Midterm 2 Textbook Notes:
-Humans see range on electromagnetic spectrum from 400nm to 700nm.
-Color is not a physical characteristic but a product of the physical stimulus interacting with na
-Steps to color perception:
-3 cone photoreceptors w/different photo pigment (different sensitivity to light of different wave-
lengths). S-Cones (Short-wavelength cones peak at 420nm), M-Cones (Medium-wavelength
cones peak at 535, L-Cones (Long-wavelength cones peak at 565nm).
-Cones work at daylight levels (photopic), Rod works in dimmer light levels (scotopic).
Problem of Univariance:Acone can only provide single type of information (neural firing rate),
where as we need it to correspond to 2 types of information, being amplitude and wavelength
corresponding to color and brightness.
-Problem of Univariance is reason for lack of color in dimly lit scenes, making it impossible to
discriminate colors under scotopic conditions.
Trichromatic Solution: Differences between wavelengths or mixtures of wavelengths are de-
tectable due to different types of cones (S, M, L).
Metamers: Mixtures of different wavelengths that look identical. (1. Mixing wavelengths does
not change physical wavelengths, the mixture is mental event no physical changes occur. 2. For a
mixture of red light and green light to look perfectly yellow, the red and green have to be adjust-
ed, whereas if not adjusted, the mixture will be greenish or reddish.)
-Light reaching the visual field is converted to 3 numbers by the three cone type, if numbers are
different from numbers in different patch, it can be discriminated, if not, it will result in
-Thomas Young and Hermann von Helmholtz worked out 3-D nature of experience of color
(Trichromatic TheoryAKAYoung-Helmholtz Theory).
-Additive color mixture: Mixture of lights (taking wavelength(s) and adding it to another).
-Subtractive color mixture: Mixture of pigments (Aand B) whereAwill absorb some wave-
lengths and B will absorb some, leaving a result of a dark color such as brown. (ex. Color filters)
-Pointillism: Method of painting used in late nineteenth century which used additive color mix-
ture (small dots and different textures of few colors).
Nervous system: To discriminate the color of the lights, the nervous system looks at the differ-
ence in activities between the 3 cones. Signals of (L-M), [(L+M)-S], and (L+M).
[S Cones make small contribution to perception of brightness.]
Cone-Opponent Cells in Retina & LGN:
LGN: The lateral geniculate nucleus is in the thalamus which receives input from retinal gan-
glion cells (maximally stimulated by spots of light), and has input and output connections to the
-Some retinal and ganglion cells are stimulated by L-cones on its center and inhibited by M-
cones on its surround. (L-M) is a type of cone-opponent cell (a cell type that subtracts one type
of cone input from another)[other ex. [(M-L), (M+L)-S, S-(M+L)]. ColorAppearance:
Color space: 3D (Hue, Saturation, and Brightness) space based on outputs of 3 cone types, that
describes set of all colors.
Achromatic: Color result that will occur (when R, G, and B are equal) which lacks chromatic
component (black, white, or grey).
-The chromatic aspect of color, shade is Hue; The potency or strength of the hue is Saturation;
The physical intensity of light is known as Brightness.
Opponent color theory: Theory which states perception of color is based on 3 mechanisms, each
resulting from opponency between 2 colors (red vs. green, blue vs. yellow, white vs. black).
Unique Hues:Any of the 4 colors which can be described by single terms (red, yellow, green,
-Color opponent processes that support color appearance are likely to be found in the visual cor-
tex (since cells in LGN are cone opponent cells).
Double opponent cell- Found in visual cortex; cell which is excited by one cone/color type and
inhibited by the opponent cone/color & the adjacent region is inhibited by 1st input and excited
by 2nd input.
Single opponent cell- Conveys information about color of broad area (cone-opponent cell).
Achromatopsia- Inability to perceive colors due to damage to CNS (brain damage).
Afterimage- Visual image seen after the visual stimuli has been removed.
Adapting Stimulus- Stimulus whose removal produces a change in visual perception/sensitivity.
Negative Afterimage-An afterimage whose polarity is opposite to original stimuli (colors com-
plementary; ex. red produces green).
Neutral Point- Point at which opponent color mechanism is not producing any signals.
