Perception I Flashcards
What are the 5 steps of visual processing?
(recap)
Photoreceptors
-> light is transduced into neural signals
-> photorecptors in retina
(rods vs cones,
fovea vs periphery,
scotopic vs photopic vision)
Ganglion Cells
-> summarize information from several photoreceptors
(Donut-shaped receptive fields as ganglion cells
centre vs surround,
on-centre vs off-centre ganglion cells)
LGN cells
-> Ganglion Cells send information to LGN cells
(same Donut-receptive fields as ganglion cells
feedback from higher-level areas)
Striate Cortex
-> cells summarize information from LGN cells
(Building “line detectors” from donuts
orientation tuning
simple vs complex cells
end stopped cells)
Higher-level brain areas
-> responding to increasingly complex and specific stimuli
(faces,
places,
body parts,
hands)
What are the different cell types in low-level vision?
(recap)
Retina
-> Ganglion Cell
-> Donut-shape
-> Dot of light surrounded by darkness or vise versa
LGN
-> LGN Cell
-> Donut-shape
-> Dot of light surrounded by darkness or vise versa
Striate Cortex
-> simple cell
-> elongated, clearly divided
into excitatory / inhibitory
-> Orientation tuning possible
-> stimulus exactly on ON7OFF region
=> edge detector
-> one, one inhibitory are
froming a clear edge
-> Edge between light and dark
=> stripe detector
-> excitatory stripe between two
inhibitory areas (or vice versa)
-> stripe of light between dark (or vice versa)
-> Complex cell
-> elongated, not clearly divided
inte excitatory / inhibitory
-> Orientation tuning possible
-> stimulus anywhere within receptive field
=> end stopped cell
-> stimulus of specific length
What are the different vision levels?
(recap)
Low-level vision
-> extracting basic image features
(edges, lines, patterns from the image)
-> simple yet abstract stimulus properties
-> in striate cortex
Mid-level vision
-> Grouping low-level features together into objects
-> identifying boundaries and surfaces in an image
High-level vision
-> identifying complex stimuli
(such as faces, places, body parts)
-> inferior temporal cortex
(FFA, PPA, EBA)
What is the key function of mid-level vision?
(mid-level vision)
Keyfunction of mid-level vision
-> figure ground assignment
(black object, white background or vice versa)
(not fully closed figure on same color background)
=> boundary detection
What attributes helps us differentiate between figure and background?
(mid-level vision)
Surroundedness
-> background usually surrounds figure
Size
-> figure is usually smaller than background
Symmetry
-> figure is more likely to be symmetrical
Parallelism
-> parallel contours more likely to be figure
What is bottom-up und top-down processing?
(mid-level vision)
Perception is always an interplay of
Bottom-up processing
-> information about the environment as
represented by the senses
(Sound, Color, Features, Location, Movement)
Top-down processing
-> Information about the environment that
an individual brings
(Knowledge, Expectation, Experience, Memory, Culture)
=> Interpretation trough Top-down and Bottom-up
processing
What is the dynamic cycle of perception?
(mid-level vision)
Stimulus
1. Environmental stimulus
2. Attended stimulus
3. stimulus on the receptors
Electricity
4. Transduction
5. Transmission
6. Processing
Experience and action
7. Perception
8. Recognition
9. Action
What can perception be defined as?
(mid-level vision)
Perception
-> process of recognizing, organizing, interpreting
from senses
-> Not an exact copy of the world
-> based on our past experience and espectations
What 2 approaches are there to understanding perception?
(mid-level vision)
Direct perception theories
-> focus on bottom-up processing
-> perception comes from stimuli in the environment
-> parts are identified and put together
then recognition occurs
constructive perception theories
-> focus on top-down processing
-> people actively construct perceptions using
information based on expectations
What are direct perception theories?
(mid-level vision)
Biederman (1987)
Recognition by components theory (RBC)
= elemental features: “geons” 3D shape
-> Objects are recognized by perceiving elementary
features
-> 36 different geons identified
-> idea that neurons along neral hierarchy process
increasingly complex stimuli
What are constructive perception theories?
(mid-level vision)
focus on top-down processing
-> basing one perception on another
-> interferences based on context
-> guessing from experiences
(making sense of ambiguous stimuli)
What is the Ebbinghaus size illusion?
(mid-level vision)
Illusion where on perception is based on another
example
size of an orange circle perceived
-> smaller when surrounded by bigger sized black circles
-> bigger when surrounded by smaller sized black circles
What is Helmholtz’ theory of unconsious inference
(Gestalt principles)
When image on the retina is ambiguous
-> perceptual system decides
Unconscious inference
->likelyhood principle
(most likely interpretation is favoured)
-> based on unconscious assumptions, inferences about
the environment
=> process is rapid, feels almost automatic
What is the Gestalt psychology?
(Gestalt principles)
Wertheimer, Köhler, Koffka
Gestalt psychology
-> the perceptual whole is more than the sum of its
sensory parts
-> the mind groups patterns according to rules of
perceptual organization
(Gestalt principles)
Gestalt psychologists
-> see gestalt principles as “intrinsic laws”
(built into the system)
-> effect of perceptual princles should be stronger
than effect of experience with the world
-> only minor role of experience
What are the gestalt principles?
(gestalt principles)
Good continuation
-> points connecting to form straight or smoothly curving
lines
-> perceived as belonging together
-> lines are seen in a way to follow the smoothes path
Prägnanz
-> principle of simplicity or good figure
-> patterns are seen in the simplest way possible
Similarity
-> similar things appear to be grouped together
-> color, size, shape, orientation
determine similarity
-> grouping also as result of enclosure or proximity
Occlusion
-> when an object is partially covered by another, we
perceive it as continuing behind that object
Familiarity
-> Elements are more likely to form groups when groups are familliar or meaningful
