PSY280 - 4. VISION IN THE CORTEX Flashcards
in the light
packets of light hit retinal - isomerization - retinal changes shape to detach to opsin
once rhodopsin is activated - cGMP becomes GMP
no cGMP means sodium channels shut
more K moves out driving potential down -70mV
thus in the light release more neurotransmitter
in the dark
CGMP: keep doors open, sodium rushes in cell, K+ channels open, K moves out
trading ions to return them, creating a dark current, -40mV in the dark which is slightly depolarized
it releases a lot of neurotransmitter
On + Off Bipolar
On bipolar: in light less neurotransmitter - spontaneously releases more in the absence of light - stimulating ganglion cell to fire more
product of strength of activation - inhibition
inhibition lateral is proportional to excitation
doesn’t reflect diff in illumination - enunciates the shift in contrast
On + Off Bipolar
lights are off - photoreceptors releasing a lot of neurotransmitter
off bipolar cells release a lot + excite ganglion cells
now it’s opposite
mirror curve - reversed
these 2 curves do not cancel each other out
at the level of cortex, they see it as a signal that there is a change in light intensity
On + Off Bipolar
Lateral connections via horizontal & amacrine cells send information to nearby bipolar & RGCs, changing the firing rate of the RGCs under non- uniform lighting conditions.
in uniform light you activate centre surround cells which cancel each other out
boost contrast signals
On + Off Bipolar
horizontal cells connecting lateral cells creating lateral inhibition
centre shine = active ganglion cell
surround = less activation on ganglion cell
whether the bipolar cells are on or off dictate whether its centre surround
retinotopic mapping & cortical magnification
Activity on the retina causes activity in the cortex.
There is disproportionately more cortex dedicated processing information from the fovea.
info adjacent on retina is processed in adjacent neurons in the cortex
retinotopic mapping & cortical magnification
beside each other in visual field, on the retina, processed by neurons in V1 that are also adjacent
cortical magnification” more cortex processing fovea
necessary for tasks that require high visual acuity - reading, object recognition
Vernier acuity & grating judgments
vernier acuity: high acuity, can sense smallest diff
how diff does reference have to be just to notice the diff
grating
both use JNDs to estimate visual acuity in participants.
• Spatial organization
way stimuli at specific locations in the environment are represented by activity at specific locations in the nervous system
Electronic map of V1
Points on the retinal image cause activity in the cortex
Retinotopic map: electronic map of the retina where locations on the cortex correspond to locations in the retina
Points that are close together on the object and on the right noble activate neurons that are close together in the brain
Electronic map of V1
More space being allocated two locations near the fovea then to locations in the peripheral retina
Cortical magnification: apportioning to the small phobia of a large area on the cortex
Extra cortical space allotted to letters and words at which person is looking provides extra neural processing needed to accomplish tasks such as reading
Visual angle
amount of space something takes up on the retina
depends on how far away it is
visual angle is absolute measure of how much space is taken up on the retina
mouse + elephant take up same VA
60 minutes of arc to 1 visual angle
cortical magnification
estimate of cortical magnification, researchers presented spots of light of varying visual angles & measured the size of the fMRI response.
how much of visual cortex activated by spot of light
1 degree VA + 1mm = 1:1
1 degree VA + 12 mm = 1:12
ratio = M for magnification
cortical magnification
higher participant’s visual acuity, the greater their magnitude of cortical magnification
low acuity: 12 minutes of arc necessary
high acuity: small diff in minutes can still reliably detect diff
negative relationship in both tasks
columns in V1 (striate cortex)
Neurons perpendicular to the surface of the cortex have RFs at approximately the same location on the retina.
highly systemic organized of preference neurons
receptive field locations on retina are same location
receptive fields are one on top of each other perpendicular of cortex
columns in V1 (striate cortex)
neurons perpendicular are interested not only to same location, but also stimuli of same orientation
move just a little bit, preference changes just a little bit
Location and orientation columns
Striate cortex is organized into location columns that are perpendicular to the surface of the cortex so that all of the neurons within a location call them have their receptive fields at the same location on the retina
Cortex is organized into orientation columns, each column containing cells that correspond best to a particular orientation
One location: many orientation columns
Neurons in that location: receive signals from a particular location on the retina, which correspond to a small area in the visual field
Hyper column: Receives info about all possible orientations that fall within a small area of the retina and therefore well-suited for processing info from a small area in the visual field
columns in V1 (striate cortex)
cut across the orientation columns, they found orientation preference changed systematically
systematic variation across orientations along the parallel of the cortex surface
Ocular dominance
neurons in striate cortex will respond to stimulation from either eye, but tend to prefer one eye over the other
LGN are eye specific, but in V1, neurons respond to both eyes
however, they tend to have a preference
ocular dominance varies specifically as well
Hypercolumns
pair of ocular dominance columns in conjunction with one complete series of orientation columns.
