Exam 1 - L5

Question 1

Signal detection theory (SDT)

Answer 1

How do we make decisions (detections) in the presence of noise? Shows how biases and expectations (top-down) impact our perception

Question 2

Signal (SDT)

Answer 2

target stimulus you are trying to detect
Ex: waldo (in where’s waldo)

Question 3

Noise (STD)

Answer 3

anything that impacts your ability to detect the signal
Ex: distractors (in where’s waldo)

Question 4

Sensitivity (SDT)

Answer 4

how easy or hard it is to distinguish signal from noise
Low sensitivity: difficult to distinguish
High sensitivity: easy to distinguish

Question 5

Bias (SDT)

Answer 5

tendency to say “yes” that a signal is present or “no” that a signal is not present
Independent of sensitivity
Determined by expectations/payoffs
A large reward for a hit will bias toward Yes
A large penalty for a false alarm will bias toward
No

Question 6

Hit (SDT)

Answer 6

there is a signal, and you correctly detect it

Question 7

Miss (SDT)

Answer 7

there is a signal, but you fail to detect it

Question 8

Correct rejection (SDT)

Answer 8

there is No signal, and you correctly say No

Question 9

False alarm (SDT)

Answer 9

there is No signal, but you say Yes

Question 10

Context effects

Answer 10

When the perception of an object is affected by the context/ environment
-Subjective contours: context causes the edge
detectors in our brain to fire even if the edges
aren’t actually there
-Comes from environment and context

Question 11

Word superiority effect

Answer 11

People recognize letters faster when they appear as part of a word than when they appear by themselves or among random letters
Higher-level (words) affect perception of lower-level (letters)
Example of a top-down effect

Pseudoword superiority effect:
Faster even if the letter string is not a word than letter alone, as long as the nonword is pronounceable

Question 12

Interactive activation model

Answer 12

Activation flows between levels in both directions
-Hierarchical structure: feature level, letter level.
and word level
-Feature nodes excite or inhibit letter nodes
(bottom-up)
-Curved features activate letter C
-Inhibit other letters
-Letter nodes compete and excite/inhibit word
nodes (bottom-up)
-Letter C activates word CAT
-Inhibits other words
-Word nodes compete and excite/inhibit letter
nodes (top-down)
-Word CAT activates letter C
-Inhibits other words/letters