What are the two problems with perception?
1. Sensory input does not contain enough information to explain our perception, and 2. The world presents us with too much sensory input to include in our coherent perceptions at any given moment
What is the goal of perception?
To take in information about the world and make sense of it
What does vision do?
It tells us what is out there, and where it is
What is the retina?
light-sensitive tissue lining the inner surface of the eye, made up of photoreceptors
What is the layer of cells that respond to light?
What are photoreceptors?
cells that respond to light
What are cells that respond to light?
What do nerve cells at the back of each eye do?
Convert light into electrochemical signals
What converts light into electrochemical signals?
nerve cells at the back of each eye
How is light transmitted to the brain?
Converted into electrochemical signals by nerve cells at the back of each eye, and then sent via optic nerves to the brain
What do the optic nerves do?
Take light in the form of electrochemical signals to the brain
What takes light in the form of electrochemical signals to the brain?
What are the optic nerves made up of?
Bundle of long axon fibers of the ganglion cells in the retina
What is the bundle of long axon fibers of the ganglion cells in the retina?
What do the axons of the optic nerves make contact with?
The neurons of the lateral geniculate nucleus
What do the neurons of the lateral geniculate nucleus make contact with?
The axons of the optic nerves
What is V1?
The primary visual cortex
What is the primary visual cortex called?
What do the axons of the lateral geniculate nucleus do?
Send signals up to the primary visual cortex
What sends signals up to the primary visual cortex?
Axons of the LGN
What is the LGN?
lateral geniculate nucleus
Describe how light gets to the brain.
Patterns of light intensity, edges, and other features form image on retina. Converted to electrochemical signals and sent via optic nerves to the LGN neurons, which send signals up to the primary visual cortex
What two main pathways go beyond the primary visual cortex?
The dorsal pathway and the ventral pathway
Where does the dorsal pathway go?
Reaches up into the parietal lobes
Which pathway goes to the parietal lobes?
The dorsal pathway
Why is the dorsal pathway important?
Helps process information about where items are located and how they might be acted upon, guiding movements
Which pathway goes to the temporal lobes?
The ventral pathway
Where does the ventral pathway go?
The temporal lobes
What is important about the ventral pathway?
Processes information that leads to the recognition and identification of objects
Which pathway is the "what" pathway?
The ventral pathway
Which pathway is the "where" pathway?
The dorsal pathway
Which pathway processes information that leads to the recognition and identification of objects?
The ventral pathway
Which pathway processes information about where items are located and how they might be acted upon?
The dorsal pathway
What does it mean that visual areas have reciprocal connections?
Most visual areas that send output to another area also receive input from that area
What is visual perception a product of?
Top-down and bottom-up processes
What is the product of top-down and bottom-up processes?
What are perceptions?
Interpretations of what we see, representations produced by the interaction of bottom-up and top-down processing
What 2 things are our interpretations of the world around us determined by?
1) Biological structure of our brains, and 2) experience, which modifies that structure
How are visual experiences formed as babies?
As a baby sees something over and over, the lingering memory of each occurrence combines with each new instance, accumulating to form lasting mental representations of people, places, and things. These representations form the basis for subsequent recognition of objects
How does the characteristics of the infants' environment at particular times influence adult capabilities?
If exposure to the natural environment is limited during the designated critical period for a given response, the animal will fail to develop that ability properly. However, the introduction of some prenatal stimulation can impair normal perceptual development later in life as well
What is the critical period for sensory processing?
Different aspects of sensory processing have different critical periods
What does it mean when it is said that there is competition in sensory input?
Different sources and different modalities of sensory input seem to compete for representation in cortex
What happens when one eye is more active than the other?
Cortical resources are redeployed in that direction, and once assigned in infancy, such resources are not easily modified later
What are visual features?
Spots, edges, colors and shapes, movements and textures. The building blocks of perception
What are spots, edges, colors and shapes, movements and textures known as?
What are the building blocks of perception?
What do photoreceptor cells do?
Convert light energy reflected from various objects in the physical world into an electrochemical signal that can travel through the nervous system
What is the visual field?
the portion of the world that is visible at the given moment
What is the receptive field?
the region of the visual field in which a stimulus will affect the activity of the cell
What is the portion of the world that is visible at the given moment?
What is the region of the visual field in which a stimulus will affect the activity of the cell?
