COGNITIVE PSYCHOLOGY

Chapter 3
Visual Perception
Sternberg & Sternberg, Cognitive Psychology (Cengage, 2017)

CHAPTER OVERVIEW
This reviewer covers every major topic in Chapter 3 — from the basics of perception to the visual
system's anatomy, theories of perception, Gestalt laws, depth cues, pattern recognition, and
perceptual deficits. Each concept includes a definition, a plain-language explanation, and a
concrete example.
Key sections:
• 1. From Sensation to Perception — Basic Concepts
• 2. How the Visual System Works
• 3. Visual Pathways: What & Where / What & How
• 4. Approaches to Perception: Bottom-Up vs. Top-Down
• 5. Perception of Objects and Forms — Gestalt Laws
• 6. Recognizing Patterns and Faces
• 7. The Environment Helps You See — Constancies & Depth
• 8. Deficits in Perception — Agnosias, Ataxias, Color Blindness
SECTION 1: FROM SENSATION TO PERCEPTION — BASIC
CONCEPTS

PERCEPTION
Definition: The set of processes by which we recognize, organize, and make sense of
sensations received from environmental stimuli.
Simply put: Your brain's job of turning raw sensory signals into something meaningful. Seeing
isn't just about light hitting your eye — your brain actively interprets what all that stimulation
means.
Example: Light hits your retina when you look at your professor's face. That raw light signal is
sensation. Recognizing it as your professor's face is perception.

Gibson's Four Elements of Perception

James Gibson (1966, 1979) gave us a framework to understand how perception works. Every
perceptual experience involves four elements:

Element Definition Example (Vision)

Distal Object The real object out in the world — Your grandma's face across the
the actual thing you're trying to room
perceive. 'Distal' means far away
(from your sense organs).
Informational Medium The carrier of information from the Light waves reflected off
object to your sense organs (light grandma's face
waves, sound waves, molecules,
etc.).
Proximal Stimulation The moment the information Photons absorbed by rod and
physically contacts your sense cone cells in your retina
receptors. 'Proximal' means
near/close.
Perceptual Object What you actually consciously Your conscious experience of
perceive — the mental result of all seeing grandma's face
that processing.

📌 Quick Tip: Sensation vs. Perception vs. Cognition

Sensation = detecting the raw stimulus (Is there a red light?). Perception = identifying and
organizing the stimulus (That's a stop sign!). Cognition = using that information for decisions (I
should brake now). These are on a continuum — not hard stops.

Sensory Adaptation
SENSORY ADAPTATION
Definition: The process by which receptor cells stop responding to a constant, unchanging
 stimulus. If stimulation doesn't change, the receptors stop firing until something new happens.
Simply put: Your senses get bored. If nothing changes, they stop reporting it — which is why
you stop 'feeling' your clothes after a few minutes.
Example: Walking into a room that smells like smoke: at first the smell hits you hard, but after a
few minutes you barely notice it. Your olfactory receptors have adapted.

GANZFELD
Definition: A uniform, unchanging visual field (e.g., a completely solid red surface with no
variation). When exposed to a Ganzfeld, perception fades and the visual field appears gray due
to sensory adaptation.
Simply put: If your entire visual field is one unbroken color with no edges or details, your visual
receptors adapt and perception shuts down — you see gray instead of the original color.
Example: Covering both eyes with half-ping-pong-balls painted red and staring at a light source
— after a minute or two, the red fades to gray.

PERCEPT
Definition: A mental representation of a stimulus that has been perceived. It is the internal
image your mind forms of something you've sensed.
Simply put: A percept is your brain's 'picture' of what it just sensed. Without a percept, you
might sense something but have no idea what it is.
Example: Looking at Dallenbach's Cow image: before you recognize it, you sense the shapes.
The moment you mentally organize it into a cow, you've formed a percept of a cow.

Optical Illusions — Why They Matter

Optical illusions demonstrate that what we sense is NOT necessarily what we perceive. Three key
observations:
• Sometimes we perceive what is NOT there (e.g., the illusory triangle in the Kanizsa figure)
• Sometimes we do NOT perceive what IS there (e.g., Dallenbach's cow hidden in shading)
• Sometimes we perceive what CANNOT be there (e.g., impossible figures like the Penrose
triangle)

📌 Why This Matters for Exams

The existence of optical illusions is direct evidence that the brain actively constructs perception
— not just passively records it. This supports top-down theories of perception.
SECTION 2: HOW THE VISUAL SYSTEM WORKS
The precondition for vision is light. Humans can only perceive wavelengths from 380–750 nanometers
(the visible spectrum). Here's what happens when light enters your eye:

Anatomy of the Eye — The Path of Light

Structure What It Is What It Does
Cornea Clear dome covering the front of Protects the eye; first thing light
the eye passes through
Pupil The opening in the center of the Controls how much light enters —
iris (the colored part) dilates in dark, constricts in light
Iris The colored ring around the pupil Muscle that adjusts pupil size
Crystalline Lens Transparent, flexible structure Focuses light onto the retina;
behind the pupil changes shape to focus near/far
objects
Vitreous Humor Gel-like substance filling most of Maintains the eye's shape; light
the eyeball passes through it
Retina Light-sensitive layer at the back of Where light is converted
the eye (transduced) into neural signals
Fovea Tiny pit in the center of the retina, The zone of sharpest, most acute
size of a pin head vision — where we look directly
Optic Nerve Bundle of ganglion cell axons Sends visual signals to the brain
exiting the eye
Blind Spot Where the optic nerve exits — no Creates a gap in vision that the
photoreceptors here brain fills in automatically

📌 Simply put — the path of light

Light → Cornea → Pupil → Lens → Vitreous Humor → Retina (Fovea) → Photoreceptors fire →
Signals travel via optic nerve → Brain processes it as vision.