-Age turns lens of eye yellow over time
-Genes that code for M & L cones are in X chromosome, males only have 1 copy as opposed to
females with 2. Damage to one copy leaves males color blind as opposed to females will per-
ceive colors normally since they'd still have one undamaged.
-Factors include: type of cone being affected, type of defect (photopigment of cone is anomalous,
or missing entirely).
-Missing a cone type means they will see in 2D instead of 3D color.
Deuteranope- Individual with color blindness due to absence of M-cones (M - 560nm and L -
610nm lights will be classified the same for the person).
Protanope- Individual with color blindness due to absence of L-cones (will have different set of
color matches based on outputs of M and S cones).
Tritanope- Individual with color blindness due to absence of S-cones.
Color Anomalous- Color blind individuals who make discriminations of color that are anom-
alous (typically 3 cone types but 2 are extremely similar which make them experience world the
same as individuals with only 2 cone types).
Cone Monochromat- Individual with color blindness and only one type of cone in retina. There-
fore only see 1D color space (shades of gray).
Rod Monochromat- Color blind individual missing all cones (difficulty seeing in daylight condi-
tions, poor acuity, lack of ability to discriminate colors). Agnosia- Individual can see image but fail to recognize them due to brain damage.
Anomia- Individual can see and recognize image but fail to name them due to brain damage.
Cultural relativism- In sensation and perception, the basic perceptual experiences may be deter-
mined in part by cultural environment (ex. each group free to make their own color space).
Color Contrast- Color perception effect color of one region induces opponent color in neigh-
Color assimilation- Color perception effect where 2 colors bleed into each other, taking on some
chromatic quality of the other.
Unrelated color- Color that can be experienced in isolation.
Related Color- Color such as brown or gray seen only in relation to other colors (ex. gray patch
in complete darkness appears white).
Color Constancy- Surface color appear same under wide range of illuminants.
Illuminant- Light that illuminates a surface.
Spectral reflectance function- The % of particular wavelength reflected from a surface.
Spectral power distribution- Physical energy in a light as function of wavelength.
Reflectance- The % of light hitting surface and reflected but not absorbed into the surface (typi-
cally a function of wavelength).
-Ability to perceive and interact with the structure of space is one of the fundamental goals of the
Realism- Philosophical position arguing that the world is real to be able to sense.
Positivism- Philosophical position that argues that all we perceive is from the evidence of our
senses, therefore for all we know everything could be an elaborate hallucination.
Euclidean- Geometry rules of the world (ex. parallel lines remain parallel extended in space, ob-
jects maintain same size and shape as moving around in space, internal angles of triangle always
add up to 180 degrees).
-How we see the world is reconstructed from two non-Euclidean inputs (two distinct retinal im-
-Human visual field 190 degrees from left to right, 110 covered by both eyes, 60 up from center
of gaze and 80 degrees down.
Binocular (with 2 eyes) Summation: Combination of signals from both eyes increases perfor-
mance on tasks rather than signals from one eye alone.
Binocular Disparity- Difference between retinal images from both eyes, this disparity is what
provides a vivid perception of the 3D world, which cannot be achieved through monocular vi-
sion (one eyed vision).
Stereopsis-Ability to use binocular disparity as a cue to depth perception.
Monocular Cues to 3D Spaces:
Monocular depth cue- Depth cue (info about 3d visual space) which is available when viewing
world with one eye.
Binocular depth cue- Depth cue that relies on information from both eyes (Stereopsis).
Occlusion- Depth cue to relative position of objects (one object obstructing view of another ex.
circle over corner of square shape). Nonmetrical depth cue- Depth cue that provides info about relative depth but not magnitude
(front or behind object).
Metrical depth cue- Depth cue that provides quantitative info about distance in the 3D.
Projective Geometry- The geometry which describes transformations that occur when 3d world
is projected to 2d surface.
Relative size- Comparison of size between items w/o knowing absolute size of either items.
Texture gradient- Depth cue based on geometric fact items of same size form smaller images
when they are further away.An array of items changing in size smoothly across image will ap-
pear to form a surface tilted in depth.
Relative height- Depth cue observation that objects at different distances from the viewer on a
ground plane will form images at different heights in retinal image, therefore the further the ob-
ject is the higher it will be in the image.