What is the ganglion cell connected to?
When will the ganglion cell respond to light?
Only when light lands on the photoreceptor cells that the ganglion cell is connected to, or the light in one specific region of the visual field
Where does the ganglion cell get its input from?
The receptors that lie in its receptive field
How does the connection from photoreceptors to the ganglion cell vary?
Light in some portions of the receptive field will excite the cell, whereas light elsewhere will inhibit the ganglion cell
Where does input excite the ganglion cell?
In the central (white) zone of the receptive field
Where does input inhibit the ganglion cell?
In the surrounding (gray) region of the receptive field
What does input to the surrounding(gray) region of the receptive field do to the ganglion cell?
What does input to the central (white) zone of the receptive field do to the ganglion cell?
What happens when the central region of the receptive field is stimulated by a gray area?
Won't be very excited
What happens when the entire receptive field is 100% bright?
The strong excitation of the center would be offset by the strong inhibition from the surrounding
When is the ganglion cell maximally excited?
When a bright spot the size of the central region falls on the central region
What happens when center-surround receptive fields receive inputs across an edge?
One receptive field is mostly on the light side, and another mostly on the dark side. Makes the former a bit lighter and the other "darker-than-dark," enhancing the contrast
What are Mach Bands?
Gray areas are uniform, but each lighter bar appears a bit lighter on the right side, and each darker bar appears a bit darker on the corresponding left side
Why is the visual system tuned to pick out edges?
edge information is important for defining the shape of objects and providing cues for where to direct action
Why are we tricked by the Mach Bands?
Not an unreasonable assumption to believe that image lighter on light side of edge than on darker side. Looks for edges because edge information is important for defining the shape of objects and providing cues for where to direct action
What does throwing out information about intensity of light at every point in space show?
Visual perception efficiently extracts visual features by ignoring some data
What is the fovea?
The part of the retina that has a high concentration of ganglion cells with tiny receptive fields
What is the part of the retina that has a high concentration of ganglion cells with tiny receptive fields?
What is the result of having a fovea?
The area is capable of high resolution, and fine details can be perceived
Where are the finest details perceived in the eye?
How do receptive fields change according to distance from the point of fixation?
The receptive fields get bigger and bigger the farther away you get
What do the optic nerves form?
The optic chiasm
What is the optic chiasm?
The point at which fibers from each optic nerve cross to opposite hemispheres of the brain
What is the point at which fibers from each optic nerve cross to opposite hemispheres of the brain called?
What do cells in the primary visual cortex respond to?
Variations in basic features such as orientation, movement and color
What responds to variations in basic features such as orientation, movement and color?
Cells in the primary visual cortex
What is the striate cortex?
The primary visual cortex
What is the extrastriate cortex?
A collection of visual areas that output from V1 feeds into
What is the collection of visual areas that output from V1 feeds into called?
The extrastriate cortex
What does the extrastriate cortex contain?
Areas whose cells appear to be specialized for the further processing of these basic features and of more elaborate representations, such as faces
What is the visual cortex divided into?
What are hypercolumns made up of?
Made up of cells that are activated by stimuli in one small part of the visual field
What are cells in the hypercolumns activated by?
activated by stimuli in one small part of the visual field. Cells in the next hypercolumn will respond to input from a neighboring portion of space
What things do hypercolumns focus on?
More hypercolumns focused on detailed processing of input to the fovea than to cruder processing of more peripheral parts of the visual information
How are hypercolumns organized?
Cells are ordered by their sensitivity to specific aspects of the visual feature, such as edges at a specific orientation
If a cell within a hypercolumn sensitive to edge orientation responds the most to vertical lines, what will the other cells be like?
The next cell over will respond the most to lines tilted a bit more off vertical, and the next one to those even more tilted
How are precise assessments of orientation made?
Made by comparing activity across a population of neurons
What does it mean when it is said that precise assessments of orientation are made by comparing activity across a population of neurons?
If there are neurons that respond to tilts 15 degrees to the left and neurons that respond to tilts 15 degrees to the right, a line that is perceived as vertical would be one that stimulates both of these populations of neurons equally
What is the tilt aftereffect?
How do we know that precise assessments of orientation are made by comparing activity across a population of neurons?
