exam 2 (lecture) Flashcards
Which genetic difference in color vision would cause someone to see the most color metamers when doing a color-matching experiment
rod monochromatism
(Brightness). At any extreme levels, you are going to have ______.
achromatic vision. If it is too bright, you lose colors.
(Brightness). If it is too dim, the visual system _____
shifts to rods, which do not detect color (grayscale vision)
_____ plays a crucial role in color constancy. Helps the brain have a ____ ______.
v4; stable perception
example of color constancy
Ensures the red apple looks red whether it is sunlight, in the shade, inside.
What does “relative metrical” mean?
Allows for comparisons of depth magnitude but does not give absolute measurements.
What is binocular disparity?
The difference in retinal images between the two eyes that the brain uses for depth perception.
3 primary psychological dimensions for color perception
Hue
Saturation
Brightness (luminance)
Explain how a
classic stereogram
works and
connect it to
binocular vision
A classic stereogram works by displaying two slightly different 2D images, one to each eye, which creates binocular disparity.
Each image is projected onto corresponding & non-corresponding points on the retinas.
The brain merges these images, using the disparities at non-corresponding points to perceive depth, creating a vivid 3D illusion
from flat images.
This process mirrors how our eyes naturally perceive depth in the
world, with each eye receiving a slightly offset view due to the distance between them. Binocular vision allows us to experience stereopsis, the perception of depth from these retinal disparities
3 STEPS to color perception
- Detection: Visible light (380–750 nm)
- Discrimination: Differentiating
wavelengths & mixtures - Appearance: maintain color constancy
Color is a _____, not a _____
brain-generated perception; property
of objects
3 cone types
S-cones, short wavelengths (blue range)
M-cones, medium wavelengths (green range)
L-cones, long wavelengths (red range)
cones respond to …
a whole range of colors, not just one specific color
Spectral
related to light wavelength
illuminant
light source
Spectral Density (Spectral Power Distribution
- Describes how light energy is distributed
across wavelengths - Determines light color and quality
- The level of detail in measurement depends on how we are dividing or “binning” the spectrum
- Human vision only uses a few bins, we divide it into red green and blue
illuminant power spectrum
Energy distribution across wavelengths
does Spectral Density use Fourier Analysis?
no - light wavelengths do not combine like sound waves
Monochromatic Light
pure Wavelengths, Single wavelength, appears as a distinct hue (e.g.,
lasers)
Hyperspectral cameras use _______ for precise measurements
hundreds or thousands of bins
Sunglasses absorb ____ ,while allowing
____-
harmful UV light; visible light through
Broad Spectrum Light
(Smooth Curves), Emits many wavelengths across a range (e.g., sunlight, incandescent bulbs
Complex Mixtures
(Uneven Spikes) Multiple intensity peaks, creating mixed color output, Common in fluorescent lights, LEDs, and sodium
lamps
color detection (step 1) Three key types of spectra
- Continuous - full visible light range (sunlight)
- Emission - bright lines show emitted wavelengths
- Absorption - dark lines across color indicate wavelengths that have been absorbed
Absorption Spectrum
How Materials Interact with Light.
Missing lines = absorbed wavelengths, creating spectral gaps.
Shows which wavelengths a material absorbs vs which pass through or reflect
Spectral Density of Reflected Light
Determines what is reflected
after light absorption.
2 main types of spectral reflectance
- Unbalanced - Selectively reflects some
wavelengths (color). some wavelengths reflected more than other, example carrots and tomatoes reflect longer wavelengths than cabbage - Flat Reflectance - Reflects all
wavelengths evenly (achromatic,
grayscale). evenly reflected across wavelengths, produces grayscale shades instead of color.
reflectance curves help predict ____
how materials appear under different lighting
Level of brightness determines ____
grayscale level
Color vision relies on signals from _____
three types of cone photoreceptors (s-cones, m-cones, l-cones)
Tetrachromacy
Superhuman color vision.
* Women with four cone types; no males yet
* Caused by mutation in one X chromosome (M & L cones are X-
linked).
* Can distinguish up to 100 million colors (vs. 1 million for trichromats).
Does Tetrachromacy Improve Vision?
- A fourth cone alone doesn’t guarantee better color vision.
- The brain must process the extra input to enhance color perception
how to test for Tetrachromacy?
- No online tests—screens can’t display extra colors. .. RGB!
- DNA tests can confirm the genetic trait.
example of Tetrachromacy
Concetta Antico, an artist with
confirmed tetrachromacy, may
see subtle shades others can’t
The Principle of Univariance
- A single photoreceptor cannot distinguish one color based on wavelength alone
- Different combinations of wavelength
(hue) and intensity (brightness) can
produce the same response in a single
photoreceptor. - To see color, we need multiple cone
types, each sensitive to different ranges
of wavelengths
Lights of 450 and 625 nm elicit _______.
the same response.
Wouldn’t be able to tell them apart, it needs to compare responses across different cones, needs at least one other one.
this is why we need multiple cones .
Univariance: The probability of
firing depends on ______
intensity and wavelength
M-cone fires at same rate for
bright Cyan (500nm)
* Dim Green (534nm)
* Bright Orange (580nm
Since a single cone can’t tell the
difference, color perception
requires _______
comparing firing rates of all three cone types.
Under photopic conditions (lots of light),
______
the S-, M-, and L-cones are all active.
Rods specialize in _____
low light vision (scotopic)
rods follow the _____
principle of univariance—they detect
brightness, not color.
All rods contain _______, a single
photopigment, making them ____
rhodopsin; equally sensitive to different wavelengths
Trichromatic Theory Of Color Vision
(aka Trichromacy - Young-Helmholtz theory)
How we perceive color based on the activation of the three different cone receptors
Color perception comes from comparing signals from three types of cone photoreceptors.
* S-cones → Peak at blue, M-cones → Peak at green, L-cones → Peak at red
in scotopic conditions, only ____ are
active, which is why night vision is
______
rods; colorblind
what cant Trichromatic Theory Of Color Vision account for?
Cannot account for afterimages, effect of surround color, or that some colors do not seem possible to mix. It has limitations.
