Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 5: Perceiving Color (color and the visual system (trichromatic…
Chapter 5: Perceiving Color
Light and Color
color vision:
the ability to see differences between lights of different wavelengths
perception of color comes from interaction between receptors in eyes and wavelengths of light reflected from surfaces of objects
spectral reflectance:
the proportion of light that a surface reflects at each wavelength
way objects reflect light depends on molecular structure of surface, which determines its spectral reflectance
some surfaces, such as white/gray/black paper have reflectance curves that are approximately horizontal lines, indicating that they reflect about the same percentage of all wavelengths
however, differ in overall amount of light they reflect, as indicated by whether their reflectance curve is high or low in graph
perceived color of things depends on SPD of light source and on how things reflect light
visible spectrum:
the portion of the electromagnetic spectrum in the range of about 400 to about 700 nm; within this range, people with normal vision perceive differences in wavelength as differences in color
color isn’t in light or material things; it is a perceptual experience evoked by the wavelengths of light reaching our eyes
spectral power distribution (SPD):
the intensity (power) of a light at each wavelength in the visible spectrum
heterochromatic light:
light that consists of more than one wavelength
white light is heterochromatic light that contains wavelengths from across the entire visible spectrum and has no really dominant wavelengths
achromatic light (or white light):
light containing wavelengths from across the visible spectrum, with no really dominant wavelengths; perceived as more or less colorless (shades of gray)
SPD of idealized white light would be horizontal line; all wavelengths across visible spectrum have exactly same power
monochromatic light:
light that consists of only one wavelength
most light sources emit light that consists of only single wavelength
SPD of monochromatic light is vertical spike
Dimensions of Color: Hue, Saturation, and Brightness
three dimensions
saturation:
the vividness (or purity or richness) of a hue
can decrease saturation of monochromatic light by adding white light to it; if enough white light is added, will appear to be nearly pure white with faintest hint of hue, but hue itself will not have changed
brightness refers to perceived amount of light
hue:
the quality usually referred to as ‘color’—that is, blue, green, yellow, red, and so on; the perpetual characteristic most closely associated with the wavelength of light
color representations
color circle:
a 2-D depiction in which hue varies around the circumference and saturation varies along any radius
most saturated hues at circumference
red and blue ends of spectrum meet at part of circumference labeled ‘nonspectral purples’; purple lights are mixtures of the shortest-wavelength violet and longest-wavelength red
called ‘nonspectral’ because don’t exist anywhere on visible spectrum as single wavelengths; white is also a nonspectral color
color solid:
a 3-D depiction in which hue varies around the circumference, saturation varies along any radius, and brightness varies vertically
central axis represents shades of gray light, since all hues are equally represented at the center of the circle
colors along this achromatic axis range from black at the bottom point to the brightest white at the top point
color circle can be produced at any given level of brightness by cutting horizontally through color solid
radius shrinks as move up/down from middle; shrinking radium implies that saturation implies that saturation varies over smaller and smaller range as brightness increases or decreases from midlevel
corresponds to perception that as colors get very dim or very bright, become less vivid
color mixtures
subtractive color mixtures: mixing substances
subtractive color mixture:
a mixture of different-color substances; called ‘subtractive’ because the light reflected by the mixture has certain wavelengths subtracted (absorbed) by each substance in the mixture
when two or more different-colored substances are mixed, reflectance curve of the mixture can be computed by multiplying the reflectances of all the substances in the mixture at each wavelength and plotting results on a graph
paint, for example, is mixture of substances called pigments, and each pigment reflects and absorbs certain wavelengths; after mixed, each pigment continues to absorb the same wavelengths as before pigments were mixed
perceived color depends on percentile of light it reflects at each wavelength, with rest being absorbed
additive color mixtures: mixing lights
additive color mixture:
a mixture of different-colored lights; called ‘additive’ because the composition of the mixture is the result of adding together all the wavelengths in all the lights in the mixture
when three equally bright hue lights projected on white screen so partially overlap, screen equally reflects all wavelengths that hit it because reflectance of white surface is might same at every wavelength
projections of wavelengths in reflected light are same as proportions in projected lights and where there is overlaps, reflected light contains sum of wavelengths
can use color circle to predict perceived color of additive mixture of any two monochromatic lights; if draw line between two hues in picture, perceived color will fall somewhere on line (exactly where will depend on intensities of each light)
