Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
39 Cards in this Set
- Front
- Back
|
Visual Processing occupies up to half of the total cortex space
|
No Way!
|
|
|
Distal Stimulus
|
Objects and events - things to be perceived (books, trees, monkeys)
|
|
|
Proximal Stimulus
|
The reception of info and its registration by a sense organ
|
|
|
Percept
|
The interpretation of the stimuli (trees, cars, etc) from the upside-down, backward, two-dimensional image you take in from your retina. This is not part of the proximal stimulus. Ex. Size Constancy
|
|
|
Size Constancy
|
Example of a percept. Moving your hand closer and farther away from you brings you proximal stimulus images of two different sized hands, but you interpret your hand as the same size.
|
|
|
Form Perception
|
Gestalt perception. The segregation of figure from ground. Ex. faces vs vases - distinguishing the separate forms in the black and white areas. The same principle applies to many Dali paintings. Also, Subjective contours images.
|
|
|
Subjective Contours
|
Complex display is subject to a simplifying interpretation the perceiver makes without being aware of making it. The perception is not completely determined by the stimulus display, it requires the perceiver's active participation. (circles with corners cut from them form a square that is not there)
|
|
|
Gestalt Principles of Perceptual Organization
|
Principle of Proximity, Similarity, Good Continuation, Closure, and Common Fate. (Particularly Sarcastically, Gill Cried "Cold, Canned Fish!")
|
|
|
Principle of Similarity
|
Gestalt - things that are similar are grouped together (in a block of light objects, a dark section/row will be considered separate from the rest of the group)
|
|
|
Principle of Proximity
|
Gestalt - closer items are grouped together (writing is seen as rows, not clumps of letters)
|
|
|
Principle of Good Continuation
|
Gestalt - We group together objects whose contours form a continuous straight or curved line. An X is seen as two intersecting lines, not two points meeting in the middle.
|
|
|
Principle of Common Fate
|
Gestalt - Elements that move together will be grouped together.
|
|
|
Law of Pragnanz
|
More general law than Gestalt - of all the possible ways of interpreting a display, we select the organization that yields the simplest and most stable shape /form. There is a perceptual immediacy of larger figures over smaller components (a large H made up of numerous S's will be seen as an H, not as S's)
|
|
|
Bottom-Up Processes
|
Data driven. Small bits of information from the environment are combined to form a percept.
|
|
|
Top-Down Processes
|
Theory or Conceptually driven. Expectations, theories, or concepts guide the selection and combination of the information in the pattern recognition process.
|
|
|
Template Matching and Problems (3)
|
Bottom-Up. Recognizing previously stored patterns, or templates. Everything we encounter forms a template in our mind that we try to match from then on.
Problems (3) 1) we would have to store an impossibly large number of templates 2) We are still capable of recognizing new technology - How and when templates are created / How we keep track of them is not explained, 3) People recognize patterns as more or less the same, even when the stimulus patterns differ greatly (different handwriting) The template only works with clean stimuli. |
|
|
Feature Analysis
|
Bottom-Up. Recognition of a whole object depends on recognition of its features. Influence of edge detectors, horizontal-line detectors, vertical-line detectors, and many more. However, there is no clear definitions of what can be a feature and what cannot.
|
|
|
Geons
|
Bottom-Up. Irving Biederman using feature analysis. People use Geons, simple geometric components, to view objects. Similarly to verbal phonemes, using a set of 36 geons can be used to represent thousands of common objects we can quickly recognize. When people see incomplete drawings of objects, but they have the ability to recognize geons, they can identify the picture. When these vertexes are deleted, we can no longer recognize the pic.
|
|
|
Visual Search Task
|
Bottom-Up. Neisser - trying to search out a Z in a grouping of letters with similar features (T, M, V, and E's) is much harder than finding it in a grouping of letters with dissimilar features (O, U, R, and G's). Part of Feature Analysis.
|
|
|
Pandemonium
|
Bottom-Up. Demons (detectors) 'scream' when they detect their particular feature. They scream louder or softer depending on the clarity of the input. Lower and higher level demons detect simpler and more complex stimuli. Some features matter more than others in pattern recognition.
|
|
|
Prototype Matching
|
Bottom-Up. Attempts to correct shortcomings of the template matching and feature analysis models. Matching input to a stored representation of information, but the representation is a loose prototype - an idealized representation of some class of objects or events (the letter M, a cup, a collie, etc.) This model does not require an object to contain any one specific feature or set of features to be recognized. The more features a particular object shares with a prototype, the higher the probability of a match. It also takes into consideration features' relationships among each other. These prototypes are formed very quickly.
|
|
|
Context Effects
|
Top-Down. This is one of the biggest problems with bottom-up processes, along with expectation effects. The context in which a pattern or object appears sets up expectations in the perceiver as to what object will occur. Both accuracy and length of time to recognize vary with context - you recognize utensils faster in a kitchen setting then in a mixed scene.