Photoreceptors: Rods vs. Cones

Within the retina are two types of photoreceptors — cells that convert light into neural signals. Each eye
has ~120 million rods and ~8 million cones.

RODS CONES
Shape: Long and thin Shape: Short and thick
Location: Concentrated in the periphery of the Location: Concentrated in the fovea (center)
retina (edges)
 Function: Night vision, detecting light/dark, Function: Color vision, fine detail, daylight
movement vision
~120 million per eye ~8 million per eye
Simply put: Your night-vision and side-vision Simply put: Your color and detail detectors
detectors
Example: You can see something moving in Example: Reading text or distinguishing a red
your peripheral vision in a dark room using rods apple from a green one uses cones

From Photoreceptors to the Brain

The signal generated by rods and cones doesn't go directly to the brain. It travels through a chain:
• Rods/Cones → Bipolar Cells → Ganglion Cells
• Ganglion cell axons bundle together → form the Optic Nerve
• The two optic nerves meet at the Optic Chiasma at the base of the brain
• At the optic chiasma: nasal (inner) half of each retina crosses to the opposite hemisphere;
temporal (outer) half stays same side
• ~90% of signals → Lateral Geniculate Nucleus (LGN) of the Thalamus → Primary Visual Cortex
(V1) in the occipital lobe

📌 Important detail for exams

The lens inverts the image — so the signal sent to your brain is literally upside-down and
backward. Your brain corrects for this automatically without you ever noticing.
SECTION 3: VISUAL PATHWAYS — WHAT, WHERE & HOW
From the primary visual cortex (V1), information splits into two major pathways (called fasciculi or fiber
bundles). These pathways process different aspects of the same visual stimulus simultaneously.

VENTRAL PATHWAY (What Pathway) DORSAL PATHWAY (Where/How

Pathway)
Direction: Goes DOWN toward the temporal Direction: Goes UP toward the parietal lobe
lobe
Also called: The 'What' Pathway Also called: The 'Where' Pathway or 'How'
Pathway
Function: Identifies WHAT an object is — color, Function: Locates WHERE objects are in
shape, identity space; guides movement
Damage result: Can't recognize what they see Damage result: Can't reach/grasp objects
(agnosia) accurately (ataxia)
Simply put: Your object-recognition system Simply put: Your spatial navigation and motor-
guidance system
Example: Recognizing that the object in front of Example: Successfully reaching out and picking
you is a coffee mug up that mug

What-Where Hypothesis vs. What-How Hypothesis

WHAT-WHERE HYPOTHESIS WHAT-HOW HYPOTHESIS
Proposed by: Ungerleider & Mishkin (1982) Proposed by: Goodale & Milner (2004)
The two pathways = 'what is it?' vs. 'where is The two pathways = 'what is it?' vs. 'how do I
it?' interact with it?'
Evidence: Monkeys with temporal lobe lesions Evidence: Spatial info is ALWAYS present in
couldn't identify objects; those with parietal both pathways; what differs is emphasis on
lesions couldn't locate them identification vs. action
Limitation: Doesn't fully account for action- The 'how' pathway controls movements relative
guidance role to identified objects
SECTION 4: APPROACHES TO PERCEPTION — BOTTOM-UP vs.
TOP-DOWN
BOTTOM-UP THEORIES (Data-Driven) TOP-DOWN THEORIES (Knowledge-
Driven)
Perception starts with raw sensory data from Perception starts with prior knowledge,
the eyes expectations, and context
The stimulus drives everything — you perceive Your brain constructs perception using what it
what hits your retina already knows
Also called: stimulus-driven theories Also called: constructive or intelligent
perception
Example: Reading text by analyzing each Example: Recognizing a partly hidden stop sign
letter's features from its shape and color
Key theorists: Gibson, Biederman, Selfridge Key theorists: Gregory, Bruner, von Helmholtz

📌 The Takeaway
Neither extreme is correct alone. The best explanation combines both: sensory data provides the
foundation, and prior knowledge fills in the gaps. A complete theory of perception needs both.

Bottom-Up Theories (4 Main Types)

1. Direct Perception (Gibson)

DIRECT PERCEPTION
Definition: Gibson's theory that all the information needed for perception is directly available in
the sensory input from the environment — no higher cognitive processes needed. Also called
ecological perception.
Simply put: The world provides enough information by itself. You don't need to think hard or
guess — just look. Your senses are biologically tuned to pick up everything you need.
Example: You perceive a rock as close because it looks textured and detailed. You perceive a
distant mountain as far because it looks smooth and hazy. No reasoning needed — the texture
gradient directly tells you the distance.