spectral power distributions (wavelengths and intensities) of lights determine perceived color of additive color mixtures
hue resulting from any mixture will be less than fully saturated because will fall inside color circle, nearer to fully unsaturated center
complementary colors
complementary colors:
pairs of colored lights that, when combined in equal proportion, are perceived as a shade of gray
50-50 combination of this pair will be perceived as some shade of gray, depending on intensity of lights
when two hues exactly opposite each other on circumference of color circle, connecting line runs through center of circle
primary colors
primary colors:
any three colors of light that can be combined in different proportions to produce a range of other colors
any three sides in color circle can be connected to form triangle enclosing other colors; means there are no unique set of primaries
however, some sets are better than others in terms of producing as many of colors in color circle as possible
red-green-blue are primaries conventionally used to produce additive color pictures
magenta-cyan-yellow conventionally used for subtractive color mixtures
based on sets of three colors because human color perception based on three types of cones, each with different sensitivities to different wavelengths of light
color and the visual system
trichromatic color representation
color matching with mixtures of three primary colors
in order to understand that mixture of three primary colors can match any other color, need to understand that this is a psychophysical color
question is whether the right mixtures of three monochromatic primary colors is perceived as identical in color to some other monochromatic light
to show perceptual equivalence of lights, conduct metameric color-matching experiments
metamers:
any two stimuli that are physically different but are perceived as identical
in experiment, shown two patches of light, called test patch and comparison patch, on perfectly white screen that reflects all wavelengths equally
test patch is single wavelength with fixed intensity while comparison patch is additive mixture of three monochromatic lights which function as primary colors
participant adjusts intensities of each of three primary colors independently until matches test patch
if observer can do this, two patches constitute a metameric color match- perceptually identical despite physical differences
still unclear if need three types of cones to match colors successfully
cones and colors
three types of cones, each containing different photopigment and that each photopigment has particular spectral sensitivity
called L-cones, M-cones, and S-cones according to whether peak sensitivity is long, medium, or short wavelength
spectral sensitivity function:
the probability that a cone’s photopigment will absorb a photon of light of any given wavelength
all three respond to wide range of wavelengths, and spectral sensitivity functions overlap considerably
principle of univariance
principle of univariance:
with regard to cones, the principle that absorption of a photon of light results in the same response regardless of the wavelength of the light
strength of response generated by cone when it transduces light depends only on amount of light transduced (number of photons), not wavelength of light
impossible to work backward from response of single cone to determine wavelength of light that caused response
cones differ in likelihood that they will absorb photon of light of particular wavelength, but once photon is absorbed, its effect is the same for all wavelengths
means color vision depends crucially on relative responses of multiple cone types
if you only had one type of cone (or only rods)
if person with only M-cone trying to mach test patch with wavelength of 615 nm and intensity of 500 photons/sec, relative sensitivity of M-cones to this light is 29%; means each M-cone absorbs and transduces 29% of photons of this wavelength that strike it
since intensity of test light is 500 photons/sec, each M-cone absorbs 145 photons/zed and produces a response with a strength corresponding to the transaction of those 145 photons/sec
tries using 525nm comparison light to create color match; relative sensitivity of M-cones is 89%
if person adjusts 525 nm light to intensity of 163 photons/sec, M-cones will absorb and transduce the same 145 photons/sec and because of univariance, will produce the same response as response to test light
person with just one type of cone can create a metameric color match between any randomly selected test light and any arbitrary comparison light as long as they are both light in visible spectrum by appropriately adjusting the intensity of the comparison light
to a person with just one time of cone, changes in wavelength of light are indistinguishable from changes in intensity, with the degree of apparent change in intensity depending on the relative sensitivity of the cone type at the given wavelength
principle of univariance also applies to rods and explains why night vision is color-blind
rods much more sensitive to light than cones and only rods active in low light
in low light, only have one type of photoreceptor- rods
lights of different wavelengths can produce identical responses from rods
if only had two types of cones
if only had M-cones and L-cones and conducted metameric color-matching experiment using monochromatic test light with wavelength of 615 nm and monochromatic comparison light with wavelength of 495 nm, intensity of test light is fixed at 1,000 photons/sec and relative sensitivity of M-cones is 29%, only absorb 290 photons/sec and relative sensitivity of L-cones is 62%, so absorb 620 photons/sec
