WEBVTT

00:00:00.080 --> 00:00:01.670
The following
content is provided

00:00:01.670 --> 00:00:03.820
under a Creative
Commons license.

00:00:03.820 --> 00:00:06.550
Your support will help MIT
OpenCourseWare continue

00:00:06.550 --> 00:00:10.160
to offer high quality
educational resources for free.

00:00:10.160 --> 00:00:12.710
To make a donation or to
view additional materials

00:00:12.710 --> 00:00:16.620
from hundreds of MIT courses,
visit MIT OpenCourseWare

00:00:16.620 --> 00:00:17.275
at ocw.mit.edu.

00:00:25.560 --> 00:00:27.060
PROFESSOR: All right
so today, we're

00:00:27.060 --> 00:00:32.020
going to have a review of the
visual and oculomotor systems

00:00:32.020 --> 00:00:34.030
that I've covered so far.

00:00:34.030 --> 00:00:35.820
And what I'm going
to do is I'm going

00:00:35.820 --> 00:00:41.030
to go over many, many
basic facts in a rather

00:00:41.030 --> 00:00:47.710
quick fashion, which will sort
of refresh your memory of what

00:00:47.710 --> 00:00:50.450
we have covered so far
and also will make you

00:00:50.450 --> 00:00:55.790
more aware of what you want to
look at carefully when you look

00:00:55.790 --> 00:01:03.290
at the material on the
website, on Stellar,

00:01:03.290 --> 00:01:07.805
and also when you read
the assigned readings.

00:01:09.380 --> 00:01:12.270
I want to remind all
of you again that you

00:01:12.270 --> 00:01:15.700
will have to put
together that paper

00:01:15.700 --> 00:01:18.960
on the accessory optic system
that I will mention very

00:01:18.960 --> 00:01:21.075
briefly at the
end of the review.

00:01:22.680 --> 00:01:25.230
And your prime task
there will be--

00:01:25.230 --> 00:01:28.800
that's an old paper
published in the 1960s, which

00:01:28.800 --> 00:01:31.080
was a major discovery
at the time--

00:01:31.080 --> 00:01:33.620
and your task will
be predominantly

00:01:33.620 --> 00:01:36.700
to say what-- well, first of all
what has been discovered there,

00:01:36.700 --> 00:01:40.580
that you can cover in a
paragraph and then to add to it

00:01:40.580 --> 00:01:45.230
what people have
contributed to the study

00:01:45.230 --> 00:01:50.990
of that area since
that original paper.

00:01:50.990 --> 00:01:54.340
All right, so anyway
then I will mention

00:01:54.340 --> 00:01:57.720
at the end or so a bit
more about the exam, which

00:01:57.720 --> 00:02:01.240
is going to take place on
Wednesday right in here, which

00:02:01.240 --> 00:02:04.220
is going to consist of
multiple choice questions.

00:02:04.220 --> 00:02:06.090
All right, so to
begin with then,

00:02:06.090 --> 00:02:11.390
let's talk about the basic
wiring of the visual system

00:02:11.390 --> 00:02:12.730
that we have covered.

00:02:12.730 --> 00:02:18.845
And that is outlined here for
the primate and for the human.

00:02:20.540 --> 00:02:25.450
And I should mention as I had
gone in the initial lecture

00:02:25.450 --> 00:02:29.400
that this is different from
many of the lower level

00:02:29.400 --> 00:02:32.750
species in which the
two eyes look sideways,

00:02:32.750 --> 00:02:38.210
and each eye sends all of its
retinal ganglion cell axons

00:02:38.210 --> 00:02:43.760
across to the other
hemisphere in the brain.

00:02:43.760 --> 00:02:45.410
Now this big change
occurred when

00:02:45.410 --> 00:02:47.060
the eyes move to the front.

00:02:47.060 --> 00:02:50.200
And we have discussed already
why that may have happened.

00:02:50.200 --> 00:02:55.220
And as a result of this, if you
imagine cutting your vertically

00:02:55.220 --> 00:02:58.380
in half, you divide it into
the nasal and temporal hemi

00:02:58.380 --> 00:02:59.210
retinae.

00:02:59.210 --> 00:03:01.170
And it so happens
that the nasal hemi

00:03:01.170 --> 00:03:03.530
retina of one eye and
the temporal retina

00:03:03.530 --> 00:03:05.340
of the other eye
goes to one side,

00:03:05.340 --> 00:03:07.780
and the obvious happens
to the other side.

00:03:07.780 --> 00:03:11.390
The connections are made to
several areas, most notably

00:03:11.390 --> 00:03:14.530
for our purposes, was the
lateral geniculate nucleus

00:03:14.530 --> 00:03:17.770
but also the superior colliculus
and several other structures

00:03:17.770 --> 00:03:19.660
that we have talked
about that include

00:03:19.660 --> 00:03:21.790
the accessory optic system.

00:03:21.790 --> 00:03:24.930
Once the connections
come up to the cortex,

00:03:24.930 --> 00:03:28.950
several cortical areas-- we'll
talk about that in a minute--

00:03:28.950 --> 00:03:34.060
have evolved that are involved
in progressively higher levels

00:03:34.060 --> 00:03:36.690
of visual analysis.

00:03:37.810 --> 00:03:42.310
Now this circle here--
hopefully you guys

00:03:42.310 --> 00:03:46.200
remember-- is either called
the Vieth-Muller circle,

00:03:46.200 --> 00:03:48.260
or it's called the horopter.

00:03:48.260 --> 00:03:50.980
And it was shown by a
clever experimentalist

00:03:50.980 --> 00:03:55.110
that if you put any
spot along that circle

00:03:55.110 --> 00:03:58.800
when the person fixes at
this point in the circle,

00:03:58.800 --> 00:04:03.740
all of those points impinge on
corresponding points in the two

00:04:03.740 --> 00:04:04.240
retinae.

00:04:05.390 --> 00:04:07.250
However, if there's
an object that

00:04:07.250 --> 00:04:11.850
seemed to be beyond or closer
to the eye than the circle,

00:04:11.850 --> 00:04:16.370
then they hit non equivalent
points in the retina.

00:04:16.370 --> 00:04:20.149
And that non equivalency
is actually used

00:04:20.149 --> 00:04:22.400
for depth perception,
as we have discussed,

00:04:22.400 --> 00:04:27.560
and I will mention again when
it comes to a stereopsis.

00:04:27.560 --> 00:04:32.060
So that is the very, very
basic wiring arrangement.

00:04:32.060 --> 00:04:34.130
And then if you
proceed from here

00:04:34.130 --> 00:04:36.780
and look at the retina and
the lateral geniculate nucleus

00:04:36.780 --> 00:04:39.830
in a bit more detail, first
starting with the retina.

00:04:39.830 --> 00:04:42.230
I wanted to point out
to you first of all

00:04:42.230 --> 00:04:45.300
that there are two different
kinds of photoreceptors.

00:04:45.300 --> 00:04:47.380
You all know this
very well by now.

00:04:47.380 --> 00:04:49.910
You knew this before
you came to class.

00:04:49.910 --> 00:04:51.900
You have the rods and the cones.

00:04:51.900 --> 00:04:56.470
There are three basic types of
cones, red, green, and blue,

00:04:56.470 --> 00:05:00.200
which more appropriately refer
to short, medium, and long

00:05:00.200 --> 00:05:02.640
wavelength selective cones.

00:05:02.640 --> 00:05:04.910
And then we have the rods.

00:05:04.910 --> 00:05:10.690
Now what happens is
that the light comes in,

00:05:10.690 --> 00:05:14.470
in this case from the bottom
if you look at yourself.

00:05:14.470 --> 00:05:17.870
The light comes in, and
it goes through many

00:05:17.870 --> 00:05:19.710
of the cells in the retina.

00:05:19.710 --> 00:05:23.040
And it impinges
on the receptors,

00:05:23.040 --> 00:05:27.490
which are facing away from
the light against the pigment

00:05:27.490 --> 00:05:28.710
epithelium.

00:05:28.710 --> 00:05:31.670
And as I've mentioned
to you, there

00:05:31.670 --> 00:05:33.620
have been some
interesting questions

00:05:33.620 --> 00:05:37.450
as to why this strange
arrangement had emerged.

00:05:37.450 --> 00:05:41.150
Nobody had predicted that
before we had any anatomy,

00:05:41.150 --> 00:05:43.150
people just thought if
there were any receptors,

00:05:43.150 --> 00:05:44.730
they would face the light.

00:05:44.730 --> 00:05:48.280
So that's certainly a strange
arrangement, an unusual one.

00:05:48.280 --> 00:05:51.030
And a lot of
speculation have been

00:05:51.030 --> 00:05:53.780
advanced as to why
this has happened.

00:05:53.780 --> 00:05:56.240
And I will just briefly
mention two of those.

00:05:56.240 --> 00:05:58.950
One is that when
these photoreceptors

00:05:58.950 --> 00:06:01.640
are right against the
pigment epithelium, which

00:06:01.640 --> 00:06:12.050
in diurnal animals is black,
and absorbs the photons thereby

00:06:12.050 --> 00:06:14.480
preventing scattering of light.

00:06:14.480 --> 00:06:16.990
As a result of which,
you can gain high acuity.

00:06:18.410 --> 00:06:23.130
And that is known by the
fact that if you assess

00:06:23.130 --> 00:06:27.510
the visual capacities of albinos
who don't have a black pigment

00:06:27.510 --> 00:06:30.120
epithelium because they
lack pigment-- that's

00:06:30.120 --> 00:06:32.140
the definition of
being an albino--

00:06:32.140 --> 00:06:35.300
most people have very poor
vision because the photons

00:06:35.300 --> 00:06:37.820
come into the eye, and they
scatter all over the place

00:06:37.820 --> 00:06:42.730
and activate many photoreceptors
rather than just those

00:06:42.730 --> 00:06:47.010
which the incoming photon
would hit directly.

00:06:47.010 --> 00:06:49.760
So that is the
arrangement for these.

00:06:49.760 --> 00:06:54.110
Another factor, which I don't
think I may not have mentioned

00:06:54.110 --> 00:07:02.050
is that what you have
in these photoreceptors

00:07:02.050 --> 00:07:05.770
are little packets, if you talk
about the rods in particular,

00:07:05.770 --> 00:07:07.710
you have little
packets, each of which

00:07:07.710 --> 00:07:12.680
has the molecules,
which are sensitive

00:07:12.680 --> 00:07:14.470
to the incoming light.

00:07:14.470 --> 00:07:17.603
And each of these
packets-- there

00:07:17.603 --> 00:07:20.435
are about 1,000 packets
in each of these rods.

00:07:21.890 --> 00:07:26.190
And each of those 1,000 packets
is about 10,000 molecules.

00:07:26.190 --> 00:07:28.470
So we're talking about
gigantic numbers.

00:07:28.470 --> 00:07:34.440
Now what happens is that given
this 1,000 of-- about 1,000

00:07:34.440 --> 00:07:38.380
packets, they are not
there for your life.

00:07:38.380 --> 00:07:42.220
What happens is that
the packets gradually

00:07:42.220 --> 00:07:45.570
disintegrate and get
replaced by a new one.

00:07:46.920 --> 00:07:51.490
It's about once every 10
days you lose a packet,

00:07:51.490 --> 00:07:53.430
and you gain a new one.

00:07:53.430 --> 00:07:56.300
Now that means that some of
this stuff gets sloughed off.

00:07:56.300 --> 00:08:01.620
And one of the reasons people
think that the photoreceptors

00:08:01.620 --> 00:08:04.430
ended up facing away from
the light is that they could

00:08:04.430 --> 00:08:07.310
be close to this inner
part of the retina

00:08:07.310 --> 00:08:11.690
where anything that's
sloughed off can be absorbed

00:08:11.690 --> 00:08:15.860
rather than being just
thrown into the eye itself,

00:08:15.860 --> 00:08:17.920
into the vitreous,
because if that

00:08:17.920 --> 00:08:20.250
were to happen over
many, many years,

00:08:20.250 --> 00:08:23.690
the vitreous would become
cloudy, and you couldn't see.

00:08:23.690 --> 00:08:25.900
So those are two
possible reasons

00:08:25.900 --> 00:08:28.820
why this strange
arrangement has evolved.

00:08:28.820 --> 00:08:31.020
And you see this in
virtually all species.

00:08:31.020 --> 00:08:35.450
There are just a few species
that have-- and most of those

00:08:35.450 --> 00:08:41.580
are actually in the sea--
who have photoreceptors

00:08:41.580 --> 00:08:43.440
that face towards the light.

00:08:43.440 --> 00:08:45.790
All right, and then
if you proceed here,

00:08:45.790 --> 00:08:48.700
the other amazing thing
that had been discovered

00:08:48.700 --> 00:08:54.410
is that all the photoreceptors
hyper polarized to light.

00:08:54.410 --> 00:08:57.060
Again, they do the opposite
of what people had thought.

00:08:57.060 --> 00:08:59.150
You'd think that
when photons come in,

00:08:59.150 --> 00:09:03.000
they would activate
the photoreceptors,

00:09:03.000 --> 00:09:08.400
and they would send the
signal down the stream

00:09:08.400 --> 00:09:10.010
through the eye.

00:09:10.010 --> 00:09:14.630
Turns out the opposite
happens that the discharge

00:09:14.630 --> 00:09:17.800
on the neurotransmitter
here occurs

00:09:17.800 --> 00:09:20.421
when there's a darkening rather
than an increase in light.

00:09:20.421 --> 00:09:22.045
That's an important
factor to remember.

00:09:23.310 --> 00:09:25.570
That's true for all
the forests accept

00:09:25.570 --> 00:09:28.250
as all photoreceptors
hyperpolarized to light.

00:09:28.250 --> 00:09:29.560
You know this well already.

00:09:29.560 --> 00:09:32.120
I must have said that
about 10 times by now.

00:09:32.120 --> 00:09:33.960
Now the amazing
thing is that when

00:09:33.960 --> 00:09:38.830
you come to the bipolar cells,
the next set of receptors here,

00:09:38.830 --> 00:09:41.600
it was discovered
that two basic types

00:09:41.600 --> 00:09:44.420
of several different types
like from for the major parasol

00:09:44.420 --> 00:09:48.970
cells, but there are two basic
types, the on and the off.

00:09:48.970 --> 00:09:55.450
And this is accomplished by
having two kinds of synapses

00:09:55.450 --> 00:09:59.510
in the on and off bipolar cells,
sign conserving ones and sign

00:09:59.510 --> 00:10:00.560
inverting ones.

