1
00:00:00,070 --> 00:00:01,670
The following
content is provided

2
00:00:01,670 --> 00:00:03,820
under a Creative
Commons license.

3
00:00:03,820 --> 00:00:06,550
Your support will help MIT
OpenCourseWare continue

4
00:00:06,550 --> 00:00:10,160
to offer high quality
educational resources for free.

5
00:00:10,160 --> 00:00:12,700
To make a donation or to
view additional materials

6
00:00:12,700 --> 00:00:16,620
from hundreds of MIT courses,
visit MIT OpenCourseWare

7
00:00:16,620 --> 00:00:17,275
at ocw.mit.edu.

8
00:00:28,417 --> 00:00:29,000
PROFESSOR: OK.

9
00:00:29,000 --> 00:00:30,220
I guess we'll get started.

10
00:00:30,220 --> 00:00:34,520
So last time we were talking
about the auditory midbrain,

11
00:00:34,520 --> 00:00:37,215
where the big structure is
the inferior colliculus.

12
00:00:38,410 --> 00:00:43,830
And we talked about
its inputs coming

13
00:00:43,830 --> 00:00:46,350
from the superior
olivary nuclei,

14
00:00:46,350 --> 00:00:49,200
like the lateral
superior olive, which

15
00:00:49,200 --> 00:00:54,050
has neuron sensitive to
enter oral level differences,

16
00:00:54,050 --> 00:00:57,710
and another input
being the MSO, where

17
00:00:57,710 --> 00:01:01,782
the neurons insensitive
to binaural stimuli

18
00:01:01,782 --> 00:01:03,156
with interaural
time differences.

19
00:01:04,180 --> 00:01:09,230
And those inputs converge
on the inferior colliculus.

20
00:01:09,230 --> 00:01:13,220
And in today's lecture,
we have a little slide

21
00:01:13,220 --> 00:01:14,540
of the inferior colliculus.

22
00:01:14,540 --> 00:01:19,480
And the big part of the inferior
colliculus, diagrammed here

23
00:01:19,480 --> 00:01:21,403
is this central
nucleus, the ICC.

24
00:01:23,100 --> 00:01:28,230
And that structure
has been well studied,

25
00:01:28,230 --> 00:01:30,170
part because it's the big part.

26
00:01:30,170 --> 00:01:34,330
It's easy to record from,
it's deep in the brain.

27
00:01:34,330 --> 00:01:36,956
It has a strong
tonotopic organization.

28
00:01:39,250 --> 00:01:42,560
We talked last time about
neurons there having

29
00:01:42,560 --> 00:01:46,830
time intensity trading because
they get in some cases, input

30
00:01:46,830 --> 00:01:50,210
from both LSO and MSO.

31
00:01:50,210 --> 00:01:53,870
We talked about coding for
stimulating head precedence

32
00:01:53,870 --> 00:01:58,980
like effects in the
inferior colliculus,

33
00:01:58,980 --> 00:02:03,720
and we talked a little bit about
sound localization in there

34
00:02:03,720 --> 00:02:06,640
in the mammal, at
least for the parts

35
00:02:06,640 --> 00:02:08,979
of the inferior colliculus
that had been studied,

36
00:02:08,979 --> 00:02:14,890
there being no really good or
obvious map of sound location

37
00:02:14,890 --> 00:02:20,710
to place in the colliculus,
and we also talked about some

38
00:02:20,710 --> 00:02:23,960
of the other edge
regions of the colliculus

39
00:02:23,960 --> 00:02:26,830
like the external nucleus
of the colliculus that

40
00:02:26,830 --> 00:02:29,870
has been explored very
nicely in the bar now.

41
00:02:31,110 --> 00:02:33,310
And we talked about
its projections

42
00:02:33,310 --> 00:02:39,040
up to the optic tectum,
or the tectum in the bar

43
00:02:39,040 --> 00:02:42,350
now, which is the analog
of the superior colliculus

44
00:02:42,350 --> 00:02:43,590
in the mammal.

45
00:02:43,590 --> 00:02:45,690
And there's this
beautiful mapping

46
00:02:45,690 --> 00:02:49,300
of auditory space
in the barn owl

47
00:02:49,300 --> 00:02:51,540
at that position in the brain.

48
00:02:51,540 --> 00:02:55,730
Now, such mappings are
not found in general

49
00:02:55,730 --> 00:02:58,750
in the auditory
system of mammals,

50
00:02:58,750 --> 00:03:01,940
with the one exception
being in the deep layers

51
00:03:01,940 --> 00:03:03,180
of the superior colliculus.

52
00:03:04,270 --> 00:03:07,770
There are neurons that are
responsive to auditory,

53
00:03:07,770 --> 00:03:09,870
as well as visual stimuli.

54
00:03:09,870 --> 00:03:12,410
Of course, we think of
the superior colliculus

55
00:03:12,410 --> 00:03:13,950
as being a visual nucleus.

56
00:03:15,250 --> 00:03:17,570
And in those deeper
layers, there

57
00:03:17,570 --> 00:03:19,770
is a mapping of
auditory space that's

58
00:03:19,770 --> 00:03:23,500
in line with the mapping
for visual space there.

59
00:03:23,500 --> 00:03:29,320
So it's as if a mapping of space
can be created in the mammal,

60
00:03:29,320 --> 00:03:32,220
but it just doesn't
appear, at least

61
00:03:32,220 --> 00:03:35,000
in the main nuclei of
the auditory pathway.

62
00:03:35,000 --> 00:03:39,130
The superior colliculus is
really not an auditory nucleus.

63
00:03:41,110 --> 00:03:41,610
OK.

64
00:03:41,610 --> 00:03:44,510
So any questions from last time?

65
00:03:44,510 --> 00:03:48,430
So today, we're going to
be talking about mainly

66
00:03:48,430 --> 00:03:52,860
the auditory cortex
and how it has

67
00:03:52,860 --> 00:03:56,220
a number of fields
in the cortex.

68
00:03:56,220 --> 00:04:00,820
Usually we talk about
fields rather than nuclei,

69
00:04:00,820 --> 00:04:04,340
and these fields
are tonotopically

70
00:04:04,340 --> 00:04:08,940
organized-- at least,
several of them are.

71
00:04:08,940 --> 00:04:10,770
And some of them
aren't, but we'll

72
00:04:10,770 --> 00:04:14,230
be emphasizing the
tonotopic ones.

73
00:04:14,230 --> 00:04:18,209
There are some very
interesting experiments

74
00:04:18,209 --> 00:04:23,420
where investigators have
explored how changes in hearing

75
00:04:23,420 --> 00:04:27,660
can affect the tonotopic
mappings in these fields

76
00:04:27,660 --> 00:04:29,540
so that they can be plastic.

77
00:04:29,540 --> 00:04:33,470
They can be molded or shaped
depending on experience.

78
00:04:34,740 --> 00:04:36,280
So we'll talk about
that plasticity.

79
00:04:38,420 --> 00:04:41,560
Because there are many
fields, the obvious question

80
00:04:41,560 --> 00:04:44,815
is, well, what does one field
do in your sense of hearing,

81
00:04:44,815 --> 00:04:46,780
and what does another field do.

82
00:04:46,780 --> 00:04:51,140
And we're generally not able
to answer those questions.

83
00:04:51,140 --> 00:04:53,800
I like to think that there
are Nobel Prizes waiting

84
00:04:53,800 --> 00:05:00,100
to be earned in this area as
that function is worked out

85
00:05:00,100 --> 00:05:01,330
for each of the fields.

86
00:05:01,330 --> 00:05:05,520
But it's clear that there's a
big role of one field, which

87
00:05:05,520 --> 00:05:09,340
we call A1, in
sound localization.

88
00:05:12,200 --> 00:05:15,653
And we'll review that
line of investigation.

89
00:05:16,670 --> 00:05:21,130
That if A1 is lost, an
experimental animal's ability

90
00:05:21,130 --> 00:05:25,740
to localize sounds
is completely gone.

91
00:05:27,601 --> 00:05:28,100
OK?

92
00:05:28,100 --> 00:05:30,190
So we'll talk about
the evidence, how

93
00:05:30,190 --> 00:05:33,580
we test animals for
sound localization.

94
00:05:33,580 --> 00:05:37,685
Then we'll talk about where
A1 is in the human auditory

95
00:05:37,685 --> 00:05:38,185
cortex.

96
00:05:39,290 --> 00:05:43,130
And we'll talk finally about
some very interesting imaging

97
00:05:43,130 --> 00:05:48,640
studies that have shown that
there is a center that lights

98
00:05:48,640 --> 00:05:51,110
up in imaging studies
when you present

99
00:05:51,110 --> 00:05:53,630
sounds with salient pitch.

100
00:05:53,630 --> 00:05:57,280
That is, very strong
sensations of pitch,

101
00:05:57,280 --> 00:06:01,850
and it's an area near A1,
but a little bit beyond it.

102
00:06:01,850 --> 00:06:03,760
So we'll talk about
those imaging studies.

103
00:06:03,760 --> 00:06:06,490
And that relates to
the paper that we

104
00:06:06,490 --> 00:06:08,620
have for assigned reading
for today's lecture.

105
00:06:09,950 --> 00:06:12,950
So that's the roadmap for today.

106
00:06:12,950 --> 00:06:17,270
So the first slide is a
very complicated slide

107
00:06:17,270 --> 00:06:20,260
that as I said, has the
inferior colliculus on it.

108
00:06:21,950 --> 00:06:27,300
And in the second row, has
the auditory thalamus on it.

109
00:06:27,300 --> 00:06:29,820
So the auditory thalamus
is the medial geniculate.

110
00:06:31,800 --> 00:06:33,330
And I'm not going
to say very much

111
00:06:33,330 --> 00:06:35,090
about the medial geniculate.

112
00:06:35,090 --> 00:06:37,140
Probably Peter Schiller
talked a lot more

113
00:06:37,140 --> 00:06:41,210
about the lateral geniculate in
the visual part of the course.

114
00:06:42,240 --> 00:06:45,280
I want to just say that
there is a large part

115
00:06:45,280 --> 00:06:48,620
of the medial geniculate called
the ventral division, indicated

116
00:06:48,620 --> 00:06:49,120
here.

117
00:06:50,300 --> 00:06:51,800
There's a couple
little symbols that

118
00:06:51,800 --> 00:06:54,130
say v. That stands
for ventral division.

119
00:06:54,130 --> 00:06:56,235
And there is a large part
of the medial geniculate

120
00:06:56,235 --> 00:06:58,240
that is tonotopically organized.

121
00:06:59,640 --> 00:07:02,230
And so it receives
a lot of input

122
00:07:02,230 --> 00:07:05,150
from the central nucleus of the
colliculus, which is tonotopic.

123
00:07:06,380 --> 00:07:07,730
It is tonotopic.

124
00:07:07,730 --> 00:07:12,460
And it, in turn, projects
2 auditory cortical fields

125
00:07:12,460 --> 00:07:14,310
that are listed up
there, which we'll

126
00:07:14,310 --> 00:07:16,750
go over in detail
in just a minute.

127
00:07:16,750 --> 00:07:19,670
And all of those
listed cortical fields,

128
00:07:19,670 --> 00:07:23,800
with the exception of the
one that says t, all of those

129
00:07:23,800 --> 00:07:25,350
are tonotopically organized.

130
00:07:27,130 --> 00:07:30,600
So to some people
than, there is a part

131
00:07:30,600 --> 00:07:34,900
of this auditory pathway
at these higher levels

132
00:07:34,900 --> 00:07:37,440
that's called the
tonotopic system.

133
00:07:37,440 --> 00:07:40,210
And that has a
bulk of the nuclei.

134
00:07:42,660 --> 00:07:44,090
There are some other systems.

135
00:07:45,120 --> 00:07:48,950
And at least in
this author's idea,

136
00:07:48,950 --> 00:07:52,680
they're labeled as diffuse
and multisensory systems.

137
00:07:52,680 --> 00:07:54,540
We're not going to
talk much about them.

138
00:07:55,630 --> 00:07:59,950
But they start out in other
parts of the colliculus,

139
00:07:59,950 --> 00:08:02,430
like the dorsal part
of the colliculus,

140
00:08:02,430 --> 00:08:04,460
and the external nucleus
of the colliculus.

141
00:08:05,690 --> 00:08:08,435
And they pass through other
divisions of the thalamus.

142
00:08:10,140 --> 00:08:13,160
And they either go to other
auditory cortical fields,

143
00:08:13,160 --> 00:08:18,440
like A2 here, or they
project diffusely

144
00:08:18,440 --> 00:08:21,030
to a whole bunch of
places in auditory cortex.

145
00:08:21,030 --> 00:08:24,580
So we're not going to emphasize
them very much at all.

146
00:08:24,580 --> 00:08:27,720
Just stick with the
tonotopic system for now.

147
00:08:29,580 --> 00:08:33,620
So in the auditory
cortex, there's

148
00:08:33,620 --> 00:08:37,799
been a lot of research done
on the cortex of the cat.

149
00:08:37,799 --> 00:08:42,179
Cat has been a standard
model for auditory cortex

150
00:08:42,179 --> 00:08:44,790
work since the beginning,
since the '60s.

151
00:08:46,230 --> 00:08:48,815
And so the fields are very
well known in the cat.

152
00:08:49,840 --> 00:08:54,000
And they're indicated in
the colored shading here.

153
00:08:54,000 --> 00:08:56,940
And this is showing
you the cat's brain

154
00:08:56,940 --> 00:08:59,660
from looking at its left side.

155
00:08:59,660 --> 00:09:05,070
So you see that the colored
areas, the auditory areas,

156
00:09:05,070 --> 00:09:08,520
are in the part of
the brain called

157
00:09:08,520 --> 00:09:13,530
the temporal cortex,
the side of the cortex.

158
00:09:13,530 --> 00:09:19,820
So way up here would be the
frontal part of the cortex,

159
00:09:19,820 --> 00:09:21,730
the frontal lobes here.

160
00:09:21,730 --> 00:09:25,130
Way back here, you're familiar
with the occipital cortex,

161
00:09:25,130 --> 00:09:26,655
which would be the visual areas.

162
00:09:27,850 --> 00:09:32,620
And these areas are on the side,
our so-called temporal cortex.

163
00:09:32,620 --> 00:09:33,885
Those are the auditory areas.

164
00:09:36,210 --> 00:09:38,350
And the cat is a nice
model for cortex.

165
00:09:38,350 --> 00:09:41,900
Because much of the big
auditory cortical field,

166
00:09:41,900 --> 00:09:45,870
A1 here, is on the surface.

167
00:09:45,870 --> 00:09:52,530
It's on a big gyrus
that's accessible.

168
00:09:52,530 --> 00:09:55,020
You can get to it, you
can record from it.

169
00:09:55,020 --> 00:09:56,560
You can make lesions in it.

170
00:09:57,600 --> 00:10:01,220
A little bit of A1 goes
down into a sulcus.

171
00:10:01,220 --> 00:10:04,760
And those sulci are
labeled by the lines.

172
00:10:04,760 --> 00:10:09,340
This one is labeled ps, the
posterior ectosylvian sulcus.

173
00:10:10,440 --> 00:10:15,700
And this is the cortex with
the cortex map pulled apart.

174
00:10:15,700 --> 00:10:19,150
So you can see the parts
of the cortex that's

175
00:10:19,150 --> 00:10:21,820
down in the sulcus.

176
00:10:21,820 --> 00:10:24,330
And those are the
dark pink here.

