1
00:00:00,090 --> 00:00:02,490
The following content is
provided under a Creative

2
00:00:02,490 --> 00:00:04,030
Commons license.

3
00:00:04,030 --> 00:00:06,330
Your support will help
MIT OpenCourseWare

4
00:00:06,330 --> 00:00:10,690
continue to offer high quality
educational resources for free.

5
00:00:10,690 --> 00:00:13,320
To make a donation or
view additional materials

6
00:00:13,320 --> 00:00:17,280
from hundreds of MIT courses,
visit MIT OpenCourseWare

7
00:00:17,280 --> 00:00:19,932
at ocw.mit.edu.

8
00:00:19,932 --> 00:00:21,390
ABBY NOYCE: One of
the things we've

9
00:00:21,390 --> 00:00:24,620
been talking about when we're
talking about how vision

10
00:00:24,620 --> 00:00:28,440
and the visual system works is
that the visual system is not

11
00:00:28,440 --> 00:00:32,310
very interested in places where
the world is all the same.

12
00:00:32,310 --> 00:00:37,467
Clear blue sky, clear smooth
surfaces, your visual system

13
00:00:37,467 --> 00:00:40,050
looks at that and says I don't
need a whole lot of information

14
00:00:40,050 --> 00:00:42,810
to give you a
perception about this.

15
00:00:42,810 --> 00:00:44,550
But it really pays
a lot of attention

16
00:00:44,550 --> 00:00:47,430
to places where things
change, where there's

17
00:00:47,430 --> 00:00:50,617
motion, where there's
edges, where color changes,

18
00:00:50,617 --> 00:00:51,450
where there's lines.

19
00:00:51,450 --> 00:00:53,200
And we're going to
talk today a little bit

20
00:00:53,200 --> 00:00:59,040
about the neural processes
that manage to grab

21
00:00:59,040 --> 00:01:01,440
onto these sorts of features.

22
00:01:01,440 --> 00:01:06,600
So you guys talked yesterday
about ganglion cells,

23
00:01:06,600 --> 00:01:08,460
right, in the retina?

24
00:01:08,460 --> 00:01:09,150
Yes, no.

25
00:01:09,150 --> 00:01:11,052
AUDIENCE: Yes.

26
00:01:11,052 --> 00:01:12,510
AUDIENCE: Are those
rods and cones?

27
00:01:12,510 --> 00:01:12,990
AUDIENCE: She mentioned it.

28
00:01:12,990 --> 00:01:13,920
ABBY NOYCE: OK.

29
00:01:13,920 --> 00:01:15,360
So what happens is--

30
00:01:15,360 --> 00:01:16,437
we'll back up then.

31
00:01:16,437 --> 00:01:19,020
So there's a bunch of different
layers of cells in the retina,

32
00:01:19,020 --> 00:01:19,520
right?

33
00:01:19,520 --> 00:01:21,750
Here's the back of an eyeball.

34
00:01:21,750 --> 00:01:25,500
Here's all of our
photoreceptors, rods and cones.

35
00:01:28,470 --> 00:01:29,910
And then in here,
there's a bunch

36
00:01:29,910 --> 00:01:34,990
of different layers of neurons.

37
00:01:34,990 --> 00:01:36,480
Neurons, and
there's some that do

38
00:01:36,480 --> 00:01:38,490
horizontal processing and more.

39
00:01:41,580 --> 00:01:42,910
Lots of neurons.

40
00:01:42,910 --> 00:01:44,460
So the retina is
this sheet, right?

41
00:01:44,460 --> 00:01:46,793
It's got photoreceptors
furthest to the back of the eye,

42
00:01:46,793 --> 00:01:49,170
and then it does this
preliminary processing.

43
00:01:49,170 --> 00:01:54,180
And one of the things
that is important

44
00:01:54,180 --> 00:01:55,890
here is that this
set of neurons that

45
00:01:55,890 --> 00:01:58,650
is closest to the
inside of the eye--

46
00:01:58,650 --> 00:02:00,985
if you look straight in
through somebody's pupil, this

47
00:02:00,985 --> 00:02:03,360
would be the surface of the
retina that's closest to you.

48
00:02:03,360 --> 00:02:05,400
Remember, the photoreceptors
are on the back,

49
00:02:05,400 --> 00:02:08,220
and the layers of
processing come forward--

50
00:02:08,220 --> 00:02:10,591
are called ganglion cells.

