Visual perception Flashcards
what is the retina?
the light-sensitive surface on 75% of the inner eye
what is the peripheral retina?
all the retina outside of the fovea. it contains rods & cones
what is the fovea?
what the image in our central vision falls on. high acuity (detailed). contains only cones
what is the optic disk?
point of exit for ganglion cell axons. it corresponds to the blind spot (no rods/cones)
what is macular degeneration?
deterioration of the retina in the macula (around the fovea). it’s a progressive illness that causes irreversible blindness & loss of central vision
short wavelengths
blue/green & high electromagnetic energy
long wavelengths
red/orange/yellow & low electromagnetic energy
process of accomodating to a far object
ciliary muscles relax & lens slims (light only bent a little)
process of accomodating to a near object
ciliary muscles tighten & lens thickens (light bent a lot)
what is accommodation?
the process by which the eye changes optical power to focus on an object as it’s the distance varies
what is myopia?
(i.e. near-sightedness) far objects are out of focus because the lens is too thick or the eyeball is too long. requires concave correction
What is hyperopia?
(i.e. far-sightedness) near objects are out of focus because the eyeball is too short. requires convex correction.
what is transduction?
the image on the retina is transformed into electrical energy
how many rods and cones?
120 million rods & 6 million cones
where are cones and rods?
only cones in the fovea, both cones and rods in the peripheral retina (except for in optic nerve)
how do cones adapt to darkness?
fast adaptation and low sensitivity (tested using only fovea)
how do rods adapt to darkness?
slow adaptation and high sensitivity (adjust more)
what is visual acuity?
can see detail (higher in cones)
what is neural convergence?
allows a neuron to receive input from many neurons in a network.
the difference between neural convergence in rods & cones
more convergence of rods than cones. 1 ganglion cell receives input from 120 rods & 6 cones (on average)
what is the spectral sensitivity of rods and cones?
rods are sensitive to shorter wavelengths (max of 500nm) & cones to longer wavelengths (max 560nm)
what is the Purkinje shift?
there’s more sensitivity to shorter wavelengths in the dark-adapted eye
what are the 3 types of cones
S cones (respond to short wavelengths, blue), M cones (respond to medium wavelengths, green), & L cones (respond to long wavelengths, red)
what is photopic vision?
cone-dominated, fovea & peripheral vision. light-adapted, high visual acuity, most sensitive to long wavelengths, the basis of colour vision, & few neural conversions
what is scotopic vision?
rod-dominated, peripheral vision, dark-adapted, low visual acuity, sensitive to short wavelengths, & no colour sensation
what is mesopic vision?
rod & cone vision together
what are ganglion cells?
receive input from photo receptors. input excites or inhibits ganglion cell.
they have on/off centre/surround receptive fields (on = excitatory when stimulated by light)
how many photo receptors do ganglion cells receive input from?
it varies, but approx 126
what’s uniform illumination?
light on on & off area of ganglion cell. doesn’t have an excitatory or inhibitory effect.
what is luminance discontinuity?
edge of light is touching the centre of a ganglion cell. this causes activation based on the surround
what are the 3 types of ganglion cells?
Magnocellular (M or parasol),
parvocellular (P or midget),
Koniocellular (K)
Magnocellular ganglion cells
get most of their input from rods & aren’t colour specific
Parvocellular ganglion cells
get input from single M or L cones. they’re colour specific (green or red on/off)
Koniocellular ganglion cells
excitatory input from S cones & inhibitory input from M & L comes (blue on)
eye anatomy
axon of ganglion cells -> retino-geniculo-striate pathway -> temporal/nasal representation -> optic nerve -> optic chasm -> optic tract -> LGN -> visual cortex
which side of the brain does visual information go to?
things in left visual field go to temporal representation in right eye or nasal representation in left eye (visa versa for right visual field).
at optic chasm, the nasal axons cross over to the other side of the brain & temporal axons stay on the same side. the visual fields are represented in the contralateral hemisphere.
what is the structure of the lateral geniculate nucleus?
6 layers of axons. 4 from P neurons, 2 from M neurons, & K neurons are interlayered.
visual cortex structure
6 areas - V1 (primary visual/striate cortex), extrastriate areas (V2-V5), & IT (inferotemporal cortex)
describe the V1 area
where optic radiation terminates. has M & P cell layers.
some V1 neurons are orientation-selective, motion-direction selective, or selective for colour/brightness.
what are orientation-selective neurons?
elongated receptive fields that are side by side. stimulated by a stimulus with the preferred orientation
who studied orientation-selective neurons in cats in the V1 area?