(Trichromatic Theory Of Color Vision) Brain interprets color by ______—not by detecting _____
analyzing the relative activation of these cones; single wavelengths
example of Trichromatic Theory Of Color Vision
Yellow light doesn’t strongly
activate S-cones, but it equally stimulates M- and L-cones, making us perceive yellow
Metamers
colors that look identical in one light but different in another
example of how different wavelength mixtures can trigger the same cone responses, making colors
appear the same
Red + Green light = Yellow (even
though no “yellow” wavelength is
present)
example of metamer
A blue shirt may match your shorts
under store lighting but be mismatched in
sunlight
Additive Color Mixing: How Light Colors Combine
Colors combine by adding wavelengths, not blending pigments.
Primary colors form basis for this.
(Additive Color Mixing) perception depends on ________
wavelength combos
Light A + Light B = new perceived color.
Subtractive Color Mixing: How Pigments Combine
Happens when pigments are combined, and certain wavelengths are absorbed (subtracted), leaving only reflected colors visible.
- Not thinking about light, think of paints, mixing paints
- Mixing pigments removes wavelengths,
- each pigment absorbs specific colors, so mixing more results in fewer reflected
wavelengths - The primary colors here are cyan, magenta, and yellow (CMY), (same as printer)
example of subtractive color mixing
Mixing blue and yellow paint absorbs blue and red, leaving green as the only reflected color.
Cone-opponent cells ______ to
process color
compare signals from different cones
Subtractive Mixing & Filters: How
Wavelengths Are Removed
- White light contains all wavelengths.
- Yellow paint or filter absorbs short
wavelengths, reflecting medium and
long wavelengths, which look yellow. - Blue paint or filter absorbs long
wavelengths, reflecting short and
medium wavelengths, which look blue. - Mixing yellow and blue leaves only
medium wavelengths, which look green.
where are cone-opponent cells found?
Found in retina ganglion cells,
LGN, and cortex, they have a
center-surround organization.
how do cone-opponent cells compute color contrast?
compute color contrast by
exciting some cone inputs and
inhibiting others. helps us see edges and color boundaries better, why we see red and green as opposing colors.
Opponent Cell processing enhances _____
color boundaries and differentiation
Single-opponent cells compare signals from
from one cone type against another cone type, distinguish broad color.
(Are circular – Retina & LGN) typically have circular fields
Double-opponent cells
once in v1 you find these.
compare contrasts across different areas of the receptive field, refining color edges and patterns.
Complex color contrasting, not only just comparing, they do it across different spatial areas of the receptive field.
Elongated oval field, processing color edges and patterns. Textures, fine details. Looking for two things. Color constancy.
Opponent Color Theory
The perception of color is based on three opponent mechanisms, each processing two opposing colors.
- Red–Green: L-M or M-L
L & M cones have opposing responses. - Blue–Yellow: (L+M) – S or S – (L+M)
S cones compare signals against the sum of L & M - Black–White: Achromatic pathway
The brain processes brightness separately from color
why cant you see a reddish-green?
Opponent colors. Because of color discrimination.
“Legal” color mixes
bluish-green (cyan), reddish-yellow (orange), and bluish-red (purple).
“Illegal” color mixes
reddish-green or bluish-yellow are impossible to perceive.
Unique hues are defined by _____
opponent processing. meaning they do not contain traces of their opposite color.
Hue cancellation method helps identify
Unique hues by
adding an opponent color until no trace of the opponent remains.
Unique blue has no
red or green tint
Unique red has no
greenish hue
Unique yellow has no
blue or purple tint
Unique green has no
reddish hue.
Unique hues act as _____- in
the color spectrum and support
opponent-color theory
perceptual anchors
afterimage
A visual image that lingers after a stimulus is removed
Negative afterimage
Colors appear as their opponent (e.g., red → green, blue→ yellow).
- Opponent colors reveal
themselves—red produces green
afterimages, and blue produces
yellow (and vice versa). - Light stimuli create dark
afterimages. - This demonstrates opponent
processing in action
fatigue Effect
occurs when prolonged exposure to a color reduces cone responsiveness.
When photoreceptors adapt, their
opponent color becomes visible once the
stimulus is removed.
M- & L-cone genes are on the ___-chromosome
X
Males (XY): One X, so M & L mutations are more common →higher color blindness rates.
* Females (XX): A second X can compensate, lowering risk a LOT
S-cone mutations happen in 2 places
Chromosome 7 (tritanopia): Equal in males & females, blue-yellow color deficiency
X-linked (S-cone monochromacy): Very Male-biased true color blindness
A more accurate term than “color blindness.
Color-anomalous. Most color-deficient individuals can still distinguish wavelengths, just differently from the norm.
Trichromats
Normal Vision
Protanope:
L-cones absent, reduced sensitivity to red
Deuteranope:
M-cones absent, hard to distinguish between red and green
Tritanope
S-cones absent, rare deficiency
Cone Monochromat
Has only one functional cone type, leading to a lack of color vision. World in 1 tint
Still uses cones for daylight vision
Can distinguish 100 shades vs average of 1 million.
S cone only, 1 in 100,000
M or l cone only, < 1 in a million
Rod Monochromat, aka Rod Achromatopsia
has no functional cones, relying only on rods.
* Truly color-blind and severely sensitive to bright light.
* Very poor visual acuity due to reliance on low-light vision
Cortical Achromatopsia
Aka Cortical Color Blindness
individuals see in shades of gray with
normal acuity and contrast sensitivity
loss of color perception due to brain damage
Caused by damage to V4 in the ventral
pathway.
can happen without damage ,disconnection syndromes.
Color Anomia
Specifically affects color
naming—individuals can see and distinguish colors but cannot name them due to a language-processing deficit.
damage to language-processing areas in the left temporal lobe or inferior parietal lobule.
synesthesia:
A condition where one sensation automatically triggers another
sensory experience.
Consistent & Involuntary – Once
formed, associations remain stable
4-5% prevalence, influenced by both
genetics & environment
example of synesthesia
Study found correlation between fisher price magnets and grapheme color synesthetic associations (letters
evoke colors)
Color space
A three-dimensional space that
describes all colors
RGB: red, green, and blue light
intensities
HSB: Hue: chromatic (color), Saturation: chromatic strength of a hue, Brightness: overall lightness of the color, ranging from black (0%) to full brightness (100%)
Spectral Colors:
Hues from approximately 380 to 750 nm correspond to single wavelengths of light
Nonspectral Colors
Some colors, like magenta and other purples, do not exist as single wavelengths but instead arise
from a mix of multiple wavelengths.
