- Images from the retina are analyzed for form/movement/color. For each point in the
outside world, there is a corresponding image point on the retina in each of the two eyes. These
images from the two eyes are then brought together and compared at higher stations in the visual
pathway for extration of information about depth.
- Development of retina: The prosencephalon protrudes laterally and enlarges to form
primary optic vesicles. These later invaginate to give double-walled optic cups, with a largely
obliterated ventricular space between the two walls. The outer wall becomes the pigment
epithelium, while the inner wall becomes the neural retina. The two are separated by the
subretinal space, the residue of the ventricular cavity.
I. Structure of retina
- The retina is composed five main types of neurons, segregated in to three main layers:
1) Outer nuclear layer: photoreceptors, no direct blood supply. Instead, they
receive nourishment from the choroidal circulation filtered thru the pigment epithelium.
2) Inner nuclear layer: bipolar (second order visual neurons), horizontal (lateral
association neurons), amacrine (ditto), Mueller glial cell bodies
3) Ganglion cell layer: retinal ganglion cells (third order visual neurons, who
axons constitute the optic nerve).
- The outer and inner plexiform layers are sites of synapses between layers
- There are two types of photoreceptors
1) rods: most sensitive in dim light, only one type
2) cones: less sensitive to light, three types mediate color vision (peaks at blue,
green and yellow).
- Distribution of photoreceptors is not uniform: fovea has high density of cones and
few rods, periphery has more rods. Blue sensitive cones are also absent from the
center of the fovea.
- The fovea has the highest visual acuity, or spatial resolution of the visual space. This
is achieved by having thinner cones and more of them per unit space than elsewhere
in the retina; with the layers of cells anterior to the receptors being pushed aside.
This clearing minimizes light scattering which tends to degrade the quality of the
- The structure of the photoreceptors:
o Outer segment: where light is absorbed and a neural signal is generated. It
contains an orderly stack of membranes in which is embedded the light
sensitive visual pigment (rhodopsin in rods, and one of three cone pigments in
cones). In cones these membranes are continuous with the plasma membrane
to form a highly convoluted surface membrane. In rods, the membranes are
completely internalized to form a stack of flattened membranous discs. The
high density of pigment provides a high probability of absorption of an
incident photon. o Inner segment: contains the metabolic machinery of the cell, while the
synaptic terminal forms chemical synaptic connections with bipolar and
o Glu is the NT released by both rods and cones.
- Pigment epithelial cells serve to (1) absorb light not absorbed by the rods/cones using
melanin, (2) phagocytose shedded fragments of rods/cones, (3) regenerate pigment by
supplying frest chromophore to the bleached pigment in the outer segment.
III. Visual pigments
- Pigment is composed of a chromophore covalently bound to an opsin 7TM protein that
is different in different pigments. The two parts are joined together by a protonated Schiff base.
- Humans have 4 pigments (1 rod, 3 cones) each maximally sensitive at different
- Upon photon absorption, the chromophore undergoes several configuration changes; the
configuration that triggers vision takes several ms to reach; the ultimate result is the release of
chromophore, and binding of fresh chromophore with opsin
- In darkness, rods and cones have a membrane potential of about –30 to –40 mV.
- Rods/cones hyperpolarize in response to light in a graded fashion with respect to
intensity. Cones require more light and their responses rise and fall more rapidly than
rod responses. Their briefer responses permit cones to have a better time resolution of
- Phototransduction mechanism: In darkness the surface membrane of the outer segment
has a higher permeability to cations, as a result of which there is a steady influx of sodium (and
Ca and Mg) ions into the outer segment, being driven by their inward-directed electrochemical
gradients. This steady influx of + charge in darkness (the dark current) maintains the cell in a
partially depolarized state and, consequently, a steady release of NT from the cell’s synaptic
terminal onto second-order neurons. In light, the ionic permeability of the outer segment is
reduced, thus decreasing the influx of cations and producing a membrane hyperpolarization,
which spreads passively to the synaptic terminal where it reduces the rate of transmitter release
from the receptor. The permeability reflects the opening of a cGMP-activated conductance,
which by itself has no intrinsic light sensitivity. Light, however, closes the conductance by
activating an enzyme cascade that leads to the lowering of the cGMP level in the outer segment.