00:10:01.650 --> 00:10:04.210
This is accomplished
in the on bipolars

00:10:04.210 --> 00:10:08.620
by virtue of the [INAUDIBLE]
six receptor site

00:10:08.620 --> 00:10:14.160
and the [INAUDIBLE] one and
two for the off bipolar cells.

00:10:14.160 --> 00:10:20.570
So that means now that you have
signals in some of these cells

00:10:20.570 --> 00:10:24.260
when there's an increase in
light and the signals in some

00:10:24.260 --> 00:10:26.220
where there's a
decrease in light.

00:10:26.220 --> 00:10:30.030
So that's the situation for
the on and off bipolars.

00:10:30.030 --> 00:10:35.860
And then when you come into the
level of the ganglion cells,

00:10:35.860 --> 00:10:40.360
two major classes of ganglion
cells are the on and the off.

00:10:40.360 --> 00:10:43.570
Now I'll talk about that in
more detail in just a minute.

00:10:43.570 --> 00:10:45.570
Now the other
interesting, curious thing

00:10:45.570 --> 00:10:49.090
is that when you
look at the rods,

00:10:49.090 --> 00:10:57.480
the rods-- and they connect
to their bipolar cells.

00:10:57.480 --> 00:11:00.380
They are all sign
inverting synapses.

00:11:01.440 --> 00:11:03.920
They only come in one
type, at least in humans

00:11:03.920 --> 00:11:05.530
and in primates.

00:11:05.530 --> 00:11:10.370
So what happens then to
create on and off section

00:11:10.370 --> 00:11:12.350
done in the inner
retina by virtue

00:11:12.350 --> 00:11:16.540
of having a synapse here
[INAUDIBLE] to amacrine cell,

00:11:16.540 --> 00:11:20.590
which is a glycinergic synapse,
and it also makes connections

00:11:20.590 --> 00:11:23.180
to the on bipolar,
which is a gap junction.

00:11:23.180 --> 00:11:25.640
And this way, it becomes
a double ended system

00:11:25.640 --> 00:11:28.140
for the rods as well
as for the cones.

00:11:28.140 --> 00:11:29.911
So hopefully you guys
all remember this.

00:11:29.911 --> 00:11:30.910
I know it's complicated.

00:11:32.110 --> 00:11:34.890
But that is something that one
doesn't have a choice about.

00:11:34.890 --> 00:11:36.690
That's how it simply is.

00:11:36.690 --> 00:11:38.590
All right, so now we move on.

00:11:38.590 --> 00:11:41.660
And we are going to look at
the lateral geniculate nucleus.

00:11:41.660 --> 00:11:42.899
Here's a cross section of it.

00:11:42.899 --> 00:11:43.815
I've shown you before.

00:11:44.950 --> 00:11:47.400
It was discovered that
the lateral geniculate

00:11:47.400 --> 00:11:50.680
nucleus in central retina--
this is a monkey retina.

00:11:50.680 --> 00:11:53.130
The human is very
similar, so input

00:11:53.130 --> 00:11:57.620
from a monkey retina to the
lateral geniculate nucleus.

00:11:57.620 --> 00:12:01.410
And the six layers consist
of two major types,

00:12:01.410 --> 00:12:04.770
the parvocellular so called
the magnocellular layers.

00:12:04.770 --> 00:12:08.160
And what was discovered is
that the parvocellular layers

00:12:08.160 --> 00:12:10.940
get input from the
midget cells that we'll

00:12:10.940 --> 00:12:12.330
talk about in just a minute.

00:12:12.330 --> 00:12:18.260
And the bottom two layers, which
are the magnocellular layers

00:12:18.260 --> 00:12:20.550
get input from
the parasol cells.

00:12:20.550 --> 00:12:25.140
And then what happens is that
when you go from central vision

00:12:25.140 --> 00:12:27.980
to peripheral vision,
you have a huge change

00:12:27.980 --> 00:12:32.150
in the productive percentage
of midget and parasol cells

00:12:32.150 --> 00:12:35.440
that you have in the retina
and in the lateral geniculate

00:12:35.440 --> 00:12:38.930
nucleus near the fovea.

00:12:38.930 --> 00:12:42.980
In the foveola itself, you
don't have any parasol cells,

00:12:42.980 --> 00:12:45.930
but in the fovea itself you do.

00:12:45.930 --> 00:12:48.070
And there's a ratio
of about eight to one.

00:12:48.070 --> 00:12:49.870
And then as you go
to the periphery,

00:12:49.870 --> 00:12:52.040
eventually they're
equally in number.

00:12:52.040 --> 00:12:55.860
So there's a huge emphasis
on the midget system

00:12:55.860 --> 00:13:00.110
in central vision and
the much increased

00:13:00.110 --> 00:13:04.060
emphasis on the parasol
cells in peripheral vision.

00:13:04.060 --> 00:13:05.160
So that's the arrangement.

00:13:05.160 --> 00:13:08.090
And this is reflected
in the geniculate, which

00:13:08.090 --> 00:13:13.760
has six layers in central
vision after about 18 degrees.

00:13:13.760 --> 00:13:16.720
And it has four layers
in the periphery where

00:13:16.720 --> 00:13:20.000
the midget and parasol
inputs are pretty much

00:13:20.000 --> 00:13:23.710
equal as reflected
by these four layers.

00:13:23.710 --> 00:13:26.950
So that's the basic arrangements
for the lateral geniculate

00:13:26.950 --> 00:13:27.800
nucleus.

00:13:27.800 --> 00:13:31.360
Now if we move on-- let
me say one more thing.

00:13:31.360 --> 00:13:36.970
The receptive field properties
of cells in the retina,

00:13:36.970 --> 00:13:39.220
in retinal ganglion
cells, I should say,

00:13:39.220 --> 00:13:43.049
and in the lateral geniculate
nucleus are highly similar.

00:13:43.049 --> 00:13:44.215
They're virtually identical.

00:13:45.550 --> 00:13:48.130
You have circular
receptive fields

00:13:48.130 --> 00:13:50.340
with centers around antagonism.

00:13:50.340 --> 00:13:52.600
All right, now if you
move on and move up

00:13:52.600 --> 00:13:55.170
to the visual
cortex, what happens

00:13:55.170 --> 00:14:00.450
is that there's a huge
change that arises,

00:14:00.450 --> 00:14:02.720
the beautiful discoveries
made by Hubel and Wiesel

00:14:02.720 --> 00:14:04.920
for which they had
received the Nobel Prize.

00:14:04.920 --> 00:14:08.790
And this is just a quick
view of the monkey brain.

00:14:08.790 --> 00:14:10.465
Here is area V1.

00:14:12.150 --> 00:14:14.560
I'll come back to the
other areas in a minute.

00:14:14.560 --> 00:14:16.390
The nice thing about
this in the monkey

00:14:16.390 --> 00:14:19.760
is that this area is
[INAUDIBLE] as I had told you.

00:14:19.760 --> 00:14:22.590
And because of that,
it's easy to study

00:14:22.590 --> 00:14:28.350
the cells and their
properties in area V1.

00:14:28.350 --> 00:14:32.240
All right, so now if one
examines the properties

00:14:32.240 --> 00:14:35.750
of single cells
in area V1, it was

00:14:35.750 --> 00:14:41.210
discovered some major
transformations had occurred

00:14:41.210 --> 00:14:44.340
in the inputs from the
lateral geniculate nucleus.

00:14:44.340 --> 00:14:48.720
And these major transforms can
be summarized in just a second.

00:14:48.720 --> 00:14:53.100
But I will first tell you
that there is a differential

00:14:53.100 --> 00:14:56.680
input from the parvocellular
and magnocellular layers, which

00:14:56.680 --> 00:15:00.374
project respectively to [? 4C ?]
beta and [? 4C ?] alpha.

00:15:00.374 --> 00:15:01.790
And then there's
yet another class

00:15:01.790 --> 00:15:07.360
of cells that originates in
the retina that are project

00:15:07.360 --> 00:15:09.530
into the inter lamina
layers, and they

00:15:09.530 --> 00:15:14.630
project into the upper
portions of the visual cortex.

00:15:14.630 --> 00:15:17.650
So now if one looks in
detail at the properties

00:15:17.650 --> 00:15:21.120
of these cells, which we
have discussed quite a bit.

00:15:21.120 --> 00:15:23.560
You can refer to
these as transforms.

00:15:23.560 --> 00:15:28.690
The transforms of the visual
input into the cortical cells.

00:15:28.690 --> 00:15:30.880
So when you're record
from these cortical cells,

00:15:30.880 --> 00:15:34.980
you'll find one big transform
is that these cells,

00:15:34.980 --> 00:15:37.190
the overwhelming
majority of these cells,

00:15:37.190 --> 00:15:38.870
become orientation selective.

00:15:43.750 --> 00:15:46.680
Many cells become
direction selective,

00:15:46.680 --> 00:15:51.630
virtually all simple cells and
about half the complex cells.

00:15:51.630 --> 00:15:53.780
So direction selectively
becomes very important.

00:15:53.780 --> 00:15:56.550
We'll talk about that in a
bit more detail later on.

00:15:56.550 --> 00:15:59.870
Then, some cells are
spatial frequency selective.

00:15:59.870 --> 00:16:02.620
Many cells get an
input from both eyes.

00:16:04.720 --> 00:16:07.640
And there's a
convergence of input

00:16:07.640 --> 00:16:10.580
from the on and off channels.

00:16:10.580 --> 00:16:12.940
This is also true for
some of the cells that

00:16:12.940 --> 00:16:16.100
get a convergent input from
the midget and parasol cells.

00:16:16.100 --> 00:16:18.590
So those are the
major transforms

00:16:18.590 --> 00:16:21.860
that you see in
the visual cortex.

00:16:21.860 --> 00:16:26.700
All right, so now as a result of
having made these discoveries,

00:16:26.700 --> 00:16:28.530
people came up with
a question of how

00:16:28.530 --> 00:16:32.050
is this organized in
the visual cortex.

00:16:32.050 --> 00:16:34.000
And the first point
that I had made

00:16:34.000 --> 00:16:37.490
is that there's a
topographic layout

00:16:37.490 --> 00:16:40.590
of the visual field
in visual cortex.

00:16:41.700 --> 00:16:43.750
But with much more
area allocated

00:16:43.750 --> 00:16:46.560
for central vision than
peripheral vision simply

00:16:46.560 --> 00:16:49.370
copying the relative
percentage of cells

00:16:49.370 --> 00:16:52.790
already in the retina that
exist in central vision

00:16:52.790 --> 00:16:56.100
and peripheral vision
and because the thickness

00:16:56.100 --> 00:17:00.050
of the gray matter in cortex is
about two millimeters roughly.

00:17:02.010 --> 00:17:03.510
And it's constant.

00:17:03.510 --> 00:17:06.300
More space has to be
allocated for central vision

00:17:06.300 --> 00:17:07.960
than peripheral vision.

00:17:07.960 --> 00:17:09.670
And as a result of
these people that

00:17:09.670 --> 00:17:14.380
studied the spacial
arrangement and organization

00:17:14.380 --> 00:17:17.680
of the visual cortex--
and the initial model

00:17:17.680 --> 00:17:19.300
that was proposed,
if you remember,

00:17:19.300 --> 00:17:21.780
is the Hubel and
Wiesel model, according

00:17:21.780 --> 00:17:23.640
to which in one
direction you have

00:17:23.640 --> 00:17:25.900
the alternation of
left and right eyes.

00:17:25.900 --> 00:17:29.210
You have column, left,
right, left, right.

00:17:29.210 --> 00:17:32.590
And in the other direction,
you have a systematic change

00:17:32.590 --> 00:17:34.770
in the orientation of cells.

00:17:34.770 --> 00:17:37.620
Now this model
didn't fare that well

00:17:37.620 --> 00:17:40.330
because it's not as neat
as has been proposed.

00:17:40.330 --> 00:17:42.880
An alternative model
was the Raidial model.

00:17:42.880 --> 00:17:44.400
And the last one
I'm showing here,

00:17:44.400 --> 00:17:47.740
which I call the Swirl
model is not really a model

00:17:47.740 --> 00:17:50.203
because some very
clever experiments that

00:17:50.203 --> 00:17:53.610
have been carried
out by [INAUDIBLE]

00:17:53.610 --> 00:17:57.680
actually did optical
recording and demonstrated

00:17:57.680 --> 00:18:00.040
that the visual
cortex from the top

00:18:00.040 --> 00:18:01.840
looks something
like this where you

00:18:01.840 --> 00:18:06.000
have indeed systematic
arrangement of orientations

00:18:06.000 --> 00:18:09.060
in left and right eye columns.

00:18:09.060 --> 00:18:12.430
But it's not a
straight linear factor,

00:18:12.430 --> 00:18:14.130
but it's kind of a
swirly arrangement.

00:18:15.300 --> 00:18:20.230
So that then established
what is the layout

00:18:20.230 --> 00:18:22.490
of the primary visual cortex.

00:18:22.490 --> 00:18:27.960
OK now the other important
thing that we had emphasized

00:18:27.960 --> 00:18:32.080
is that contrary to some
of the popular ideas

00:18:32.080 --> 00:18:37.090
that people have had that
the cells in the brain

00:18:37.090 --> 00:18:40.870
are feature selective,
meaning that they'll extract

00:18:40.870 --> 00:18:46.140
specific features from the
visual scene like say one cell

00:18:46.140 --> 00:18:49.610
extracts color, another cell
extracts a particular face,

00:18:49.610 --> 00:18:50.400
and so on.

00:18:50.400 --> 00:18:53.330
It turns out that that's
a false impression

00:18:53.330 --> 00:18:54.820
that people had gained.

00:18:54.820 --> 00:18:58.150
And instead what is
happening that any given one

00:18:58.150 --> 00:19:03.685
cell processes many different
kinds of visual information.

00:19:04.770 --> 00:19:08.450
And it's the activity of
thousands and thousands

00:19:08.450 --> 00:19:15.840
of cells in a network that
can come up with the percepts

00:19:15.840 --> 00:19:18.150
that you perceive.

00:19:18.150 --> 00:19:21.640
Now that's extremely
complicated, 10 times more

00:19:21.640 --> 00:19:23.660
complicated than any computer.

00:19:23.660 --> 00:19:26.720
And it is something
that to a larger extent

00:19:26.720 --> 00:19:28.450
still has not been solved.

00:19:28.450 --> 00:19:31.970
You don't know how does a
person recognize a face.

00:19:31.970 --> 00:19:34.960
You can tell oh, it takes place
in various parts of the brain

00:19:34.960 --> 00:19:35.680
and so on.