If we fixate our gaze on a pattern of lines with the same right tilt, this tires out the neurons specialized for that. Thus, when gaze is switched to a vertical line, it will appear to be tilting to the left because there is more activity in the left neurons. The right neurons are too fatigued to sustain equal activity
How is the value of a particular feature in vision determined?
Determined by comparison between two or more sets of neurons--with different sensitivities--responding to that stimulus
Where is motion detected?
What does V5 detect?
What do the cells in V5 detect?
Objects moving in a particular direction
What detects objects moving in a particular direction
cells in area V5
What does damage to area V5 result in?
What is akinetopsia a result of?
Damage to area V5
What is akinetopsia?
The loss of the ability to see objects move
What is the loss of the ability to see objects move?
What does V4 perceive?
What area perceives color?
What is achromatopsia a result of?
Damage to area V4
What does damage to area V4 result in?
What is achromatopsia?
Cortical color blindness
What is cortical color blindness called?
What are the rules that govern the complex process by which we infer the contents of the visual world called?
Gestalt grouping principles
What do the Gestalt grouping principles do?
Guide the visual system and produce our perception of what goes with what
What produces our perception of what goes with what?
Gestalt grouping principles
What are the 5 grouping principles?
Proximity, uniform connectedness, similarity, good continuation, closure
What principle overrides proximity?
Principle of uniform connectedness
What is the principle of proximity?
All things being equal, things that are closer to one another are more likely to be grouped together than things farther apart
Name this principle. Things that are closer to one another are more likely to be grouped together than things farther apart
Principle of proximity
What principle does the principle of uniform connectedness override?
The principle of proximity
What is the principle of colinearity?
Lines group when their orientations are close to that of a neighbor's
What is it called when lines group when their orientations are close to that of a neighbor's?
Principle of colinearity
What does it mean to be occluded?
Why do our brains perceive things that are not actually there (i.e., in the case of occlusion)?
It is unlikely that white rectangles and a black one would align just so to produce the image given Instead, it is more likely that one surface (the black rectangle) might occlude another (the long white one)
What is a subjective contour?
a contour that is not physically present int he stimulus, but is filled in by the visual system
What is an illusory contour?
a contour that is not physically present int he stimulus, but is filled in by the visual system
What is a contour that is not physically present int he stimulus, but is filled in by the visual system called?
subjective (illusory) contour
What is the binding problem?
the question of how we associate different features such as shape, color, and orientation so that we perceive a single object
What is the question of how we associate different features such as shape, color, and orientation so that we perceive a single object called?
What is part of the answer to the binding problem?
Spatial location can serve as the required "glue"
What is recognition?
The process of matching representations of organized sensory input to stored representations in memory
What is the process of matching representations of organized sensory input to stored representations in memory?
What is agnosia?
The inability to recognize objects
What is the inability to recognize objects called?
What is it called when sight is unimpaired, but recognition failed?
What is viewpoint dependence?
An object can be viewed from an infinite combination of possible angles and possible distances, each of which projects a slightly different two-dimensional image on a plane
What two challenges are there in recognizing objects?
Viewpoint dependence and exemplar variation
What is exemplar variation?
There are many different instances of each object category
What does an object category consist of?
Many possible examples
What is the term for many different instances of each object category?
What four models have been proposed for recognition?
1) Template-matching models, 2) Feature-matching models, 3) Recognition-by-components, 4) Configural models
In short, what is the template matching model?
Matches a whole image to a stored representation of a whole object
In short, what is the feature-matching model?
Extracts important or discriminating features from the image and match them with known features of objects
In short, what is the recognition-by-component model?
Represents three-dimensional structure of objects by specifying their parts and the spatial relations among those parts
In short, what are configural models?
Distinguish among objects that share the same basic parts and overall structure by coding each exemplar according to how it deviates from the average or prototypical object
What model distinguishes among objects that share the same basic parts and overall structure by coding each exemplar according to how it deviates from the average or prototypical object
What model represents three-dimensional structure of objects by specifying their parts and the spatial relations among those parts?
What model xtracts important or discriminating features from the image and matches them with known features of objects
What modelmatches a whole image to a stored representation of a whole object?
Template matching model
What is a template?
a pattern that can be used to compare individual items to a standard
What is a pattern that can be used to compare individual items to a standard
When is the template matching model most useful?