Key concept within Direct Perception — Texture Gradients: Objects that are close appear to have
coarser, more detailed textures; objects farther away look smoother and finer. Gibson argued we use
these gradients directly as distance cues without needing inference.
Mirror Neurons: Neuroscience supports direct perception in social settings. About 30–100 ms after a
visual stimulus, mirror neurons fire. These neurons activate both when you do an action AND when you
observe someone else do it — allowing almost instantaneous understanding of others' expressions and
emotions without conscious reasoning.

2. Template Theories
 TEMPLATES
Definition: Highly detailed mental models stored in memory for patterns we might recognize.
We recognize something by comparing it against our internal templates until we find an exact
match.
Simply put: Your brain has saved copies of patterns it has seen before. To recognize
something, you flip through those saved copies and find the one that matches exactly.
Example: Bank ATMs use template matching to read the printed numbers on checks.
Fingerprint scanners match your print against stored templates. This only works when the
pattern is always identical.

Limitation of Template Theories: We can recognize the letter 'A' written in dozens of different fonts,
sizes, and handwriting styles — but template theory would require a separate template for EVERY
possible version. This is impractical. Template theories cannot explain how we read 'THE CAT'
correctly even though the H in THE is physically identical to the A in CAT.

3. Feature-Matching Theories

FEATURE-MATCHING THEORIES
Definition: Instead of matching an entire pattern to a stored template, we break patterns down
into basic features (lines, angles, curves) and match those individual features to what's stored in
memory.
Simply put: Rather than saving a whole 'A', your brain saves the components: two diagonal
lines meeting at a point, with a horizontal crossbar. Any configuration with those features gets
recognized as 'A'.
Example: The letter 'R' has one vertical line, two horizontal lines, one oblique line, right angles,
acute angles, and one discontinuous curve. The brain combines those features to identify 'R'.

The Pandemonium Model (Selfridge, 1959)

The most famous feature-matching model, using four types of metaphorical 'demons':

Demon Type Role Simply Put

Image Demons Receive the raw sensory input Like a camera that captures what
(retinal image) comes in
Feature Demons Each detects a specific feature Specialists who each check for
(horizontal lines, curves, angles, one specific thing
etc.) and 'shouts' when their
feature is detected
Cognitive Demons Listen to the feature demons; Consultants who say 'could be D,
shout out possible letters/patterns P, or R!'
that match the features detected
Decision Demon Listens to the cognitive demons; The boss who makes the final call
picks the identity based on which based on most evidence
cognitive demon is 'shouting
loudest' (most matches)
Global vs. Local Features
Global features: The overall shape/form of a pattern. Local features: The small-scale details.
Global Precedence Effect: When local letters are small and tightly spaced, we identify the global letter
faster. Example: a large H made of tiny H's — you see the big H first.
Local Precedence Effect: When local letters are widely spaced, we identify local features faster.
Example: a large H made of widely-spaced tiny S's — you see the small S's first.

Neuroscience & Feature Theories (Hubel & Wiesel)

Research using single-cell recording found that neurons in the visual cortex respond only to specific
features at specific locations on the retina. This creates a hierarchy:
• Simple cells: respond to lines of a specific orientation (e.g., only vertical lines)
• Complex cells: respond to more complex patterns (e.g., moving edges, angles)
• Higher-order cells / 'grandmother cells': respond to complex objects like faces or hands
A disproportionately large section of the visual cortex is devoted to the foveal region — the area of
sharpest vision.

4. Recognition-by-Components (RBC) Theory — Biederman (1987)

RECOGNITION-BY-COMPONENTS (RBC) THEORY

Definition: We recognize 3-D objects by decomposing them into basic geometric shapes called
geons (geometrical ions), then matching that geon arrangement to stored object representations.
Simply put: Everything you see can be broken down into a small set of basic 3D shapes (like
cylinders, bricks, cones). Once your brain identifies those shapes and how they're arranged, it
recognizes the object — just like how a small set of letters can make millions of words.
Example: A cup = cylinder (body) + curved handle (partial torus). A lamp = flat disk (base) +
cylinder (pole) + cone (shade). Even if the cup is partially hidden, you can still infer what it is
from the visible geons.

Key feature of geons: They are VIEWPOINT-INVARIANT — you recognize a cylinder as a cylinder
whether you see it from the front, side, or above. This explains why we can recognize objects from
many different angles.
Limitation: RBC explains recognizing general object categories (a chair vs. a lamp) but NOT individual
instances. Your face and your friend's face use the same geons — RBC can't explain how you tell them
apart. Also can't fully account for context effects.

Top-Down Theories — Constructive/Intelligent Perception

CONSTRUCTIVE PERCEPTION
Definition: The perceiver actively builds (constructs) a perceptual understanding of a stimulus
by combining sensory data with prior knowledge, context, and expectations. Also called
intelligent perception.
Simply put: Your brain is like a detective: it doesn't just record what's in front of it — it takes
incomplete sensory evidence and fills in the gaps using everything it already knows to construct
the most probable interpretation.
Example: Seeing a partially covered stop sign (only 'ST_P' visible): pure sensory data is
ambiguous, but your prior knowledge tells you it's a stop sign. You brake. That's constructive
 perception at work.

Key concept: Unconscious Inference

Unconscious inference: The process by which we automatically and unconsciously combine multiple
sources of information to construct a perception. We make perceptual judgments without being aware
we're making them.
Example: You see something on rail tracks. Without consciously thinking, you perceive it as a train.
Your brain has already combined 'object on rails + large shape + context' to form a perception — all
before conscious awareness.

