1
00:00:00,450 --> 00:00:03,736
The following content is
provided by MIT OpenCourseWare

2
00:00:03,736 --> 00:00:06,278
under a Creative Commons
license.

3
00:00:06,278 --> 00:00:09,329
Additional information about
our license and MIT

4
00:00:09,329 --> 00:00:13,904
OpenCourseWare in general is
available at ocw.mit.edu.

5
00:00:13,904 --> 00:00:18,640
PROFESSOR: What I'm going to do
for the next three lectures

6
00:00:18,640 --> 00:00:23,050
is to talk about how we see.

7
00:00:23,050 --> 00:00:27,210
What I'm going to do today
is talk about how we take

8
00:00:27,210 --> 00:00:29,320
information in.

9
00:00:29,320 --> 00:00:33,390
Next Tuesday, I'll talk about
the fact that we take in too

10
00:00:33,390 --> 00:00:36,820
much information -- more
information than we can use

11
00:00:36,820 --> 00:00:40,800
and we have to select amongst
that information.

12
00:00:43,940 --> 00:00:48,620
On Thursday of next week, what
I'll talk about is the fact

13
00:00:48,620 --> 00:00:55,020
that perception is never just a
direct, somehow registration

14
00:00:55,020 --> 00:00:56,190
of the outside world.

15
00:00:56,190 --> 00:00:59,335
It's always an active
interpretation of the world,

16
00:00:59,335 --> 00:01:02,900
and I'll talk about how we
go about interpreting the

17
00:01:02,900 --> 00:01:06,370
information that we've gotten
from the outside.

18
00:01:06,370 --> 00:01:09,300
I will be barely able
to scratch the

19
00:01:09,300 --> 00:01:11,040
surface of the topic.

20
00:01:11,040 --> 00:01:14,900
I will abandon all discussion
of the other senses, for

21
00:01:14,900 --> 00:01:16,800
instance, which is
a great pity.

22
00:01:16,800 --> 00:01:19,640
Now, it's true that I can barely
scratch the surface of

23
00:01:19,640 --> 00:01:23,790
almost any topic in this course,
but here the danger is

24
00:01:23,790 --> 00:01:25,170
greater because this
is actually

25
00:01:25,170 --> 00:01:27,620
what I do for a living.

26
00:01:27,620 --> 00:01:35,650
So, on any little bit of this,
I can talk forever.

27
00:01:35,650 --> 00:01:39,790
So, exactly how much of the
surface I scratch will be

28
00:01:39,790 --> 00:01:43,740
determined by how far afield
I end up going.

29
00:01:43,740 --> 00:01:50,210
This slide is up there to
illustrate, in some sense,

30
00:01:50,210 --> 00:01:51,440
those three bits.

31
00:01:51,440 --> 00:01:56,690
Let's see, if I take the stage,
chalkboard light off --

32
00:01:56,690 --> 00:01:59,170
there we go, you
want to go off?

33
00:01:59,170 --> 00:02:00,840
Then you can see a little
bit better.

34
00:02:04,180 --> 00:02:05,510
You can sort of see stuff.

35
00:02:09,820 --> 00:02:16,620
That's this Bruegel's painting
of the battle between the good

36
00:02:16,620 --> 00:02:20,440
angels and the bad angels.

37
00:02:20,440 --> 00:02:23,640
For me, the purpose is to say
look, it's clear that you see

38
00:02:23,640 --> 00:02:26,980
something, it's clear that
what you are seeing is an

39
00:02:26,980 --> 00:02:28,380
active interpretation.

40
00:02:28,380 --> 00:02:30,910
In fact, you may not have
particularly known that it was

41
00:02:30,910 --> 00:02:34,270
a battle between good angels
and bad angels

42
00:02:34,270 --> 00:02:35,890
before you look at it.

43
00:02:35,890 --> 00:02:39,230
It's also clear that even though
you, in some sense, see

44
00:02:39,230 --> 00:02:42,990
the whole thing, you're not
processing it fully, at least

45
00:02:42,990 --> 00:02:44,090
not all at once.

46
00:02:44,090 --> 00:02:47,060
So, for instance, you may not
have particularly noticed --

47
00:02:47,060 --> 00:02:49,330
I don't know what he is, he's
a really cool beast --

48
00:02:49,330 --> 00:02:52,460
I'm not sure if that's a bad
angel falling or something.

49
00:02:52,460 --> 00:02:54,350
But you probably hadn't
particularly

50
00:02:54,350 --> 00:02:56,950
noticed this fish guy.

51
00:02:56,950 --> 00:02:59,260
Actually, in this picture you're
probably still not

52
00:02:59,260 --> 00:03:02,470
noticing the fish guy -- he
just looks like a blob.

53
00:03:02,470 --> 00:03:06,440
It's a great picture,
look at it sometime.

54
00:03:06,440 --> 00:03:11,340
And this guy's got very bad
stomach problems, whatever

55
00:03:11,340 --> 00:03:12,010
he's doing.

56
00:03:12,010 --> 00:03:14,720
OK.

57
00:03:14,720 --> 00:03:18,380
Rather than continuing to try
to look at a murky picture,

58
00:03:18,380 --> 00:03:22,890
let's go to a less murky picture
and let's start by

59
00:03:22,890 --> 00:03:27,520
talking about how you actually
get information in from the

60
00:03:27,520 --> 00:03:31,790
visual system, which
obviously involves

61
00:03:31,790 --> 00:03:36,790
going through the eye.

62
00:03:36,790 --> 00:03:39,850
On the handout, which --
do I have a copy of

63
00:03:39,850 --> 00:03:42,350
it here, yes, OK.

64
00:03:42,350 --> 00:03:44,870
A lot of terminology.

65
00:03:44,870 --> 00:03:46,130
Are you responsible for it?

66
00:03:46,130 --> 00:03:47,690
Yes, of course.

67
00:03:47,690 --> 00:03:51,365
If you miss some little bit of
terminology, that's not going

68
00:03:51,365 --> 00:03:54,390
to sink you in the course,
either on this topic or on any

69
00:03:54,390 --> 00:03:56,080
other topic.

70
00:03:56,080 --> 00:03:59,090
You want to put your effort
into thinking about the

71
00:03:59,090 --> 00:04:04,630
broader topic, but it's useful
in being able to talk

72
00:04:04,630 --> 00:04:07,630
intelligently about it to get
a grip on some of this

73
00:04:07,630 --> 00:04:11,460
terminology, and hence, the tour
of the visual system that

74
00:04:11,460 --> 00:04:13,730
I will now embark on.

75
00:04:13,730 --> 00:04:19,240
From the outside-in, the purpose
of the eye, or at

76
00:04:19,240 --> 00:04:23,370
least the front half of the eye
is to form an image on the

77
00:04:23,370 --> 00:04:25,920
back half of the eye.

78
00:04:25,920 --> 00:04:29,890
So all of this, sometimes called
the anterior segment of

79
00:04:29,890 --> 00:04:33,700
the eye, is there to form an
image, like a camera would, on

80
00:04:33,700 --> 00:04:36,140
the back surface of the eye.

81
00:04:36,140 --> 00:04:41,090
Most of the optical power of the
eye is at the very front

82
00:04:41,090 --> 00:04:41,990
surface of it.

83
00:04:41,990 --> 00:04:45,250
That front surface is
called the cornea.

84
00:04:45,250 --> 00:04:51,760
It has, apart from its virtues
as a lens focusing light, it

85
00:04:51,760 --> 00:04:56,440
is also one of the two surfaces
in the body most

86
00:04:56,440 --> 00:04:58,870
sensitive to pain.

87
00:04:58,870 --> 00:05:02,950
You may try this out if
you want at home.

88
00:05:02,950 --> 00:05:06,460
I don't recommend it, but if
you scratch your cornea you

89
00:05:06,460 --> 00:05:09,230
will know it and you
will be happy.

90
00:05:09,230 --> 00:05:11,540
What's the other one?

91
00:05:11,540 --> 00:05:15,050
What's the other surface,
do you think?

92
00:05:15,050 --> 00:05:19,360
Anybody care to venture
a guess?

93
00:05:19,360 --> 00:05:20,610
Thumbs?

94
00:05:23,460 --> 00:05:26,720
No, that's a novel thought.

95
00:05:26,720 --> 00:05:28,250
No, look --

96
00:05:28,250 --> 00:05:30,530
he's thinking like the cartoons,
right, you hit the

97
00:05:30,530 --> 00:05:32,500
thumb with the hammer
and ow, it hurts.

98
00:05:32,500 --> 00:05:34,960
Let me tell you, if you hit your
cornea with the hammer it

99
00:05:34,960 --> 00:05:36,640
hurts a lot more.

100
00:05:36,640 --> 00:05:37,930
Yes?

101
00:05:37,930 --> 00:05:38,780
AUDIENCE: Some place
inside the mouth?

102
00:05:38,780 --> 00:05:40,360
PROFESSOR: Some place
inside the mouth.

103
00:05:40,360 --> 00:05:40,740
No.

104
00:05:40,740 --> 00:05:44,950
Again, you think about that
inside the mouth is a place

105
00:05:44,950 --> 00:05:48,760
where pain happens, but that's
because you do stupid things

106
00:05:48,760 --> 00:05:52,690
with your mouth, like hey,
that pizza looks good.

107
00:05:52,690 --> 00:05:52,980
Ow.

108
00:05:52,980 --> 00:05:56,370
You would never stick the
pizza on your eyeball.

109
00:05:59,760 --> 00:06:05,950
So, again, nicely innervated for
pain but not particularly.

110
00:06:05,950 --> 00:06:07,640
Not the other great location.

111
00:06:07,640 --> 00:06:08,830
I'll take one more here.

112
00:06:08,830 --> 00:06:10,425
AUDIENCE: [INAUDIBLE]

113
00:06:10,425 --> 00:06:12,020
at the bottom of your foot?

114
00:06:12,020 --> 00:06:14,260
PROFESSOR: Yeah, yeah, same
sort of thing, right.

115
00:06:14,260 --> 00:06:16,370
You step on nails
or something.

116
00:06:16,370 --> 00:06:18,220
The bottom of your foot's
a good place.

117
00:06:21,220 --> 00:06:23,720
There's two people here, I've
just gotta take their answers

118
00:06:23,720 --> 00:06:24,990
because they're so eager.

119
00:06:24,990 --> 00:06:27,085
AUDIENCE: Yeah, that's what I
was going to say, somewhere

120
00:06:27,085 --> 00:06:27,850
inside your ear.

121
00:06:27,850 --> 00:06:31,230
PROFESSOR: Somewhere inside your
ear is the point, not the

122
00:06:31,230 --> 00:06:33,820
chunk that so many people
are thrilled with.

123
00:06:33,820 --> 00:06:35,930
Oh, let's put a hole in
this -- oh, that was

124
00:06:35,930 --> 00:06:37,900
fun, let's do it again.

125
00:06:37,900 --> 00:06:40,280
That obviously hurts, I guess.

126
00:06:40,280 --> 00:06:41,820
I've never gone in
for this myself.

127
00:06:45,060 --> 00:06:48,490
One of the reasons your mother
said don't put anything inside

128
00:06:48,490 --> 00:06:52,700
your ear that's bigger than your
elbow is because if you

129
00:06:52,700 --> 00:06:55,230
put something small and pointy
in there, it eventually

130
00:06:55,230 --> 00:06:58,320
reaches your eardrum and that's
the other surface.

131
00:06:58,320 --> 00:07:05,610
When you had an ear infection as
a kid and the pressure from

132
00:07:05,610 --> 00:07:09,050
your middle ear, the fluid in
your ear, stretched that, it

133
00:07:09,050 --> 00:07:11,770
hurt, and that hurts a lot.

134
00:07:11,770 --> 00:07:16,060
So, don't put the hot pizza
in your ear either.

135
00:07:16,060 --> 00:07:16,900
OK.

136
00:07:16,900 --> 00:07:20,600
The cornea is just a relatively
thin layer.

137
00:07:20,600 --> 00:07:25,990
It's held in shape by fluid, a
watery-like fluid, which is

138
00:07:25,990 --> 00:07:28,810
known technically as the aqueous
humor -- humor in this

139
00:07:28,810 --> 00:07:30,360
case has nothing to do
with being funny.

140
00:07:30,360 --> 00:07:37,460
It's an old word for a fluid,
It forms a space that

141
00:07:37,460 --> 00:07:41,080
separates the cornea
from the iris.

142
00:07:41,080 --> 00:07:46,070
The iris is the bit that gives
your eye its color, but it's

143
00:07:46,070 --> 00:07:50,550
basically a sphincter of muscle
with a hole in it, and

144
00:07:50,550 --> 00:07:55,350
that hold is the pupil of the
eye, which is the aperture of

145
00:07:55,350 --> 00:07:56,210
the camera.

146
00:07:56,210 --> 00:08:03,600
That's where light's getting in
from the outside into the

147
00:08:03,600 --> 00:08:06,250
inner portions of the eye.

148
00:08:06,250 --> 00:08:08,470
OK.

149
00:08:08,470 --> 00:08:11,590
The pupil changes size, because
you've got this

150
00:08:11,590 --> 00:08:14,440
sphincter of muscle that either
relaxes or contracts.

151
00:08:14,440 --> 00:08:16,050
What makes it change size?

152
00:08:16,050 --> 00:08:17,760
AUDIENCE: Light.

153
00:08:17,760 --> 00:08:18,240
PROFESSOR: Light.

154
00:08:18,240 --> 00:08:18,980
We know that.

155
00:08:18,980 --> 00:08:20,510
More light makes it do what?

156
00:08:20,510 --> 00:08:21,550
AUDIENCE: Get smaller.

157
00:08:21,550 --> 00:08:22,350
PROFESSOR: Get smaller.

158
00:08:22,350 --> 00:08:23,040
Why?

159
00:08:23,040 --> 00:08:25,680
AUDIENCE: [INAUDIBLE].

160
00:08:25,680 --> 00:08:27,750
PROFESSOR: Yeah, so there's
less light.

161
00:08:27,750 --> 00:08:28,910
Sure, it's going to
be less light, but

162
00:08:28,910 --> 00:08:31,800
why do you have this?

163
00:08:31,800 --> 00:08:33,420
What did they teach
you in 3rd grade?

164
00:08:33,420 --> 00:08:34,950
AUDIENCE: [INAUDIBLE].

165
00:08:34,950 --> 00:08:36,290
PROFESSOR: Never stick your
elbow in your eye.

166
00:08:36,290 --> 00:08:38,810
Yes, I know that.

167
00:08:38,810 --> 00:08:39,820
No, I need a hand here.

168
00:08:39,820 --> 00:08:41,920
There's too many mutterings.

169
00:08:41,920 --> 00:08:44,910
There's a hand.

170
00:08:44,910 --> 00:08:44,995
AUDIENCE: Because you don't
want to go blind.