The item to be recognized and the template to which the system compares it to are almost identical and different from others
Where do templates fail?
Cannot accommodate variations in object size and orientation
What is the main disadvantage of the template matching model?
Recognition often demands great flexibility, but templates aren't flexible
What is a feature?
Any attribute that distinguishes one object from others
What do feature-matching models do?
Search for simple but characteristic features of an object, and the presence signals a match
How is the feature-matching model more flexible than template-matching models?
As long as the features are present it will work, even if the object has parts that may be rearranged. It also requires less storage space than template models because relatively few features would render recognizable many objects of the same category that are not identical
What sort of processing does the feature-matching model support?
Parallel, distributed processing
How does the feature-matching model work in a neural-net model?
Excitatory connections increase the activity of a unit, inhibitory connections decrease it In a letter-recognition model, units representing different line segments are connected to units in the next level that represent letters. A connection is excitatory if the letter has the feature specified by that line segment, inhibitory if it does not. Some features have no additional features beyond what an A has, but lack some feature of A, so are only partially active. Other letter units will share some of those features and also have another feature, meaning they, too, will only be partially active. Only the representation of the letter that matches all the features will be maximally active, and go on to influence recognition at the next level of the net, where units representing individual letters excite or inhibit units representing words
What does the fact that neurons are selective for an array of different features suggest?
Particular features that are important for recognition may vary with the level of detail required at the moment
When does our use of feature matching rather than template matching depend on?
How difficult it is to see, how closely the object matches canonical pictures. e.g., template matching for a robin, feature-matching for an ostrich
What is the problem with feature-matching models?
They cannot distinguish objects with the same component features, but arranged in a different spatial relationship, like V and X
What is the recognition-by-component model?
Any three-dimensional object can be generally described according to its parts and the spatial relationships among the parts
What components are present in the recognition-by-component model?
24 geometrical, 3-dimensional shapes known as geons
What are geons?
geometrical, 3-dimensional component shapes used in the recognition-by-component model
Which model uses geons?
The recognition-by-component model
Why are geons useful for describing objects?
Their properties are viewpoint invariant
What is the opposite of viewpoint dependent?
What is the opposite of viewpoint invariant?
What uniquely specify geons from each other?
Each geon is associated with a set of viewpoint-invariant properties that uniquely specify it from other geons
What is the evidence for the RBC model?
Recognition is faster when the second image shown has the same geons as the first
What is evidence against the RBC model?
Many neurons fail to generalize across all possible views, and although RBC can account for man-made objects, it cannot be applied as well to natural objects, like plants and animals, and especially faces
Why is the RBC model not good for faces?
The RBC model would construct the same arrangement of geons for every face, and so would not detect individual differences between one face and another. Good at identifying most commonly used categories, but cannot identify a specific exemplar
What are configural models?
Objects that share the same parts and a common structure are recognized according to both the spatial relations among those parts and the extent to which those spatial relations deviate from the prototype object
How do RBC and configural models differ?
Configural models also take into account the extent to which those spatial relations deviate from the prototype object
How are specific faces described in the configural model?
Specific faces are described by their deviations from the prototypical face, as defined by quantified average proportions in a population. All faces have the same component parts in the same spatial arrangement, but their relative sizes and distances make them unique
What support is there for the configural theory of facial recognition?
We are better at recognizing caricatures of famous faces, which accentuate the differences from the average face. Also, when instructed to stare at a particular face and then look at an average face, will briefly experience a visual aftereffect, in which they perceive the "opposite" or "anticaricature" of the original face
What constrains our ability to process faces?
Participants only able to recognize faces in the configural theory if the faces are upright. Cannot if inverted, must do in piecemeal manner
How do we process noses?
Do better at recognizing differences in noses if shown in context of face as compared to in isolation. However, benefit only applies if face upright, not if inverted
What is the evidence for the configural theory of facial recognition in terms of disorders?
People with prosopagnosia can differentiate faces from other objects, but cannot differentiate one face from another
What is the expertise hypothesis?
A specialized neural system develops that allows expert visual discrimination, and is required to judge subtle differences within any particular visual category
What does damage to the fusiform face area result in?
What is prosopagnosia a result of?
Damage to the fusiform face area
What is a contrasting view to the expertise hypothesis?
