Cognitive psychology 3 PDF

Preview

Summary

Chapter 3...

Description

 Theory: An object, such as a letter, is recognized by comparing it to the templates stored in the brain, much like fingerprints are matched  Do we have templates for every possible size, orientation, and form of each object we encounter? Highly unlikely  This overly simple model did, however, suggest the idea of a perceptual system based on features  A bottom-up theory  Word superiority effect: letters are easier to recognize when they are contained in a word (a) than in a non-word (b) or even when presented alone (c)  Task is to pick the letter seen in the original stimulus K PARK XXXX XXXX M K ARPK XXXX XXXX M K K XXXX XXXX M original stimulus masking stimulus test stimulus  This behavioral study demonstrates feedback activation from the word to the letter level  Theory: Objects are analyzed into individual features in the preattentive stage; in the focused attention stage, the features are combined into an object that can be perceived  The phenomenon of illusory conjunctions provides support for FIT  Behavioral studies of this phenomenon demonstrate that early in the perceptual process features of a visual stimulus may become separated from each other. i.e., free-floating Stimuli Response: “small, red triangle and large, blue circle”  FIT mostly involves bottom-up processing  The types of features in FIT do not explain how we recognize three-dimensional objects in our world  Also a feature theory but the features are three-dimensional volumes called geons  Biederman has proposed that there are 36 different geons  Geons can be identified when viewed from different angles  Geons, therefore, have view invariant properties  Two other important properties of geons:  Discriminability: each geon can be distinguished from each other from almost all viewpoints  Resistance to visual noise: geons can be perceived even under poor viewing conditions, e.g., fog or partial obstruction  Therefore, if enough information is available to distinguish an object’s geons, the object will be identifiable  Like FIT, RBC is based on analysis of objects into parts early in the perceptual process  The early Gestalt psychologists, believing that the whole is greater than the sum of its parts, asked how elements in the environment are organized into elements 

 http://www.moillusions.com/wpcontent/uploads/i207.photobucket.com/albums/bb234/vurdlak8/illusions/5.jpg  The Gestalt “laws” of perceptual organization are really heuristics, or “rules of thumb,” that provide best guess predictions about what is perceived in the environment  Law of Pragnanz  Also known as the law of “good figure” or the law of simplicity  Most people will describe this figure as three overlapping squares, not as six capital Ls, which was how the figure was created  Law of Similarity  Items that are similar in some way tend to be grouped together  Law of Good Continuation  Objects arranged in either a straight line or a smooth curve tend to be seen as a unit  Law of Proximity  Objects that are near to each other appear to be grouped together  Law of Common Fate  Objects that are moving in the same direction appear to be grouped together  Law of Familiarity  Objects are more likely to be grouped together if the group formed appears familiar or meaningful  An example is a “man” seated on the surface Mars:  The Gestalt laws of perceptual organization seem like pure common sense. Duh. Of course, we group things together that are similar, moving in the same direction, etc.  Recall, however, the concept of experience-dependent plasticity from Chapter 2. The Gestalt laws are analogous in that we group objects the way we do because of the characteristics of our environment. These heuristics are correct most of the time—in this world  Thought experiment: Imagine a world where the Gestalt laws would be of no help to you ...  Despite astounding advances in other areas, computers still have difficulty identifying specific objects.  Why computers seem to lack perceptual intelligence:  Inverse projection problem: a particular image on the retina could be made by an infinite number of objects  Objects need to be distinguished from their surroundings and from one another  Example: speech segmentation, distinguishing individual words when the speech signal is actually continuous  Objects can be partially hidden or blurred  Difficulty determining whether a border is caused by shadows or a change in the parts of a scene  These situations pose few problems for people, who are able to apply their knowledge of the real world—and their Gestalt heuristics  Segmentation is also aided by our knowledge of transitional probabilities, i.e., the chances of one sound following another sound  Process of learning about these probabilities is called statistical learning  Infants as young as 8 months can learn the transitional probabilities of the language spoken in their families

 “I scream, you scream, we all scream for ice cream.”  “I scream” and “ice cream” have the same pronunciation but are heard differently because we understand the meaning of the sentence  Speech segmentation, or perceptual organization of a continuous stream of sounds is facilitated by our knowledge of the meaning of the sounds  Experiments involving simultaneous events  Failure to perceive an event while attending to another event  Person in gorilla suit walks unnoticed through a basketball game  Change detection experiments  Failure to see obvious changes from one shot to another of the same scene  A woman’s scarf disappears from shot to another without notice from research participants asked to note any changes to objects, body position , or clothing (Levin & Simons, 1997)   This difficulty is known as change blindness  If you are unaware of this difficulty, you may be suffering from the dreaded condition:  change blindness blindness ...