Visual Perception

B&C

Axons

Transmit electrical potential down to terminal buttons, fat surrounded = conduction. ranvier= conduction jumps

-70 mv resting membrane potensial

depolarise (na+ in)… if cross trigger zone= burst of action potentials.

Excitatory neurotransmitters

Open NA chanels- depolarises

inhibitory neurotransmitters

cl- in, less likely to depolarise

Action potential reaches terminal button

Neurotansmitters released to synapse which excite/inhibit post synaptic neuron

First visual receptor cells

CONES= colour and detail perception, located in fovea (centre of field)

Second visual receptor cell

Rods= vision in dim light, located in periphery

How Retinal ganglion cells receive input

from a few cones/hundred rods

Reception (body interacts with enviroment)

Rods/cones (photoreceptor cells) absorb physical light energy

Transduction (physical energy converted to form nervous system understands)

Photoreceptor cells turn absorbed light energy into electrochemical patterns and nervous impulses that travel along neurons via optic nerve

Coding- Mapping where specific neural signals represent specific physical properties

brain establishes one to one correspondence between incoming neural firing pattern and the visual Left visual of the original light stimulus

Left visual field processed by?

Right hemisphere

Parvocellular pathway

Processes colour, fine sharp input from cones

Magnocellular

Processes movement + rapid changed in environment info from rods

Vertical axis measuring time in milliseconds (looking at how visual info flows upwards through the diff visual areas of the brain over a tiny fraction of a second)

regions lower (LGN) response 30 ms after light hits eye. Regions higher (IT) 100 ms— brain understands what object is

Box size when trackingg route of visual information flowing through diff visual areas of brain

Size indicated how much surface areas that specific part of brain takes up, large= massive amount of brain V1,V2 dedicated to early visual processing

arrows show direction, what about the %?

% show proportion of nerves fibres taking that exact route (where most electrical data flows)

Step 1 how visual signals travel from eyes to brains thinking centres

Eye / Retina. light hits photoreceptors passes thru bipolar cells and reaches retinal ganglion cells pack up forming optic nerve , leaves eyeball

2

LGN (Lateral geniculate Nucleus)- ganglion cell fibres leave eye, cross over synapse in thalamus

3

striate cortex- V1 primary sorting hub at back of head4

4

Extrastriate cortex- high level processing v2,v3,v4,v5

Zeki functional specialisation theory

Visual brain processes diff attributes of visual scene in separate regions of cortex

V1 and V2

basic visual processing

V3/ V4A

Form perception ( of moving stimuli especially)

v4

Colour and shape perception

V5/ Mt in humans

Motion perception

LOC

Object perception

OFA

Face perception

FFA

Face perception

imge

Cells in each area responding seleticelt to four stimulus feautures

Each columns represents a feature of an object, height= % of cells that care about that feature

Orientation, V1,V3,V2 are high meaning

Early brain areas highly specialised at detecting the basic angled, lines and borders of objectsM

MT high up 90%

Direction= proves MT area is brains motion tracker

Depth

Shows V2 MT help calculate distance between eyes

V4 an VP highest in column?

V4 heavily specialised in processing colds and hues

Positves of Zeki functional speculation theory

Highly ambitious+influential way to explain how brain divides

Bad Zeki

Real brains messier areas are less specialised than Zeki thought.

Leaved binding problem unresolved.

binding problem

Zekis thing Fails to explain how the brain stitches separate features (like color from V4 and motion from MT) back together into dingle object so how is visual info combined and integrated?,

Binding by synchrony hypothesis

Neurons that code different dimensions start to oscillate in synchrony, buts its unclear how and why such synchrony might occur as precise synchrony is implausible … So if one module codes shape, and other codes color, how do we bind them to get representation?

perception action model

Brain processes visual info through two distinct pathways

Ventral stream What pathway Connects V4,LOC and IT

Vision for perception- regognize objects tells mind what. an object is, Allocentric (objects are in relation to each other), Charecteristics (info lasts in memory) = sustained representation

intention of parvocellular pathway

The dorsal stream How/where path CONNECTS V5

Purpose= vision for action so provide real time spatial info needed to physically interact with enviroment (like reaching to grab coffee cup)=

