COGNITIVE PSYCHOLOGY

CHAPTER 3
VISUAL PERCEPTION

PERCEPTION – process that we recognize, organize and ASSIGNMENT:

make sense we receive from our environment
Watch how the participants react while experiencing the
James Gibson Ganzfeld effect.

- provided a framework for studying FORMAT:

perception
- introduced the concepts of distal (external) A4, NORMAL, NORMAL, ARIAL 12
object, informational medium, proximal REMOVE PARAGRAPH, SPACING,TABING,JUSTIFIED
stimulation, and perceptual object.
- DISTAL – (far) object in the external world. QUIZ:
- PROXIMAL (near) stimulation.
From localization of functions up to here
SENSORY ADAPTATION – already adapted to the scent

- Stop detecting the presence of the stimulus

HOW THE VISUAL SYSTEM WORK?
GANZFELD - “complete field”
1. Light passes through the eye
- Blocking the sense of sight and sound 2. Cornea – clear dome that protects the eye
- They started hallucinating 3. Pupil – opening center of the iris
4. Vitreous humor – gel-like that makes up the
PERCEPTUAL ILLUSION eyes
- what we sense (in our sensory organs) is not 5. Retina – contains photoreceptors (rods&cones)
necessarily what we perceive (in our minds). 6. Fovea – small region of retina
7. Optic nerve –
Different wavelengths in the spectrum of visible light 8. Optic chiasm
9. Lateral geniculate
10. PVC – Primary visual cortex

SENSATION – stimulation

PERCEPTION –identify and form pattern, movement.

COGNITION – used to determine further goals.

SENSORY ADAPTATION – adapt constant stimulation.

PERCEPT – stimulus is perceived

- MERNETHE
Perception Definition:

- The process of recognizing, organizing, and Distal (Far) Object - This is the object in the real world,
interpreting sensory information from the like a falling tree.
environment.
Informational Medium - This is the way information
Focus on Visual Perception: about the object is transmitted, like sound waves, light,
or other environmental signals.
- Visual perception is the most studied and
widely recognized type of perception. Proximal Stimulation - This happens when the
information (like light) reaches your sensory receptors,
such as the cells in your retina absorbing light waves.
Example of Perception Perceptual Object - This is the mental image or
- Distant objects appear smaller due to the perception of the object, like seeing the falling tree in
effects of distance, illustrating how your mind based on the information received.
perception differs from reality. Perception Varies by Sense- The way we perceive things
Optical Illusions changes depending on which sense (sight, sound, smell,
etc.) we use.
- Used to demonstrate the complexities and
quirks of human perception. The Tree Falling Example - If no one is around to hear a
tree fall, it still makes sound waves (yes), but there is no
Biology of Visual System perceived sound without someone to hear it (no).

- Sensory receptors in the retina send signals Perception - We identify objects and patterns (e.g., Is
to the brain, which organizes and interprets that an apple?)
these sensations into a coherent image.
Cognition - decide what to do next (e.g., Is it edible?
Complexity of Perception Should I leave the forest?).

- Perception involves intricate processes and Variation in Perception - Every sensory experience (like
can be influenced by various factors, leading seeing an apple or hearing a voice) is slightly different
to phenomena like missing something right each time.
in front of you or mishearing lyrics.
Perceptual Stability - Despite sensory changes, we
Perceptual Phenomena maintain a stable perception of the world.

- Perception helps explain how we interpret Sensory Adaptation - Our sensory receptors stop
and make sense of the world around us. responding to constant stimuli, so we don’t notice
things like background noise or a steady visual field.

Stabilized Images - Scientists use stabilized images that

BASIC CONCEPTS OF PERCEPTION:
don’t move on the retina to study perception. These
James Gibson's Framework images disappear after constant exposure.

- Gibson introduced concepts to help study Ganzfeld Effect - When exposed to a uniform visual
perception, including distal (external) field (like a plain red surface), we stop perceiving it after
objects, informational media, proximal a while and see just a gray field.
stimulation, and perceptual objects.
HOW THE VISUAL SYSTEM WORK Photopigments - Both rods and cones contain
photopigments,chemicals that react to light and convert
RETINA > RODS AND CONES > HORIZONTAL CELLS > it into neural impulses for the brain.
AMACRINE CELLS > BIPOLAR CELLS > GANGLION CELLS
> FOVEA > OPTIC NERVE > OPTIC CHIASM > LATERAL Neurochemical Messages - Rods and cones send
GENICULATE > PVC neurochemical messages to bipolar cells, which then
transmit them to ganglion cells.
Vision - requires light, which is electromagnetic
radiation. Humans can perceive a small range of
wavelengths (380 to 750 nanometers).
Optic Nerve - The axons of the ganglion cells form the
optic nerve, which carries visual information to the
brain.
Cornea - The cornea is the clear, protective dome
covering the eye through which light first passes.