Hubel & Wiesel (1959)
what are motion-direction selective neurons?
respond when a stimulus in their receptive field matches their preferred motion direction.
describe the V2 area
receptive fields 2x as large as in V1
respond to basic stimulus features (orientation, motion-direction, brightness etc.). integrate more info than in V1, so respond to more complex features, too (e.g. arcs, lengths, angle, shapes, texture)
who found evidence for the dorsal and temporal stream?
Ungerleider & Mishkin (1982) - studied monkeys with lesions.
whats the difference between the parietal and temporal stream?
parietal (dorsal) - processes object location (where) - V3 & V5
temporal (ventral)- processes object identity (what) - V4
describe the V4 area
5x larger receptive fields than in V1
neurons respond to object-defining features, e.g. colour, orientation, complex shapes, & texture
what neurons converge onto V4?
mainly P & K
what neurons converge onto V3?
mainly M
what is sensation?
the uninterpreted sensory impressions created by detecting a stimulus
what is perception?
the cognitive processes of making sense of sensations
what is structuralism?
studies the elements of consciousness. based on the idea that conscious experience can be broken down into sensory elements
who came up with structuralism?
Wundt (1879)
what is introspection?
someone describes a description of the sensory elements they experience (Titchener, 1901)
what did Fechner do?
1860-aims to measure the elements of perception.
how did Fechner measure the absolute perception threshold?
Method of adjustment
method of limits
method of constant stimuli
what is the absolute perception threshold?
how much stimulus intensity we need to perceive a stimulus.
what is the method of adjustment?
participants adjust intensity of a test light until they’re just able to perceive it (min threshold).
the more time in the dark, the less stimulus light needed for perception
what is the method of limits?
participants presented with trails of increasing/decreasing light intensity & asked if they can see the stimuli.
the trials were done with ascending/descending light intensity
mean absolute perception threshold = 2.25 candela
what is the method of constant stimuli?
same task as method of limits. more trials per light intensity & light intensity’s randomised for each trial.
80% of stimuli detected at 4 candela, & 25% at 2 candela
50% detected at intensity 3 (change level)
What did Weber do?
1834- aimed to measure the difference threshold (minimum intensity to discriminate between 2 stimuli)
what’s Weber’s law?
the change in a stimulus needed to discriminate it from another stimulus is a constant ratio of the original stimulus (difference threshold = 8%)
brighter stimulus 1 is, the greater required brightness for stimulus 2 to be perceived as different
what did Stevens do?
1957- aimed to come up with a subjective magnitude estimation to measure the relationship between stimulus intensity & perceived intensity.
an increase in the perceived stimulus intensity can be larger (response expansion), or smaller (response compression), than the increase in the measured stimulus intensity
what did Gibson do?
1966- ecological theory of perception. Assumed perception’s direct (perception= sensation). & real-world provides sufficient context for visual systems to directly perceive what’s there.
perception’s based on invariant info extracted by observer’s movement. when observer moves, optic array becomes ambient, but some info remains invariant
what are some examples of invariant visual info?
optic flow pattern
texture gradient
horizon ratio
gravity, straight lines, & reflectance on diff areas
what is optic flow pattern?
focus point remains motionless, whilst rest of visual field moves away from this point
what is texture gradient?
the elements of a textured group appear denser in the distance.
what is horizon ratio?
the proportion of objects above & below the horizon is constant for objects of the same size standing on the same ground
what is the affordance of objects?
combine physical & psych states with optic arrays to recognise what an object does
common visual illusions/ambiguity
fraser spiral illusion, vertical-horizontal illusion jastrow illusion irradiation illusion figure/ground ambiguity feature ambiguity depth ambiguity
what is erroneous perception?
when what we perceive isn’t represented on our retina
what is equivocal perception?
when we perceive different objects from the same retinal image
common real-world illusion
ecological validity
moon illusion
waterfall illusion
wagon wheel effect
what is the moon illusion?
the moon appears larger when it’s closer to the horizon
what is the waterfall illusion?
motion aftereffect. after observing motion in one direction, stationary objects appear to move in the opposite direction
what is the wagon wheel effect?
stroboscopic effect where a moving wheel appears to stand still/move slowly, or in opposite direction to its true rotation
what did Gregory do?