177
00:10:24,330 --> 00:10:28,450
And in the front part, rostral
part of the auditory cortex,

178
00:10:28,450 --> 00:10:31,970
there's another anterior
ectosylvian sulcus.

179
00:10:31,970 --> 00:10:34,740
And that's shown by
the dark pink here.

180
00:10:34,740 --> 00:10:37,700
So much of A1 is on the
surface, but a little bit of it

181
00:10:37,700 --> 00:10:40,075
dives down into a sulcus.

182
00:10:43,670 --> 00:10:48,000
Now, when you go with
microelectrodes and record

183
00:10:48,000 --> 00:10:53,350
from the auditory cortex,
you take off the skull,

184
00:10:53,350 --> 00:10:55,910
you take off the dura,
and you see the cortex.

185
00:10:55,910 --> 00:10:57,410
And you can take
your microelectrode

186
00:10:57,410 --> 00:11:00,170
and go right into the cortex.

187
00:11:00,170 --> 00:11:02,940
And of course, the
cortex has layers.

188
00:11:02,940 --> 00:11:06,590
So you'd be going, starting
in layer 1, and going down.

189
00:11:06,590 --> 00:11:09,140
How many layers are
there in cortex?

190
00:11:09,140 --> 00:11:10,700
There's 6, right?

191
00:11:12,120 --> 00:11:12,830
OK.

192
00:11:12,830 --> 00:11:16,030
And you find that if you go
down through all those layers

193
00:11:16,030 --> 00:11:18,500
and make recordings in
the different layers,

194
00:11:18,500 --> 00:11:21,790
all of the neurons in
one cortical penetration

195
00:11:21,790 --> 00:11:23,574
with your electrode
have the same CF.

196
00:11:24,936 --> 00:11:29,120
So of course, we want to measure
the characteristic frequency

197
00:11:29,120 --> 00:11:31,905
from the tuning curve, just
like we do in other places.

198
00:11:33,410 --> 00:11:37,150
And if you do that,
in these recordings

199
00:11:37,150 --> 00:11:39,045
you find that they
all have the same CF.

200
00:11:39,045 --> 00:11:44,090
And that's like going down
through a cortical column.

201
00:11:45,700 --> 00:11:51,510
So columnar organization is
very important in cortex.

202
00:11:56,960 --> 00:12:00,685
And if we draw it from the
side, this would be the surface.

203
00:12:02,860 --> 00:12:06,080
And you'd have these 6
layers usually labeled

204
00:12:06,080 --> 00:12:07,100
by Roman numerals.

205
00:12:14,250 --> 00:12:18,150
And your electrode would be
coming down here and sampling

206
00:12:18,150 --> 00:12:21,900
from the various layers,
single neurons at a time.

207
00:12:21,900 --> 00:12:24,540
Of course, you'd have to
maybe record from layer 1

208
00:12:24,540 --> 00:12:27,590
first, then layer 2, and
then so on, and so forth.

209
00:12:28,940 --> 00:12:34,110
And as long as you stay
within a column of cortex,

210
00:12:34,110 --> 00:12:37,640
the neurons have the same
characteristic frequency.

211
00:12:37,640 --> 00:12:41,190
So looking at it from
the surface here then,

212
00:12:41,190 --> 00:12:43,210
one column would be like a dot.

213
00:12:43,210 --> 00:12:44,950
You'd be looking
down at the very top

214
00:12:44,950 --> 00:12:49,190
of the column from the capital
down to the base of the column,

215
00:12:49,190 --> 00:12:50,890
if it were an
architectural column.

216
00:12:52,420 --> 00:12:56,600
When you do that, you
can make a CF mapping

217
00:12:56,600 --> 00:12:59,250
for these tonotopic fields.

218
00:12:59,250 --> 00:13:03,720
And you find that if you're
way over here, rostral in A1,

219
00:13:03,720 --> 00:13:06,455
you get very high CFs.

220
00:13:06,455 --> 00:13:11,170
And so high CFs for a cat would
be like 40 kilohertz, maybe

221
00:13:11,170 --> 00:13:13,220
even 50 kilohertz.

222
00:13:13,220 --> 00:13:17,270
An octave beyond the upper
limit of human hearing.

223
00:13:17,270 --> 00:13:21,210
But if you move your
electrode, more caudally,

224
00:13:21,210 --> 00:13:24,020
the CFs get lower,
and lower, and lower

225
00:13:24,020 --> 00:13:26,790
until you reach the
posterior edge of A1.

226
00:13:27,870 --> 00:13:29,860
And this little
word, "low" means,

227
00:13:29,860 --> 00:13:35,130
that's where the lowest
CFs in A1 would be found.

228
00:13:35,130 --> 00:13:40,805
And the low CFs would be like
0.1 kilohertz, 100 Hertz.

229
00:13:43,150 --> 00:13:48,010
If you went perpendicular to
that so-called tonotopic axis,

230
00:13:48,010 --> 00:13:52,301
you can have recordings that go
one here, one there, one there,

231
00:13:52,301 --> 00:13:52,800
one there.

232
00:13:52,800 --> 00:13:55,270
If you're moving up
from ventral to dorsal,

233
00:13:55,270 --> 00:13:57,610
then you find they
all have the same CFs.

234
00:13:57,610 --> 00:13:59,153
So that's an iso CF.

235
00:14:00,452 --> 00:14:02,170
Lamina, if you will.

236
00:14:03,570 --> 00:14:05,000
And it starts right here.

237
00:14:05,000 --> 00:14:06,930
This line right here
is an iso CF lamina.

238
00:14:06,930 --> 00:14:08,200
This would be the highest CFs.

239
00:14:10,400 --> 00:14:12,020
And then as you go
caudally the CFs

240
00:14:12,020 --> 00:14:13,353
get lower, and lower, and lower.

241
00:14:14,900 --> 00:14:19,280
Then, when you keep moving
your electrode, more and more

242
00:14:19,280 --> 00:14:23,510
caudally, you find further CFs.

243
00:14:23,510 --> 00:14:26,870
And then the
progression changes.

244
00:14:26,870 --> 00:14:29,575
The CFs start to become
higher, and higher, and higher.

245
00:14:31,720 --> 00:14:35,410
That's the signal that you've
entered another auditory

246
00:14:35,410 --> 00:14:38,840
cortical field right behind A1.

247
00:14:40,410 --> 00:14:42,616
And it gets its name
from being behind it,

248
00:14:42,616 --> 00:14:44,910
behind an anatomical
terminology is posterior.

249
00:14:46,100 --> 00:14:49,590
So this field is p,
or sometimes called

250
00:14:49,590 --> 00:14:52,620
PAF, posterior auditory field.

251
00:14:53,770 --> 00:14:57,270
And if you keep going,
more and more caudally.

252
00:14:57,270 --> 00:14:59,840
In this case, the
thing takes a turn.

253
00:14:59,840 --> 00:15:02,830
So you go, caudally and
ventrally, the CFs then

254
00:15:02,830 --> 00:15:09,880
start to get higher, until
you approach the boundary of p

255
00:15:09,880 --> 00:15:13,470
with the next auditory
cortical field, which is VP.

256
00:15:13,470 --> 00:15:17,580
Ventral posterior
auditory field.

257
00:15:17,580 --> 00:15:21,070
Then the CFs go
from high to low.

258
00:15:23,410 --> 00:15:29,290
Same thing happens where A1
meets the anterior auditory

259
00:15:29,290 --> 00:15:31,780
field, indicated by a here.

260
00:15:31,780 --> 00:15:36,013
Those two fields share a high
frequency, high CF boundary.

261
00:15:37,190 --> 00:15:38,704
And then, as you
go more rostral,

262
00:15:38,704 --> 00:15:40,370
the CFs get lower,
and lower, and lower.

263
00:15:41,800 --> 00:15:46,600
Sometimes this organization,
at the edges of the fields,

264
00:15:46,600 --> 00:16:02,780
is called, mirror image
tonotopy in the auditory system.

265
00:16:02,780 --> 00:16:04,660
And where else have
you guys seen this?

266
00:16:09,340 --> 00:16:12,410
Did you go over this in the
vision part of the course

267
00:16:12,410 --> 00:16:17,220
where you have the mapping
of the retina on V1

268
00:16:17,220 --> 00:16:19,150
in the occipital cortex, right?

269
00:16:19,150 --> 00:16:20,935
And it has a retina
topic mapping.

270
00:16:22,570 --> 00:16:25,890
And that has some image where
the nasal part of the retina

271
00:16:25,890 --> 00:16:29,560
is over here, and the temporal
part of the retina's over here.

272
00:16:29,560 --> 00:16:34,580
And where V1 abuts V2, you also
have a retina topic mapping.

273
00:16:34,580 --> 00:16:37,975
But it's a mirror
image reversal of V1.

274
00:16:39,600 --> 00:16:41,190
So you should have
gone over that

275
00:16:41,190 --> 00:16:42,565
in the visual part
of the course.

276
00:16:44,030 --> 00:16:50,590
You find such mirror image flips
in the somatosensory cortex.

277
00:16:51,690 --> 00:16:53,950
In the somatosensory
cortex, there's

278
00:16:53,950 --> 00:16:57,260
a mapping of the body surface.

279
00:16:57,260 --> 00:16:59,470
If you touch here
on the body surface,

280
00:16:59,470 --> 00:17:02,524
a certain part of the
somatosensory cortex responds.

281
00:17:02,524 --> 00:17:04,524
If you touch up here, a
different part responds.

282
00:17:04,524 --> 00:17:08,650
And there's a mapping
of the body surface

283
00:17:08,650 --> 00:17:10,770
onto the surface of the cortex.

284
00:17:10,770 --> 00:17:15,310
And where S1 meets S2,
there's also a mapping.

285
00:17:15,310 --> 00:17:16,465
But it's a mirror image.

286
00:17:17,619 --> 00:17:18,119
OK.

287
00:17:18,119 --> 00:17:20,569
So this is not a
surprise in terms

288
00:17:20,569 --> 00:17:23,990
of general cortical
organization.

289
00:17:23,990 --> 00:17:28,950
And it's also not a surprise
in the visual periphery.

290
00:17:28,950 --> 00:17:31,030
You have the retina
topic mapping.

291
00:17:31,030 --> 00:17:33,216
In the auditory periphery,
you have the CF mapping.

292
00:17:34,320 --> 00:17:36,925
And you have this nice CF
mapping along the cortex.

293
00:17:40,070 --> 00:17:40,570
OK.

294
00:17:40,570 --> 00:17:44,520
And in the cat, you have
these four tonotopically

295
00:17:44,520 --> 00:17:46,520
organized system fields.

296
00:17:46,520 --> 00:17:53,080
And you have several
fields; A2, DPV, and T,

297
00:17:53,080 --> 00:17:56,686
where the tonotopy is either
nonexistent, or much less

298
00:17:56,686 --> 00:17:57,185
obvious.

299
00:17:58,860 --> 00:18:01,920
And there are some
challenges here

300
00:18:01,920 --> 00:18:06,380
to exploring responses
in these other areas.

301
00:18:06,380 --> 00:18:18,305
For example, tuning curves in
A2 can sometimes look like this.

302
00:18:20,520 --> 00:18:25,145
And it's very difficult
then, to assign a CF

303
00:18:25,145 --> 00:18:28,880
to such a broad bowl
shaped tuning curve.

304
00:18:28,880 --> 00:18:30,790
You could do it, if you
really were pressed.

305
00:18:30,790 --> 00:18:33,990
But there's not much difference
between that frequency

306
00:18:33,990 --> 00:18:34,920
and that frequency.

307
00:18:34,920 --> 00:18:39,170
So this could be an
octave or more difference.

308
00:18:39,170 --> 00:18:42,740
And so it's hard to
assign a CF to some

309
00:18:42,740 --> 00:18:45,330
of the neurons in
these other fields.

310
00:18:45,330 --> 00:18:47,480
That's not true in A1.

311
00:18:47,480 --> 00:18:50,170
In the tonotopically
organize fields in general,

312
00:18:50,170 --> 00:18:53,110
there's very precise,
sharp frequency tuning.

313
00:18:55,571 --> 00:18:56,070
OK.

314
00:18:56,070 --> 00:18:59,030
So the tuning curves,
there's a little table here.

315
00:18:59,030 --> 00:19:04,080
They're usually sharp in these
tonotopically organized fields.

316
00:19:04,080 --> 00:19:05,940
Yes, there's tonotopic
organization.

317
00:19:05,940 --> 00:19:06,865
The latency is short.

318
00:19:09,430 --> 00:19:11,940
The response is brisk
or robust compared

319
00:19:11,940 --> 00:19:15,420
to response that might be
called poor or insecure.

320
00:19:15,420 --> 00:19:20,080
That just simply means, every
time you turn on a sound,

321
00:19:20,080 --> 00:19:22,340
and people remember,
typically tend

322
00:19:22,340 --> 00:19:24,590
to measure histograms
in response

323
00:19:24,590 --> 00:19:28,900
to hundreds of sound bursts
every time there's a response.

324
00:19:28,900 --> 00:19:30,550
But here, there
might be a response

325
00:19:30,550 --> 00:19:32,740
to the first tone
burst in a train.

326
00:19:32,740 --> 00:19:35,240
And then the neuron might shut
down and not respond anymore.

327
00:19:36,570 --> 00:19:41,290
So these diffuse areas
are hard to investigate.

328
00:19:44,110 --> 00:19:47,290
Most of these cortex
recordings I'm talking about

329
00:19:47,290 --> 00:19:49,940
have been done in
anesthetized animals.

330
00:19:49,940 --> 00:19:51,800
And so there's
always a question of,

331
00:19:51,800 --> 00:19:56,080
how much has anesthesia
changed the response patterns?

332
00:19:56,080 --> 00:19:59,190
Of course, anesthesia
has a big effect

333
00:19:59,190 --> 00:20:01,300
on these higher levels
of the nervous system.

334
00:20:01,300 --> 00:20:05,190
So it's not always clear
how much of these properties

335
00:20:05,190 --> 00:20:08,015
or change in these properties
has been due to anesthesia.

336
00:20:11,510 --> 00:20:14,640
So back to the
somatosensory cortex.

337
00:20:16,620 --> 00:20:20,180
Here's a mapping of the
somatosensory cortex.

338
00:20:20,180 --> 00:20:23,520
And this is the
surface of the body map

339
00:20:23,520 --> 00:20:26,120
onto the surface of
the cortex that I

340
00:20:26,120 --> 00:20:28,030
was referring to earlier.

341
00:20:29,090 --> 00:20:33,670
So if you stimulate
the surface of the body

342
00:20:33,670 --> 00:20:35,710
over here, the tips
of the toes, you

343
00:20:35,710 --> 00:20:38,330
get a response here in
the somatosensory cortex.

344
00:20:39,460 --> 00:20:42,200
If you stimulate the
fingers, you get a response

345
00:20:42,200 --> 00:20:45,480
here in the
somatosensory cortex.

346
00:20:45,480 --> 00:20:49,040
You stimulate the facial region,
you get a response over here.

347
00:20:50,320 --> 00:20:55,090
And this distorted
mapping or caricature

348
00:20:55,090 --> 00:20:58,620
of the body's surface,
sometimes called a homunculus.

349
00:20:58,620 --> 00:21:02,460
And it shows you
that a lot of cortex

350
00:21:02,460 --> 00:21:05,940
is devoted to certain
important parts of your body,

351
00:21:05,940 --> 00:21:06,665
like the face.