when 495 nm comparison light has same intensity as test light (1,000 photons/sec), M-cones absorb greater number of photons than they do with test light while L-cones absorb fewer photons than they do with test light
since responses of two cone types differ for test and comparison lights, they will look different
if reduces intensity of comparison light to 763 photons/sec so response of M-cones is the same, L-cones will absorb even fewer
if increases intensity of comparison light so response of L-cones is the same, M-cones will absorb even more
cannot adjust intensity of single arbitrary comparison light to match the color of a test light with a different wavelength
can match monochromatic test light of any wavelength if they have mixture of two monochromatic comparison lights to work with
number of comparison lights required to produce match with any arbitrary test light tells is how many cone types observer has
physiological evidence for trichromacy
Wald developed method for measuring amount of light at each wavelength absorbed by foveal cone, which enabled the determination of the spectral sensitivities of the three types of cones
mosaic of three types of cones within human retina can be directly visualized using technique called retinal densitometry, which produces high-resolution images of the retina
retina contains only 5% S-cones while relative number of M- and L-cones can differ greatly from one person to the next
meaning of trichromacy
trichromacy representation of wavelength can be thought of as form of data compression
results in some loss of information; reason for many pairs of tights that are physically different in wavelength composition being perceived as identical
evolution as resulted in extremely efficient means of encoding color- just three types of cones, each with a spectral sensitivity function that spans most visible wavelengths, suffice to create fine enough distinctions in color for humans to thrive in varied visual environment
opponent color representation
four basic colors in two pairs of opposites
Hering’s observations were evidence for two important ideas
in some respects, the human visual system operates as if there were fur basic colors
four basic colors can be divided into two pairs of colors, with members of each being in some sense opposites/opponents- red-green pair and blue-yellow pair
when people given stacks of cards, each with different patch of color, and asked to sort into similar colors, usually create four piles aligning with red, green, blue, yellow
color afterimages will often have other member of pair than one that was shown
colors often appear to be mixtures of two non-opponent colors but never appear to be mixtures of two opponent colors
hue cancellation
hue cancellation:
an experimental technique in which the person cancels out any perception of a particular color (yellow) in a test light by adding light of another color (blue)
person is shown monochromatic test light and instructed to neutralize, or cancel out, a perceived basic color in the light by adding its opponent color
intensity of color needed to cancel out one in test light corresponds to perceived amount of color in test light
there is no unique red, meaning that blue-yellow curve only meets once, at unique green
all hues of red perceived to have bit of yellow in them
two colors that vertical line would intersect at any given point (except unique points) represent proportional makeup of hue
physiological evidence of opponency
in 1950s, techniques for measuring the responses of single cells in the retina and then the brain were developed
first, single cones in the retina were found that respond in opposite ways to wavelengths from different parts of the visible spectrum, providing the first physiological evidence for opponency
followed by measurements of neurons in LGN that also responded to color in opponent fashion
later confirmed existence of neural circuits underlying opponent color representations- circuits supporting color vision with four basic colors grouped in two pairs of opposites, red-green and blue-yellow
nerve impulses from three types of cones are processed by networks of other retinal neurons, resulting in combinations of excitatory and inhibitory inputs to RGCs
patterns of neural signals create four different types of ‘opponent color circuits’ in which RGCs function as ‘opponent neurons’ where + is excitatory, - is inhibitory, S is short-wavelength or bluish-to-greenish, M is medium-wavelength or greenish-to-yellowish, and L is long-wavelength or yellowish-to-reddish
+S-ML circuit- RGC is this circuit fires above baseline rate in response to short-wavelength light and below baseline rate in response to medium- and long-wavelength light
+ML-S circuit
+L-M circuit
+M-L circuit
+S-ML and +ML-S neurons respond oppositely to blue and yellow light, so their outputs support the perception of blue and yellow as basic and opposite colors
+L-M and +M-L neurons respond oppositely to green and red light, so their outputs support the perception of green and red as basic and opposite colors
color-opponent neurons in the visual pathway
color-opponent RGC respond with increase in firing rate to one range of wavelengths and with a decrease in firing rate to the complementary range of wavelengths
receptive fields of these neurons are spatially uniform; respond most strongly to a uniform patch of light having the preferred wavelength and are maximally suppressed by a uniform patch of light having the complementary wavelength