|
|
|
David Marr
|
Top-Down. People first make a primal sketch (brightness and darkness on a 2D image with basic geometric shape - you see boundaries, but don't know what it means). Then there is a 2 1/2 D sketch adding shading, texture, etc.to see depth somewhat. Then the 3D sketch is made, and is the only step to use top-down processing. Here, you recognize objects and understand the meaning of the scene.
|
|
|
Perceptual Learning
|
Top-Down. Perception changes with practice. Ex. A person is shown a card, then is told to identify any copies of the original card in a deck. People made errors dependent upon the number of similarities with the original. Over time, people become more familiar with the stimuli and make less errors. (expert vs novice) pg 84
|
|
|
Change Blindness
|
Top-Down. An inability to detect changes in an object or scene, especially when given different views of that object or scene. Ex. Continuity errors in film. People may make of the visual representation as a 'gist' of the scene, but not the specific details. Our visual percepts are not precise copies of our visual world. This is to help our perceptual system from becoming overwhelmed by the sheer amount of information available in any one glance or view.
|
|
|
Word Superiority Effect
|
Top-Down. Reicher. Letters are easier to perceive in a familiar context than in an unfamiliar context or no context at all. Participants asked to identify either a D or a K performed better when they were presented in the forms of WORD and WORK, rather than alone.
|
|
|
Missing Letter Effect
|
Top-Down - part of the Word Superiority Effect. The ability to detect letters in enhanced by word familiarity when words appear in isolation, but inhibited by increased familiarity or role when a word appears in real text. When people are asked to cross out all the f's in a text, they distinguish function words (of, for) from content words (fortune, fort). People miss f's when they appear in function words.
|
|
|
Connectionist Model of Word Perception
|
The model assumes that input, whether written, spoken, or higher level, is processed at several levels (features, letters, phonemes, etc) which feed into one another. Nodes associated with features can be either inhibitory or excitatory. You can perceive only one word per instant. Elaborate web of connections.
|
|
|
Neuroscientific Perspective on Word Perception
|
Words and pseudowords produce different PET scans from those produced when participants saw random letter strings or false fonts. Activity in the left hemisphere in regions outside the primary visual cortex. This is involved in semantic processing.
|
|
|
Direct Perception
|
The act of perception is the construction of mental representations of objects. The proximal stimulus does not contain all the info needed to identify objects , so we use knowledge to fill in the gaps. We still recognize a tune in the key of G, even if it is usually presented in the key of C.
|
|
|
Constructivist Approach to Perception
|
Involves Direct Perception. People add to and distort the information in the proximal stimulus to obtain a percept, a meaningful interpretation of incoming information. People are active selectors, integrators, and constructors of information.
|
|
|
Gibson
|
Part of Direct Perception. Patterns of motion provide much information to people. Ex. motion detectors on models - when standing still, they just look like a collection of dots, but when walking, it is obviously a person. Pilots watch texture and motion as they land - they move more quickly as you come up to them, and become more detailed.
|
|
|
Affordances
|
Gibsonean thought. Acts or behavior permitted by objects, places, and events. We perceive the affordances of the environment along with with other factors, such as movement, shape, texture, etc. It keeps us from running into walls. There is no sharp definition of what an affordance is, however.
|
|
|
Schema
|
From Neisser. They derive from the knowledge base and contian expectations derived from context, guiding the perceiver to explore the environment in particular ways. Helping the perceiver modify expectations. Must pay attention to the relationship between the way info is processed and the goals and needs of the organism doing the processing.
|
|
|
Visual Agnosia
|
Impairments in the ability to interpret visual information in a meaningful way. There is no visual problem - can reproduce images with fine detail - and there is no memory problem, but they cannot recognize a drawing of a dog as a dog. They cannot recognize things by sight, but can by sound, tough, or smell.
|
|
|
Apperceptive Agnosia
|
Process limited visual information. They see contours/outlines of a drawing, but cannot match it with another or categorize objects. Some cannot recognize X from O, or match the same object from two diff points of view. Associated with damage to the right hemisphere.
|
|
|
Associative Agnosia
|
they can match objects or drawings, and copy drawings,but do so slowly and extremely carefully (point by point instead of drawing big features and filling in the details). Associated with bilateral brain damage.
|
|
|
Prosopagnosia
|
visual agnosia for faces. Associated with damage to the right hemisphere. May not be able to recognize faces of family, politicians, or themselves, but their object recognition is unaffected. Can see details of faces, but cannot but them together into a face.
|
|
|
Unilateral Neglect
|
Associated with damage to the parietal cortex, it causes you to ignore stimuli on the opposite side. You may only brush one side of your hair, wash one side of your face, etc.
|