columns of neurons that prefer info from the right eye beside columns that prefer left eye
respond more strongly to preference
Hypercolumns
cubes in V1 that are 1/4 mm x 1/2 mm
ice cube tray model
retina organization then will also be the same
columns prefer a certain area of the visual field
Tilting
working together these columns cover the entire visual field
Any scene is presented in the striate cortex by amazingly complex pattern of firing
Hypercolumns
hypercolumns in cortical region on the fovea have a small receptive field
large receptive fields give sensitivity not detail
small receptive field means you need more (hypercolumns) cortex which explains cortical magnification
midget cells receiving info from fovea cones to details (stationary, colour contrast) all of which importnant for object identification
Parallel visual channels
Pathways from LGN to V1 stay segregated
compartementalization allows brain to be more efficient due to specialization
parallel visual channels form basis for seperation of duties at early stage of visual processing
Magnocellular
retinotopic emphasis: peripheral receptive field size: large contrast sensitivity: high color sensitivity: low spatial resolution: low temporal resolution: high motion sensitivity: high neural transmission: rapid
Parvocellular
retinotopic emphasis: central receptive field size: large contrast sensitivity: low color sensitivity: high spatial resolution: high temporal resolution: low motion sensitivity: low neural transmission: sluggish
extrastriate cortex
borders V1 & is also important for visual processing:
Broadmann’s Area 18 & 19
broadmann identified regions as distinct
association areas
Using electrophysiology, behavior & lesion studies, researchers determined these pathways extend beyond the striate cortex into the association areas.
areas of cortex that connect visual info with meaning
Monkey Cortex
Magnocelluar - V1 - area MT Parvocellular - V1 - V4
neurons in this MT region prefer object motion + movement detail
v4 - object form, colour, textures - variables important for identifying objects
V5 in humans instead of MT
Dorsal
area MT - parietal lobe
pathway is responsible for determining where in space an object is
Ventral
V4 - temporal lobe
pathway responsible for determining the identity of an object.
Where pathway
halfway leading from the striate cortex to the parietal lobe
the upper part of the brain, where the parietal lobe is located is the dorsal part of the brain
Ventral pathway [what]
lower part of the brain where Temporel lobe is located is ventral part of the brain
pathway leading from the striate cortex to the Temporel lobe
pathways
highly interactive
segregation done in parallel with cortex
dorsal - body + eye movements
dependent on visual input based on magno function
but what object is guides how you should act on it
these provide top down processes
pathways
Properties of the ventral and dorsal streams are established by two different types of ganglion cells in the retina which transmit signals to different layers of the LGN
processed in the what pathway, not where pathway
dorsal stream provides info about how to direct action with regard to stimulus
Neurons in the parietal cortex that respond when a monkey look set an object and when it reaches toward the object
Ablation
remove it and see how the system now functions.
refers to distraction or removal of tissue in the nervous system
Landmark description problem
goal in this task is to remove food well cover that is closest to the cylinder.
parietal lobe removed had difficulty solving the landmark discrimination problem
Object discrimination problem
shown one object and was been presented with two choice task which included the target and another stimulus
part of the Temporel lobe, object discrimination problem was very difficult
Neuropsychology
researchers look for dissociations in functional deficits.
naturally occurring lesions understand cognitive function of those areas
pairs of patients with opposing deficits in function
study of behavioral effects of brain damage in humans
Feedback
backward flow of info provides information from higher centers that can influence the signals blowing into the system
double dissociations
Involve two people: in one person, damage to one area of the brain causes function A to be absent while function B is present, and the other person damage to B causes function B to be absent function a is present
orientation distinguishes the identity of these 2 shapes
processed in the what pathway, not where pathway
D.F
DF couldn’t match the orientation* of the card to the slot.