51
00:02:10,591 --> 00:02:12,090
And ganglion cells
are the ones that

52
00:02:12,090 --> 00:02:17,100
actually bundle all their axons
together and line up and form

53
00:02:17,100 --> 00:02:18,240
the optic nerve, right?

54
00:02:20,790 --> 00:02:25,320
So ganglion cells
have a particular kind

55
00:02:25,320 --> 00:02:27,550
of receptive field.

56
00:02:27,550 --> 00:02:31,590
A cell's receptive field is
the part of the field of vision

57
00:02:31,590 --> 00:02:33,220
that it responds to.

58
00:02:33,220 --> 00:02:34,410
So for a given--

59
00:02:34,410 --> 00:02:37,290
any given photoreceptor
is responding usually

60
00:02:37,290 --> 00:02:39,330
to a very small part
of the visual field.

61
00:02:39,330 --> 00:02:42,510
These have this very small and
very simple receptive field.

62
00:02:42,510 --> 00:02:45,300
They respond-- if
there is light there,

63
00:02:45,300 --> 00:02:46,650
they change their response.

64
00:02:46,650 --> 00:02:50,187
If there is not, they go
back to their baseline.

65
00:02:50,187 --> 00:02:52,020
And some of them respond
to different colors

66
00:02:52,020 --> 00:02:54,386
preferentially, but
for the most part,

67
00:02:54,386 --> 00:02:56,260
they have these very
simple receptive fields.

68
00:02:59,480 --> 00:03:04,040
Ganglion cells,
on the other hand,

69
00:03:04,040 --> 00:03:06,350
because of what's called
lateral inhibition,

70
00:03:06,350 --> 00:03:08,800
have center-surround
receptive fields.

71
00:03:08,800 --> 00:03:11,870
So each ganglion cell is
looking at a little bit

72
00:03:11,870 --> 00:03:18,990
of the visual world
and this is kind

73
00:03:18,990 --> 00:03:25,360
of the classic representation
of an on-center ganglion cell.

74
00:03:25,360 --> 00:03:28,050
So this is a cell
that would respond

75
00:03:28,050 --> 00:03:31,830
if it's looking at a dark
surface or a dark screen

76
00:03:31,830 --> 00:03:35,460
like this, and it had
a little white dot

77
00:03:35,460 --> 00:03:39,120
right where the center of
the receptive field is--

78
00:03:39,120 --> 00:03:43,350
this is an excitatory center
and an inhibitory surround.

79
00:03:43,350 --> 00:03:47,310
So this ganglion cell
would respond a whole lot

80
00:03:47,310 --> 00:03:49,680
to this sort of stimulus.

81
00:03:49,680 --> 00:03:57,730
Whereas if it's looking
at just a blank field,

82
00:03:57,730 --> 00:04:01,240
then the excitatory center is
still going to get stimulated,

83
00:04:01,240 --> 00:04:04,240
but the inhibitory surround is
also going to get stimulated.

84
00:04:04,240 --> 00:04:07,420
And so the excitatory
and inhibitory stuff

85
00:04:07,420 --> 00:04:10,790
that's feeding in cancels out.

86
00:04:10,790 --> 00:04:13,330
So you can think of this as
if we're looking in the fovea

87
00:04:13,330 --> 00:04:15,871
in the center of your eye, where
all the receptive fields are

88
00:04:15,871 --> 00:04:18,250
really small because your
visual system is doing detail,

89
00:04:18,250 --> 00:04:19,666
it might look
something like this.

90
00:04:19,666 --> 00:04:22,737
We've got one cone.

91
00:04:22,737 --> 00:04:24,320
That's not what it
looks like, I know.

92
00:04:24,320 --> 00:04:25,930
They've got all
that squiggly stuff.

93
00:04:25,930 --> 00:04:29,350
And this cone-- and we've got
our ganglion cell down here.

94
00:04:32,210 --> 00:04:36,140
And this cone is feeding
in an excitatory impulse

95
00:04:36,140 --> 00:04:38,244
to that ganglion
cell, and that's

96
00:04:38,244 --> 00:04:40,160
going to be the center
of this ganglion cell's

97
00:04:40,160 --> 00:04:42,620
receptive field.

98
00:04:42,620 --> 00:04:52,000
But all of the cones
that surround this cone

99
00:04:52,000 --> 00:04:58,031
would be feeding inhibitory
impulses to the ganglion cell.