Metarhodopsin II, an intermediate photoproduct of rhodopsin, catalyzes the
activation of GPCR (rod/cone transducin) thru GTP binding G-prot activates cGMP
phosphodiesterase which hydrolyzes cGMP GMP.
Shut off of the cascade involves: 1) phos of the photisomerized rhodopsin,
rendering less effective in activating the G-prot, followed by final capping due to binding of
another protein called arrestin to the phos rhodopsin. 2) deactivation of the active G-prot thru its
intrinsic GTPase activity, which converts the bound GTP to GDP. 3) turn off of the active
phosphodiesterase by rebinding of an inhibitory subunit of the enzyme.
Summary - In dark, cGMP keeps a nonselective cation channel open, and cell is
depolarized. Rhodopsin is coupled to G-protein that activates phosphodiesterase and lowers
[cGMP]. - Ca influx plays a key negative feedback function and mediates light adaptation.
In dark, there is a circulation of Ca++ at the surface membrane of the outer segment, consisting
of an influx thru the cGMP-gated channels and an efflux thru a transport mechanism involving
an exchange of cations. In the light, the Ca++ influx stops due to closure of the channels, but the
efflux continues, thus resulting in a decrease in the cytosolic Ca++ concentration. This decrease
leads to 1) an increase in guanylate cyclase activity 2) a more effective phosphorylation of the
photoexcited rhodopsin and 3) a higher likelihood of channel opening by cGMP. These effects
all antagonize the action of illumination, and underlie the ability of photoreceptors to adapt to
V. Synaptic connections in the retina (see Figure on p. 12)
- The synaptic terminals of rods and cones are morphologically different with rods ending
in spherules (smaller and round) and cones ending in pedicles (larger and with a flat base).
These terminals form connections with horizontal cells and bipolar cells.
- Throughput pathway: photoreceptor bipolar retinal ganglion cell optic nerve
- Lateral associations: between photoreceptors (via gap junctions), and via horizontal and
- There are distinct synapse morphologies in the retina
- Synaptic triad: photoreceptor with horiztonal-bipolar-horizontal cells, and contains a
- Dyad: proximal bipolar end synapses with two cells
- Primary nxt’s in the retina are Glu (excitatory) and GABA (inhibitory, released by some
horizontal cells). DA is released from interplexiform cells.
VI. Information processing in the retina
- Only amacrine and ganglion cells give all-or-none impulses. The rest of the cells in the
retina have graded responses.
- The receptive field of a single photoreceptor is bigger than itself due to connections
with neighboring photoreceptors; for the same reason, the receptive field of horizontal
cells is bigger than the multiple photoreceptors it contacts
- Since bipolar cells receive synapses from both photoreceptors and horizontal cells, their
receptive field is divided into a center (dominated by photoreceptor) and surrounding
antagonistic ring (dominated by horizontal cells)
- In light, on-bipolars depolarize in response to light in the center of its receptive field
and hyperpolarize in response to light on the surrounding ring; off-bipolars are vice versa.
In terms of synaptic organization, an on-bipolar cell’s receptive field is derived from
sign-inverting (i.e. hyperpolarizing) synapses from the receptors (in the field center) and
sign-preserving (i.e. depolarizing) synapses from the horizontal cells (for the surround).