352
00:21:07,720 --> 00:21:11,710
And less important parts,
like the trunk of your body,

353
00:21:11,710 --> 00:21:16,240
receives much smaller
representation in the cortex.

354
00:21:17,710 --> 00:21:19,760
In the auditory
cortex, if you draw

355
00:21:19,760 --> 00:21:24,380
the mapping-- so
here's a mapping of A1.

356
00:21:25,970 --> 00:21:28,505
In this case, it's from
the guinea pig cortex.

357
00:21:29,900 --> 00:21:32,230
And in this case, the
mapping is plotted

358
00:21:32,230 --> 00:21:36,620
with the CF on the
y-axis, and the distance

359
00:21:36,620 --> 00:21:40,640
along the cortex on the x-axis.

360
00:21:40,640 --> 00:21:42,300
And these are the data.

361
00:21:42,300 --> 00:21:44,830
Each dot indicates
a recording site

362
00:21:44,830 --> 00:21:47,470
from a single column
in the cortex.

363
00:21:47,470 --> 00:21:48,970
And if you could
read these numbers,

364
00:21:48,970 --> 00:21:50,255
they would indicate the CFs.

365
00:21:51,830 --> 00:21:55,952
And these lines are
the ISO CF contours.

366
00:21:57,550 --> 00:22:02,160
And this CF axis is plotted
along this distance here.

367
00:22:02,160 --> 00:22:05,230
So this distance
is going like this,

368
00:22:05,230 --> 00:22:07,290
and this distance is
going along like this.

369
00:22:07,290 --> 00:22:10,923
And there's a very nice,
almost linear relationship.

370
00:22:12,530 --> 00:22:15,450
And it shows you
something quite different

371
00:22:15,450 --> 00:22:17,900
from the somatosensory
mapping, which

372
00:22:17,900 --> 00:22:20,786
is that there aren't any
really important frequencies.

373
00:22:22,630 --> 00:22:26,130
They all have about the same
representation in the cortex.

374
00:22:26,130 --> 00:22:30,170
It's a pretty boring,
straight line, if you will.

375
00:22:30,170 --> 00:22:34,460
It's not as interesting
as the homunculus

376
00:22:34,460 --> 00:22:35,905
in the somatosensory cortex.

377
00:22:37,110 --> 00:22:39,300
So this is true in
the general mammal.

378
00:22:40,380 --> 00:22:44,270
Next time, when we talk about
the auditory cortex in the echo

379
00:22:44,270 --> 00:22:48,210
locating bat, we'll have
quite a different finding.

380
00:22:48,210 --> 00:22:50,480
There are some very
important frequencies

381
00:22:50,480 --> 00:22:54,780
in certain types of bats that
relate to the echo locating

382
00:22:54,780 --> 00:22:57,730
signal that they
emit, and the echo

383
00:22:57,730 --> 00:23:00,900
that comes back to them so
that they can find targets,

384
00:23:00,900 --> 00:23:01,700
even in the dark.

385
00:23:02,910 --> 00:23:06,880
But most of general mammals
do not echolocate, of course.

386
00:23:06,880 --> 00:23:10,900
And so their mapping
of the sensory cortex

387
00:23:10,900 --> 00:23:15,180
is pretty linear and
boring, if you will.

388
00:23:15,180 --> 00:23:19,460
So today's reading
comes from what

389
00:23:19,460 --> 00:23:23,110
motivates the next
experiment that I'll show.

390
00:23:23,110 --> 00:23:24,900
For a long time,
these mappings were

391
00:23:24,900 --> 00:23:29,230
thought to be laid down at
birth and not changeable.

392
00:23:29,230 --> 00:23:30,865
So they were just immutable.

393
00:23:31,970 --> 00:23:34,290
But some very
interesting experiments

394
00:23:34,290 --> 00:23:39,260
by Dexter Irvine and Don
Robertson in the 1980s

395
00:23:39,260 --> 00:23:41,320
showed that was not true.

396
00:23:41,320 --> 00:23:44,770
And they were not the
pioneers in showing

397
00:23:44,770 --> 00:23:47,745
that cortex can change as
a result of experiment.

398
00:23:48,960 --> 00:23:52,240
Rather, some people who
were working, especially

399
00:23:52,240 --> 00:23:55,550
in the somatosensory
cortex, were the first.

400
00:23:55,550 --> 00:23:59,770
And so I didn't have
a reading for today.

401
00:23:59,770 --> 00:24:03,650
So right before I came over, I
pulled up the Wikipedia entry

402
00:24:03,650 --> 00:24:05,340
on the Silver Spring monkeys.

403
00:24:05,340 --> 00:24:08,320
So has anybody heard of
the Silver Spring monkeys?

404
00:24:09,890 --> 00:24:15,850
So Silver Spring monkeys were
in the news in the 1980s.

405
00:24:15,850 --> 00:24:17,810
They got their name
from the Institute

406
00:24:17,810 --> 00:24:22,180
of Behavioral Research in Silver
Springs, Maryland-- Silver

407
00:24:22,180 --> 00:24:24,230
Spring, Maryland.

408
00:24:24,230 --> 00:24:28,890
And from 1981 to
1991, they became

409
00:24:28,890 --> 00:24:33,560
what one writer called, the most
famous lab animals in history,

410
00:24:33,560 --> 00:24:37,930
as a result of battle between
animal researchers, animal

411
00:24:37,930 --> 00:24:40,650
advocates, politicians,
and courts.

412
00:24:40,650 --> 00:24:45,340
So there was a researcher
whose name was Edward Taub.

413
00:24:45,340 --> 00:24:49,720
And he was experimenting on the
somatosensory sensory cortex.

414
00:24:49,720 --> 00:24:51,730
And he was taking
the monkeys, and he

415
00:24:51,730 --> 00:24:54,950
was denervating
the sensory input

416
00:24:54,950 --> 00:24:56,860
from certain parts
of their limbs.

417
00:24:56,860 --> 00:25:01,110
So for example, he would cut the
nerves that carried information

418
00:25:01,110 --> 00:25:04,330
from the middle
finger of the monkeys.

419
00:25:04,330 --> 00:25:09,210
And he was studying to see
if the somatosensory cortex

420
00:25:09,210 --> 00:25:12,520
remapped, and he was
finding small effects.

421
00:25:12,520 --> 00:25:19,300
But in May, 1981, Alex Pacheco,
from the animal rights group

422
00:25:19,300 --> 00:25:22,250
PETA, began working
undercover in his lab,

423
00:25:22,250 --> 00:25:26,816
alerted the police to what PETA
viewed as unacceptable living

424
00:25:26,816 --> 00:25:27,940
conditions for the monkeys.

425
00:25:29,180 --> 00:25:31,650
And there was a long battle.

426
00:25:31,650 --> 00:25:35,060
Initially, the researcher was
convicted of animal cruelty,

427
00:25:35,060 --> 00:25:37,610
and these charges were
subsequently overturned.

428
00:25:41,410 --> 00:25:44,740
But anyway, the monkeys
were held in limbo

429
00:25:44,740 --> 00:25:48,770
for in some cases, many
years because his research

430
00:25:48,770 --> 00:25:49,530
was put on hold.

431
00:25:50,540 --> 00:25:53,030
During the subsequent
experiments

432
00:25:53,030 --> 00:25:55,900
on the monkeys after the
court battles were all done,

433
00:25:55,900 --> 00:25:59,710
it was discovered that
significant cortical remapping

434
00:25:59,710 --> 00:26:00,350
had occurred.

435
00:26:02,110 --> 00:26:04,210
This is evidence of
the brain's plasticity,

436
00:26:04,210 --> 00:26:06,980
and it helped to overturn the
widely held view that the old

437
00:26:06,980 --> 00:26:10,250
adult brain cannot reorganize
itself in response to its

438
00:26:10,250 --> 00:26:11,480
environment.

439
00:26:11,480 --> 00:26:16,090
So the analogous experiments
in the auditory system

440
00:26:16,090 --> 00:26:17,980
have been done in small animals.

441
00:26:17,980 --> 00:26:22,930
And maybe it's a result
of much decline and use

442
00:26:22,930 --> 00:26:24,525
of primates in research.

443
00:26:24,525 --> 00:26:27,420
There was hardly
any auditory work

444
00:26:27,420 --> 00:26:29,470
done on primates these days.

445
00:26:29,470 --> 00:26:32,505
This reorganization work
is done in the guinea pigs.

446
00:26:33,850 --> 00:26:35,805
And the experiments
are done like this.

447
00:26:38,320 --> 00:26:43,530
There's a peripheral lesion
made in the cochlear.

448
00:26:43,530 --> 00:26:47,560
In the Guinea pig
cochlear, it's very easy

449
00:26:47,560 --> 00:26:50,250
to make a little hole
in the middle layer,

450
00:26:50,250 --> 00:26:51,500
and look down on the cochlear.

451
00:26:51,500 --> 00:26:53,405
And the cochlear's
a bony structure.

452
00:26:55,260 --> 00:26:58,950
The most accessible part is
the basal turn of the cochlear.

453
00:26:58,950 --> 00:27:01,260
And you can go right
through the round window

454
00:27:01,260 --> 00:27:04,435
and see the basilar
membrane, and the hair cells.

455
00:27:05,610 --> 00:27:09,820
And you can make a
little tiny pinpoint

456
00:27:09,820 --> 00:27:15,500
opening in the organ of
corti with a fine metal pick,

457
00:27:15,500 --> 00:27:17,880
and create a
substantial hearing loss

458
00:27:17,880 --> 00:27:19,935
in one little place
of the cochlear.

459
00:27:21,150 --> 00:27:25,380
So here is indicated a
graph of the compound action

460
00:27:25,380 --> 00:27:27,080
potential threshold.

461
00:27:27,080 --> 00:27:29,820
This is a response from
the auditory nerve.

462
00:27:29,820 --> 00:27:32,090
Action potential,
obviously, is an impulse

463
00:27:32,090 --> 00:27:34,030
from single auditory
nerve fibers.

464
00:27:34,030 --> 00:27:37,960
Compound means it's a
recording from many, many,

465
00:27:37,960 --> 00:27:41,730
if not all of the
auditory nerve in response

466
00:27:41,730 --> 00:27:46,190
to a tone burst of the
different frequencies.

467
00:27:46,190 --> 00:27:48,560
And if you make a small
lesion at the basal turn,

468
00:27:48,560 --> 00:27:50,950
remember the frequency
organization of the cochlear

469
00:27:50,950 --> 00:27:55,860
is such that the basal turn
processes, high frequencies.

470
00:27:57,190 --> 00:28:01,400
And so instead of the normal
curve in the lesion animal,

471
00:28:01,400 --> 00:28:05,710
you have a big increase in
threshold, maybe 60, or 70,

472
00:28:05,710 --> 00:28:08,820
or 80 dB in the lesion case.

473
00:28:10,280 --> 00:28:13,760
And that lesion goes
from about 10 kilohertz

474
00:28:13,760 --> 00:28:15,075
to about 20 kilohertz.

475
00:28:16,460 --> 00:28:19,240
And in other parts
of the cochlear,

476
00:28:19,240 --> 00:28:21,090
the hearing is normal.

477
00:28:21,090 --> 00:28:23,315
So this is a peripheral
hearing loss.

478
00:28:26,020 --> 00:28:29,720
And now, we're going to
then look in the cortex

479
00:28:29,720 --> 00:28:32,714
and see if the tonotopy
of the auditory cortex

480
00:28:32,714 --> 00:28:33,755
is the same or different.

481
00:28:33,755 --> 00:28:35,740
And obviously, my
big build up here,

482
00:28:35,740 --> 00:28:38,850
that there's plasticity
of tonotopy, is found.

483
00:28:39,900 --> 00:28:43,880
This is the normal
mapping that we first saw.

484
00:28:43,880 --> 00:28:47,460
This is the mapping in
the lesioned animal.

485
00:28:47,460 --> 00:28:50,610
And in this case, it's a mapping
where each of these dots,

486
00:28:50,610 --> 00:28:52,590
of course, is a recording site.

487
00:28:52,590 --> 00:28:54,175
These are very high CFs.

488
00:28:55,720 --> 00:28:58,390
You march along here, and
the CF gets a little lower.

489
00:28:59,590 --> 00:29:04,850
But there's a big region
of about 20 kilohertz--

490
00:29:04,850 --> 00:29:08,230
big, long distance in the
cortex-- when all you get

491
00:29:08,230 --> 00:29:09,915
is CFs of 20 kilohertz.

492
00:29:11,900 --> 00:29:14,535
Notice that that is right
at the edge of the lesion.

493
00:29:17,160 --> 00:29:21,120
In the lesion, between
20 and 10 kilohertz,

494
00:29:21,120 --> 00:29:23,050
you don't get any response.

495
00:29:23,050 --> 00:29:25,000
Well, there's a huge
hearing loss there.

496
00:29:25,000 --> 00:29:27,670
It's no surprise that there's no
response to those frequencies.

497
00:29:28,790 --> 00:29:32,440
The auditory periphery is
not sending you any messages,

498
00:29:32,440 --> 00:29:35,260
or sending you very few messages
about those frequencies.

499
00:29:36,740 --> 00:29:39,385
Then you jump a little
bit in distance.

500
00:29:40,680 --> 00:29:45,030
And the CF jumped from
20 down to 10 kilohertz.

501
00:29:45,030 --> 00:29:47,370
And there's a long
region of cortex

502
00:29:47,370 --> 00:29:49,400
at which the CFs are
all 10 kilohertz.

503
00:29:51,050 --> 00:29:53,730
And notice that 10 kilohertz
is a very important frequency

504
00:29:53,730 --> 00:29:56,130
in the audiogram of this
animal, and that it's

505
00:29:56,130 --> 00:29:58,790
right at the edge, the low
frequency edge of the hearing

506
00:29:58,790 --> 00:29:59,290
loss.

507
00:30:00,840 --> 00:30:02,940
After that extensive
region then, you

508
00:30:02,940 --> 00:30:07,400
pick up your normal
tonotopy of auditory cortex.

509
00:30:08,760 --> 00:30:12,700
So this is clearly a
massive reorganization

510
00:30:12,700 --> 00:30:20,190
compared to the normal
of this lesioned animal's

511
00:30:20,190 --> 00:30:21,110
cortical mapping.

512
00:30:21,110 --> 00:30:23,250
So couple comments about this.

513
00:30:25,100 --> 00:30:31,520
If you do the mapping
right after the lesion,

514
00:30:31,520 --> 00:30:33,875
this reorganization
is not found.

515
00:30:35,990 --> 00:30:37,135
So it takes some time.

516
00:30:41,350 --> 00:30:42,970
In the case of the
auditory system,

517
00:30:42,970 --> 00:30:49,026
it takes more than three weeks
to see the reorganization.

518
00:30:58,960 --> 00:31:00,886
What's the mechanism
for this reorganization?

519
00:31:09,240 --> 00:31:17,510
Well, we have input coming
up to here from the thalamus,

520
00:31:17,510 --> 00:31:19,580
where the inputs
from the 20 kilohertz

521
00:31:19,580 --> 00:31:22,535
place of the thalamus,
did they come up here?

522
00:31:23,950 --> 00:31:28,650
And did they grow into a
large part of the cortex

523
00:31:28,650 --> 00:31:31,250
where they weren't
present before?

524
00:31:31,250 --> 00:31:35,640
And did they do that growth
because that part of the cortex

525
00:31:35,640 --> 00:31:38,790
had gone silent because
of the hearing loss?