171
00:08:44,995 --> 00:08:46,090
PROFESSOR: Because you don't
want to go blind.

172
00:08:46,090 --> 00:08:47,310
Why would you go blind?

173
00:08:47,310 --> 00:08:48,530
AUDIENCE: There's
too much light.

174
00:08:48,530 --> 00:08:49,390
PROFESSOR: There's
too much light.

175
00:08:49,390 --> 00:08:51,070
See, that's what they teach
you in 3rd grade.

176
00:08:51,070 --> 00:08:53,390
It's bogus.

177
00:08:53,390 --> 00:09:01,040
It sounds good and it seems to
make sense, but ask yourself

178
00:09:01,040 --> 00:09:06,410
what level of protection that
pupil can give you.

179
00:09:06,410 --> 00:09:09,150
Right, so how much light
is it going to cut out?

180
00:09:09,150 --> 00:09:14,220
The answer is it cuts out a bit
more than a factor of 10.

181
00:09:14,220 --> 00:09:17,560
From the largest it can be to
the smallest it'll be if you

182
00:09:17,560 --> 00:09:19,470
stare at -- you know,
if you're out in

183
00:09:19,470 --> 00:09:20,660
really bright light.

184
00:09:20,660 --> 00:09:23,210
So it can take it can change
light levels by

185
00:09:23,210 --> 00:09:24,770
about a factor of 10.

186
00:09:24,770 --> 00:09:27,890
If you go from in here right now
-- actually, it's fairly

187
00:09:27,890 --> 00:09:29,210
dim in here --

188
00:09:29,210 --> 00:09:33,870
but if you go from here to
just go outside into the

189
00:09:33,870 --> 00:09:38,050
courtyard, the change in light
level is going to be on the

190
00:09:38,050 --> 00:09:40,880
order of a factor of
1,000 right there.

191
00:09:40,880 --> 00:09:43,680
The change in light level that
your visual system can deal

192
00:09:43,680 --> 00:09:48,830
with from, say, a moonlit night
to bright sun on a snow

193
00:09:48,830 --> 00:09:52,250
field is on the order of 10 to
the twelfth or 10 to the

194
00:09:52,250 --> 00:09:53,490
thirteenth.

195
00:09:53,490 --> 00:09:58,930
So there's no way that the need
to protect yourself is

196
00:09:58,930 --> 00:10:02,460
going to be taken care of
by the pupil of the eye.

197
00:10:02,460 --> 00:10:05,400
There's just not enough
range there.

198
00:10:05,400 --> 00:10:09,290
The pupillary constriction, what
it's probably doing for

199
00:10:09,290 --> 00:10:13,030
you, those of you who are
photographers, is it's

200
00:10:13,030 --> 00:10:15,980
increasing the depth of
field on your camera.

201
00:10:15,980 --> 00:10:19,900
A pinhole camera will
focus many different

202
00:10:19,900 --> 00:10:21,460
distances at once.

203
00:10:21,460 --> 00:10:26,840
A larger aperature will only
focus one plane at one time.

204
00:10:26,840 --> 00:10:29,830
And so if you've got lots of
light, you crank down the

205
00:10:29,830 --> 00:10:33,210
pupil in order to be able to
focus -- in order to get

206
00:10:33,210 --> 00:10:35,270
better focus, basically.

207
00:10:35,270 --> 00:10:38,110
But the protection thing
just doesn't --

208
00:10:38,110 --> 00:10:40,470
it's not a stupid answer because
it's what everybody

209
00:10:40,470 --> 00:10:46,430
believes, but it doesn't turn
out to be a correct answer.

210
00:10:46,430 --> 00:10:49,290
Oh, now there's several other
things that will change the

211
00:10:49,290 --> 00:10:52,100
size of your pupil, but one
other of interest, what else

212
00:10:52,100 --> 00:10:53,470
changes the size
of your pupil?

213
00:10:53,470 --> 00:10:55,210
Anybody know anything
else that they

214
00:10:55,210 --> 00:10:56,280
can change the size--?

215
00:10:56,280 --> 00:10:56,830
Yup?

216
00:10:56,830 --> 00:10:58,980
AUDIENCE: Looking all the
way down there at you.

217
00:10:58,980 --> 00:11:01,990
PROFESSOR: Looking all the
way down there at me.

218
00:11:01,990 --> 00:11:07,440
Well, yes, but for reasons
that are--.

219
00:11:07,440 --> 00:11:08,690
Why?

220
00:11:12,350 --> 00:11:13,990
Well, no, well you're
changing --

221
00:11:13,990 --> 00:11:16,770
sorry, that's the next
bullet here.

222
00:11:16,770 --> 00:11:19,710
What you're changing when you
change focus of the eye is

223
00:11:19,710 --> 00:11:22,850
you're changing the shape of the
lens -- at least you are.

224
00:11:22,850 --> 00:11:28,130
I'm not because at my age the
lens has basically become a

225
00:11:28,130 --> 00:11:34,270
rigid lump, and so when I want
to look from there to here, I

226
00:11:34,270 --> 00:11:36,390
use an external lens.

227
00:11:36,390 --> 00:11:37,620
Yes, back to the pupil.

228
00:11:37,620 --> 00:11:38,780
AUDIENCE: When you lie.

229
00:11:38,780 --> 00:11:39,570
PROFESSOR: When you lie.

230
00:11:39,570 --> 00:11:44,270
Yes, that will work a little bit
maybe, not reliably enough

231
00:11:44,270 --> 00:11:47,510
to -- you know, the CIA or
something every now and then

232
00:11:47,510 --> 00:11:50,380
gets the idea that that
really works.

233
00:11:50,380 --> 00:11:52,380
It's on the right
general track.

234
00:11:52,380 --> 00:11:52,760
Yeah?

235
00:11:52,760 --> 00:11:55,280
AUDIENCE: Some drugs do that?

236
00:11:55,280 --> 00:11:55,710
PROFESSOR: Yup.

237
00:11:55,710 --> 00:12:02,100
Some drugs will certainly
modulate it.

238
00:12:02,100 --> 00:12:03,000
All right, [UNINTELLIGIBLE].

239
00:12:03,000 --> 00:12:05,630
AUDIENCE: When you look at the
object of your affection.

240
00:12:05,630 --> 00:12:06,470
PROFESSOR: There you go.

241
00:12:06,470 --> 00:12:11,750
That's particularly the
one I was fishing for.

242
00:12:11,750 --> 00:12:16,510
Your level of arousal will
change the size of your pupil.

243
00:12:16,510 --> 00:12:20,680
Notably, if you look at
something that you like or

244
00:12:20,680 --> 00:12:25,180
someone that you like, your
pupils get bigger.

245
00:12:25,180 --> 00:12:28,420
Obviously, not everybody here
knew that explicitly, but you

246
00:12:28,420 --> 00:12:32,280
sort of knew that implicitly,
because the result of that is

247
00:12:32,280 --> 00:12:36,190
that you think that things with
big pupils are cuter than

248
00:12:36,190 --> 00:12:39,550
things a little pinhole
pupils, right.

249
00:12:39,550 --> 00:12:44,850
That's why sappy cartoons or
cards have big pupils.

250
00:12:44,850 --> 00:12:50,930
It is also why in years gone
by, women would put a drug

251
00:12:50,930 --> 00:12:54,440
known as atropine or known as
a cosmetic as belladonna for

252
00:12:54,440 --> 00:12:57,100
pretty woman, pretty girl.

253
00:12:57,100 --> 00:13:01,730
In their eyes what it did was
it paralyzed and relaxed the

254
00:13:01,730 --> 00:13:04,485
muscles of the irish and
made a great big pupil.

255
00:13:04,485 --> 00:13:06,560
It didn't do anything good for
their vision because it made

256
00:13:06,560 --> 00:13:08,290
things kind of blurry.

257
00:13:08,290 --> 00:13:10,860
It also paralyzing the lens.

258
00:13:10,860 --> 00:13:14,590
But it gave them great big
pupils, and so they're looking

259
00:13:14,590 --> 00:13:18,390
at the guy who may or may not be
the object of their desire

260
00:13:18,390 --> 00:13:21,120
-- they're looking at some guy,
they can't see him so

261
00:13:21,120 --> 00:13:25,190
well, which who knows,
that may have helped.

262
00:13:25,190 --> 00:13:30,210
But he's looking at her thinking
implicitly her pupils

263
00:13:30,210 --> 00:13:33,960
are really big, I know
what that means.

264
00:13:33,960 --> 00:13:36,110
Well, it turns out, following
up on whoever it was over

265
00:13:36,110 --> 00:13:37,900
there, it actually means
she's on drugs.

266
00:13:40,590 --> 00:13:47,700
In any case, pupil size is
modulated by emotional state

267
00:13:47,700 --> 00:13:50,500
and it's one of these things
that we sort of knew

268
00:13:50,500 --> 00:13:53,080
implicitly, people know
implicitly, and think big

269
00:13:53,080 --> 00:13:57,200
pupils are attractive
for that reason.

270
00:13:57,200 --> 00:13:59,210
It's only when you show up here
you find out these things

271
00:13:59,210 --> 00:14:00,530
explicitly.

272
00:14:00,530 --> 00:14:03,200
All right, I've already said
something about the lens.

273
00:14:03,200 --> 00:14:06,270
The most notable thing about
the lens is it gets harder

274
00:14:06,270 --> 00:14:08,170
with age and --

275
00:14:08,170 --> 00:14:09,840
you could try this out.

276
00:14:09,840 --> 00:14:15,870
In fact, it's a good time in
your life to start this

277
00:14:15,870 --> 00:14:17,690
experiment.

278
00:14:17,690 --> 00:14:21,410
Take something like the handout
and ask yourself how

279
00:14:21,410 --> 00:14:22,840
close you can get and
have the letters

280
00:14:22,840 --> 00:14:26,860
still be in sharp focus.

281
00:14:26,860 --> 00:14:29,760
This is so sad it like
makes me weep.

282
00:14:29,760 --> 00:14:32,430
But I know you're getting
there, too.

283
00:14:32,430 --> 00:14:34,330
Now what you can do --

284
00:14:34,330 --> 00:14:36,620
you could actually, if you're
compulsive about these sort of

285
00:14:36,620 --> 00:14:38,500
things, write it down
somewhere and

286
00:14:38,500 --> 00:14:40,660
just check every year.

287
00:14:40,660 --> 00:14:44,190
This is one of those things
where the signs of aging start

288
00:14:44,190 --> 00:14:46,610
showing up now.

289
00:14:46,610 --> 00:14:49,250
You will be able to see, even
during the course of your

290
00:14:49,250 --> 00:14:51,820
twenties that that so-called
near point is

291
00:14:51,820 --> 00:14:54,430
moving further out.

292
00:14:54,430 --> 00:14:59,310
For me without my glasses we're
out here somewhere.

293
00:14:59,310 --> 00:15:03,600
And eventually you're locked
at one plane of focus, and

294
00:15:03,600 --> 00:15:06,930
depending on where that is,
determines which particular

295
00:15:06,930 --> 00:15:08,995
set of glasses you need -- do
you need glasses to read, do

296
00:15:08,995 --> 00:15:10,530
you need bifocals, whatever.

297
00:15:10,530 --> 00:15:12,450
It's because you've lost
the ability to

298
00:15:12,450 --> 00:15:15,050
fine-tune that focus.

299
00:15:15,050 --> 00:15:17,690
Now, you need something
to hold the rest

300
00:15:17,690 --> 00:15:20,710
of the eye in shape.

301
00:15:20,710 --> 00:15:26,630
What you've got in there is it's
really a jelly called the

302
00:15:26,630 --> 00:15:31,970
vitreous humor for the
glass-like substance.

303
00:15:31,970 --> 00:15:34,080
For our purposes it's not
desperately interesting, it's

304
00:15:34,080 --> 00:15:38,410
there to hold your eye out so
that the front end of the eye

305
00:15:38,410 --> 00:15:40,990
can make an image on
this back surface.

306
00:15:40,990 --> 00:15:49,090
The back surface of the eye
where the transduction from

307
00:15:49,090 --> 00:15:52,670
photons, the light energy, into
signals in the nervous

308
00:15:52,670 --> 00:15:55,730
system where that takes place,
that tissue is known as the

309
00:15:55,730 --> 00:15:58,790
retina -- the word comes from
the word for a net.

310
00:16:04,420 --> 00:16:08,850
The fovea is the point --

311
00:16:08,850 --> 00:16:10,580
fovea comes the word
for pit --

312
00:16:10,580 --> 00:16:16,670
it's the point on the retina
where you have the highest

313
00:16:16,670 --> 00:16:19,460
visual aquity, the greatest
ability to

314
00:16:19,460 --> 00:16:22,660
resolve small details.

315
00:16:22,660 --> 00:16:28,160
You don't particularly realize
in the course of your normal

316
00:16:28,160 --> 00:16:31,980
day-to-day existence just
how bad vision gets

317
00:16:31,980 --> 00:16:36,130
away from the fovea.

318
00:16:36,130 --> 00:16:39,250
You have this notion of a highly
detailed world that's

319
00:16:39,250 --> 00:16:42,090
in focus all over the place.

320
00:16:42,090 --> 00:16:45,160
But that, to jump ahead to the
third lecture in this series,

321
00:16:45,160 --> 00:16:46,780
is a construction.

322
00:16:46,780 --> 00:16:47,910
It's not reality.

323
00:16:47,910 --> 00:16:49,820
And that you can prove
to yourself.

324
00:16:49,820 --> 00:16:53,430
Take the handout again, hold it
at some comfortable reading

325
00:16:53,430 --> 00:16:58,960
distance, fixate on the l on
that second line there where

326
00:16:58,960 --> 00:17:01,720
it says Lecture 4.

327
00:17:01,720 --> 00:17:06,100
If you're fixating on the l, the
l is forming an image on

328
00:17:06,100 --> 00:17:06,950
your fovea.

329
00:17:06,950 --> 00:17:10,160
Ask yourself while you're
fixated on the l, how many

330
00:17:10,160 --> 00:17:13,830
letters off to the right
you can actually read.

331
00:17:13,830 --> 00:17:18,170
And the answer is going
to be not many.

332
00:17:18,170 --> 00:17:20,130
If you think you can read to the
end of line it's because

333
00:17:20,130 --> 00:17:21,970
you're not fixating --

334
00:17:21,970 --> 00:17:24,460
you're moving your
eyes around.

335
00:17:24,460 --> 00:17:26,730
Again, like the pupil thing,
this is something you knew

336
00:17:26,730 --> 00:17:27,630
implicitly.

337
00:17:27,630 --> 00:17:33,420
Nobody sits there and says I'm
going to read this handout.

338
00:17:33,420 --> 00:17:38,180
I can see it, it's there, but
I can't read it because the

339
00:17:38,180 --> 00:17:43,270
outside of the fovea, I simply
don't have the resolving

340
00:17:43,270 --> 00:17:47,870
ability to do it.