Perception of Magenta & Purple
These colors are perceived when S- and L-cones are strongly activated, with little to no intermediate M-cone activation, leading the brain to interpret a color that does not exist in the visible spectrum
Color contrast
occurs when one color induces its
opponent color in a neighboring region due to opponent processing. (A red square appears brighter on a green
background than on a red one)
Color assimilation AKA Spread Effect
illusion that occurs when colors blend
together locally, making adjacent colors take on qualities of each other
-causes colors to merge.
* Alters perceived hue, brightness, or
saturation
Absolute (Unrelated) Color:
color can be perceived in isolation, without contex.
Best experienced against a white
background.
- Do not shift appearance due to
color contrast effects or assimilation
Yellow
can be both spectral and a non spectral color.
There is yellow on the rainbow. 570 nm. Yellow is perceived as yellow when we look at that.
But also because opponent processing, if we have equal red and green, our system does not know what to do, so we see yellow, that is a non-spectral color.
Related Color:
A color perceived only in relation to others
includes Brown, Grey, Yellow
Color constancy
Our ability to perceive an object’s color as stable even when lighting conditions change.
- achieved through context and
prior knowledge of how light behaves
To maintain color constancy, the brain ______
estimates how an illuminant affects the
wavelengths reaching our retina.
allows us to interpret the true
color of objects, even though the raw
sensory input varies under different
lighting.
ex of color constancy depending on individual perception
the dress (white/gold or black/blue).
Your brain adjusts for lighting conditions, making the same colors look different.
*People’s past experiences with light affect whether they see white & gold or black & blue.
*Some individuals can actively switch their perception by reconsidering the light source
our perception obeys ______—a survival mechanism that usually helps but can lead to illusions
physical principles
from Retina to LGN: How Color Signals Are Processed
primate LGN organizes visual info
into 6 layers, each processing
different visual info
Magnocellular (Layers 1-2): Primarily
processes motion and brightness via rod-
dominated M cells; not involved in color.
Parvocellular (Layers 3-6): Receives input
from P cells, processes M- & L-cone
opponent signals for fine details and red-
green color vision.
Koniocellular (Between M & P): Role is less
understood but known to process S-cone
(blue/yellow) opponent signals.
Color perception extends beyond the retina and LGN requiring _______ to create color appearance—how we experience color in different contexts
cortical processing
_____ plays a crucial role in color constancy, ensuring colors ____
v4; appear stable despite changes in lighting.
types of Depth Cues
Nonmetrical depth cues show depth order
but not exact distance
Metrical depth cues provide measurable
depth info
- Relative: Indicate how far objects are
from one another but not exact distances
- Absolute: Give precise distance measurements, though human accuracy
is limited compared to some animals
Depth cue
provide info about
spatial relationships in visual
perception.
2 Kinds
1. Oculomotor – use eye muscle
feedback to estimate
- Retinal Image Based – use info
from 1 or both eyes
- Monocular Cues - rely on 1 eye. 2 kinds: static & dynamic
- Binocular Cues - rely on differences between images from both eyes
Oculomotor cues
rely on eye muscle feedback & can provide absolute metrical depth info for close objects – but we don’t use it that way. Is relative for most, at best.
for depth perception brain Relies on
2 non matching retinal images, each with their own distortion and blind spots
Accommodation
ciliary muscle adjust the lens to focus on near objects. Provides depth info but is
limited beyond 1–2 meters.
Convergence
Eyes rotate inward to focus on close objects. Stronger than accommodation but only effective within 2 meters
9 Static Monocular Cues (2D
Pictorial) in 3 Categories
position based
- partial Occlusion
- Relative height
size based
- Familiar size
- Relative size,
-Texture gradients,
- Linear perspective
lighting based
- shading
- Cast shadows
Partial Occlusion
- Universally used depth cue
- Nonmetrical, Ordinal - Only gives depth order, not relative or absolute distances
- Most reliable depth cue – Works in nearly all visual environments with 1 eye & no movement
Relative Height AKA Relative Position
Objects lower in the visual field on the
ground appear closer; those higher appear farther.
Works with Relative Size to enhance
depth perception.
horizon line effect
Near the horizon = farther; farther from the horizon = closer.inverted for Ceiling Objects: Higher in the visual field = closer.
Relative Size
compares objects without having any prior knowledge applied to it about their actual size.
Smaller objects are perceived as farther away
* Example: If object A appears twice as large as object B, we infer their relative distance without knowing exact
measurements.
Familiar Size
relies on knowing the object’s actual measurements. If we recognize and know its real word dimensions, we can judge its depth more accurately.
Motion-based depth cues arise from ___
from retinal image changes as we move; are all relative metrical
Texture Gradient
When surface elements maintain consistent size and spacing, their retinal
image shrinks with distance, creating depth perception.
The farther objects are, the smaller, less
detailed, and more densely packed they
appear
As it recedes off into distance, we get less detail, it gets smaller.
Strongly influenced by relative size and
relative height
Applying to a continuous surface. Important for natural environments.
Cast Shadows
Shadows provide depth cues by indicating an object’s position relative to a light source in real world situations
Effectiveness depends on the viewer’s assumptions
about light direction and object size
Long shadow – farther from low light source
Objects farther from a light source or in
shadowed areas appear darker due to reduced illumination
Close shadow – light directly overhead / close to light
linear Perspective AKA Perspective
Convergence
Parallel lines appear to converge toward a
vanishing point in the distance
Works even though the lines remain parallel in reality
Common in railways, roads, and architecture
Allows us to judge some distance info. In man made areas, we combine with texture gradients.