526
00:31:38,790 --> 00:31:45,780
So one mechanism
could be growth,

527
00:31:45,780 --> 00:31:48,065
if you will, sideways
growth of axons.

528
00:31:52,870 --> 00:31:56,580
Another mechanism
could be the inputs

529
00:31:56,580 --> 00:31:58,270
from the thalamus
are coming up here.

530
00:31:58,270 --> 00:32:00,830
And even in the normal
case, they don't just

531
00:32:00,830 --> 00:32:03,890
go to one place, but they have
all sorts of side branches.

532
00:32:05,160 --> 00:32:09,470
And these side branches in
the normal case are inhibited,

533
00:32:09,470 --> 00:32:12,550
or they're not
strong to begin with.

534
00:32:12,550 --> 00:32:15,290
Because there's a lot of
other stuff coming in here,

535
00:32:15,290 --> 00:32:17,640
and the other stuff is
saying, you guys be quiet.

536
00:32:17,640 --> 00:32:18,650
I'm the main input.

537
00:32:19,940 --> 00:32:22,010
And maybe after
the hearing loss,

538
00:32:22,010 --> 00:32:23,440
that other stuff is silenced.

539
00:32:24,800 --> 00:32:29,050
And these previously weak
inputs become stronger.

540
00:32:29,050 --> 00:32:31,700
So you could say then
that another mechanism

541
00:32:31,700 --> 00:32:47,210
is strengthening of
preexisting inputs

542
00:32:47,210 --> 00:32:50,930
that are weak before
the perturbation.

543
00:32:50,930 --> 00:32:52,810
And I wrote mechanism, ?

544
00:32:52,810 --> 00:32:55,900
because we don't know
what the mechanism is.

545
00:32:55,900 --> 00:32:57,160
And it may be both.

546
00:32:57,160 --> 00:32:58,655
They're not mutually exclusive.

547
00:33:01,770 --> 00:33:04,390
And we're thinking here, cortex.

548
00:33:04,390 --> 00:33:05,800
This is a lecture on the cortex.

549
00:33:07,270 --> 00:33:13,973
But I should bring up, where is
the locus of this plasticity?

550
00:33:17,110 --> 00:33:21,280
So in the auditory system, we
have a rich array of nuclei.

551
00:33:21,280 --> 00:33:24,950
We have the cochlear nucleus,
we have the superior olivary

552
00:33:24,950 --> 00:33:28,053
complex, we have the
inferior colliculus.

553
00:33:29,560 --> 00:33:32,920
We've just learned a little
about the auditory thalamus.

554
00:33:32,920 --> 00:33:37,730
Well, could one of
those lower level nuclei

555
00:33:37,730 --> 00:33:41,610
have reorganized, and
then just passively spit

556
00:33:41,610 --> 00:33:44,290
their reorganized
input up to the cortex?

557
00:33:44,290 --> 00:33:47,490
And the answer is yes, and that
actually has been looked at.

558
00:33:50,100 --> 00:33:53,915
So there's no reorganization
in the cochlear nucleus.

559
00:33:56,640 --> 00:34:01,670
The part of the cochlear nucleus
that processes 20 kilohertz

560
00:34:01,670 --> 00:34:03,640
is supposed to be normal.

561
00:34:03,640 --> 00:34:07,940
In between 20 and 10 kilohertz,
you have a normal region.

562
00:34:07,940 --> 00:34:12,505
But it's completely silent in
these types of hearing loss.

563
00:34:13,850 --> 00:34:20,969
You have a small amount
of reorganization

564
00:34:20,969 --> 00:34:23,310
in the inferior colliculus.

565
00:34:23,310 --> 00:34:25,585
But it's not as big as
what we see in the cortex.

566
00:34:27,409 --> 00:34:30,650
And maybe in the
medial geniculate body,

567
00:34:30,650 --> 00:34:37,330
we have a larger reorganization,
but probably not quite as much

568
00:34:37,330 --> 00:34:38,949
as you do in the
auditory cortex.

569
00:34:38,949 --> 00:34:42,139
So there may be a little
bit of reorganization,

570
00:34:42,139 --> 00:34:48,139
an IC, a little bit in MGB,
and then further reorganization

571
00:34:48,139 --> 00:34:49,195
in auditory cortex.

572
00:34:50,320 --> 00:34:51,469
So that has been looked at.

573
00:34:51,469 --> 00:34:55,830
And the answer is at all
of these higher levels,

574
00:34:55,830 --> 00:34:57,010
there's reorganization.

575
00:34:57,010 --> 00:34:59,330
But there's maybe more as
you go higher in the pathway.

576
00:35:01,290 --> 00:35:02,070
Now, who cares?

577
00:35:02,070 --> 00:35:05,710
We usually don't have
hearing losses, right?

578
00:35:05,710 --> 00:35:08,230
Well I should remind
you that we had

579
00:35:08,230 --> 00:35:09,840
a big lecture on hearing loss.

580
00:35:09,840 --> 00:35:14,470
Then let me show you one
of the slides from it.

581
00:35:14,470 --> 00:35:20,760
So we had this slide under
the lecture on hearing loss.

582
00:35:20,760 --> 00:35:24,360
Well, I've given
it a new title now

583
00:35:24,360 --> 00:35:28,360
because I was thinking
about getting old this week.

584
00:35:29,930 --> 00:35:35,120
And the type of hearing loss
that you have when you get old

585
00:35:35,120 --> 00:35:36,900
is called presbycusis.

586
00:35:36,900 --> 00:35:39,140
Presbycusis is the
age related loss

587
00:35:39,140 --> 00:35:41,985
of hearing, especially
at high frequencies.

588
00:35:41,985 --> 00:35:44,360
And almost all of us are going
to go through presbycusis.

589
00:35:46,370 --> 00:35:51,680
This is a normal audiogram
or threshold of hearing curve

590
00:35:51,680 --> 00:35:54,810
when you're young, and this
is one when you get older.

591
00:35:56,130 --> 00:35:58,990
And invariably, we lose
our high frequency hearing.

592
00:36:01,080 --> 00:36:02,440
The causes are not known.

593
00:36:02,440 --> 00:36:06,340
But clearly it takes
place in the periphery,

594
00:36:06,340 --> 00:36:10,440
which is very similar to the
lesion study we just went over.

595
00:36:10,440 --> 00:36:12,850
It's a peripheral
hearing loss, what

596
00:36:12,850 --> 00:36:14,700
happens to your central pathway.

597
00:36:14,700 --> 00:36:22,940
It probably results in cortical
reorganization in the human.

598
00:36:22,940 --> 00:36:24,120
And again, I can't spell.

599
00:36:24,120 --> 00:36:25,205
There's a missing t here.

600
00:36:28,730 --> 00:36:31,800
So we probably all-- and
we have plenty of time.

601
00:36:31,800 --> 00:36:33,594
This hearing loss
doesn't happen in days.

602
00:36:33,594 --> 00:36:35,385
It happens over the
course of our lifetime.

603
00:36:36,670 --> 00:36:39,830
There's plenty of time
to reorganize the cortex.

604
00:36:39,830 --> 00:36:41,500
Now, let me go back
a little bit here

605
00:36:41,500 --> 00:36:44,720
and ask an interesting
question which

606
00:36:44,720 --> 00:36:48,465
has some negative answers,
and some positive answers.

607
00:36:49,750 --> 00:36:51,330
People have said, well, wow.

608
00:36:51,330 --> 00:36:53,490
This animal-- you
have a lot of cortex.

609
00:36:54,550 --> 00:36:59,205
Half a millimeter devoted
to CFs of 20 kilohertz.

610
00:36:59,205 --> 00:37:03,430
And you have a lot of cortex
devoted to CF of 10 kilohertz.

611
00:37:03,430 --> 00:37:07,630
Does this animal do something a
lot better at those frequencies

612
00:37:07,630 --> 00:37:11,000
than this normal animal
which just has a little bitty

613
00:37:11,000 --> 00:37:13,310
part of cortex devoted
to 20 kilohertz,

614
00:37:13,310 --> 00:37:15,790
and a little bit
devoted to 10 kilohertz.

615
00:37:15,790 --> 00:37:20,690
So it's not clear what the
answer is to that question yet.

616
00:37:20,690 --> 00:37:23,100
And people have
speculated in the normal--

617
00:37:23,100 --> 00:37:26,480
if you train a normal
person or a normal animal

618
00:37:26,480 --> 00:37:28,790
to do a task at 10 kilohertz.

619
00:37:28,790 --> 00:37:31,360
So they're listing to 10
kilohertz over and over again.

620
00:37:32,370 --> 00:37:35,220
It's clearly a training
effect for many tasks.

621
00:37:35,220 --> 00:37:36,485
You get better with training.

622
00:37:37,700 --> 00:37:39,860
Does that mean in
the normal case

623
00:37:39,860 --> 00:37:43,530
that we enlarge the 10
kilohertz part of our cortex?

624
00:37:43,530 --> 00:37:44,980
Well, it's not known.

625
00:37:44,980 --> 00:37:48,400
Does that means this
big area of 10 kilohertz

626
00:37:48,400 --> 00:37:52,370
and 20 kilohertz in the
lesioned animals cortex

627
00:37:52,370 --> 00:37:54,750
enable that animal to
do something better?

628
00:37:54,750 --> 00:37:56,047
That's not clear.

629
00:37:58,970 --> 00:38:02,200
There's evidence for both
answers to those questions.

630
00:38:05,950 --> 00:38:08,740
Let's move on to some
other properties that

631
00:38:08,740 --> 00:38:11,930
have been observed
in auditory cortex.

632
00:38:11,930 --> 00:38:16,490
And we've been stuck in
this mode and this course

633
00:38:16,490 --> 00:38:19,530
on frequency organization.

634
00:38:19,530 --> 00:38:23,160
It's clearly a very
strong component

635
00:38:23,160 --> 00:38:25,390
of central auditory nuclei.

636
00:38:25,390 --> 00:38:29,830
And here are some tuning curves
from auditory cortex neurons.

637
00:38:30,970 --> 00:38:33,370
You have the x-axis
being frequency,

638
00:38:33,370 --> 00:38:36,550
and the y-axis being
sound pressure level,

639
00:38:36,550 --> 00:38:38,900
response being inside
the tuning curve.

640
00:38:38,900 --> 00:38:41,500
In this case, they've
embellished the tuning curve

641
00:38:41,500 --> 00:38:44,100
a little bit by plotting
the biggest, best

642
00:38:44,100 --> 00:38:46,820
response in real black shading.

643
00:38:46,820 --> 00:38:49,890
And these are the
tuning curves we've

644
00:38:49,890 --> 00:38:52,330
been talking about in
the course all along.

645
00:38:52,330 --> 00:38:54,800
You might see goes
from the auditory nerve

646
00:38:54,800 --> 00:38:56,780
from cochlear nucleus
and lower levels.

647
00:38:58,560 --> 00:39:01,427
In the auditory
cortex, you see some

648
00:39:01,427 --> 00:39:02,760
of those kinds of tuning curves.

649
00:39:02,760 --> 00:39:05,070
But you also see from
different neurons,

650
00:39:05,070 --> 00:39:06,570
other types of tuning curves.

651
00:39:07,600 --> 00:39:10,700
And these are seen to a greater
extent in the auditory cortex,

652
00:39:10,700 --> 00:39:12,710
although you do see some,
to a certain extent,

653
00:39:12,710 --> 00:39:14,630
in the colliculus, and
to a certain extent,

654
00:39:14,630 --> 00:39:15,940
in the thalamus.

655
00:39:15,940 --> 00:39:19,150
So you see them in a numerically
greater extent in the cortex.

656
00:39:20,440 --> 00:39:23,575
And here is a tuning curve
that sure has a best frequency.

657
00:39:24,770 --> 00:39:27,555
But at that best frequency, if
you get it higher and higher

658
00:39:27,555 --> 00:39:29,880
in level, the neuron
actually stops

659
00:39:29,880 --> 00:39:34,380
responding at a pretty moderate,
and certainly at a high level.

660
00:39:35,750 --> 00:39:39,840
And if you study its response
inside the black shading,

661
00:39:39,840 --> 00:39:42,450
you can say, well, that
neuron has the best

662
00:39:42,450 --> 00:39:44,370
frequency, or
characteristic frequency.

663
00:39:44,370 --> 00:39:46,155
It also has a
characteristic level.

664
00:39:47,170 --> 00:39:51,660
It likes it a lot when the
tone frequency is 10 kilohertz,

665
00:39:51,660 --> 00:39:54,190
and the tone level is 40 dB.

666
00:39:54,190 --> 00:39:57,140
That's its maximal response.

667
00:39:57,140 --> 00:39:59,000
That's the one it
prefers, if you will.

668
00:40:00,010 --> 00:40:01,610
Here's a different one.

669
00:40:01,610 --> 00:40:03,970
A best level is somewhere
in the middle here.

670
00:40:03,970 --> 00:40:08,500
Here's a best level here in
a very narrow response area.

671
00:40:10,450 --> 00:40:14,460
So many of the neurons
in the auditory cortex

672
00:40:14,460 --> 00:40:16,620
has to have these not
only characteristic

673
00:40:16,620 --> 00:40:19,040
frequencies, but
also best levels.

674
00:40:19,040 --> 00:40:23,560
If you, at their CF,
raise the sound level,

675
00:40:23,560 --> 00:40:24,930
you get this type of pattern.

676
00:40:24,930 --> 00:40:27,720
These are rate level functions.

677
00:40:27,720 --> 00:40:31,480
So this is the firing rate, and
this is the tone level or tone

678
00:40:31,480 --> 00:40:32,579
intensity.

679
00:40:32,579 --> 00:40:34,120
And this is from a
number of neurons.

680
00:40:34,120 --> 00:40:35,650
But just concentrate
on this one.

681
00:40:35,650 --> 00:40:39,700
The firing rate goes up with
level, reaches a maximum,

682
00:40:39,700 --> 00:40:41,010
and then it declines.

683
00:40:41,010 --> 00:40:45,050
And at the highest level,
it doesn't respond anymore.

684
00:40:45,050 --> 00:40:47,520
So we don't know
what this means.

685
00:40:47,520 --> 00:40:51,420
It means somehow that
these neurons could

686
00:40:51,420 --> 00:40:56,920
tell the animal or the person,
well, the sound level is x dB.

687
00:40:58,630 --> 00:41:00,340
They can tell you
the sound frequency

688
00:41:00,340 --> 00:41:04,900
is a certain number of
kilohertz in the sound level

689
00:41:04,900 --> 00:41:08,490
if they're responding
maximally, is a certain SPO.

690
00:41:10,220 --> 00:41:12,780
It's clearly very different.

691
00:41:12,780 --> 00:41:18,820
Now, I brought that
up because people

692
00:41:18,820 --> 00:41:23,180
have looked at that
in terms of coding

693
00:41:23,180 --> 00:41:28,686
for preferred areas of space
where the neuron's response is

694
00:41:28,686 --> 00:41:29,185
maximum.

695
00:41:30,630 --> 00:41:40,000
And these are some data from
Clarey et al on azimuth level

696
00:41:40,000 --> 00:41:42,220
response areas for
cortical neurons.

697
00:41:42,220 --> 00:41:43,515
So what does that mean?

698
00:41:43,515 --> 00:41:46,990
Well, they're recording from
a single neuron in the cortex.

699
00:41:48,290 --> 00:41:50,710
And they have the
animal-- in this

700
00:41:50,710 --> 00:41:53,760
it's, a cat-- in
an anechoic room.

701
00:41:53,760 --> 00:41:57,015
And they move the sound
source around in azimuth.

