The Retina II (Lecture 6)
Photoreceptors representing lighter regions ___________________ lateral inhibition
Photoreceptors representing lighter regions ____generate more___ lateral inhibition
What does contralateral mean with respect to vision?
Each hemisphere of the brain responds to the opposite side of the visual field
The dorsal stream processes __________
The ventral stream processes ___________
The dorsal stream processes spatial and action-related information--the "where" stream
The ventral stream processes object information--the "what" stream
What are 2 reasons that it is hard to tell what we're looking at?
•Objects can be partially hidden by other objects (occlusion)
•Objects can be in different orientations
•Objects can be seen in different lighting and shadow
What is the term for figuring out what is the object versus background?
Figure-ground segregation
__________ allow photoreceptors to send signals to each other.
__________ allow ganglion cells to send signals to each other.
Horizontal cells
Amacrine cells
Name 2 conditions that could be caused by a stroke to area V1 in the right hemisphere, and which portion of the visual field is affected
Homonymous hemianopia: half of the visual field missing. In this case, it would be left visual field
Homonymous quadrantanopia: 1/4 of the visual field missing. In this case, it would be either top left or bottom left visual field
Name the region primarily responsible for object recognition, and the larger region it is a part of
Lateral occipital complex
Inferotemporal cortex (within ventral stream)
Describe the 2 competing theories of how we resolve ambiguity
•Structuralism: our perception results from just the adding up of visual information. The whole is the sum of its parts; we don’t make any inferences beyond what we see
•Gestalt psychology: the whole is different than the sum of its parts. We “add something” to the basic visual information to create our perception
Name 3 ways in which objects differ from scenes
Object: a single entity, compact, acted upon
Scene: both objects and background, extended, acted within
Describe 3 properties of infant vision and/or photoreceptors.
Infant cones are shorter and have less surface area for visual pigments
Infants' central vision is poor, but their peripheral vision is adult-like
Name 3 steps of the path of the visual signal after it leaves the eye, in the correct order
Optic nerve
Optic chiasm
Thalamus (most of the signal)*
Area V1
Ventral and dorsal streams
*The other 10% goes to superior colliculus
Describe the brain region and the disorder related to understanding written words
Visual word form area
Alexia
Describe 3 pieces of evidence for the gestalt view
Apparent movement: Different objects moving in a sequence produce the illusion that one object is moving (like animation)
Illusory contours: perceived edges that do not physically exist
The fact that your brain fills in the blind spot
What is scene gist?
What does it imply for theories of how we analyze scenes?
Scene gist: the general description or basic properties of a scene. typically includes information about:
•The scene category
•What the scene is generally about (its meaning)
•The general layout and spatial structure
Implies that scene recognition does not depend on object recognition
How do neural convergence and spatial summation differ between the fovea and the rest of the retina?
What is the consequence of this for light sensitivity and resolution?
Neural convergence is lower in the fovea, meaning spatial summation is lower in the fovea
Fovea is less light sensitive but higher resolution
Define a spatial receptive field, and list 2 other types of receptive fields
Spatial receptive field: Where a stimulus needs to be in the visual field to cause a given neuron to change its activity
and: the area of the retina that, when stimulated, causes that neuron to change its response.
Other types of receptive fields: orientation, motion, size
What area processes faces?
What disorder results from damage to this area?
How do we process faces?
What is the alternative hypothesis for this area?
Fusiform face area
Prosopagnosia
Holistically (processed as a whole, not a collection of features)
Expertise hypothesis: the FFA actually processes anything we have a lot of expertise for
Describe the 2 core issues our visual system needs to solve to resolve ambiguity
1. Many different objects can generate the same image on the retina
And on the flip side:
2. The same object can generate very different images on the retina
Name 2 examples of physical regularities
Name 2 examples of semantic regularities
Define a scene schema
Physical regularities: Light from above assumption; more horizontal and vertical lines in the world than other orientations
Semantic regularities: gravity keeps objects down, an oven is in a kitchen
Scene schema: a collection of semantic regularities; what a specific type of scene typically contains, and where elements are placed
If a light is shined on a visual receptor, what happens to nearby receptors?
What is this phenomenon called?
What does it do for vision?
If a light is shined on a visual receptor, nearby receptors decrease neural firing.
This is called lateral inhibition
It enhances vision for edges and increases contrast
According to the retinotopic map, if objects are close together in your visual field: _________________________
What does cortical magnification refer to?
The neurons that represent them are close together in Area V1
Cortical magnification: Information in foveal vision is way over-represented in terms of how much space it takes up on the retinotopic map
Name 2 types of information processed by the dorsal stream
Name 2 types of information processed by ventral stream
Describe which parts of the brain they go to
Dorsal: spatial processing, location, movement, spatial transformations, spatial relations.
- Parietal lobe
Ventral: object processing, color, texture, pictorial detail, shape, size.
- Temporal lobe
Describe 5 gestalt principles
- Proximity: Things that are near each other appear to be grouped together
- Similarity: Things that are similar appear to be grouped together
- Good continuation: figures with edges that are smooth are more likely seen as continuous than edges that have abrupt or sharp angles. Also, objects that are partially covered by other objects are seen as continuing behind the covering object
- Pragnanz: (principle of simplicity): Things are seen in a way such that the resulting structure is as simple as possible
- Common fate: Things that are moving in the same direction appear to be grouped together
- Common region: Elements within the same region of space appear to be grouped together
- Uniform connectedness: A connected region of the same visual properties is perceived as a single unit
Describe the 3 components of Bayesian inference, and the goal of Bayesian inference in visual perception
Components:
1. Prior: our initial estimate of the probability of an outcome (pre-existing beliefs/knowledge)
2. Likelihood: the extent to which the current available evidence is consistent with an outcome (current evidence)
3. Inference: The result of combining the prior and the likelihood; what we infer
Goal: resolve ambiguity in what we're looking at