Shading
Allow us to process 3d shapes quickly. Depth perception is strongly influenced by this.
the brain naturally assumes light
comes from above, as from the sun
- Objects with light on top and
shadow below appear raised, while
the reverse looks indented
This 2D cue is crucial in interpreting
3D shapes and navigation
Deletion & accretion
Occur when objects gradually disappear behind another object (deletion) and gradually reappear on the other side (accretion)
Basing depth cue on speed at which it becomes covered and then uncovered
Faster deletion/accretion suggests
closer objects, while slower changes
indicate greater distance
Atmospheric Perspective AKA Aerial
Perspective
Relies on understanding that as the scene expands outward, we are going to get less detail and scattering.
Objects that are further away, air between it will have more particles causing less contrast.
Nonmetrical, Ordinal - Distant objects
appear hazy due to moisture, dust, and
particles in the air
Blurred edges, lower contrast, less color, less sharp lines. Blurred edges and lower contrast make far objects seem farther away
Large scale distances, such as separation between hills.
Optic Flow
Describes how objects in a scene shift on retina as we move through it
Focus of expansion, point in the visual field where there is no motion. It is going to constantly shift.
Occurs when going forward and backwards
Objects in our view appear to expand
outward as we move forward from a central point; moving backward contracts the scene
Objects closer appear to move faster, while distant objects move slower
Motion Parallax
Uses relative movement to get distance info
occurs when observer moves sideways or turns their head, away from optic flow
Head movements and relative motion between objects reveal depth
Makes close objects appear as if they are moving quickly compared to the distant ones
Binocular Vision
Static Depth Cues from two eyes
Binocular Vision nonmetrical
Specialized neurons in the ventral
(what) pathway categorize depth by coding near vs. far relationships without precise measurements.
Binocular Vision Mostly Relative Metrical
Hyperacuity cells in the dorsal (where/how) pathway process disparity
with precision finer than individual photoreceptor spacing, for exact depth calculations.
Gives Absolute depth if paired with certain
other cues
Binocular Disparity
The brain calculates depth based on the
degree of retinal disparity between the two images from each eye.
- Greater disparity = closer objects; smaller disparity = farther objects.
Binocular disparity provides relative
depth, but combined with vergence or
familiar size, it can yield absolute depth
Processing of can lead to stereopsis
stereopsis
The brain’s ability to vividly perceive depth using disparity, creating a strong 3D effect.
Horopter
an empirically measured curve that
varies depending on actual biological factors. Has Corresponding & Noncorresponding Points
Panum’s fusional area
The region of space, in front of and behind the horopter, within which binocular single vision is possible
Objects ON the horopter ___
fall on corresponding points, appearing at the same depth (no disparity).
Are seen as single images when viewed
with both eyes
Stereoscope:
A device for presenting one image to one eye and a slightly different image to the other eye
Creates depth illusion from two 2D
photos shown into separate eyes to
mimic natural binocular disparities of
scene onto both retinas
modern stereoscope
virtual reality
The Oculus Rift VR gaming headset is a modern stereoscope that renders real-time, dynamic images
Objects off the horopter ___
fall on noncorresponding points, creating
binocular disparity (2 images), which the
brain uses for stereopsis (depth
perception).
Greater disparity = closer objects;
smaller disparity = farther objects
if visible in both eyes, stimuli falling
outside of Panum’s fusional area will ____
appear diplopic
Diplopia: Double vision.
3 types of Binocular Disparity
- Zero disparity: object at horopter; no
disparity - Crossed disparity: objects in front of the
horopter.
Larger disparity for closer objects
Objects closer than the horopter shifted
right in left eye, and left in right eye - Uncrossed disparity: objects behind the
horopter.
Larger disparity for farther objects
Objects farther out than horopter shifted
left in left eye, shifted right in right eye
Free fusion
converging (crossing) or diverging (uncrossing) the eyes to view a stereogram without a stereoscope
Reveals how binocular disparity alone can drive depth perception
Magic Eye” - Rely on free fusion
Stereoacuity
A measure of the smallest binocular disparity that can generate a perception of depth
Shows up suddenly in infants 3 – 5 months
Stereoacuity is often tested using dichoptic stimuli
How is stereopsis implemented in Brain?
Disparity-detecting neurons found V1,
V2, V3 and extends to dorsal stream
(MT/V5, parietal lobe).
these neurons detect differences in object position between two eyes
Tuned for both amount & type of
disparity (crossed vs. uncrossed)
Only fire when retinal images have the
right disparity, aiding depth perception
Random dot stereogram (RDS):
A stereogram composed of randomly placed dots
Implications: Correspondence precedes
object recognition in visual processing.
Correspondence problem
In binocular vision, the problem of
figuring out which bit of the image
in the left eye should be matched
with which bit in the right eye
- Algorithm used by brain is unknown – however:
- Correspondence necessary
- Don’t need actual objects to do it – Random Dot stereograms work well
There are several ways to solve the
correspondence problem & aid
smooth disparity:
Blur high-frequency details to
make large structures easier to
match.
- Match each feature once per eye
(Uniqueness Constraint). - Assume smooth depth except at
object edges (Continuity Constraint)
Cyclopean
defined by binocular disparity alone
Demonstrates stereopsis without monocular depth cues
Stereo blindness
An inability to make use of binocular disparity as a depth cue
- 3-5% of the population
- Free fusing does not give depth, although in focus
- Usually from childhood visual disorder, such as strabismus (cross-eyed).
- Most people who are stereo blind do
not realize it. Not necessary for modern life
Dichoptic Stimuli
Referring to the presentation of two stimuli, one to each eye
Binocular cooperation
Both eyes normally work together to create a unified depth perception
_____ enables stereopsis, while ____ reveals competition in vision that can disrupt normal development.
Cooperation; rivalry
Binocular rivalry
When each eye sees different images, perception alternates rather than
merging
-In misaligned eyes, rivalry leads to chronic suppression, affecting stereo vision
Dominant eye effect:
the stronger image is prioritized while the other is temporarily suppressed
Strabismus
A misalignment of the two eyes
such that a single object in space is imaged on the fovea of one eye, and on a non-foveal area of the other (turned) eye
Exotropia
Strabismus in which one eye deviates outward
Depth perception relies on
multiple cues working together
Esotropia
Strabismus in which one eye deviates inward
Suppression
In vision, the inhibition of an unwanted image. (stereo blindness)
If strabismus is present during the critical period, it can lead to
stereo blindness
Illusions often arise from
perceptual committees resolving ambiguity
Depth perception combines
sensory input with prior knowledge.
