Transcript Chapter 10

Chapter 10
The Ear and Auditory System
Sound waves
If a tree falls in the woods and no one is
around to hear it ….
Distinguish between perceptual qualities
and physical qualities
The physical quality relevant for hearing is
mechanical disturbances in particular
media.
This is acoustic energy.
Robert Boyle shows that typical sounds
are vibrations in the molecules of air.
Speed of sound
343 meters/second in air
1,500 meters/second in water
5,000 meters/second in steel
Speed is constant in any given medium,
although sounds fade with increasing
distance. Signal strength declines with the
square of the distance. (double distance =
4x reduction in sound)
Echoes
Sound, unlike light, can propagate around
and through objects. This makes it more
difficult to block out sounds than light.
Objects do reflect sound waves as
echoes.
These echoes can be used in sonar
(sound navigation ranging)
Echoes
Direction of reflections make a vast
difference in acoustic properties of rooms,
e.g. concert halls.
Plaster or tile absorbs about 3% of incident
sound waves.
Carpet absorbs about 25% of incident sound
waves.
Echoes
Anechoic chambers use foam wedges to
eliminate all echoes.
These rooms have a “dead” feel to them.
Echoes do appear to enable humans to
navigate and can be used to identify
materials.
Nature of sound waves
Sound waves are variations in the density of
molecules in the air.
Sounds levels
Measured in decibels (dB).
Decibel scale is logarithmic.
dB = 20 log (p1/p0)
20 μPa ≈ softest sound humans can hear.
The Auditory System: The Ear
Outer Ear Trivia
Pinna “colors” the sounds we hear.
Ear canal is about 2.5 cm long x 7mm in
diameter.
Ear drum (tympanic membrane) can
detects sounds using a displacement of
one millionth of a centimeter.
Ear drum has surface area of 68mm2


Gerbil area = 15 mm2
Elephant area = 450 mm2
Middle Ear
Each bone is about the size of a letter
on a printed page.
Ossicles vary in size across species:
Human: 28.5 mg
Gerbil: 1.15 mg
Elephant: 335 mg
Note: Error in Figure 10.9, p. 362,
Labeling of middle ear.
Why have the ossicles?
Sound waves in air do not transmit well to
sound waves in water.
The ossicles decrease the loss in signal
strength.
The Eustachian tube maintains air
pressure on both side of the ear drum.
The acoustic reflex
The tensor tympani connects to the ear drum.
The stapedius connects to the stapes.
These muscles flex during loud noises to reduce
the response of the ossicles.
This reduces intensity of sound transmission by
the equivalent of 30 dB.
The Acoustic Reflex
Is more effective at damping low
frequencies than at damping high
frequencies.
Takes roughly 1/20 of a second to take
effect.
Perhaps reduces ability to hears one’s
own voice.
The Inner Ear
Cochlea
Three chambers:



Vestibular canal (aka scala vestibuli)
Cochlear duct (aka scala media)
Tympanic Canal (aka scala timpani)
http://www.vimm.it/cochlea/index.htm
Hair cell trivia
~20,000 total hair cells
~4,500 Inner hair cells
~15,500 outer hair cells
OHC arranged in “v”s
IHC are linearly arranged.
OHC attached to tectorial membrane;
IHC are not attached.
OHC amplify signals of IHC.
Background to Bekesy
Temporal theory: Basilar membrane
vibrates at frequency of incoming sound
waves.
Place theory: Basilar membrane vibrates
at a place corresponding to the frequency
of incoming sound.
Temporal theory
Temporal theory: Basilar membrane
vibrates at frequency of incoming sound
waves.
Rutherford: A 500 Hz sound would cause
basilar membrane to vibrate at 500 Hz; a
1,200 Hz sound would cause basilar
membrane to vibrate at 1,200 Hz.
Problems for temporal theory
Basilar membrane varies in width and stiffness
over its length, so cannot vibrate uniformly over
its length.
Nerve cells cannot fire faster than about 1,000
Hz, but sounds are much high pitched than this.