Optic Chiasma - At the optic chiasma, the ganglion cells

from the nasal (inner) part of the retina cross to the
Pupil and Iris - Light then passes through the pupil, the opposite hemisphere, while those from the temporal
opening in the center of the iris. (outer) part stay on the same side.

Inverted Image - The lens in each eye inverts the image,

Crystalline Lens and Vitreous Humor - Light continues so the message sent to the brain is upside-down and
through the crystalline lens and the vitreous humor, a backward.
gel-like substance that makes up most of the eye. Thalamus and Visual Cortex - About 90% of the
Retina and Transduction - Light focuses on the retina, ganglion cells' signals go to the lateral geniculate
where it is converted into neural electrochemical nucleus of the thalamus, then to the primary visual
impulses. cortex (V1) in the occipital lobe.

Fovea - Vision is sharpest in the fovea, a small, thin area Visual Processing Areas - The visual cortex has several
on the retina the size of a pinhead. areas that process different aspects of visual
information, such as color, location, depth, pattern, and
Photoreceptors - (rods and cones) that convert light into form.
electrochemical energy for the brain.
Bottom-Up Theories - Perception begins with external
Rods and Cones stimuli, such as light, entering the eye.

Rods - About 120 million in each eye, long and thin,

concentrated in the periphery of the retina. They are
responsible for night vision and light/dark sensitivity. Stimulus-Driven - These theories are data-driven,
meaning perception is based on the raw sensory
Cones - About 8 million in each eye, short and thick, information received from the environment.
concentrated in the fovea.

- responsible for color perception.

Example: Looking at a cityscape, the light information is
processed by the brain to form perception.
Top-Down Theories - Perception is driven by high-level 2. TEMPLATE THEORIES – Suggest that our minds store
cognitive processes, prior knowledge, and expectations. many templates (detailed models) to recognize patterns
by comparing them to these templates.
Cognitive Processes - These theories suggest that
expectations shape our perception before sensory data
is processed.
Pattern Recognition - We recognize patterns by
matching them to the exact template that fits what we
observe (e.g., fingerprint matching, UPC scanning).
BOTTOM – UP THEORIES:

1.Direct Perception - Recognizing objects, involves

perceiving its distinct features. Chunk-Based Theories: Expertise is developed by
acquiring chunks of knowledge in long-term memory,
Gibson’s Theory of Direct Perception - According to which can be accessed for fast recognition (e.g., chess
Gibson, all the information needed for perception is players recalling previous games).
provided by our sensory receptors and the environment,
without the need for higher cognitive processes.

Limitations of Template Matching:

Also called Ecological Perception - This view suggests

that our environment provides all the necessary cues for
perception, meaning we don’t need to rely on beliefs or - Requires exact matches, which works for
inferential thinking to interpret sensory data. machines (e.g., numeral recognition in
banks), but is impractical for human
Perceptual judgments - made using contextual perception.
information from the environment, which we are - Human perception can adapt to variations
biologically tuned to respond to. in size, orientation, and form, unlike strict
template matching.
Texture Gradients - We use texture gradients
(variations in detail based on distance) to perceive Challenges for Template Matching:
depth and distance.
-It’s impractical to have templates for every possible
Depth Cues - When objects cover parts of other objects, variation of an object, such as a face or a letter.
we infer that the covered objects are farther away.
-Template theories struggle to explain how we can
Neuroscience and Direct Perception - Direct perception recognize letters (like A and H) from just one physical
is involved in person perception, with mirror neurons form or handle variations in size and orientation.
firing 30-100 milliseconds after a visual stimulus.
- Storing and retrieving a vast number of templates for
every possible variation would be difficult and
inefficient.
Mirror Neurons - These neurons activate when a person
performs or observes an action, helping us quickly
understand expressions, emotions, and movements.