100
00:04:58,031 --> 00:05:00,280
And this is just straight
up excitatory and inhibitory

101
00:05:00,280 --> 00:05:03,010
synapses, right, like we
talked about last week?

102
00:05:03,010 --> 00:05:06,280
So that if just this
cone gets stimulated,

103
00:05:06,280 --> 00:05:10,390
then there's only an excitatory
input coming into that ganglion

104
00:05:10,390 --> 00:05:13,862
cell, and the ganglion
cell will, in turn, fire.

105
00:05:13,862 --> 00:05:15,820
If there's input coming
in from both the center

106
00:05:15,820 --> 00:05:19,840
and the surround, then the mix
of excitatory and inhibitory

107
00:05:19,840 --> 00:05:24,040
is going to actually decrease,
probably average out to zero,

108
00:05:24,040 --> 00:05:26,690
the way most of
these are weighted.

109
00:05:26,690 --> 00:05:28,930
If the cell is looking at
a stimulus that just looks

110
00:05:28,930 --> 00:05:34,840
like that, so that only the
inhibitory part is stimulated

111
00:05:34,840 --> 00:05:37,750
and the excitatory part
isn't getting anything,

112
00:05:37,750 --> 00:05:41,001
then it's going to actually
decrease its baseline firing

113
00:05:41,001 --> 00:05:41,500
rate.

114
00:05:41,500 --> 00:05:43,960
So all these cells kind
of have a base firing rate

115
00:05:43,960 --> 00:05:46,450
that if they're not getting
any kind of stimulation,

116
00:05:46,450 --> 00:05:49,240
they just tick, tick, tick.

117
00:05:49,240 --> 00:05:51,200
They just go every so often.

118
00:05:51,200 --> 00:05:54,820
And then that rate will
increase as the cell gets

119
00:05:54,820 --> 00:05:57,430
more excitatory stimulation,
or it can actually

120
00:05:57,430 --> 00:06:00,359
decrease as the cell gets
more inhibitory stimulation.

121
00:06:04,580 --> 00:06:07,030
So ganglion cells,
we get on-center.

122
00:06:07,030 --> 00:06:10,600
We also get off-center
ganglion cells.

123
00:06:10,600 --> 00:06:14,140
These are inhibitory in
the center and excitatory

124
00:06:14,140 --> 00:06:16,990
in the surround.

125
00:06:16,990 --> 00:06:21,550
Just like the on-center,
the off-center cell,

126
00:06:21,550 --> 00:06:27,370
if it's looking at a
plain field like this,

127
00:06:27,370 --> 00:06:28,870
it's not going to
respond very much.

128
00:06:28,870 --> 00:06:32,080
The excitatory and the
inhibitory parts cancel out.

129
00:06:32,080 --> 00:06:36,980
But this cell will be inhibited
by the single white dot

130
00:06:36,980 --> 00:06:41,759
and excited by the
ring with the dark dot.

131
00:06:41,759 --> 00:06:44,050
So they're responding to
opposite sorts of simulations,

132
00:06:44,050 --> 00:06:47,380
but in both cases, they're
not responding to these kind

133
00:06:47,380 --> 00:06:48,430
of featureless fields.

134
00:06:48,430 --> 00:06:50,596
They're responding to places
where the visual system

135
00:06:50,596 --> 00:06:54,550
is doing something more-- where
the visual stimulus is doing

136
00:06:54,550 --> 00:06:55,884
something more interesting.

137
00:07:08,250 --> 00:07:12,830
So like I said, so ganglion
cells are the cells whose axons

138
00:07:12,830 --> 00:07:15,980
run all the way down
the optic nerve,

139
00:07:15,980 --> 00:07:19,760
back into the lateral geniculate
nucleus of the thalamus,

140
00:07:19,760 --> 00:07:22,876
and from there, they get
passed to a new set of--

141
00:07:22,876 --> 00:07:23,750
they terminate there.

142
00:07:23,750 --> 00:07:26,180
They synapse onto more
neurons, and those neurons

143
00:07:26,180 --> 00:07:27,565
go to primary visual cortex.

144
00:07:33,700 --> 00:07:39,900
So did Xander show you
guys the scintillating grid

145
00:07:39,900 --> 00:07:42,240
illusion with the black
squares and the white grid

146
00:07:42,240 --> 00:07:43,470
in between them?

147
00:07:43,470 --> 00:07:45,780
Did you talk about
what causes that?

