to test if we had done so,

300
00:12:37,037 --> 00:12:38,341
we can take our animals once again

301
00:12:38,365 --> 00:12:40,287
and place them back in the blue box.

302
00:12:40,311 --> 00:12:43,026
Again, we had just reactivated
the memory of the blue box

303
00:12:43,050 --> 00:12:45,431
while the animal got a couple
of mild foot shocks,

304
00:12:45,455 --> 00:12:47,562
and now the animal suddenly freezes.

305
00:12:47,586 --> 00:12:50,920
It's as though it's recalling being
mildly shocked in this environment

306
00:12:50,944 --> 00:12:53,766
even though that never actually happened.

307
00:12:53,790 --> 00:12:55,628
So it formed a false memory,

308
00:12:55,652 --> 00:12:57,700
because it's falsely
fearing an environment

309
00:12:57,724 --> 00:12:59,058
where, technically speaking,

310
00:12:59,082 --> 00:13:01,242
nothing bad actually happened to it.

311
00:13:01,266 --> 00:13:03,695
XL: So, so far we are only talking about

312
00:13:03,719 --> 00:13:06,061
this light-controlled "on" switch.

313
00:13:06,085 --> 00:13:09,349
In fact, we also have
a light-controlled "off" switch,

314
00:13:09,373 --> 00:13:11,429
and it's very easy to imagine that

315
00:13:11,453 --> 00:13:13,907
by installing this
light-controlled "off" switch,

316
00:13:13,931 --> 00:13:19,495
we can also turn off a memory,
any time, anywhere.

317
00:13:19,519 --> 00:13:21,709
So everything
we've been talking about today

318
00:13:21,733 --> 00:13:26,386
is based on this philosophically
charged principle of neuroscience

319
00:13:26,410 --> 00:13:30,504
that the mind, with its seemingly
mysterious properties,

320
00:13:30,528 --> 00:13:34,149
is actually made of physical
stuff that we can tinker with.

321
00:13:34,173 --> 00:13:35,624
SR: And for me personally,

322
00:13:35,648 --> 00:13:37,410
I see a world where we can reactivate

323
00:13:37,434 --> 00:13:39,321
any kind of memory that we'd like.

324
00:13:39,345 --> 00:13:42,619
I also see a world where we can
erase unwanted memories.

325
00:13:42,643 --> 00:13:44,786
Now, I even see a world
where editing memories

326
00:13:44,810 --> 00:13:46,094
is something of a reality,

327
00:13:46,118 --> 00:13:47,799
because we're living in a time
where it's possible

328
00:13:47,823 --> 00:13:50,252
to pluck questions from the tree
of science fiction

329
00:13:50,276 --> 00:13:52,444
and to ground them
in experimental reality.

330
00:13:52,468 --> 00:13:54,327
XL: Nowadays, people in the lab

331
00:13:54,351 --> 00:13:56,713
and people in other
groups all over the world

332
00:13:56,737 --> 00:14:00,530
are using similar methods
to activate or edit memories,

333
00:14:00,554 --> 00:14:04,371
whether that's old or new,
positive or negative,

334
00:14:04,395 --> 00:14:07,043
all sorts of memories so
that we can understand

335
00:14:07,067 --> 00:14:08,907
how memory works.

336
00:14:08,931 --> 00:14:10,691
SR: For example, one group in our lab

337
00:14:10,715 --> 00:14:13,329
was able to find the brain cells
that make up a fear memory

338
00:14:13,353 --> 00:14:16,104
and converted them into a pleasurable
memory, just like that.

339
00:14:16,128 --> 00:14:19,271
That's exactly what I mean about editing
these kinds of processes.

340
00:14:19,295 --> 00:14:21,534
Now one dude in lab
was even able to reactivate

341
00:14:21,558 --> 00:14:23,479
memories of female mice in male mice,

342
00:14:23,503 --> 00:14:26,474
which rumor has it
is a pleasurable experience.

343
00:14:26,498 --> 00:14:30,591
XL: Indeed, we are living
in a very exciting moment

344
00:14:30,615 --> 00:14:34,396
where science doesn't have
any arbitrary speed limits

345
00:14:34,420 --> 00:14:37,583
but is only bound by our own imagination.

346
00:14:37,607 --> 00:14:39,750
SR: And finally, what do
we make of all this?

347
00:14:39,774 --> 00:14:41,701
How do we push this technology forward?

348
00:14:41,725 --> 00:14:43,916
These are the questions
that should not remain

349
00:14:43,940 --> 00:14:45,213
just inside the lab,

350
00:14:45,237 --> 00:14:47,809
and so one goal of today's talk
was to bring everybody

351
00:14:47,833 --> 00:14:50,214
up to speed with the kind
of stuff that's possible

352
00:14:50,238 --> 00:14:51,488
in modern neuroscience,

353
00:14:51,512 --> 00:14:52,998
but now, just as importantly,

354
00:14:53,022 --> 00:14:56,330
to actively engage everybody
in this conversation.

355
00:14:56,354 --> 00:14:59,116
So let's think together as a team
about what this all means

356
00:14:59,140 --> 00:15:01,133
and where we can and should go from here,

357
00:15:01,157 --> 00:15:03,231
because Xu and I think we all have

358
00:15:03,255 --> 00:15:05,767
some really big decisions ahead of us.

359
00:15:05,791 --> 00:15:06,892
Thank you. XL: Thank you.

360
00:15:06,916 --> 00:15:08,550
(Applause)