Separate Neural Pathways - The lateral occipital area

processes form, color, and texture separately in the
brain.
Neuroscience and Template Theories - Letters are Local vs. Global Features:
simpler than faces and other complex stimuli, but the
brain still needs a way to recognize them

Human Handwriting Recognition - Humans easily Local Features -Small, detailed parts of a pattern (e.g.,
understand handwriting, even when letters and small "H"s).
numbers vary in appearance. Global Features - The overall shape or form of the
Automated Check Processing - Machines can easily read pattern (e.g., the big letter "H").
printed numbers on checks but face difficulty when Example: In two examples of the letter H:
trying to decipher handwritten text.

3.FEATURE-MATCHING THEORIES:
In panel (a): Small "H"s make up the large "H" (local
Suggest that pattern recognition occurs by matching features match global ones).
features of a stimulus to features stored in memory,
rather than matching a whole pattern to a template. In panel (b): Small "S"s do not form the letter "H" (local
features don’t match the global ones).
Pandemonium Model - A feature-matching model
where metaphorical "demons" analyze and process Pandemonium model - helps explain how we break
different features of a stimulus. down complex visual information into simple features
for recognition.
Demons' Duties - Each "demon" has a specific role in
receiving and analyzing features of the stimulus to Neuroscience Support for Feature-Matching Theories:
recognize patterns.

The term "pandemonium" reflects the chaotic, noisy

- Research on single-cell recordings with
nature of this process.
animals supports feature-matching theories.
Pandemonium Model - Created by Oliver Selfridge, it - Individual neurons in the visual cortex
explains how we recognize patterns through different respond to specific visual stimuli, like
"demons" (metaphorical agents) working together. horizontal lines, in certain regions of the
retina.

Neuron Mapping:
Four Kinds of Demons:

1.Image Demons - Receive the image from the retina

and pass it on to the next level. - Neurons in the visual cortex are mapped to
receptive fields on the retina.
2.Feature Demons - Identify specific features of the - A larger portion of the visual cortex is
image (e.g., lines, angles) and shout when they find a dedicated to the fovea, the region
match. responsible for sharp, detailed vision.
3.Cognitive Demons -Listen to the feature demons and
shout out possible patterns stored in memory that
match the features.

4.Decision Demon - Decides what has been seen by

choosing the pattern with the most matching features,
based on which cognitive demon is shouting the most.
Neuron Response: Recognition by Components Theory:

- Cells in the cortex do not respond to simple - After analyzing features like orientation, the
light spots but to specifically oriented line brain integrates them to recognize
segments. complete objects.
- Neurons are organized hierarchically:
- Lower-level cells respond to lines. Recognition-by-Components Theory (RBC):
- Higher-level cells respond to corners, edges, - Objects are recognized by breaking them
shapes, etc. down into simple 3-D shapes called geons
Gnostic Units: (e.g., bricks, cylinders, cones).

Geons - are viewpoint-invariant, meaning they can be

recognized from any angle or perspective.
-Higher-level neurons that can recognize complex
objects (e.g., a face) are called gnostic units or RBC theory - explains how we recognize objects quickly
"grandmother cells." and accurately, even if part of the object is obscured or
degraded.
- These neurons don't respond to just one specific
object but are involved in recognizing complex stimuli. Geons and Object Recognition:

- A small set of geons can combine to form

many different objects, similar to how
Neuron Selectivity and Face Recognition: letters combine to form words.

- Some neurons fire when a human face is Recognition- is automatic and happens even when the
viewed, showing selectivity in recognition. object is partly hidden, as long as some geons are
visible.
Visual System and Receptive Fields:
Face Recognition:
- As visual information moves through the
brain, receptive fields get larger and require RBC theory - can explain general face recognition (e.g.,
more complex stimuli to trigger a response. distinguishing faces from non-faces), but struggles with
- Initially, it was believed there were only two recognizing specific faces, like your own or a friend’s.
types of neurons: simple and complex cells,
but this was oversimplified. Faces - are made up of the same geons (eyes, nose,
mouth), so RBC theory doesn't explain how we tell one
Feature Detectors: face apart from another.

