Cognitive neuroscience: Perception and visual Recognition

primary pathway

retina → optic nerve → optic chiasm → optic tract → LGN (in thalamus) → optic radiations → V1 (primary visual/striate cortex)

visual field

left visual field is processed by the right hemisphere. and vice versa

each LGN receives input from both eyes but represents the contralateral visual field

circle shaped

optic chiasm

the crossing of the nasal half fibers

LGN (lateral geniculate nucleus)

a 6 layered thalamic structure between retina and V1. It receives feedback from cortex and other modulatory inputs

What layers of LGN receive input from the contralateral eye

contralateral - opposite eye

layers 1, 4, 6

What layers receive input from the ipsilateral eye?

ipsilateral - same side

Layers 2, 3, 5

Magnocellular (M)

• in layers 1-2

• likes motion, rapid temporal canges, coarse spatial information

Parvocellular (P)

• layers 3-6

• likes fine detail and color

• Fovea has many of these

• smaller in size than magnocells

Koniocellular (K)

thin layers between the six main layers

Optic radiations

• bundle of white matter axons connecting the LGN to the primary visual cortex

Meyer’s loop

• travels through the temporal lobe and carries upper-visual-field information

• the parietal toute carries loser-visual-field information

Lesion in the optic nerve causes

vision loss in one eye

Lesion in the optic chiasm causes

bitemporal hemianopia - loss of outer halves of both visual fields

lesion to the optic tract causes

contralateral homonymous hemianopia - damage to opposite side of brain, loss of vision in the same half of the visual field for both eyes

Lesion to the temporal optic radiations/Meyer’s loop

contralateral upper quadrantanopia - pie in the sky

Lesion to the parietal optic radiations

contralateral lower quadrantanopia

Lesion to V1

contralateral visual-field loss, extent depends on lesion

receptive field

region of space in which a stimulus changes a neuron’s response

V1

processes early features: orientation, contrast, spatial frequency, and location

retinotopic mapping

nearby locations in the visual field map to nearby locations in visual cortex

fovea and nearby central vision receive a disproportionately large V1 representation (cortical magnification)

fMRI mapping expanding ring stimulus and rotating wedge stimulus

outward ring mapping and angular position around fixation. What is on the retina corresponds with activity of fMRI mapping

orientation columns

neighboring neurons prefer similar orientations

ocular dominance columns

groups of neurons preferential to one eye

eye inputs remain separate in parts of layer 4 and combine in other cortical layers

hypercolumn

block of neurons in V1 that processes orientation, color, and both eye inputs for a small region of the visual field

not all neurons in a hypercolumn encode the same features

blobs

neurons that are color selective, not orientation

extrastriate cortex

region of the occipital cortex directly outside of striate cortex

Dorsal stream

• where - spatial properties, motion-guided action (grasping for things)

• projects toward parietal cortex

ventral stream

what - object identity and recognition

projects toward temporal cortex

supports recognition of objects, faces, and categories

MT/V5

Area V5 (also known as MT or the middle temporal area ) is a region of the extrastriate visual cortex specialized in visual motion perception (direction, speed, and depth)

Gets information from V1

Lateral Occipital Complex (LOC)

object shape and form, responses to complex objects

FFA (fusiform face areas)

face processing

PPA (Parahippocampal Place Area)

activates strongly to locations and buildings

receptive fields tend to become ___ larger in later stages of visual processing

larger

ventral areas emphasize

foveal vision and recognition-related features

Avoid assuming every neuron in a stream has identical properties

dorsal areas often emphasize

spatial properties relevant to action

Avoid assuming every neuron in a stream has identical properties

apperceptive agnosia

difficulty building an accurate visual representation of a form

ex: difficulty coping or matching an object

involved perceptual-form processing

associative agnosia

difficulty connecting a visual representation to a name or explain what the object does

ex: can copy an object but cannot identify it visually

involves recognition/association

prosopagnosia

difficulty recognizing familiar faces

optic ataxia

difficulty using vision to guide action/motion

ex: can see an object, but difficulty reaching for it

visual agnosia

impaired access to object identity/knowledge despite preserved basic vision

Tong et al. (1998) findings

during binocular rivalry between face and house images, fMRI activity in FFA and PPA tracked the participant’s reported percept more than the unchanging retinal input

supports a relationship between neural activity and conscious visual perception

Gauthier et al. (2000)

the FFA responded to trained visual categories and expertise effects

supporting the idea that experience can influence fusiform responses

study did not prove that FFA processes only faces

Chang and Tsao

monkey face-patch neurons represented dimensions of face appearance

findings support a population-based code rather than requiring a “grandma neuron” for every individual face

motion aftereffect

prolonged viewing of motion in one direction can make stationary objects move in the opposite direction

imaging work found MT activity related to the aftereffect, consistent with MT’s role in motion perception

blindsight

some people with V1 damage respond to visual stimuli without visual awareness

upper-left visual field loss poiniwer to

the right hemisphere, below the calcarine sulcus

lower-left visual-field loss points to

the right hemisphere, above the calcarine sulcus

LGN neurons receptive field are combined in the simple cell in V1. LGN neurons project to simple

simple neurons activation timing of receptive field gives complex cells motion

quadranopsia

quadrant cannot be seen

hemianopsia

low vision of half visual field

scotoma

small part of visual field is lost

What causes Motion after effect?

• perceptive system is designed to find changes

• the illusion shows the adaptation

• neurons and their preferences calibrate during an experience

[ Neurons preferring LEFT ] 🔄 Balanced Baseline 🔄 [ Neurons preferring RIGHT ] → Stare at continuous LEFT motion → [ Neurons preferring LEFT ] 📉 Become Fatigued 📉 [ Neurons preferring RIGHT ] → Look at a STATIONARY object → [ Neurons preferring LEFT ] (Reduced Output) < [ Neurons preferring RIGHT ] (Normal Baseline) → 🧠 Brain perceives RIGHT motion!

lesion to the parietal results in

struggling spatial task (are they in the same location)

fine in object task

lesion in temporal results in

struggling in object task (are they the same 3 objects)

fine in spatial task

ventral stream

receptive fields get larger as you move from V1 cortex to TE

receptive fields include more of foveal region than dorsal

neurons become sensitive to more complex objects further along stream

some neurons in TE response selectively to faces

some neurons in TE are invariant to changes in position, orientation, size, etx

mechanism of object recognition

lines and colors represented by V1 neurons are assembled by higher visual areas into recognizable objects

reccognizing objects is based on testing hypothesis. Looping in the brain. High level requires multiple brain areas to understand what you’re looking at

binding problem

how are the images put back together

How do you decide what is a part of the same object

binding criteria

nearness of lines

nearness of color

coherent motion

experience

hierarchical coding of an object

features (sides) → conjunction of features (corners) → component shapes → gnostic cell (the cell for recognition)

prosopagnosia

impairs a person's ability to recognize familiar faces, despite having normal vision

superior temporal sulcus (STS)

sees where people are looking and looks for biological motion

dorsal stream

receptive fields get larger as you move from V1 cortex to PPC

receptive fields include less of foveal region than ventral

neurons are sensitive to visual motion of a particular direction

neurons in dorsal stream are used to guide motor behaior

some neurons in PPC are invariant to changes stimulus features but code for position