Context Effects

CONTEXT EFFECTS
Definition: The influence of surrounding environment and prior information on how we perceive
an object.
Simply put: What you see changes depending on what's around it. The same stimulus can be
interpreted differently based on context.
Example: The H in 'THE' and the A in 'CAT' are physically identical marks, but context
(surrounding letters) makes you read them as different letters. In Palmer's (1975) study, a loaf of
bread was recognized faster in a kitchen scene than out of context.

Key Context Effects

Configural-Superiority Effect: Objects in certain groupings/configurations are easier to spot than
those in isolation, even when the configurations are more complex. Example: It is faster to spot the
odd-one-out among triangles vs. among single diagonal lines.

Object-Superiority Effect: A target line within a drawing of a 3-D object is identified more accurately
than the same line in a disconnected 2-D pattern.

Word-Superiority Effect: It is easier to identify a single letter when it appears in a real word than in a
nonsense letter string. Example: The letter 'O' is recognized faster in 'HOUSE' than in 'HUSEO'.
SECTION 5: PERCEPTION OF OBJECTS AND FORMS

Viewer-Centered vs. Object-Centered Representation

VIEWER-CENTERED REPRESENTATION OBJECT-CENTERED REPRESENTATION
Stores how an object looks from YOUR Stores the object's structure independent of
perspective your viewpoint
The stored shape changes depending on your The stored shape remains stable across
angle of view different orientations
To recognize the object from a new angle, you Recognition is based on the object's inherent
mentally rotate it to match a stored image spatial axes, not your position
More strongly supported by current research Less supported — neurons DO respond
differently to different views
Example: You recognize your laptop differently Example: (Theoretical) You'd recognize your
from the front vs. the side, and mentally 'rotate' laptop the same way regardless of angle
to match

A third type — Landmark-Centered Representation: Location is encoded relative to a prominent

landmark. Example: navigating a new city by remembering streets relative to your hotel.

The Gestalt Laws of Perception

Developed in early 20th-century Germany by Koffka, Köhler, and Wertheimer, the Gestalt approach
explains how we organize visual elements into wholes. The core principle: the whole is different from
the sum of its parts.

LAW OF PRÄGNANZ (Overarching Gestalt Law)

Definition: We tend to perceive visual arrays in the simplest, most stable, and most coherent
way possible. All other Gestalt principles flow from this law.
Simply put: Your brain is lazy in the best way — it always takes the simplest possible
interpretation of what it sees. It automatically organizes chaos into order.
Example: When you look at a circle made of dots, you see 'a circle' not 'a collection of individual
dots.' Simplest interpretation wins.

Gestalt Principle Definition (Parsimonious) Example

Figure-Ground We automatically separate visual Rubin's Vase: you see either two
scenes into a prominent figure faces OR a vase — never both
(the thing we focus on) and a simultaneously. Whichever you
receding background. Simply: focus on becomes the 'figure'; the
There's always something that other becomes 'ground.'
'pops out' and something that
'falls back.'
Proximity Objects that are physically close Three pairs of dots are seen as
 to each other are grouped three columns, not six individual
together. Simply: Close = same dots, because each pair is close
group. together.
Similarity Objects that look alike are A grid of dots alternating filled and
grouped together. Simply: Alike = empty is seen as rows (grouped
same group. by color similarity), not columns.
Continuity We tend to perceive smooth, An S-shape partially covered by a
continuous lines rather than rectangle is still perceived as a
broken or abrupt ones. Simply: complete 'S' because continuity
Your brain connects the dots to tells you the line continues behind
make the smoothest path. the bar.
Closure We mentally 'close' or complete A circle drawn with gaps is still
objects that are not fully drawn. perceived as a circle. A series of
Simply: Your brain fills in the gaps disjointed lines may be seen as a
to see a complete shape. cube.
Symmetry We perceive objects as forming The symbol '[ ]' is perceived as
mirror-image pairs around a one matching pair of brackets, not
center. Simply: Your brain prefers two separate marks.
balanced, symmetric
interpretations.

📌 Gestalt Laws and Species

Gestalt principles appear to be uniquely human. In one experiment, baboons did NOT fall for the
Ebbinghaus illusion (where surrounding circles distort perceived size of a center circle), while
humans consistently did. This suggests humans uniquely attend to the surrounding visual
context.
SECTION 6: RECOGNIZING PATTERNS AND FACES

Two Pattern Recognition Systems (Martha Farah)

FEATURE ANALYSIS SYSTEM (System 1) CONFIGURATIONAL SYSTEM (System 2)
Specializes in analyzing individual PARTS and Specializes in recognizing overall
assembling them into wholes CONFIGURATIONS — the whole gestalt of a
pattern
Used for detailed, part-by-part analysis Used for holistic, 'at a glance' recognition
Example: In biology class, examining each part Example: Admiring the overall beauty and form
of a tulip — stamen, pistil, petals of a tulip in a garden
Used in face recognition when someone looks Dominantly used for recognizing familiar faces
vaguely familiar — you analyze features one by — you recognize your friend's face as a whole,
one to identify them not piece by piece
Damage: Affects reading disability (trouble with Damage: Causes prosopagnosia (inability to
letter recognition) recognize faces)

Neuroscience of Face Recognition

FUSIFORM GYRUS
Definition: A region in the temporal lobe that is especially active when recognizing faces. It
responds more intensely to faces than to other objects.
Simply put: Think of the fusiform gyrus as your brain's 'face recognition module.' It lights up
specifically when you're processing faces.
Example: Brain scans show the fusiform gyrus activates strongly when you see human faces,
but barely responds when you look at objects like chairs or chairs.