00:19:35.680 --> 00:19:38.540
But exactly physically
how that's done

00:19:38.540 --> 00:19:42.150
is something that's still
remains largely a mystery.

00:19:42.150 --> 00:19:45.780
All right, so now
let's move on and talk

00:19:45.780 --> 00:19:47.960
about extrastriate cortex.

00:19:47.960 --> 00:19:53.190
And extrastriate cortex-- here's
a diagram of the monkey brain

00:19:53.190 --> 00:19:54.040
again.

00:19:54.040 --> 00:19:57.170
Now I'll point out
here's area V1.

00:19:57.170 --> 00:20:00.270
And once you get close to
the lunate sulcus here,

00:20:00.270 --> 00:20:03.190
V2 begins and folds under.

00:20:03.190 --> 00:20:06.290
And then inside
there we have V3.

00:20:06.290 --> 00:20:09.500
And then actually make
folds back out again.

00:20:09.500 --> 00:20:10.995
You have area V4 here.

00:20:12.320 --> 00:20:16.140
And then you have areas
MT and MST right here.

00:20:16.140 --> 00:20:18.400
And then, in addition,
you have, of course,

00:20:18.400 --> 00:20:20.990
your infertemporal
cortex area, which

00:20:20.990 --> 00:20:25.255
plays a very important role in
complex analyses such as faces.

00:20:26.650 --> 00:20:30.390
And then you come
to the frontal lobe

00:20:30.390 --> 00:20:33.010
here, in which you have
the frontal eye fields

00:20:33.010 --> 00:20:36.890
and medial eye fields that
also process visual information

00:20:36.890 --> 00:20:40.410
but mostly for eye
movements that I

00:20:40.410 --> 00:20:42.700
will talk about later on.

00:20:42.700 --> 00:20:46.840
So that then is in a
nutshell the arrangement.

00:20:46.840 --> 00:20:51.480
And much of the work that has
been done in the past dozen

00:20:51.480 --> 00:20:56.055
years or so was to examine what
these extrastriate areas do

00:20:56.055 --> 00:20:57.000
for vision.

00:20:57.000 --> 00:20:59.460
And I'll come back
to that when we

00:20:59.460 --> 00:21:01.970
talk about higher level
visual processing.

00:21:01.970 --> 00:21:05.030
Now basically the
fact is that there

00:21:05.030 --> 00:21:11.010
are more than 30 visual
areas and that there

00:21:11.010 --> 00:21:13.810
are more than 300
interconnections among them.

00:21:15.240 --> 00:21:18.890
Initially the idea was--
the feature detection idea--

00:21:18.890 --> 00:21:22.800
that each of these areas
is specific for analyzing

00:21:22.800 --> 00:21:25.000
a particular type of percept.

00:21:25.000 --> 00:21:28.180
But then it became more evident,
increasingly more evident,

00:21:28.180 --> 00:21:32.690
that these areas tremendously
interact with each other

00:21:32.690 --> 00:21:37.020
and perform these
very complex analyses

00:21:37.020 --> 00:21:41.290
based on networks being active.

00:21:41.290 --> 00:21:44.670
Now the basic major
cortical visual areas,

00:21:44.670 --> 00:21:49.420
V1 I just talked about, V2
I mentioned, V3, V4, MT.

00:21:49.420 --> 00:21:52.900
Then when you come to
the temporal cortex,

00:21:52.900 --> 00:21:56.740
you come to infertemporal
region that I just mentioned.

00:21:56.740 --> 00:21:59.020
And then in the
parietal cortex, we

00:21:59.020 --> 00:22:00.900
have the lateral
parietal sulcrus,

00:22:00.900 --> 00:22:04.340
the ventral interparietal, and
the medial superior temporal

00:22:04.340 --> 00:22:05.140
sulcrus.

00:22:05.140 --> 00:22:07.060
So those are some
of the major areas.

00:22:07.060 --> 00:22:10.270
And then as I've already
noted, in the frontal cortex,

00:22:10.270 --> 00:22:11.950
we have the frontal eye fields.

00:22:11.950 --> 00:22:15.560
And then even we had the medial
eye fields, which are not

00:22:15.560 --> 00:22:19.210
listed here that
also play a role

00:22:19.210 --> 00:22:22.140
in eye movements,
perhaps a lesser

00:22:22.140 --> 00:22:24.030
extent in visual
analysis as such.

00:22:24.030 --> 00:22:28.260
But many of the cells there too
have visual receptive fields,

00:22:28.260 --> 00:22:31.180
although they are
very hard to discern.

00:22:31.180 --> 00:22:35.900
They much more clearly have
motor fields than visual field.

00:22:35.900 --> 00:22:38.100
All right so now what
we are going to do is

00:22:38.100 --> 00:22:40.540
we are going to go
back to the beginning

00:22:40.540 --> 00:22:44.490
and look at the so-called
on and off channels briefly.

00:22:44.490 --> 00:22:45.890
We talked about that a lot.

00:22:46.980 --> 00:22:49.930
Again, to reemphasize,
all photoreceptors

00:22:49.930 --> 00:22:51.690
hyperpolarize to light.

00:22:51.690 --> 00:22:57.650
And then because the two major
classes of neurotransmitter

00:22:57.650 --> 00:23:00.050
receptor sites in
the bipolar cells,

00:23:00.050 --> 00:23:02.440
you create a double ended
system from a single

00:23:02.440 --> 00:23:05.340
ended one creating the
so-called on and off.

00:23:05.340 --> 00:23:08.240
Now these systems were
discovered initially

00:23:08.240 --> 00:23:10.480
by Keffer Hartline,
who received the Nobel

00:23:10.480 --> 00:23:12.590
Prize for that
remarkable discovery.

00:23:12.590 --> 00:23:16.250
And he thought at the
time that the on system

00:23:16.250 --> 00:23:18.590
signaled when a
stimulus came on.

00:23:18.590 --> 00:23:21.610
And the off channel
signaled when it went off.

00:23:21.610 --> 00:23:23.540
That was his idea,
which turned out

00:23:23.540 --> 00:23:28.160
to be all wrong because that's
not what these cells are about.

00:23:28.160 --> 00:23:31.390
What these cells are about, as
I've pointed out repeatedly,

00:23:31.390 --> 00:23:36.920
is that they can process both
light increment and light

00:23:36.920 --> 00:23:40.660
decrement with an
excitatory response.

00:23:40.660 --> 00:23:44.060
That means because of the
nature, the physical nature,

00:23:44.060 --> 00:23:48.630
of light that some objects
in the world reflect light,

00:23:48.630 --> 00:23:51.650
and some objects in
the world absorb light.

00:23:53.860 --> 00:23:56.340
Because of this,
as you look around,

00:23:56.340 --> 00:23:58.270
some objects look
black, and some objects

00:23:58.270 --> 00:23:59.940
look white, or whatever.

00:23:59.940 --> 00:24:04.980
And because of that, to
be able to rapidly process

00:24:04.980 --> 00:24:09.200
something that is a dark object
as well as a light object,

00:24:09.200 --> 00:24:11.610
you need to have
excitatory signals

00:24:11.610 --> 00:24:14.520
to go to the central nervous
system to process that.

00:24:14.520 --> 00:24:17.440
So therefore, we can
say, first of all,

00:24:17.440 --> 00:24:19.370
that we have these cell types.

00:24:19.370 --> 00:24:21.910
And they won't have sensor
surround antagonism.

00:24:21.910 --> 00:24:23.990
And let me add
one more fact here

00:24:23.990 --> 00:24:26.770
is that they're comfortable
with adaptation,

00:24:26.770 --> 00:24:31.760
that the average firing rate,
average maximum firing rate,

00:24:31.760 --> 00:24:38.395
of a retinal ganglion cell is
maybe about 400 to 600 hertz.

00:24:39.670 --> 00:24:43.380
And that is a rather
limited frequency range.

00:24:43.380 --> 00:24:46.710
And yet, you have to
analyze practically

00:24:46.710 --> 00:24:50.630
over 10 log units of
light information.

00:24:50.630 --> 00:24:53.530
And because of that, the
sensor surround antagonism

00:24:53.530 --> 00:24:55.780
has evolved so that
these cells always

00:24:55.780 --> 00:25:03.380
look at local contrast
changes rather than absolutes.

00:25:03.380 --> 00:25:07.190
So then, if you look at
the on and off cells,

00:25:07.190 --> 00:25:10.400
I've told you, in accordance
with the sensor surround

00:25:10.400 --> 00:25:13.834
antagonism, if you split
a small spot of light

00:25:13.834 --> 00:25:15.250
in the center of
receptive fields,

00:25:15.250 --> 00:25:17.460
on cells fire when
you increase it.

00:25:17.460 --> 00:25:20.080
Off cells fire when
you decrease it.

00:25:20.080 --> 00:25:22.400
But when you use a
much larger spot,

00:25:22.400 --> 00:25:24.240
you get a lesser
response because

00:25:24.240 --> 00:25:26.100
of the surround antagonism.

00:25:26.100 --> 00:25:30.420
So that's the basic principle
of these two types of cells.

00:25:30.420 --> 00:25:34.500
And then, I told you
about these experiments,

00:25:34.500 --> 00:25:37.465
in which two
2-amino-4-phosphonobutyrate had

00:25:37.465 --> 00:25:42.470
been used, which is for
brief purposes, called APB.

00:25:42.470 --> 00:25:45.140
And I told you about two
types of experiments,

00:25:45.140 --> 00:25:49.290
one doing single cell recordings
in various parts of the brain

00:25:49.290 --> 00:25:51.920
and the others to do
behavioral studies.

00:25:51.920 --> 00:25:55.230
And what the signal cell
recordings had shown

00:25:55.230 --> 00:26:02.060
is that the-- let me
first say what APB does.

00:26:02.060 --> 00:26:03.190
APB is what?

00:26:03.190 --> 00:26:04.540
Anybody remember?

00:26:04.540 --> 00:26:07.215
It's a neurotransmitter analog.

00:26:08.780 --> 00:26:10.460
And what neurotransmitter is it?

00:26:12.010 --> 00:26:12.885
AUDIENCE: [INAUDIBLE]

00:26:14.340 --> 00:26:15.310
PROFESSOR: Very good.

00:26:15.310 --> 00:26:15.910
Glutamate.

00:26:15.910 --> 00:26:20.520
All right so what you do is
when you inject this substance

00:26:20.520 --> 00:26:25.460
into the eye-- this is
an artificial substance--

00:26:25.460 --> 00:26:30.130
it blocks the on
cells from being

00:26:30.130 --> 00:26:32.940
able to respond
to incoming light

00:26:32.940 --> 00:26:35.240
but does nothing
to the off cells.

00:26:36.910 --> 00:26:42.260
So if you do this and study
the responses of single neurons

00:26:42.260 --> 00:26:44.250
in various parts of
the brain-- there

00:26:44.250 --> 00:26:46.090
have been all these
different hypotheses

00:26:46.090 --> 00:26:50.730
as to why we had the
on and off channels.

00:26:50.730 --> 00:26:53.366
One of them was to create
sensor surround antagonism.

00:26:55.280 --> 00:27:02.110
And the other one was to
create orientation direction

00:27:02.110 --> 00:27:03.400
selectivism in the cortex.

00:27:04.550 --> 00:27:09.070
But it turned out that when
you injected APB into the eye,

00:27:09.070 --> 00:27:13.870
and you blocked the on channel,
the off input to the cells,

00:27:13.870 --> 00:27:16.670
and the off cells
therefore, still

00:27:16.670 --> 00:27:18.970
had sensor surround antagonism.

00:27:18.970 --> 00:27:21.005
And the cells in
the cortex still

00:27:21.005 --> 00:27:23.080
had orientation and
direction selectivities.

00:27:24.100 --> 00:27:26.710
So these two systems
did not arise

00:27:26.710 --> 00:27:30.040
for the purpose of creating
those basic attributes, which

00:27:30.040 --> 00:27:33.350
are so central for being able
to analyze the visual scene.

00:27:33.350 --> 00:27:36.410
Now the second
important finding was

00:27:36.410 --> 00:27:39.000
that when you did
a behavioral study

00:27:39.000 --> 00:27:41.035
and asked monkeys to
detect light increment

00:27:41.035 --> 00:27:43.450
and detect light
decrement, there

00:27:43.450 --> 00:27:48.590
was a huge deficit in detecting
light increments but no deficit

00:27:48.590 --> 00:27:50.860
in detecting light decrement.

00:27:50.860 --> 00:27:53.480
So these observations
and many other studies

00:27:53.480 --> 00:27:56.570
analyzing why there
are on and off channels

00:27:56.570 --> 00:27:58.990
came up with the
conclusion, which I think

00:27:58.990 --> 00:28:02.280
is quite valid that
these two systems

00:28:02.280 --> 00:28:07.620
have evolved to enable
organisms to quickly respond

00:28:07.620 --> 00:28:11.130
to both light detrimental
and light incremental input.

00:28:14.220 --> 00:28:18.530
And you probably remember
the little quick movie

00:28:18.530 --> 00:28:21.040
I showed you that you
have a fish in the ocean.

00:28:21.040 --> 00:28:23.140
Fish also have on and
off channels of course.

00:28:24.410 --> 00:28:29.760
If there's a bird in the
sky, like an osprey that

00:28:29.760 --> 00:28:32.220
is seen by virtue
of light decrement,

00:28:32.220 --> 00:28:34.630
your off system tells
that fish, oh, there's

00:28:34.630 --> 00:28:36.810
a bird up there
so it can escape.

00:28:36.810 --> 00:28:39.670
And if a predator
from below that is

00:28:39.670 --> 00:28:45.480
lit up by the sunshine, the
on system in response to that

00:28:45.480 --> 00:28:48.470
and enables this fish to escape.

00:28:48.470 --> 00:28:54.770
So that's one example of the
function, the prime function,

00:28:54.770 --> 00:28:57.430
of the on and off channels.

00:28:57.430 --> 00:28:59.360
All right so that's
the basic fact then.

00:28:59.360 --> 00:29:01.960
So to conclude then,
the on and off channels

00:29:01.960 --> 00:29:03.970
have emerged in the
course of evolution

00:29:03.970 --> 00:29:08.370
to enable organisms to process
both light incremental and

00:29:08.370 --> 00:29:12.016
light decremental information
rapidly and effectively.

00:29:13.670 --> 00:29:17.080
So that's the conclusion
then in a nutshell

00:29:17.080 --> 00:29:20.240
of the on and the off channels.