- The above property of bipolars, propagated to on- and off-center ganglion cells,
probably helps mediate contrast and edge detection
- Ganglion cells have a center-surround receptive field organization like bipolar cells. In
terms of synaptic connections, the on-center ganglion cell probably receives sign-preserving
inputs from on-bipolars in its field center. The on-bipolar cell already has a center-surround
organization, but, in addition, the on-bipolar cells in the surround probably excite amacrine cells,
which in turn have an inhibitory input on the ganglion cell. They can also be classified as X
(slow adapting) and Y (fast adapting) types. - Amacrine cell firing pattern is best fit to respond to moving objects. When light is
shown anywhere in their receptive field, these cells usually depolarize transiently at the on and
the off of the illumination. The do not have a center-surround organization.
Early Vision: Retina and LGN
I. Ganglion cells:
- M Ganglion cells: large cells bodies with widely branching dendrites and a large
myelinated axon; large, center-surround receptive fields but with no preference for
wavelength; display a brisk response so long as an appropriate stim is moving,
responding to both the leading and trailing edge; rapid adaptation is stimulus is held
in place; projects to LGN.
- P Ganglion cells: smaller cell bodies with narrow dendrites and smaller myelinated
axon; small, center-surround receptive fields, with single cone projecting to center
and many to surround, which are specific for wavelength (e.g. Red-ON center); will
continue to dire as long as appropriate stim is present in receptive field; projects to
- Small bistratified cells (SBS): fairly large receptive field that is ON to short
wavelengths (blue) and OFF to both long and middle (Red and Green); center only,
no surround; extremely sensitive to insult to nervous system (acquired tritanopia);
projects to LGN.
- Melanopsin containing ganglion cells project to the SCN in hypothalamus to entrain
circadian rhythms and are capable of transducing light in the absence of rods and
- Ganglion cells that send axons to regions in the midbrain: Superior colliculus deals
with reflexive eye movements; pretectum plays a role in controlling the size of the
pupils as the level of light increases or decreases. These cells have small cell bodies
but large dendritic fields and receptive fields.
II. Specialization of the retina:
- At the fovea, all ganglion cells, bipolar cells and their processes are swept to the side
creating a pit where light can get to the outer segments of the cones without having to
travel thru a bunch of junk.
- Each cone in the fovea synapses directly onto 5 bipolar cells, and each of those
bipolar cells synapse on a single ganglion. So you have five ganglion cells carrying
separate signals to the brain from a single cone.
- Ganglion cell axons penetrate thru the retina at the optic disc (blind spot) to form the
III. Optic nerve and tract
- Axons in the optic nerve cross at the optic chiasm so that each optic tract
represents the contralateral visual field. Axons from nasal side of retina
cross midline; temporal side axons remain ipsilateral and each hemifield info
travels together in the optic tract. Information from each visual hemifield is
processed by the contralateral cerebral hemisphere. - Only the extreme periphery of each hemifield is processed by a single eye
- Damage in optic tract and back (caudal to chiasm) results in damage to vision
in contralateral visual hemifield.
- There are three major visual pathways:
1) optic tract LGN primary visual cortex (higher visual processing)
2) optic tract superior colliculus (saccades)
3) optic tract pretectum (pupillary light reflex)
IV. Lateral Geniculate Nucleus (LGN)
- Visuotopic map: central retina posterior (caudal) LGN; peripheral retina anterior
(rostral) LGN; superior retina lateral LGN; inferior retina medial LGN.
- LGN is made of 6 layers: magnocellular 1,2 and parvocellular 3-6
- There are 4 Parvo layers because there is two for each eye, with one for ON-center and
one for OFF-center receptive field ganglion cells each.
- Ipsilateral retina maps to layers 2,3,5 (prime!) and contralateral to 1,4,6
- LGN neurons have receptive fields that resemble on- and off-center ganglion cells
- Intralaminar cells respond to blue light only, may be responsible for color
V. LGN to primary visual cortex
- LGN projects dorsally to V1 in the occipital lobe with inferior-superior
inversion with respect to visual field (this makes the object appear right-side
up to the cortex since the image was initially inverted by the lens).