702
00:41:57,950 --> 00:42:00,450
We've talked about this with
experiments in the [INAUDIBLE].

703
00:42:03,140 --> 00:42:05,830
And they study the
neuron's response

704
00:42:05,830 --> 00:42:07,680
as a function of azimuth.

705
00:42:07,680 --> 00:42:11,195
In this particular study, they
also varied the sound level.

706
00:42:12,720 --> 00:42:15,150
And so that's what
the y-axis is here.

707
00:42:15,150 --> 00:42:16,560
This is the sound level axis.

708
00:42:18,030 --> 00:42:21,380
And what we just said is
that many neurons in cortex

709
00:42:21,380 --> 00:42:23,630
have a preferred sound level.

710
00:42:23,630 --> 00:42:26,510
They start to
increase their firing.

711
00:42:26,510 --> 00:42:28,960
And that's what's meant
by this shading here,

712
00:42:28,960 --> 00:42:32,380
the dark of the shading,
the higher the firing rate.

713
00:42:32,380 --> 00:42:36,420
And then at high sound
levels, the firing trails off,

714
00:42:36,420 --> 00:42:37,880
or it goes down to zero.

715
00:42:37,880 --> 00:42:41,096
This is a level function
for these different neurons.

716
00:42:42,968 --> 00:42:47,720
These are recordings now
from the auditory cortex

717
00:42:47,720 --> 00:42:50,750
within a single column.

718
00:42:50,750 --> 00:42:56,350
So shown here is the electrode
penetration going from unit

719
00:42:56,350 --> 00:42:58,960
one down two unit 10.

720
00:42:58,960 --> 00:43:03,415
And those are indicated here,
unit 1, down to unit 10.

721
00:43:05,030 --> 00:43:08,380
And the cortical layers are
indicated by the Roman numerals

722
00:43:08,380 --> 00:43:10,140
1 through 6 here.

723
00:43:10,140 --> 00:43:12,950
And they're recording
these 10 different neurons

724
00:43:12,950 --> 00:43:14,650
from a given cortical column.

725
00:43:17,110 --> 00:43:20,300
And what's impressive
about these studies is

726
00:43:20,300 --> 00:43:24,260
that the azimuth
of each of these

727
00:43:24,260 --> 00:43:27,230
neurons, where it
prefers in space,

728
00:43:27,230 --> 00:43:31,740
it is it a certain
azimuth, 45 to 90 degrees.

729
00:43:33,610 --> 00:43:35,670
And that's as if
you were recording

730
00:43:35,670 --> 00:43:37,890
from the left auditory cortex.

731
00:43:37,890 --> 00:43:41,214
45 degrees would be
over here, to 90 degrees

732
00:43:41,214 --> 00:43:42,380
would be straight over here.

733
00:43:43,580 --> 00:43:45,170
So these neurons
are going to respond

734
00:43:45,170 --> 00:43:48,980
when the speaker is in
a position over here.

735
00:43:48,980 --> 00:43:53,970
And they're going to respond
when there's a moderate sound

736
00:43:53,970 --> 00:43:55,110
level.

737
00:43:55,110 --> 00:43:58,980
Not the lowest level, and
not the highest level.

738
00:43:58,980 --> 00:44:01,990
So they prefer a
certain sound level.

739
00:44:01,990 --> 00:44:06,100
And within a given column,
almost all of the neurons

740
00:44:06,100 --> 00:44:09,335
have similar types of
azimuth level functions.

741
00:44:11,030 --> 00:44:12,950
And remember, before
we said that these all

742
00:44:12,950 --> 00:44:14,262
have about the same CF.

743
00:44:15,468 --> 00:44:20,490
So a second thing that is common
to units in a given column

744
00:44:20,490 --> 00:44:24,760
and auditory cortex is their
azimuth level response areas.

745
00:44:24,760 --> 00:44:26,180
And that's shown
very nicely here.

746
00:44:28,490 --> 00:44:33,360
These type of data suggests
that maybe these neurons

747
00:44:33,360 --> 00:44:36,590
play some kind of
role in telling us

748
00:44:36,590 --> 00:44:38,290
where a sound is coming from.

749
00:44:39,600 --> 00:44:41,630
Like without this
column, maybe we really

750
00:44:41,630 --> 00:44:44,100
wouldn't know that sound
sources were located

751
00:44:44,100 --> 00:44:47,980
at 45 to 90 degrees over there
on the contralateral hemifield.

752
00:44:49,680 --> 00:44:55,435
So there's been a lot of work
on auditory cortex and sound

753
00:44:55,435 --> 00:44:55,976
localization.

754
00:44:58,340 --> 00:45:00,420
And I want to get into it here.

755
00:45:00,420 --> 00:45:07,290
So how do experimenters
test behaviorally

756
00:45:07,290 --> 00:45:10,040
for how an animal
can localize sound?

757
00:45:11,280 --> 00:45:13,980
Well, this is the
formal way to do it.

758
00:45:13,980 --> 00:45:15,455
Here's an experimental animal.

759
00:45:16,490 --> 00:45:19,715
It's going to a speaker
that emitted a sound.

760
00:45:21,370 --> 00:45:25,600
Before, it had been sitting
in this central position

761
00:45:25,600 --> 00:45:27,110
in the testing cage here.

762
00:45:28,640 --> 00:45:30,980
Waiting maybe cued by
a light that's saying,

763
00:45:30,980 --> 00:45:32,460
the trial's about to start.

764
00:45:32,460 --> 00:45:33,925
And the animal then listens.

765
00:45:35,800 --> 00:45:40,300
And this speaker up here
near b didn't emit the sound.

766
00:45:40,300 --> 00:45:42,650
But the speaker down
here did emit the sound.

767
00:45:42,650 --> 00:45:44,980
And then the animal is
trained to go to the speaker

768
00:45:44,980 --> 00:45:47,540
that it had heard
emit the sound.

769
00:45:47,540 --> 00:45:51,690
If it does that correctly,
there's a little food reward

770
00:45:51,690 --> 00:45:54,745
area down below the speaker, and
the animal gets a food reward.

771
00:45:54,745 --> 00:45:56,290
And the animals
are food deprived,

772
00:45:56,290 --> 00:45:58,015
so they're motivated
to do this task.

773
00:45:59,250 --> 00:46:01,430
In this case, the
speakers can be moved.

774
00:46:03,340 --> 00:46:05,710
So you have removable speakers.

775
00:46:05,710 --> 00:46:10,060
Or you can have an array of
speakers, as indicated here.

776
00:46:11,450 --> 00:46:15,160
And the animal has to choose
which of the several speakers

777
00:46:15,160 --> 00:46:19,350
emitted the sound, and
go to the correct one

778
00:46:19,350 --> 00:46:21,480
to get the food reward.

779
00:46:21,480 --> 00:46:23,260
And this has been
done with a variety

780
00:46:23,260 --> 00:46:24,360
of experimental animals.

781
00:46:25,690 --> 00:46:27,390
In this case, it
looks like a cat.

782
00:46:27,390 --> 00:46:29,818
But they've also tested
rats and monkeys.

783
00:46:33,610 --> 00:46:36,960
Now, there's a couple things
you have to worry about here.

784
00:46:36,960 --> 00:46:40,460
You have to worry if your
sound is on a long time

785
00:46:40,460 --> 00:46:42,750
that the animal isn't cheating.

786
00:46:42,750 --> 00:46:44,990
And one way of cheating
would be for the animal

787
00:46:44,990 --> 00:46:48,510
to sit still here, and
listen to the sound,

788
00:46:48,510 --> 00:46:50,710
and then move a
little bit, maybe just

789
00:46:50,710 --> 00:46:52,620
by bending over, saying, OK.

790
00:46:52,620 --> 00:46:55,270
If I moved over here,
did the sound get louder?

791
00:46:55,270 --> 00:46:57,480
Oh, that means the sound
is coming from this side.

792
00:46:59,490 --> 00:47:02,000
So generally, these
tests are using

793
00:47:02,000 --> 00:47:05,555
pretty short stimulae--
50, 100 milliseconds.

794
00:47:06,740 --> 00:47:08,930
And during that
period, the animal

795
00:47:08,930 --> 00:47:13,560
doesn't have a chance to move
and sample the sound field.

796
00:47:15,630 --> 00:47:19,060
So what are the data in
normal hearing animals?

797
00:47:20,260 --> 00:47:26,200
75% correct at distinguishing
which of the speakers

798
00:47:26,200 --> 00:47:29,530
have emitted the
sound at 5 degrees.

799
00:47:29,530 --> 00:47:33,430
So in that case, it's
when the movable speaker's

800
00:47:33,430 --> 00:47:34,480
just 5 degrees.

801
00:47:35,820 --> 00:47:38,110
With a, in this
case, 500 millisecond

802
00:47:38,110 --> 00:47:40,420
long spectrally complex stimuli.

803
00:47:40,420 --> 00:47:44,980
So animals are not as good at
this sound localization task.

804
00:47:44,980 --> 00:47:48,530
In human performance, I think
back a few lectures ago,

805
00:47:48,530 --> 00:47:50,960
we talked about the minimum
audible angle in humans

806
00:47:50,960 --> 00:47:53,230
being a couple of degrees.

807
00:47:53,230 --> 00:47:54,155
Maybe 1 degree.

808
00:47:55,370 --> 00:47:57,475
In this case, the animal's
going to 5 degrees.

809
00:47:57,475 --> 00:47:59,900
So it's not quite as good.

810
00:48:01,340 --> 00:48:07,400
The animal can still do
the task surprisingly,

811
00:48:07,400 --> 00:48:10,790
even if it has just
one functioning ear.

812
00:48:12,350 --> 00:48:13,615
And how is that possible?

813
00:48:13,615 --> 00:48:15,570
Well, the animal's pinna.

814
00:48:15,570 --> 00:48:18,090
We talked about the
external ear or pinna,

815
00:48:18,090 --> 00:48:20,345
providing you some
nice spectral queues.

816
00:48:21,740 --> 00:48:25,860
And it looks like the spectrally
complex stimuli are being used.

817
00:48:25,860 --> 00:48:27,455
So those spectra
cues are available.

818
00:48:28,470 --> 00:48:31,940
But the minimum audible angle's
more like 10 to 12 degrees,

819
00:48:31,940 --> 00:48:34,690
with just one ear with
a good pinna on it.

820
00:48:36,150 --> 00:48:38,310
So the best performance
is with two ears.

821
00:48:39,400 --> 00:48:41,600
Now, why am I going
over this paradigm?

822
00:48:41,600 --> 00:48:46,230
Well, people have then
taken experimental animals

823
00:48:46,230 --> 00:48:47,810
and studied them after lesions.

824
00:48:49,450 --> 00:48:52,740
And we're talking about
the auditory cortex.

825
00:48:52,740 --> 00:48:58,670
So let's look at the results of
lesioning the auditory cortex

826
00:48:58,670 --> 00:49:00,130
on sound localization.

827
00:49:00,130 --> 00:49:03,190
So looks like in
this study, they're

828
00:49:03,190 --> 00:49:04,706
using the array of loudspeakers.

829
00:49:07,990 --> 00:49:12,190
The lesion is located
in the auditory cortex.

830
00:49:12,190 --> 00:49:15,980
And this is the right
side of the cortex.

831
00:49:15,980 --> 00:49:18,880
This is the occipital or
back part of the cortex.

832
00:49:18,880 --> 00:49:20,360
This is the front of the cortex.

833
00:49:20,360 --> 00:49:22,300
So the lesion is made
on the right side.

834
00:49:24,940 --> 00:49:27,460
When a lesion is made
on the right side

835
00:49:27,460 --> 00:49:30,480
of the auditory
cortex, the animal

836
00:49:30,480 --> 00:49:34,710
has problems localizing sound
in the opposite hemifield.

837
00:49:36,470 --> 00:49:38,670
So the lesion is made
on the right side.

838
00:49:40,110 --> 00:49:44,110
The animal doesn't know if it's
this speaker, this speaker,

839
00:49:44,110 --> 00:49:46,250
this speaker, or
this speaker that's

840
00:49:46,250 --> 00:49:50,170
emitting the sound as a random
performance on that side.

841
00:49:50,170 --> 00:49:53,250
However, the animal
can distinguish

842
00:49:53,250 --> 00:49:57,630
between the speakers in
the hemifield ipsilateral

843
00:49:57,630 --> 00:50:04,280
to the lesion, which suggests
that the intact auditory

844
00:50:04,280 --> 00:50:06,990
cortex is mediating
that behavior.

845
00:50:06,990 --> 00:50:09,240
So there's a deficit
contralaterally

846
00:50:09,240 --> 00:50:10,520
in the opposite hemifield.

847
00:50:11,900 --> 00:50:15,240
And as you can see from this
lesion, which was located smack

848
00:50:15,240 --> 00:50:20,660
in the middle of A1,
A1 lesions effectively

849
00:50:20,660 --> 00:50:22,810
knock out sound
localization behavior.

850
00:50:25,530 --> 00:50:28,050
So these early studies
suggested that A1

851
00:50:28,050 --> 00:50:32,020
is critically important, and
is necessary for correct sound

852
00:50:32,020 --> 00:50:33,125
localization behavior.

853
00:50:34,370 --> 00:50:36,770
And in these early
studies, the lesions

854
00:50:36,770 --> 00:50:41,610
were actually made surgically by
taking out some cortex tissue.

855
00:50:42,990 --> 00:50:47,510
And they became, as time
went on in the mid1980s,

856
00:50:47,510 --> 00:50:49,580
the lesion studies
became more elegant

857
00:50:49,580 --> 00:50:56,275
in that before the lesion was
made, frequency mapping of A1

858
00:50:56,275 --> 00:50:56,775
was made.

859
00:50:59,180 --> 00:51:01,680
And this frequency
mapping is shown here.

860
00:51:01,680 --> 00:51:04,766
This is best frequency, or CF.

861
00:51:04,766 --> 00:51:06,065
Those terms are synonymous.

862
00:51:07,770 --> 00:51:10,015
And this is distance
along the cortex.

863
00:51:11,580 --> 00:51:16,440
And not the entire auditory
cortex, field A1, was removed,

864
00:51:16,440 --> 00:51:21,470
but just a particular part,
just a little distance here.

865
00:51:21,470 --> 00:51:22,930
And it was known
from the mapping

866
00:51:22,930 --> 00:51:26,690
what CF was affected
by the lesion.

867
00:51:26,690 --> 00:51:28,915
And the other CFs
were left intact.

868
00:51:30,850 --> 00:51:34,570
You can test the animal for
any frequency you want to.

869
00:51:34,570 --> 00:51:38,830
So you can test for
frequencies in the intact area,

870
00:51:38,830 --> 00:51:42,430
or you can test for frequencies
in the lesioned area.

871
00:51:42,430 --> 00:51:48,180
And it was shown very clearly
then by plotting performance.

872
00:51:48,180 --> 00:51:50,390
This is a performance
axis, where

873
00:51:50,390 --> 00:51:52,640
downward is very accurate.

874
00:51:52,640 --> 00:51:56,070
This must be the number
of mistakes made.

875
00:51:56,070 --> 00:51:58,930
So 0 is no mistakes made.

876
00:51:58,930 --> 00:52:01,725
At the low frequencies
where the cortex is intact.

877
00:52:03,540 --> 00:52:06,910
At the midfrequencies, where
the cortex is lesioned,

878
00:52:06,910 --> 00:52:08,210
performance goes to chance.