00:29:20.240 --> 00:29:23.170
Now we can move on and look
at the so-called midget

00:29:23.170 --> 00:29:29.120
and parasol cells that had
been discovered initially

00:29:29.120 --> 00:29:30.190
in the cat.

00:29:30.190 --> 00:29:31.950
And they were called
the x and y cells.

00:29:33.840 --> 00:29:36.440
In the monkey, it's
called midget and parasol

00:29:36.440 --> 00:29:39.420
because when you look
at them anatomically,

00:29:39.420 --> 00:29:40.910
the midget cells are small.

00:29:40.910 --> 00:29:42.900
And they're very
small dendritic arbors

00:29:42.900 --> 00:29:45.230
where the parasols
cells are much bigger

00:29:45.230 --> 00:29:48.634
and have much larger
dendritic arbors that

00:29:48.634 --> 00:29:49.550
look like an umbrella.

00:29:51.070 --> 00:29:53.880
So those two systems
were discovered,

00:29:53.880 --> 00:29:57.080
and statistical analysis
revealed that they are totally

00:29:57.080 --> 00:29:59.230
separate types of cells.

00:29:59.230 --> 00:30:00.800
They're not a continuum.

00:30:00.800 --> 00:30:06.430
So the question then became why
did these two systems evolve?

00:30:06.430 --> 00:30:08.800
And why did nature
go to such trouble

00:30:08.800 --> 00:30:11.276
as to make sure that they
were separate in retina

00:30:11.276 --> 00:30:13.150
and separate in the
geniculate to the monkey?

00:30:14.680 --> 00:30:17.360
And then, in the
cortex, sometimes it

00:30:17.360 --> 00:30:20.840
remains separate-- sometimes the
two systems remain separate--

00:30:20.840 --> 00:30:23.730
and sometimes they
converge as I had

00:30:23.730 --> 00:30:26.450
noted in those
transforms in area V1.

00:30:28.320 --> 00:30:30.100
So now if you look
at that, you've

00:30:30.100 --> 00:30:31.630
seen this several times now.

00:30:31.630 --> 00:30:34.310
The midget system, the
center in central retina

00:30:34.310 --> 00:30:36.320
consists of just a single cone.

00:30:36.320 --> 00:30:38.370
And therefore,
this system should

00:30:38.370 --> 00:30:41.400
be able to tell you about color
whereas the parasol system has

00:30:41.400 --> 00:30:44.420
mixed inputs both in the
center and the surround.

00:30:44.420 --> 00:30:46.890
Furthermore, the parasol
system response much more

00:30:46.890 --> 00:30:49.490
trangently than
the midget system.

00:30:49.490 --> 00:30:54.440
So temporal information can
be processed more effectively

00:30:54.440 --> 00:30:57.120
by the parasol system
than the midget system.

00:30:57.120 --> 00:31:00.670
So those are the initial
observations at the single cell

00:31:00.670 --> 00:31:01.380
level.

00:31:01.380 --> 00:31:05.760
And then behavioral
studies were carried out

00:31:05.760 --> 00:31:08.600
in which either the midget
or the parasol system

00:31:08.600 --> 00:31:10.580
were selectively blocked.

00:31:10.580 --> 00:31:14.810
And then performance was
tested where those systems had

00:31:14.810 --> 00:31:17.900
been blocked and where
the systems were intact.

00:31:17.900 --> 00:31:23.050
And when this was done,
some major findings emerged.

00:31:23.050 --> 00:31:25.700
Before I tell you about
that, let me just reiterate

00:31:25.700 --> 00:31:28.365
again what these
connections are.

00:31:28.365 --> 00:31:30.240
Here we have the midget
and the parasol cells

00:31:30.240 --> 00:31:33.490
as well as the cornea
cellular cells.

00:31:33.490 --> 00:31:35.240
They project through
the geniculate

00:31:35.240 --> 00:31:37.030
up to the visual cortex.

00:31:37.030 --> 00:31:38.545
And then from there,
there has been

00:31:38.545 --> 00:31:42.430
lots of debate as to what is
the nature of the connections

00:31:42.430 --> 00:31:45.060
to higher areas in the brain.

00:31:45.060 --> 00:31:47.082
And we talked about
that quite a bit.

00:31:47.082 --> 00:31:48.540
And some beautiful
studies had been

00:31:48.540 --> 00:31:52.260
carried out showing
that the input to area

00:31:52.260 --> 00:31:55.110
MT in the parietal
lobe is dominated

00:31:55.110 --> 00:31:58.130
by the parasol
system, but the input

00:31:58.130 --> 00:32:00.540
to V4 in the temporal
lobe is about

00:32:00.540 --> 00:32:02.760
equal for the two systems.

00:32:02.760 --> 00:32:05.110
So that was the
basic factor then.

00:32:05.110 --> 00:32:07.280
And so now the
question then comes up,

00:32:07.280 --> 00:32:09.620
what is the contribution
of these two

00:32:09.620 --> 00:32:11.670
systems, the midget
and the parasol?

00:32:11.670 --> 00:32:13.681
And so experiments
are carried out

00:32:13.681 --> 00:32:16.305
where lesions are made in either
parvocellular or magnocellular

00:32:16.305 --> 00:32:17.570
geniculate.

00:32:17.570 --> 00:32:22.441
And then the monkey was tested,
as I've said, in intact areas,

00:32:22.441 --> 00:32:24.440
in areas where the midget
system and areas where

00:32:24.440 --> 00:32:26.070
the parasol system
had been blocked.

00:32:27.590 --> 00:32:30.430
Now one additional
fact is that when

00:32:30.430 --> 00:32:33.030
you block both of
these by lesion

00:32:33.030 --> 00:32:35.850
in the lateral geniculate
nucleus, for the most part,

00:32:35.850 --> 00:32:37.930
the monkey becomes blind.

00:32:37.930 --> 00:32:40.350
OK, so these two
systems are really

00:32:40.350 --> 00:32:45.000
central for being able to
process visual information.

00:32:45.000 --> 00:32:49.380
All right so now, if one looks
at what kinds of deficits

00:32:49.380 --> 00:32:52.470
arise, a monkey can be
trained in a whole bunch

00:32:52.470 --> 00:32:53.960
of different tasks.

00:32:53.960 --> 00:32:55.420
It talked about these.

00:32:55.420 --> 00:32:57.900
Color vision, texture
perception, pattern perception,

00:32:57.900 --> 00:33:01.510
shape perception, brightness,
[INAUDIBLE] scotopic vision,

00:33:01.510 --> 00:33:04.680
contrast sensitivity,
stereopsis, motion perception,

00:33:04.680 --> 00:33:05.660
flicker perception.

00:33:05.660 --> 00:33:07.260
We'll talk about those first.

00:33:07.260 --> 00:33:11.890
So it was found that
there was severe deficits

00:33:11.890 --> 00:33:17.350
after a parvocellular lesion,
meaning when the midget

00:33:17.350 --> 00:33:20.690
system was blocked in color
vision, and texture perception,

00:33:20.690 --> 00:33:23.170
pattern perception,
and shape perception.

00:33:23.170 --> 00:33:26.690
Also in contrast sensitivity
and severe in stereopsis.

00:33:27.830 --> 00:33:34.180
None of those cause a deficit
with the magnocellular region,

00:33:34.180 --> 00:33:36.820
mean eliminating
the parasol system.

00:33:36.820 --> 00:33:41.680
But when examine motion
perception and flicker

00:33:41.680 --> 00:33:47.840
perception, there was a
moderate to major deficit

00:33:47.840 --> 00:33:50.470
where that system was missing.

00:33:50.470 --> 00:33:52.500
So that's then established.

00:33:52.500 --> 00:33:53.780
I'll come back to these later.

00:33:54.850 --> 00:33:58.980
Established at least
in some people's mind

00:33:58.980 --> 00:34:00.890
why these two
systems have emerged

00:34:00.890 --> 00:34:02.420
in the course of evolution.

00:34:02.420 --> 00:34:07.300
And so a summary statement
to that effect is shown here.

00:34:07.300 --> 00:34:11.670
If you look at the ability
to process spatial frequency

00:34:11.670 --> 00:34:13.659
by the midget and
parasol system,

00:34:13.659 --> 00:34:15.989
the midget system
can process it up

00:34:15.989 --> 00:34:17.935
to much higher
spatial frequencies.

00:34:19.710 --> 00:34:23.170
The obverse is the case when
it comes to temporal frequency.

00:34:23.170 --> 00:34:25.929
The parasol system can
process to much higher levels

00:34:25.929 --> 00:34:31.159
of rapid motion
or flicker, as you

00:34:31.159 --> 00:34:32.659
can see in this little diagram.

00:34:32.659 --> 00:34:35.380
So the midget system
extends the range

00:34:35.380 --> 00:34:39.719
of vision in the spatial
frequency wavelength range.

00:34:39.719 --> 00:34:41.560
And the parasol
system extends it

00:34:41.560 --> 00:34:43.510
in the temporal frequency range.

00:34:43.510 --> 00:34:46.120
So that's why these two
systems have evolved.

00:34:46.120 --> 00:34:49.100
And then if you look
at this in terms

00:34:49.100 --> 00:34:54.449
of the relative number of cells
in the retina that are devoted

00:34:54.449 --> 00:34:56.960
to these two attributes--
I told you that

00:34:56.960 --> 00:35:05.540
in the foveola, there no input
at all to the parasol system.

00:35:05.540 --> 00:35:07.940
So therefore, what
about this fine vision

00:35:07.940 --> 00:35:12.740
that the fovea makes possible
for you is due to the fact

00:35:12.740 --> 00:35:18.740
that area is dominated
by the midget system.

00:35:18.740 --> 00:35:21.430
Then as you go progressive
to the periphery,

00:35:21.430 --> 00:35:24.200
that ratio changes as
I had just shown you

00:35:24.200 --> 00:35:27.020
because increased
emphasis has to be placed

00:35:27.020 --> 00:35:29.955
a seeing motion and rapid
changes in the periphery.

00:35:31.080 --> 00:35:38.470
So that's what happens with
the parasol system's increased

00:35:38.470 --> 00:35:41.230
number of cells in
the periphery that

00:35:41.230 --> 00:35:43.150
can handle that requirement.

00:35:44.290 --> 00:35:46.450
So now we're going
to move on and talk

00:35:46.450 --> 00:35:51.670
about various aspects
of visual processing.

00:35:51.670 --> 00:35:54.405
And we'll start first with
color vision and adaptation.

00:35:56.090 --> 00:36:01.040
As I've shown you before, one
of the beautiful advances that

00:36:01.040 --> 00:36:03.410
had been made initially
actually, believe it or not,

00:36:03.410 --> 00:36:06.460
by Newton-- I mentioned
that I think--

00:36:06.460 --> 00:36:09.760
was the discovery of-- I
shouldn't say discovery--

00:36:09.760 --> 00:36:13.180
the invention on
the color circle.

00:36:13.180 --> 00:36:18.970
Now this invention arose in part
because it was established--

00:36:18.970 --> 00:36:21.140
it's a well-known
fact-- that we don't

00:36:21.140 --> 00:36:24.950
have opposites along these axes.

00:36:24.950 --> 00:36:28.450
You don't have a
yellowish blue color.

00:36:28.450 --> 00:36:30.930
You don't have a
reddish green color.

00:36:30.930 --> 00:36:33.920
But anything that's not an
opposite in this color circle,

00:36:33.920 --> 00:36:34.540
you do have.

00:36:35.550 --> 00:36:39.520
So you have yellowish red,
or you have yellowish green,

00:36:39.520 --> 00:36:40.780
and so on.

00:36:40.780 --> 00:36:45.180
So the color circle
was then elaborated

00:36:45.180 --> 00:36:47.020
upon over many years.

00:36:47.020 --> 00:36:49.340
This is a slightly
modified version

00:36:49.340 --> 00:36:52.310
from what Newton had invented.

00:36:52.310 --> 00:36:54.680
And this is set up
in such a fashion

00:36:54.680 --> 00:36:57.690
that when you go
around this circle--

00:36:57.690 --> 00:37:01.951
I should say disk I suppose--
you change the hue of course.

00:37:01.951 --> 00:37:04.200
And then when you go from
the center of the periphery,

00:37:04.200 --> 00:37:05.575
you increase saturation.

00:37:07.450 --> 00:37:10.510
This is not the perfect
display, especially

00:37:10.510 --> 00:37:13.580
because the projector
isn't perfect.

00:37:13.580 --> 00:37:16.510
But the center is
supposed to be white.

00:37:16.510 --> 00:37:19.140
And all this is
fairly equal luminent.

00:37:19.140 --> 00:37:21.560
And so you go from
unsaturated to saturated.

00:37:22.940 --> 00:37:26.990
Now I will say
already at this point

00:37:26.990 --> 00:37:32.250
another very important
factor in appreciating

00:37:32.250 --> 00:37:37.740
the beauty of the
color circle is

00:37:37.740 --> 00:37:42.250
that when you
analyze after images,

00:37:42.250 --> 00:37:49.814
it was found that if you adapt
to something that's yellow,

00:37:49.814 --> 00:37:51.355
you adapt the eye
to this wavelength.

00:37:52.550 --> 00:37:56.110
And then you shift
it to white, then

00:37:56.110 --> 00:37:58.880
you get an after-effect,
which is blue.

00:37:58.880 --> 00:38:01.830
And if you do that for red,
the after-effect is green.

00:38:01.830 --> 00:38:04.560
And the same thing is all
the way around the circle.

00:38:04.560 --> 00:38:07.010
If you have this one,
the after-effect is here.

00:38:07.010 --> 00:38:09.970
So the color circle
perfectly predicts

00:38:09.970 --> 00:38:16.870
what you're after images
are due to adaptation, which

00:38:16.870 --> 00:38:23.060
occurs as a result of
having bleached selectively

00:38:23.060 --> 00:38:31.530
the molecules in the various
cone types that we have,

00:38:31.530 --> 00:38:33.560
the three cones,
red, green, and blue.

00:38:33.560 --> 00:38:36.030
So that's the basic rule
of the color circle,

00:38:36.030 --> 00:38:37.545
which can be used extensively.

00:38:37.545 --> 00:38:38.920
And I think you
yourself can have

00:38:38.920 --> 00:38:43.560
a lot of fun studying this
in your off time, which you

00:38:43.560 --> 00:38:45.220
don't have too much of I'm sure.

00:38:45.220 --> 00:38:48.310
But it's really a wonderful
thing to play around with.