148
00:07:45,780 --> 00:07:47,535
AUDIENCE: She said no one knows.

149
00:07:47,535 --> 00:07:48,660
AUDIENCE: No, not that one.

150
00:07:48,660 --> 00:07:50,100
That was the other one.

151
00:07:50,100 --> 00:07:53,460
AUDIENCE: No, she said no
one knows for that one.

152
00:07:53,460 --> 00:07:55,390
Yeah.

153
00:07:55,390 --> 00:07:58,060
ABBY NOYCE: So to
some extent, it

154
00:07:58,060 --> 00:08:00,760
can be explained by looking
at the ganglion cells.

155
00:08:00,760 --> 00:08:02,770
The other kind of
classic example of this--

156
00:08:02,770 --> 00:08:05,174
I wish I had my slides--

157
00:08:05,174 --> 00:08:07,090
is what's called the
mock band illusion, where

158
00:08:07,090 --> 00:08:10,270
if you put rectangles of
light gray and dark gray

159
00:08:10,270 --> 00:08:13,110
next to each other, then right
at the border between them,

160
00:08:13,110 --> 00:08:15,610
the light gray one looks lighter
and the dark gray one looks

161
00:08:15,610 --> 00:08:18,190
darker compared to the
rest of the rectangle.

162
00:08:18,190 --> 00:08:21,340
And again, this kind of
ganglion cell response

163
00:08:21,340 --> 00:08:23,110
can explain some of that.

164
00:08:23,110 --> 00:08:26,920
So that there scintillating
grid, if you've got--

165
00:08:26,920 --> 00:08:28,990
new piece of board.

166
00:08:28,990 --> 00:08:32,390
You've got your boxes, right?

167
00:08:32,390 --> 00:08:42,320
Box, et cetera.

168
00:08:42,320 --> 00:08:44,130
That's a lot of chalk dust.

169
00:08:44,130 --> 00:08:47,242
So think about
what's happening--

170
00:08:49,900 --> 00:08:52,560
think about what's
happening to ganglion cells

171
00:08:52,560 --> 00:08:55,620
with-- let's just think about
on-center ganglion cells

172
00:08:55,620 --> 00:08:57,070
as they hang out here.

173
00:08:57,070 --> 00:09:00,720
So an on-center
ganglion cell that's

174
00:09:00,720 --> 00:09:03,650
scooched right up against the
edge of one of these boxes,

175
00:09:03,650 --> 00:09:08,400
so center-surround, its
center is going to get

176
00:09:08,400 --> 00:09:11,274
inhibited by the white part
of the grid that it's on.

177
00:09:11,274 --> 00:09:13,440
The center is going to get
excited by the white part

178
00:09:13,440 --> 00:09:15,600
of the grid that it's on, right?

179
00:09:15,600 --> 00:09:18,660
And half of the
surround is going

180
00:09:18,660 --> 00:09:20,430
to be inhibited but
the other half isn't,

181
00:09:20,430 --> 00:09:27,060
so the excitatory part still
trumps the inhibitory part.

182
00:09:27,060 --> 00:09:29,400
So this ganglion
cell will actually

183
00:09:29,400 --> 00:09:35,530
respond more than
a ganglion cell

184
00:09:35,530 --> 00:09:39,600
that's only looking
at the white part

185
00:09:39,600 --> 00:09:44,370
because of how part of
its inhibitory surround

186
00:09:44,370 --> 00:09:45,870
is not being exposed
to the white.

187
00:09:45,870 --> 00:09:47,286
Part of its
inhibitory surround is

188
00:09:47,286 --> 00:09:49,380
looking at the black
portion of the square.

189
00:09:51,915 --> 00:09:53,290
And one of the
things that people

190
00:09:53,290 --> 00:09:55,460
think is happening with
this here scintillating grid

191
00:09:55,460 --> 00:10:04,540
illusion is that ganglion cells
that are here and here and all

192
00:10:04,540 --> 00:10:07,870
around it, that
are schooched right

193
00:10:07,870 --> 00:10:11,290
up against the
edges of the squares

194
00:10:11,290 --> 00:10:15,430
are firing more than
a ganglion cell that's

195
00:10:15,430 --> 00:10:17,260
out here in the middle
of the intersection.

196
00:10:20,154 --> 00:10:21,820
So it ends up looking
like the bits that

197
00:10:21,820 --> 00:10:24,970
are between the squares are
brighter than the bit that's

198
00:10:24,970 --> 00:10:26,466
in the middle.

