Color vision
Theories
120 mil rods, scotopic vision, most active in dark;
7 mil cones allow perceiving color
Trichromatic eye -> 3 cones.
5% s-cones (pigment cyanolabe) blue sensitive 420nm (390-500)
35% m-comes (pigment chlorolabe) green sensitive 530nm (430-620)
60% l-cones (pigment erythrolabe) red sensitive 560nm (450-650)
Visible light -> 390-700nm
Color is a psychophysical property
Thomas Young came up with trichromatic theory of color vision. 3 primary colors.
Combination of these produces every perceived color. Young-Helmholtz theory. Test
wavelength match with mixed wavelength. Result: 3 primary colors enough for all colors.
Some only needed 2. There is 3 receptor mechanisms. Combining inputs of these three ->
produce any color.
Light enters outer segment of photoreceptors, hitting visual pigments and triggering
nervous response. Light that enters has 2 properties. Light that goes in has intensity
(amount of light, photons), and wavelengths (color). Electrical potential difference between
inside and outside the cell. Photons hit the pigment of cone they create single electrical
response. Principle of univariance: all wavelength information is lost when light enters
photoreceptor.
x
Peak sensitivity but normal (100 quanta) amount would cause 50% absorption, while
less/more than peak sensitivity and same amount – 25%. If less/more than peak is more
quanta like 200, then 50% absorption.
Monochromatism: only one (type) photorecptor in retina (either one or no photopigment).
Only perceived as brighter or darker, so cannot make wavelength-based discrimination. At
least 2 cones are needed to perceive color.
Hering theory challenged trichromatic theory with opponency theory. Experiment: green
over red, X in middle, stare for 30 seconds, look at X at white background caused red on top,
green-blue on bottom. Same but blue on yellow, yellow up and blue below. Illusion: color
after effect. Result: viewing a red field generates green afterimage and vice versa. Same is
true for blue/yellow. Has to do with nervous system adaptation. Look at red, adapt to red
and perceive opposite – green. Conclusions: color is processed by bipolar channels (color
opponency). 3 opponent mechanisms. Black vs white, red vs green and blue vs yellow. Does
not match up with trichromatic theory.
,Merge. First s, m and l-cones detect light. Opponency first occurs at the ganglion cell and
LGN level: red-green channel input from m & l cone; blue-yellow channel input from s cone
and combined m+l cone for yellow. through channels is how color information is
transmitted from retina to visual cortex in brain and allows us to see color. Individual
photoreceptor fits into the ganglion cell.
Red-green channel: m and l cone responses fit into ganglion cell (red on). Excitory from l and
inhibitory from m. so, red light on retina->strong response form l and weaker from m. after
this info goes to ganglion cell, positive response from ganglion. If green light, strong
response from m cone, weaker from l, therefore inhibitory response from ganglion and no
red perceived. Basically, ganglion cells receive information from cones, simplify color
information and convert it into opponency signal.
Opponent cells have receptive fields both in retina and LGN. There are red on also green on
cells. In blue yellow pathway receptive fields in ganglion cells are not center-surround,
organization is random. Blue on cell has contributions from the s cone and combination of m
and l cones. When excited, we perceive blue light. In yellow on cells receptive fields are
identified at the LGN (koniocellular layer). Ganglion not identified.
Why are r-g and b-y pathways different? Evolution. First color vision with 2 cones only, s and
l, i.e. blue and yellow. Medium evolved out of l cones. Sensitivity split due to evolution
between m and l cones so that’s why center-surround receptive fields arised. We are not
just detecting difference between brighter and darker, but also in wavelength so larger
receptive fields required. Amount of light should fall on both s and l cones and with the
responses the color can be perceived not just dark bright discrimination.
From retina to LGN, parvocellular pathway (r-g chromatic channel; achromatic luminance
channel) and koniocellular pathway (b-y channel). Then visual cortex, V1 the start of visual
processing in cortex. Cells in V1 are organized into hypercolumns – each correspond to
retinal point. Neurons from koniocellular layers synapse in layer 3, known as blobs. Neurons
from parvocellular layers synapse in layer 4CB(beta) then post-synaptic projections to layer
3 (separation of achromatic and chromatic information).
Analysis of object color occurs in blobs, composed of r-g and b-y opponent cells. Analysis of
object form happens in interblobs. Achromatic and chromatic information processed in
parallel. All information, inputs from blobs and interblobs, then passed to V4, which is color
analysis destination. Responsible for color constancy. Color constancy is a process that
enables us to see colors being the same despite the different background colors, in different
illuminations. It assesses color of overall visual field, and subtracts color from whole image.