Expert-Individuation Hypothesis: An alternative explanation of fusiform gyrus activity — it activates

not exclusively for faces, but for any category you have VISUAL EXPERTISE in. Bird experts' fusiform
gyri activate when identifying individual birds; car experts' activate for cars. We are all, in a sense, face
'experts' from years of experience.

PROSOPAGNOSIA
Definition: A neurological disorder characterized by a severely impaired ability to recognize
faces — even one's own face in a mirror. Caused by damage to the configurational system,
specifically the right fusiform gyrus / right temporal lobe.
Simply put: The person's eyes work fine and they can see a face clearly — they just can't
attach an identity to it. They know it's a face, but they don't know WHOSE face it is.
Example: A person with prosopagnosia might be able to describe a colleague's facial features in
detail but still fail to recognize them. They may not recognize their own mother by face alone,
relying instead on voice or clothing.

📌 Schizophrenia and Face Recognition

People with schizophrenia often struggle to recognize emotions in faces. Research shows they
scan faces differently — looking at fewer salient features and making fewer long fixations —
rather than the typical inverted-triangle scan pattern (eyes → nose → mouth) used by
neurotypical individuals.
SECTION 7: THE ENVIRONMENT HELPS YOU SEE

Perceptual Constancies
PERCEPTUAL CONSTANCY
Definition: The tendency to perceive an object as remaining stable and unchanging even
though the proximal stimulus (retinal image) of the object changes with distance, angle, or
lighting.
Simply put: Your brain compensates for changes in the retinal image so you perceive a stable
world. The object hasn't changed — your viewing conditions have — but you still see it as the
same.
Example: As a friend walks toward you, their image on your retina gets larger and larger. But
you do NOT perceive your friend as growing. You perceive them as the same size, just getting
closer. That's size constancy.

SIZE CONSTANCY SHAPE CONSTANCY

Definition: Perceiving an object as the same Definition: Perceiving an object as the same
size despite changes in the size of its retinal shape despite changes in its orientation or the
image shape of its retinal image
Simply: You don't think people shrink as they Simply: A door doesn't become a trapezoid just
walk away from you because you opened it
Example: A car driving away doesn't seem to Example: A circular plate looks oval when
shrink — you perceive it as a normal-sized car viewed from an angle, but you perceive it as
getting farther away circular
Illusion: Müller-Lyer — two equal lines look Brain region: Extrastriate cortex handles shape
different in length because of arrow cues at analysis
their ends
Note: Not fully innate — must be partially Easier to achieve when the object is
learned. Judging size accurately develops symmetrical (symmetry provides orientation
through childhood cues)

📌 Müller-Lyer Illusion
Two lines of identical length appear different when one has inward-pointing arrowheads and the
other has outward-pointing arrowheads. This illusion shows that our size constancy mechanisms
use contextual cues (angles that look like corners of buildings) to estimate size — which can
mislead us in 2-D images. Brain regions activated: right posterior parietal cortex and right
temporo-occipital cortex.

Depth Perception
DEPTH PERCEPTION
Definition: The ability to perceive 3-D space — to judge how far objects are from us and from
each other — even though the retinal image is only 2-D.
Simply put: Your retina is flat, like a camera sensor. But you see the world in 3D. You
 accomplish this by using depth cues — environmental clues that signal distance.
Example: Reaching for a cup of tea, driving a car, catching a ball — all require accurate depth
perception. Without depth cues, you'd misjudge distances constantly.

Monocular Depth Cues (work with just ONE eye)

These cues can be represented in a flat 2D image and detected with just one eye. Artists use them to
create the illusion of depth in paintings.

Monocular Cue How It Signals Depth Example

Texture Gradients Objects nearby appear to have A brick road: bricks near you look
coarse, detailed textures; farther large and detailed; bricks far away
objects look smoother and finer. look tiny and uniform.
Relative Size Bigger objects appear closer; Two people in a photo: the one
smaller objects appear farther. appearing larger is perceived as
closer.
Interposition (Overlap) When one object partially blocks A tree partially blocking a building:
another, the blocking object is the tree is in front of the building.
perceived as closer.
Linear Perspective Parallel lines appear to converge Railroad tracks appear to meet on
toward a vanishing point as they the horizon, even though they're
recede into distance. always the same distance apart.
Aerial Perspective Distant objects appear fuzzier and Mountains far away look blurry
less sharply defined due to and bluish compared to nearby
atmospheric haze. trees.
Location in Picture Below the horizon: closer objects In a seascape, ships close to the
Plane are lower in the picture; above horizon appear higher in the
horizon: closer objects are higher. frame.
Motion Parallax As you move, closer objects Looking out a car window: nearby
appear to move faster across your telephone poles whiz by; distant
visual field than distant objects. mountains barely move.
Requires real movement —
cannot be in a still image.