341
00:17:47,870 --> 00:17:48,720
Why not?

342
00:17:48,720 --> 00:17:50,620
Well, I can give you at
least one reason.

343
00:17:50,620 --> 00:17:52,320
You have about a
hundred million

344
00:17:52,320 --> 00:17:55,040
photoreceptors in your retina.

345
00:17:55,040 --> 00:18:00,370
Photoreceptors are the cells
that convert light energy into

346
00:18:00,370 --> 00:18:01,960
a signal in the nervous
system, they're the

347
00:18:01,960 --> 00:18:04,920
transduction elements.

348
00:18:04,920 --> 00:18:10,810
You have only about a quarter
of a million to a million

349
00:18:10,810 --> 00:18:16,460
axons, nerve fibers running
from the eye to the brain.

350
00:18:16,460 --> 00:18:21,730
So that hundred million has got
to get boiled down to the

351
00:18:21,730 --> 00:18:25,750
much smaller number that you've
got available to you.

352
00:18:25,750 --> 00:18:46,530
So that means you can't treat
each photoreceptor as a pixel

353
00:18:46,530 --> 00:18:49,720
with a private line
to the brain.

354
00:18:49,720 --> 00:18:52,450
You do have something
approximating this in the

355
00:18:52,450 --> 00:18:56,450
fovea where you have fine
detailed resolution.

356
00:18:56,450 --> 00:19:01,520
But in the periphery, away from
the fovea, what you have

357
00:19:01,520 --> 00:19:06,580
is many, many, many, many
photoreceptors that are all

358
00:19:06,580 --> 00:19:10,860
ganged together to send their
signal off to the central

359
00:19:10,860 --> 00:19:11,910
nervous system.

360
00:19:11,910 --> 00:19:20,000
So you lose resolution out
there in the periphery.

361
00:19:20,000 --> 00:19:22,545
Let's say a little more about
the details of that retina.

362
00:19:22,545 --> 00:19:27,310
Oh, the optic nerve is the
bundle of fibers going off

363
00:19:27,310 --> 00:19:31,580
from the eye to the brain.

364
00:19:31,580 --> 00:19:34,800
I'll say a bit more about
that in a minute or so.

365
00:19:34,800 --> 00:19:39,280
A quick tour of the
retina itself.

366
00:19:39,280 --> 00:19:43,930
Up at the top there you've got
the photoreceptors, these

367
00:19:43,930 --> 00:19:45,420
transduction elements.

368
00:19:45,420 --> 00:19:49,310
There are two flavors of them
called rods and cones -- named

369
00:19:49,310 --> 00:19:52,040
rods and cones because under
the microscope rods are

370
00:19:52,040 --> 00:19:53,320
rod-shaped and cones
are cone-shaped.

371
00:19:56,360 --> 00:20:01,140
They are different in their
functions and capabilities.

372
00:20:01,140 --> 00:20:06,060
Cones operate in bright light
at daylight levels.

373
00:20:06,060 --> 00:20:10,640
They do mediate color
vision and they are

374
00:20:10,640 --> 00:20:13,150
concentrated at the fovea.

375
00:20:13,150 --> 00:20:18,150
They are most densely
represented at the fovea.

376
00:20:18,150 --> 00:20:22,380
Rods work in dim light levels.

377
00:20:22,380 --> 00:20:29,080
They are not able to mediate
color vision, which is why on

378
00:20:29,080 --> 00:20:31,550
a moonlit night you don't see
colors -- it's not because

379
00:20:31,550 --> 00:20:33,730
color's somehow drained
out of the world.

380
00:20:33,730 --> 00:20:36,730
It's because the visual system
that's looking at that world

381
00:20:36,730 --> 00:20:41,930
can no longer analyze the input
for color information,

382
00:20:41,930 --> 00:20:46,360
but only for grey scaled
information.

383
00:20:46,360 --> 00:20:49,110
Rod photoreceptors are
concentrated away

384
00:20:49,110 --> 00:20:50,100
from the retina --

385
00:20:50,100 --> 00:20:53,250
I'm sorry, away from
the fovea.

386
00:20:53,250 --> 00:20:57,260
They are, in fact, absent
from the central fovea.

387
00:21:00,610 --> 00:21:04,190
The best place to see this
actually is if your room gets

388
00:21:04,190 --> 00:21:10,930
dim enough at night that you
can't see color, you're

389
00:21:10,930 --> 00:21:14,990
working in the rod realm, and
then you will discover that if

390
00:21:14,990 --> 00:21:18,430
you fixate on a dim spot of
light you can't see it.

391
00:21:18,430 --> 00:21:20,620
That there might be something
out here -- you notice it and

392
00:21:20,620 --> 00:21:23,700
then you move your eye to look
at it and it'll disappear,

393
00:21:23,700 --> 00:21:26,890
because the image, if it's small
enough, will now fall

394
00:21:26,890 --> 00:21:30,690
into the fovea and there
are no rods there,

395
00:21:30,690 --> 00:21:33,170
you can't see it.

396
00:21:33,170 --> 00:21:36,970
This is why if you're a star
gazer and you're looking at a

397
00:21:36,970 --> 00:21:40,290
dim star, you want to look at
that star out of the so-called

398
00:21:40,290 --> 00:21:41,930
corner of your eye.

399
00:21:41,930 --> 00:21:44,690
The corner of your eye is about
20 degrees away from

400
00:21:44,690 --> 00:21:47,620
straight ahead because that's
where the density of rod

401
00:21:47,620 --> 00:21:50,660
photoreceptors turns
out to be greatest.

402
00:21:54,750 --> 00:21:57,880
I was going to say something
else about rod photoreceptors,

403
00:21:57,880 --> 00:21:59,600
but that has disappeared
from my mind.

404
00:21:59,600 --> 00:22:06,410
So I will say that the rest of
this is a collection of nerve

405
00:22:06,410 --> 00:22:08,200
cells, of neurons.

406
00:22:08,200 --> 00:22:10,390
It's really a little
chunk of brain.

407
00:22:10,390 --> 00:22:13,310
Developmentally the retina is a
chunk of brain that's pushed

408
00:22:13,310 --> 00:22:18,070
out into the eye, and this is
the start of the visual

409
00:22:18,070 --> 00:22:20,320
nervous system.

410
00:22:20,320 --> 00:22:24,220
For present purposes it's
worth thinking about two

411
00:22:24,220 --> 00:22:28,290
aspects of this organization.

412
00:22:28,290 --> 00:22:37,480
One is a through path that runs
from the photoreceptors

413
00:22:37,480 --> 00:22:42,450
through so-called bipolar
cells to ganglion cells.

414
00:22:42,450 --> 00:22:45,870
There's sort of a vertical path
in this picture heading

415
00:22:45,870 --> 00:22:49,320
from the photoreceptors
out to the brain.

416
00:22:49,320 --> 00:22:55,530
Then there are cells whose
processes are sitting here and

417
00:22:55,530 --> 00:22:59,610
here making horizontal
connections so that

418
00:22:59,610 --> 00:23:03,150
photoreceptors don't behave like
isolated pixels so that

419
00:23:03,150 --> 00:23:06,030
they can talk to each other,
and so you can start doing

420
00:23:06,030 --> 00:23:09,260
some sort calculations,
if you like.

421
00:23:09,260 --> 00:23:11,680
Those are cells like horizontal
cells that have

422
00:23:11,680 --> 00:23:15,120
their processes up here, and
amacrine cells that have their

423
00:23:15,120 --> 00:23:16,880
processes down here.

424
00:23:16,880 --> 00:23:19,770
So those are things making
horizontal connections across

425
00:23:19,770 --> 00:23:20,980
the retina.

426
00:23:20,980 --> 00:23:24,420
Photoreceptors, bipolar cells,
ganglion cells are forming

427
00:23:24,420 --> 00:23:29,950
this through pathway going
from eye to brain.

428
00:23:29,950 --> 00:23:35,350
Now, if you were to look at this
and I was to say where is

429
00:23:35,350 --> 00:23:39,630
light coming from in this
picture, the reasonable

430
00:23:39,630 --> 00:23:41,210
answer would be?

431
00:23:41,210 --> 00:23:42,460
AUDIENCE: [INAUDIBLE].

432
00:23:49,180 --> 00:23:50,850
PROFESSOR: It would seem like
a reasonable answer would be

433
00:23:50,850 --> 00:23:52,980
like it's coming from the top
hit in the photoreceptors and

434
00:23:52,980 --> 00:23:54,440
heading off to the brain
and out the bottom.

435
00:23:54,440 --> 00:23:57,390
But in fact, light's
coming from there.

436
00:23:57,390 --> 00:24:00,560
The retina is put on backwards
in an interesting kind of a

437
00:24:00,560 --> 00:24:05,200
way, or intuitively backwards at
least, so that light shines

438
00:24:05,200 --> 00:24:11,740
through all of this stuff before
actually getting to the

439
00:24:11,740 --> 00:24:13,210
photoreceptors.

440
00:24:13,210 --> 00:24:15,190
Why is that?

441
00:24:15,190 --> 00:24:17,650
Well, there are a variety
of possible answers.

442
00:24:17,650 --> 00:24:22,300
One of the answers is -- let's
see, the remaining term on the

443
00:24:22,300 --> 00:24:28,610
list that I haven't mentioned in
that retina list, which is

444
00:24:28,610 --> 00:24:30,370
actually at the top ofo the
list, it says pigment

445
00:24:30,370 --> 00:24:31,620
epithelium.

446
00:24:36,420 --> 00:24:42,180
Up here there would be a layer
of what boils down to black

447
00:24:42,180 --> 00:24:46,220
gunk coating the back surface
of the eye, and the

448
00:24:46,220 --> 00:24:49,420
photoreceptors are
stuck into that.

449
00:24:49,420 --> 00:24:51,110
That's the pigment epithelium.

450
00:24:51,110 --> 00:24:57,460
What it's there for is -- well,
let's suppose that all

451
00:24:57,460 --> 00:25:03,150
of you guys are photoreceptors
and here I am, I'm a photon

452
00:25:03,150 --> 00:25:08,480
and I'm part of this image of
the world on the outside and

453
00:25:08,480 --> 00:25:12,910
I'm coming flying in, and with
luck I'll be absorbed by Mr.

454
00:25:12,910 --> 00:25:15,010
Photoreceptor here.

455
00:25:15,010 --> 00:25:16,470
But suppose I don't?

456
00:25:16,470 --> 00:25:18,750
Suppose I hit over here.

457
00:25:18,750 --> 00:25:24,090
Well, photons don't have
much conscious life

458
00:25:24,090 --> 00:25:25,850
as far as we know.

459
00:25:25,850 --> 00:25:28,690
You, the owner of this visual
system, don't want that photon

460
00:25:28,690 --> 00:25:30,040
bouncing around anywhere else.

461
00:25:30,040 --> 00:25:33,710
You want it either to be
absorbed by a photographers or

462
00:25:33,710 --> 00:25:34,890
to be gone.

463
00:25:34,890 --> 00:25:37,360
Because if it bounces and hits
up there somewhere it's going

464
00:25:37,360 --> 00:25:38,800
to degrade the image, right.

465
00:25:38,800 --> 00:25:41,550
It's going to be like haze
across the image if it's not

466
00:25:41,550 --> 00:25:47,610
landing where it should to
make a nice starp image.

467
00:25:47,610 --> 00:25:55,640
So this black gunk is there to
soak up extra photoreceptors.

468
00:25:55,640 --> 00:26:00,330
A very good idea if you are a
diurnal animal, an animal who

469
00:26:00,330 --> 00:26:02,720
works out in bright light.

470
00:26:02,720 --> 00:26:07,320
Not such a good idea if you're
an animal who is nocturnal or

471
00:26:07,320 --> 00:26:09,180
running around in dim light.

472
00:26:09,180 --> 00:26:12,320
There what you want to do is
catch every photon that you

473
00:26:12,320 --> 00:26:14,010
can possibly catch.

474
00:26:14,010 --> 00:26:17,600
So if you're a cat, for
example, you do this

475
00:26:17,600 --> 00:26:18,590
differently.

476
00:26:18,590 --> 00:26:20,860
Instead of a pigment epithelium
at the back of your

477
00:26:20,860 --> 00:26:28,370
eye, you have a structure
known as the tapetum --

478
00:26:28,370 --> 00:26:31,120
it's actually very beautiful, it
looks like Mother of Pearl,

479
00:26:31,120 --> 00:26:35,440
but you only get to see that if
you take it out of the cat,

480
00:26:35,440 --> 00:26:39,220
which makes the cat
less beautiful.

481
00:26:39,220 --> 00:26:41,970
But it's a reflective surface.

482
00:26:41,970 --> 00:26:48,490
This is why if you get a cat in
your flashlight beam, its

483
00:26:48,490 --> 00:26:53,430
eyes seem to glow at you,
because the light comes in --

484
00:26:53,430 --> 00:26:56,370
if it doesn't get absorbed by,
say, the photoreceptors, it

485
00:26:56,370 --> 00:26:59,780
can bounce and an amount of
it will bounce back out.

486
00:26:59,780 --> 00:27:02,200
Some amount of light will bounce
back out of your eyes.

487
00:27:02,200 --> 00:27:07,680
That's what you see as
red eye in an unlucky

488
00:27:07,680 --> 00:27:09,510
photograph, for example.

489
00:27:09,510 --> 00:27:10,270
Is that a hand?

490
00:27:10,270 --> 00:27:15,940
AUDIENCE: Wouldn't that
distort the image?

491
00:27:15,940 --> 00:27:17,170
PROFESSOR: To the cat.

492
00:27:17,170 --> 00:27:17,390
AUDIENCE: Yeah.

493
00:27:17,390 --> 00:27:18,550
PROFESSOR: Yeah, absolutely.

494
00:27:18,550 --> 00:27:21,230
That's why you almost never
see cats reading

495
00:27:21,230 --> 00:27:23,170
The New York Times.

496
00:27:23,170 --> 00:27:27,510
No, cat visual aquity
is pretty lousy.

497
00:27:27,510 --> 00:27:33,760
A cat has different visual
desires than you and I do.

498
00:27:33,760 --> 00:27:38,470
The cat's not big on sweating
the visual aquity thing.

499
00:27:38,470 --> 00:27:40,290
The cat wants to know,
did it move?

500
00:27:40,290 --> 00:27:41,270
Can I jump on it?

501
00:27:41,270 --> 00:27:43,340
Can I bite it?

502
00:27:43,340 --> 00:27:44,260
Is it another cat?

503
00:27:44,260 --> 00:27:46,110
Can we do stuff?

504
00:27:46,110 --> 00:27:47,790
All this sort of stuff.