Reference frame- body centred so maps where an object is relative to own body guiding limbs

Characteristics- relies on shortt lived representations

extension magnocellular pathway

Optic ataxia/ Perenin & Vighetto (1988)

Vision for perception is working (✅), but Vision for action is broken

James et al. (2003) / Visual Form Agnosia

Vision for perception is broken (❌), but Vision for action is working

Concept

Double dissociation- brain damage can destroy one visual path while leaving other normal

Trick

• Optic Ataxia: Think of an Attack (Ataxia). You can see the target fine, but your hand's aim is completely broken.

• Visual Agnosia: "Agnosia" sounds like Ag-NO-sis (Diagnosis). You cannot diagnose or recognize what the shape is, but your hands still know how to hold it.

Visual illusions- both vertical lines are same length findings

Illusion size was only 5.5% so in terms of vision for perception, illusion size jumped conscious fooled so visual illusions mostly affected our perception system not action system

visual illusions. on the circles where both middle ones are the same size

Conscious eyes are fooled by the context but your motor system ignores it to guide accurate movements

Complexities of grasping movements - it requires ventral and dorsal stream when..

1. Information from memory to guide u how to hold stuff

2. Conceptual Knowledge; when u need to undertsand what the object is to hold it correctly

Perception action model evaluation: Very influential theoretical approach

Major influence on psychology and how scientist viewed vision

Two systems less independent and more interactive than assumed theoretically

2 systems aren’t as independent as thought

Patients with optic ataxia and visual form agnosia have problems with BOTH vision-for-perception and vision-for-action: no clear double dissociation

Double dissociation= one impaired other intact= so harder to argue they’re operate

Visual form agnosia

A person struggles to recognise objects by sight, even though some basic visual abilities are intact.

Optic ataxia

A person struggles to use visual information to guide their hand movements accurately.

Findings with visual illusion are variable

Two dorsal systems (from perception action model evaluation)

1. Dorso dorsal systems used to grasp object rapidly , venture dorsal system using memorised object knowledge to use object appropriately

Row 1 related ti perception and actins

traditional view where vision creates a conscious perception which then directly triggers physical action

Row 2

Vision splits into 2 independent systems, ventral for perception, dorsal for action

Row 3

An updated framework showing that the perceptual and action systems are not completely isolated; they continuously interact and share information during processing.

Row 4

Final processing in the motor cortex is preceded by complex interactions between dorsal and ventral pathways

Colour Perception, Hue

Colour itself, property distinguishes red from yellow or blue

Brightness

The perceived intensity of light

Saturation

Dictates weather colour is vivid or pale- influenced by amount of white present

Dual process theory

A theory combining trichromacy and opponent-process ideas, stating that signals from three cone classes feed into three neural channels to determine perceived object colours.t

three Cone classes

• The photoreceptors responsible for early-stage colour processing: Red (long-wavelength), Green(medium-wavelength), and Blue (short-wavelength).

three opponent channels

Neural pathways that process cone inputs by subtraction or addition: Red-Green, Blue-Yellow, and Light-Dark(achromatic).

image

Colour constancy

Perceived colour stays same even if changes in wavelength/light source

Chromatic adaptation

Sensitivity ti illuminant of any given colour decreases over time

Retinex theory

We compare light reflected from surface against light reflected from adjacent surfaces to c

Colour constancy evaluation artificial visual environments

focusses on simple artificial setups for it over natural complex world

Objective vs subjective judgement colour

Observers show less constancy when making objective judgements ("what colour is it?") than when making subjective ones ("what colour does it look like?").

individual differences in colour constancy

Large variations in performance between different human observers that remain poorly understood by researchers.

Monocular cues- Linear perspective

Brain sees two railways and they merge into one — parallel lines appear to converge.