D.F Carbon monoxide poisoning - damage to temporal lobe
D.F has difficulty finding orientation of slot
Dorsal pathway should be how pathway or the action pathway because it determines how a person carries out an option
D.F
she can’t match the orientation, but she can push the card in the slot
action pathway was preserved, but her perception pathway was compromised.
dorsal stream is for taking action
provide info on how to interact with object
interaction betw what object is + how to interact with it
Modularity: structures for faces, places and bodies
structure specialized to process info about a particular type of stimulus
Specific areas in the Temporel lobe respond best to particular types of stimuli
Behavior of people without brain damage
Length estimation task: indicate how perceive the lines length by spreading their thumb and index finger
Involves the ventral what stream
Grasping task: reach toward the lines and grasp each line bites and
Involves the dorsal how stream
Perception + action are served by different mechanisms
In the length estimation task, subjects judged 1 was looking shorter than line 2 but in the grasping task they separated their fingers farther apart
action stream deficits
perceptual streams are preserved.
difficulty acting on object because action stream is damaged
Face neurons in the Monkeys IT cortex
Inferotemporal cortex
many neurons that respond it best to faces
The areas in the monkey Temporel lobe that are particularly rich in these neurons
rod & frame illusion
is it just the action/perception stream fooled
Using the same stimuli, researchers had participants do a action task & a perceptual task.
grasping - change the rod position to match stimuli
matching task - say left or right to match
we assume visual illusions occur in what pathway
rod & frame illusion
The results of this experiment were consistent with Milner & Goodale, but recent evidence suggests there is still room for Ungerleider & Mishkin.
illusion is huge in matching task
magnitude of grasping task illusion is smal
action pathway seperate from basic perception?
diff parts of parietal in knowing where things are in space to act + just plain knowing where it is
inferior temporal cortex
visual areas in temporal lobe
posterior IT - early complex representations with specific configurations
central IT + anterior IT}late complex representation
back half close to occipital: combos of line + basic shape
central + anterior IT - more toward front - more complicated, doesn’t matter if its upside down
anterior IT
Neurons in the
anterior IT
cortex in monkeys show a preference for faces.
neurons in this area that prefer face
fusiform face area.
Broadmann’s area 37 corresponds to human analog
betw parahippocampal + entorhinal - respond to various representations of ppl
neurons in middle area activate when they recognize her, but also to recall who she is
• Fusiform face area
located in the fusiform gyrus’s on the underside of the brain directly below the inferotemporel cortex
• Parahippocampal place area
activated by pictures depicting indoor and outdoor scenes
Important for spatial layout because increased activation occurs both the empty rooms into rooms that are completely furnished
Extrastriate body area
activated by pictures of bodies and parts of bodies
The stimuli also activate other areas of the brain as well
Fusiform face area
Object such as faces may cause a large focus of activity in the FFA but also cause additional activity that is distributed over wide area of the cortex
Visual system is organized functionally with different streams for what and where/how it with specific cortical areas that are rich in neurons that respond to specific types of stimulus such as faces, places and bodies
Fusiform face area
betw parahippocampal + entorhinal - respond to various representations of ppl
neurons in middle area activate when they recognize her, but also to recall who she is
Broadmann’s area 37 corresponds to the human analog
Modules
While there is some specialization through modules, categorical stimuli still activate other areas of the brain.
areas of specialization can be misleading
various stimuli produce activation in various parts of the brain
lots of overlap activation for face, house, chairs
there is distribution of representation
Face recognition
holistic process
we don’t recognize indiv elements of person + assembleing them later
it needs to be in a particular configuration
relationship of features
configuration: distance betw eyes, + their distance to middle of mouth
similar configuration adds to their similarity
FFA
processing holistic + expertly info, we are just all face recognition experts
Novices tend to focus on one aspect of a stimulus for discrimination.
Experts tend to focus on the whole stimulus for discrimination
Greebles
features vary in a familiar way; holistic processing is required to discriminate individuals.
forcing ppl to rely on holistic processing
trained ppl to be greeble experts
got really good at distinguishing them
Greebles
FFA responded to greebles once participants became experts
after training nearly identical response rates to both greebles + faces
3 bifurcations
3 splits in processing
retina - Sc, most to LGN
LGN - MAGNO + PARVO - V1
V1 - dorsal, ventral