other color-selective neurons in visual pathway, including RGCs, LGN cells, cortical cells have receptive fields that produce more elaborate patterns of response
some V1 cells have single-opponent center-surround receptive fields- respond with an increase in firing rate when the receptive field center is stimulated with long-wavelength light and with a decrease in its firing rate when the receptive field surround is stimulated by medium-wavelength light
responses of neurons with double-opponent center-surround receptive fields do provide information about color edges
seen in many color-selective V1 neurons
produces a strong increase in firing rate when the center is stimulated with long-wavelength light and surround is stimulated with medium-wavelength light
response is strongly suppressed when the center is stimulated with medium-wavelength light and the surround with long-wavelength light
complement neurons that provide information about edges between regions that differ in brightness
help us perceive objects in scenes where objects and their background have the same brightness but are of different colors
responses of single-opponent neurons carry information about the wavelength of light within uniformly colored regions of the visual scene but don’t provide much information about color edges, locations where adjacent regions are illuminated by different wavelengths
color afterimages and opponency
one of primary mechanisms of dark and light adaptation is photopigment bleaching
photopigment bleaching:
a photopigment molecule’s loss of ability to absorb light for a period after undergoing photoisomerization
time must pass before molecule changes back in shape and again becomes light sensitive
bleached means it can’t respond to light by absorbing photons with the wavelengths to which the molecule is sensitive in its unbleached state
chromatic adaptation:
a kind of photopigment bleaching that results from exposure to relatively intense light consisting of a narrow range of wavelengths
results in color afterimages
if relatively intense light of one particular wavelength strikes there retina for an extended time, the photopigment molecules in the type of cones that are most sensitive to those wavelengths become bleached, rendering the visual system temporarily less sensitive to those wavelengths
would bleach one time of cone and when look at white surface, which would normally elicit equal responses of cones, only one will respond more intensely because of bleaching of other type, producing perception of the opposite color
meaning of opponency
three different types of cones with three different photopigments that respond preferentially to different wavelengths of light explains results of psychophysical experiments in color matching, showing that we can mix three primary colors of monochromatic light to match light of any color
opponent color representation expense other psychophysical phenomenon, such as categorization, color afterimages, and the results of hue-cancellation experiments; imply that people operate with four basic colors grouped in pairs of two
evidence comes from detection of opponent neurons- RGCs and neurons in LGN and visual cortex whose neural circuitry supports the four-colors-in-two-pairs aspects of color vision
likely evolved opponent representation of color because it’s a matter of efficient transmission of information
circuits provide information about the differences between responses of M- and L-cones rather than exact values of responses of each
color opponency provides extremely efficient code for representing the colors found in natural scenes, lending support to the idea that opponency is an evolutionary adaptation
trichromatic representation not questioned for a long time; once it was, two-stage process developed
first stage characterized by trichromatic representation, up to point where cones transduce light into neural signals
second stage is opponent color representation, describes how regional ganglion cells and color-selective neurons in the brain process cone signals
color contrast and color assimilation
color contrast:
the perception of a surrounded color as shifted toward the compliment of the surrounding color
perception of color when viewed in juxtaposition to other colors depends on specific combination and spatial arrangement of colors
provides further support for opponent mechanisms in color vision
color assimilation:
the perception of a surrounded color as shifted toward a non complimentary surrounding color; also known as the “spreading effect”
in color contrast, the difference in color between surrounded and surrounding elements is perceptually accentuated, whereas in color assimilation, the difference is perceptually reduced
color constancy
color constancy:
the tendency to see a surface as having the same color under illumination by lights with different spectral power distributions
base perception of color of an object on the reflectance of the object and not on the SPD of the light reflected by the object
wavelengths entering eyes don’t depend only on spectral reflectance of paper but also on spectral power distribution (SPD) of illuminating light (relative intensities of all the wavelengths present in the light)
the amount of each wavelength reflected into your eyes by the paper (the SPD of that reflected light) is determined by multiply9ing the relative intensity of the illuminating light at each wavelength by the reflectance of the paper at each wavelength
