Textbook Notes (290,000)
CA (170,000)
McMaster (10,000)
PSYCH (1,000)
Chapter 11

PSYCH 2H03 Chapter Notes - Chapter 11: Outline Of Object Recognition, Distributed Knowledge, Agnosia

Course Code
Ellen Maclellan

This preview shows half of the first page. to view the full 2 pages of the document.
Intro to Object Recognition
- Recognition of visual inputs begins with features:
o But its not just the features that matter
o How easily people recognize a pattern also depends on how frequently / recently
they’ve viewed the patterned / whether the pattern is well formed
- A feature net:
o A network of detectors, each of which is primed according to how often / recently
it has fired
o Relies on distributed knowledge to make inferences
o A process that gives up accuracy to gain efficiency
o Can be extended to other domains (such as recognition of 3D objects) BUT
recognition of faces requires a different model
- Top down influences on recognition:
o Tells us that object recognition is not a self-contained process
Patient D.F
- Suffered from apperceptive agnosia
- Could see, but couldn’t organize the elements he saw in order to perceive an entire object
- When asked to copy drawings by looking at them, the drawing replications were poor
o But the problem was not in drawing ability which is backed up by her
performance being much better when asked to draw the same objects from
memory (not looking at them)
Early considerations of recognition
- Object recognition involves complexity
- You recognition of various objects is influenced by the context in which you encounter
o Recognition being influenced by one’s knowledge and exepctations
o Concept-drive
o Top-down processing
- Processes directly shaped by the stimulus = data driven’
o Bottom up processing
The importance of features
- Recognition begins with the identification of visual features in the input pattern
o Ex: vertical lines, curves, etc.
o Once these features are appropriately catalogued, you can start assembling the
larger units
You're Reading a Preview

Unlock to view full version