Transcript Striate cortex April 2009

Striate cortex
April 14, 2009
The primary visual cortex (V1) is one of the most
studied areas in the mammalian cortex.
Pubmed search for striate
cortex yielded 35,911 hits
Important concepts
• General flow of information within the striate
cortex
• Receptive field characteristics is cortical
neurons
• Architecture of the striate cortex.
• INFORMATION FLOW
Information flow
Resident cells
• There are three basic
types of neurons in the
primate V1:
– Spiny pyramidal
cells (excitatory)
– Spiny stellate cells
(excitatory)
– Smooth stellate
(almost all are
GABAergic).
Single-unit recording
Receptive field properties of V1 cells
• Hubel & Wiesel (1959; 1968)
Simple cells
• Similar to most retinal ganglion cells and LGN cells – have
distinct excitatory and inhibitory subregions within the
receptive field.
• Orientation selective
Receptive field mapping: Find where the
neuron is sensitive
Find optimal orientation
Optimal orientation
Receptive field mapping
_
+
_
+
_
+
_
• Hubel and Wiesel movie
Hubel and Wiesel model of
Simple cells
• Simple cell receptive
fields could be 'built' in
the cortex by collecting
responses from LGN cells
whose receptive fields fall
along a line across the
retina
• In every animal that has been examined
there are neurons in V1 that respond to
images of bars or edges at a particular
orientation.
Complex cells
– Complex cells are the most numerous
in V1 (perhaps making up threequarters of the population).
– Orientation selective
– No center/surround receptive field
organization
– Thus, not sensitive to the polarity
(bright or dark) of a stimulus.
– Many complex cells are also
direction-selective, in the sense that
they respond only when the stimulus
moves in one direction and not when
it moves in the opposite direction.
H & W model of
complex cells
• Complex cells receive
integrated responses
from a collection of
simple cells.
ANOTHER TYPE OF COMPLEX CELL
• Many cortical neurons modulate their
responses based on stimulation outside
the central receptive field.
• Initial experiments of surround
modulation measured responses to high
contrast bars or gratings of varying
lengths.
• Some cells called end-stopped cells,
decrease their responses when the bars
or gratings become larger than the
central receptive field.
Length-summing cells
• Continue to their responses as bars or grating
length increases.
• All examined mammals, including primates,
cats, (Gilbert, 1977), rodents (Girman et al.,
1999), and marsupials (Oliveira et al., 2002)
have cells that show stopped- and length
summing behavior.
Columnar organization
• All mammals have some degree of columnar
organization: neurons below a particular position
on the cortical surface have similar functional
properties, and these properties tend to change
smoothly as one moves across the cortical surface.
• Many mammals including
have smooth, repeating maps of
orientation selectivity in V1.
Orientation columns
• These orientation maps repeat at 550 micrometer intervals in
macaque and tree shrews and 800-900 microns in cat
Ocular dominance columns
• Another major determinant of cell response is
eye-of-origin. Most cortical cells can be
driven by stimuli presented in either eye, but
they generally prefer (ie. respond more to)
one eye or the other - a property called
'ocular dominance'.
• Some cells prefer the right eye, and others
prefer the left eye. Hubel and Wiesel
discovered that ocular dominance is
organized in a similar way to orientation
preference - dominance is unchanging
vertically but alternates as one moves
horizontally across the cortex
• In a given band, the majority of cells
receive monocular visual input that arise
from the ipsi or contralateral eye BUT not
both.
• In the upper and lower layers of cortex, the
input becomes mixed.
• When V1 is stained for cytochrome oxidase,
an enzyme involved in metabolism, distinct
blobs show up which are most clearly seen
in layers 2 and 3. The blobs are arranged in
rows that line up above the centers of the
ocular dominance columns in layer 4.
• Livingstone and Hubel recorded from cells
inside the blob regions, and found that they
showed no orientation preference, but
instead were concentric, and over half of
them responded to color variation
The Hubel
and Wiesel
“ice-cube”
model of
visual cortex:
simplified
diagram.
(Cortical
modules are
not REALLY
square!).
No blobs
in layer
4C
Pinwheels
•The pinwheel
fashion of
orientation
columns was first
found using
optical recording
•Some people say
blobs are in the
center of the
pinwheels/orientat
ion columns
•Some people
disagree
• Columns in V1 have spatial frequency
preferences.