visual system achieves color constancy in one way by comparing the wavelength distributions in the SPDs of the light reflected from the various surfaces in a scene to estimate the SPD of the illuminating light and the reflectance of each surface
ability to do this depends on the presence of multiple surfaces with varied reflectances, which is usually the case in everyday scenes
not entirely sure how illumination and reflectance work, but there are some theories
one is discounting the illuminance- visual system automatically determines the amount of each wavelength reflected from all surfaces on average and uses this as its estimate of the SPD of the illuminant
if SPD of illuminant is known, then it is possible to determine the reflectance of a surface simply by dividing the amount of light reflected by that surface at each wavelength by the amount of light present in the illuminant at each wavelength
another way visual system achieves color constancy is through chromatic adaptation- cones adapt to continuous, intense stimulation by a particular wavelength so that their response to that wavelength becomes weaker
by adjusting your cones to compensate for the imbalance of wavelengths in the illuminating light, chromatic adaptation ‘compensates’ for non-white illumination, which helps make more accurate judgments about the actual color of surfaces
lightness constancy
the lightness of a gray surface appears about the same regardless of the intensity of the illuminant
lightness:
the perceived reflectance of a surface (the proportion of the illumination that the surface appears to be reflecting)
distinct from brightness, the perceived amount of light it might be reflecting
lightness constancy:
the tendency to see a surface as having the same lightness under illumination by very different amounts of light
if illumination across a scene is uniform, lightness constancy explained by ratio principle which says that the perceived lightness of a region is based not on the absolute amount of light reflected from the region, but on the relative amounts reflected from the region and its surround
when illumination isn’t uniform across scene, suggested that ratio principle be supplemented with two-part anchoring rule
in any given scene, the region that reflects the most light is perceived as white (or the lightest shade of gray in the scene) and the lightness of every other region is perceived in relation to that anchor point
if the scene consists of regions under different amounts of illumination, the visual system applies the anchoring rule separately in each illumination zone
color perception is two-stage process
in first, referred to as trichromatic color representation, light evokes different responses from three different types of cone photoreceptors in the retina
in second, known as opponent color representation, responses from the cones are combined and processed by subset of retinal ganglion cells and by color-selective neurons in brain
color vision deficiencies
most people who have color vision deficiencies are not entirely insensitive to differences in wavelengths of light
inherited deficiencies of color vision
two broad categories of inherited color vision deficiencies- monochromacy and dichromacy
dichromacy: partial color blindness
dichromacy:
a condition in which a person has only two types of cones, instead of the normal three; in all such cases, the person has a limited form of color vision but cannot discriminate as many colors as a person with all three cone types
can discriminate more colors than monochromat, but would still confuse some others that wouldn’t be confused by others
Ishihara color vision test:
a test using configurations of multicolored disks with embedded symbols; the symbols can be seen by people with normal color vision but not by people with particular color vision deficiencies
three types of dichromacy
protanopia:
a condition in which a person has M-cones and S-cones but lacks L-cones
1% males, .02% females
deuteranopia:
a condition in which a person has L-cones and S-cones but lacks M-cones
1% males, .01% females
tritanopia:
a condition in which a person has L-cones and M-cones but lacks S-cones
.002% males and females
monochromacy: total color blindness
monochromacy:
a condition in which a person has only rods or has only rods and one type of cone; in either case, the person is totally color-blind, perceiving everything in shades of gray
two inherited conditions that result in monochromacy- both rare
rod monochromacy:
a condition in which a person has rods only, with no cones
about .002% of population
not able to perceive color and hyper sensitive to light
rods much more sensitive to light, so rod photopigment is fully bleached and not able to respond to light
have relatively low visual acuity
cone monochromacy:
a condition in which a person has rods and only one type of cone
even more rare than rod monochromacy
have rods and only one type of cone, can be any one of three; use rids to see in dim light and cones to see in bright
entirely lack color vision
inherited deficiencies occur when person born without one or more of three types of cones in retina
affect males much more because lack of M- and L-cones caused by specific defect in gene on X chromosome; women have backup
cortical achromatopsia: color blindness from brain damage
achromatopsia:
loss of color vision caused by brain damage
even less frequent than inherited color deficiencies
activity in several regions of temporal and occipital lobes, including V4, was greater when looking at color patches, concluded that these areas contribute to perception of color
part of ventral ‘what’ system