The visual system automatically _____
, sometimes leading to errors
integrates cues
Types of Attention
Exogenous (stimulus-driven): Sudden
events capture attention (e.g., loud sound).
Endogenous (goal-driven): We can choose
where to focus (e.g., reading)
The Bayesian approach explains how
we estimate probability in depth
perception.
Our brains favor the most likely interpretation of ambiguous depth
cues.
Prior experience influences whether
we perceive depth accurately or see
illusions
Perceptual Committee Goals
Size Constancy: Object size remains
stable, even as distance (retinal image
size) varies
* Perceived size & perceived distance
related
Shape Constancy: Object shape remains
stable, even retinal image shape varies
Shape-Slant Invariance: Shape perception adjusts based on slant cues.
* Without these constancies, objects
would seem to change size and shape
as they move
Selective Attention
helps prioritize
We cannot process everything at
once → leads to sensory overload.
- Selective attention filters info
based on relevance
We can only attend to one task
at a time.
- Guides eye movements toward
what matters next
Four Selective Attention Processes
- External: Attending to stimuli in
the environment (sensations) - Internal: Focusing on thoughts or
decisions (perceptions, planning,
problem-solving) - Overt: Shifting gaze or body toward
a stimulus (visible movement) - Covert: Attending without outward
signs (hidden focus)
2 Types of Attention Management
- Divided – splitting focus between multiple tasks.
* Humans cannot divide simultaneously; we switch attention rapidly instead - Sustained – Maintaining focus over a prolonged period. AKA “vigilance”
Humans not great at sustaining attention
* Certain animals, like raptors & pigeons, excel at it
Cognitive Load Theory (1988):
Limited working memory reduces performance when attention is divided
Switching Costs
task-switching slows speed and increases errors
accuracy & Speed inversely related to
number of tasks.
* Impinges on learning & productivity
* Habitual multitasking reduces &
impairs sustained focus abilities
Driving requires
sustained attention; divided attention
increases crash risk.
Generating words while driving impaired performancemore than simple word repetition (2001)
- Talking to a passenger impairs driving less than talking on a cell phone (2008)
- Drivers were inattentive in some way 3 seconds before 80% of crashes (100 car naturalistic study)
- Drivers on the phone missed twice as many red lights as those not on the phone (lab study)
treisman’s Attenuation Model of Attention (1964)
Suggests that instead of blocking unattended info, it’s attenuated (weakened).
* All sensory input reaches further
processing, but unattended info is reduced.
* Explains why unattended info can still be
processed.
* Less clear on how the attenuation mechanism functions in the brain
Sustained Attention lasts between
20 and 30 minutes, but varies greatly by
task
Attention is
highly task-dependent and
cannot be accurately reduced to a
single, universal measure
Broadbent’s Filter Model (1958)
suggests attention acts as a bottleneck, filtering info early based on physical characteristics.
Only attended information reaches
higher-level processing, while
unattended stimuli are blocked
Explains why we cannot process all
incoming information at once
Cannot explain how unattended stimuli
can still influence behavior (e.g., Cocktail
Party Effect)
Posner’s Spotlight Model (1980)
Attention enhances processing in a focused area, i.e., spotlight metaphor.
* Selective, voluntary, and effortful: Explains attentional shifts but oversimplifies them
Zoom Lens Model (1985):
Attention expands/contracts from fixation
“Transporter” Perception (1995):
Attention jumps between locations rather than moving continuously.
Reaction time (RT)
Interval between stimulus and response
posner Cueing Paradigm
method to examine how cues affect spatial attention. Uses valid, invalid, and
neutral cues to measure attention shifts
Demonstrates how spatial cues impact response time and visual processing
Inhibition of return
Difficulty revisiting a recently attended location
cue:
A stimulus that hints where/what a target will be; can be valid, invalid,
or neutral
stimulus Onset Asynchrony
Measures how quickly attention shifts.
Endogenous cue
voluntarily; slower than Exogenous cues
* AKA Symbolic cue
Exogenous cue
involuntary; driven by physical salience
* 100 – 150 ms for cues in Periphery to be fully effective
* AKA Peripheral cue
Visual search:
Looking for a target in a
display containing distracting elements
exs of visual search
Finding weeds in your lawn or
the remote control on the coffee table
visual search elements
Target: The goal of a visual search.
- Distractor: any stimulus other than the
target. - Set size: The number of items in view
2 kinds of visual search
Feature & Conjunction Search
Feature search
Finding a target with a single attribute, such as a SALIENT color or orientation
Parallel Feature Search:
processing multiple stimuli at the same time
Single features seem to be
processed in parallel b/c
the number of distractors does not
impact reaction time (RT)
(feature search) High Efficiency:
RT x Set Size Slope = 0 ms
Conjunction (or Serial) Search
Target lacks salience, requiring multiple
attributes for identification.
- Search time increases with set size; even
steeper slope when target is absent
Ann Treisman’s Feature integration Theory (1988)
explains how we bind separate features into unified objects.
object -> preattentive stage -> focused attention stage -> perception
Features (color, shape, etc.) are initially
processed separately, requiring attention for correct integration
Serial self-terminating search
Examines items one at a time, stopping when the target is found
Unfamiliar stimuli:
Search is much harder if characters or symbols are unknown.
(FIT) Preattentive Stage
Features like color, shape, and orientation are processed separately
Parallel Processing: Early processing
happens automatically without focused
attention
(FIT) Focused Attention Stage
attention binds features together to form a cohesive perception.
- Selective Attention: Required for feature
binding and object recognition
(FIT) Binding Problem: How does the brain combine features into a single percept?
Selective Attention ensures features are
bound correctly, preventing misperceptions.
- This challenge occurs in both vision and
audition—misheard words may result from incorrect binding
(FIT) Illusory conjunctions
occur when features from different objects are mistakenly combined.
Demonstrates why focused
attention is essential for perception
accuracy
Evidence for FIT: Illusory conjunctions show that attention is needed to correctly bind features
Guided Search Theory
Attention is directed to a subset of possible items based on basic features (e.g., color, shape)
Rapid parallel feature processing
occurs first, followed by slower serial
processing using top-down cognition.