Is mediated by double opponent cells. Double opponent cells: two central cells. S on cell
excited, l is also on, overlapped at same point. They compete, whichever has strongest
output, s or l wavelength light perceived. These cells are influenced by surrounding cells. S
on cell has cells around, if these are l on cells in surrounding (different background) that
excites the cell. Same surrounding -> same background -> inhibits cell.
Monochromat puts on a pair of red sunglasses. Behave as monochromat or dichromat?
Monochromat has a single photoreceptor, single photopigment sensitive to certain
, wavelength of light. Dichromat has two different visual pigments detecting 2 different
wavelength of color. Red sunglasses are gonna change spectral composition of light reaching
retina. Person will still only have one type of pigment. They won’t be able to tell colors due
to single photoreceptor. Sensitivity of a single pigment to 2 different lights (red vs normal)
but cant distinguish because could be just bigger quanta. Univariance principal. Will behave
as monochromat and no sensitivity to color.
(cerebral) Achromatopsia: clinical syndrome caused by damage to area V4—partial or
complete loss of colour vision, despite normal cones (but lesions involve V4+other areas)
Summary:
Color vision theories: trichromacy and opponency.
Color vision requires at least 2 cone receptors with different sensitivity.
Comparing responses of 2 different cones results in color perception.
Trichromacy is at photoreceptor level – 3 types of cones.
Color opponent fields are at ganglion, LGN and cortical level.
Final color vision analysis occurs in the cortex V4.
Any color can be made by mixing primaries, and the value of each primary – tristimulus
value. For 450-550 negative reds. Bc 3 cones have overlapping sensitivity. Imaginary
primaries theoretical components X,Y,Z made so tristimulus values x,y,z are positive. Y
relates to brightness, z similar to blue and x red. Mathematically x y and z=1.
3 steps to color perception: detection, discrimination, appearance.
Color vision defects
Two main types of color vision defects:
1) Congenital
More common
8% males (mainly red-green)
0.4% females
Genetic, onset at birth and does not change
Type & severity constant
Easy to classify
Both eyes affected equally
VA and visual fields normal (except monochromat)
2) Acquired
Equal prevalence in male/female
Onset after birth, often due to disease, injury, medication, toxicity
Deficiency varies
Not easy to classify
Monocular differences
VA and visual fields might be affected
Congenital defects:
1. Cone absent/not functioning (dichromat)
Depends on which cone is missing
Protanopia (L cone)
Theories
120 mil rods, scotopic vision, most active in dark;
7 mil cones allow perceiving color
Trichromatic eye -> 3 cones.
5% s-cones (pigment cyanolabe) blue sensitive 420nm (390-500)
35% m-comes (pigment chlorolabe) green sensitive 530nm (430-620)
60% l-cones (pigment erythrolabe) red sensitive 560nm (450-650)
Visible light -> 390-700nm
Color is a psychophysical property
Thomas Young came up with trichromatic theory of color vision. 3 primary colors.
Combination of these produces every perceived color. Young-Helmholtz theory. Test
wavelength match with mixed wavelength. Result: 3 primary colors enough for all colors.
Some only needed 2. There is 3 receptor mechanisms. Combining inputs of these three ->
produce any color.
Light enters outer segment of photoreceptors, hitting visual pigments and triggering
nervous response. Light that enters has 2 properties. Light that goes in has intensity
(amount of light, photons), and wavelengths (color). Electrical potential difference between
inside and outside the cell. Photons hit the pigment of cone they create single electrical
response. Principle of univariance: all wavelength information is lost when light enters
photoreceptor.
x
Peak sensitivity but normal (100 quanta) amount would cause 50% absorption, while
less/more than peak sensitivity and same amount – 25%. If less/more than peak is more
quanta like 200, then 50% absorption.
Monochromatism: only one (type) photorecptor in retina (either one or no photopigment).
Only perceived as brighter or darker, so cannot make wavelength-based discrimination. At
least 2 cones are needed to perceive color.
Hering theory challenged trichromatic theory with opponency theory. Experiment: green
over red, X in middle, stare for 30 seconds, look at X at white background caused red on top,
green-blue on bottom. Same but blue on yellow, yellow up and blue below. Illusion: color
after effect. Result: viewing a red field generates green afterimage and vice versa. Same is
true for blue/yellow. Has to do with nervous system adaptation. Look at red, adapt to red
and perceive opposite – green. Conclusions: color is processed by bipolar channels (color
opponency). 3 opponent mechanisms. Black vs white, red vs green and blue vs yellow. Does
not match up with trichromatic theory.
,Merge. First s, m and l-cones detect light. Opponency first occurs at the ganglion cell and
LGN level: red-green channel input from m & l cone; blue-yellow channel input from s cone
and combined m+l cone for yellow. through channels is how color information is
transmitted from retina to visual cortex in brain and allows us to see color. Individual
photoreceptor fits into the ganglion cell.