199
00:10:26,466 --> 00:10:27,340
So that's one theory.

200
00:10:30,150 --> 00:10:33,060
And I like it as
an example of how

201
00:10:33,060 --> 00:10:35,070
you can take this
really simple properties

202
00:10:35,070 --> 00:10:37,854
of a particular class of nerve
cells and what it responds to,

203
00:10:37,854 --> 00:10:39,270
line a couple of
them up together,

204
00:10:39,270 --> 00:10:41,837
and hey, look, a
perceptual effect,

205
00:10:41,837 --> 00:10:43,170
and perceptual effects are cool.

206
00:10:47,670 --> 00:10:49,670
AUDIENCE: Do we know
which ones are on-center

207
00:10:49,670 --> 00:10:51,274
and which are off-center?

208
00:10:51,274 --> 00:10:52,440
ABBY NOYCE: You can test it.

209
00:10:52,440 --> 00:10:55,620
I mean, the way
people find these is--

210
00:10:55,620 --> 00:10:57,130
and you've got both.

211
00:10:57,130 --> 00:10:59,160
The way people
have found them is

212
00:10:59,160 --> 00:11:01,860
you've got your classic
anesthetized cat lying

213
00:11:01,860 --> 00:11:04,290
on a table, and you're showing
it little dots of light.

214
00:11:04,290 --> 00:11:06,960
And you're sticking an
electrode into the cat's retina,

215
00:11:06,960 --> 00:11:10,170
actually or-- yeah, probably
directly into the retina

216
00:11:10,170 --> 00:11:13,530
or maybe into the optic nerve
and picking up the axons--

217
00:11:13,530 --> 00:11:16,410
and measuring what sorts of
stimuli are causing these cells

218
00:11:16,410 --> 00:11:18,210
to fire and which aren't.

219
00:11:18,210 --> 00:11:21,800
Which is how you find that each
cell has a very specific region

220
00:11:21,800 --> 00:11:23,456
of the visual field
that it looks at,

221
00:11:23,456 --> 00:11:25,955
and it responds in one of these
two characteristic patterns.

222
00:11:36,100 --> 00:11:39,820
So ganglion cells,
scintillating grid.

223
00:11:46,370 --> 00:11:48,069
So people knew for
quite some time,

224
00:11:48,069 --> 00:11:49,610
probably from the
early 20th century,

225
00:11:49,610 --> 00:11:52,940
that ganglion cells have
this receptive field pattern.

226
00:11:52,940 --> 00:12:01,660
And people were trying to figure
out how cells in visual cortex

227
00:12:01,660 --> 00:12:03,479
respond and how
cells in the thalamus

228
00:12:03,479 --> 00:12:05,270
and the lateral geniculate
nucleus respond.

229
00:12:05,270 --> 00:12:12,660
Remember, your information
pathway goes like eye to V1,

230
00:12:12,660 --> 00:12:13,160
right?

231
00:12:13,160 --> 00:12:18,325
So the thalamus here is
taking in the sensory input,

232
00:12:18,325 --> 00:12:19,700
doing a little
bit of processing,

233
00:12:19,700 --> 00:12:22,280
and passing it along to cortex.

234
00:12:22,280 --> 00:12:26,630
There's also actually an
awful lot of information

235
00:12:26,630 --> 00:12:30,020
that comes from all of these
other parts of visual cortex

236
00:12:30,020 --> 00:12:32,340
that goes back to the thalamus.

237
00:12:32,340 --> 00:12:35,450
So there's a lot of top-down
influence on the thalamus,

238
00:12:35,450 --> 00:12:40,040
and nobody is exactly sure
what that processing is doing.

239
00:12:40,040 --> 00:12:42,320
Some theories are
kind of like it

240
00:12:42,320 --> 00:12:45,175
lets you compare what you're
currently looking at to what

241
00:12:45,175 --> 00:12:46,550
you were looking
at a moment ago,

242
00:12:46,550 --> 00:12:48,590
so you can see things
like motion and changes

243
00:12:48,590 --> 00:12:52,070
and direct your attention to it.

244
00:12:52,070 --> 00:12:54,470
LGN also seems to be involved.

245
00:12:54,470 --> 00:12:56,387
You've got a reflex,
more or less.