Conjunction Search: A type of guided
search where the target is defined by a
combination of attributes rather than a
single feature
(GST) Scene-based guidance
Prior knowledge about a scene helps
locate objects efficiently
(gst) Whole Scene Perception
A mix of attention, eye movements,
and memory shape how we
process visual scenes
(gst) Context Matters:
Objects are expected in specific locations
(e.g., books on horizontal surfaces, paintings on vertical
surfaces)
Rapid serial visual presentation
(RSVP)
an experimental procedure where stimuli appear in a rapid sequence at a single location.
- Presented at ~8 items per second;
no eye movements needed. - Used to study temporal attention
dynamics – how attention shifts
over time. - Helps investigate Attentional Blink,
a lapse in detecting the second of
two rapid targets
Attentional Blink (AB) is modulated by
emotion and salience.
less blink for emotional words & own’s
name
analogy: cocktail party effect where we
hear our name in an unattended
conversation in a noisy environment
Visual attention performance can be
improved with practice.
* Green and Bavelier (2003) reported
first-person shooter video game players
have reduced attentional blink
Attentional Blink
Difficulty perceiving a second target amid an RSVP stream
Occurs if the second target appears within 200-500 ms after the first.
- Reflects limits of attention shifting and consolidation.
ex of attentional blink
Missing a fish while catching another—attention is occupied
Attention enhances
neural activity in specific parts of
the visual field
Neurons coding attended locations
show increased activation
Neural response size depends on attentional focus
biased Competition Theory
Many neurons have large
receptive fields, allowing
multiple objects to compete for
processing
Attention resolves competition
by enhancing responses to
relevant objects
Neurons respond more strongly
to effective stimuli when
attended and weakly when
ignored
Three ways responses of a cell could be changed by attention
- Response enhancement – Neurons fire more intensely when attention is directed at a stimulus.
Example: Focusing on a bright object increases neural firing in brightness-sensitive neurons - sharper tuning – Neurons become more selective for a specific feature (e.g., color, orientation).
Example: Attending to a red object sharpens a neuron’s response to that shade - Altered tuning – Attention shifts a neuron’s sensitivity to prioritize certain stimuli. Example: Searching for vertical lines shifts neurons to prefer vertical orientations.
3 Highly Interactive Attention Networks
- Subcortical System – Involuntary Orienting - rapid shifts to new location/object
* Superior Colliculi: Shifting Attention
* Pulvinar of Thalamus: Engaging Attention - Ventral Cortical System – Bottom-up, Stimulus- Driven for unexpected, salient stimuli
* Temporal-Parietal Junction & Ventral Prefrontal Cortex - Dorsal Cortical System – Top-down, Goal-Directed. Voluntary, Sustained attention.
* Superior Prefrontal & Posterior Parietal Cortex
____ Controls Covert Orienting
Parietal Lobe.
moving attention without moving the eyes
Same-object advantage
occurs when attention spreads within an object, enhancing detection at multiple locations within that object.
* Reaction times are fastest at the cued
location but are also improved at the
uncued end of the same object.
* This suggests attention is not limited to a
single point but can extend along an entire object.
Selective pathway
Bottlenecked—only a few
objects processed at once
Uses feature binding (color,
depth, motion) for recognition
Crucial for searching specific
items in cluttered scenes.
Example: Looking for a red
apple in a mixed fruit basket
object-based attention,
perception is influenced by the structure of an object, not just spatial location
Two pathways process scenes
Selective: Recognizes individual
objects but is limited by attention.
- Nonselective: Provides a broad
scene “gist” instantly
Nonselective pathway
processes global scene properties without
identifying objects
parallel processing—not limited by
attentional focus.
- Extracts spatial layout, texture, and
color distribution instantly. - Provides rapid scene understanding
(e.g., indoor vs. outdoor). - Example: Walking into a café and
assessing available seats without
focusing on details
The nonselective pathway
extracts _______ across
a scene.
summary statistics
Gist perception occurs within
milliseconds
Objects recognized better in
expected scene
Scenes recognized better with
expected object
- Top-down (expectations) &
bottom-up (stimulus-driven)
processes interact. - Processing balance depends on
object type, scene type, and time
constraints
Spatial Layout & Gist
the nonselective pathway processes
spatial layout automatically
Scene structure is processed
holistically (open/closed,
natural/urban)
Openness & closeness guide scene
understanding, affecting recognition
speed.
Ensemble statistics
Perception of global properties (e.g.,
average color, size, motion).
- Rapid, without focusing on
specific objects. - Helps detect trends in groups,
like a school of fish or a forest’s
color distribution.
When attention works well:
We can recognize differences between
new and old information very well
Change Blindness
failure to notice a change between two scenes.
Memory for scene changes is surprisingly poor.
Attention is naturally limited,
requiring us to focus on one area
at a time.
- Large changes can go unnoticed if
they don’t disrupt scene meaning. - This phenomenon demonstrates
how our perception prioritizes gist
over details
Inattentional blindness
Failure to notice an unexpected stimulus in plain sight due to focused attention elsewhere
Attention is limited—what we don’t focus
on can go unseen
attention is ____ for awareness
required
seeing does not equal perception
Visual-field defect
A portion of vision is
missing due to damage in the visual system, affecting perception regardless of attention.
Neglect
Patients ignore one side of space
despite intact vision, failing to respond to
stimuli on the affected side
Extinction
occurs when a stimulus is ignored due to a competing stimulus in the opposite visual field
The ignored stimulus is detected alone
but not when paired with another
2 types of neglect in extinction
Spatial Neglect: Failure to perceive
stimuli on one side of fixation
- Object-Centered Neglect: Ignoring one
side of an object, no matter its position
does neglect equal blindness?
no, patients unconsciously respond to stimuli they don’t report seeing
Contralesional field
Side opposite the lesion, where attention is impaired.
Ipsilesional field:
Same side as the lesion, where attention is biased
Damage to the right parietal lobe
often causes substantial left-side neglect.
left hemisphere damage results in
milder right neglect.