1970- developed constructive theory of perception.
perception’s indirect & a constructive process of hypothesis testing.
these hypotheses help us understand sensory info based on our previous experiences (explains how perceptual errors occur)
what is the gestalt psychologist theory?
presented 2 flashes of light in rapid alternating fashion to create an illusory movement. the whole is more than the sum of it’s parts.
we achieve perception of an object (i.e. gestalt) by perceptual organisation of the environments, e.g. by grouping elements & segmenting elements.
what are the principles of perceptual organisation?
proximity- elements close together grouped together
similarity- elements that look similar grouped together
common fate- elements that move together are grouped together
good continuation- elements that continue are grouped together
closure- elements that close a figure are grouped together
relative size- elements relatively smaller are grouped together
surroundedness- elements in a surrounded area grouped together
orientation- elements horizontally/vertically orientated grouped together
symmetry- elements symmetrical grouped together
what is the law of pragnaz?
of several geometrically possible organisations, we will perceive the object which possesses the best, simplest, & most stable shape.
what does the use of colour enhance?
perceptual organisation,
top-down control,
automatic bottom-up processing
what colour are short & medium wavelengths?
blue
what colour are medium wavelengths?
green
what colour are medium & long wavelengths?
yellow
what colour are long wavelengths?
red
what is chromatic colour?
solid objects - the result of selective reflection, where some wavelengths are reflected & others are absorbed
what is an achromatic colour
black, white, & grey - the result of equal reflection
what are the 3 colour dimensions?
hue (colour value),
saturation,
brightness
what is hue?
different shades & represented on the colour wheel.
red, green, blue, & yellow are pure colours
what is saturation?
the amount of white added to the colour
what is brightness?
the amount of light reflected
how many colours are there?
Bureau of Standards identified 7500 colour names (1976)
what is additive mixing?
mixing the light of colours
all the wavelengths added are reflected
what is subtractive mixing?
mixing the pigment of colours
the pigments absorb the same wavelengths they absorb when they’re alone
only wavelengths that both pigments reflect remain
who developed the trichromatic theory of perceiving colour?
Young, 1802.
was defined by von Helmholtz, 1852.
what is the trichromatic theory of colour?
any colour can be matched correctly by adjusting the proportions of 3 different wavelengths. Colour’s based on 3 different types of receptor mechanisms
evidence- we have 3 types of cones for different wavelengths
what is monochromatism?
colour blindness
have no cones, so you only have rod vision (scotopic).
can perceive shades of brightness, & vision is slightly blurred
what is scotopic vision?
only rod vision
what is dichromatism?
colour deficiency
have only 2 types of cones
what is the opponent-process theory of perceiving colour?
Colours have a complementary afterimage
if we are responding to one colour, we can’t respond to the afterimage one at the same time
who developed the opponent-process theory of perceiving colour?
Ewald Hering, 1878
what is the afterimage of adapting to blue?
yellow
what is the afterimage of adapting to green?
red
what is the afterimage of adapting to black?
white
what is some physiological evidence for the opponent-process theory of perceiving colour?
opponent neurons in V1, V4, & inferotemporal cortex can be either excited or inhibited by the wavelength that matches the colour channel they’re receptive to
what are vertical object features?
depth & size (can verify/measure)
what are oculomotor cues?
body-related depth cue
the state of the eyes- accommodation & convergence
how your eyes can inform you if an object’s close/near
what is accommodation?
the relaxation of the ciliary muscles when an object’s far away & tightening when an object’s near
what is convergence?
the eyes are parallel when looking at something far away & less so when close
what are monocular cues?
stimulus-related depth cue
they work with 1 eye
we can extract 3D info from still 2D images using pictorial cues
what are pictorial cues?
a sub-category of monocular cues occlusion relative hight relative size perspective convergence familiar size atmospheric perspective texture gradient shadows motion parallax deletion & accretion
what is occlusion?
partially hidden (occluded) objects are further away than occluding objects
what is relative hight?
objects higher in the visual field are further away than objects lower in the visual field
objects with their bases closer to the horizon are further away than objects with their bases in front
what is relative size?
when 2 objects are equal size, the one further away will take up less of the visual field
what is perspective convergence?
converging parts of an object are further away
what is familiar size?
knowledge about the physical size influences how far away we perceive an object as
what is atmospheric perspective?
objects further away are less sharp & have a blue tint
what is texture gradient?
denser textures indicate further away objects
what are shadows?
location of the shadow impacts how far away an object looks
what is motion parallax?
when in a moving object, nearby object glide away quickly, & far objects glide slowly.
the image of the far object travels a short distance on the retina, whereas near objects travel a short distance
what is deletion & accretion?
an object being covered & uncovered is further away (occlusion in motion)
what are binocular cues?
stimulus-related depth cues
depend on info from both eyes
what is retinal disparity?
binocular cue
retinal image on left & right eye are slightly shifted. image falls into fovea in both eyes when fixated
all objects with the same distance as the fixation point fall on corresponding retinal points on the horopter
if an object isn’t on the horopter, there’s a difference between the object’s corresponding & actual retinal point (retinal disparity)
what is the horopter?
an imaginary plane through the fixation point connecting all objects of which retinal images fall on corresponding points (all objects that have equal distance from the observer than the fixated object)
what is stereopsis?
the ability to use retinal disparity as a cue to perceive depth
each distance from fixation creates a different amount of retinal disparity
the larger the disparity, the further the object is from fixation
what is evidence disparity’s enough to create depth perception?