Red-green channel: m and l cone responses fit into ganglion cell (red on). Excitory from l and
inhibitory from m. so, red light on retina->strong response form l and weaker from m. after
this info goes to ganglion cell, positive response from ganglion. If green light, strong
response from m cone, weaker from l, therefore inhibitory response from ganglion and no
red perceived. Basically, ganglion cells receive information from cones, simplify color
information and convert it into opponency signal.
Opponent cells have receptive fields both in retina and LGN. There are red on also green on
cells. In blue yellow pathway receptive fields in ganglion cells are not center-surround,
organization is random. Blue on cell has contributions from the s cone and combination of m
and l cones. When excited, we perceive blue light. In yellow on cells receptive fields are
identified at the LGN (koniocellular layer). Ganglion not identified.
Why are r-g and b-y pathways different? Evolution. First color vision with 2 cones only, s and
l, i.e. blue and yellow. Medium evolved out of l cones. Sensitivity split due to evolution
between m and l cones so that’s why center-surround receptive fields arised. We are not
just detecting difference between brighter and darker, but also in wavelength so larger
receptive fields required. Amount of light should fall on both s and l cones and with the
responses the color can be perceived not just dark bright discrimination.
From retina to LGN, parvocellular pathway (r-g chromatic channel; achromatic luminance
channel) and koniocellular pathway (b-y channel). Then visual cortex, V1 the start of visual
processing in cortex. Cells in V1 are organized into hypercolumns – each correspond to
retinal point. Neurons from koniocellular layers synapse in layer 3, known as blobs. Neurons
from parvocellular layers synapse in layer 4CB(beta) then post-synaptic projections to layer
3 (separation of achromatic and chromatic information).
Analysis of object color occurs in blobs, composed of r-g and b-y opponent cells. Analysis of
object form happens in interblobs. Achromatic and chromatic information processed in
parallel. All information, inputs from blobs and interblobs, then passed to V4, which is color
analysis destination. Responsible for color constancy. Color constancy is a process that
enables us to see colors being the same despite the different background colors, in different
illuminations. It assesses color of overall visual field, and subtracts color from whole image.
Is mediated by double opponent cells. Double opponent cells: two central cells. S on cell
excited, l is also on, overlapped at same point. They compete, whichever has strongest
output, s or l wavelength light perceived. These cells are influenced by surrounding cells. S
on cell has cells around, if these are l on cells in surrounding (different background) that
excites the cell. Same surrounding -> same background -> inhibits cell.
Monochromat puts on a pair of red sunglasses. Behave as monochromat or dichromat?
Monochromat has a single photoreceptor, single photopigment sensitive to certain
, wavelength of light. Dichromat has two different visual pigments detecting 2 different
wavelength of color. Red sunglasses are gonna change spectral composition of light reaching
retina. Person will still only have one type of pigment. They won’t be able to tell colors due
to single photoreceptor. Sensitivity of a single pigment to 2 different lights (red vs normal)
but cant distinguish because could be just bigger quanta. Univariance principal. Will behave
as monochromat and no sensitivity to color.
(cerebral) Achromatopsia: clinical syndrome caused by damage to area V4—partial or
complete loss of colour vision, despite normal cones (but lesions involve V4+other areas)
Summary:
Color vision theories: trichromacy and opponency.
Color vision requires at least 2 cone receptors with different sensitivity.
Comparing responses of 2 different cones results in color perception.
Trichromacy is at photoreceptor level – 3 types of cones.
Color opponent fields are at ganglion, LGN and cortical level.
Final color vision analysis occurs in the cortex V4.
Any color can be made by mixing primaries, and the value of each primary – tristimulus
value. For 450-550 negative reds. Bc 3 cones have overlapping sensitivity. Imaginary
primaries theoretical components X,Y,Z made so tristimulus values x,y,z are positive. Y
relates to brightness, z similar to blue and x red. Mathematically x y and z=1.
3 steps to color perception: detection, discrimination, appearance.
Color vision defects
Two main types of color vision defects:
1) Congenital
More common
8% males (mainly red-green)
0.4% females
Genetic, onset at birth and does not change
Type & severity constant
Easy to classify
Both eyes affected equally
VA and visual fields normal (except monochromat)
2) Acquired
Equal prevalence in male/female
Onset after birth, often due to disease, injury, medication, toxicity
Deficiency varies
Not easy to classify
Monocular differences
VA and visual fields might be affected
Congenital defects:
1. Cone absent/not functioning (dichromat)
Depends on which cone is missing
Protanopia (L cone)