Orientation columns
Ocular dominance columns
Spatial frequency columns
• Spatial frequency tuning may change with depth of
column.
Cortical maps: what are they
good for?
• the map of V1 is:
– Uniform: making sure the cortex can respond to all possible
stimulus properties at every point in visual space
• We perceive the environment as uniform
– We are not more sensitive to red than to green in one visual area.
– Minimizes wiring length
• Remember: there are a lot of neurons in the brain and most make
local connections in order to conserve space.
• Neurons with similar response properties are closer to each other in
cortical space.
• Neuroscientist once hoped that the columnar organization of V1
would help in understanding the specific rules of one module would
generalize to the entire cortex.
• However, many studies have failed to identify any role of columnar
organization in determining functional properties of single cells
(Purves et al., 1992; Horton & Adams, 2005).
• NO ocular dominance columns in many mammals (mice, rabbits,
squirrels, tree shrews) and vary from animal to animal in squirrel
monkeys.
• Similar story applies for orientation columns.
• May be an aftereffect of development
Important stuff!
• Flow of information from LGN to V1 and within
V1.
• Different types of functional properties of V1
cells: simple, complex, and end-stopped cells.
• Models of simple and complex cells.
END
Cortical Magnification
Factor!
Cortical magnification describes how many
neurons in an area of the visual cortex are
'responsible' for processing a stimulus of a given
size, as a function of visual field location. In the
center of the visual field, corresponding to the
fovea of the retina, a very large number of neurons
process information from a small region of the
visual field. If the same stimulus is seen in the
periphery of the visual field (i.e. away from the
center), it would be processed by a much smaller
number of neurons
Chiasm crosses fibers
from left and right, no
crossing up/down.
• The map of the visual world does not exist on the surface of V1
because the brain needs to look at a copy of the image.
• Rather the retinotopic map exist
– to keep things organized during development
– processing visual information usually involves adjacent
points in an image.
• In primates virtually all visual information that
gets to cortex must first go through primary
visual cortex. Four main synonyms:
•
•
•
•
Striate cortex
Brodmann’s area 17
Primary visual cortex
V1
But why is it called striate cortex?
• In 1782 a young
medical student named
Francesco Gennari
described a heavy band
of myelinated axons in
layer 4B, the stria of
Gennari (named after
its discoverer), about
the only landmark you
can see in cortex
without special stains.
• This striation gave rise
to its name: striate
cortex
Why area 17?
• Brodmann (1868-1918)
defined areas by
cytoarchitecture.
• Cytoarchitecture is the
organization of the cortex
as observed when a tissue
is stained for neurons.
Why is it called primary?
• It’s essential.
• First part of cortical vision
• If area 17 is ablated – you are cortically
blind!
Why V1?
• There are several cortical areas involved in
vision
• This is the first one.
• V1 is found in all
mammals from
highly visual
primates to animals
with subcutaneous
eyes such as the
blind mole rat (Krubitzer
& Kaas, 2005; Cooper et al., 1993)
ANATOMY
Cortex!
• Cortex is a general term, meaning outer
covering.
• Cerebral cortex consists of two separate flat
sheets of cells, one on each side, folded to
fit inside the skull.
• Gray matter consist of
cell bodies and glial
cells.
• White matter consists
of myelinated axons.
Laminar organization
•Need differential stains to see anything…what you see depends on the staining method
•Nissl stains RNA inside of cells
•CO stain metabolically active neurons.
•Golgi uses potassium chromate and silver nitrate to stain the cell body and dendrites
•Weigert stain for myelin (Wertige has many stains)
• This operation is analogous to the
operation of "edge detectors" in image
processing, which process an image by
spatial convolution with an edge
"kernel" consisting of adjacent positive
and negative values.
• Most cells of the mammalian visual system from the
retina to the cortex respond primarily to local
stimulus contrast rather than absolute brightness.
• In V1, the tuning width of orientation selective
neurons remains constant as stimulus contrast is
increased (Sclar & Freeman, 1982).
• The influence of spatial frequency on cortical
responses is similar to that of stimulus orientation.
• The majority of V1 spatial frequency tuning curves
have a characteristic band-pass shape.
V1
LGN
Heimel et al., 2005
Peak preferred temporal
frequencies are similar between
LGN and V1, but the highest
temporal frequency that
neurons can follow is on avg
slightly lower in the cortex as
compared to LGN.