00:38:50.840 --> 00:38:55.840
Since this course is rather
heavily fact oriented,

00:38:55.840 --> 00:38:59.060
I want you to remember
these basic facts

00:38:59.060 --> 00:39:02.430
that I had listed before.

00:39:06.610 --> 00:39:09.520
I just noted along
the color circle,

00:39:09.520 --> 00:39:12.480
you have three attributes, hue,
brightness, and saturation.

00:39:14.780 --> 00:39:17.300
And then I also mentioned
to you that there

00:39:17.300 --> 00:39:21.230
is a distinction between
the psychological

00:39:21.230 --> 00:39:26.575
and the physical
attribute of images.

00:39:28.610 --> 00:39:34.310
And this arrangement is such
that I gave you an example of.

00:39:34.310 --> 00:39:40.050
For example, when a tree
falls in the forest,

00:39:40.050 --> 00:39:44.700
is there a sound when
there's nobody around?

00:39:45.920 --> 00:39:49.110
And the answer is
a distinctive no.

00:39:50.420 --> 00:39:51.160
Why?

00:39:51.160 --> 00:39:53.970
Because sound is a
psychological attribute.

00:39:53.970 --> 00:39:55.390
If you're on the
other hand, you'd

00:39:55.390 --> 00:39:59.400
have said well, if a
tree falls in the forest,

00:39:59.400 --> 00:40:03.430
through some
wavelength result that

00:40:03.430 --> 00:40:05.699
are in the range of hearing.

00:40:05.699 --> 00:40:06.740
And that, of course, yes.

00:40:07.900 --> 00:40:11.380
But if you say sound,
that's something you hear.

00:40:11.380 --> 00:40:14.400
It's not something
that's a physical thing.

00:40:14.400 --> 00:40:20.670
So that applies to
many aspects of vision,

00:40:20.670 --> 00:40:23.450
as well as audition
and many other senses

00:40:23.450 --> 00:40:26.890
that you must make a
distinction between what

00:40:26.890 --> 00:40:30.160
your psychological
disposition is as opposed

00:40:30.160 --> 00:40:32.040
to what a physical fact is.

00:40:34.280 --> 00:40:37.120
The next thing here
is that we have

00:40:37.120 --> 00:40:40.770
three types of photoreceptors,
a short, medium,

00:40:40.770 --> 00:40:43.920
and long wavelength
for the cones.

00:40:43.920 --> 00:40:46.740
And then we have also
a different wavelengths

00:40:46.740 --> 00:40:48.210
peak for the rods.

00:40:48.210 --> 00:40:52.460
All of these peaks
are broadly tuned

00:40:52.460 --> 00:40:57.480
to enable you to--
some are in the brain--

00:40:57.480 --> 00:40:59.837
examine the relative
amounts of information

00:40:59.837 --> 00:41:01.045
from these three wavelengths.

00:41:04.870 --> 00:41:07.230
That then enables
you to perceive

00:41:07.230 --> 00:41:13.150
many, many other colors partly
because of color [INAUDIBLE]

00:41:13.150 --> 00:41:16.180
and partly because of variety
of amount of activity of them.

00:41:16.180 --> 00:41:18.410
And then I mentioned
Grassman's laws.

00:41:18.410 --> 00:41:21.780
Every color has a
complementary, which when mixed

00:41:21.780 --> 00:41:23.920
properly yields gray.

00:41:23.920 --> 00:41:28.100
That should do with again
with the color circle.

00:41:28.100 --> 00:41:30.330
So in other words,
to go back to that,

00:41:30.330 --> 00:41:34.770
if you mix yellow and
blue in equal amounts,

00:41:34.770 --> 00:41:37.650
you get white or
gray I should say.

00:41:37.650 --> 00:41:40.050
And the same thing for
anything that's and opposite.

00:41:40.050 --> 00:41:43.480
But then if you mix
things which are not

00:41:43.480 --> 00:41:48.220
at diagonals to each other, then
you get an in between color.

00:41:48.220 --> 00:41:52.230
So if you mix yellow and
green, you get yellowish green.

00:41:52.230 --> 00:41:54.650
All right so that's
Grassman's laws.

00:41:54.650 --> 00:41:58.190
So if you have non complementary
colors, you get intermediate.

00:41:58.190 --> 00:42:01.460
And if you get complementary
colors, you get gray.

00:42:01.460 --> 00:42:04.700
Again to make sure that
you understand this,

00:42:04.700 --> 00:42:07.600
complementary means this
and that, this and that,

00:42:07.600 --> 00:42:16.020
this and that, which are on
the opposites on the lines that

00:42:16.020 --> 00:42:20.100
intersect the center
of the color circle.

00:42:22.700 --> 00:42:23.740
We move on.

00:42:23.740 --> 00:42:25.560
And we talk about Abney's law.

00:42:25.560 --> 00:42:27.450
That is not very important.

00:42:27.450 --> 00:42:28.980
And you don't have
to remember that.

00:42:28.980 --> 00:42:31.313
The luminance of a mixture
of differently colored lights

00:42:31.313 --> 00:42:33.950
is equal to the sum of the
luminance of its components.

00:42:33.950 --> 00:42:37.472
That another fact, but you
don't need to know that.

00:42:37.472 --> 00:42:39.620
The last thing that
I want to mention

00:42:39.620 --> 00:42:46.570
is so-called metamers, which
are stimuli which look the same

00:42:46.570 --> 00:42:52.050
but are the product of different
subcompositions of wavelengths.

00:42:52.050 --> 00:43:00.260
So because we only have three
different cone photoreceptors,

00:43:00.260 --> 00:43:01.710
you can, in a
sense, if you will,

00:43:01.710 --> 00:43:04.780
fool them a little bit by
very carefully mixing things

00:43:04.780 --> 00:43:08.810
up with different wavelengths
to activate them equally.

00:43:08.810 --> 00:43:11.450
So that is what is
called a metamer when

00:43:11.450 --> 00:43:15.330
you can't tell the difference
between two stimuli.

00:43:15.330 --> 00:43:18.310
They look identical even though
their wavelength compositions

00:43:18.310 --> 00:43:19.410
are different.

00:43:19.410 --> 00:43:23.620
OK so now another factor
that I should note here

00:43:23.620 --> 00:43:27.590
when it comes to color
vision is that when

00:43:27.590 --> 00:43:32.290
you look at the response
characteristics of cells

00:43:32.290 --> 00:43:36.630
in the retina-- and I'm talking
about the retinal ganglion

00:43:36.630 --> 00:43:39.470
cells-- when we look at the
cells in the geniculate, which

00:43:39.470 --> 00:43:46.820
is here, what you find actually
is just a few major categories.

00:43:46.820 --> 00:43:49.100
This is a color circle here.

00:43:49.100 --> 00:43:52.090
And one presents the
stimuli around the circle

00:43:52.090 --> 00:43:54.640
and see how the cell responds.

00:43:54.640 --> 00:43:56.590
And what you're see
here is one cell,

00:43:56.590 --> 00:44:01.860
which is a blue on cell, a green
off cell, a yellow on cell,

00:44:01.860 --> 00:44:03.080
and a green on cell.

00:44:04.310 --> 00:44:07.290
Now it turns out if you record
from hundreds and hundreds

00:44:07.290 --> 00:44:09.780
of cells, you only
get these categories.

00:44:09.780 --> 00:44:13.410
You don't ever get any cells
which are at the diagonals.

00:44:13.410 --> 00:44:16.050
So to see the
diagonals, something

00:44:16.050 --> 00:44:20.420
has to be taking place in the
cortex on the basis of what

00:44:20.420 --> 00:44:24.250
is coming in from the retina and
the lateral geniculate nucleus.

00:44:26.510 --> 00:44:30.870
Now when you come to adaptation,
we talk about that quite a bit.

00:44:30.870 --> 00:44:33.740
And also with after images, I'll
come back to that in a minute.

00:44:34.930 --> 00:44:37.720
It was discovered in
some very nice example

00:44:37.720 --> 00:44:40.520
that you take a cell,
and you adapt it

00:44:40.520 --> 00:44:43.710
to various levels of
overall illumination,

00:44:43.710 --> 00:44:45.470
and then see how the
cell responds to it,

00:44:45.470 --> 00:44:48.100
and then you stimulate
the receptive field.

00:44:48.100 --> 00:44:51.590
What you find is that
here's the same cell.

00:44:51.590 --> 00:44:56.210
Here's a background illumination
of a huge range over five log

00:44:56.210 --> 00:44:56.710
units.

00:44:58.080 --> 00:45:00.680
And what you find is that when
most of the cell is adapted,

00:45:00.680 --> 00:45:02.910
it responds always the same.

00:45:02.910 --> 00:45:07.410
So it's looking not at over
all levels of illumination,

00:45:07.410 --> 00:45:10.039
it's looking at differences
in illumination.

00:45:10.039 --> 00:45:11.080
It's looking at contrast.

00:45:14.970 --> 00:45:18.380
Now how many of you remember
the formula for contrast?

00:45:20.491 --> 00:45:20.990
Anybody?

00:45:22.974 --> 00:45:26.805
All right, I think that's a
really good thing to remember.

00:45:26.805 --> 00:45:28.430
I'm sure when you go
to a party, people

00:45:28.430 --> 00:45:29.950
would be fascinated
by you knowing

00:45:29.950 --> 00:45:31.375
the formula for contrast.

00:45:32.485 --> 00:45:36.370
OK, so contrast
equals-- you take

00:45:36.370 --> 00:45:38.885
the stimulus,
which is call it x.

00:45:41.160 --> 00:45:44.820
And you take the background,
which you called y.

00:45:44.820 --> 00:45:47.010
You subtract one from the other.

00:45:47.010 --> 00:45:48.510
Then you add the two up.

00:45:50.380 --> 00:45:53.700
And then you
multiply this by 100.

00:45:53.700 --> 00:45:55.290
So that your contrast.

00:45:55.290 --> 00:45:59.030
What this formula
means that this applies

00:45:59.030 --> 00:46:02.630
to endless levels of
overall illumination.

00:46:02.630 --> 00:46:04.020
You can do this in the sunshine.

00:46:04.020 --> 00:46:06.680
You can do this in the
moonshine because you're

00:46:06.680 --> 00:46:09.770
looking at the differences
between the background

00:46:09.770 --> 00:46:13.025
and the stimulus itself.

00:46:16.566 --> 00:46:17.915
AUDIENCE: What is x and y?

00:46:20.190 --> 00:46:22.730
PROFESSOR: As I've said, x
is the illumination level

00:46:22.730 --> 00:46:24.795
for the target.

00:46:25.880 --> 00:46:34.290
Suppose you take a cell, and
you shine a spot of light on it

00:46:34.290 --> 00:46:35.130
like that.

00:46:36.960 --> 00:46:43.990
Then you remove it, and you
measure the background level.

00:46:43.990 --> 00:46:46.567
And so x is a visual stimulus.

00:46:46.567 --> 00:46:47.400
y is the background.

00:47:03.040 --> 00:47:05.330
We talk about light adaptation.

00:47:05.330 --> 00:47:07.865
Again, I want you to
know a few basic facts.

00:47:09.010 --> 00:47:11.500
The overall level
of illumination

00:47:11.500 --> 00:47:12.915
is close to 10 log units.

00:47:14.730 --> 00:47:17.460
But in contrast to
that, if you just

00:47:17.460 --> 00:47:19.860
look at reflected
light, that varies

00:47:19.860 --> 00:47:22.930
over a much smaller
range because on the very

00:47:22.930 --> 00:47:25.810
bright illumination
conditions, even a black object

00:47:25.810 --> 00:47:27.120
will reflect some light.

00:47:27.120 --> 00:47:31.000
So you're talking about direct
light versus reflected light.

00:47:31.000 --> 00:47:34.270
So when you do reflected
light, you get a smaller range.

00:47:34.270 --> 00:47:38.400
Now the pupil plays a role
in the amount of light

00:47:38.400 --> 00:47:41.550
it controls getting
into the eye.

00:47:41.550 --> 00:47:44.800
But it can only do that
over a range of 16 to one.

00:47:44.800 --> 00:47:48.200
Now because of that, the
major role of adaptation

00:47:48.200 --> 00:47:53.660
has to do with
the photoreceptors

00:47:53.660 --> 00:47:55.720
in your rods and your cones.

00:47:55.720 --> 00:47:58.270
And the way that works
is, if you remember,

00:47:58.270 --> 00:48:03.660
is that you can think
of your molecules

00:48:03.660 --> 00:48:07.785
in your photoreceptors as
existing in two forms, bleached

00:48:07.785 --> 00:48:08.410
and unbleached.

00:48:09.500 --> 00:48:11.260
And because of the
millions and millions

00:48:11.260 --> 00:48:13.700
of them that I told
you already about,

00:48:13.700 --> 00:48:19.450
1,000 times 10,000 in
just a single cone,

00:48:19.450 --> 00:48:24.670
there's a relative percentage
of bleached and unbleached

00:48:24.670 --> 00:48:29.085
molecules in each
cone and in each rod.

00:48:31.050 --> 00:48:36.170
And so what is happening is
that during dark adaptation,

00:48:36.170 --> 00:48:38.070
there's a huge increase
in the unbleached,

00:48:38.070 --> 00:48:39.760
and during light
adaptation, there's

00:48:39.760 --> 00:48:41.900
an increase in the
bleached molecules.

00:48:43.590 --> 00:48:47.450
So therefore, any
increase in the rate

00:48:47.450 --> 00:48:49.840
of at which quanta are
delivered to the eye

00:48:49.840 --> 00:48:51.780
is also in the
proportion of decrease

00:48:51.780 --> 00:48:53.500
in the number of
pigment molecules

00:48:53.500 --> 00:48:56.020
available to absorb
those quanta.

00:48:56.020 --> 00:48:58.980
Retina ganglion
cells are selected

00:48:58.980 --> 00:49:01.360
sensitive to local
contrast differences

00:49:01.360 --> 00:49:03.470
not absolute levels
of illumination.

00:49:03.470 --> 00:49:05.720
I've said that many
times over again.

00:49:05.720 --> 00:49:09.980
OK, and that's why this
formula, this contrast formula,

00:49:09.980 --> 00:49:16.110
is one that's the most useful
in being able to depict

00:49:16.110 --> 00:49:20.460
what kind of input
these cells are getting.

00:49:20.460 --> 00:49:25.030
So that then is the arrangement
about light adaptation.

00:49:26.720 --> 00:49:30.550
Now let's move on to depth
perception, which is one

00:49:30.550 --> 00:49:35.410
the most intriguing capacities
that we have since our retinae

00:49:35.410 --> 00:49:37.750
essentially are like a
two dimensional surface.