Binocular Depth Cues (require BOTH eyes)

BINOCULAR DISPARITY BINOCULAR CONVERGENCE
Each eye receives a slightly different view of As objects get closer, both eyes must turn
the same object inward (converge) to focus on them
The closer the object, the MORE different the The closer the object, the MORE your eye
two retinal images are muscles have to work to converge
Your brain interprets the DEGREE of disparity Your brain interprets the DEGREE of muscle
as a distance cue strain as a distance cue
Simply: Close objects look very different to Simply: Close objects make your eyes cross
each eye; distant objects look nearly identical more; distant objects let your eyes relax
to each eye outward
Example: Hold your finger close to your nose Example: Looking at your nose vs. a distant
— cover one eye, then the other. The finger wall feels different in terms of eye strain —
'jumps.' That's disparity. that's convergence.

📌 Interesting finding on depth perception

Perceived distance isn't just about actual distance — it's influenced by EFFORT. People wearing
a heavy backpack perceive distances as farther than those not wearing one. The more effort
required to reach something, the farther away it seems (Proffitt et al., 2003). This is a top-down
effect on a seemingly bottom-up process.
SECTION 8: DEFICITS IN PERCEPTION
Studying people with perceptual deficits tells us a great deal about how normal perception works.
Deficits often confirm the existence of separate processing pathways.

Agnosias — Difficulties with the WHAT Pathway

AGNOSIA
Definition: A condition in which a person has trouble perceiving sensory information despite
having normally functioning sense organs. Most agnosias are caused by damage to the border
of the temporal and occipital lobes.
Simply put: The eyes work fine. The signals reach the brain fine. But the brain can't figure out
what the object IS. It's like having a camera that works but no software to decode the photos.
Example: A patient who can describe a pair of eyeglasses as 'two circles connected by a
crossbar' but guesses it's a bicycle — the visual input is received but meaning cannot be
extracted.

Type of Agnosia What Is Impaired Example

Visual-Object Agnosia Cannot recognize what objects Sees a key but can only describe
are, even though they can see 'a small, thin silver object with
them clearly. The what pathway is teeth.' Cannot identify it as a key.
damaged.
Simultagnosia Cannot attend to more than ONE If shown overlapping drawings of
object at a time. Temporal region a hammer, cup, and scissors,
disturbance. they can only see the hammer at
any given moment.
Prosopagnosia Cannot recognize faces — Cannot identify a close friend or
including their own face in a family member by face alone.
mirror. Right fusiform gyrus / right May rely on voice or clothing to
temporal lobe damage. recognize people. May struggle to
follow movies because characters
are unrecognizable.

Ataxia — Difficulties with the HOW Pathway

OPTIC ATAXIA
Definition: An impaired ability to use visual information to guide physical movements. Caused
by processing failure in the posterior parietal cortex (how pathway).
Simply put: The person can SEE the object fine — they know what it is and where it is — but
they cannot successfully reach for or grasp it. The disconnect is between vision and action.
Example: Imagine trying to pick up your phone but your hand consistently misses it or grasps it
at the wrong angle, even though you can see it clearly right in front of you. That's optic ataxia.

📌 Agnosia vs. Ataxia — Key Comparison

 Agnosia = CAN'T recognize WHAT the object is (ventral/what pathway damaged). Ataxia =
CAN'T use vision to guide HOW to reach/grasp it (dorsal/how pathway damaged). A person
could theoretically have one without the other — supporting the two-pathway model.

📌 Are Perceptual Processes Independent? — Modular Processes

The specificity of perceptual deficits (e.g., some people can't name colors but can recognize
faces; others can see a mug but can't grasp it) suggests MODULAR PROCESSES — separate,
specialized processing centers for different perceptual tasks. For a process to be truly modular, it
must be domain-specific and information must NOT freely flow between modules.

Anomalies in Color Perception

Color perception deficits are far more common in males than females (genetic link on X chromosome)
but can also result from brain lesions in the ventromedial occipital and temporal lobes.

Type What's Impaired Simply Put

Rod Monochromacy Complete absence of color vision The world looks like a black-and-
(Achromacy) — the rarest and only 'true' color white film. No color at all.
blindness. Cones are
nonfunctional; only rods work.
Dichromacy Only TWO of the three color Partial color blindness — some
mechanisms work (one colors are confused.
malfunctions). Three subtypes:
Protanopia Red-green color blindness Red and green look very similar
(extreme form) — can't or identical.
distinguish red from green
Deuteranopia Trouble seeing greens — similar Green range of colors is confused
symptoms to protanopia or missing.
Tritanopia Confusion of blues/greens; Blue-yellow range is disrupted;
yellows disappear or appear rare condition.
reddish
SECTION 9: WHY DOES IT MATTER? PERCEPTION IN PRACTICE

Perception and Traffic Safety

About 50% of all collision accidents result from missing or delayed perception. Two-wheeled vehicles
(motorcycles, bicycles) are disproportionately involved in 'looked-but-failed-to-see' accidents — where
drivers looked in the right direction but didn't consciously register the two-wheeled vehicle.

Why? Drivers develop a 'scanning strategy' that focuses on the most common threats. Unusual or
small objects fall outside this strategy. People also tend to fail to notice new objects after blinking or
rapid eye movements (saccades).

