Week 2 Flashcards
Dorsal vs. ventral
Dorsal = where stream (depth)
To posterior parietal cortex
Ventral = what stream
To infotemporal cortex (below temporal)
Binding problem - spatial position
Different systems “remember” where in space those properties came from.
Assumes if properties come from the same space they belong to the same object.
Binding problem - rhythm
If the different neural systems processing the stimulus are firing in synchrony, one can assume they are processing the same environment.
- partially due to attention
Perceptual constancies
figures perceived as constant despite changes in retinal images
- size (as things move away from us)
- shape (when viewed from diff angles)
- brightness (compared to background like the checkerboard shadow thing)
- colour (when in diff lighting)
Gestalt principles; form perception
how we group and organize input into objects
- figure-ground
- similarity
- closure
- proximity
- continuity (X)
- common fate (move together like a school of fish)
- simplicity (made of simpler forms)
Oculomotor vision cues
The actual info from eyeball muscles
Accommodation: Lens bending for near vs. far things
- near spaces
Convergence: angle between eyeballs for near vs far things
- up to 10m
Monocular vs binocular cues
monocular
- can see the illusion with only one eye
binocular
- must use both eyes
Binocular cues
convergence
- angle at which eyes point to each other
- up to 10m
binocular disparity
- eyes have slightly different views of the world
Monocular cues - Motion
- motion parallax/movement gradient
(as an observer moves, the retinal images of nearby objects move more rapidly than do the retinal images of objects farther away) - structure through motion (ex. glow in the dark dots on people)
Monocular - Static (pictorial cues)
- interposition
- linear perspective (converging lines as they get further into the distance)
- reduced clarity (of further things)
- height in the visual field (further = higher)
relative size - shading (ex. concavity)
- textural gradients (closer things bigger and more detailed)
Bottom-up processing
take in raw data
Top-down
use prior knowledge, conceptual
Global and local
Global to local (recognize whole, then parts)
Local to global (recognize parts, then whole)
Reify
The constructive or generative aspect of perception
Perceived spatial info is more explicit than the sensory stimulus itself
Reify = to make real
- ex. when you see things in the blank spaces of a drawing
Optical illusion examples
Ames room
- messes with depth perception
Disney castles
- small windows make them look taller
Lord of the rings
- position hobbits to make them look smaller then gandalf
- Cornsweet effect
- Shadow on checkerboard
The educated eye
are some experts able to “see” better
Ex. cinematographers, snipers
Due to training or natural talent
Object recognition
- first organize by our perceptual system
- then identify the object
recognition =/= perception
recognition depends on contextual effects
and constraint
satisfaction
Object agnosia
Apperceptive agnosia
= Perceptual representation where you can’t organize parts into a whole
- Can use context clues
Ex. can’t find dalmatian in that spotty image
Associative agnosia
- Can’t do identification; meaning
Prosopagnosia
face blindness
- can’t recognize their identity
- can usually tell it’s a face and the gender and age of a person, but can’t tell you who it belongs to
Akinetopsia
- Patient LM developed this disorder at 43 bc of a blood clot
- Can perceive objects normally but can’t perceive motion
- Dees nothing during the movement, but sees the object before and after it moved
- She can infer change but can’t perceive the motion
It was hard for her to cross the street bc she didn’t know which cars were parked or moving, had a hard time following conversations because she couldn’t see lip movement, was anxious with people moving bc she couldn’t see their movement, pouring cups of coffee
Eye path
Cornea -> lens -> retina
Retina - rods and cones
Rods
- Rods used for semidarkness
- Can’t see colour
- Can see different intensities of light
Cones
- Less sensitive so they need more light
- See colour
- More acuity = can see detail
- Closer to the fovea
Lateral inhibition
Active cells INHIBIT their neighbouring cells
Cell path
Photoreceptors → bipolar cells → ganglion cells
Ganglion cells
Their axons make up the optic nerve
Optic nerve
Leaves the eyeball and goes to the LGN in the thalamus
Wavelength of light + visible spectrum
- 360-750 nm is the visible spectrum
- Wavelength is distance from crest to crest
Parallel processing
- processing in different areas at the same time
- as opposed to serial processing
Advantages of parallel processing
- Speed; don’t have to wait for one analysis to be complete
- Mutual influence; they fill in each other’s blanks
Conjunction errors
Binding
- When we have an overload of info we might bind incorrectly
Ex. see a red K and a blue T but report seeing a red T and a blue K
Unconscious inference of size
- used when keeping size constancy
- calculating relative size of objects with other unchanging objects
(ex. Dogs size vs chairs) - size of image on retina x distance between you and the object
Optic flow
The pattern of change in the retinal image in which the image grows larger as the viewer approaches an object and shrinks as the viewer retreats from it