Balint’s Syndrome is caused by damage
to both parietal lobes. It is characterized by 3 main symptoms
- Simultanagnosia: Inability to
perceive multiple objects at once,
disrupting scene comprehension - Oculomotor Apraxia: Difficulty
voluntarily shifting gaze between objects - Optic Ataxia: Difficulty reaching for objects using visual guidance
Patients with Balint’s Syndrome
name objects but fail to grasp the full scene
is one of the most common attention disorders.
Attention Deficit Hyperactivity Disorder
(ADHD)
ADHD marked by
Impulsivity
* Hyperactivity
* Inattentiveness
Despite attention deficits, basic visual
processing remains largely intact.
- Research suggests differences in
attention lapses, visual search, and
distractibility but no major
impairments in visual perception
Self-motion
optic flow
Absolute motion
aids depth perception when no background reference exists – no depth cues
Example: Detecting speed &
direction of a flying bird against a blank sky
Visual ambiguity
Without reference points, absolute motion
can be difficult to interpret, leading
to perceptual illusions like the
spinning dancer, where motion
direction becomes ambiguous.
Motion Aftereffects (MAE)
After prolonged motion exposure, stationary objects appear to move in the opposite direction.
due to adaptation in direction-selective
neurons, reducing sensitivity to sustained
motion
Opponent process
Like color aftereffects,
adaptation shifts perception in the opposite direction
MAE ____ when switching eyes,
indicating higher-level motion
processing
persists
a key motion processing area.
MT/V5
Interocular Transfer & MAE
Occurs in neurons responding to
both eyes, meaning in V1 or beyond
Reichardt Detectors explain how
neurons compare motion signals from
two locations
Uses excitatory & inhibitory
interactions to compute motion
* Responds to both real & apparent
motion
Motion detection relies on
spatially separated receptive fields
Motion is a change in position over
time. Double dissociation reveals 2
kinds
- First-order motion: Detected by
luminance changes, like objects
moving against a background.
* V1 → MT - Second-order motion: Perceived via
contrast, texture, or flicker, - not
defined edges.
* Extrastriate cortex beyond MT
delay mechanism allows
signals from adjacent fields to be integrated, detecting motion over time
Neural adaptation leads to
MAE
Fatigue in one direction
causes perception of opposite
motion
Opponent motion detectors
compare
leftward vs. rightward motion signals
Excitatory-inhibitory interactions suppress
responses to stationary
objects.
- Motion opponency enhances
detection of true movement
Apparent motion
the illusion of smooth motion from separate static images
Movies, stop-motion, & illusions rely
on this effect
Apparent motion. in 1878
1st movie made with 16 sequential horse-riding images
The brain treats apparent & real
motion
as the same process. fMRI studies
show similar V1 activation for
both
Apparent motion quartet
Perceived motion depends on object
spacing
Closer vertical spacing → Perception
favors vertical motion
Closer horizontal spacing →
Perception favors horizontal motion
Related to Gestalt Proximity
Principle—motion perception
minimizes distance traveled
Beta Motion
Sequential lights create smooth
motion perception
- Basis of movie projection & digital
animation
Sensitive to timing & spacing between
flashes
Phi Phenomenon
Alternating objects appear
to “jump” between positions
Creates the illusion of an object moving
when none is present
Relies on the same motion-detection
circuits as real motion
Correspondence problem (motion)
The challenge the motion detection
system faces in determining which feature in frame 2 corresponds to
which feature in frame 1.
Ambiguous motion: Are the dots moving horizontally or vertically?
Aliasing problem
When the visual system (or a camera)
samples motion too slowly, it
creates perceptual errors
ex: Tire spokes or helicopter blades may
appear still or move in the
wrong direction due to a
mismatch between motion
speed & frame rate
Aperture Problem
When viewing
motion through a small window
(e.g., receptive fields in V1), the
true direction of motion is
ambiguous.
ex. Example: The Barber Pole
Illusion—motion appears to
move upward even though
stripes are moving diagonally.
Neural Solution for aperture problem
The brain integrates motion signals across
larger receptive fields (MT)
V1 motion components: 1 set of RFs
detects horizontal stripe movement &
another detects vertical
Plaid Motion Perception: combining
signal sets allows integrating them into a
single motion perception, including
Diagonal
Neural Integration: V1 neurons have small
RF, detecting only partial movement
(horizontal or vertical).
* MT neurons integrate signals from
multiple V1 RFs, reconstructing the
actual motion direction
Motion binding
integrates separate
motion signals into a unified perception
of an object
In complex scenes, perceiving
connected motion requires first
binding individual object features.
- The brain assumes hidden parts are
continuous & fills in the missing
information. - Essential for object recognition &
distinguishing self-motion from object
motion
MT (middle temporal area) integrates
local motion signals to perceive global
motion.
Akinetopsia
rare disorder where motion perception is completely lost.
- Caused by severe bilateral
damage to MT (middle temporal
area). - Brain fails to integrate motion
signals, making life appear as a
sequence of still images. - Less severe MT damage →
Motion appears blurred instead
of absent
Lesion studies reveal MST’s role in motion
perception
Radial motion (expansion/contraction along a radius).
- Circular motion
(clockwise/counterclockwise). - MST processes optic flow, detecting self-
motion in an environment
___ resolves motion ambiguity & the
aperture problem.
MT
Ambient optic array:
The structure of light in the environment that changes with movement
_____
processes more complex motion patterns than MT.
MST (medial superior temporal area)
Optic flow:
The pattern of motion
perceived as we move, providing
direction & speed cues
Focus of expansion (FOE):
The point in optic flow indicating
heading direction
Motion reshapes the ____,
creating optic flow patterns that
______
optic array; guide navigation & balance
processes motion, spatial location, & visually guided actions
Dorsal Stream
Tracks Tau & Guides motor actions to support interactions with moving object
Fed by Magnocellular Pathway: prioritizes motion & contrast (Parvocellular Pathway processes color).
Biological motion processing occurs in the
posterior superior temporal sulcus (STSp)
fMRI evidence:
* STSp activation when viewing point-
light walker displays.
* TMS disruption: Temporary STSp
inactivation impairs biological
motion perception.
