central visual pathways Flashcards
visual pathway from retina to cortex
retina –> thalamus (LGN) –> visual cortex
how does the brain get visual signals
the retina transmits
synaptic transduciton through retina
photoreceptors detect photons –> bipolar cells –> ganglion cells
what is the output neuron of the retina
ganglion cells
projections of the visual field onto retina
projects backwards and upside down
brain has to flip back
binocular field of view
left visual hemi-field is “seen” by right visual cortex
- projection of binocular field of view relates to crossing of fibers in otptic chiasm
- visual signals from each retina cross over
thalamus
inputs from 2 eyes are relayed to visual cortex via dorso-lateral geniculate nucleus
how are inoputs from 2 eyes segregated
cell specific manner in each layer of LGN in primates/humans
- maintains organization even at thalamus
(contra/nasal: 1,4,6 and ipsi/temporal: 2,3,5)
layers of thalamus
magnocellular: 1,2
parvocellular: 3,4,5,6
contralateral: 1.4.6
ipsilateral: 2,3,5
mouse LGN inputs
infected with AAV to back trace inputs to thalamus
- different types of neurons all tend to come from same visual area
optic radiations course to visual cortex
lateral geniculate nucleus –> meyers loop: superior and inferior retinal quadrants
retinotopic organization in right occipital lobe
top of visual cortex to inferior visual field (superior retina)
- no integration of both eyes in V1
overrepresentation of inputs from fovea- high density of photoreceptors
layer 4 of visual cortex (unsure)
magno, parvo and konicellular pathways remain seperated
in layer 4 of visual cortex
columnar organization of ocular dominance
as you move across cortex there are specific columns of neurons that respond only to left or right alternating
- inputs from l/r eye are still segregated in layer 4
primary visual cortex neurons
neurons in V1 respond only to oriented edges
- have different orientation preferences
- fire most at certain length and orientation
organization of V1: orderly progression of columnar response
IV: left ear/ right eye (ocular dominance columns)
blobs
orientation columns
how are orientation units formed
LGN cells: linear alignment with overlapping receptive field responds best to bar extend over fields
summed output from LGN cells each with center surround organizations results in orientation selectivity
–> V1 cell
outputs of striate cortex
1,2,3,4a: extrastriate cortex
5: superior colliculus
6: LGN
inputs: mostly layer 4
extra striate cortical areas
IT: inferior temporal cortex
receives input through cascade, complicated visual info
MT: maintaining retinal stability
what do neurons in higher visual areas respond best to
synthesize random images and record neuorn, see what neuron responds best to adn combine images and mutate and test again til you get evolving image to see if it iis better or not until you get best stimulus
- neurons can represent high order features
neural network
algorithms that take inputs (image) and represent using fewer but higher order features of it by filtering image in diff ways and assignming weights to filters so you have fewer features but they are more imformative
- each layer gets more abstract