Maybe two or more neurons act together to coarse
code frequency information. (Volley theory)
Place Theory
Frequency information is encoded by the place
along the basilar membrane disturbed by the
fluid vibration.
Hermann von Helmholtz
Problems for Place Theory
Basilar membrane is not composed of
fibers along its width. It is a continuous
strip.
Basilar membrane is not under tension.
Bekesy & traveling waves
1920’s Georg von Bekesy could not image
the cochlea or basilar membrane in action.
So, von Bekesy built a model of the
cochlea.
Waves travel along the basilar membrane.
A wave reaches a peak, then quickly dissipates.
This peak is the peak sensitivity.
Lower tones travel farther.
This yields tonotopic representation (cf.
retinotopic representation).
Loudness
Louder noises correspond to sound waves
of higher amplitude.
In the ear, this leads to waves of greater
amplitude in the basilar membrane.
In the IHC, this leads to a larger neural
response.
Cochlear emissions
Sound in air
Movement of ear drum
Movement of ossicles
Movement of oval window
Fluid-borne pressure waves
Displacement of basilar membrane
Stimulation of hair cells
(cf. p. 374).
Cochlear emissions
Typically have a narrow band of
frequencies.
Roughly 66% of those tested display
cochlear emissions.
Frequency of emission is idiosyncratic.
More prevalent and stronger in women.
Dogs, cats, and birds have cochlear
emissions.
Cochlear emissions
Can be induced by clicks near the ear.
Can be diagnostic of early ear damage.
Tinnitus
Ringing in the ears
Occurs in about 35% of people at some point in
their lives.
Can be caused temporarily by large dose of
aspirin.
Appears to have a cortical basis.
Auditory system:
The auditory pathways
Feedforward: auditory nerve, superior
olive, medial geniculate nucleus and
inferior colliculus, auditory cortex.
Feedback:


Medial portion of superior olivary complex to
OHCs.
Lateral portion of superior olivary complex to
auditory nerve.
Auditory system:
The auditory pathways
Auditory system:
The auditory pathways
The auditory nerve
The “what” pathway
The “where” pathway

These last two are analogous to those found
in vision.
The auditory nerve
~20,000 total hair cells
~4,500 Inner hair cells
~15,500 outer hair cells
~30,000 nerve fibers from these cells
constitute the auditory nerve
Most (~95%) auditory nerve fibers connect
to IHCs.
Frequency tuning curves
Different fibers have different characteristic
frequencies.
All fibers are asymmetric (sharp high dropoff, flat low drop-off).
Narrow characteristic frequencies.
The auditory nerve coarse codes
information from the IHC along the basilar
membrane.
This is because a single fiber is
ambiguous as to what combination of
intensity and frequency of sound wave is
impinging on the ear.
Sound localization:
The “where” pathway
Works primarily on two types of localization
cues: interaural time differences and interaural
intensity differences.
Cells in the cochlear nucleus are monaural.
Cells beyond the cochlear nucleus are bimaural.
Some binaural cells act in complementary
fashion.

These prefer low frequencies.
Other binaural cells act antagonistically.

These prefer high frequencies.
Some binaural cells differ in characteristic
frequency for left and right ears.
Some binaural cells are sensitive to the speed of
moving sounds.
Binaural timing differences
Some binaural cells respond maximally to
simultaneous combination of inputs from
both ears.
Axons can differ in length, so use this to
measure time between signals.
This is Jeffress, (1948), delay line theory.
Binaural intensity differences
“When sound energy passes through a dense barrier –
such as your head – some sound energy is lost” (Blake
& Sekuler, 2005, p. 382).
Some neurons respond maximally to slightly different
intensities of sound.
An ear plug in one ear tends to distort sound localization
abilities.
Organisms with large olivary structures are also better at
sound localization.
Primary auditory cortex:
Laid out in concentric rings, beginning with A1 (aka
Brodmann’s 41 & 42)
A1 is comparable to V1 and S1.
In the “core,” which includes A1, single-cell recordings
show that sounds are represented tonotopically (3
times).
There is cortical magnification in the auditory “core,”
especially human speech and animal vocalizations.
The spatial arrangement of the whiskers on a rat’s face is illustrated
above (A), as well as the corresponding matrix of cell rings in the somatosensory
cortex (B). Actual barrels from layer IV are shown as well (C). (From Blakemore, 1977)
Secondary auditory cortex
(a.k.a. “belt” areas)
Single-cell recordings show that these cells
respond relatively weakly to pure tones.
They respond to more complicated features of
sounds, e.g. frequency modulations, intensity
modulations.
Natural sounds are non-harmonic; animal
sounds are harmonic. (To be explained in
Chapter 11)
Human auditory cortex
Has tonotopic maps revealed by fMRI.
There are “phantom” tones associated
with particular regions of the brain.
Sometimes cortex deprived of input of a
given tone will adjust its sensitivity to
include other tones.
Music and speech have idiosyncratic
auditory features, to be discussed later.