00:49:37.750 --> 00:49:40.170
So whatever comes
onto the retina,

00:49:40.170 --> 00:49:43.920
some mechanisms have
to be able to tell you

00:49:43.920 --> 00:49:46.210
where things are in depth.

00:49:46.210 --> 00:49:48.630
And because it's such
a complex problem,

00:49:48.630 --> 00:49:50.910
quite a number of
different mechanisms

00:49:50.910 --> 00:49:54.490
have evolved to make it
possible for you to do that.

00:49:54.490 --> 00:49:58.330
And that means that first
we have oculomotor cues.

00:49:58.330 --> 00:50:00.210
We don't need to
talk about those.

00:50:00.210 --> 00:50:02.750
But we have visual cues,
which have binocular cues,

00:50:02.750 --> 00:50:04.810
stereopsis we talked
about quite a bit.

00:50:04.810 --> 00:50:08.080
And the binocular cues are
motion parallax, shading,

00:50:08.080 --> 00:50:10.810
interposition size,
and perspective.

00:50:10.810 --> 00:50:13.510
All these cues we can
utilize to tell us

00:50:13.510 --> 00:50:14.935
where things are in depth.

00:50:15.940 --> 00:50:19.890
So now if we look
at stereopsis--

00:50:19.890 --> 00:50:22.430
I've handed out to you some
of these autostereograms.

00:50:23.770 --> 00:50:26.250
If you look at these, you
can't see it looking at that.

00:50:26.250 --> 00:50:28.170
You have to do it
on those sheets

00:50:28.170 --> 00:50:29.600
that I handed out to you.

00:50:29.600 --> 00:50:31.550
You can see something in depth.

00:50:31.550 --> 00:50:34.510
And this arises by
virtue of the fact

00:50:34.510 --> 00:50:38.330
that stimuli are arranged
in such a fashion

00:50:38.330 --> 00:50:44.020
that they selectively activate
neurons in the visual cortex

00:50:44.020 --> 00:50:46.790
that code depth by
virtue of the fact

00:50:46.790 --> 00:50:50.550
that they get disparity
inputs from the two eyes.

00:50:50.550 --> 00:50:54.330
Now another central
mechanism-- I

00:50:54.330 --> 00:50:56.747
should add one more
thing about stereopsis.

00:50:56.747 --> 00:50:57.830
I think I've mentioned it.

00:50:57.830 --> 00:51:02.360
In fact, 10% of the population
in the United States lack

00:51:02.360 --> 00:51:09.590
stereopsis in most cases due to
either misalignment of the two

00:51:09.590 --> 00:51:16.320
eyes or do to ambliopia,
meaning one eye doesn't see

00:51:16.320 --> 00:51:18.160
as well as the other.

00:51:18.160 --> 00:51:22.510
But those people can
still do many things

00:51:22.510 --> 00:51:24.370
and do depth quite well.

00:51:24.370 --> 00:51:26.730
They can't thread
needles, but they

00:51:26.730 --> 00:51:29.420
can do courser depth quite well.

00:51:29.420 --> 00:51:32.010
And one of those is
due to motion parallax.

00:51:32.010 --> 00:51:34.520
Now the basic rule
about motion parallax

00:51:34.520 --> 00:51:37.840
that cause the brain
to evolve to analyze it

00:51:37.840 --> 00:51:39.980
is that when objects
are different distances

00:51:39.980 --> 00:51:42.730
from the eyes as depicted
here, the objects that

00:51:42.730 --> 00:51:44.950
are closer to the
eye when this object

00:51:44.950 --> 00:51:49.940
moves over a greater range
on the retinal surface

00:51:49.940 --> 00:51:51.400
than those that
are further apart.

00:51:51.400 --> 00:51:55.590
You can see the
green versus the red.

00:51:55.590 --> 00:52:03.470
So therefore, the
system is such that it

00:52:03.470 --> 00:52:06.920
has evolved to be able
to see small differences

00:52:06.920 --> 00:52:11.130
in the relative motion of
objects in the retinal surface.

00:52:11.130 --> 00:52:13.530
And I showed you
an example of that.

00:52:13.530 --> 00:52:16.660
And I'll show it to you
again because this is fun.

00:52:16.660 --> 00:52:20.440
This is essentially similar
to the random autostereogram

00:52:20.440 --> 00:52:23.730
except it's just a single
bunch of random dots.

00:52:23.730 --> 00:52:25.810
And as soon as I
set this in motion,

00:52:25.810 --> 00:52:29.630
you see them in three dimensions
beautifully because these move

00:52:29.630 --> 00:52:32.490
over a greater range than these.

00:52:32.490 --> 00:52:34.500
And these move even less so.

00:52:34.500 --> 00:52:38.440
So this differential
motion commands

00:52:38.440 --> 00:52:39.760
you to see it in depth.

00:52:41.170 --> 00:52:43.900
So that's quite a
remarkable ability.

00:52:43.900 --> 00:52:47.880
And monkeys are even
better at it than we are.

00:52:47.880 --> 00:52:51.630
And even fish have
this kind of capacity

00:52:51.630 --> 00:52:53.440
as do many, many other species.

00:52:53.440 --> 00:52:57.150
It's so central to our
ability to process depth.

00:52:57.150 --> 00:53:00.660
OK

00:53:00.660 --> 00:53:03.910
Studies have been
carried out to determine

00:53:03.910 --> 00:53:08.760
where and how
these are analyzed.

00:53:08.760 --> 00:53:10.650
And when we came
to where, here's

00:53:10.650 --> 00:53:14.350
an example of looking at a
brain in a normal stereo-blind

00:53:14.350 --> 00:53:14.850
subject.

00:53:16.020 --> 00:53:18.690
When you only present
motion parallax,

00:53:18.690 --> 00:53:19.970
you only presents stereo.

00:53:21.210 --> 00:53:24.040
And when you do the
stereo, monocularly, you

00:53:24.040 --> 00:53:27.370
don't see depth, and
the brain is not active.

00:53:27.370 --> 00:53:29.410
So this tells you
which part of the brain

00:53:29.410 --> 00:53:35.110
is active and involved
in analysis of stereopsis

00:53:35.110 --> 00:53:38.850
and which one is involved in
the analysis of motion parallax.

00:53:38.850 --> 00:53:41.800
I showed you this picture and
several others telling you

00:53:41.800 --> 00:53:43.240
which areas it is.

00:53:43.240 --> 00:53:47.520
The limitation of that
is that it can tell you

00:53:47.520 --> 00:53:49.030
where it takes
place in the brain,

00:53:49.030 --> 00:53:51.460
but it doesn't tell
you how it takes place.

00:53:51.460 --> 00:53:53.150
So because of that,
many studies have

00:53:53.150 --> 00:53:55.040
been carried out
doing single cell

00:53:55.040 --> 00:53:58.630
recordings in these
cortical areas.

00:53:58.630 --> 00:54:00.990
And it was discovered
that the there are indeed

00:54:00.990 --> 00:54:03.730
cells already in
area 17 that get

00:54:03.730 --> 00:54:05.345
disparate input
from the two eyes.

00:54:06.840 --> 00:54:11.810
Beautiful work by
[INAUDIBLE] showing this.

00:54:11.810 --> 00:54:15.930
And establish therefore
that already in area 17,

00:54:15.930 --> 00:54:19.105
you have neurons that tell
you about stereoscopic depth.

00:54:20.520 --> 00:54:24.635
And then it was also discovered,
especially in area MT,

00:54:24.635 --> 00:54:27.370
to a lesser extent
already in V1 also,

00:54:27.370 --> 00:54:31.630
that you have cells that
respond to differential motion.

00:54:31.630 --> 00:54:33.400
And so those cells
are presumptively

00:54:33.400 --> 00:54:37.820
involved in the processing
of depth information

00:54:37.820 --> 00:54:42.100
based on motion parallax.

00:54:42.100 --> 00:54:44.000
Now another mechanism
involved that we

00:54:44.000 --> 00:54:46.790
talked about is shading.

00:54:46.790 --> 00:54:50.510
Light coming from
above like from the sun

00:54:50.510 --> 00:54:55.580
had been incorporated
into the visual system

00:54:55.580 --> 00:55:01.660
to tell you whether an object
is towards you or away from you.

00:55:01.660 --> 00:55:03.310
And this is an example of that.

00:55:03.310 --> 00:55:06.950
Here the light is from above,
and the darkness is below.

00:55:06.950 --> 00:55:08.390
This is reversed here.

00:55:08.390 --> 00:55:10.760
And because of that, you
see this as protruding,

00:55:10.760 --> 00:55:12.640
and you see this as receding.

00:55:12.640 --> 00:55:16.945
And so I showed you several
examples, some in the handout,

00:55:16.945 --> 00:55:22.200
of the fact that even shading
is a hue that's used quite

00:55:22.200 --> 00:55:28.540
extensively in depth perception.

00:55:29.570 --> 00:55:31.390
Now we come to form perception.

00:55:31.390 --> 00:55:33.290
I'll talk about this briefly.

00:55:33.290 --> 00:55:35.610
I mentioned three
kinds of theories.

00:55:35.610 --> 00:55:38.790
One is that the former
is due to the fact

00:55:38.790 --> 00:55:40.670
that neurons respond
selectively to line

00:55:40.670 --> 00:55:43.340
segments of different
orientations in V1.

00:55:43.340 --> 00:55:47.170
Another theory was that
they have a spatial mapping

00:55:47.170 --> 00:55:49.760
of the stimuli on to the
visual cortex since you

00:55:49.760 --> 00:55:50.650
have topography.

00:55:50.650 --> 00:55:53.390
And the third one is
that form perceptions

00:55:53.390 --> 00:55:55.500
are accomplished by
Fourier analysis.

00:55:55.500 --> 00:55:57.170
We talked about each of these.

00:55:57.170 --> 00:56:00.020
And I pointed out to
you that even when

00:56:00.020 --> 00:56:03.345
there are no orientation
segments in the display,

00:56:03.345 --> 00:56:08.340
you can still see and
identify faces quite well,

00:56:08.340 --> 00:56:13.490
as seen in the Wall
Street Journal where

00:56:13.490 --> 00:56:18.780
these kinds of pictures appear
every day in the paper itself.

00:56:18.780 --> 00:56:21.360
Now then if you move
on, and you look

00:56:21.360 --> 00:56:24.710
at the layout of how the
cortexes-- this is a monkey

00:56:24.710 --> 00:56:25.340
cortex here.

00:56:25.340 --> 00:56:26.435
This is a visual field.

00:56:27.480 --> 00:56:30.760
If you present these three
stimuli in the visual field,

00:56:30.760 --> 00:56:33.600
this is the area that's
activated in the cortex

00:56:33.600 --> 00:56:37.140
because more area is
allocated to central vision

00:56:37.140 --> 00:56:38.210
and peripheral vision.

00:56:38.210 --> 00:56:40.461
And so you say oh, this
is much bigger than those.

00:56:40.461 --> 00:56:41.460
But that's not the case.

00:56:41.460 --> 00:56:42.959
You can tell that
they're identical.

00:56:43.680 --> 00:56:45.440
Now even more
dramatic is the fact

00:56:45.440 --> 00:56:50.320
that if you put these three
disks centered-- OK, so half

00:56:50.320 --> 00:56:55.080
of it goes to the
ipsilater and half of it

00:56:55.080 --> 00:56:58.290
to the contralateral
visual hemisphere.

00:56:58.290 --> 00:57:01.010
What you get are a bunch
of half circles like that.

00:57:01.010 --> 00:57:03.200
And it doesn't look
anything like that.

00:57:03.200 --> 00:57:07.070
So the idea that somehow
images are laid down

00:57:07.070 --> 00:57:09.580
in the visual cortex, and
the mind then looks at it

00:57:09.580 --> 00:57:10.700
is totally wrong.

00:57:10.700 --> 00:57:12.920
It's wrong to the extent
that it's ridiculous.

00:57:14.160 --> 00:57:20.310
The last analysis theory
is accepted by some people.

00:57:20.310 --> 00:57:26.290
And doing computer analyses has
revealed that system actually

00:57:26.290 --> 00:57:28.670
can be mimicked
extremely well based

00:57:28.670 --> 00:57:30.750
on what we know about
the organization

00:57:30.750 --> 00:57:35.455
of the visual cortex.

00:57:36.990 --> 00:57:38.950
It has all the basic
attributes that you need,

00:57:38.950 --> 00:57:41.840
orientation,
direction selectivity,

00:57:41.840 --> 00:57:44.990
and phase that enable
you to break down

00:57:44.990 --> 00:57:48.710
the visual scene in an
analytical fashion, which

00:57:48.710 --> 00:57:50.891
is kind of foreign to our
thinking, namely Fourier

00:57:50.891 --> 00:57:51.390
analysis.

00:57:53.570 --> 00:57:57.670
Then we spent some time talking
about prosthesis, which you're

00:57:57.670 --> 00:57:59.380
going to hear quite
a bit about actually

00:57:59.380 --> 00:58:03.570
when Chris Brown is going to
lecture because that has been

00:58:03.570 --> 00:58:09.030
so successful in
the auditory system

00:58:09.030 --> 00:58:13.980
with the cochlear implant, which
is a remarkable achievement.

00:58:13.980 --> 00:58:17.780
We have many more
than 50,000 by now

00:58:17.780 --> 00:58:20.240
in the United States who
have cochlear implants.

00:58:20.240 --> 00:58:23.034
And they can talk and do all
kinds of remarkable things.

00:58:23.034 --> 00:58:24.700
We don't have this
in the visual system.

00:58:25.950 --> 00:58:28.600
And I've mentioned to you that
one of the big differences

00:58:28.600 --> 00:58:32.450
is that in the retina, there
are more than a million fibers

00:58:32.450 --> 00:58:36.780
in each eye that come from the
ganglion cells that project

00:58:36.780 --> 00:58:43.130
into the brain, whereas when you
talk about the auditory system,

00:58:43.130 --> 00:58:46.260
you only have about
30,000 fibers.

00:58:46.260 --> 00:58:49.960
So the magnitude is much less.

00:58:49.960 --> 00:58:51.400
But also there
are other factors,

00:58:51.400 --> 00:58:54.060
namely the retina is a
very difficult structure

00:58:54.060 --> 00:58:55.190
to work with.

00:58:55.190 --> 00:59:00.580
And also when people become
blind, most cases the retina

00:59:00.580 --> 00:59:01.520
degenerates.