505
00:27:47,790 --> 00:27:50,370
This sort of thing doesn't
require a lot of visual aquity

506
00:27:50,370 --> 00:27:54,960
and the cat wants to be able
to do it in dim light.

507
00:27:54,960 --> 00:27:59,390
So, we have this notion that
cats can see in the dark,

508
00:27:59,390 --> 00:28:01,300
which, of course, is
not really true --

509
00:28:01,300 --> 00:28:04,520
no light, no vision.

510
00:28:04,520 --> 00:28:07,330
People think cats can see in the
dark because cats can see

511
00:28:07,330 --> 00:28:10,200
stuff in dim light that you and
I can't, and part of the

512
00:28:10,200 --> 00:28:15,490
reason for that is that they
are willing to let stray

513
00:28:15,490 --> 00:28:16,630
photons go where they will.

514
00:28:16,630 --> 00:28:20,230
Because, in fact, I gave a sort
of a radical example.

515
00:28:20,230 --> 00:28:25,290
If my hypothetical photon
missed my hypothetical

516
00:28:25,290 --> 00:28:29,225
photoreceptor here, odds are
it's not going to go out to

517
00:28:29,225 --> 00:28:32,690
outer Mongolia, it's going to
get absorbed someplace nearby,

518
00:28:32,690 --> 00:28:35,910
and so the effect will be a
blurring, not a complete

519
00:28:35,910 --> 00:28:37,710
degradation of the image.

520
00:28:37,710 --> 00:28:39,430
I think the cat can cope.

521
00:28:39,430 --> 00:28:42,220
The cat doesn't mind too much.

522
00:28:42,220 --> 00:28:51,660
By the way, if you're interested
in design, taking a

523
00:28:51,660 --> 00:28:59,280
look at the design of eyes
across the animal kingdom is a

524
00:28:59,280 --> 00:29:01,040
beautiful--.

525
00:29:01,040 --> 00:29:06,040
The number of different ways
that evolution or the hand of

526
00:29:06,040 --> 00:29:08,980
God, you can take your pick,
has chosen to solve the

527
00:29:08,980 --> 00:29:13,130
problem of collecting light,
an amazing array of

528
00:29:13,130 --> 00:29:14,450
possibilities.

529
00:29:14,450 --> 00:29:15,790
Once you start --

530
00:29:15,790 --> 00:29:22,560
cats are one thing, but you go
off into insects and other

531
00:29:22,560 --> 00:29:24,100
beasties and it's great.

532
00:29:24,100 --> 00:29:25,580
Wonderful stuff.

533
00:29:25,580 --> 00:29:31,360
If you want to be an engineer,
probably lots of good

534
00:29:31,360 --> 00:29:33,520
industrial design there.

535
00:29:33,520 --> 00:29:37,310
All right, let me take a quick
trip up into the rest of the

536
00:29:37,310 --> 00:29:40,690
visual system here.

537
00:29:40,690 --> 00:29:45,660
This picture that's up on the
screen now is mimicking the

538
00:29:45,660 --> 00:29:46,980
picture on the upper
left of the

539
00:29:46,980 --> 00:29:49,840
second page of the handout.

540
00:29:49,840 --> 00:29:53,000
It's showing the connections
-- oh, I just remembered.

541
00:29:53,000 --> 00:29:55,770
I want to do a quick demo
here before I do that.

542
00:29:55,770 --> 00:29:57,020
Let's go back and do that.

543
00:29:59,590 --> 00:30:03,100
If the retina is on
backwards, you got

544
00:30:03,100 --> 00:30:07,100
this interesting problem.

545
00:30:07,100 --> 00:30:12,110
So, there are those nerve fibers
to the optic nerve.

546
00:30:12,110 --> 00:30:15,960
How are they going to get out of
the eye if light is coming

547
00:30:15,960 --> 00:30:18,370
from there?

548
00:30:18,370 --> 00:30:25,460
What they do is they run across
the surface of the

549
00:30:25,460 --> 00:30:29,090
retina until they reach the
optic nerve and then they go

550
00:30:29,090 --> 00:30:31,350
out up to the brain.

551
00:30:31,350 --> 00:30:34,270
Where are the photoreceptors
at the optic nerve?

552
00:30:34,270 --> 00:30:36,290
The answer is there aren't
any photo receptors

553
00:30:36,290 --> 00:30:37,410
at the optic nerve.

554
00:30:37,410 --> 00:30:39,510
If there aren't any
photoreceptors at the optic

555
00:30:39,510 --> 00:30:42,040
nerve, what do you see there?

556
00:30:42,040 --> 00:30:46,830
The answer isn't nothing because
you're not aware of

557
00:30:46,830 --> 00:30:50,310
the fact that there's some big
hole in your visual field.

558
00:30:50,310 --> 00:30:53,250
But there is.

559
00:30:53,250 --> 00:30:57,700
You will automatically fill it
in because if it's green here

560
00:30:57,700 --> 00:31:00,540
and green here and there's no
information there, it's a good

561
00:31:00,540 --> 00:31:02,650
bet that it's probably green
all the way across and your

562
00:31:02,650 --> 00:31:04,540
visual system can
figure this out.

563
00:31:04,540 --> 00:31:08,040
But it's worth it to prove it
to yourself that there is a

564
00:31:08,040 --> 00:31:09,240
blind spot there.

565
00:31:09,240 --> 00:31:10,740
Let's see, on the handout --

566
00:31:10,740 --> 00:31:13,000
well, flip to the back of the
handout where you got a little

567
00:31:13,000 --> 00:31:23,390
extra white space, and put an
x on handout here like this,

568
00:31:23,390 --> 00:31:26,070
and a little ways away like
about like that, put a little

569
00:31:26,070 --> 00:31:27,730
black dot or a little dot --

570
00:31:27,730 --> 00:31:28,980
I don't care if it's black.

571
00:31:32,440 --> 00:31:37,070
So the dot is to the
right of the x.

572
00:31:37,070 --> 00:31:40,710
Close the left eye, cover
your left eye.

573
00:31:40,710 --> 00:31:44,630
Look at the x, hold it so that
the dot is straight out to the

574
00:31:44,630 --> 00:31:49,860
right and then move the paper
back and forth and you will

575
00:31:49,860 --> 00:31:53,710
find that there's a spot where
the dot completely disappears.

576
00:31:53,710 --> 00:31:59,120
What you've done at that point
is move the dot into the optic

577
00:31:59,120 --> 00:32:01,440
nerve head -- the place
where the optic nerve

578
00:32:01,440 --> 00:32:02,690
is leaving the eye.

579
00:32:07,590 --> 00:32:11,140
If you're saying I can't do
this, I don't see a blind

580
00:32:11,140 --> 00:32:14,960
spot, it's not because you're
special and don't have one,

581
00:32:14,960 --> 00:32:16,450
it's because you're
doing it wrong.

582
00:32:19,120 --> 00:32:22,160
Well, there is an alternative.

583
00:32:22,160 --> 00:32:25,460
I'm given to understand that the
octopus retina is not put

584
00:32:25,460 --> 00:32:28,460
on backwards in this way and so
perhaps you're an octopus.

585
00:32:33,630 --> 00:32:38,470
OK, let's go back to this
wiring diagram here.

586
00:32:46,850 --> 00:32:49,150
You can play with this later.

587
00:32:49,150 --> 00:32:50,310
It's cool.

588
00:32:50,310 --> 00:32:53,620
Do things like -- well, if you
flip it around you can find

589
00:32:53,620 --> 00:32:55,160
that you got one in the
other eye, too.

590
00:32:55,160 --> 00:32:59,210
You can also do things like draw
a line with a hole in it.

591
00:32:59,210 --> 00:33:01,450
If you draw a line with a gap
in it and put the gap in the

592
00:33:01,450 --> 00:33:04,310
hole, the line will appear
to complete.

593
00:33:04,310 --> 00:33:06,990
If you get really adept at
this you can do what King

594
00:33:06,990 --> 00:33:10,260
Charles the Second of England
was reputed to have liked to

595
00:33:10,260 --> 00:33:14,430
do, which is you can look at
your friends and put their

596
00:33:14,430 --> 00:33:19,720
heads in blind spots and watch
their heads disappear.

597
00:33:19,720 --> 00:33:23,220
Charles the Second didn't get
to chop off as many heads as

598
00:33:23,220 --> 00:33:25,230
his predecessors had,
so this was the best

599
00:33:25,230 --> 00:33:26,670
he could do I guess.

600
00:33:26,670 --> 00:33:28,690
All right.

601
00:33:28,690 --> 00:33:36,260
How do the axons from the optic
nerve feed the visual

602
00:33:36,260 --> 00:33:39,270
centers of the brain?

603
00:33:39,270 --> 00:33:43,720
You will recall, from the first
lecture perhaps, that

604
00:33:43,720 --> 00:33:51,630
everything from the right side
of my body on the skin senses

605
00:33:51,630 --> 00:33:54,320
gets represented in the sensory
homunculus on the

606
00:33:54,320 --> 00:33:57,530
left, and everything from the
left in the sensory homunculus

607
00:33:57,530 --> 00:34:00,550
on the right.

608
00:34:00,550 --> 00:34:04,980
The way to not do this
individual system would be to

609
00:34:04,980 --> 00:34:08,040
have let's have everything
from the right, that's my

610
00:34:08,040 --> 00:34:10,300
right eye, my right eye go to
my left hemisphere, and

611
00:34:10,300 --> 00:34:12,950
everything from the left eye
go to the right hemisphere.

612
00:34:12,950 --> 00:34:14,900
That wouldn't be any good.

613
00:34:14,900 --> 00:34:17,890
Isaac Newton figured out without
bothering to do any

614
00:34:17,890 --> 00:34:19,760
anatomy, Isaac Newton figured
out that that

615
00:34:19,760 --> 00:34:20,900
wouldn't be any good.

616
00:34:20,900 --> 00:34:24,070
Because I want to look out at
the world with two eyes and I

617
00:34:24,070 --> 00:34:25,870
want to see one world.

618
00:34:25,870 --> 00:34:28,260
So I'm going to have to do
something that brings the

619
00:34:28,260 --> 00:34:31,280
information from the
two eyes together.

620
00:34:31,280 --> 00:34:36,250
What you actually do
is the following.

621
00:34:36,250 --> 00:34:41,690
Everything from the left side of
each retina ends up in the

622
00:34:41,690 --> 00:34:44,020
left hemisphere, and everything
from the right side

623
00:34:44,020 --> 00:34:48,930
of each retina ends up in
the right hemisphere.

624
00:34:48,930 --> 00:34:53,350
The image on eye is flipped --

625
00:34:53,350 --> 00:34:56,120
simple lenses do that, right.

626
00:34:56,120 --> 00:35:00,010
So I've got a simple lens
in that anterior

627
00:35:00,010 --> 00:35:01,250
portion of my eye.

628
00:35:01,250 --> 00:35:06,180
That means that the image of you
guys landing on my retain

629
00:35:06,180 --> 00:35:09,120
has your feet up, your
head down and

630
00:35:09,120 --> 00:35:10,900
left and right reversed.

631
00:35:10,900 --> 00:35:19,680
So what that means here is if I
stare at the person wearing

632
00:35:19,680 --> 00:35:22,550
yellow here --

633
00:35:22,550 --> 00:35:24,250
if I stare at the person wearing
yellow she'll look

634
00:35:24,250 --> 00:35:25,170
embarrassed.

635
00:35:25,170 --> 00:35:27,480
It always works but that's
a different phenomenon.

636
00:35:27,480 --> 00:35:29,410
All right, but I'm staring at
you but I'm not actually

637
00:35:29,410 --> 00:35:30,600
paying any attention to you.

638
00:35:30,600 --> 00:35:32,890
What I'm actually doing
is looking at --

639
00:35:32,890 --> 00:35:37,000
my aquity's so bad, it's a black
blob right about there.

640
00:35:37,000 --> 00:35:40,180
I assume she's a person --

641
00:35:40,180 --> 00:35:42,780
yeah, she's a person, OK.

642
00:35:42,780 --> 00:35:44,240
Anyway, acquity's very bad.

643
00:35:47,090 --> 00:35:52,080
If you're on my fovea, the woman
to the right is in the

644
00:35:52,080 --> 00:35:55,540
right half of my visual
field landing on the

645
00:35:55,540 --> 00:35:57,170
left side of my retina.

646
00:35:57,170 --> 00:36:00,590
She lands on the left side in
both eyes, and as a result,

647
00:36:00,590 --> 00:36:02,770
she's landing in my
left hemisphere.

648
00:36:02,770 --> 00:36:06,040
If I'm looking at you, some
arbitrary blob -- there's a

649
00:36:06,040 --> 00:36:07,970
blob that just moved
over there.

650
00:36:07,970 --> 00:36:10,460
Oh, she's a person, too.

651
00:36:10,460 --> 00:36:13,820
This person in the left visual
field ends up on the right

652
00:36:13,820 --> 00:36:17,930
retain in each eye and in
the right hemisphere.

653
00:36:17,930 --> 00:36:20,890
So there's only one of her, in
a sense, up in my brain, and

654
00:36:20,890 --> 00:36:26,220
only one of her up in
my brain and they're

655
00:36:26,220 --> 00:36:27,320
getting together in that--.

656
00:36:27,320 --> 00:36:30,600
And the result is that half the
fibers from each retina

657
00:36:30,600 --> 00:36:32,490
have to cross to
the other side.

658
00:36:32,490 --> 00:36:40,230
That's what this cute little map
on page 2 is telling you.

659
00:36:40,230 --> 00:36:44,850
But the important thing not to
get confused about is don't

660
00:36:44,850 --> 00:36:48,590
tell us on some exam that your
left eye goes to your right

661
00:36:48,590 --> 00:36:50,260
hemisphere or something.

662
00:36:50,260 --> 00:36:56,370
It's left half of each retina
goes to left hemisphere, right

663
00:36:56,370 --> 00:36:59,860
half of each retina goes to
the right hemisphere.

664
00:36:59,860 --> 00:37:06,690
What it does once it gets there
is -- this is a monkey

665
00:37:06,690 --> 00:37:11,090
brain, and what you're
looking at is the

666
00:37:11,090 --> 00:37:15,200
cortical surface here.

667
00:37:15,200 --> 00:37:17,660
That must be the outside surface
and this is the inner

668
00:37:17,660 --> 00:37:21,050
surface of the hemisphere, if
there are two hemispheres.

669
00:37:21,050 --> 00:37:23,910
The one on the bottom is looking
at the inside surface.

670
00:37:23,910 --> 00:37:26,660
The colored areas are ones that
are important in vision.

671
00:37:26,660 --> 00:37:29,640
I talked at the beginning about
the fact that primary

672
00:37:29,640 --> 00:37:32,360
visual cortex is right here at
the back of the brain, but

673
00:37:32,360 --> 00:37:38,220
it's amazing how much of the
brain is devoted to vision.