Texture gradient monocular cues

• A depth cue where details become denser and less distinct as a textured surface slants or stretches away into the distance.

(like pebbles on floor)

Monocular cues Interposition

An overlapping cue where a nearer object blocks the view of a farther object, showing which one is closer. (Coffee cup blocks laptop screen so brain knows its closer)

Monocues familiar size

sing our prior knowledge of an object's actual size to accurately gauge how far away it is.

Blur monocular

An ambiguous depth cue where objects appear fuzzy, which can signify either that they are very far away or located in our peripheral vision.

Monocular - Motion parallax

• A dynamic depth cue where, as you move, closer objects appear to move faster and farther across your retina than distant objects.

• 
  

• (Looking out train window fence goes quickly mountains go slow)

Binocular cue

Stereopsis

What does stereopsis enhance

Depth perception

Stereopsis based on?

Sllight diference / disparity between two retinal images ppl w lazy eye

Three ways information from different cues combined

1. Additivity (info from all cues combined)

2. Selection(info from single cue use with info from other cues ignored)

3. Weighting of cues (more reliable cues)

Depth cues

Visual signals your eyes and brain collect from your surroundings to determine how far away objects are. these split into two cues- binocular both eyes and monocular cues work with one eye

Combining information from different

cues in depth perception

breaks down how your brain calculates depth when different visual signals (cues) give slightly different information.

Small Conflicts vs. Large Conflicts

Small= addivity and weighting (if cues are only slightly off brain combines them (addivity) Gives higher status (weighting) to trustworthy cueu

Large- if two cues give unrealistic stories brain realises can’t average them so choses one single cue

Depth perception is close to optimal

Brain is mathematically efficient at combining these signals to give most accurate possible view of world

Global assessment

IRL surroundings are rich with information instead of treating every depth cue like an isolated math problem and trying to glue them together later brain takes a step back and judges overall structural layout of environment all at once

Object size measured in degrees of visual angle

if objects are nearer they take up more degrees of visual angle like moving hand nearer and farther from eye- u see it occludes more or less background but if we move back we reach size constancy

Size constancy

example of perceptual constancy (perceive an object as remaining the same physical size regardless of changes in its distance from the observer.izes

size distance invariance hypothetical

1. Perceived size is proportional to retinal size and perceived distance

size distance what do we have

retinal size, we can infer distance from the depth cues

We can infer distance from the depth cues, two objects are the same distance away one has larger retinal size.

AMES ROOM back wall sloped but perceived distance on the left and right corner is the same because most depth cues are blocked, if u had depth cues the ppl wold look the same size and the one on the left would look nearer

Familiar size

size perception of objects depends on memory of their familiar size . Judgements are significantly more accurate for invariantt size objects than variable size objects

Invariant size- when u look at an objects that never changes size in the real world, brain uses mental template and accuracy stays perfect in invariant size

Variable size graph

if an object can come in any size - memory template can’t help as much, brain relies on environment so judgement is lesss accurate

unfamiliar objects graph - lowest accuracy

without a memory trmplatte to guide u, size constancy suffers most especiallyy at longer distances

Blindsight: perception without

awareness—— paradox

People with blindsight are functionally blind in parts of their visual field due to brain damage (V1), yet if you throw a ball at them or ask them to point to a light, they can catch or locate it accurately without "seeing" it

Type-1 vs. Type-2 Blindsight:

type 1- patient not conscious experience of visual stimuli, subconscious guesses that happens to be correct. “Swear I can’t see anything but if I have to guess the object is moving to left) and there right

Type 2- Patient has some residual awareness, might feel a vague change in enivroment without seeing an object “I still can’t see a picture of anything but I have a feeling that x”

Blindsight patients

often show evidence of degraded conscious vision when sensitive measured of conscious awareness are used, reorganisation of brain connectivity in blindsight patients explains many findings”

what is blindsight

Person damages to primary visual cortex (V1) part of brain that makes conscious visual , coz its damaged patients are cliciallt blind in parts of their vision like show them black apple they’ll get it right by chance