- Neurons in V1 are the first to respond not to spots of light, but to lines, bars
- Neurons in V1 are first to respond to activity in both eyes; they have binocular
Visual Cortex: Organization and Plasticity
I. Primary visual cortex
- Located in both banks of calcarine sulcus
- Consists of 6 layers and the physiological properties in V1 vary from one layer to the
- Neurons above layer 4 send their axons to other areas of cerebral cortex whereas
below layer 4 send their axons to subcortical targets.
- Foveal region has high cortical magnification in dorsal/caudal V1. Parts of the visual
field progressively farther away from the foveal representation are mapped at progressively more
rostral parts of V1.
- All axons of neurons in both Magno and Parvo layers of the LGN terminate layer 4 of
V1. The bigger cells in the upper half of layer 4 receive all axonal input from Magno
LGN and the smaller cells in the lower half from Parvo LGN. Layer 4 amplifies
imput and sends it along to layers 2 and 3. - Layer 4 is organized in alternating ocular dominance columns
- Columns in LGN project to same part area of V1 (recall that layer 4 receives
- Parvocellular layers 4Cβ and 4A in cortex
- Magnocellular layers 4Cαand 4B in cortex
- Intralaminar projects to layer 3
- Layers 2 and 3 respond preferentially to bars, edges and contours, not just spots.
Moreover, the neurons respond selectively for the orientation of a bar, with responses
maximum at one angle but still responds +/- 20/30 degrees.
- Layers 2 and 3 communicate, in part, with layers 5 and 6.
- In V1, input is sent to stellate interneurons, with output via pyramidal cells
- Input and output comes from/goes to LGN, pulvinar, superior colliculus, other cortical
- A cortical column is a collection of neurons across all layers that are similar to one
another in some functional property and that differs from neurons in adjacent columns
by that same property.
- Ocular dominance and orientation selectivity properties stand out dramatically in V1.
o Ocular dominance:
Neurons in layer 4 are monocular: they are driven exclusively by
stimuli falling on one retina. Their projections are generally vertical
(radial connectivity), neurons throughout V1 create columns about 1
mm wide that display dominant responses to one eye. The columns
alternate between contralateral and ipsilateral eyes. But some
projections from 4 cross borders such that neurons in layers 2 and 3
are the first with binolcular receptive fields. Ocular dominance
columns are the means whereby signals from the two eyes are kept
separate initially and then combined in a precise manner to provide
subsequent stages a strong signal about depth.
Any perturbation in the vision of one eye disrupts the normal pattern
of ocular dominance organization, leading to expansion of columns
dominated by the normal eye and shrinkage of columns dominated by
the deprived eye.
o Orientation selectivity:
Layers 2 and 3 show fairly sharp tuning for the orientation of a
stimulus and their vertical projections create columns in V1 down thru
layers 5 and 6 that are all tuned to the same orientation. Over the
distance of 1 mm, a full coverage of all stimulus orientation is made
III. Early Development of visual cortex organization
- Despite the diminished light level and the absence of patterned visual input in the
uterus, babies are born with ocular dominance columns already established.
- In both the LGN and V1, competition for trophic factors appears to be the principal
means for establishing segregated inputs. Coordinated activity among adjacent cells
is the means whereby neighboring cells come to terminate next to one another. - The neurotrophins are normally released in tiny amounts by cortical neurons, thereby
setting up a competition among axons that innervate cortical neurons. When a super-
abundance of the neurotrophins is experimentally presented into the V1 of a neonate,
geniculocortical axons no longer segregate into ocular dominance columns. This
failure occurs because the competition between axons driven by the ipsilateral and
contralateral eye has been eliminated by excess neurotrophin.
IV. Later Development and the critical period
- Deprivation by lid-suture: If suturing is done early enough in the postnatal period, the
result is a dramatic rearrangement of ocular dominance columns with the normal eye
expanding (not abnormal retention) to take up >90% of the total cortical volume. The
columns driven by the deprived eye shrink as a result of loss in synaptic sites.