- Some neurons (feature detectors) respond Limitations of RBC Theory:

to specific shapes like corners, angles, or
triangles. - The theory doesn't fully explain how we
- Complex cells can fire in response to shapes recognize specific objects or faces, as it
such as a hand or face, regardless of size. focuses on general object categories.
- RBC theory doesn't account for how prior
Multiple Functions and Parallel Processing: knowledge or context affects how we
perceive patterns and objects.
- Neurons can perform multiple functions
and work simultaneously in parallel (e.g.,
processing spatial and contour information
at the same time).
Confirmation of Biederman's Theory Example of Constructive Perception

- Biederman's theory suggests that geons - When driving and approaching a new
(the basic shapes of objects) are viewpoint- intersection, you can infer that an octagonal
invariant, meaning they should be red sign means "Stop", even if part of the
recognizable from any angle. sign is obscured.
- If the theory were correct, we would expect
neurons to respond to properties of objects Perceptual Constancies
that remain constant regardless of the - We perceive things like size and color as
viewpoint. constant, even when they change on the
Neuroscientific Findings retina due to things like distance or lighting
changes.
- Studies found neurons in the inferior
temporal cortex that react to properties of Example: A car looks bigger as it approaches, but we still
objects that stay the same across different perceive it as the same size.
viewpoints (supporting Biederman's
theory).
- However, some neurons primarily respond Hypothesis Testing in Perception:
to specific views of objects and their
- Our brains quickly form and test
response decreases as the object is rotated,
hypotheses about what we perceive based
which contradicts Biederman's idea of
on:
viewpoint-invariant geons.
-Sensory data (what we see/hear/etc.)
Top-Down Perception
-Knowledge (what we already know)
- Involves a constructive approach, where
perception is built using sensory data along -Inference (using our thinking and reasoning skills).
with higher-level cognitive processes.
- Cognitive processes like knowledge and
expectations influence what we perceive,
Constructive vs. Bottom-Up Perception:
not just the sensory input.
- Bottom-up theories (data-driven) don’t
Key Features of Constructive Perception
fully explain how context affects perception.
-Intelligent perception: Perception is influenced by our For example, context can change how we
intelligence, not just basic sensory data. interpret objects (like "THE CAT").
- Context effects - The environment can
-Reciprocal relationship: The world affects perception, dramatically influence how we perceive
and our perception shapes how we experience the something. Studies show that people
world. identify objects differently depending on
whether they are seen in an appropriate or
-Learning's role: What we learn and know guides how
inappropriate context.
we perceive things.
Prior Expectations Example: It’s easier to identify the letter "o" in "house"
than in "huseo."
-We recognize familiar things faster by using what we
expect. For example, you can quickly spot a friend from Constructive Perception:
afar when you’re expecting to meet them.
- Intelligence plays a key role in perception.
Context and Inference: - We don’t just perceive based on sensory
input; our expectations, knowledge, and
-We also rely on context. For instance, if something is intelligence shape how we interpret what
approaching on rail tracks, we infer it must be a train. we see.
We use prior knowledge and reasoning to make sense Perception and intelligence work together to help us
of what we see, often without realizing it. make sense of the world.
Unconscious Inference:

- We make judgments based on multiple Bottom-Up vs. Top-Down Theories

sources of information without being aware
of it. This is called unconscious inference.
Example: When seeing a partially obscured
stop sign, we still recognize it because of Bottom-up: Focuses on sensory input being enough for
our prior knowledge, even if some of the perception (direct perception).
letters are hard to read. Top-down: Emphasizes the role of prior knowledge and
Intelligence in Perception: expectations in shaping perception (constructive
perception).
- Successful perception requires combining
sensory data with knowledge from past Complementary Approaches:
experiences. - These theories aren't entirely incompatible
Configural-Superiority Effect: but work together to explain perception.
- Sensory info can be richer and more
- Objects are easier to recognize when they informative than bottom-up theories
are part of a configuration than when they suggest, but it's still not always complete.
are isolated, even if the objects are more
complex. Perception with Degraded Stimuli:
Example: Participants can spot the odd one - When stimuli are unclear or degraded (e.g.,
out more easily when it’s part of a shape poor lighting or distractions), we rely on
(like triangles) than when it's just a line. both sensory input and past knowledge to
Object-Superiority Effect: understand what we see.

- A target line in a 3D object is recognized Visual Pathway and Attention:

more accurately than a line in a - Early stages of vision focus on basic features
disconnected 2D pattern. (e.g., color, motion).
Word-Superiority Effect: - Later stages are influenced by what we're
focused on or paying attention to.
- It’s easier to recognize a letter in a word
than in a random sequence of letters.
Vision for Action vs. Vision for Perception: Research Findings:

- Perception: When we simply see an object - People are more likely to use viewer-
(e.g., a cell phone). centered representation.
- Action: When we intend to interact with the - Neurons respond to changes in view,
object, like picking it up, which requires a supporting the viewer-centered approach.
more analytical process.