674
00:37:38,220 --> 00:37:40,520
It's very hard to look at in a
picture -- well, it's very

675
00:37:40,520 --> 00:37:42,440
hard look at, period.

676
00:37:42,440 --> 00:37:46,210
What vision researchers like
to do is to make maps like

677
00:37:46,210 --> 00:37:48,520
this where you take the
cortical surface

678
00:37:48,520 --> 00:37:50,220
and flatten it --

679
00:37:52,770 --> 00:37:58,670
it's all this wrinkley involuted
structure --

680
00:37:58,670 --> 00:38:01,840
use a variety of techniques to
flatten it out like a map.

681
00:38:01,840 --> 00:38:04,420
Then you see that this is the
piece that's at the back of

682
00:38:04,420 --> 00:38:06,730
the head, the one primary
visual cortex.

683
00:38:06,730 --> 00:38:11,750
Then you've got all these
gazillion other regions that

684
00:38:11,750 --> 00:38:16,150
are important in visual
processing, some of them for

685
00:38:16,150 --> 00:38:17,500
rather precise sort of things.

686
00:38:17,500 --> 00:38:19,020
Where did mt go?

687
00:38:23,510 --> 00:38:28,620
I think right there is mt, for
example, which is important in

688
00:38:28,620 --> 00:38:30,620
motion processing.

689
00:38:30,620 --> 00:38:32,710
Lots of work being done trying
to figure out what different

690
00:38:32,710 --> 00:38:34,210
bits of it do.

691
00:38:34,210 --> 00:38:38,990
Look, let me give you a broad
organizing scheme for how to

692
00:38:38,990 --> 00:38:41,860
think about this.

693
00:38:41,860 --> 00:38:44,810
It's not perfect but
it ain't bad.

694
00:38:44,810 --> 00:38:48,650
From primary visual cortex where
very basic information

695
00:38:48,650 --> 00:38:53,170
about spots and lines and
primitive bits of motion gets

696
00:38:53,170 --> 00:38:59,010
pulled out, more advanced
processing can be thought of

697
00:38:59,010 --> 00:39:01,990
as in two large streams.

698
00:39:01,990 --> 00:39:07,330
One large stream going up
towards the parietal lobe is

699
00:39:07,330 --> 00:39:12,280
telling you things about where
stuff is in the world.

700
00:39:12,280 --> 00:39:14,600
How is the world laid out?

701
00:39:14,600 --> 00:39:20,890
So, my notion that this space
is laid out as a big slanted

702
00:39:20,890 --> 00:39:28,190
plane in front of me is probably
a product of this

703
00:39:28,190 --> 00:39:30,770
sort of a pathway.

704
00:39:30,770 --> 00:39:38,090
A second pathway going down into
the temporal lobe is very

705
00:39:38,090 --> 00:39:42,260
concerned with what stimuli
it might be.

706
00:39:42,260 --> 00:39:48,100
So there, if you're a monkey and
I spear individual cells,

707
00:39:48,100 --> 00:39:50,830
it might be where I would find
a cell that was particularly

708
00:39:50,830 --> 00:39:56,030
interested in a hand, a monkey
hand, or a hairy leg or

709
00:39:56,030 --> 00:39:58,570
something like that.

710
00:39:58,570 --> 00:40:01,370
It turns out that the hairy
part can be important.

711
00:40:03,960 --> 00:40:06,960
Some of the work on things like
monkey hand detectors was

712
00:40:06,960 --> 00:40:12,150
being done by one of my
professors at Princeton, and

713
00:40:12,150 --> 00:40:17,790
actually my TA in intro psych,
when I was an undergrad, was a

714
00:40:17,790 --> 00:40:20,660
grad student who was working on
this, and he reported to us

715
00:40:20,660 --> 00:40:22,480
that he was recording --

716
00:40:22,480 --> 00:40:24,860
you go spear a cell.

717
00:40:24,860 --> 00:40:26,760
It's one thing if you're just
trying to figure out if it

718
00:40:26,760 --> 00:40:30,000
likes spots of light, but how do
you figure out if the cell

719
00:40:30,000 --> 00:40:32,630
likes something like a hand.

720
00:40:32,630 --> 00:40:37,520
Well, they had in those days
drawers full of stuff and you

721
00:40:37,520 --> 00:40:40,480
waved all this stuff in front
of the monkey and see what

722
00:40:40,480 --> 00:40:43,940
makes the cell go.

723
00:40:43,940 --> 00:40:46,490
These experiments in the good
old days tended to run for

724
00:40:46,490 --> 00:40:50,120
very long time, because this was
an anesthetized monkey and

725
00:40:50,120 --> 00:40:51,620
the monkey wasn't going
to wake up from

726
00:40:51,620 --> 00:40:52,900
this particular operation.

727
00:40:52,900 --> 00:40:56,680
You ran the experiment for days
and days and days and you

728
00:40:56,680 --> 00:40:58,340
stayed up the whole time.

729
00:40:58,340 --> 00:41:03,870
So, Bob Desimone, now a famous
visual neuroscientist is up

730
00:41:03,870 --> 00:41:08,260
late at night with this stupid
monkey with his stupid cell

731
00:41:08,260 --> 00:41:10,180
and the cell isn't
doing anything.

732
00:41:10,180 --> 00:41:11,660
He's waving everything at it.

733
00:41:11,660 --> 00:41:16,940
So finally, he took off his
pants, wrapped a towel around

734
00:41:16,940 --> 00:41:20,120
his waist and did a little dance
in front of the monkey.

735
00:41:20,120 --> 00:41:21,540
The cell went nuts.

736
00:41:25,700 --> 00:41:29,240
Being a good scientist, he then
narrowed this down to the

737
00:41:29,240 --> 00:41:31,870
notion that this was a cell
that actually seem to be

738
00:41:31,870 --> 00:41:36,810
interested in hairy legs, and
the hairy part turned out to

739
00:41:36,810 --> 00:41:38,980
be important.

740
00:41:38,980 --> 00:41:43,470
I feel free to tell this story
about Bob Desimone because Bob

741
00:41:43,470 --> 00:41:47,730
Desimone is running around the
country with a talk at the

742
00:41:47,730 --> 00:41:50,030
moment that has a picture of
me in it kissing a wombat.

743
00:41:53,540 --> 00:41:55,760
Don't ask.

744
00:41:55,760 --> 00:41:58,570
It all seems fair to me.

745
00:41:58,570 --> 00:42:03,300
All right, so if you go up into,
way down here in the

746
00:42:03,300 --> 00:42:09,070
temporal lobe, you might find
cells that respond only to

747
00:42:09,070 --> 00:42:11,220
something like a hairy leg
or a monkey hand or

748
00:42:11,220 --> 00:42:12,400
something like that.

749
00:42:12,400 --> 00:42:17,010
If you're hanging out
back here in primary

750
00:42:17,010 --> 00:42:19,990
visual cortex --

751
00:42:19,990 --> 00:42:22,450
well, actually let's go
back and hang out

752
00:42:22,450 --> 00:42:24,300
in the retina even.

753
00:42:24,300 --> 00:42:26,290
Look at that, we'll hang
out right there.

754
00:42:26,290 --> 00:42:33,410
If you're hanging out in the
retina, if you like much

755
00:42:33,410 --> 00:42:38,610
simpler needs than this hairy
leg detector somewhere up in

756
00:42:38,610 --> 00:42:46,170
the temporal lobe, I mentioned
before that you might be

757
00:42:46,170 --> 00:42:52,340
bringing together a whole
bunch of photoreceptors

758
00:42:52,340 --> 00:42:59,230
through, say, a bipolar cell to
a ganglion cell, to bring

759
00:42:59,230 --> 00:43:01,480
together a whole bunch
of them to send a

760
00:43:01,480 --> 00:43:03,460
signal off to the brain.

761
00:43:03,460 --> 00:43:05,990
But what I didn't mention
is that not all the

762
00:43:05,990 --> 00:43:09,370
photoreceptors that are hooked
up to stimulated ganglion cell

763
00:43:09,370 --> 00:43:11,690
are going to excite that cell.

764
00:43:11,690 --> 00:43:15,740
Let's draw some pictures
over here.

765
00:43:15,740 --> 00:43:21,620
What actually happens, and I'm
just going to mimic this over

766
00:43:21,620 --> 00:43:30,240
here, is that you'd have some
photoreceptors who are set up

767
00:43:30,240 --> 00:43:35,910
to excite a ganglion cell.

768
00:43:35,910 --> 00:43:43,160
Then some other guys around them
who would be set up to

769
00:43:43,160 --> 00:43:45,100
inhibit it.

770
00:43:52,230 --> 00:43:56,460
The result is that
what this cell --

771
00:43:56,460 --> 00:43:58,690
that's a one-dimensional slice
and this would be a

772
00:43:58,690 --> 00:44:00,890
two-dimensional picture --
the retina's after all a

773
00:44:00,890 --> 00:44:02,280
two-dimensional surface.

774
00:44:02,280 --> 00:44:12,080
So what you'd really get is some
region of photoreceptors

775
00:44:12,080 --> 00:44:17,930
that excite the cell, and some
surrounding region where light

776
00:44:17,930 --> 00:44:21,660
inhibits the subsequent
ganglion cell.

777
00:44:21,660 --> 00:44:27,150
Then light out here
does nothing.

778
00:44:27,150 --> 00:44:30,370
So this is the receptive field
of this cell -- this is the

779
00:44:30,370 --> 00:44:33,145
only part of the retina
that this cell

780
00:44:33,145 --> 00:44:37,590
over here cares about.

781
00:44:37,590 --> 00:44:41,160
It's going to send its signal
off to the brain.

782
00:44:41,160 --> 00:44:42,830
What does this cell do?

783
00:44:42,830 --> 00:44:45,180
Well, among the things that
this cell can do is it can

784
00:44:45,180 --> 00:44:47,980
tell you I am most
excited by a spot

785
00:44:47,980 --> 00:44:52,100
that's exacty this size.

786
00:44:52,100 --> 00:44:55,310
If the spot gets any bigger it's
going to encroach on this

787
00:44:55,310 --> 00:44:58,720
inhibitory stuff and reduce
my excitation.

788
00:44:58,720 --> 00:45:04,020
If the spot's any smaller, I'm
not going to get all of my

789
00:45:04,020 --> 00:45:08,200
excitatory guys excited, and the
response will get smaller.

790
00:45:08,200 --> 00:45:12,300
So I'm most excited by a spot
of exactly this size.

791
00:45:12,300 --> 00:45:17,770
If I build receptors with
different size centers, this

792
00:45:17,770 --> 00:45:20,020
guy's only going to be excited
by a little spot and I could

793
00:45:20,020 --> 00:45:23,150
make one that would be excited
by a bigger spot.

794
00:45:23,150 --> 00:45:27,100
So, even at the level of the
retina, you can start to get

795
00:45:27,100 --> 00:45:30,880
cells that are telling you more
than just light on, light

796
00:45:30,880 --> 00:45:33,900
off, they're starting to tell
you something about the size

797
00:45:33,900 --> 00:45:39,240
of things in the visual field.

798
00:45:39,240 --> 00:45:46,570
When you get up to the cortex,
you find that now cells come

799
00:45:46,570 --> 00:45:52,450
to be interested in things like
lines -- not just a spot

800
00:45:52,450 --> 00:45:53,370
but a line.

801
00:45:53,370 --> 00:45:58,970
And this cell will way, I like
lines in this orientation.

802
00:45:58,970 --> 00:46:01,020
I don't like lines in
this orientation.

803
00:46:01,020 --> 00:46:03,370
In fact, I like lines in this
orientation only if they move

804
00:46:03,370 --> 00:46:05,140
this direction.

805
00:46:05,140 --> 00:46:08,202
So you can imagine that what you
start to do early in the

806
00:46:08,202 --> 00:46:15,410
visual system is to pull out
a whole bunch of little

807
00:46:15,410 --> 00:46:20,570
features, if you like, that you
can then use subsequently

808
00:46:20,570 --> 00:46:25,480
to try to figure out is there
a monkey hand here.

809
00:46:25,480 --> 00:46:31,160
Is there a laptop computer
or something like that.

810
00:46:31,160 --> 00:46:37,680
Early in the system you've got
basic features being pulled

811
00:46:37,680 --> 00:46:38,480
out of the image.

812
00:46:38,480 --> 00:46:41,790
Now, this presents a problem.

813
00:46:41,790 --> 00:46:45,000
Suppose that what you
want to do --

814
00:46:45,000 --> 00:46:46,650
Let's start with orientation.

815
00:46:46,650 --> 00:46:49,800
Suppose you want to decide that
a line is a particular

816
00:46:49,800 --> 00:46:50,840
orientation.

817
00:46:50,840 --> 00:46:52,790
How good are you at that?

818
00:46:52,790 --> 00:46:58,030
The answer turns out to be that
you are good enough to be

819
00:46:58,030 --> 00:47:03,610
able to tell roughly 1
degree increments.

820
00:47:03,610 --> 00:47:04,490
How are you going to do that?

821
00:47:04,490 --> 00:47:07,880
Well you could have a cell
specifically designed to look

822
00:47:07,880 --> 00:47:11,150
for zero degrees vertical, 1
degree, 2 degree, 3 degree, 4

823
00:47:11,150 --> 00:47:14,880
degree, and you need one of
those everywhere in the visual

824
00:47:14,880 --> 00:47:17,380
field, and at all different
sizes --

825
00:47:17,380 --> 00:47:19,990
a little line that's vertical,
a big line that's vertical.

826
00:47:19,990 --> 00:47:22,930
You're going to need an awful
lot of cells then.

827
00:47:22,930 --> 00:47:25,070
That's not what we do.

828
00:47:25,070 --> 00:47:28,180
What we seem to do all over
the place in the visual

829
00:47:28,180 --> 00:47:32,260
system, and in sensory systems
in general, is to make

830
00:47:32,260 --> 00:47:35,450
comparisons.

831
00:47:35,450 --> 00:47:40,320
That comparisons are very
powerful in this game.

832
00:47:40,320 --> 00:47:43,390
So how do you decide that
something's vertical?

833
00:47:43,390 --> 00:47:46,300
To a first approximation what
you do is you look at whether

834
00:47:46,300 --> 00:47:49,900
or not the lift-tilted guys and
the right-tilted guys, the

835
00:47:49,900 --> 00:47:52,940
cells that like stuff --

836
00:47:52,940 --> 00:47:54,780
the cells that look for
something that's tilted a

837
00:47:54,780 --> 00:47:57,100
little to the right, cells that
are interested in stuff

838
00:47:57,100 --> 00:47:58,060
that are tilted to the left.

839
00:47:58,060 --> 00:48:01,140
Are they roughly equally
excited?