Moreover, the cells that are still driven by this eye are weakly active and fatigue
easily with repeated presentation of stimulus (amblyopia).
- Deprivation by strabismus: If one of the muscles that moves an eye is damages, that
eye looks inward or outward, but it is aimed differently than the other, normal eye.
The result is the same level of light and the same extent of patterned visual input
reaches the two eyes, but the two see different things. As a result, connectivity and
column development is normal, but all neurons are monocular (even in layers 2, 3, 5)
The most likely explanation is that cells in layer 4 no longer maintain some axonal
connection across ocular dominance columns.
- Critical periods: first 22 weeks in rhesus monkeys, first 22 months in humans.
- Reverse-Deprivation experiments ask what happens if, during the critical period, a
deprived ye is returned to normal and the formerly normal eye is deprived/patched?
Development of connections and columns can be normal is timed right, but cortical
cells with binocular receptive fields fail to develop. Therefore, the use of stereoptic
cues for depth perception is permanently lost.
Extrastiate Visual Processing
- Higher visual areas can be broadly divided into two parts
- Dorsal pathway (where): spatial position and motion, input primarily
- Ventral pathway (what): shape, texture, color of objects, input from all layers of
- Secondary visual cortex can be stained to reveal a thick-pale-thin-pale stripe pattern
- Color sensitive neurons in V1 CO blobs thin stripe V2 ventral pathway,
- Neurons with fine spatial sensitivity in V1 layer 2/3 interblob pale/inter stripe
V2 ventral pathway, V4
- Motion sensitive neurons inV1 layer 4B (magno.) thick stripe V2 dorsal
pathway, MT I. Introduction to the ventral pathway
- The ventral pathway processes shape and color
- It is composed of V1, V2, V4, V8 (most color sensitive region in brain), and
inferotemporal regions (ant., post., central)
- Other miscellaneous regions
- Lateral occipital (LO): object/shape
- Fusiform face area (FFA): face stimuli or complex shape stimuli
- Prospagnosia: the selective inability to recognize individual faces; a
result of FFA lesion.
- Parahippocampal place area (PPA): landscapes and building interiors
- Extrastriate body area (EBA): bodies and body parts
- As you progress in an anterior direction toward higher levels in the ventral pathway,
receptive fields become larger, more bilateral, more concentrated near the fovea, and cells
become increasingly sensitive to large-scale complex shape properties.
- The neural mechanism underlying shape perception is not fully understood but it
appears to be a parts-bases processing system in which the capacity of encode a large number of
objects as combinations of parts.
II. V1: simple orientation tuning
- V1 cells are tuned to the orientation of lines/edges
- Simple cells: tuned to orientation, position variant
- Have on/off receptive fields structured to recognize edges, perhaps as a result of
convergent input of multiple LGN cells whose receptive fields are aligned
- Complex cells: also tuned to orientation, but position invariant
- This property may be a result of convergent input of multiple simple cells
- Hypercomplex cells: added property of end-stopping (tuned to short lines)
- Columnar organization: each column is orientation tuned, and exact orientation shifts
gradually from column to column
- Color contrast is mediated by single- or double-opponent receptive fields
III. V2: more complex visual processing than V1
- V2 has a thick-thin-pale-thin stripe pattern
- Thick: tuned to motion, depth
- Thin: tuned to color and is organized in a hue map
- Pale: tuned to shape
- V2 cells have property of figure ground segmentation: tuned to properties of object
outside receptive field (e.g. what side is the object on?)
