Transcript The Visual Cortex

The Visual Cortex: Anatomy
The LGN projects to ipsilateral primary visual cortex (area 17,
striate cortex). Remember that this means that left visual field
goes to right area 17. Also remember that left visual field gets
input from left eye (nasal retina) and right eye (temporal retina).
This cortical region is often called striate cortex because of the
unique striations caused by the dense cells in layer 4. LGN
fibers end in layer 4 c in a complex pattern that I will briefly
discuss.
Bear et al.
The Visual Cortex: Circuitry
Pyramidal cells (excitatory) of
area 17. The stellate cells of
layer 4 C are unique modified
pyramidal cells.
It is important to remember
that there are many classes
of inhibitory (GABA)
LGN projects to layer 4 stellate cells.
interneurons that are not
These cells project to cells in the
shown here. There are
upper layer; cells in the upper layers
perhaps as many as 30
project to the lower layers and out.
interneurons in cortex.
Note the vertical (columnar)
organization.
Bear et al.
Upper cortical layers project to
cortex; in this case to areas 18,19.
Lower layers (5,6) project to
brainstem and thalamus. Layer 5
projects to superior colliculus (control
of eye movements). Layer 6 projects
topographically to LGN.
This is a highly simplified view of cortical circuitry.
Bear et al.
The Visual Cortex: Spatial Organization 1
The above represents the projection of one eye
to the striate cortex (layer 4C). Note the stripes.
The blank regions are the stripes representing
the other eye. So, in layer 4C, the eyes
representing a visual field remain separated:
because this separation extends through a
vertical column, these are sometimes called the
oculur dominance columns.
A stain for cytochrome oxidase reveals an
unexpected columnar organization: the
CO blobs. It is independent of the OD
columns. We’ll see the blobs functional
significance later on.
The Visual Cortex: Spatial Organization 2
Bear et al
The M (movement) and P (high resolution) pathways are kept separate in striate cortex.
Both pathways converge on the blobs. Note the vertical organization of the intrinsic circuitry.
Now what does the cortex do with this input?
The Visual Cortex: Functional Organization- Simple Cells
Many pyramidal neurons in striate cortex respond to oriented
moving bars and not to spots of light (remember that LGN relay
cells respond to spots of light. So there are two important
transformations that occur in striate cortex:
Most cells are selective for a particular orientation and they require
some movement.
Orientation selectivity depends on the convergence of input from an
appropriate subset of LGN relay cells as indicated above.
Orientation selectivity may also be enhanced by cortical circuitry.
The cell illustrated here responds to an oriented bar in one region of
space; this type of cell is called a simple cell.
Bear et al.
The Visual Cortex: Functional Organization- Complex Cells
Bear et al.
Complex cells are typically found in the upper and lower layers of striate cortex. They are
very varied in their properties but always respond to oriented moving bars. The difference
between complex cells and simple cells is that complex cells respond over a wide area.
Some complex cells also respond to to movement in one direction only.
The Visual Cortex: Overall Organization
Bear et al.
The figure to the left shows optical imaging of ocular dominance
columns (mid) and orientation columns (bottom). You can see
that the orientation stripes cut through the OD stripes. The
figure at the top shows the conceptual diagram of this
relationship. For each bit of striate cortex (a small region of the
visual field) each eye representation has complete
representation of all possible orientations (layer 4C); the upper
layers combine these representions to allow the visual system
to estimate depth (stereopsis).
The blobs represent a completely different information channel.
Cells in the blobs are not sensitive to orientation or movement.
The blobs neurons respond to colour (primates); this
channel is not found in cats, rodents etc.
Plasticity in the Visual Cortex 1
Ocular dominance stripes with
normal rearing.
Ocular dominance stripes with one
eye occluded during a critical
period. The occluded eye does not
capture as much cortical territory
as the normal eye.
This molecular biology of this effect
has been intensely studied. A
variety of factors, including
neurotrophins, are involved but a
complete picture is lacking.
Kandel et al.
Plasticity in the Visual Cortex 2
Kandel et al.
This developmental sequence can be
disrupted by eye closure and
depends on complex molecular
interactions driven by correlated input
from the two eyes.
Development of ocular dominance columns in
cats.
Plasticity in the Visual System: Competition for Targets
Normal amphibian tectum: tracer injection in the
contralateral eye.
Amphibian tectum with third eye grafted on: the
two eyes compete for tectal target space and
produce a columnar arrangement- artificial ocular
dominance.
NMDA receptors are critically involved in this
process.
Kandel et al.
The Visual Cortex: Estimating Depth
Kandel et al.
Objects at the eyes fixation point are fused; those behind or in front of the fixation point
(retinal disparity) are interpreted as closer or further away from you. A population of cells in
striate cortex combines the images from the 2 eyes in such a way that they respond to
retinal disparity and thus signal the relative depth of objects. This system is very important
for many animals such as carnivores (estimating distance to prey) and primates (estimating
distance to insect or fruit).
The Visual Cortex: Projections- Two Processing Streams 1
Bear et al.
There are two main processing streams for primate vision. A dorsal stream goes through the
middle temporal lobe and onto the parietal cortex. It deals with movement (the M channel to
a large extent) and is used for navigation and motion perception.
The ventral stream goes into the temporal lobe and deals with identification of visual input
(the P channel is a big contributor).
The Visual Cortex: Projections- Two Processing Streams 2
Bear et al.
Kandel et al.
PET image of activation of the dorsal stream
in response to visual motion.
Response of a neuron in temporal cortex
(monkey, ventral stream) to faces.
Recently far more specific cells have
been found in humans- responding to
pictures of Bill Clinton etc. Specific
cortical areas respond to such
“communication” signals.
Different neurons in the visual system respond to different features: motion, shape, colour etc.
How are these different attributes combined to form a single percept- this is the binding
problem. One controversial hypothesis is that synchronized activity in neurons representing
these different features are the basis of the unity of perception.