879
00:52:09,720 --> 00:52:11,210
And then at the
high frequencies,

880
00:52:11,210 --> 00:52:13,910
again, where the
cortex is intact,

881
00:52:13,910 --> 00:52:17,310
the performance gets
very few errors.

882
00:52:17,310 --> 00:52:18,010
It's very good.

883
00:52:20,820 --> 00:52:24,760
So this elegant
experiment shows then

884
00:52:24,760 --> 00:52:29,250
that sound localization
proceeds by frequency

885
00:52:29,250 --> 00:52:30,210
independent channels.

886
00:52:32,430 --> 00:52:35,830
That is, the part of A1 that's
responsive to low frequencies

887
00:52:35,830 --> 00:52:39,850
is mediating low sound
frequency localization.

888
00:52:39,850 --> 00:52:42,670
And the part that's
responsive to high frequencies

889
00:52:42,670 --> 00:52:44,749
is mediating high frequency
sound localization.

890
00:52:44,749 --> 00:52:46,165
So a very beautiful
demonstration.

891
00:52:49,640 --> 00:52:54,050
Now, these lesions were
done with the techniques

892
00:52:54,050 --> 00:52:55,390
available at the time.

893
00:52:55,390 --> 00:52:59,005
And it's very simple to go in
and destroy a part of cortex.

894
00:53:00,450 --> 00:53:04,660
And for one reason
or another, people

895
00:53:04,660 --> 00:53:10,620
decided to revisit these
lesion experiments,

896
00:53:10,620 --> 00:53:12,370
even though they
were very convincing,

897
00:53:12,370 --> 00:53:14,595
done in many, many
different species.

898
00:53:15,850 --> 00:53:19,540
They decided to revisit them
with a completely different way

899
00:53:19,540 --> 00:53:20,395
of making a lesion.

900
00:53:22,510 --> 00:53:32,105
And this is the method of
inactivation by cooling.

901
00:53:34,610 --> 00:53:37,330
And maybe many of you
have heard of this.

902
00:53:37,330 --> 00:53:40,130
This might be the auditory
cortex, the surface.

903
00:53:40,130 --> 00:53:42,630
This is layer 1 and the
different cortical layers.

904
00:53:44,820 --> 00:53:48,590
The way you can
inactivate the cortex

905
00:53:48,590 --> 00:54:00,690
by cooling is by taking a piece
of tubing that has cooling

906
00:54:00,690 --> 00:54:02,820
fluid, or if you will, coolant.

907
00:54:06,020 --> 00:54:11,930
And you can put this coolant
with a pump through this tube,

908
00:54:11,930 --> 00:54:14,895
and you can lay the tube right
on the surface of the cortex.

909
00:54:16,010 --> 00:54:19,150
And obviously, as the
coolant comes through here,

910
00:54:19,150 --> 00:54:22,220
it's going to cool down
first the top layer

911
00:54:22,220 --> 00:54:25,270
of the cortex, and
then the lower layers.

912
00:54:25,270 --> 00:54:28,540
And finally, all of the cortex.

913
00:54:28,540 --> 00:54:33,760
And you can assure yourself that
this cooling has inactivated

914
00:54:33,760 --> 00:54:36,655
the cortex by doing things like
recording evoked potentials.

915
00:54:38,990 --> 00:54:42,260
And what's elegant about
the cooling experiments

916
00:54:42,260 --> 00:54:46,120
is that you can reverse them.

917
00:54:46,120 --> 00:54:48,230
So I don't know
if this is a word,

918
00:54:48,230 --> 00:54:53,410
but instead of coolant,
you can use a warmant,

919
00:54:53,410 --> 00:54:55,320
and restore this to
body temperature.

920
00:54:58,286 --> 00:54:59,910
And responses come back.

921
00:55:02,310 --> 00:55:06,600
And what's very impressive is
that these kinds of experiments

922
00:55:06,600 --> 00:55:09,570
can be done in animals
that are actually doing

923
00:55:09,570 --> 00:55:13,030
a behavioral task,
localizing sound.

924
00:55:15,650 --> 00:55:20,310
And clearly, those experiments
confirm these earlier lesion

925
00:55:20,310 --> 00:55:23,380
experiments that
if you cool A1, you

926
00:55:23,380 --> 00:55:26,510
get a deficit for a
localization ability

927
00:55:26,510 --> 00:55:27,890
in the contralateral hemifield.

928
00:55:29,040 --> 00:55:34,180
However, they have also come
up with an interesting result

929
00:55:34,180 --> 00:55:40,450
in that if you cool
some other fields,

930
00:55:40,450 --> 00:55:42,975
you also change sound
localization behavior.

931
00:55:45,820 --> 00:55:53,220
Field PAF, when cooled, also
interrupts sound localization

932
00:55:53,220 --> 00:55:53,820
behavior.

933
00:55:53,820 --> 00:55:55,190
So that's posterior to A1.

934
00:55:57,070 --> 00:56:01,440
And a small field
that's not named, but is

935
00:56:01,440 --> 00:56:06,320
right on top of the anterior
ectosylvian sulcus, the AES.

936
00:56:08,000 --> 00:56:11,800
When that area is cooled,
sound localization behavior's

937
00:56:11,800 --> 00:56:12,445
also disrupted.

938
00:56:14,010 --> 00:56:16,620
So I used to be
able to say, anyone

939
00:56:16,620 --> 00:56:20,770
who performs this critical
function, sound localization,

940
00:56:20,770 --> 00:56:23,260
well, not so sure
about it anymore.

941
00:56:23,260 --> 00:56:25,940
Because if you cool
these other fields,

942
00:56:25,940 --> 00:56:29,870
you have a disruption of
sound localization behavior.

943
00:56:29,870 --> 00:56:34,430
If you cool any of the other
fields like A2, of VPAF,

944
00:56:34,430 --> 00:56:36,482
or most of the
AAF, you don't get

945
00:56:36,482 --> 00:56:37,815
an interruption of the behavior.

946
00:56:38,920 --> 00:56:43,070
So what do we take
home from that?

947
00:56:43,070 --> 00:56:46,430
Well, it seems like there
are several fields that

948
00:56:46,430 --> 00:56:48,830
are important in sound
localization behavior.

949
00:56:50,320 --> 00:56:59,230
Looks like A1, p, or PAF
as it's sometimes called,

950
00:56:59,230 --> 00:57:07,565
and region near the
anterior ectosylvian sulcus

951
00:57:07,565 --> 00:57:08,315
are all important.

952
00:57:09,530 --> 00:57:11,590
And it's a little
bit controversial,

953
00:57:11,590 --> 00:57:14,720
why the old lesions
didn't actually show this.

954
00:57:14,720 --> 00:57:16,140
It seemed like in
the old lesions,

955
00:57:16,140 --> 00:57:19,060
there were some studies which
said, if you leave A1 intact,

956
00:57:19,060 --> 00:57:22,540
and you lesion all the
other cortical fields,

957
00:57:22,540 --> 00:57:24,670
the animal can
still do the task.

958
00:57:24,670 --> 00:57:27,100
That doesn't really fit with
the cooling results, which

959
00:57:27,100 --> 00:57:29,600
has several fields
that are important.

960
00:57:31,000 --> 00:57:35,230
So don't let me leave
you with the idea

961
00:57:35,230 --> 00:57:40,976
that what these fields do
is only sound localization

962
00:57:40,976 --> 00:57:41,475
behavior.

963
00:57:42,840 --> 00:57:45,790
So A1 may be involved
then hundreds

964
00:57:45,790 --> 00:57:48,600
of other tasks related
to our sense of hearing.

965
00:57:49,690 --> 00:57:51,545
It's also involved in
sound localization.

966
00:57:53,160 --> 00:57:56,400
So that's the right
way to think about it,

967
00:57:56,400 --> 00:57:58,910
that these fields
probably do many things.

968
00:57:58,910 --> 00:58:03,310
And I think if you got the gist
of what Doctor Schiller talked

969
00:58:03,310 --> 00:58:07,140
about in vision, he's not a
big fan of this little area

970
00:58:07,140 --> 00:58:09,090
of cortex does this
little function.

971
00:58:09,090 --> 00:58:11,250
And over here, this little
area does this function.

972
00:58:11,250 --> 00:58:16,950
He's more of a believer in
holistic cortex function

973
00:58:16,950 --> 00:58:20,780
where to do a task, you
employ a lot of cortex;

974
00:58:20,780 --> 00:58:23,195
auditory cortex if you're
doing an auditory task.

975
00:58:24,210 --> 00:58:26,840
And perhaps the more
difficult a task is,

976
00:58:26,840 --> 00:58:28,230
the motor cortex you use.

977
00:58:29,460 --> 00:58:30,940
We'll see an evidence of that.

978
00:58:30,940 --> 00:58:35,110
And next time, when we talk
about language processing

979
00:58:35,110 --> 00:58:36,800
in humans from imaging studies.

980
00:58:38,960 --> 00:58:44,760
So what else does cortex do
besides sound localization?

981
00:58:51,424 --> 00:58:53,870
Oh, I forgot to talk
about auditory cortex

982
00:58:53,870 --> 00:58:57,300
in humans, how many
tonotopic fields there are.

983
00:58:57,300 --> 00:59:02,895
So I brought this nice
model of the primate brain.

984
00:59:04,170 --> 00:59:07,265
And where is auditory
cortex in humans?

985
00:59:08,470 --> 00:59:11,510
So this is a slice
of the brain, as

986
00:59:11,510 --> 00:59:13,320
if you were to cut it like this.

987
00:59:13,320 --> 00:59:15,320
And look at one slice.

988
00:59:15,320 --> 00:59:19,000
And this is the right
and left temporal lobes.

989
00:59:19,000 --> 00:59:22,640
So in the primate, you have
actually a separate lobe

990
00:59:22,640 --> 00:59:24,320
of the brain called
the temporal lobe.

991
00:59:25,860 --> 00:59:30,810
And in the temporal lobe,
you have to-- the primate

992
00:59:30,810 --> 00:59:32,880
is a little bit more
difficult to examine

993
00:59:32,880 --> 00:59:37,490
than the cat-- you have to pull
down the sylvian fissure, which

994
00:59:37,490 --> 00:59:39,660
is between-- separates
the temporal lobe

995
00:59:39,660 --> 00:59:42,150
from the parietal
cortex up here,

996
00:59:42,150 --> 00:59:47,470
and look on the superior
surface of the temporal lobe,

997
00:59:47,470 --> 00:59:48,740
and find the sight of A1.

998
00:59:50,720 --> 00:59:55,560
And on that superior
temporal lobe surface,

999
00:59:55,560 --> 01:00:00,150
you have a little gyrus that was
examined by an animus Heschl.

1000
01:00:00,150 --> 01:00:02,810
And Heschl's gyrus in
humans is the site of A1.

1001
01:00:05,750 --> 01:00:10,670
Some humans actually
have 2 Heschl's, gyri,

1002
01:00:10,670 --> 01:00:14,710
and they have their
A1 either on one

1003
01:00:14,710 --> 01:00:19,710
or both of the Heschl's gyri.

1004
01:00:20,890 --> 01:00:26,090
Now, looking at it from the
side view, in the temporal lobe,

1005
01:00:26,090 --> 01:00:29,060
you have the 3 big gyri.

1006
01:00:29,060 --> 01:00:33,520
Superior temporal gyrus,
inferior temporal gyrus--

1007
01:00:33,520 --> 01:00:36,980
sorry, middle temporal gyrus,
and inferior temporal gyrus.

1008
01:00:36,980 --> 01:00:40,090
And so A1 is on the
superior surface

1009
01:00:40,090 --> 01:00:43,725
of the superior temporal
gyrus on a little bitty gyrus

1010
01:00:43,725 --> 01:00:45,140
called Heschls.

1011
01:00:45,140 --> 01:00:48,050
So I'm going to pass
this model around.

1012
01:00:48,050 --> 01:00:51,060
And A1 is indicated by a little
piece of yellow tape there.

1013
01:00:51,060 --> 01:00:53,460
You can take a look at it.

1014
01:00:53,460 --> 01:00:58,740
And so that area,
Heschl's gyrus,

1015
01:00:58,740 --> 01:01:01,350
lights up very nicely
in imaging studies

1016
01:01:01,350 --> 01:01:04,070
when you present
auditory stimuli.

1017
01:01:04,070 --> 01:01:08,890
And so here's an imaging study
where the imaging plane was

1018
01:01:08,890 --> 01:01:13,530
parallel to the sylvian
fissure or sylvian sulcus,

1019
01:01:13,530 --> 01:01:16,750
and it's capturing just this
superior temporal gyrus.

1020
01:01:17,900 --> 01:01:19,620
And the plane is
looking down here.

1021
01:01:19,620 --> 01:01:23,200
And on Heschl's gyrus, you
see the left and the right

1022
01:01:23,200 --> 01:01:27,020
in this case, and just the
right in this case lighting up.

1023
01:01:29,020 --> 01:01:33,220
And you can use, of course,
different frequency sounds.

1024
01:01:36,260 --> 01:01:39,700
And change the
frequencies of those

1025
01:01:39,700 --> 01:01:44,640
sounds in a progression
from high frequencies

1026
01:01:44,640 --> 01:01:45,590
to low frequencies.

1027
01:01:46,630 --> 01:01:48,650
And you can draw
the progressions

1028
01:01:48,650 --> 01:01:51,140
that you see in imaging
signals, and that's

1029
01:01:51,140 --> 01:01:52,595
what's drawn with these arrows.

1030
01:01:53,930 --> 01:01:56,785
This is a MIT thesis
by Tom Talavage.

1031
01:01:58,350 --> 01:02:00,520
And he showed that
there were at least 1,

1032
01:02:00,520 --> 01:02:07,110
2, 3, 4, 5 clear progressions
of frequency sensitivity,

1033
01:02:07,110 --> 01:02:09,800
as if you were progressing
along tonotopically

1034
01:02:09,800 --> 01:02:13,660
mapped auditory cortical
fields in the human.

1035
01:02:13,660 --> 01:02:19,130
So remember, we saw 4
tonotopically mapped fields

1036
01:02:19,130 --> 01:02:19,860
in the cat.

1037
01:02:19,860 --> 01:02:24,300
And here we have at least
4, perhaps 5 in humans.

1038
01:02:25,760 --> 01:02:28,935
This one is labeled HG,
that's Heschl's gyrus.

1039
01:02:30,670 --> 01:02:33,100
And that's probably
primary auditory cortex.

1040
01:02:34,800 --> 01:02:38,690
This is a view looking down
on the superior surface

1041
01:02:38,690 --> 01:02:39,720
of the temporal lobe.

1042
01:02:39,720 --> 01:02:42,380
And this is what's
called an inflated view.

1043
01:02:43,420 --> 01:02:46,020
So it's like taking
that cortex model

1044
01:02:46,020 --> 01:02:48,370
and blowing it up
like a balloon.

1045
01:02:48,370 --> 01:02:52,260
And so the gyri are indicated
by the lighter shading,

1046
01:02:52,260 --> 01:02:57,125
and the sulci are indicated
by the more dark shading.

1047
01:02:57,125 --> 01:02:59,910
So that's what that is.

1048
01:02:59,910 --> 01:03:01,260
These are the dimensions here.

1049
01:03:01,260 --> 01:03:03,740
Posterior lateral
is that direction.

1050
01:03:03,740 --> 01:03:08,270
So we have multiple
tonotopically organized areas

1051
01:03:08,270 --> 01:03:09,955
in human auditory cortex.