- Unlike V1, will respond to illusionary contours
IV. V4: curvature tuning
- V4 cells are tuned to curvature, the start of part-based coding
- Color tuning in V4 is complex
V. Inferior temporal regions: highest level of ventral pathway, parts-based coding
- Integrates edge and contour information to code objects as a collection of geometrical
- Even complex objects like faces can be represented in a parts-based way I. Introduction to the dorsal pathway
- The dorsal pathway processes depth and motion
- Components of the dorsal pathway are V1, V2, V3, middle temporal (MT), medial
superior temporal (MST), intraparietal regions (lat., vent., med., ant.), and area 7a
- MT is buried in the superior temporal sulcus and receives inputs from layer 4B in V1
and the thick stipes in V2, which contain a high percentage of cells tuned for motion
direction and cells tuned for position in depth. MT neurons show strong tuning for
motion and depth.
- Dorsal division of MST, MSTd, is specialized for processing optic flow, movement
fields produced when the observer translates thru the environment. Some neurons in
MST show complex interaction between motion direction and stereoscopic depth. This
type of interaction is thought to relate to MST’s role in representing optic flow. If we
walk while gazing off to the left, the optic flow of near objects across the retina will be
leftward, while the optic flow of far objects will be rightward.
- Lateral division of MST, MSTl, is specialized for smooth pursuit eye movements.
- LIP is associated with eye position and saccadic eye movements.
- VIP processes both visual and somatosensory information about peripersonal space.
Many of these neurons are sensitive to both visual and tactile motion.
- MIP is associated with arm reach movements.
- AIP is associated with spatially precise grasping of objects with the hand.
- The sequence LIP MIP AIP is responsible for grabbing objects of interest
- 7a, on the surface of the inferior parietal lobe, is considered the highest stage in the
dorsal pathway, and may be the site of convergence of multiple kinds of spatial
- Parietal cortex is also thought to be critical for controlling spatial attention.
- Lesions in the human parietal cortex can cause
- Attentional hemineglect: neglect of contralateral space, also manifested in
- Constructional apraxia: inability to reproduce complex objects
- Lesions to MT and MST (MT+) can cause akinetopsia, the inability to perceive motion.
II. V1, V2: lower level processing of stereoscopic vision
- 3D depth perception is achieved by stereopsis: there is a slight disparity between the
images on the two retinas
- Both V1 and V2 are tuned to binocular image disparity and motion direction
III. MT: intermediate level processing
- MT cells respond to motion direction and speed, as well as to binocular disparity. V1
and V2 cells (especially in the thick CO stripes) tuned for binocular disparity project
preferentially to area MT.
- MT receptive fields are larger than in lower levels
- MT columns are organized in smooth gradients of direction preference and stereoscopic
- Microstimulation of MT cells can bias perception of motion direction and depth IV. Note on visuospatial coordinate systems
- Visual system must deal with information coded in various coordinate systems (e.g.
retinotopic, head-centered, body-centered, etc.)
- How coordinates systems are dealt with
- Sequential transformation from one system to another as one progresses along the visual
pathway, so any given level uses only one coord sys: evidence for this in lower
- Integration of info in multiple coordinate systems, resulting in a mix of spatial
sensitivity: most cells in higher areas do this
The Auditory Periphery
I. Introduction to hearing
- Sound is a pressure wave, and pure tones can be characterized by amplitude (measured
indirectly as dB) and frequency
- Dynamic range of hearing in human is about 0-100 dB (whisper to pain) and 20-20K Hz
- Types of damage to ear
- Chronic severely loud sound exposure kills hair cells
- Ototoxic antibiotics, e.g. streptomycin or gentamicin, kills hair cells
- Hearing loss of high frequencies with age, aka presbycusis
I. Anatomy of the ear
A) Outer ear
- Pinna: collects and funnels sound toward the opening to the auditory canal, thus
amplifying sound; sound shadows from sounds coming from different positions in space provide
resolution of sound origin
- External auditory canal: acts as resonant tube for 4 kHz tones
B) Middle ear
- Impedance matching of air to water sound transmission by three mechanisms
(1) Area of tympanic membrane is 20x that of stapes footplate
(2) Lever advantage of ossicles (malleus, incus, stapes)
(3) Oval window of stapes is smaller than eardrum
- Middle ear muscles (tensor tympani, stapedius): restrict ossicle movement in response
to loud sound to prevent damage. The tensor tympani reduces the area of the flexible
membrane in the eardrum; the stapedius restricts the amplitude of stapes motion.