246
00:12:56,387 --> 00:12:58,220
So if you hear a loud
noise off to one side,

247
00:12:58,220 --> 00:13:00,750
you'll usually turn
and look at it.

248
00:13:00,750 --> 00:13:04,550
And that seems to be
auditory information coming

249
00:13:04,550 --> 00:13:07,064
into the lateral
geniculate nucleus,

250
00:13:07,064 --> 00:13:08,480
coming into the
thalamus, at least

251
00:13:08,480 --> 00:13:12,740
from the auditory system, which
tells the lateral geniculate

252
00:13:12,740 --> 00:13:14,332
nucleus to send you
a signal to direct

253
00:13:14,332 --> 00:13:16,790
your visual attention over
where you heard the noise coming

254
00:13:16,790 --> 00:13:17,290
from.

255
00:13:21,214 --> 00:13:23,090
All right.

256
00:13:23,090 --> 00:13:25,850
So cells in the lateral
geniculate nucleus

257
00:13:25,850 --> 00:13:30,520
have on-center and
off-center receptive fields

258
00:13:30,520 --> 00:13:33,860
with, this is called
center-surround antagonism,

259
00:13:33,860 --> 00:13:36,350
that pattern, where the
center is doing one thing

260
00:13:36,350 --> 00:13:39,180
and the surround is
doing the opposite.

261
00:13:39,180 --> 00:13:43,082
And the cells in the
lateral geniculate nucleus

262
00:13:43,082 --> 00:13:45,040
have pretty much that
same pattern of response.

263
00:13:45,040 --> 00:13:47,690
So a given cell will
respond to a given thing.

264
00:13:47,690 --> 00:13:50,900
Cells in the visual
cortex, however, don't.

265
00:13:50,900 --> 00:13:54,810
They do something different.

266
00:13:54,810 --> 00:13:57,210
You guys, OK?

267
00:13:57,210 --> 00:14:01,620
So the classic example
of this is two guys

268
00:14:01,620 --> 00:14:04,680
named David Hubel and--

269
00:14:04,680 --> 00:14:06,020
shoot, what was his name?

270
00:14:06,020 --> 00:14:07,770
Like Thorston or something.

271
00:14:07,770 --> 00:14:09,850
Thorin, no, that's
somebody else--

272
00:14:09,850 --> 00:14:10,640
and Wiesel.

273
00:14:10,640 --> 00:14:12,420
Hubel and Wiesel
were researchers

274
00:14:12,420 --> 00:14:16,530
in the '50s, who were working
with anesthetized cats,

275
00:14:16,530 --> 00:14:20,610
trying to find receptive
fields for cells in the cat's

276
00:14:20,610 --> 00:14:22,700
primary visual cortex.

277
00:14:22,700 --> 00:14:28,410
So cat on table so that
it's facing a screen.

278
00:14:28,410 --> 00:14:31,470
The cat's anesthetized
so it holds still,

279
00:14:31,470 --> 00:14:33,420
and it has to look
at the screen.

280
00:14:33,420 --> 00:14:35,250
And they put little
microelectrodes

281
00:14:35,250 --> 00:14:37,020
into these parts
of visual cortex

282
00:14:37,020 --> 00:14:42,720
and tried showing the cat
different kinds of stimuli,

283
00:14:42,720 --> 00:14:46,766
dots, lights dots,
dark dots, dots

284
00:14:46,766 --> 00:14:48,390
in different parts
of the visual field.

285
00:14:48,390 --> 00:14:51,014
all of the stuff that will make
cells in the lateral geniculate

286
00:14:51,014 --> 00:14:53,670
nucleus so the ganglion cells
in the retina fire like mad,

287
00:14:53,670 --> 00:14:55,080
right?

288
00:14:55,080 --> 00:14:57,300
And it didn't work.

289
00:14:57,300 --> 00:15:00,360
And eventually, they
managed to find one cell

290
00:15:00,360 --> 00:15:03,570
that sometimes when
they showed it a dot,

291
00:15:03,570 --> 00:15:07,162
it responded to different
parts of the visual field.

292
00:15:07,162 --> 00:15:08,620
And they didn't
really think it was

293
00:15:08,620 --> 00:15:13,020
the dot of light
they were showing it

294
00:15:13,020 --> 00:15:14,910
that was the key stimulus.