Many visual representations are spatially distributed throughout the ventral pathway, meaning that sufficient visual information is analyzed outside the fusiform face area to distinguish faces from other objects
What explains why prosopagnosiacs can differentiate faces from other objects, but not faces from other faces?
All ventral areas are involved in object recognition and provide useful information for categorization, but certain distinct systems are necessary for performing fine-tuned discriminations within a category
How do we manage to deal with sensory overload?
Information is interpreted relative to context across all levels of perceptual representation and processing
What is a brightness illusion?
Perceiving an object to have the same color all throughout, even despite changes in illumination across its surface
What is a size illusion?
We assume objects maintain their "true" size across changes in apparent distance from the observer
What is the effect of grouping?
Makes it difficult to see each item independently, but allows us to see common attributes of many items at once, and then perform operations on the group as a whole
What are the context effects for object recognition?
Recognition of an object may be improved if it is seen in an expected or customary context, as compared to an unexpected or inconsistent with previous experience context
What do context effects in recognition reflect?
Reflect the information that is important for and integral to the representation of objects
What is the word superiority effect?
the phenomenon in which a letter is seen better in the context of a word than alone; surrounding letters may suggest a word, and therefore influence the perception of the central letters
What is the phenomenon in which a letter is seen better in the context of a word than alone known as?
word superiority effect
Why does the word superiority effect exist?
The recognition of objects is not strictly a matter of putting together the pieces via bottom-up processing. the whole word is recognized by the combined influence of all the letters, thus supporting the identification of each letter because of its context
What is the face superiority effect?
the finding that people are better able to discriminate between two features of a face when they are shown within the context of the rest of a face than when the parts are shown alone
What is the the finding that people are better able to discriminate between two features of a face when they are shown within the context of the rest of a face than when the parts are shown alone known as?
face superiority effect
What are the two approaches to modeling the influence of top-down effects?
Network-feedback models and Bayesian approaches
What is the Bayesian approach?
the influence of stored information is probabilistic; it reflects what has often happened in the past and therefore likely to happen again
What is Bayem's Theorem?
It is possible to estimate reverse possibility from the available probabilities from previous experience
What increases the accuracy of perception?
Knowing more from experience about what patterns are likely to appear
What is bottom-up processing determined by?
Information from the external environment
What is top-down guided by?
Internal knowledge, beliefs, goals, expectations
When are bottom-up processing techniques used? Top-down?
Usually used simultaneously to establish best available solution for object recognition
What is bistable perception?
We can perceive both interpretations, but only one at a time.
What are the two ways we resolve ambiguity of stimulI?
bistable perception and binocular rivalry
What does bistable perception lead to?
Spontaneous alternations between the two interpretations, even when we keep our eyes focused on a fixation point
What two principles do neural networks rely on in explaining bistable perception?
Competition and adaptation
How does bistable perception work?
If one of two possible interpretations induces a stronger pattern of activation, it will suppress the other, producing a single winning interpretation. However, the ability of the winner to suppress the loser gradually adapts or weakens over time, until the "loser" can dominate
What is figure-ground relations?
What is binocular rivalry?
A state in which individual images to each eye compete. If a different monocular image is viewed in the fovea of each eye, you spontaneously alternate between the two images, reversing every few seconds and never seeing both at the same time
Why does binocular rivalry exist?
What does spatial processing of location rely on?
Dorsal "where" pathway
What does the dorsal pathway project onto?
V1 and parietal lobes
What does the ventral visual pathway project onto?
V1 to ventral areas like V4 and inferior temporal cortex
What does object recognition rely on?
Ventral visual pathway
What is the dorsal pathway involved in?
Both planning visually guided actions as well as localizing objects
What is apperceptive agnosia?
Impairment in judging even basic aspects of the form or shape of objects
What is the impairment in judging even basic aspects of the form or shape of aspects?
What is apperceptive agnosia a result of?
Damage to ventral pathway
What does damage to the ventral pathway lead to?
What does damage to the dorsal pathway lead to?
What is apraxia?
the inability to make voluntary movements
What is the inability to make voluntary movements?
What is apraxia the result of?
Damage to the dorsal pathway
What can apraxics do? What can they not?
Can perform actions from memory and can describe what they see, but cannot perform new actions on what they see
What can apperceptive agnosics do? What can they not?
Cannot judge aspects of form or shape of objects, but can perform objects with them