CHANGE BLINDNESS BLINDNESS

Definition: People's unawareness of their own susceptibility to change blindness. Most people
believe they would notice all changes in their visual field — but research consistently shows they
don't.
Simply put: We overestimate how much we actually see. We think we're taking in everything
around us, but we miss huge changes when our attention isn't focused on them.
Example: In famous demonstrations, people fail to notice an actor being replaced by a different
person mid-conversation (while attention is briefly diverted). Most people watching the video for
the first time are shocked they missed it.

Perception in Photography & Design

Understanding depth cues allows practical application:
• A long nose appears shorter when photographed from slightly below the facial midline (depth
recession)
• Leaning forward makes the upper body look larger; leaning back makes it look smaller
• In group photos, standing slightly in front of others makes you appear larger; standing behind
makes you smaller
• Fashion designers use depth cues (color blocking, stripes, cuts) to visually alter perceived body
proportions

Applications to Machine Vision

Research on human perception directly informs AI and computer vision:
• Postal services use machines to read ZIP codes — they can't use pure template matching
because people write numbers differently, so feature analysis is needed
• CAPTCHAs (Completely Automated Public Turing Test to Tell Computers and Humans Apart)
exploit the fact that computers rely heavily on template matching — making distorted text easy
for humans but hard for machines
• Computers still lag behind humans at recognizing ambiguous real-world images (reflections,
shadows, occlusions)
SECTION 10: MASTER KEY TERMS LIST
Quick reference for all key terms from Chapter 3:

Term Short Definition Page Ref

Agnosia Impaired ability to recognize p.111
sensory information despite
normal sensory function
Binocular depth cues Depth cues requiring information p.108
from both eyes (disparity +
convergence)
Bipolar cells Neurons connecting p.80
photoreceptors to ganglion cells in
the retina
Bottom-up theories Stimulus-driven theories where p.81
perception starts with raw sensory
input
Cones Photoreceptors for color and p.80
detail vision; concentrated in the
fovea
Constructive perception Top-down view: perceivers BUILD p.91
perception using sensory data +
prior knowledge
Context effects Influence of surrounding p.92
environment/information on object
perception
Depth Distance from a surface; judged p.106
using monocular and binocular
cues
Direct perception Gibson's view: all needed p.82
perceptual info is in the sensory
environment itself
Feature-matching Recognition by matching p.85
theories individual features (lines, curves)
against stored features
Figure-ground Gestalt separation of visual scene p.97
into a prominent figure and
receding background
Fovea Small central region of retina with p.78
sharpest vision; cones
concentrated here
Ganglion cells Retinal neurons whose axons p.80
form the optic nerve
Gestalt approach Perception theory emphasizing p.97
whole forms over individual parts;
law of Prägnanz
Landmark-centered Representing locations relative to p.97
a well-known reference point
Law of Prägnanz Overarching Gestalt law: we p.97
perceive the simplest, most stable
interpretation
Monocular depth cues Depth cues observable with one p.107
eye (texture gradients, size,
overlap, etc.)
Object-centered Storing an object's shape p.96
representation independent of the viewer's angle
Optic ataxia Impaired ability to use vision to p.111
guide hand/reaching movements
Optic nerve Bundle of ganglion cell axons p.80
carrying visual signals from eye to
brain
Percept A mental representation of a p.77
stimulus that has been perceived
Perception Processes by which we p.72
recognize, organize, and make
sense of sensations
Perceptual constancy Perceiving an object as stable p.104
despite changes in its retinal
image
Photo pigments Chemical substances in p.79
rods/cones that react to light and
trigger neural signals
Photoreceptors Rods and cones; convert light p.79
energy into electrochemical
neural impulses
Recognition-by- We recognize objects by p.90
components (RBC) decomposing them into basic 3D
shapes called geons
Retina Light-sensitive inner layer of the p.78
eye where photoreceptors are
located
Rods Photoreceptors for night vision / p.79
light-dark detection; concentrated
in periphery
Templates Detailed mental models against p.84
which perceived patterns are
matched exactly
Top-down theories Knowledge-driven theories: p.81
 existing knowledge/expectations
shape perception
Viewer-centered Storing an object as it appears p.96
representation from YOUR specific viewpoint
SECTION 11: CONCEPT CHECK QUESTIONS & ANSWERS

Concept Check 1 (pp. 81)

Q1: What is the difference between sensation and perception?
Sensation = the raw detection of stimuli by sense receptors (noticing qualities like
brightness, loudness). Perception = the brain's interpretation and organization of those
sensations into meaningful objects/events (recognizing what that brightness is — a
flashlight, a star, a headlight). They exist on a continuum, not as hard stops.
Q2: What is the difference between the distal and the perceptual object?
The distal object is the actual real-world object (e.g., a tree). The perceptual object is your
mental representation of it — what you consciously experience after your brain processes
the sensory information.
Q3: How are rods and cones similar and different?
Both are photoreceptors that contain photo pigments and convert light into neural signals.
Rods: thin, in periphery, for night/low-light vision, black-and-white only (~120 million).
Cones: thick, concentrated in fovea, for color and detail in bright light (~8 million).
Q4: Major parts of the eye and their functions?
Cornea (protects, first stop for light) → Pupil (opening that controls light entry) → Lens
(focuses light) → Retina (where light is converted to neural signals) → Fovea (sharpest
vision zone) → Optic nerve (sends signals to brain).
Q5: What is the what-where hypothesis?
The proposal that the brain has two distinct visual pathways: the ventral (what) pathway in
the temporal lobe identifies WHAT objects are, and the dorsal (where) pathway in the
parietal lobe determines WHERE objects are located in space.

