CHP. 4 - Recognizing Objects

Question 1

Associative agnosia

Answer 1

Problem of recognition
Can pick up info & process it, but it can’t recognize or make sense of the info coming in.
* Perception occurs, recognition doesn’t
* Problems with linking input to visual knowledge
* “What” pathway

Can copy an image with detail, but can’t tell you what it is.

Question 2

Apperceptive agnosia

Answer 2

Problem of perception
* Senses work, but the brain can’t process the info
* Problem with organization of elements

Can identify an image from memory, but can’t replicate it in a drawing.

Question 3

Feature detection

Answer 3

Recognition begins with ID of visual features in input pattern
* Smaller units -> larger units

Question 4

Visual search task

ex. Where’s Waldo?

Answer 4

Participants examine a display & judge whether a particular target is present
* Efficient when target is defined by a simple feature
* Slow when target is defined by a combo of features

Question 5

Word superiority effect (WSE)

Answer 5

It’s easier to perceive & recognize letters-in-context (ie. a word) than if they appear in isolation
* Word = faster/better than individual letters
* Word = faster/better than non-words

Question 6

Word-formedness

Answer 6

How closely a letter sequence conforms to the typical patterns of spelling in the language
* More well-formed = easier to recognize sequence
* Influences errors (like DPUM misread at DRUM)

ex. HZYQ vs. FIKE vs. HIKE

Question 7

Activation level

Answer 7

Measure of the current status for a node or detector
* Increased if the node/detector receives appropriate input from associted nodes/detectors
* High if frequency & recency are high

Question 8

Response threshold

Answer 8

Quanitiy of info or activation needed to trigger a response in the node/detector (neuron)

Question 9

Recency

Answer 9

Detectors that have fired recently will have higher activation levels
* Warm-up effect

Question 10

Frequency

Answer 10

Detectors that have fired frequently will have higher activation levels
* Exercise effect

Question 11

Bigram

Answer 11

Pair of 2 letters

ex. HJ or ET

Question 12

Digraph

Answer 12

Pair of 2 letters = one sound

ex. PH or SH

Question 13

Ambiguous inputs

Answer 13

Weak signal will likely be enough to trigger only a well-primed detector

Question 14

Local representation

Answer 14

Info is encoded in a small number of identifiable nodes
* “1 idea per node”

Question 15

Distributed representation

Answer 15

Ideas or contents in which there is no 1 node (or specific group) representing the content & where it’s stored.
* Content is represented via a pattern of simulatenous activity across multiple nodes

Question 16

Efficiency vs. accuracy

Word recognition

Answer 16

Perfect accuracy is sacrificed for effiency

Question 17

McClelland & Romehart model

Feature nets

Answer 17

Emphasizes the role of excitary & inhibitory connections
* Info flows bottom-up
* Higher-level detectors can influence lower-level detectors

Question 18

Recognition by components (RBC) model

Answer 18

Applies the feature net model to recognition of 3D objects

Question 19

Hierarchy of detectors

Objects

Answer 19

1. Feature detectors (e.g. curves, edges)
2. Geon detectors
3. Geon assemblies
4. Object model

Question 20

Inferotemporal (IT) cortex

Answer 20

Important for object recognition; part of the lower temporal lobe
* “What” pathway
* IT cortex lesioned = agnosia

Question 21

Prosopagnosia

Answer 21

Inability to recognize individual faces despite normal vision

Question 22

Holistic perception

Hugh Jackman & Ryan Reynolds example

Answer 22

Perception of the overall configuration rather than an assemblage of parts