Julesz, 1971- random dot stereogram with 1 dot shifted. this can be perceived as 3D using a stereoscope (no pictorial cues)
what is size perception?
the size of retinal object representations are measured in degrees of the visual angle (depends on physical object size & physical distance from observer)
what is size consistency?
perception of an object’s size is relatively consistent even when viewed from different distances (creates different visual angles
this is because of the size-distance scaling equation
what is the size distance scaling equation?
S = K(R x D)
s=perceived size
r=size of retinal image
d=perceived distance
when do visual illusions occur?
there’s a lack of pictorial depth cues
what are the types of motion?
real (produces retinal motion) & illusory (apparent motion perceived in a static stimulus)
what are 3 examples of illusory motion?
induced motion
motion aftereffects
visual illusions
what is induced motion?
a moving frame of reference creates the illusion that a stationary object is moving in the opposite direction
what is motion aftereffects?
after adapting to motion in 1 direction, stationary objects seem to move in the opposite direction
who studied middle temporal neuron’s firing rate, which depended on motion direction?
Newsome et al., 1995- monkeys watched moving dot display with random -> moving in the same direction
at 12.8% coherence, monkeys judged motion direction accurately almost all the time
what is the aperture problem?
the direction of a moving stimulus through an aperture (e.g. receptive field) is ambiguous
solutions to the aperture problem
pooling the responses of multiple neurons
end-stopped cells signalling the end of a stimulus to define stimulus direction
role of STS region
cross-talk between dorsal & ventral stream
combines input about object form & motion direction to create biological motion
what is biological motion?
self-produced motion of a biological being (produces retinal motion.
we are experts at perceiving & recognising biological motion
what is retinal motion?
a successive receptor stimulation
motion perception depends on retinal motion & eye movements
what are the eye muscles?
4 at top, bottom, left, & right of each eye, & 2 wrapped around it
what is the corollary discharge theory?
3 relevant signals (image displacement signal, motor signal, & corollary discharge signal)
these signals are compared to see if real motion’s occurred
what is the image displacement signal?
image moves across retina & successively stimulates receptors (retinal motion). the network that processes this is the comparator
what is the motor signal?
a motion command is sent from the brain to eye muscles
what is the corollary discharge signal?
copy of the motor signal is sent to the comparator
Corollary discharge theory- when there’s a static stimulus in a moving eye…
there’s 1 of each signal (no real motion)
Corollary discharge theory- when viewing a moving stimulus…
there’s an IDS signal, but no others
Corollary discharge theory- when your eye is following a moving stimulus…
there’s a CDS & MS signal
neuronal evidence for the Corollary discharge theory
Galletti & Fattori (2003)- real-motion neurons respond to moving stimuli (not static) in moving eye
explains why a moving stimulus generates motion perception & why a moving eye doesn’t
approach to object recognition- template-matching
object representations = templates. visual input’s transformed till it matches a template (e.g. rotation)
(problem- requires lots of templates)
what is object recognition?
when visual input matches an internal object representation
approach to object recognition- feature analysis
object representations = list of features
problem- spatial arrangement isn’t taken into account
approach to object recognition- structural analysis
objects can differ in 2D appearance but become equivalent in 3D space (visual input must be analysed at different levels)
approach to object recognition- structural analysis levels
Marr (1982)- visual object analysis goes through 4 levels of representations: raw primal sketch full primal sketch 2 1/2D sketch 3D model
Structural analysis- raw primal sketch
processing intensity (brightness) changes across retina
Structural analysis- full primal sketch
geometric organisation of intensity changes (defines edges/contours)
Structural analysis- 2 1/2D sketch
processing spatial locations of visible surfaces. includes depth, but only represents object from observer’s viewpoint (not generalisable)
Structural analysis- 3D model
viewpoint-independent representation
can match with stored object representation
model of object recognition
Ellis & Young (1988) initial representation -> viewer-centred representation -> object-centred representation or object recogniton units -> semantic system -> name retrieval
Case studies supporting model of object recognition
Mr S (Benson & Greenberg, 1969)- could maintain fixation, desc colours, etc. but couldn't perceive shape/form or name objects (all stages that depend on viewer-centred representations impaired AB (Warrington, 1975)- couldn't name common objects (model cut off at object recognition units)
model of object recognition in patients with right posterior lesions
Warrington, 1982- couldn’t match objects from unusual views (object-centred representation cut off)
model of object recognition & apperceptive agnosia
impaired perceptual stages - can’t match/copy simple shapes
model of object recognition & associative agnosia
impaired semantic stages - intact percept has no meaning
what is perceptual categorisation deficit?
can’t create object-centred representations