1052
01:03:11,580 --> 01:03:17,190
Now, the paper that we read for
today's class by Penagos et al

1053
01:03:17,190 --> 01:03:25,800
talks about a center near A1,
near Heschl's gyrus, which

1054
01:03:25,800 --> 01:03:30,380
lights up in imaging studies
when the subject is presented

1055
01:03:30,380 --> 01:03:33,880
with sounds that have a
strong cessation of pitch.

1056
01:03:34,990 --> 01:03:37,250
So we talked about pitch
a little bit earlier

1057
01:03:37,250 --> 01:03:37,800
in the class.

1058
01:03:37,800 --> 01:03:40,130
And this is the
slide that I showed.

1059
01:03:40,130 --> 01:03:42,420
And it has a
different title now.

1060
01:03:42,420 --> 01:03:47,622
I think it was titled something
like complicated sounds,

1061
01:03:47,622 --> 01:03:48,580
or something like that.

1062
01:03:48,580 --> 01:03:52,510
Well, a complex sound
is simply a sound

1063
01:03:52,510 --> 01:03:54,035
that has multiple frequencies.

1064
01:03:55,560 --> 01:03:57,970
And so the stimulate
used in the paper

1065
01:03:57,970 --> 01:04:00,890
are complex sounds in that
they have multiple frequencies.

1066
01:04:02,750 --> 01:04:05,060
Earlier, we talk
about this in terms

1067
01:04:05,060 --> 01:04:07,700
of the context of
musical sounds.

1068
01:04:07,700 --> 01:04:11,796
Musical sounds almost always
have a fundamental frequency,

1069
01:04:11,796 --> 01:04:13,296
and then a whole
bunch of harmonics.

1070
01:04:15,120 --> 01:04:19,600
And to have a strong
sensation of pitch,

1071
01:04:19,600 --> 01:04:24,930
these musical sounds have
a very tight relationship

1072
01:04:24,930 --> 01:04:27,050
of the fundamental
and the harmonics.

1073
01:04:27,050 --> 01:04:29,550
They can't be a
random relationship.

1074
01:04:29,550 --> 01:04:32,150
They actually have to be
multiples of the fundamental.

1075
01:04:33,260 --> 01:04:52,140
For example, this complex of
tones, 100 hertz, 200 hertz,

1076
01:04:52,140 --> 01:04:59,850
300 hertz, 400, and so on,
are multiples of one another.

1077
01:04:59,850 --> 01:05:04,080
But if you had the
fundamental be 100,

1078
01:05:04,080 --> 01:05:10,070
the next harmonic be 150,
the next harmonic be 190,

1079
01:05:10,070 --> 01:05:13,070
the next harmonic be
230, where they're not

1080
01:05:13,070 --> 01:05:15,870
multiples of one another,
that stimulus would not

1081
01:05:15,870 --> 01:05:18,576
have a strong pitch
associated with it.

1082
01:05:21,240 --> 01:05:25,680
These musical sounds are
interesting because they

1083
01:05:25,680 --> 01:05:27,580
have a strong pitch.

1084
01:05:27,580 --> 01:05:29,790
The pitch is almost
always related

1085
01:05:29,790 --> 01:05:32,920
to the lowest, or fundamental
frequency of them.

1086
01:05:34,060 --> 01:05:36,050
And the pitch is very invariant.

1087
01:05:36,050 --> 01:05:39,860
As long as you have this nice
pattern of harmonics that

1088
01:05:39,860 --> 01:05:43,000
are related to each
other by multiples,

1089
01:05:43,000 --> 01:05:48,140
the pitch of this, no, which
I think is a piano note,

1090
01:05:48,140 --> 01:05:52,660
and the pitch of this note,
which-- let's do these two.

1091
01:05:52,660 --> 01:05:55,924
This is a guitar sound, and
this is an alto saxophone sound

1092
01:05:55,924 --> 01:05:57,340
where the fundamental
is the same.

1093
01:05:58,800 --> 01:06:01,760
The harmonic amplitude
is completely different.

1094
01:06:03,550 --> 01:06:06,755
But we recognize them as
playing the same pitch.

1095
01:06:08,690 --> 01:06:11,240
You can play a lot
around with the amplitude

1096
01:06:11,240 --> 01:06:12,250
of these harmonics.

1097
01:06:12,250 --> 01:06:13,415
I drew them all the same.

1098
01:06:14,640 --> 01:06:16,995
But clearly, they
can be any jumble

1099
01:06:16,995 --> 01:06:21,080
of pattern, as long as they're
multiples of one another.

1100
01:06:21,080 --> 01:06:24,290
You hear this as having
the same pitch as that.

1101
01:06:24,290 --> 01:06:26,440
And pitch is very
invariant to things

1102
01:06:26,440 --> 01:06:28,710
like where the sound
is coming from.

1103
01:06:28,710 --> 01:06:34,340
Pitch is invariant to how
high and level the sound is.

1104
01:06:34,340 --> 01:06:38,400
So pitch is a very fundamental
attribute of the sound.

1105
01:06:38,400 --> 01:06:41,340
Defined in the
psychophysics textbook

1106
01:06:41,340 --> 01:06:46,320
is, pitch is that attribute
of auditory sensation in terms

1107
01:06:46,320 --> 01:06:49,540
of which sounds may be
ordered on a musical scale.

1108
01:06:49,540 --> 01:06:53,570
So this one's low, this one's
middle, and this one's high.

1109
01:06:53,570 --> 01:06:56,910
And so when cochlear implant
users are programmed first,

1110
01:06:56,910 --> 01:07:00,780
they take this electrode,
and they stimulate it.

1111
01:07:00,780 --> 01:07:03,240
And the user says, yeah,
that sounds like a low one.

1112
01:07:03,240 --> 01:07:06,140
Then they activate
the next electrode.

1113
01:07:06,140 --> 01:07:08,765
And the audiologist
says, is this one higher,

1114
01:07:08,765 --> 01:07:10,720
or is this one lower?

1115
01:07:10,720 --> 01:07:15,090
And if it's higher, then they
route their speech processor,

1116
01:07:15,090 --> 01:07:16,875
higher frequencies,
into that electrode.

1117
01:07:19,650 --> 01:07:22,455
So they do a pitch ranking
in an auditory implants.

1118
01:07:24,240 --> 01:07:27,100
Now, pitch of the
complicated sound

1119
01:07:27,100 --> 01:07:29,130
depends strongly on the
fundamental frequency.

1120
01:07:29,130 --> 01:07:30,470
Everybody knows that.

1121
01:07:30,470 --> 01:07:34,880
Wow, you can play little
tricks in these stimuli.

1122
01:07:34,880 --> 01:07:38,180
You can do something like remove
the fundamental frequency.

1123
01:07:40,000 --> 01:07:41,450
How does that change the pitch?

1124
01:07:41,450 --> 01:07:44,335
Well, this guy becomes
the new fundamental.

1125
01:07:46,080 --> 01:07:47,510
That's what you might think.

1126
01:07:47,510 --> 01:07:50,520
But actually, removing
this fundamental

1127
01:07:50,520 --> 01:07:54,250
is just like playing around
with the amplitude of the higher

1128
01:07:54,250 --> 01:07:54,840
frequencies.

1129
01:07:54,840 --> 01:07:56,430
It doesn't change
the pitch at all.

1130
01:07:57,530 --> 01:07:59,730
And this is called the
missing fundamental.

1131
01:08:07,930 --> 01:08:11,600
And that's actually lucky
for cheap speakers that

1132
01:08:11,600 --> 01:08:12,970
might not have a very good base.

1133
01:08:14,320 --> 01:08:16,120
The fundamental is
hardly there at all,

1134
01:08:16,120 --> 01:08:18,810
but the piece still
sounds musical,

1135
01:08:18,810 --> 01:08:23,000
and it's not changed a lot.

1136
01:08:23,000 --> 01:08:24,810
Or why does that happen?

1137
01:08:24,810 --> 01:08:31,050
Well, this very nice multiples
of 100 is still present.

1138
01:08:31,050 --> 01:08:36,060
And so the temporal pattern
of all these harmonics, if you

1139
01:08:36,060 --> 01:08:43,640
add them up and look at this
thing in the time domain--

1140
01:08:43,640 --> 01:08:45,610
remember, this was
a graph frequency.

1141
01:08:47,970 --> 01:08:51,680
Whatever this looks
like, it's going

1142
01:08:51,680 --> 01:08:54,439
to repeat after 10 milliseconds.

1143
01:08:54,439 --> 01:08:56,890
Because it's period
is still 100 hertz.

1144
01:08:58,029 --> 01:09:00,660
I'm not a very good artist,
but it's going to be the same.

1145
01:09:00,660 --> 01:09:02,939
And it's going to
be the same here.

1146
01:09:02,939 --> 01:09:07,170
So each 10 milliseconds, it's
going to repeat its pattern.

1147
01:09:07,170 --> 01:09:12,989
It has the same regularity, even
if you remove the fundamental.

1148
01:09:14,300 --> 01:09:15,520
Now, you may not believe me.

1149
01:09:15,520 --> 01:09:16,978
So let me give you
a demonstration.

1150
01:09:18,040 --> 01:09:19,580
And in this
demonstration, there's

1151
01:09:19,580 --> 01:09:21,925
a whole bunch of harmonics
presented at first.

1152
01:09:23,250 --> 01:09:25,180
And then in the
second presentation,

1153
01:09:25,180 --> 01:09:27,140
they remove the fundamental.

1154
01:09:27,140 --> 01:09:29,490
And you should listen
to see if the pitch

1155
01:09:29,490 --> 01:09:32,990
that your ears hear
changes at all.

1156
01:09:32,990 --> 01:09:37,720
On the second presentation,
they remove this next harmonic,

1157
01:09:37,720 --> 01:09:39,010
and so on, and so forth.

1158
01:09:39,010 --> 01:09:41,310
And I think they
end up with removing

1159
01:09:41,310 --> 01:09:44,069
four different
harmonics after they

1160
01:09:44,069 --> 01:09:46,760
present the complete stimulus.

1161
01:09:46,760 --> 01:09:50,395
So listen to this demonstration
of the missing fundamental.

1162
01:09:50,395 --> 01:09:51,311
[AUDIO PLAYBACK]

1163
01:09:51,311 --> 01:09:54,007
-Pitch of the missing
fundamental, or virtual pitch.

1164
01:09:55,410 --> 01:09:59,470
You will hear a complex tone
with 10 harmonics, first

1165
01:09:59,470 --> 01:10:03,380
complete, and then with the
lower harmonics successively

1166
01:10:03,380 --> 01:10:04,165
removed.

1167
01:10:04,165 --> 01:10:05,852
Does the pitch of
the complex change?

1168
01:10:06,890 --> 01:10:08,740
The demonstration
is repeated once.

1169
01:10:11,680 --> 01:10:16,090
[SERIES OF PITCHES]

1170
01:10:32,260 --> 01:10:33,290
[END AUDIO PLAYBACK]

1171
01:10:33,290 --> 01:10:35,930
PROFESSOR: I think
it's pretty clear.

1172
01:10:35,930 --> 01:10:38,130
Does everybody want
to discuss this?

1173
01:10:38,130 --> 01:10:40,810
So when you lose
these first 1 or 2,

1174
01:10:40,810 --> 01:10:43,190
the pitch doesn't
change a great deal.

1175
01:10:43,190 --> 01:10:45,420
But by the end of
the demo, this pitch

1176
01:10:45,420 --> 01:10:46,905
is starting to
sound a lot higher.

1177
01:10:48,460 --> 01:10:51,570
So if you move some of
these around or decrease

1178
01:10:51,570 --> 01:10:54,190
their amplitude, this
one's a low fundamental.

1179
01:10:54,190 --> 01:10:55,890
The pitch doesn't
change a great deal.

1180
01:10:57,060 --> 01:11:01,820
Now, that is what
they did in the paper

1181
01:11:01,820 --> 01:11:02,895
that we read for today.

1182
01:11:04,790 --> 01:11:07,190
What they do is they
have a complex tone

1183
01:11:07,190 --> 01:11:08,650
with a whole bunch of harmonics.

1184
01:11:11,010 --> 01:11:13,830
And they do a clever
thing like they've

1185
01:11:13,830 --> 01:11:15,340
done on this demonstration.

1186
01:11:15,340 --> 01:11:17,570
They just select
some of the harmonics

1187
01:11:17,570 --> 01:11:21,470
to present to the observers
in the imaging study.

1188
01:11:23,610 --> 01:11:26,740
And the cleverness
of this study is

1189
01:11:26,740 --> 01:11:32,380
that by clever filtering
of this harmonic pattern,

1190
01:11:32,380 --> 01:11:34,720
they can give you
some stimulae that

1191
01:11:34,720 --> 01:11:37,330
have really strong
sensations of pitch.

1192
01:11:38,400 --> 01:11:40,680
Or in this one
case, which I think

1193
01:11:40,680 --> 01:11:46,040
is condition 2, a very weak
sensation of pitch because

1194
01:11:46,040 --> 01:11:48,130
of the particular
harmonics they've chosen.

1195
01:11:50,170 --> 01:11:52,450
So they have 3 stimulate.

1196
01:11:52,450 --> 01:11:54,710
They giving you a strong
sensation of pitch,

1197
01:11:54,710 --> 01:11:57,520
and one that has a very weak
sensation of pitch because

1198
01:11:57,520 --> 01:12:00,340
of the clever way
they've filtered it.

1199
01:12:00,340 --> 01:12:04,690
And further clevering,
all of these stimulae

1200
01:12:04,690 --> 01:12:06,570
have the same regularity.

1201
01:12:06,570 --> 01:12:10,000
They have the same
regularity in terms

1202
01:12:10,000 --> 01:12:11,340
of their temporal waveform.

1203
01:12:12,360 --> 01:12:14,560
So the cleverness
of this study then

1204
01:12:14,560 --> 01:12:16,660
is that the temporal
waveform hasn't

1205
01:12:16,660 --> 01:12:19,800
changed in terms
of its regularity.

1206
01:12:19,800 --> 01:12:22,270
But the subject's
impression of the pitch,

1207
01:12:22,270 --> 01:12:27,590
whether it's a strong pitch
or a weak pitch, has changed.

1208
01:12:27,590 --> 01:12:29,790
So by weak pitch,
I mean something

1209
01:12:29,790 --> 01:12:32,620
that sounds like a
noise, or a click.

1210
01:12:32,620 --> 01:12:35,230
Those stimulate don't have
strong sensation of pitch

1211
01:12:35,230 --> 01:12:36,390
because they're random.

1212
01:12:36,390 --> 01:12:39,170
They don't have this nice
pattern of harmonics.

1213
01:12:40,460 --> 01:12:46,770
So I wasn't convinced by
this verbiage and the figure.

1214
01:12:46,770 --> 01:12:51,730
So I decided I wanted to
listen to these stimuli myself.

1215
01:12:51,730 --> 01:12:54,940
And it was convenient, because
I know all three authors.

1216
01:12:54,940 --> 01:12:58,810
So Hector Penagos, when
he wrote this paper,

1217
01:12:58,810 --> 01:13:02,300
was a graduate student
in the speech and hearing

1218
01:13:02,300 --> 01:13:03,925
bioscience and
technology program.

1219
01:13:05,730 --> 01:13:08,040
Jennifer Melcher
is a faculty member

1220
01:13:08,040 --> 01:13:09,450
over at the Eaton-Peabody Lab.

1221
01:13:09,450 --> 01:13:11,860
Her office is
right next to mine.