- Damage in the middle ear is called “conductive.” A common cause is
otosclerosis in which bony growth impedes the movement of the middle ear ossicles.
C) Inner ear (more on cochlea below)
- Damage in the inner ear is called “sensorineural”
- One can distinguish between conductive and sensorineural hearing loss by
comparing the audibility of a tuning fork held in the air, or pressed against the skull. In
conductive hearing loss, the latter position is effective at presenting sound by bone conduction,
thus overcoming the conductive loss that pertains to air-borne sound. III. The Cochlea
- Memorize cochlear structure (e.g. figure on p. 5)
- The stapes footplate moving in and out if the oval window sets up a fluid wave within
the scala vestibule. The motion of this wave causes the cochlear partition to move alternatively
up and down within the cochlear duct. The cochlear partition of the basilar membrane and the
hair cells, supporting cells and extracellular structures lying upon it, and the overlying tectorial
membrane. These all lie just under the scala media, the central membraneous tube flanked on
top and bottom by the scala vestibule and scala tympani, respectively. The scala media is
bounded by Reisnner’s membrane on the top and is actually delimited by the apical surface of
the hair cell epithelium on the bottom, not by the basilar membrane. This distinction is important
because the scala media is filled with endolymph (high K, low Na), which is produced by a
secreting epithelium called the stria vascularis that lines the outer wall of scala media. The s.
vestibule and tympani contain perilymph (low K, high Na). Since the basilar membrane itself
does not present a diffusion barrier, the basal surface of the hair cells is bathed by low K
perilymph while their apical, hair-bearing surface faces high K endolymph. This disposition of
ionic media is important for hair cell function. Since endolymph is 80 mV positive to perilymph,
this endolymphic potential provides additional driving force for K+ ions to flow into hair cells.
- Meniere’s disease arises under conditions in which the normal circulation of endolymp
is blocked or altered resulting in sensorineural hearing loss.
- There are two classes, inner and outer, of hair cells. In a cochlear cross section one can
see the single row of flask-shaped inner hair cells, and three rows of columnar outer hair cells.
- Stereocilia of hair cells are arranged in rows of increasing height and are embedded in
the tectorial membrane; shear of basilar membrane with respect to tectorial membrane
causes displacement of hair bundle
- Tip links join nonselective cation channels on one stereocilium to the next tallest one
tilting of the stereocilia causes channel to open or close open channel allows K, the
primary cation of endolymph to travel down steep electrical gradient into hair cell and
depolarize it if depolarized, hair cell opens voltage gated Ca channels, allows Ca in,
and releases Glu onto the nerve ending (see also p. 9)
- Hair cell adaptation may occur through two suggested negative feedback mechanisms
(1) end of tip link is attached to myosin-like motor (myosin 1C) on stereocilia,
movement up or down changes the set point of system. Myosin 1C exerts steady force on
the transduction channel.
(2) Ca entry feeds back to regulate the mechanotransduction channel
- Synaptic transmission: At its basal end the hair cell is contacted by afferent fibers
whose cell bodies are located in the spiral ganglion and which send an axon centrally to end on
the cochlear nucleus of the brainstem. When the hair cell is depolarized by opening of the
transducer channel, voltage dependant calcium channels in the basolateral membrane are opened
and synaptic vesicles fuse with plasma membrane to release Glu, which binds to AMPA
receptors causing excitation of the afferent axon. This depolarizing effect of v-gated Ca+ influs
is counterbalanced by v-gated K+ channels that dominate the basolateral conductance of the hair
cell and allow outward flow of K into low K perilymph.
