Psych 207 - Module 3: Perception

Bottom-up Processes

• Refers to ability to take basic data and form a percept

• Operates in one direction, reflexively and passively, from stimulus to output

• Involves Prototype Matching, Feature Analysis, Template Matching

Template Matching

Template Matching

- Barcodes

- Proximal stimulus in retina, compared to templates in memory until a match is found or and object can be named

- Operates with all senses

- Perception depends on physical match between stimulus and stored representation in memory

Problem:

○ If identification requires an exact match, impossibly large data base is needed

○ Model cannot explain how we recognize new object, one that we have never encountered before,

○ How do we deal with surface variation in stimuli

Feature Analysis

Feature Analysis is a theory in cognitive psychology and perception that explains how we recognize and process visual or auditory stimuli by breaking them down into their basic components (features) before combining them into a complete perception.

A theory of perception where complex patterns are recognized by first breaking them down into simpler, individual features

Objects recognized by component parts or features, and manner in which they are combined

Selfridge's Pandemonium model

○ Dumb/feature Demons --> scream when they recognize the stimuli they are designed to detect

○ First, Image demons, then feature demons, then cognitive demons, then decision

The model has four levels of demons, each with a different role:

Image Demon

• Receives the raw sensory input (like the visual image on your retina).

• Passes the stimulus to the next level without analysis.

Feature Demons

• Each is specialized to detect a specific visual feature (line, angle, curve).

• If a feature is present, that demon “shouts”.

Cognitive Demons

• Represent entire patterns or letters.

• Each cognitive demon listens to the feature demons and shouts louder if more of its required features are present.

Decision Demon

• Listens to all cognitive demons and selects the loudest one as the recognized object or letter.

Prototype Matching

Attempts to correct some of the problems associated with rigidity of both template matching and feature analysis model

○ Doesn't require exact match, rather a best fit is good enough

All of these approaches differ in how the comparisons are made, what they are based on, and how flexible it is

Top-Down Processes

• Come from us, the observer, instead of environment

• Consist of world knowledge, theories, expectations, etc

Disruptions of Perception - Visual Agnosia, Apperceptive Agnosia, Associative Agnosia

Visual Agnosia: inability to identify object by sight

Apperceptive Agnosia

• Can see and interpret contours and outlines

• Have a difficult time maintaining basic representations in memory to even match objects or to distinguish amongst them

Associative Agnosia

• can match objects and copy drawing but do so slowly

• Cannot name objects just seen or even just copied

• It is not that they cannot see the object or do not know what it is -- if tested in a different way the person will show that they can identify it

• Cannot access meaning or semantics from a visual description alone

Put differently, Associative agnosics can match one object to another but do so very slowly and much more difficulty matching one object to another due to only being able to process a very limited visual information

Perception Prosopagnosia

• specific to faces: can recognize visual objects and can see details of faces, but cannot recognize a face as a coherent unit, even those of loved ones and friends or famous people

• Can use other information, such as voice, hair, posture, gait, and the like to recognize people

• Lack explicit (aware) face recognition and cannot overtly name someone from looking at a photograph of a face, that same photograph can be used to demonstrate implicit (unaware) face recognition abilities that are preserved

• Galvanic skin response (GSR) recorder will show different responses to pictures of loved ones

• Thus they can't recognize faces consciously

• Impaired explicit or overt face recognition, but preserved implicit or covert face recognition

Capgras Syndrome

• individuals have preserved explicit or overt face recognition, but impaired implicit or covert face recognition

• Can look at picture and tell you whom it looks like, but no GSR (Galvanic Skin Response)

• Claim person is imposter

  ○ Deficit in reasoning

  ○ Attempt to reconcile the face with the lack of warmth one normally feels when seeing a loved one