* Lesion studies: STSp damage leads
to difficulty perceiving social motion
cues
Biological motion
Perception of movement
unique to living organisms
Gender cues: Subtle motion differences help identify male vs. female gait.
- Body size: We detect body composition
through movement patterns. - Species differences: Humans rely on
biological motion cues, but some birds (e.g., pigeons) use global motion. - Point-Light-Walker Display: Minimal cues
(joint lights) still reveal movement
coordination.
How do we estimate time to collision (TTC) when depth cues fail?
Tau (τ): Retinal expansion rate signals
approach speed
TTC is proportional to tau—rapid image
expansion means impact is imminent.
- MT detects motion speed & direction; MST integrates optic flow to predict motion trajectory.
- Example: A car ahead grows rapidly larger → imminent collision → brake now
MT tracks
object motion (speed &direction)
MST integrates
global motion (optic flow, large-scale movement).
Motion perception struggles with ___ stimuli, because M path processing relies on contrast
isoluminant
Motion-Induced Blindness (MIB)
Stationary objects vanish when viewed against a moving background.
Brain prioritizes motion, filtering out
unchanging stimuli (perceptual suppression).
- Isoluminant stimuli disappear more easily due to reduced contrast in motion
processing. - Fixation increases suppression, making
peripheral objects vanish
Eye movements are essential to view a scene due to limited high-resolution vision in the fovea. 4 Types:
- (Con/Di) Vergence – Depth Control. Aligns eyes on moving object.
- Vestibulo-ocular Reflex (VOR)– Stabilizes gaze. Reflexively moves eyes opposite to head
- Smooth pursuit – Tracks object. Is fluid
voluntary eye movements. - Saccades – Reset Gaze. Rapid, most are
involuntary eye shifts between fixations
these 3 processes
ensure motion
perception is stable &
accurate while reducing
blur caused by both eye
& object movements
- Tracking & Eye Compensation –
Adjusts eye position to maintain
stable vision as objects or the head
move (includes VOR). - Object vs. Self-Motion – Distinguishes
external movement (motion blur
expected) & self-movement (scene
stabilized). - Saccadic Suppression – Briefly blocks
visual input during saccades to
prevent motion blur.
___ occurs when objects move relative to the eye (or camera!)
Blur
Fixating on a moving object blurs the
background; fixating on a stationary object blurs motion
Vestibular System
Inner ear system detecting motion, aiding in vision & balance
Motion is detected only when the object
moves across direction-sensitive receptive fields (RFs)
When the eyes are moving, the visual system compensates for this, & motion is not perceived unless there is relative motion between the object
& the receptive fields
___ rapidly shift fixation between
objects, occurring voluntarily &
involuntarily.
Saccades. lasting 20–40 ms; ~3 per second
We perceive the world in _____, as
visual input is suppressed during eye
movements
snapshots
Intrasaccadic motion streaks
link an object’s pre- & post-saccadic positions, aiding spatiotemporal continuity
Fixations last
50 ms to several seconds; 200–250 ms ave
Time Spent in Saccades
6–12% of an hour (3.6–7.2 min/hour) is
spent in saccadic suppression – meaning
the brain fills in visual gaps.
Similar to film frames: 24 FPS appears
smooth because brief visual gaps are
imperceptible
MOTION is generally more ____ than
other features
salient
At least ___ needed for conscious visual awareness
~100 ms
Saccades and reading
Reading relies on saccades (~9 characters on average), moving the eyes across
text
Fixations (~200–250 ms) allow for word recognition, while saccadic suppression
prevents blur
English readers process more text to right of fixation due to asymmetric
perceptual span
Lab demonstrations (e.g., disappearing text) reveal how vision is restricted to
fixation-based processing
Top-down attention
Actively directs gaze based on motion expectations—tracking moving objects, predicting trajectories, &
prioritizing relevant motion cues
Attention selects where to ____ next &
determines ____ duration
saccade; fixation
Bottom-up attention:
Driven by stimulus salience
Corollary Discharge Signal (CDS)
informs the brain of upcoming eye
movements before they occur,
preventing motion blur.
Saccadic Suppression & CDS process
Sent from Superior Colliculus →
Medial Dorsal Nucleus (MDN) →
Frontal Eye Fields (FEF)
Comparator:
Predicts retinal shifts & stabilizes vision. Prevents the world from appearing to
‘jump’ with each eye movement.
Superior Colliculus
Midbrain hub integrating visual, auditory, & motor signals. Initiates saccades.
Sends motor command to eye
muscles & CDS to the comparator
(via MDN → FEF) to stabilize
vision
Combines inputs for rapid,
reflexive orienting.
- Guides attention shifts
CDS is linked to ____
Receptive fields. RFs in
the frontal & parietal lobes adjust
before a saccade
before Saccades. CDS & Receptive Fields.
- Saccade Planning: Neurons shift
receptive fields in preparation. - Saccade Execution: Incoming visual
info is processed mid-saccade. - Frontal Eye Fields (FEF) fine-tune
receptive field updates for
seamless perception
intraparietal sulcus (IPS
Plans & guides eye movements,
allocates spatial attention
frontal eye fields (FEF)
Initiates voluntary saccades &
refines eye control
explaining Lilac Chaser Illusion
Phi Phenomenon: Apparent motion illusion. The brain fills in the gap left by disappearing lilac dot, creating a perceived moving object rather than empty space.
- Retinal Neuronal Fatigue: Staring at the lilac dots fatigues S (blue) & L (red) cones, making them less responsive.
- Negative Afterimage (Color Opponent
Process): Fatigued S & L cones leave M cones (green) responsive – creating a green afterimage in open space - Motion-Induced Blindness (MIB):
Stationary lilac dots in the periphery
fade because the brain prioritizes
movement. - Isoluminance: Lilac dots have similar
brightness (isoluminant—low luminance
contrast) & are stationary, making them
prone to MIB due to the motion system’s
insensitivity to color
Perception emerges from interaction
between motion & color, creating an
illusion of fading & movement
FEF + IPS interact to
suppress distractions & maintain gaze stability
_____ plays a minor role
but assists in targeting saccades
Superior Colliculus
____ allow you to see a three-dimensional shape while looking at a two-dimensional image
Stereograms