00:59:01.520 --> 00:59:06.430
So you can't put a
device into the eye

00:59:06.430 --> 00:59:11.580
very effectively in most
blind people to create vision.

00:59:11.580 --> 00:59:15.090
So another alternative is
to try some other regions.

00:59:15.090 --> 00:59:18.360
Some people have advocated to
do this in the visual cortex.

00:59:18.360 --> 00:59:23.640
And the problem there is we have
the huge magnification factor.

00:59:23.640 --> 00:59:27.940
So if you put 256 stimuli
like this in the visual scene,

00:59:27.940 --> 00:59:29.825
this is the actual
physical activation.

00:59:31.420 --> 00:59:33.570
And once you know
what this layout is,

00:59:33.570 --> 00:59:37.180
then you can put electrodes
in, which are spaced like this.

00:59:37.180 --> 00:59:39.140
Then, if you were
to stimulate these,

00:59:39.140 --> 00:59:42.480
then you would create an image,
which is at least moderately

00:59:42.480 --> 00:59:46.480
similar to this that would
be in slightly different

00:59:46.480 --> 00:59:49.480
washed out colors, but which
would still essentially

00:59:49.480 --> 00:59:50.660
be a square.

00:59:50.660 --> 00:59:54.370
So if you do that then,
and you take a camera,

00:59:54.370 --> 00:59:58.180
and you take the
input to the camera

00:59:58.180 --> 01:00:04.130
and connect it selectively
to this proportional implant.

01:00:04.130 --> 01:00:07.060
If you put the word, fiat
lux-- remember what that is?

01:00:07.060 --> 01:00:08.440
Let there be light.

01:00:08.440 --> 01:00:10.440
You get a pretty
good reproduction

01:00:10.440 --> 01:00:12.060
of what has been put in there.

01:00:12.060 --> 01:00:19.560
But by contrast, if you take
a ray of electrodes, which

01:00:19.560 --> 01:00:23.020
are equally spaced-- then,
if you activated all those,

01:00:23.020 --> 01:00:24.670
you would get a butterfly image.

01:00:24.670 --> 01:00:29.200
And if you then put in the
fiat lux to the camera,

01:00:29.200 --> 01:00:30.900
it would look like that.

01:00:30.900 --> 01:00:32.460
So that would mean
that they would

01:00:32.460 --> 01:00:34.470
get a pretty false
impression of the world,

01:00:34.470 --> 01:00:36.390
and you wouldn't be
able to even read.

01:00:36.390 --> 01:00:40.990
So therefore, it would be very
important to take into account

01:00:40.990 --> 01:00:43.520
the functioning of
the visual system,

01:00:43.520 --> 01:00:48.040
as well as functioning
of the individual neurons

01:00:48.040 --> 01:00:50.920
if you are going to create
a prosthetic device.

01:00:50.920 --> 01:00:53.120
So now we will move on.

01:00:53.120 --> 01:00:55.635
And I will say a few
words about illusions.

01:00:56.637 --> 01:00:58.220
We talked about quite
a few illusions,

01:00:58.220 --> 01:01:00.190
and you got some of
those in the hand out.

01:01:00.190 --> 01:01:04.130
The one I mentioned to you
that I think all of you enjoyed

01:01:04.130 --> 01:01:05.990
is the Hermann grid
illusion, which

01:01:05.990 --> 01:01:08.840
shows the smudges at the
smudges at the intersections.

01:01:08.840 --> 01:01:12.470
And the famous theory that
was advanced by Baumgartner

01:01:12.470 --> 01:01:16.770
is that it's due to the fact
that if you have a cell that

01:01:16.770 --> 01:01:18.720
is centered around
here, as opposed

01:01:18.720 --> 01:01:22.120
to not at the intersections,
this cell would

01:01:22.120 --> 01:01:24.150
be inhibited more than this one.

01:01:24.150 --> 01:01:31.070
So this hypothesis had been
accepted by many people

01:01:31.070 --> 01:01:34.660
a few years back,
and it has appeared

01:01:34.660 --> 01:01:36.430
in many, many textbooks.

01:01:36.430 --> 01:01:38.520
It turns out this
theory is all wrong

01:01:38.520 --> 01:01:41.150
if you remember
because first of all,

01:01:41.150 --> 01:01:43.780
here you just make
a small change

01:01:43.780 --> 01:01:48.380
in the physical
layout of the lines.

01:01:48.380 --> 01:01:50.270
And you don't get
the effect at all,

01:01:50.270 --> 01:01:56.270
even though if you put a cell
here, as opposed to here,

01:01:56.270 --> 01:01:58.110
the arrangement
is still the same.

01:01:58.110 --> 01:02:00.950
So consequently,
that theory is wrong.

01:02:00.950 --> 01:02:04.110
And it's even further proven by
the fact that when you analyze

01:02:04.110 --> 01:02:08.630
it physically to see what the
number of cells is in this area

01:02:08.630 --> 01:02:11.360
here-- and this is
for parasol cells.

01:02:11.360 --> 01:02:13.060
And this is for
the midget cells.

01:02:13.060 --> 01:02:14.830
You have a huge number of cells.

01:02:14.830 --> 01:02:16.810
And this is shown
only for the on cells.

01:02:16.810 --> 01:02:20.900
You can double that
for the off cells.

01:02:20.900 --> 01:02:23.400
You would activate
this teeny area here,

01:02:23.400 --> 01:02:26.380
five degrees from the fixation.

01:02:26.380 --> 01:02:32.690
You would activate 365 midget
cells and 50 parasol cells,

01:02:32.690 --> 01:02:35.450
half of which would be on and
half of which would be off.

01:02:35.450 --> 01:02:40.270
So this theory is
just incorrect.

01:02:40.270 --> 01:02:43.330
And so alternative ideas
have been developed still

01:02:43.330 --> 01:02:44.790
sort of under debate.

01:02:44.790 --> 01:02:47.910
And one is that this takes place
because of the simple cells

01:02:47.910 --> 01:02:50.107
in the visual cortex as
we have talked about it.

01:02:52.970 --> 01:02:58.570
Now then, another set of
illusions we talked about

01:02:58.570 --> 01:03:00.540
are the after-effect illusions.

01:03:00.540 --> 01:03:04.490
And the experiments that
we so informally did here

01:03:04.490 --> 01:03:07.390
asked if you look at
a particular display,

01:03:07.390 --> 01:03:10.597
you fix it for a while, and
then you change the display,

01:03:10.597 --> 01:03:11.930
you have an after-effect, right?

01:03:11.930 --> 01:03:13.600
A very dramatic after-effect.

01:03:13.600 --> 01:03:18.325
And one of those was the
rotating dots in the circle.

01:03:21.330 --> 01:03:22.970
And I showed it to you.

01:03:22.970 --> 01:03:26.700
The experiment was that you
adapt to it with one eye,

01:03:26.700 --> 01:03:30.170
and then you look at the display
afterwards with the other eye,

01:03:30.170 --> 01:03:32.070
and then you would
have no effect, which

01:03:32.070 --> 01:03:34.980
proves that this takes
place in the retina

01:03:34.980 --> 01:03:38.317
and proves it is due
to the adaptation that

01:03:38.317 --> 01:03:39.733
takes place in the
photoreceptors.

01:03:41.090 --> 01:03:44.170
All right so those were
the so-called interlocking

01:03:44.170 --> 01:03:45.475
experiments we had discussed.

01:03:46.900 --> 01:03:50.580
So now let me move
on and talk some more

01:03:50.580 --> 01:04:01.010
about the deficits in vision
arise as a function of lesions.

01:04:02.300 --> 01:04:04.010
And I already showed
you a whole set

01:04:04.010 --> 01:04:08.650
of those when we talked
about the lesions

01:04:08.650 --> 01:04:11.390
of the midget and
parasol systems.

01:04:11.390 --> 01:04:14.490
And now if you look at
this in more detail,

01:04:14.490 --> 01:04:17.400
we add to this,
what happens when

01:04:17.400 --> 01:04:20.480
you remove V4 and remove MT?

01:04:20.480 --> 01:04:24.030
And it's quite striking
that the deficits are far,

01:04:24.030 --> 01:04:31.425
far less when you take
out the midget system.

01:04:32.520 --> 01:04:35.830
You have very mild
deficits with V4 lesions

01:04:35.830 --> 01:04:37.420
for most of these up here.

01:04:37.420 --> 01:04:39.650
These are basic
visual capacities.

01:04:41.740 --> 01:04:45.430
But MT lesions do give you
pretty much the same deficits

01:04:45.430 --> 01:04:51.840
as a magnocellular lesion that
blocks the parasol system.

01:04:51.840 --> 01:04:53.810
Now then when
higher level visual

01:04:53.810 --> 01:04:56.740
capacities we have now
analyzed-- I showed you those

01:04:56.740 --> 01:04:58.950
as well-- you found
that there was

01:04:58.950 --> 01:05:06.370
some dramatic deficits with
V4 lesions when monkeys

01:05:06.370 --> 01:05:13.230
had to choose less a stimuli and
had to learn visual percepts,

01:05:13.230 --> 01:05:17.160
they had severe deficits
with a V4 lesion.

01:05:17.160 --> 01:05:21.490
So that suggests
that an area like V4

01:05:21.490 --> 01:05:24.410
plays a very important role in
higher level visual processing.

01:05:24.410 --> 01:05:25.770
Yes?

01:05:25.770 --> 01:05:27.710
AUDIENCE: What does
the pronounced mean?

01:05:27.710 --> 01:05:28.284
Is that--

01:05:28.284 --> 01:05:29.200
PROFESSOR: Pronounced?

01:05:29.200 --> 01:05:31.340
It means like a strong deficit.

01:05:31.340 --> 01:05:32.980
AUDIENCE: So more than severe?

01:05:32.980 --> 01:05:33.730
PROFESSOR: No, no.

01:05:35.770 --> 01:05:37.860
You can see by the color also.

01:05:37.860 --> 01:05:39.380
Severe is the strongest.

01:05:39.380 --> 01:05:41.170
Pronounced is strong.

01:05:41.170 --> 01:05:42.720
Moderate is weaker.

01:05:42.720 --> 01:05:48.430
And mild is mild. [CHUCKLES]

01:05:48.430 --> 01:05:52.745
OK so now, next I want to
turn to eye movement control.

01:05:53.970 --> 01:05:57.680
And when we do that
I want to remind you

01:05:57.680 --> 01:06:01.460
that the many cortical areas as
well as the subcortical areas

01:06:01.460 --> 01:06:05.570
that play a significant role
in eye movement control.

01:06:05.570 --> 01:06:08.370
And one way to test this is
to electrically stimulate

01:06:08.370 --> 01:06:10.430
various regions in
the brain and see

01:06:10.430 --> 01:06:12.110
if you get any eye movements.

01:06:12.110 --> 01:06:13.830
And this happens in many areas.

01:06:13.830 --> 01:06:17.740
The ones we have here are
superior colliculus, of course,

01:06:17.740 --> 01:06:18.830
we talked a lot about.

01:06:18.830 --> 01:06:22.750
V1, LIP, the medial eye fields,
and the frontal eye fields.

01:06:24.230 --> 01:06:28.630
Now in all but one
of these you get

01:06:28.630 --> 01:06:31.977
a constant vector [INAUDIBLE] at
any sight where you stimulate,

01:06:31.977 --> 01:06:34.560
meaning no matter where the eye
is looking when you stimulate,

01:06:34.560 --> 01:06:38.700
you've got a particular
vector as depicted here.

01:06:38.700 --> 01:06:42.160
The exception to that
is the medial eye fields

01:06:42.160 --> 01:06:43.880
where you have a place code.

01:06:43.880 --> 01:06:47.200
The result of stimulating
any given area

01:06:47.200 --> 01:06:49.790
is to bring the eye,
normally where the eye is,

01:06:49.790 --> 01:06:51.485
into that motor field.

01:06:52.510 --> 01:06:56.310
Now different regions
obviously in these areas

01:06:56.310 --> 01:06:57.880
have different vectors.

01:06:57.880 --> 01:06:59.520
So that's the basic layout.

01:06:59.520 --> 01:07:05.200
And then the question arose how
do these get down to the brain

01:07:05.200 --> 01:07:10.660
stem ocularmotor complex that
drives the eye muscles that we

01:07:10.660 --> 01:07:11.650
had talked about.

01:07:11.650 --> 01:07:14.110
Well, the way the
experiment was done then

01:07:14.110 --> 01:07:16.470
is to remove the
superior colliculus.

01:07:16.470 --> 01:07:18.670
And when that was
done, what you found

01:07:18.670 --> 01:07:21.470
was really quite
dramatic, namely

01:07:21.470 --> 01:07:23.820
that you could no
longer drive cells

01:07:23.820 --> 01:07:25.870
from the posterior
part of the cortex,

01:07:25.870 --> 01:07:28.710
but you could still drive
them from the anterior part.

01:07:28.710 --> 01:07:32.390
This led to the idea that
you have two major systems

01:07:32.390 --> 01:07:36.500
in saccadic eye
movement generation,

01:07:36.500 --> 01:07:39.120
the so-called posterior system
and the anterior system.

01:07:41.610 --> 01:07:45.567
And then when people looked
at the question of well,

01:07:45.567 --> 01:07:46.650
we have these two systems.

01:07:46.650 --> 01:07:47.730
What do they do?

01:07:47.730 --> 01:07:49.970
It was discovered that
the posterior system

01:07:49.970 --> 01:07:53.330
is very important for generating
quick saccades, especially

01:07:53.330 --> 01:07:56.250
express saccades, because when
you remove the colliculus,

01:07:56.250 --> 01:07:58.510
you never got an
express saccade again.

01:07:58.510 --> 01:08:03.770
And the anterior system plays a
very important role in stimulus

01:08:03.770 --> 01:08:07.580
selection and the
sequencing of eye movements

01:08:07.580 --> 01:08:10.800
because you make so many eye
movements in rapid succession,

01:08:10.800 --> 01:08:14.042
you have to make
plans ahead to decide

01:08:14.042 --> 01:08:15.750
where you're going to
look in a sequence.

01:08:15.750 --> 01:08:18.939
And that was found to be very
important for the frontal eye

01:08:18.939 --> 01:08:21.300
fields because when
you remove that, there

01:08:21.300 --> 01:08:25.649
was a major deficit in target
selection and in sequencing.

01:08:25.649 --> 01:08:28.180
So then when this
was done, we also

01:08:28.180 --> 01:08:31.960
examined, if you remember
the question, of what

01:08:31.960 --> 01:08:40.050
is the role of these various
areas when you block inhibition

01:08:40.050 --> 01:08:43.729
or you increase inhibition.