- Individual afferent neurons makes only a single contact with a single inner hair cell. At
the point of contact, the hair cell places a synaptic ribbon, an electron-dense body to which a
number of transmitter-containing vesicles are tethered. It is thought the ribbon serves to motivate vesicles for release during stimulation by sound and during the ongoing spontaneous
activity that occurs in silence.
- Nearly all afferent nerves end on inner hair cells, whereas efferents end on outer hair
cells. The crossed olivo-cochlear bundle arises from the contralateral superior olivary
complex and synapses massively on outerhair cells. This pathway is inhibitory and when
activated, causes a loss of sensitivity and freq selectivity in afferent fiber response.
- Outer hair cells exhibit electromotility (contract on depolarization and vice versa) to
regulate distance of tectorial membrane to inner hair cells and adjust their sensitivity; this
mechanism depends on a motor protein called prestin. Targeted disruption of prestin in
mice causes deafness.
- OHCs express a Ca channel opened by ACh from efferent nerves; the Ca causes K
channels to open and results in hyperpolarization, which ultimately results in loss of
- Otoacoustic emissions: active movement of basilar membrane can generate sound.
- Active motions of OHCs are driven by the sound-evoked change in membrane potential.
These active motions enhance the overall vibration pattern of the cochlear partition, thereby
causing larger deflections of the IHC bundle. Cochlear signaling is enhanced by OHC
electromotility. If OHCs are damaged, the cochle becomes much less sensitve. Thus, the
efferent innervation of the cochlea, which causes a suppression of cochlear sensitivity, must do
so by inhibiting OHCs. This occurs via cholinergic inhibition of cochlear hair cells from
olivocochlear efferent activity. Hair cells are depolarized by nicotinic AChR, but only briefly.
Ca++ enters the cell thru the receptor and causes the opening of Ca++ dependant K+ channels, of
which there are many, causing a net hyperpolarization of the hair cell’s membrane.
The Peripheral Auditory System – Auditory I
I. Human perceive the frequency and loudness of sound
- The dynamic range of human perception is about 100-16,000 Hz and 0-120 dB
- The dynamic range of conversational speech is a small subset of total perception
- One way to measure a tuning curve is to measure the loudness needed to mask a probe
of single tone as a function of frequency; obviously, minimum loudness is needed when
mask frequency is same as probe
II. Speech consists of a mixture of tones of different frequencies
- Different vowel sounds are the result of the shape of the vocal tract and how it modifies
the tone produced by the vocal folds
- The information needed to recognize vowel sounds is encoded as peaks of energy at
certain frequencies; these peaks are called formants
- The precise value of formants can vary from person to person (e.g. due to size of vocal
tract), but are still recognized because vowel sounds cluster in formant-space III. Basilar membrane (BM) is responsible for tuning: the separation of sound into its component
- The most important aspect of BM motion is that it is tuned, meaning that points on the
BM are differentially sensitive to different frequencies in the sound stimulus.
- Sound results in a traveling waveform in the BM, and the position of maximum
displacement depends on frequency (high freq near stapes, low freq near apex)
- Since hair cells sense motion of the BM at one point only, each hair cell inherits from
the BM a selectivity for frequency, in which the hair cell is most sensitive to the freq that
causes maximum BM motion at the point of the hair cell’s innervation. Likewise, since
auditory nerve fibers are distributed down length of cochlea, each ANF has a different
best frequency, giving rise to a tonotopic representation of sound
- Outer hair cells are responsible for sound sensitivity: they amplify BM movement in
response to soft sounds and partially damp BM movement in response to loud sounds;
loss of OHCs (e.g. due to kanamycin) results in loss of sensitivity and broadened tuning
IV. Auditory nerve fiber is phase-locked (there is a close relationship between the stimulus
waveform (produced by the back-and-forth motion of hair cell flexion) and the acti