840
00:48:01,140 --> 00:48:04,180
If they're roughly equally
excited in balance one with

841
00:48:04,180 --> 00:48:13,950
the other, then odds are
that you're looking

842
00:48:13,950 --> 00:48:16,320
at a vertical stimulus.

843
00:48:16,320 --> 00:48:18,150
OK?

844
00:48:18,150 --> 00:48:22,460
If the right-tilted stuff is
stronger than the left-tilted

845
00:48:22,460 --> 00:48:24,840
stuff, so that balance gets
tiled, you're going to say

846
00:48:24,840 --> 00:48:27,920
that the stimulus is tilted.

847
00:48:27,920 --> 00:48:36,540
It turns out that if you
do something to make --

848
00:48:36,540 --> 00:48:40,880
let's say you go off and you
make the left-tilted stuff

849
00:48:40,880 --> 00:48:44,640
weak by tiring it out -- how do
you tire it out, you show

850
00:48:44,640 --> 00:48:47,610
it a bunch of left, show it a
bunch of left lines, that

851
00:48:47,610 --> 00:48:48,570
makes this weaker.

852
00:48:48,570 --> 00:48:50,090
Then something that's vertical's
going to turn out

853
00:48:50,090 --> 00:48:52,310
to look tilted to the --

854
00:48:52,310 --> 00:48:53,350
you're not going to
believe that.

855
00:48:53,350 --> 00:48:55,110
Let me show this to you.

856
00:48:55,110 --> 00:48:56,670
I'll show it to you in --

857
00:48:56,670 --> 00:49:01,640
I'm going to jump to here.

858
00:49:01,640 --> 00:49:05,160
Well, we'll jump here first.

859
00:49:05,160 --> 00:49:05,890
OK.

860
00:49:05,890 --> 00:49:11,280
Here are a very boring
picture.

861
00:49:11,280 --> 00:49:14,310
It will be even more boring if
I turn out a little more of

862
00:49:14,310 --> 00:49:15,560
the lights here.

863
00:49:18,650 --> 00:49:20,800
OK.

864
00:49:20,800 --> 00:49:24,320
How you know that something's
white or grey or achromatic?

865
00:49:24,320 --> 00:49:28,330
Again, in this case you've got
color mechanisms that say how

866
00:49:28,330 --> 00:49:33,480
red or green is this?

867
00:49:33,480 --> 00:49:36,520
If the answer is well, it's
balanced between red and

868
00:49:36,520 --> 00:49:44,030
green, and you also ask how
blue or yellow is this?

869
00:49:44,030 --> 00:49:46,640
If it's not blue and it's not
yellow and it's not red and

870
00:49:46,640 --> 00:49:50,200
it's not green, well, you know
what it's probably white.

871
00:49:50,200 --> 00:49:54,600
OK, well, so here let's take
this red/green thing and let's

872
00:49:54,600 --> 00:49:59,620
make the green side weak
by wearing it out.

873
00:49:59,620 --> 00:50:02,370
We'll show you some green.

874
00:50:02,370 --> 00:50:04,700
Now I'm going to show you
something white -- the green

875
00:50:04,700 --> 00:50:08,700
is weak, the red is strong, the
balance goes this way, and

876
00:50:08,700 --> 00:50:10,980
you think that something
that was once white now

877
00:50:10,980 --> 00:50:14,480
looks kind of green.

878
00:50:14,480 --> 00:50:17,300
There, see, isn't
that amazing?

879
00:50:17,300 --> 00:50:19,000
They're not convinced.

880
00:50:19,000 --> 00:50:22,260
The ones who are convinced
are confused.

881
00:50:22,260 --> 00:50:26,650
Stare at the black dot
at the center here.

882
00:50:26,650 --> 00:50:29,960
You just want to stare at the
black dot and think about,

883
00:50:29,960 --> 00:50:32,700
say, that red dot up
at 12 o'clock.

884
00:50:32,700 --> 00:50:34,900
What you're doing is you're
tiring out the red side of

885
00:50:34,900 --> 00:50:38,480
this red/green equation in this
particular case, and just

886
00:50:38,480 --> 00:50:43,170
keep staring there, and then
I'm going to go boink.

887
00:50:51,030 --> 00:50:56,340
So, the people who went "oh,"
presumably saw colors.

888
00:50:56,340 --> 00:50:59,470
Let's do that again, here.

889
00:50:59,470 --> 00:51:09,080
What you want to notice is
that the color is this

890
00:51:09,080 --> 00:51:11,470
opposite color, it's going
the other direction.

891
00:51:11,470 --> 00:51:13,550
So the red one will turn green,
the yellow one will

892
00:51:13,550 --> 00:51:16,880
turn bluish, the green one
will turn reddish, right.

893
00:51:16,880 --> 00:51:18,370
So stare at the center, boink.

894
00:51:21,410 --> 00:51:23,710
It works.

895
00:51:26,960 --> 00:51:29,310
That's what's known as a
negative after image.

896
00:51:31,850 --> 00:51:33,110
You can just toggle
back and forth.

897
00:51:33,110 --> 00:51:41,070
If you look at the black dot
on the right and then move

898
00:51:41,070 --> 00:51:43,890
your eyes to the left, you'll
see the colors show up in the

899
00:51:43,890 --> 00:51:46,720
left hand circles, because
that after image is

900
00:51:46,720 --> 00:51:48,580
essentially cooked onto
your retina --

901
00:51:48,580 --> 00:51:51,090
it's there in retinal
coordinates and

902
00:51:51,090 --> 00:51:54,830
moves with your eyes.

903
00:51:54,830 --> 00:51:58,610
Alright, now let me illustrate
to you that this is a

904
00:51:58,610 --> 00:52:02,420
ubiquitous phenomenon that shows
up for basic phenomena

905
00:52:02,420 --> 00:52:08,120
all over the place by turning
this back on.

906
00:52:08,120 --> 00:52:09,480
Is that going to be enough
light for this?

907
00:52:12,600 --> 00:52:15,400
So, I won't show you the
orientation one, even though I

908
00:52:15,400 --> 00:52:19,930
managed to get a PhD studying
it, but I'll show you the

909
00:52:19,930 --> 00:52:22,250
motion equivalent of it.

910
00:52:22,250 --> 00:52:23,500
Here's what you want to do.

911
00:52:26,220 --> 00:52:28,110
You'll still see it over there,
except for the guy

912
00:52:28,110 --> 00:52:30,350
who's asleep behind the pillar,
that's his tough luck.

913
00:52:38,800 --> 00:52:42,600
So, look at this, and we're
doing the same game.

914
00:52:42,600 --> 00:52:45,280
We're just taking the game and
transferring from the color

915
00:52:45,280 --> 00:52:47,780
realm, or the orientation
realm, down

916
00:52:47,780 --> 00:52:48,590
to the motion realm.

917
00:52:48,590 --> 00:52:51,740
So this is contracting, right?

918
00:52:51,740 --> 00:52:53,450
Stare at the center,
that's important.

919
00:52:53,450 --> 00:52:54,910
So stare at the center.

920
00:52:54,910 --> 00:52:57,250
Again, think about 12 o'clock.

921
00:52:57,250 --> 00:52:59,940
All the contours are
moving down.

922
00:52:59,940 --> 00:53:02,670
You know that something's
stationary because it's not

923
00:53:02,670 --> 00:53:04,220
moving up, down,
left or right.

924
00:53:04,220 --> 00:53:07,650
But at 12 o'clock I'm wearing
out your down detectors, and

925
00:53:07,650 --> 00:53:11,620
at 6 o'clock I'm wearing out
your up detectors and so on.

926
00:53:11,620 --> 00:53:17,360
The consequences of this evil
act of mine will be seen

927
00:53:17,360 --> 00:53:20,160
vividly if, when I take
it away, what you do

928
00:53:20,160 --> 00:53:22,460
is look at my nose.

929
00:53:41,350 --> 00:53:47,020
Of course, what really worries
me here is several people's

930
00:53:47,020 --> 00:53:48,930
pupils got much larger, which
means they think I

931
00:53:48,930 --> 00:53:51,340
look better that way.

932
00:53:53,890 --> 00:53:57,730
Now, a word of warning.

933
00:53:57,730 --> 00:53:58,480
Two words of warning.

934
00:53:58,480 --> 00:54:00,160
This is the best demo
in the whole class.

935
00:54:00,160 --> 00:54:02,610
It's all downhill from here.

936
00:54:02,610 --> 00:54:06,450
The other word of warning is
it's a very salient demo, but

937
00:54:06,450 --> 00:54:10,250
it's there for a purpose.

938
00:54:10,250 --> 00:54:14,290
It's really lame on the mid-term
or the final when

939
00:54:14,290 --> 00:54:16,800
some motion -- this is called a
motion after effect -- when

940
00:54:16,800 --> 00:54:20,550
a motion after effect question
shows up to say oh, that was

941
00:54:20,550 --> 00:54:24,190
cool, man, his head was
shrinking and stuff.

942
00:54:24,190 --> 00:54:26,740
Because you don't get much
points for that.

943
00:54:26,740 --> 00:54:30,810
What you want to know is why
this is actually interesting,

944
00:54:30,810 --> 00:54:32,930
which I will reiterate by going
the other direction

945
00:54:32,930 --> 00:54:36,290
since it seems a pity
not to, right?

946
00:54:36,290 --> 00:54:40,040
OK, so again, the way you figure
out that something is,

947
00:54:40,040 --> 00:54:43,850
in this case, stationary is by
making a set of comparisons,

948
00:54:43,850 --> 00:54:48,330
and Iam using this clever little
tool to systematically

949
00:54:48,330 --> 00:54:51,500
distort the comparisons
that you're making.

950
00:54:51,500 --> 00:54:53,710
By the way, I'm also causing
you to make one of these

951
00:54:53,710 --> 00:54:57,860
inferences that we'll talk
about next week.

952
00:54:57,860 --> 00:55:00,480
By seeing all these various bits
of odd motion, you are

953
00:55:00,480 --> 00:55:03,220
making the conclusion that my
head is shrinking or growing

954
00:55:03,220 --> 00:55:04,610
or something like that.

955
00:55:04,610 --> 00:55:06,640
So this is expanding, right,
so this is the

956
00:55:06,640 --> 00:55:08,310
amazing pinhead version.

957
00:55:08,310 --> 00:55:09,560
Ready?

958
00:55:19,540 --> 00:55:25,260
The original version of this,
by the way, is known in the

959
00:55:25,260 --> 00:55:29,300
literature as sometimes it's
the waterfall illusion,

960
00:55:29,300 --> 00:55:32,810
because it gets described
multiple times in the

961
00:55:32,810 --> 00:55:39,190
literature, but one of them is
by a Scotsman named Adams who

962
00:55:39,190 --> 00:55:43,300
was staring at a waterfall and
then noticed that the rocks on

963
00:55:43,300 --> 00:55:46,610
the side of the waterfall
seem to be drifting up.

964
00:55:46,610 --> 00:55:50,230
This is a known waterfall in a
known spot in Scotland and

965
00:55:50,230 --> 00:55:52,880
there's a group of vision
researchers who meet there

966
00:55:52,880 --> 00:55:57,370
every year and the entry
requirement for joining this

967
00:55:57,370 --> 00:56:01,760
particular elite group is a
bottle of scotch of a variety

968
00:56:01,760 --> 00:56:04,570
that they have not tried before,
which I think helps

969
00:56:04,570 --> 00:56:08,760
make the rocks move in all sorts
of interesting ways.

970
00:56:08,760 --> 00:56:13,320
They've never invited me
so how would I know.

971
00:56:13,320 --> 00:56:15,410
OK, this is what I'm
going to do.

972
00:56:15,410 --> 00:56:17,700
Let me follow up.

973
00:56:17,700 --> 00:56:21,620
So you've seen a negative
color after image.

974
00:56:21,620 --> 00:56:25,780
You've seen a negative
motion after effect.

975
00:56:25,780 --> 00:56:29,020
I've asserted that these
exist in orientation.

976
00:56:29,020 --> 00:56:32,510
You can show this often -- you
show somebody big stuff and

977
00:56:32,510 --> 00:56:33,920
then medium stuff looks small.

978
00:56:33,920 --> 00:56:34,870
It's ubiquitous.

979
00:56:34,870 --> 00:56:36,450
It shows up all over
the place.

980
00:56:36,450 --> 00:56:39,000
I want to show you one more
version of this kind of a

981
00:56:39,000 --> 00:56:42,830
effect, and that's going to
take a certain amount of

982
00:56:42,830 --> 00:56:45,210
build-up here.

983
00:56:45,210 --> 00:56:48,320
So the first thing to do
is say the vertical and

984
00:56:48,320 --> 00:56:52,650
horizontal stripes, do they
look different to you?

985
00:56:52,650 --> 00:56:54,300
If they look different somebody
raise their hand and

986
00:56:54,300 --> 00:56:55,110
tell me what the
difference is.

987
00:56:55,110 --> 00:56:56,290
What's the difference?

988
00:56:56,290 --> 00:56:57,820
AUDIENCE: The horizontal
ones are horizontal.

989
00:56:57,820 --> 00:57:02,290
PROFESSOR: Yeah, alright.

990
00:57:02,290 --> 00:57:05,150
In their color or shading
in some fashion.

991
00:57:05,150 --> 00:57:05,490
Yeah?

992
00:57:05,490 --> 00:57:07,310
AUDIENCE: The vertical
ones look blurry.

993
00:57:07,310 --> 00:57:09,280
PROFESSOR: The vertical ones
look a little -- they do look

994
00:57:09,280 --> 00:57:12,330
a little blurry.

995
00:57:12,330 --> 00:57:12,660
Yeah?

996
00:57:12,660 --> 00:57:15,265
AUDIENCE: The vertical ones --
the black lines look like

997
00:57:15,265 --> 00:57:16,450
they're thicker than
the ones--.

998
00:57:16,450 --> 00:57:19,030
PROFESSOR: Well, we're not
getting a lot of vivid color

999
00:57:19,030 --> 00:57:19,830
reports here.

1000
00:57:19,830 --> 00:57:22,570
AUDIENCE: The horizontal
lines look brighter.

1001
00:57:22,570 --> 00:57:23,970
PROFESSOR: Brighter?

1002
00:57:23,970 --> 00:57:26,890
Nobody knows what the
word color means.

1003
00:57:26,890 --> 00:57:27,630
Last try.

1004
00:57:27,630 --> 00:57:33,237
AUDIENCE: The horizontal lines
are greyer and the vertical

1005
00:57:33,237 --> 00:57:34,110
lines are black.

1006
00:57:34,110 --> 00:57:38,040
PROFESSOR: OK, look, this makes
the point just fine.

1007
00:57:38,040 --> 00:57:40,190
Never mind guys.

1008
00:57:40,190 --> 00:57:42,840
There is no in general
consensus here.

1009
00:57:42,840 --> 00:57:45,830
I'm about to produce a
general consensus.

