Vision and Eye Movements
2012-02-02
Stanford
“All these topics are covered in the book.”
“You have an upside down action-potential.”
“The permeability for different ions never goes to zero.”
“Just memorize the anatomy and understand the concepts.”
“Don’t memorize everything in the book.”
“Perception is an interpretation of reality.”
“This is the mantra driving cognitive neuroscience today.”
“There is no good theory for how we see colors.”
“These tables are the same size, but through experience we don’t see them that way.”
“We’ve learned through experience that those are two shapes because of the way space distorts things.”
Hyperacuity is our ability to see better somehow than our retina can tell us through the resolution it has.
“We get better at this through time.”
“Vernier acuity continues to improve late in life.”
Your ability to detect faces improves with age.
“Perception is an active process.”
“It’s not too hard to get a picture of the cones in the retina.”
The camera solution is very costly in terms of energy.
Dim objects are easer to see from an angle.
To see something clearly in light you have to project it on the fovea.
fovea is Latin for pit
three sets of eye muscles: up/down. in/out, cw/ccw
“We look a lot at the eyes, mouth and nose to figure out what a person looks like.”
Eye movements
stabilizing movements (smooth pursuit)
vestibulo-ocular reflex (head movement reflex)
opto-kinetic nystagmus (watching trees in the car)
orienting movements
“These have been extremely well studied in neuroscience for a long time.”
saccades
rapid (<100ms)
blind during movements
unaware of blurred input
compensation in the brain
brain edits the input
superior colliculous
"Very well studied."
"It's easy to study eye movements."
rostral, back to front
cottal, front to back
"How do we generate this visual scene?"
top-down (goal driven)
eg delayed saccade task
bottom-up (stimulus-driven)
eg reward
(beautiful people)
Activity scales with expected reward.
There is accumulating reward information, a tickle.
saliency map
We want to look for important things in our world.
We only see a very small part of the world in high resolution.
drawn to different information depending on scale
Perceptual consequences
control of eye movement is controlled by superior colliculous, common output structure
necessary to view scene with detail
movements guided by salience, top-down, bottom-up
"We don't really understand how salience is determined."
We are blind to visual input that are not sufficiently salient.