1222
01:13:11,860 --> 01:13:16,157
And Andrew Oxenham was a
faculty member here at MIT,

1223
01:13:16,157 --> 01:13:18,115
and has since moved to
University of Minnesota.

1224
01:13:19,320 --> 01:13:23,290
So I started asking the
authors, because they're

1225
01:13:23,290 --> 01:13:25,877
all friends of mine, if
I could have the demos.

1226
01:13:25,877 --> 01:13:27,960
And one of them said, well,
it's been a long time.

1227
01:13:27,960 --> 01:13:29,350
I'm not sure I still have them.

1228
01:13:30,380 --> 01:13:32,600
And the other author,
is the second author

1229
01:13:32,600 --> 01:13:34,300
I went to-- I won't
say who it is,

1230
01:13:34,300 --> 01:13:35,680
said, I got them right away.

1231
01:13:35,680 --> 01:13:39,200
So he sent them-- that author
sent them to me right away.

1232
01:13:40,300 --> 01:13:43,420
And so I have them, and
I'll play them for you.

1233
01:13:43,420 --> 01:13:46,000
Now, you'll listen
to these stimuli.

1234
01:13:46,000 --> 01:13:47,594
And what was
surprising to me that I

1235
01:13:47,594 --> 01:13:49,260
didn't get from the
methods of the paper

1236
01:13:49,260 --> 01:13:52,540
is that they don't just keep
presenting the same thing over

1237
01:13:52,540 --> 01:13:53,060
and over.

1238
01:13:53,060 --> 01:13:54,970
The pitch actually moves around.

1239
01:13:54,970 --> 01:13:58,000
And that's one of the
nice parts of this demo

1240
01:13:58,000 --> 01:14:01,720
is that you can actually tell
that the pitch is moving around

1241
01:14:01,720 --> 01:14:04,270
in the ones with strong
sensation of pitch.

1242
01:14:05,430 --> 01:14:08,480
Second thing that
they did was, they

1243
01:14:08,480 --> 01:14:12,470
added a little bit of
background noise to these.

1244
01:14:12,470 --> 01:14:14,700
And it turns out
that when you present

1245
01:14:14,700 --> 01:14:17,410
a whole bunch of harmonics
with the speaker,

1246
01:14:17,410 --> 01:14:19,955
the speaker will introduce
a little distortion.

1247
01:14:19,955 --> 01:14:22,230
Your ear introduces distortion.

1248
01:14:22,230 --> 01:14:25,160
And they wanted to mask
that distortion out.

1249
01:14:25,160 --> 01:14:27,780
And the distortion is
pretty low in level.

1250
01:14:27,780 --> 01:14:30,240
And so this noise that's
a continuous background

1251
01:14:30,240 --> 01:14:32,570
is a pretty effective mask.

1252
01:14:32,570 --> 01:14:36,580
I think you can still hear that
these stimuli, in some cases,

1253
01:14:36,580 --> 01:14:39,517
have pretty strong
sensations of pitch.

1254
01:14:39,517 --> 01:14:41,850
So I'm going to start out
with condition one, which they

1255
01:14:41,850 --> 01:14:43,915
say has a strong
sensation of pitch.

1256
01:14:43,915 --> 01:14:46,406
And you can judge for
yourself, whether you

1257
01:14:46,406 --> 01:14:48,604
hear the pitch moving around.

1258
01:14:48,604 --> 01:14:52,076
[PITCH WITH STATIC]

1259
01:15:13,404 --> 01:15:15,388
PROFESSOR: Maybe it
will go on forever.

1260
01:15:15,388 --> 01:15:17,372
I don't know how long
it would go on for.

1261
01:15:18,390 --> 01:15:20,575
Well, anyway, could you
hear those moving around?

1262
01:15:20,575 --> 01:15:21,982
You could rank them.

1263
01:15:23,460 --> 01:15:24,460
Here's number 2.

1264
01:15:24,460 --> 01:15:27,890
[PITCH WITH STATIC]

1265
01:15:27,890 --> 01:15:30,510
PROFESSOR: It's
moving around, right?

1266
01:15:30,510 --> 01:15:32,280
Number 3.

1267
01:15:32,280 --> 01:15:35,837
[LOWER PITCH WITH STATIC]

1268
01:15:35,837 --> 01:15:36,420
PROFESSOR: OK.

1269
01:15:36,420 --> 01:15:40,170
Now, to me, those have strong
sensations of pitch, I believe.

1270
01:15:40,170 --> 01:15:41,314
I'm a believer.

1271
01:15:41,314 --> 01:15:42,730
Now, here's the
last one that they

1272
01:15:42,730 --> 01:15:44,302
say has a weak [INAUDIBLE].

1273
01:15:44,302 --> 01:15:48,166
[PITCH WITH STATIC]

1274
01:15:54,837 --> 01:15:55,420
PROFESSOR: OK.

1275
01:15:55,420 --> 01:15:59,890
So at first, I was expecting to
hear no change in pitch at all.

1276
01:15:59,890 --> 01:16:02,090
But actually, the pitches
change a little bit.

1277
01:16:02,090 --> 01:16:05,970
And so when you go back and say,
it's a weak sensation of pitch,

1278
01:16:05,970 --> 01:16:06,711
OK.

1279
01:16:06,711 --> 01:16:07,210
Right?

1280
01:16:07,210 --> 01:16:08,001
So we're believers?

1281
01:16:08,940 --> 01:16:14,350
Or does anybody want
to say-- all right.

1282
01:16:14,350 --> 01:16:16,500
What happened in
their imaging study?

1283
01:16:16,500 --> 01:16:19,380
Well, this is
pretty small figure.

1284
01:16:19,380 --> 01:16:22,160
But it summarizes the
results in that they

1285
01:16:22,160 --> 01:16:25,970
had an activation of
the circled area, which

1286
01:16:25,970 --> 01:16:31,870
is near Heschl's gyrus-- just,
I believe, anterior to it.

1287
01:16:31,870 --> 01:16:35,550
And this is the place that
had a high activation.

1288
01:16:36,630 --> 01:16:38,905
In the cases of the
stimuli, was strong.

1289
01:16:40,400 --> 01:16:43,750
Psychophysical
sensations of pitch,

1290
01:16:43,750 --> 01:16:46,270
but had a low
activation in the case

1291
01:16:46,270 --> 01:16:49,390
where there was a weak
sensation of pitch.

1292
01:16:49,390 --> 01:16:51,900
Other areas of the
brain-- for example,

1293
01:16:51,900 --> 01:16:56,860
Heschl's gyrus lit up for
all of the conditions.

1294
01:16:58,250 --> 01:17:00,540
And what's interesting
about this study

1295
01:17:00,540 --> 01:17:04,520
is they examined some of
the subcortical nuclei

1296
01:17:04,520 --> 01:17:05,770
that we've been talking about.

1297
01:17:05,770 --> 01:17:10,480
For example, the inferior
colliculus and the cochlear

1298
01:17:10,480 --> 01:17:10,980
nuclei.

1299
01:17:12,280 --> 01:17:16,750
And these are the activations
for those centers.

1300
01:17:16,750 --> 01:17:19,310
Cochlear nucleus activates
pretty much the same

1301
01:17:19,310 --> 01:17:21,680
for all of the conditions.

1302
01:17:21,680 --> 01:17:23,800
This black one is the
one that was associated

1303
01:17:23,800 --> 01:17:26,000
with the weak
sensation of pitch.

1304
01:17:26,000 --> 01:17:28,250
The inferior colliculus,
perhaps a little bit less

1305
01:17:28,250 --> 01:17:33,170
activation for that stimulus,
but not significantly so.

1306
01:17:33,170 --> 01:17:37,430
In the case of this center
for pitched salience

1307
01:17:37,430 --> 01:17:40,355
in the auditory cortex, clearly
there's a lot less activation.

1308
01:17:41,750 --> 01:17:45,320
For that stimulus with the
weak sensation of pitch,

1309
01:17:45,320 --> 01:17:48,210
about the same amount
of activation as you'd

1310
01:17:48,210 --> 01:17:50,737
find with noise bursts there.

1311
01:17:52,380 --> 01:17:54,700
So that, I think,
clearly demonstrates,

1312
01:17:54,700 --> 01:17:56,790
and other imaging
studies have clearly

1313
01:17:56,790 --> 01:18:00,690
demonstrated that this area
is the region of cortex that

1314
01:18:00,690 --> 01:18:04,180
becomes very active when
we experience stimuli

1315
01:18:04,180 --> 01:18:05,710
with strong sensations of pitch.

1316
01:18:07,220 --> 01:18:10,480
In experimental animals,
recordings from, for example,

1317
01:18:10,480 --> 01:18:13,430
the marmoset auditory cortex.

1318
01:18:13,430 --> 01:18:17,690
You find neurons in
an equivalent area

1319
01:18:17,690 --> 01:18:22,880
that respond very nicely
to harmonic complexes that

1320
01:18:22,880 --> 01:18:25,190
have strong
sensation of pitches.

1321
01:18:25,190 --> 01:18:29,030
If you remove too
many of the harmonics

1322
01:18:29,030 --> 01:18:32,905
where the pitch changes a
lot, the neurons fire less.

1323
01:18:35,070 --> 01:18:38,030
And clearly, in those
cases, remember,

1324
01:18:38,030 --> 01:18:40,540
most neurons in the
cortex are finally

1325
01:18:40,540 --> 01:18:42,550
tuned to sound frequency.

1326
01:18:42,550 --> 01:18:45,070
You can remove a whole bunch
of these lower harmonics,

1327
01:18:45,070 --> 01:18:48,970
and not change the response,
suggesting that there is really

1328
01:18:48,970 --> 01:18:53,100
signaling, that there is pitch
associated with that stimulus

1329
01:18:53,100 --> 01:18:55,614
rather than there's a
certain kind of frequency.

1330
01:18:58,850 --> 01:19:01,720
So that's the bottom
line for this study.

1331
01:19:01,720 --> 01:19:03,910
Let me just mention
a couple things that

1332
01:19:03,910 --> 01:19:07,270
make auditory
experiments difficult

1333
01:19:07,270 --> 01:19:08,920
when you're trying
to image the brain.

1334
01:19:08,920 --> 01:19:12,605
Has anybody listened
to an fMRI machine?

1335
01:19:12,605 --> 01:19:13,105
An imager?

1336
01:19:14,700 --> 01:19:16,515
Very loud, right?

1337
01:19:17,740 --> 01:19:21,310
So you have problems
with the subjects

1338
01:19:21,310 --> 01:19:25,050
listening to the stimulus that
you intend to present to them

1339
01:19:25,050 --> 01:19:29,435
rather than listening to
the imaging noise itself.

1340
01:19:29,435 --> 01:19:35,500
And so in this study,
they went to great extent

1341
01:19:35,500 --> 01:19:38,265
to try to reduce
the imaging noise.

1342
01:19:39,460 --> 01:19:45,660
The subjects were wearing
protective earmuffs,

1343
01:19:45,660 --> 01:19:47,950
and the stimulae
were loudspeakers

1344
01:19:47,950 --> 01:19:51,600
that led to those
earmuffs in long tubes.

1345
01:19:51,600 --> 01:19:55,100
Of course, you can't have a
speaker right near the ear

1346
01:19:55,100 --> 01:19:57,445
because there's a magnet
associated with the speaker.

1347
01:19:57,445 --> 01:19:59,570
So you have to have the
speaker outside the imager.

1348
01:20:01,110 --> 01:20:05,115
So they actually turned one
of the imaging pumps off.

1349
01:20:07,350 --> 01:20:11,950
There is a lot of challenge in
imaging such small structures

1350
01:20:11,950 --> 01:20:14,270
as the cochlear nuclei
in inferior colliculus.

1351
01:20:15,350 --> 01:20:18,150
And they said to improve
the detection of activation

1352
01:20:18,150 --> 01:20:20,780
in these brainstem
[? structures, ?]

1353
01:20:20,780 --> 01:20:24,050
the data were activated
using cardiac triggering.

1354
01:20:24,050 --> 01:20:25,790
So does anybody know
what that means?

1355
01:20:28,490 --> 01:20:33,630
Well, when your heart beats,
it pulses on all the arteries,

1356
01:20:33,630 --> 01:20:36,080
and actually moves
the brainstem.

1357
01:20:36,080 --> 01:20:38,670
Brainstem is so small,
it can be moved.

1358
01:20:38,670 --> 01:20:41,920
Cortex is moving too, but the
areas are generally bigger.

1359
01:20:41,920 --> 01:20:46,930
And so if you weren't going
to take care of the heartbeat,

1360
01:20:46,930 --> 01:20:48,590
the brainstem might
be imaged when

1361
01:20:48,590 --> 01:20:50,280
it was in this
position at one point.

1362
01:20:50,280 --> 01:20:53,270
And the next image might,
when it was over here.

1363
01:20:54,490 --> 01:20:57,650
So what cardiac
triggering is, they

1364
01:20:57,650 --> 01:20:59,840
record the EKG from the subject.

1365
01:21:01,050 --> 01:21:04,620
And when they see the
QRS complex, or whatever

1366
01:21:04,620 --> 01:21:07,260
waveform from the
EKG, and they say,

1367
01:21:07,260 --> 01:21:09,035
that's the time
to take the image.

1368
01:21:09,035 --> 01:21:11,720
So they only take the
image at a certain point

1369
01:21:11,720 --> 01:21:13,920
relative to the cardiac cycle.

1370
01:21:13,920 --> 01:21:15,930
So the brain, even
though it's moving,

1371
01:21:15,930 --> 01:21:17,875
it's always moved in
this certain position.

1372
01:21:19,200 --> 01:21:22,640
So that's a challenge
that's associated

1373
01:21:22,640 --> 01:21:26,270
with imaging small structures
like these brainstem nuclei.

1374
01:21:33,470 --> 01:21:35,250
So then, another
property, at least

1375
01:21:35,250 --> 01:21:37,350
of this field of
the auditory cortex,

1376
01:21:37,350 --> 01:21:42,130
is to process stimulae that
to have high pitch salience.

1377
01:21:42,130 --> 01:21:45,140
So we've had two functions
associated with auditory cortex

1378
01:21:45,140 --> 01:21:47,220
then, just as a summary here.

1379
01:21:47,220 --> 01:21:51,470
One is processing stimuli
with high pitch salience,

1380
01:21:51,470 --> 01:21:54,170
and the other is
processing sound stimuli

1381
01:21:54,170 --> 01:21:55,580
that change in location.

1382
01:21:56,770 --> 01:22:00,180
So those are the two things you
can really hang your hat on,

1383
01:22:00,180 --> 01:22:05,220
and what is done at the auditory
cortex in terms of function.

1384
01:22:06,350 --> 01:22:08,625
And for the pitch
sensitive area,

1385
01:22:08,625 --> 01:22:10,790
you have this area near A1.

1386
01:22:12,760 --> 01:22:18,700
For the localization, you
have A1, posterior field,

1387
01:22:18,700 --> 01:22:22,180
and a field near the
anterior ectosylvian sulcus,

1388
01:22:22,180 --> 01:22:23,375
as we know currently.

1389
01:22:27,280 --> 01:22:27,780
All right.

1390
01:22:27,780 --> 01:22:28,279
Questions?

1391
01:22:30,730 --> 01:22:32,735
If not, have a
good Thanksgiving.

1392
01:22:33,850 --> 01:22:37,610
Don't eat too much, or enjoy
eating too much, I guess.

1393
01:22:39,120 --> 01:22:41,410
I'll see you on Monday.