01:08:43.729 --> 01:08:46.790
So we use muscimal
and bicuculline to do

01:08:46.790 --> 01:08:48.950
that as shown here.

01:08:48.950 --> 01:08:54.930
And it showed that with V1,
you get a strong interference

01:08:54.930 --> 01:08:58.510
with both, and you also
get a strong deficit

01:08:58.510 --> 01:09:00.910
in visual discrimination
because to be

01:09:00.910 --> 01:09:02.960
able to analyze
the visual scene,

01:09:02.960 --> 01:09:05.960
you need to have interaction
between excitation

01:09:05.960 --> 01:09:07.960
and inhibition both
for eye movements

01:09:07.960 --> 01:09:10.250
and for visual discrimination.

01:09:11.649 --> 01:09:14.970
And then with a frontal
eye field lesion,

01:09:14.970 --> 01:09:18.750
you've got facilitation as
you did in the colliculus

01:09:18.750 --> 01:09:23.220
when you put in bicuculline,
which eliminates inhibition.

01:09:23.220 --> 01:09:25.286
The monkey couldn't
help but makes saccades.

01:09:25.286 --> 01:09:27.035
But you've got
interference with muscimal.

01:09:28.300 --> 01:09:30.149
LIP had no effect.

01:09:30.149 --> 01:09:33.905
So that's then in a summary
was what we had discussed.

01:09:33.905 --> 01:09:35.279
And this is
something, of course,

01:09:35.279 --> 01:09:38.670
you need to go over
again in your notes,

01:09:38.670 --> 01:09:42.660
and in Stellar, and in
the assigned readings

01:09:42.660 --> 01:09:46.310
so that you can remember
this for the exam.

01:09:46.310 --> 01:09:51.260
OK and then I pointed out to
you that even though we never

01:09:51.260 --> 01:09:53.460
think of eye movements, we
have an incredible number

01:09:53.460 --> 01:09:56.000
of structures and
a number of tasks

01:09:56.000 --> 01:10:00.050
to be able to make
each eye movement.

01:10:00.050 --> 01:10:01.560
We have to select a target.

01:10:01.560 --> 01:10:04.810
We have to decide what each--
every time you move your eye,

01:10:04.810 --> 01:10:06.050
there are dozens of targets.

01:10:06.050 --> 01:10:07.460
We have to select one of those.

01:10:07.460 --> 01:10:09.644
Then we have to
decide what they are.

01:10:09.644 --> 01:10:11.310
Then we have to decide
which one to look

01:10:11.310 --> 01:10:12.840
at which one not to look at.

01:10:12.840 --> 01:10:16.090
And then we have to
use our system, which

01:10:16.090 --> 01:10:20.860
is a spatial organization
of the motor fields

01:10:20.860 --> 01:10:23.570
to eventually generate
an eye movement.

01:10:23.570 --> 01:10:27.280
Now in reality then what happens
is that many other systems--

01:10:27.280 --> 01:10:30.430
I showed you this before, too--
many other systems are involved

01:10:30.430 --> 01:10:33.990
in generational eye movements,
hearing, touch, so on.

01:10:35.020 --> 01:10:37.950
And we had generated
all sorts of systems

01:10:37.950 --> 01:10:39.830
to enable for you to do this.

01:10:39.830 --> 01:10:43.170
the so-called anterior
and the posterior systems

01:10:43.170 --> 01:10:48.580
that reach the brain stem
through various channels here.

01:10:48.580 --> 01:10:51.740
This is available for
you on the internet.

01:10:51.740 --> 01:10:54.310
It's also available to you
on the assigned reading.

01:10:55.390 --> 01:10:59.020
So now lastly we'll turn
to motion perception.

01:10:59.020 --> 01:11:01.200
And when we talked
about motion perception,

01:11:01.200 --> 01:11:04.615
I pointed out to you
that in the area of V1,

01:11:04.615 --> 01:11:08.300
we have simple cells,
and complex cells,

01:11:08.300 --> 01:11:10.420
several different
classes of simple cells.

01:11:10.420 --> 01:11:13.630
And almost all of these cells,
if you look at their responses

01:11:13.630 --> 01:11:15.770
to light increment
and light decrement,

01:11:15.770 --> 01:11:18.140
meaning light edges
and dark edges,

01:11:18.140 --> 01:11:23.030
almost every one of these
cells is direction selective.

01:11:23.030 --> 01:11:25.000
And it's also true for
most complex cells.

01:11:25.000 --> 01:11:27.440
About half the complex
cells, maybe more,

01:11:27.440 --> 01:11:29.850
are also direction selective.

01:11:29.850 --> 01:11:34.450
So direction selectivity is one
of the most central features

01:11:34.450 --> 01:11:38.040
in the visual system
that we use extensively

01:11:38.040 --> 01:11:40.230
not just to analyze
motion but also

01:11:40.230 --> 01:11:44.000
to be able to see depth
by virtue of motion.

01:11:44.000 --> 01:11:45.540
Paradox

01:11:45.540 --> 01:11:49.930
So now we can say, because of
all those little experiments

01:11:49.930 --> 01:11:54.040
I had shown you, that the
parasol system and because

01:11:54.040 --> 01:11:57.960
of the lesion experiments
plays a central role in motion

01:11:57.960 --> 01:11:58.930
analysis.

01:11:58.930 --> 01:12:01.930
And when we do those experiments
with a apparent motion, when

01:12:01.930 --> 01:12:08.270
we moved little spots in
color or in small differences

01:12:08.270 --> 01:12:13.870
in shape, color and
small difference in shape

01:12:13.870 --> 01:12:19.550
didn't matter, indicating
that the parasol system plays

01:12:19.550 --> 01:12:24.040
a central role in us seeing
apparent motion the way we

01:12:24.040 --> 01:12:25.030
see it.

01:12:25.030 --> 01:12:25.530
All right.

01:12:25.530 --> 01:12:29.330
So now, last very briefly
I want to say here

01:12:29.330 --> 01:12:32.770
is about the
accessory optic system

01:12:32.770 --> 01:12:35.060
because this is what you're
going to be writing about.

01:12:35.060 --> 01:12:38.850
And I just wanted to remind you
that the basic discovery was

01:12:38.850 --> 01:12:42.460
that in the retina, the
cells of [INAUDIBLE] that

01:12:42.460 --> 01:12:45.960
feed into the accessory
optic system come in three

01:12:45.960 --> 01:12:50.440
different direction
selectivities as shown here

01:12:50.440 --> 01:12:53.720
and that these three direction
selectivities correspond

01:12:53.720 --> 01:12:56.820
to the direction selectivity
of semicircular canals.

01:12:56.820 --> 01:12:58.320
That's quite a
remarkable discovery.

01:13:00.020 --> 01:13:02.600
And this then
enables the organism

01:13:02.600 --> 01:13:04.990
through the system,
which by the way

01:13:04.990 --> 01:13:09.120
these cells respond to all
the slow movements, that's

01:13:09.120 --> 01:13:12.830
prime function is
to-- so they claim,

01:13:12.830 --> 01:13:16.340
and I think that's
correct-- is to stabilize

01:13:16.340 --> 01:13:18.230
the eye with respect
to the world.

01:13:18.230 --> 01:13:20.490
So when you walk
around, what happens

01:13:20.490 --> 01:13:23.820
is you can still see the world
very clearly with no blurring

01:13:23.820 --> 01:13:28.220
because the accessory optic
system adjusts the eye

01:13:28.220 --> 01:13:31.900
to keep it stable with respect
to the world out there.

01:13:31.900 --> 01:13:34.235
And in fact, I can't remember
if I told you this story.

01:13:40.400 --> 01:13:45.400
Way back when in Germany
when some people were treated

01:13:45.400 --> 01:13:50.950
for pneumonia, they used the
drug-- I forget the name of it

01:13:50.950 --> 01:13:57.570
right now-- that
caused malfunctioning

01:13:57.570 --> 01:13:59.290
in the semicircular canals.

01:14:00.380 --> 01:14:04.250
As a result of
that, that system,

01:14:04.250 --> 01:14:11.260
which co-exists with the
accessory optic system

01:14:11.260 --> 01:14:13.830
no longer was able to
stabilize the eyes.

01:14:13.830 --> 01:14:19.184
And so here was this guy in
Munich living in a neighborhood

01:14:19.184 --> 01:14:20.850
where he had lived
for many, many years.

01:14:20.850 --> 01:14:25.490
And he realized that he
can't see anything clearly

01:14:25.490 --> 01:14:26.890
when he's walking on the street.

01:14:26.890 --> 01:14:28.190
Everything was blurry.

01:14:28.190 --> 01:14:29.840
And so he said oh, my god.

01:14:29.840 --> 01:14:31.510
I won't be able to
recognize my friend.

01:14:31.510 --> 01:14:33.360
I won't be able to
say hello to him.

01:14:33.360 --> 01:14:34.990
And so what he
learned to do is this.

01:14:38.510 --> 01:14:39.985
Hi, Joe.

01:14:39.985 --> 01:14:41.060
Like that.

01:14:41.060 --> 01:14:44.560
He stabilized his
head by holding it.

01:14:44.560 --> 01:14:46.870
So that highlights
for you the fact

01:14:46.870 --> 01:14:52.690
that this system of
stabilizing the eye

01:14:52.690 --> 01:14:55.200
through the accessory
optic system

01:14:55.200 --> 01:14:57.370
plays a very
important central role

01:14:57.370 --> 01:14:59.620
in your being able to
move around in the world

01:14:59.620 --> 01:15:02.337
and being able to analyze
the visual scene in spite

01:15:02.337 --> 01:15:03.920
of the fact that
you're moving around.

01:15:05.210 --> 01:15:08.196
So the last thing I wanted
to show you-- first,

01:15:08.196 --> 01:15:09.820
I'm going to show
you one more picture.

01:15:09.820 --> 01:15:13.360
But first, let me say a couple
of words about the exam again.

01:15:13.360 --> 01:15:17.360
I've told you the exam is
a multiple choice exam,

01:15:17.360 --> 01:15:20.155
probably something about
100 or so questions.

01:15:23.120 --> 01:15:26.270
Almost all of them
deal with basic facts,

01:15:26.270 --> 01:15:27.755
I should say basic facts.

01:15:27.755 --> 01:15:29.130
So you've got to
know your facts.

01:15:30.690 --> 01:15:35.180
And what you need to do
is read each question,

01:15:35.180 --> 01:15:36.230
circle the choice.

01:15:37.270 --> 01:15:40.391
You don't get punished
extra for being wrong.

01:15:40.391 --> 01:15:41.640
If you're wrong, you're wrong.

01:15:41.640 --> 01:15:43.350
But I'm not going
to subtract on top

01:15:43.350 --> 01:15:47.405
of that the wrong answers
from the right answers.

01:15:48.810 --> 01:15:52.530
So choose an answer,
even if you don't know it

01:15:52.530 --> 01:15:53.485
for every question.

01:15:55.070 --> 01:15:59.030
And you'll have a probability of
one in four, maybe one in five

01:15:59.030 --> 01:16:03.700
of getting the right answer
if you're totally ignorant.

01:16:03.700 --> 01:16:05.650
So that's what the exam is.

01:16:05.650 --> 01:16:08.200
It's going to take about an
hour or so, hour and a half

01:16:08.200 --> 01:16:10.050
maybe, depending on
how fast you read

01:16:10.050 --> 01:16:12.150
and how fast you make decisions.

01:16:12.150 --> 01:16:18.820
And that's going to take
place this coming Wednesday

01:16:18.820 --> 01:16:19.830
right in this room.

01:16:20.900 --> 01:16:24.640
Now the last thing I
wanted to show you is-- I

01:16:24.640 --> 01:16:26.750
mean I seem to be so
certain about everything

01:16:26.750 --> 01:16:28.230
being right and wrong here.

01:16:28.230 --> 01:16:30.655
I just wanted to tell you
one thing, a note of caution.

01:16:31.890 --> 01:16:33.780
And the caution
is this one here.

01:16:33.780 --> 01:16:36.325
This is a wonderful
sculpture by Naum Gabo.

01:16:39.010 --> 01:16:41.470
I don't know if you've
ever heard of Naum Gabo.

01:16:41.470 --> 01:16:45.090
Anyway, this is obviously
you can almost instantly

01:16:45.090 --> 01:16:49.350
say it's an upper body
and a face, right?

01:16:49.350 --> 01:16:52.640
But the fact is
that, as I say here,

01:16:52.640 --> 01:16:55.980
as many scientific
hypotheses of brain function

01:16:55.980 --> 01:16:59.590
are appealing but a far
cry from the real McCoy.

01:17:00.620 --> 01:17:01.820
So we are still groping.

01:17:02.980 --> 01:17:07.610
And yes, we are a long
way from phrenology,

01:17:07.610 --> 01:17:10.920
but still many of the hypotheses
and ideas that we have

01:17:10.920 --> 01:17:15.900
are wrong and are more like
a cartoon of what it really

01:17:15.900 --> 01:17:16.600
is like.

01:17:19.880 --> 01:17:23.320
And of course, the
further up, in my opinion,

01:17:23.320 --> 01:17:26.380
you go from the
retina, the higher

01:17:26.380 --> 01:17:28.830
the fancifulness of the ideas.

01:17:28.830 --> 01:17:30.380
At least when it
comes to the retina,

01:17:30.380 --> 01:17:32.240
I think we are
reasonably comfortable

01:17:32.240 --> 01:17:34.730
that we know a lot
about the photoreceptors

01:17:34.730 --> 01:17:36.030
and how they interact.

01:17:36.030 --> 01:17:38.520
And that's fairly close
to the way it really is.

01:17:39.700 --> 01:17:44.605
So that may be more like
a photograph of Obama.

01:17:46.390 --> 01:17:49.020
But when it comes to the
cortex of the higher areas,

01:17:49.020 --> 01:17:51.050
things are a bit more like that.

01:17:51.050 --> 01:17:52.730
So that's the end of it then.

01:17:52.730 --> 01:17:53.930
Thank you very much.

01:17:53.930 --> 01:18:00.170
And I wish you the best of
luck on your exam on Wednesday.

01:18:00.170 --> 01:18:01.087
[APPLAUSE]

01:18:01.087 --> 01:18:01.670
Oh, thank you.

01:18:01.670 --> 01:18:01.970
Thank you.

01:18:01.970 --> 01:18:02.678
That's very nice.

01:18:02.678 --> 01:18:04.120
I appreciate it.