Landmark-Centered Representation:
PERCEPTION OF OBJECTS AND FORMS:
- Information is stored based on its relation to
Viewer-Centered Representation: a prominent or well-known item (e.g., a
hotel).
- The object is stored based on how it looks - Example: In a new city, you might navigate
from the viewer’s perspective. by referencing your hotel as a landmark.
- The shape of the object changes depending
on the angle. Brain Activation:
- Multiple views are stored, and we mentally
rotate the object to recognize it. - Studies show that people can switch
between viewer-centered, object-centered,
Object-Centered Representation: and landmark-centered representations.
- Different brain regions are activated
- The object is stored based on its structure, depending on which strategy is used.
independent of the viewer’s perspective.
- The shape remains stable across different
orientations. Gestalt Laws of Perception:
- Major and minor axes help define the
object’s properties. - Perception helps organize confusing stimuli
into coherent groups to reduce the number
Key Difference: of things we process.
- Objects or parts of objects are grouped
Viewer-Centered: Object’s appearance depends on the together to form a whole.
viewer's perspective.
Law of Prägnanz:
Object-Centered: Object’s structure stays the same no
matter the viewpoint. - We tend to perceive visual arrays in the
simplest, most organized form possible.
- We avoid seeing a jumble of disorganized
Example (Computer): sensations.

Viewer-Centered: Parts of the computer (screen,

keyboard, mouse) are stored based on their relation to
the viewer.

Object-Centered: Parts are stored in fixed positions

relative to each other, independent of the viewer.
Gestalt Principles:

-Figure-Ground Perception: We focus on a main object

(figure) while other elements serve as background.

-Proximity: Objects that are close together are

perceived as a group.

-Similarity: Objects that look alike are grouped together. We tend to perceive smoothly flowing or continuous
forms rather than disrupted or discontinuous ones.
-Continuity: We perceive continuous lines or patterns,
not broken ones.

-Closure: We see incomplete objects as whole (filling in

missing parts).

-Symmetry: We tend to perceive symmetrical shapes as

a group.
We tend to perceptually close up, or complete, objects
Example: When walking into a room, we notice some
that are not, in fact, complete.
items (e.g., faces in photos) as the figure and the rest as
the background.

The principle of figure-ground can be seen when light-

colored text (figure) stands out against a dark
background (ground).

Gestalt Principles of Visual Perception:

We tend to perceive objects as forming mirror images

about their center.

We group objects together if they are close to each

other.

Figure-ground

When perceiving a
visual field, some
objects (figures) seem

prominent, and other

aspects of the field
We tend to group objects on the basis of their similarity.
recede into the
background.
Two Pattern Recognition Systems: Distorted Face Effect:

-Feature Analysis System: Recognizes parts of objects - Staring at a distorted face for a while can
and assembles them into wholes. Used when focusing cause a normal face to appear distorted in
on specific elements (e.g., parts of a flower). the opposite direction. This demonstrates
how our brain processes faces in a
-Configurational System: Recognizes larger patterns or configurational way.
whole configurations. Important for recognizing faces
and overall shapes (e.g., a tulip in full view). Cognitive and Emotional Interaction:

- Recognition of faces can be influenced by

emotions and age, such as an age-related
Face Recognition: "face positivity" effect
-We mainly use the configurational system to recognize
faces.
UIIAI
-Feature analysis system can also help with face
recognition when we break down specific features (like
facial hair or glasses).

-Both systems may be used together for difficult

recognitions.

Face Recognition and the Brain:

- Occurs mainly in the fusiform gyrus of the

temporal lobe.
- This brain area responds strongly to faces
but not to other objects.

Special Nature of Face Recognition:

- Infants can track human faces more easily

than other complex stimuli (e.g., non-faces).
- Studies show people have difficulty
recognizing parts of faces but recognize
whole faces easily, suggesting that face
recognition relies on a special system, the
configurational system.

Study Example:

- Participants recognized parts of houses

easily, but had difficulty recognizing parts of
faces. This suggests face recognition relies
more on the whole face rather than parts.