Textbook Notes (280,000)
CA (160,000)
Western (10,000)
PSYCH (5,000)
Chapter 3

Psychology 2135A/B Chapter Notes - Chapter 3: Connectionism, Word Superiority Effect, Change Blindness


Department
Psychology
Course Code
PSYCH 2135A/B
Professor
Robert Brown
Chapter
3

This preview shows page 1. to view the full 5 pages of the document.
Chapter 3 - Object Perception and Pattern Recognition
- perception: interpretation of sensory information to yield a meaningful description
- areas of brain responsible for visual processing occupy up to half of total cortex space
- perception divided into visual, auditory, gustatory, olfactory, haptic
- "classic" approach to defining perception:
- distal stimulus: stimulus that exists in the world
- proximal stimulus: reception of information and its registration by a sense organ
- percept: outcome of a perceptual process; meaningful interpretation of incoming
information
- retina: back of eye
- retinal image: two-dimensional image; closer to object, larger the image; upside down and
reversed with respect to left and right; formed on retina
- size constancy: phenomenon that one's perception of an object remains constant even as the
retinal image of the object changes size
- pattern recognition: classification of a stimulus into a category
- most, if not all, instances of perception involve pattern recognition
Gestalt
- we interpret stimulus arrays as consisting of objects and backgrounds
- form perception: process by which brain differentiates objects from their backgrounds
- objects are figures and the background is ground
- figure is seen as having a definite shape, better remembered
- ground is seen as more shapeless, less formed, and farther away in space
- subjective (illusory) contours: illusory outline created by certain visual cues that lead to
erroneous form perception
- subjective contours come to be through a simplifying interpretation the perceiver makes;
perception not completely determined by the stimulus display
- Gestalt psychologists assert whole is not the same as the sum of its parts; we recognize objects
as a whole instead of individual features
- Gestalt principles of perceptual organization: laws that explain regularities in the way
people come to the perceptual interpretations of stimuli
- principle of proximity (nearness): group together things that are nearer to each other
- principle of similarity: grouping elements that are similar
- principle of good continuation: group together objects whose contours form a continuous
straight or curved line
- principle of closure: mentally filling in gaps; creates subjective contours
- principle of common fate: group elements that move together (includes luminance change even
if stimuli differ in brightness)
- law of Pragnanz: most Gestalt principles subsumed here; of all the possible ways of interpreting
a display, we tend to select organization that yields the simplest and most stable shape or form
- simple and symmetry forms are seen more easily
- infants as young as three demonstrate use of some Gestalt principles
- formalization of Gestalt law of Pragnanz creates minimal model theory
- Gestalt approach does not answer how these principles are translated into cognitive or
physiological processes
You're Reading a Preview

Unlock to view full version

Only page 1 are available for preview. Some parts have been intentionally blurred.

- law of Pragnanz can be circular (why do we see two triangles? interpretation makes for a
simple, stable figure. How do we know this figure is simple and stable? Because we so readily
see it)
Bottom-up Processes
- bottom-up (data-driven) process: perceiver starts with small bits of information from
environment and combines them in various ways to form a percept
- system has no way of going back to an earlier point to make adjustments
- lower level of processing is taking information about a stimulus
- higher level process includes expectations or previous learning
- bottom-up processes are relatively uninfluenced by expectations or previous learning
- bottom-up processes involve automatic, reflexive processing that occurs even when perceiver is
passively regarding information
- two biggest problems are context effects and expectation effects
- context effects: effect on a cognitive process due to information surrounding target object or
event; sometimes called expectation effect since context is thought to set up certain expectations
- three examples of bottom-up models: template matching, featural analysis, prototype matching
Template Matching
- templates: previously stored patterns
- every stimulus is compared to some previously stored pattern, or template
- if a number of templates match, further processing needed to sort out which template is most
appropriate
- model implies we have stored millions of different templates
- model cannot provide complete explanation
- first, for complete explanation, need to have stored impossibly large amount of templates
- second, does not explain how and when templates are created and how we keep track of an
ever-growing number of templates
- third, people recognize many patterns as more or less the same thing, even when stimulus
patterns differ greatly i.e. handwriting
- must of stimulus information we perceive is far from regular i.e. alteration, degradation,
unfamiliar orientation (upside-down); is separate template needed for each variation; how do we
know how an object should be before matching to a template
- cannot know before whether an input pattern should be adjusted before matching to templates
- good for relatively clean stimuli, but does not explain how we perceive effectively among all
the "noise" in everyday life
Featural Analysis
- features: parts searched for and recognized
- recognition of a whole object depends on recognition of its features
- certain detectors in senses appear to scan input patterns, looking for a particular feature i.e.
horizontal line
- visual search task: task where subjects asked to detect presence of particular target against an
array of similar stimuli
- harder to find target among similar stimuli
- Pandemonium: model of letter perception based on a bottom-up hierarchy of feature detectors
You're Reading a Preview

Unlock to view full version