Familiar size- Depth cue based on knowledge of typical size of objects like pennies.
Relative metrical depth cue- Depth cue that can specify the distance of objectArelative to B w/o
providing absolute distance of eitherAor B.
Absolute metrical depth cue- Depth cue providing quantifiable info about distance in 3D (nost
sticks out 3 cm in front of his face).
Aerial Perspective/haze- Depth cue based on implicit understanding that light is scattered by the
atmosphere, therefore the more distant the object, the more the object appears faint and less dis-
Linear Perspective- Depth cue where lines that are parallel in 3D world appear to converge in
Vanishing point- The apparent point at which parallel lines receding in depth converge.
Pictorial depth cue- Depth cue for distance or depth used by artists to depict 3D depth in 2D pic-
Anamorphosis- Use of rules of linear perspective to create a 2D distorted image that it looks cor-
rect only when viewed from special angle or mirror that counters the distortion.
Motion parallax- Depth cue based on head movement; Geometric info obtained from an eye in 2
different positions at 2 different times is similar to info from 2 eyes in different positions of the
head at the same time (This is a nonpictorial depth cue).
Accomodation- Process by which the eye changes focus (lens gets fatter as gaze is directed to-
wards nearer objects).
Convergence-Ability for both eyes to turn inward, this happens to place 2 images of a feature in
the world on corresponding locations in the 2 retinal images (when you shift focus from a far ob-
ject to a nearer object;Also it reduces disparity of feature to zero or nearly zero).
Divergence-Ability for bother eyes to turn outward, this happens in order to place 2 images of a
feature in the world on corresponding locations in the 2 retinal images (when you shift focus
from a near image to an image further away;Also reduces disparity of feature to zero or nearly
-Convergence, divergence and familiar size cues can tell us exact distance to an object.
Corresponding retinal points- Geometric concept stating points on retina of each eye where
monocular retinal images of single object are formed are at the same distance from the fovea in
each eye (and that the 2 foveas are corresponding points). -Objects falling on corresponding retinal locations have zero binocular disparity.
Vieth-Muller Circle- Location of object fall on geometrically corresponding points of two reti-
Horopter- Location of objects whose images lie on corresponding points; surface of zero dispari-
Diplopia- Double vision; if visible in both eyes, stimuli falling outside of Panum's fusional area
will appear diplopic.
Panum's fusional area- Region of space, in front and behind the horopter, within which binocu-
lar single vision is possible.
Crossed disparity- Sign of disparity created by objects in front of plane of fixation (horopter).
Uncrossed disparity- Sign of disparity created by objects behind plane of fixation (horopter).
Stereoscope- Device for simultaneously presenting one image to an eye and a different image to
the other eye (used to present dichoptic stimuli for stereopsis and binocular rivalry.
-Viewers were cards with double images 2 inches apart (distance of human eye) to view the im-
age as if it was a vivid 3D space.
Free fusion- The technique of converging or diverging eyes to view stereogram w/o stereoscope.
Stereoblindness- Inability to make use of binocular disparity as a depth cue (someone who has
lost vision in one eye is not described as stereoblind).
-Stereoblindness is usually due to childhood visual disorders such as strabismus (misaligned
Random dot stereogram-Astereogram made of a large number of randomly placed dots with no
monocular cues to depth. The stimuli available are Cyclopean stimuli.
Cyclopean- Stimuli defined by binocular disparity alone.
Correspondence problem- The problem of figuring out, which bit of image in left eye should be
matched with which bit in the right eye, in binocular vision.
Uniqueness constraint- In stereopsis, the observation that a feature in the world is represented
exactly once in each retinal image (this constraint simplifies the correspondence problem).
Continuity constraint- In stereopsis, the observation that neighbouring points in the world lie at
similar distances from viewer other than edges of objects.
-Binocular neuron (2 receptive fields); the receptive fields in binocular striate cortex neurons are
similar in both eyes (preferred speed and direction of motion, orientation, spatial-frequency, tun-
ing), therefore are well suited to match images in the two eyes.
-Stereopsis can be used for metric as well as nonmetric depth cues.
Bayesian approach- Ide