Unit 4 lecture 2 - signal detection theory

0.0(0)
studied byStudied by 0 people
0.0(0)
full-widthCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/11

encourage image

There's no tags or description

Looks like no tags are added yet.

Study Analytics
Name
Mastery
Learn
Test
Matching
Spaced

No study sessions yet.

12 Terms

1
New cards

3 domains of expericne as humans

Sensory: five ways of bringing info in

○ Cognition: information processing &

decision making

○ Action: motor system

2
New cards

what is the signal detection theory

A framework for understanding accuracy of decision making is quantified into...

4 outcomes

1. Hit (signal w right response)

2. Miss (signal wrong response)

3. False alarm (no signal wrong response) TYPE 1 ERROR

4. Correct negative (no signal right response) TYPE 2 ERROR

<p>A framework for understanding accuracy of decision making is quantified into...</p><p>4 outcomes</p><p>1. Hit (signal w right response)</p><p>2. Miss (signal wrong response)</p><p>3. False alarm (no signal wrong response) TYPE 1 ERROR</p><p>4. Correct negative (no signal right response) TYPE 2 ERROR</p>
3
New cards

SDT is the ability to...

differentiate between

"information-bearing patterns" (stimulus/signals) and "random patterns" that distract from the information

*during unpaced task just focus on accuracy bBUT uncertainty is present

4
New cards

Applying SDT to motor learning

helps understand how accurately and consistently someone makes decision s while their learning a new skill

Eg. athlete must decide is that movement signal important (signal) or irrelevant (noise)

Recall: selective attention

5
New cards

explain the sdt gaph theoretically if both categoiescan be detected w 100% accuracy

x axis - OUR internal sense of what we think the dot on the radar is

*yellow line...

ß = Criterion/cut off - this is our descision to choose one over the other --> past the line will decide its a airplane and before it will descide as bird

d' = dscriminibility/perceptual sensitivity

--> The distance between those two curves, The larger the d', the easier it is to distinguish

<p>x axis - OUR internal sense of what we think the dot on the radar is</p><p>*yellow line...</p><p>ß = Criterion/cut off - this is our descision to choose one over the other --&gt; past the line will decide its a airplane and before it will descide as bird</p><p>d' = dscriminibility/perceptual sensitivity</p><p>--&gt; The distance between those two curves, The larger the d', the easier it is to distinguish</p>
6
New cards

beta shifts

changes the number of H, M, CN, FA

*we only have control over the criterion not the discrimintibility

beta shift left - lax/liberal

(okay with getting more FA (calling it a plane when its a bird) if that means getting less misses)

eg. looking attb xray call anything thats even a little blur a concern so you dont miss but might have false negative

beta shift right - strict

only calls it a plane if theyre very sure so they are less false alarms but may be misses

eg. in medical situations to diagnose only if very sure of a hit

<p>changes the number of H, M, CN, FA</p><p>*we only have control over the criterion not the discrimintibility</p><p>beta shift left - lax/liberal</p><p>(okay with getting more FA (calling it a plane when its a bird) if that means getting less misses)</p><p>eg. looking attb xray call anything thats even a little blur a concern so you dont miss but might have false negative</p><p>beta shift right - strict</p><p>only calls it a plane if theyre very sure so they are less false alarms but may be misses</p><p>eg. in medical situations to diagnose only if very sure of a hit</p>
7
New cards

Perceptual performance depends on.. ● Two major factors:

- Perceptual sensitivity of the performer (more sensistive = more appropriate response and less false alarms)

- Expectancies about the performer

and the rewards and penalties inherent in the situation (essentially what ß shift they wanna do)

**these factors are independant of eachother

8
New cards

perceptual sensistivity (d')

With high sensitivity there are more correct decisions (hits and correct negatives) and fewer errors (false alarms and misses) made

● Reverse is true for low sensitivity

● Note that correct decisions and errors are

related. If one changes so does the other.

9
New cards

what might cause a change is perceptual sensiotiviety d'

- environemntal conditions eg wetaher

- individual diferences eg. sight, experience (novice vs expert)

10
New cards

observers criterion ß

determeined byt the performer and reflects their expectation and rewardsa and penalties in the situation eg. whats on the line if we miss is it worth gett more false alarms

The position of the criterion affects the number of correct decisions and errors in the same manner as d'

11
New cards

d' and ß simmilarities/dif

both affects performance but change independant of eachother

eg. d could be the same in both novice and expert but but b is more lax in experts

D' is more about environmental noise and beta is more internal

12
New cards

B Shifts in drivers

noobs = strict shift to right

- inc miss less false alarm bc too puss to respond to hazardous dituation

expert = lax shift to left

- decr miss, inc fals alarm

better safe than sorry and react to hazardous situation even if tis not acc there