Concept Check 2 (pp. 104)

Q1: What are the major Gestalt principles?
Law of Prägnanz (overarching), Figure-ground, Proximity, Similarity, Continuity, Closure,
Symmetry.
Q2: What is recognition-by-components theory?
Biederman's theory that we recognize 3-D objects by breaking them down into simple
geometric components called geons. A small set of geons can combine to create countless
objects, just as letters combine to form words.
Q3: Difference between top-down and bottom-up theories?
Bottom-up: perception starts with and is driven by raw sensory data. Top-down: perception
is driven by prior knowledge, expectations, and cognitive processes that work 'downward' to
interpret incoming sensory data. Both are needed for a complete explanation.
Q4: Difference between viewer-centered and object-centered perception?
Viewer-centered: mental representation stores how the object looks from YOUR perspective
— varies with viewpoint. Object-centered: mental representation stores the object's structure
independently — remains stable across viewpoints. Current evidence favors viewer-
centered.
 Q5: What is prosopagnosia?
A neurological condition characterized by an impaired ability to recognize faces — even
one's own. Caused by damage to the configurational system, particularly the right fusiform
gyrus. The person can see faces clearly and may even describe emotions, but cannot attach
an identity to the face.

Concept Check 3 (pp. 113)

Q1: What is shape constancy?
The perceptual tendency to perceive an object as maintaining the same shape despite
changes in its orientation (and thus changes in its retinal image). A door looks rectangular
whether it's closed or open, even though the retinal image changes from a rectangle to a
trapezoid.
Q2: Main cues for depth perception?
Monocular: texture gradients, relative size, interposition, linear perspective, aerial
perspective, location in picture plane, motion parallax. Binocular: binocular disparity
(difference in retinal images) and binocular convergence (degree of inward eye-turning).
Q3: What is visual agnosia?
A condition in which a person can see objects normally but cannot recognize what they are.
The what pathway (ventral stream) is damaged. They can see all parts of the visual field but
cannot extract meaning from what they see.
Q4: Difference between monochromacy and dichromacy?
Monochromacy (rod monochromacy/achromacy): complete absence of color vision — cones
are nonfunctional, only rods work, everything appears as shades of gray. Dichromacy: only
TWO of the three color mechanisms work (one malfunctioning), resulting in partial color
confusion (e.g., red-green color blindness).
SECTION 12: CHAPTER 3 AT A GLANCE — SUMMARY TABLE
Major Topic Core Concept Key Takeaway
Perception Set of processes for recognizing Perception is active and
and organizing sensations constructive — not passive
recording
Gibson's Framework Distal object → Informational Every perceptual act involves a
Medium → Proximal Stimulation real object, a carrier, a receptor
→ Perceptual Object event, and a mental result
Sensory Adaptation Receptors stop firing to constant Variation is essential for
stimuli perception — without change, we
stop noticing
Ganzfeld Uniform visual field causes Demonstrates that constant
perception to fade to gray stimulation leads to adaptation
and perceptual disappearance
Visual Pathway Cornea → Pupil → Lens → Light is transduced in the retina;
Retina → Optic Nerve → LGN → brain adds meaning
V1 → Ventral/Dorsal streams
Rods vs. Cones Rods: night/periphery; Cones: Two complementary systems for
color/detail/fovea different lighting and task
demands
What vs. Where/How Ventral = identity; Dorsal = Two dissociable pathways
Pathways location/action confirmed by specific perceptual
deficits
Bottom-Up Theories Direct Perception, Templates, Perception explained by analyzing
Feature-Matching, RBC incoming sensory data upward
Top-Down Theories Constructive/Intelligent Prior knowledge, expectations,
Perception; context effects and intelligence shape what we
see
Gestalt Laws Prägnanz, Figure-Ground, The brain organizes stimuli into
Proximity, Similarity, Continuity, the simplest coherent whole
Closure, Symmetry automatically
Pattern Recognition Feature Analysis System vs. We use two systems — parts
Configurational System analysis and holistic configuration
— for different tasks
Face Recognition Fusiform gyrus; Expert- Faces have a special processing
Individuation Hypothesis; advantage; damage causes
Configurational system prosopagnosia
Perceptual Constancy Size constancy; Shape constancy Brain compensates for retinal
changes to give stable perception
Depth Cues Monocular (7 types) and 3D perception from a 2D retinal
Binocular (2 types) image using environmental and
binocular cues
Agnosias Visual-object, Simultagnosia, What-pathway damage; can see
Prosopagnosia but can't recognize
Ataxia Optic ataxia — can't use vision to How-pathway damage; can see
guide movement but can't act on what they see
Color Vision Deficits Monochromacy (no color) vs. More common in males; genetic
Dichromacy (partial loss) or lesion-based; varies in severity

Good luck on your exam! 🎓

Remember: perception is not passive. Your brain actively constructs every experience. You
are not just recording reality — you are building it.