1010
00:57:49,160 --> 00:57:51,520
That's the motivational part
of today's lecture.

1011
00:57:51,520 --> 00:57:57,100
No, actually what I have to
do now is for the next few

1012
00:57:57,100 --> 00:57:59,570
minutes what you want to do
is keep your head upright.

1013
00:57:59,570 --> 00:58:02,200
You don't have to stare
at any one spot.

1014
00:58:02,200 --> 00:58:03,830
You don't even have to your head
upright, just keep it in

1015
00:58:03,830 --> 00:58:05,460
the same orientation.

1016
00:58:05,460 --> 00:58:08,780
If you flop all over the
place and fall asleep

1017
00:58:08,780 --> 00:58:10,250
nothing good happens.

1018
00:58:10,250 --> 00:58:12,840
So you're going to look at these
green vertical lines and

1019
00:58:12,840 --> 00:58:14,630
these red horizontal lines.

1020
00:58:20,250 --> 00:58:23,980
All I need to do here
is toggle back and

1021
00:58:23,980 --> 00:58:25,550
forth between them.

1022
00:58:25,550 --> 00:58:28,470
Now you can see the negative
after image here.

1023
00:58:28,470 --> 00:58:31,310
If you stare at this green --
oh, I should have done that

1024
00:58:31,310 --> 00:58:32,010
quickly before.

1025
00:58:32,010 --> 00:58:34,810
So stare at this green
thing for a while.

1026
00:58:34,810 --> 00:58:38,850
You get an nice purply
red thing?

1027
00:58:38,850 --> 00:58:39,750
OK.

1028
00:58:39,750 --> 00:58:42,490
So that's just the negative
after image.

1029
00:58:42,490 --> 00:58:45,400
But we're not interested in the

1030
00:58:45,400 --> 00:58:48,670
negative after image anymore.

1031
00:58:48,670 --> 00:58:50,290
Actually, we are.

1032
00:58:50,290 --> 00:58:53,870
There's an aspect of the
negative after image that's

1033
00:58:53,870 --> 00:59:01,310
really cool here, which you may
start to notice shortly.

1034
00:59:01,310 --> 00:59:04,800
During the blank period,
shout out the

1035
00:59:04,800 --> 00:59:06,730
orientation that you see.

1036
00:59:06,730 --> 00:59:07,980
AUDIENCE: [INAUDIBLE PHRASE]

1037
00:59:14,680 --> 00:59:18,540
PROFESSOR: That's
a little weird.

1038
00:59:18,540 --> 00:59:20,890
Wouldn't you think?

1039
00:59:20,890 --> 00:59:26,060
Now that's a little weird
because if you stare at these

1040
00:59:26,060 --> 00:59:28,260
horizontal stripes -- in fact,
I should be able to get you

1041
00:59:28,260 --> 00:59:29,050
back to seeing horizontal.

1042
00:59:29,050 --> 00:59:30,310
So, stare at these --

1043
00:59:30,310 --> 00:59:34,080
fixate, don't move your eyes
around and we'll just stare at

1044
00:59:34,080 --> 00:59:37,450
that for a minute or so.

1045
00:59:40,230 --> 00:59:43,530
Has it been long enough?

1046
00:59:43,530 --> 00:59:44,810
Yeah.

1047
00:59:44,810 --> 00:59:46,950
Does it look horizontal?

1048
00:59:46,950 --> 00:59:49,230
Still looks vertical, oh boy.

1049
00:59:49,230 --> 00:59:58,430
Well, here's the get rich quick
part of this lecture,

1050
00:59:58,430 --> 01:00:02,490
assuming that you're
unscrupulous, you can use this

1051
01:00:02,490 --> 01:00:04,990
to make your fortune claiming
that you can teach people to

1052
01:00:04,990 --> 01:00:07,180
be psychic, right.

1053
01:00:07,180 --> 01:00:10,800
What's the next orientation
I'm going to show?

1054
01:00:10,800 --> 01:00:12,700
What's the next orientation
I'm going to show?

1055
01:00:16,560 --> 01:00:21,640
If you can make your fortune out
of that, just send me 10%.

1056
01:00:21,640 --> 01:00:27,570
But what's really going on here,
what this is actually

1057
01:00:27,570 --> 01:00:31,170
illustrating, this aspect of
it is illustrating, is that

1058
01:00:31,170 --> 01:00:34,600
you're seeing with your whole
visual system, not with your

1059
01:00:34,600 --> 01:00:37,440
retina all by itself or
something like that.

1060
01:00:37,440 --> 01:00:41,400
So, each time you look, say,
the red horizontal, you're

1061
01:00:41,400 --> 01:00:44,530
building up a green horizontal
negative after image on your

1062
01:00:44,530 --> 01:00:46,760
retina, right?

1063
01:00:46,760 --> 01:00:49,560
And this is building up a red
vertical kind of after image

1064
01:00:49,560 --> 01:00:50,570
on your retina.

1065
01:00:50,570 --> 01:00:58,640
If you go back and forth, you
build up sort of a plaid.

1066
01:00:58,640 --> 01:01:02,720
So think about your retina
as having a plaid on it.

1067
01:01:02,720 --> 01:01:07,090
Now you don't see it here,
but the plaid is --

1068
01:01:07,090 --> 01:01:09,160
when you're looking at a blank
screen, it's just that plaid

1069
01:01:09,160 --> 01:01:12,290
that's getting shipped up to
your brain, basically.

1070
01:01:12,290 --> 01:01:14,290
OK, well what else
is happening?

1071
01:01:14,290 --> 01:01:17,470
Well, when you're staring at
this green vertical thing, up

1072
01:01:17,470 --> 01:01:20,700
in your brain the cells that are
interested in vertical are

1073
01:01:20,700 --> 01:01:22,460
saying we're getting tired.

1074
01:01:22,460 --> 01:01:26,900
Go to the blank part, tired
vertical guys and more awake

1075
01:01:26,900 --> 01:01:29,290
horizontal guys are looking at
this weak plaid and they're

1076
01:01:29,290 --> 01:01:30,880
only seeing the horizontal
part.

1077
01:01:30,880 --> 01:01:32,860
Now you look at the horizontal,
the horizontal

1078
01:01:32,860 --> 01:01:34,790
guys get tired out -- oh,
we can only see the

1079
01:01:34,790 --> 01:01:36,480
vertical, and so on.

1080
01:01:39,410 --> 01:01:42,720
You may ask isn't it about
time we saw something

1081
01:01:42,720 --> 01:01:45,040
interesting here?

1082
01:01:45,040 --> 01:01:48,910
The answer is undoubtedly so,
but this particular effect

1083
01:01:48,910 --> 01:01:55,900
takes a few minutes to build
up, and so you gotta keep

1084
01:01:55,900 --> 01:01:59,710
looking at this.

1085
01:01:59,710 --> 01:02:06,250
Oh, you've got to ask yourself
what the effects ought to be.

1086
01:02:06,250 --> 01:02:08,460
What I'm building up here
is an effect called the

1087
01:02:08,460 --> 01:02:13,150
McCullough Effect, named after
Celeste McCullough, who unlike

1088
01:02:13,150 --> 01:02:18,940
most people who discover visual
effects, was actually

1089
01:02:18,940 --> 01:02:21,440
looking for this.

1090
01:02:21,440 --> 01:02:23,530
She knew what she was looking
for and she knew what she

1091
01:02:23,530 --> 01:02:24,120
could find.

1092
01:02:24,120 --> 01:02:26,380
Now I already showed you that
the pattern you're going to

1093
01:02:26,380 --> 01:02:28,470
look at again, right, it was
that black and white pattern

1094
01:02:28,470 --> 01:02:31,530
of vertical and horizontal
lines.

1095
01:02:31,530 --> 01:02:34,660
What do you think the vertical
lines should look like when

1096
01:02:34,660 --> 01:02:37,090
we're done?

1097
01:02:37,090 --> 01:02:38,930
They should look kind
of reddish maybe.

1098
01:02:38,930 --> 01:02:40,310
The horizontal should look?

1099
01:02:40,310 --> 01:02:41,460
AUDIENCE: Green.

1100
01:02:41,460 --> 01:02:41,800
PROFESSOR: Green.

1101
01:02:41,800 --> 01:02:43,730
Let's see, well we might have
done this long enough.

1102
01:02:43,730 --> 01:02:46,350
This is one of these things
where if I do it long enough I

1103
01:02:46,350 --> 01:02:51,300
can hit absolutely everybody,
and if I quit too soon not

1104
01:02:51,300 --> 01:02:52,640
everybody will see the effect.

1105
01:02:58,160 --> 01:02:59,260
All right, here we go.

1106
01:02:59,260 --> 01:03:02,600
So, close your eyes, because
that will let me get to the

1107
01:03:02,600 --> 01:03:05,330
right slide, but it will also
let you perhaps see that

1108
01:03:05,330 --> 01:03:08,010
checkerboard floating
there in the dark?

1109
01:03:08,010 --> 01:03:09,430
Murmur if you can see
a checkerboard

1110
01:03:09,430 --> 01:03:11,330
floating in the dark.

1111
01:03:11,330 --> 01:03:11,850
OK, good.

1112
01:03:11,850 --> 01:03:13,560
That's that's the negative
after image thing.

1113
01:03:13,560 --> 01:03:16,450
OK, now take a look at this
test pattern again.

1114
01:03:20,700 --> 01:03:22,820
How many people think that the
vertical and horizontal look

1115
01:03:22,820 --> 01:03:27,310
different now in terms
of their color?

1116
01:03:27,310 --> 01:03:30,350
How many people think it might
look a little sort of maybe

1117
01:03:30,350 --> 01:03:32,240
kind of different, but
it's not exactly

1118
01:03:32,240 --> 01:03:35,160
biting me on the posterier?

1119
01:03:35,160 --> 01:03:37,070
Alright, that's a better test.

1120
01:03:37,070 --> 01:03:39,510
Now, the way to find out whether
you've actually got

1121
01:03:39,510 --> 01:03:43,770
the right effect is without
knocking out your neighbor,

1122
01:03:43,770 --> 01:03:45,540
tilt your head 90 degrees.

1123
01:03:50,990 --> 01:03:52,240
So that's cooler.

1124
01:04:10,080 --> 01:04:13,000
I'll say a word more about this,
but I won't say a word

1125
01:04:13,000 --> 01:04:17,110
more about this until you've
had a chance to get up and

1126
01:04:17,110 --> 01:04:18,420
stretch for a second.

1127
01:05:11,320 --> 01:05:16,160
OK.

1128
01:05:28,050 --> 01:05:37,060
How many people when they look
at these dots still see

1129
01:05:37,060 --> 01:05:38,900
reasonably compelling colors?

1130
01:05:42,330 --> 01:05:43,240
Not many.

1131
01:05:43,240 --> 01:05:46,430
How many people when they
look at my head

1132
01:05:46,430 --> 01:05:47,930
still see it shrinking?

1133
01:05:50,690 --> 01:05:52,470
Yeah, except for the
weird people.

1134
01:05:55,180 --> 01:06:03,250
That's because your visual
system is continuously

1135
01:06:03,250 --> 01:06:05,490
re-generating itself.

1136
01:06:05,490 --> 01:06:10,850
You go stare at the red thing,
you wear out, in a sense, the

1137
01:06:10,850 --> 01:06:14,020
red detecting chunk of your
visual system, and then if you

1138
01:06:14,020 --> 01:06:16,010
flip your eyes over you find
out that you see a

1139
01:06:16,010 --> 01:06:17,860
white thing is green.

1140
01:06:17,860 --> 01:06:19,850
But then you recover
from it, right?

1141
01:06:19,850 --> 01:06:21,350
Good.

1142
01:06:21,350 --> 01:06:22,710
How many people still
see this effect?

1143
01:06:26,240 --> 01:06:29,680
Tilt your head just to make sure
if you think it's there.

1144
01:06:29,680 --> 01:06:31,990
A little weak today.

1145
01:06:31,990 --> 01:06:34,620
We probably should have gone
a little bit longer.

1146
01:06:34,620 --> 01:06:36,810
Those of you who are seeing it,
and perhaps even those of

1147
01:06:36,810 --> 01:06:37,530
you who are not --

1148
01:06:37,530 --> 01:06:41,940
did I put a test pattern on
the handout, I forget?

1149
01:06:41,940 --> 01:06:42,430
No, OK.

1150
01:06:42,430 --> 01:06:45,470
I'll just bring the Powerpoint
back next week.

1151
01:06:45,470 --> 01:06:49,620
Not only are you still seeing
this some minutes afterwards,

1152
01:06:49,620 --> 01:06:51,720
but particularly if I don't show
you the test pattern a

1153
01:06:51,720 --> 01:06:57,640
lot you may see it for
days and weeks.

1154
01:06:57,640 --> 01:07:00,210
The longest report in the
literature for the McCullough

1155
01:07:00,210 --> 01:07:04,620
effect is three months.

1156
01:07:04,620 --> 01:07:09,780
The only reason it's only three
months is because at

1157
01:07:09,780 --> 01:07:12,440
that point the undergraduates,
who, of course, were the

1158
01:07:12,440 --> 01:07:15,930
subjects, went away
on vacation and

1159
01:07:15,930 --> 01:07:17,100
the experiment ended.

1160
01:07:17,100 --> 01:07:21,410
The evidence suggests that if we
locked you in the dark that

1161
01:07:21,410 --> 01:07:24,580
the McCullough effect
would last forever.

1162
01:07:24,580 --> 01:07:26,310
We won't lock you in the dark.

1163
01:07:30,370 --> 01:07:32,410
I teach this up at Harvard
where there are many more

1164
01:07:32,410 --> 01:07:35,930
pre-law students, and the
immediate reaction to this

1165
01:07:35,930 --> 01:07:38,480
news is half of them consider
filing suit.

1166
01:07:41,260 --> 01:07:43,910
This will not, in fact, ruin
your life in any fashion.

1167
01:07:43,910 --> 01:07:46,980
In fact, it'll only be
noticeable if you happen to

1168
01:07:46,980 --> 01:07:49,370
end up looking at a test pattern
like this or perhaps

1169
01:07:49,370 --> 01:07:55,030
serving time in the
near future.

1170
01:07:55,030 --> 01:07:58,820
But this was something that
McCullough was not expecting.

1171
01:07:58,820 --> 01:08:03,590
McCullough thought that what she
had done here was, OK, you

1172
01:08:03,590 --> 01:08:06,750
can fatigue color, you can
fatigue orientation, well I'll

1173
01:08:06,750 --> 01:08:08,990
fatigue the cells in the brain
and there are cells in the

1174
01:08:08,990 --> 01:08:12,760
brain that are interested
in red vertical lines.

1175
01:08:12,760 --> 01:08:15,850
They'll get tired, I'll see
white vertical as green,

1176
01:08:15,850 --> 01:08:18,070
they'll recover, I'll see
white vertical as white.