295
00:15:14,910 --> 00:15:18,600
So eventually, they
managed to figure out

296
00:15:18,600 --> 00:15:23,040
that what was triggering
the cell to respond was,

297
00:15:23,040 --> 00:15:25,680
as they pulled slides in and
out of the side projector

298
00:15:25,680 --> 00:15:26,900
they were using--

299
00:15:26,900 --> 00:15:28,700
old school technology here--

300
00:15:28,700 --> 00:15:30,930
that as the edge of
the slide kind of

301
00:15:30,930 --> 00:15:33,940
cast a shadow that
moved across the screen,

302
00:15:33,940 --> 00:15:36,810
the cell was responding
to that, to a line,

303
00:15:36,810 --> 00:15:38,850
the particular angle,
in a particular part

304
00:15:38,850 --> 00:15:42,960
of the receptive field, moving
in a particular direction

305
00:15:42,960 --> 00:15:46,420
or moving perpendicular to
how the line was oriented.

306
00:15:46,420 --> 00:15:49,830
So and once they had found
that, they said, hey, look,

307
00:15:49,830 --> 00:15:52,500
these cells are responding to
something completely different

308
00:15:52,500 --> 00:15:55,530
than everything in the
LGN and in the retina.

309
00:15:55,530 --> 00:15:59,190
And pretty quickly, other
people started working on this.

310
00:15:59,190 --> 00:16:03,240
And they've managed to find
three different main types

311
00:16:03,240 --> 00:16:08,340
of cells in V1, and all of them
are dependent, really picky

312
00:16:08,340 --> 00:16:09,480
about orientation.

313
00:16:09,480 --> 00:16:11,010
They're all basically
line detectors

314
00:16:11,010 --> 00:16:11,980
of one kind or another.

315
00:16:18,170 --> 00:16:24,970
And the simplest one are cells
called simple cells, which

316
00:16:24,970 --> 00:16:30,010
respond to a line of a
particular orientation

317
00:16:30,010 --> 00:16:33,010
or an edge of a
particular orientation

318
00:16:33,010 --> 00:16:36,230
in a particular part
of the visual field.

319
00:16:36,230 --> 00:16:38,770
So they're like all
of-- like the cells

320
00:16:38,770 --> 00:16:41,355
earlier in the system,
like the ganglion cells

321
00:16:41,355 --> 00:16:42,730
and the lateral
geniculate cells,

322
00:16:42,730 --> 00:16:48,880
they're very picky about where
in space their stimuli are.

323
00:16:48,880 --> 00:16:56,170
And these cells look for
lines of a particular angle.

324
00:16:56,170 --> 00:17:01,600
So how do you go from
cells that look for this--

325
00:17:01,600 --> 00:17:04,150
how do you take that
kind of information

326
00:17:04,150 --> 00:17:08,140
and pass it along and
build a line detector using

327
00:17:08,140 --> 00:17:10,650
this kind of preliminary cell?

328
00:17:16,470 --> 00:17:20,530
AUDIENCE: Say if you're looking
at one thing and then a line

329
00:17:20,530 --> 00:17:26,250
moving, I guess, would cause
multiple cells to fire,

330
00:17:26,250 --> 00:17:27,960
like just one after another.

331
00:17:27,960 --> 00:17:29,220
ABBY NOYCE: Yeah, good.

332
00:17:29,220 --> 00:17:30,920
So yeah, it looks
like a lot of what's

333
00:17:30,920 --> 00:17:32,336
happening with
motion detecting is

334
00:17:32,336 --> 00:17:35,510
where you get sequential
firing over a particular area

335
00:17:35,510 --> 00:17:36,650
of the cortex.

336
00:17:36,650 --> 00:17:38,930
Frogs actually have
cells in their retina,

337
00:17:38,930 --> 00:17:40,640
really early in their
visual processing

338
00:17:40,640 --> 00:17:44,210
system that look for just
a small dot moving around.

339
00:17:44,210 --> 00:17:47,485
They've got fly detectors to
help them find their food.

340
00:17:47,485 --> 00:17:48,860
So they've got
cells that respond

341
00:17:48,860 --> 00:17:51,350
to motion of a small stimulus.

342
00:17:51,350 --> 00:17:53,522
We don't have those.

343
00:17:53,522 --> 00:17:55,940
AUDIENCE: I think I read
that frogs also they

344
00:17:55,940 --> 00:18:00,640
can't detect their food
if it's absolutely still.

345
00:18:00,640 --> 00:18:04,100
So if it's a fly on
the wall and it's not--