1177
01:08:18,070 --> 01:08:21,330
But what happened was you looked
at red vertical lines

1178
01:08:21,330 --> 01:08:23,950
for a while, white vertical
looked green and white

1179
01:08:23,950 --> 01:08:27,110
vertical looked green for like
a really, really long time

1180
01:08:27,110 --> 01:08:29,080
thereafter.

1181
01:08:29,080 --> 01:08:34,540
What seems to be going on here
is something like this.

1182
01:08:38,590 --> 01:09:03,160
Let's suppose that the way you
decide that something is white

1183
01:09:03,160 --> 01:09:10,320
vertical is by looking at the
balance of -- well here, let's

1184
01:09:10,320 --> 01:09:11,630
draw it this way.

1185
01:09:15,430 --> 01:09:19,450
You've got red vertical input
and you've got green vertical

1186
01:09:19,450 --> 01:09:28,150
input, and they're being like
let's put a little dial here.

1187
01:09:28,150 --> 01:09:31,190
If they're roughly equal you
decide something looks white.

1188
01:09:31,190 --> 01:09:34,330
So I show you red vertical for
a while and this thing gets

1189
01:09:34,330 --> 01:09:37,380
weak and cruddy, and now
something that should be

1190
01:09:37,380 --> 01:09:40,600
producing equal input gets more
from there and it ends up

1191
01:09:40,600 --> 01:09:42,670
looking a little green.

1192
01:09:42,670 --> 01:09:45,270
Alright, that's fine.

1193
01:09:45,270 --> 01:09:53,120
What happens if this cell gets
sick -- it's not going to be

1194
01:09:53,120 --> 01:09:57,120
one cell, but suppose that this
process gets weak in some

1195
01:09:57,120 --> 01:10:02,400
fashion, not because you were
looking at some stimulus, but

1196
01:10:02,400 --> 01:10:09,380
because you went out and ate
the mushroom that you

1197
01:10:09,380 --> 01:10:12,940
shouldn't have eaten or whatever
and this cell just

1198
01:10:12,940 --> 01:10:13,700
took a hit.

1199
01:10:13,700 --> 01:10:16,350
And so now this is kind
of a cruddy cell.

1200
01:10:19,390 --> 01:10:21,780
Well, there are a couple
of possibilities.

1201
01:10:21,780 --> 01:10:26,620
One of them is that for the rest
of time you're going to

1202
01:10:26,620 --> 01:10:31,200
end up seeing white things
as looking green.

1203
01:10:31,200 --> 01:10:33,790
That's not going to
be good right.

1204
01:10:33,790 --> 01:10:38,480
You don't want that happen.

1205
01:10:38,480 --> 01:10:45,820
So if you decided that this was
the situation, what you

1206
01:10:45,820 --> 01:10:50,930
would want to do is change
the gain on here.

1207
01:10:50,930 --> 01:10:54,990
So if this was previously just
equal gain, let's crank this

1208
01:10:54,990 --> 01:10:58,960
up a little bit so that
we can drag that

1209
01:10:58,960 --> 01:11:03,940
back up to being equal.

1210
01:11:03,940 --> 01:11:05,180
That sounds OK.

1211
01:11:05,180 --> 01:11:06,910
That seems like a clever
enough thing to do.

1212
01:11:06,910 --> 01:11:09,320
But how are you gonna do that?

1213
01:11:09,320 --> 01:11:16,460
How are you going to figure
out that you're broken.

1214
01:11:16,460 --> 01:11:20,020
How would you know that your
visual system was not

1215
01:11:20,020 --> 01:11:22,570
reporting the truth to you.

1216
01:11:22,570 --> 01:11:25,460
Well, I mean one way would be
you're saying hey, that looks

1217
01:11:25,460 --> 01:11:27,780
white to me, and somebody else
says that looks green and they

1218
01:11:27,780 --> 01:11:29,490
laugh at you or something.

1219
01:11:29,490 --> 01:11:31,280
Nah, it doesn't work.

1220
01:11:31,280 --> 01:11:33,500
No evidence that that works.

1221
01:11:33,500 --> 01:11:41,180
What you do is that you seem to
have some sort of notion of

1222
01:11:41,180 --> 01:11:46,570
the statistics of the world, the
way the world ought to be.

1223
01:11:46,570 --> 01:11:52,420
If the world is systematically
different from that, you draw

1224
01:11:52,420 --> 01:11:58,650
the conclusion that you've
got trouble here of

1225
01:11:58,650 --> 01:11:59,980
some variety or other.

1226
01:11:59,980 --> 01:12:02,270
Well not necessary, you don't
necessarily draw the -- let me

1227
01:12:02,270 --> 01:12:04,120
give you a non-troubled
version of this.

1228
01:12:04,120 --> 01:12:06,380
How many people have been
to one of these omni

1229
01:12:06,380 --> 01:12:09,810
theater kind of movies?

1230
01:12:09,810 --> 01:12:14,150
The main reason for going to an
omni theater movie is what?

1231
01:12:14,150 --> 01:12:17,740
You want to feel like you're
moving, right,

1232
01:12:17,740 --> 01:12:19,360
when you're not moving.

1233
01:12:19,360 --> 01:12:21,990
Because all the movies that
they ever show you at the

1234
01:12:21,990 --> 01:12:26,330
science museum are things where
you go driving around in

1235
01:12:26,330 --> 01:12:31,100
race cars, stuff like that.

1236
01:12:31,100 --> 01:12:33,300
Why does this work?

1237
01:12:33,300 --> 01:12:42,100
You're sitting at this huge
screen, and everything in your

1238
01:12:42,100 --> 01:12:45,640
visual field is moving
in the same way.

1239
01:12:45,640 --> 01:12:47,850
You know something
about the world.

1240
01:12:47,850 --> 01:12:52,030
The world on average
is stationary.

1241
01:12:52,030 --> 01:12:57,300
If the whole world is moving
there's a good explanation for

1242
01:12:57,300 --> 01:12:58,840
that typically.

1243
01:12:58,840 --> 01:13:02,430
The answer is not the
world, it's you.

1244
01:13:02,430 --> 01:13:05,280
What is it that causes the whole
world to slide across

1245
01:13:05,280 --> 01:13:05,810
your retina?

1246
01:13:05,810 --> 01:13:07,540
This is what causes
the whole world to

1247
01:13:07,540 --> 01:13:09,040
slide across the retina.

1248
01:13:09,040 --> 01:13:13,000
Sit in the omni and the world is
sliding across your retina.

1249
01:13:13,000 --> 01:13:15,470
Your brain says uh-huh, the
world isn't moving, I'm

1250
01:13:15,470 --> 01:13:18,310
moving, and I think I'll throw
up now for a while or

1251
01:13:18,310 --> 01:13:19,960
something like that.

1252
01:13:19,960 --> 01:13:23,990
That's a separate interesting
question.

1253
01:13:23,990 --> 01:13:26,820
What is it here that
could tell you that

1254
01:13:26,820 --> 01:13:27,860
you're out of whack?

1255
01:13:27,860 --> 01:13:31,260
Well, what is the normal
relationship between color and

1256
01:13:31,260 --> 01:13:32,710
orientation in the world?

1257
01:13:36,890 --> 01:13:40,760
Typically in the world if you
see something that's vertical,

1258
01:13:40,760 --> 01:13:43,090
what do you know about
its color?

1259
01:13:43,090 --> 01:13:45,610
That's right, nothing.

1260
01:13:45,610 --> 01:13:49,500
So what's going on here?

1261
01:13:49,500 --> 01:13:50,740
It's vertical.

1262
01:13:50,740 --> 01:13:51,730
It's green.

1263
01:13:51,730 --> 01:13:54,200
If it's horizontal it's red.

1264
01:13:54,200 --> 01:13:56,400
If it's vertical, it's green.

1265
01:13:56,400 --> 01:13:58,480
If it's horizontal, it's red.

1266
01:13:58,480 --> 01:14:00,680
You do this for long enough
and a little chunk of your

1267
01:14:00,680 --> 01:14:02,720
brain, rather like the little
chunk of your brain that's

1268
01:14:02,720 --> 01:14:06,630
saying I ate the cupcakes,
I threw up, I'd better do

1269
01:14:06,630 --> 01:14:08,400
something about this.

1270
01:14:08,400 --> 01:14:11,510
This little chunk of your brain
is saying, oh, there's a

1271
01:14:11,510 --> 01:14:14,350
correlation between color
and orientation.

1272
01:14:14,350 --> 01:14:21,500
The only way that happens is
if I'm broken, and the only

1273
01:14:21,500 --> 01:14:25,860
thing to do about that is to
adjust the gain on chunks of

1274
01:14:25,860 --> 01:14:27,370
my visual system.

1275
01:14:27,370 --> 01:14:33,390
So, everything looks green.

1276
01:14:33,390 --> 01:14:34,930
All the vertical things
look green.

1277
01:14:34,930 --> 01:14:36,690
Oh my goodness, what
am I going to do?

1278
01:14:36,690 --> 01:14:40,290
Well I better crank down the
gain on the green side, it's

1279
01:14:40,290 --> 01:14:43,840
obviously too high, and I'm want
to crank up the gain on

1280
01:14:43,840 --> 01:14:45,440
the red side.

1281
01:14:45,440 --> 01:14:49,130
Unfortunately, that
was a goof.

1282
01:14:49,130 --> 01:14:52,740
And it was wrong, so then when
you look at a white thing it

1283
01:14:52,740 --> 01:14:54,780
now looks red because you
cranked the gain up

1284
01:14:54,780 --> 01:14:56,820
too much over here.

1285
01:14:56,820 --> 01:14:58,300
How do you recover from this?

1286
01:14:58,300 --> 01:15:03,680
Well, if I now show you the
opposite one -- what were we

1287
01:15:03,680 --> 01:15:04,350
doing before?

1288
01:15:04,350 --> 01:15:05,350
What was it, green vertical?

1289
01:15:05,350 --> 01:15:07,695
If I show you red verticals and
green horizontals for a

1290
01:15:07,695 --> 01:15:09,710
while, that'll push it
back the other way.

1291
01:15:09,710 --> 01:15:12,140
But that doesn't typically
happen in the world.

1292
01:15:12,140 --> 01:15:14,450
Simply being exposed to a
world where there's no

1293
01:15:14,450 --> 01:15:17,880
correlation will slowly
push you back.

1294
01:15:17,880 --> 01:15:22,590
But the reason the effect lasts
so long is that the

1295
01:15:22,590 --> 01:15:25,600
evidence from the real world
is kind a weak compared to

1296
01:15:25,600 --> 01:15:30,750
this artificial situation
that I cooked up.

1297
01:15:30,750 --> 01:15:32,300
The way I'm thinking about
this is like --

1298
01:15:32,300 --> 01:15:33,920
here's another way of
thinking about this.

1299
01:15:33,920 --> 01:15:45,210
Suppose you're measuring average
weight of people.

1300
01:15:45,210 --> 01:15:51,600
I'm just going to keep weighing
people and what's the

1301
01:15:51,600 --> 01:15:52,770
average weight of people.

1302
01:15:52,770 --> 01:15:54,440
It'll bounce around
a little bit.

1303
01:15:54,440 --> 01:15:55,010
Alright.

1304
01:15:55,010 --> 01:15:58,220
This is how much people
on average weigh.

1305
01:15:58,220 --> 01:16:02,790
Now what happens is the amazing
1,000 pound guy comes

1306
01:16:02,790 --> 01:16:04,500
into my experiment.

1307
01:16:04,500 --> 01:16:06,160
What happens to the average?

1308
01:16:09,200 --> 01:16:12,770
Now I start weighing normal
people again.

1309
01:16:12,770 --> 01:16:16,120
Does it just go -- no.

1310
01:16:16,120 --> 01:16:19,440
Weighing regular people it'll
take a comparatively long time

1311
01:16:19,440 --> 01:16:21,180
for me to get back to normal.

1312
01:16:21,180 --> 01:16:23,020
So that's what's
going on here.

1313
01:16:23,020 --> 01:16:26,220
The McCullough effect adapting
stimulus is the 1,000 pound

1314
01:16:26,220 --> 01:16:27,980
guy that's causing
you to think that

1315
01:16:27,980 --> 01:16:29,460
your brain is broken.

1316
01:16:29,460 --> 01:16:34,860
You go and re-adjust your brain
and now slowly it will

1317
01:16:34,860 --> 01:16:37,380
get better.

1318
01:16:37,380 --> 01:16:40,280
How's it look there?

1319
01:16:40,280 --> 01:16:41,020
Not better yet.

1320
01:16:41,020 --> 01:16:43,810
How many people have a
nice after effect?

1321
01:16:43,810 --> 01:16:46,410
Still, nice McCullough effect?

1322
01:16:46,410 --> 01:16:50,390
Celeste McCullough would
be so proud.

1323
01:16:50,390 --> 01:16:54,280
You will notice that I have not
said a word about signal

1324
01:16:54,280 --> 01:16:55,140
detection theory.

1325
01:16:55,140 --> 01:16:58,160
I noticed that, too.

1326
01:16:58,160 --> 01:16:59,850
There is a chunk in
[? Gleichman ?] about it.

1327
01:16:59,850 --> 01:17:02,800
I might decide to talk
about it on Thursday.

1328
01:17:02,800 --> 01:17:05,500
Oh, you may also notice that you
have lost your cell phone.

1329
01:17:05,500 --> 01:17:07,550
If that is true come and
talk to me about it

1330
01:17:07,550 --> 01:17:09,520
because we found it.

1331
01:17:09,520 --> 01:17:11,480
One last thing.

1332
01:17:11,480 --> 01:17:13,020
How many people know
what that is?

1333
01:17:16,540 --> 01:17:18,590
I didn't ask you what it was,
I just wanted to know if you

1334
01:17:18,590 --> 01:17:19,670
knew what it was.

1335
01:17:19,670 --> 01:17:23,210
How many people recognize
this?

1336
01:17:23,210 --> 01:17:24,390
Only a few.

1337
01:17:24,390 --> 01:17:26,290
OK, now you can shout out.

1338
01:17:26,290 --> 01:17:27,270
AUDIENCE: It's a dog.

1339
01:17:27,270 --> 01:17:28,070
PROFESSOR: It's a dog.

1340
01:17:28,070 --> 01:17:29,330
How many people does
that help?

1341
01:17:35,490 --> 01:17:38,750
It's head is -- well,
wait a second.

1342
01:17:38,750 --> 01:17:44,820
There's its head, foot, back
paw, it's back, it's sort of

1343
01:17:44,820 --> 01:17:47,960
sniffing at a ground plane
that's like here.

1344
01:17:55,470 --> 01:17:58,030
You can store that one away
till next week and

1345
01:17:58,030 --> 01:17:59,980
we'll talk about it.