Calculation of signal detection theory measures

20 important questions on Calculation of signal detection theory measures

Where can signal detection theory be applied (two possible stimulus types)

1) psychologists first applied the theory in studies of perception, where subjects discriminated between signals (stimuli) and noise (no stimuli)

2) recognition memory (old and new) and lie detection

What are the three most popular tasks?

- yes/no task; involves signal trials, which present one or more signals and noise trails, which present one or more noise stimuli

- rating task

- forced choice task

Give an example for a decision situation that is suitable for applying SDT

A physician interprets an X-ray picture A crime witness tries to identify a criminal in a line-up A jury has to decide whether a suspect is guilty A smoke detector beeps when a slice of toast burns in the toaster You see a man who looks familiar and try to remember whether you know him or not A proofreader looks for spelling mistakes in a book manuscript A spider phobic discovers what looks like a spider on the wall
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

According to SDT, what are the two factors that affect human discrimination decisions?

Sensitivity: How good am I at identifying/discriminating? Decision criterion / Response criterion: How do I prefer to respond, if I am in doubt?

Why is it important to consider both factors when analyzing decisions

Because the responses always depend on two factors

Which two features should an experiment have, so it can be analyzed with SDT?

Repeated presentation/ not presentation' of stimuli must be a difficult decisions

In SDT, what does 'decision criterion' mean?

Consequences of decisions (pay-off matrix), frequency of signal (u know 50/50 chance, u do 50/50)

Use an experimental example to explain what hits, false alarms, correct rejections and misses are?

Hit -> Signal + Response (Spider present and Spider seen) Miss -> Signal + No Response (Spider present but Spider overlooked) False Alarm -> No Signal + Response (no Spider present but spider assumed) Correct Rejection -> No Signal + No Response (no Spider present and no spider seen)

What are the theoretical assumptions that explain differences in sensitivity?

In reality, there is a signal or not (dichotomous events)l, however subjective experience varies

subjective experiences vary and are normally distributed around the means.

1) data follow a Gaussian distribution 2) the variances of two distributions are equal

What does it mean to say that someone has high or low sensitivity?

People for which it is easy to tell things apart have a high sensitivity (bell shapes further apart) and people that have troubles discriminating have low sensitivity

What does it mean to say that someone responds liberally, neutrally or conservatively

Liberal: a lot of false alarms, but hardly any or no misses Neutral: respond with yes equally as often with no Conservative: many misses but no false alarms

What is the meaning of the SDT values d' (d prime) and B (beta)?

Distance between standardized means of the N distribution and the S+N distribution more liberal, neutral or more conservative

What does a low/high value of d' mean?

Low = 0 (person cannot discriminate at all)
high = 4.66 (almost perfect at discriminating; 99% correct)

What does a low/high value of B mean?

Low = .50 (liberal Yes responder)
High = 1.99 (conservative no responder)

Which values do you need to compute a person's B value

Yes responses (hits and false alarms)
B= yhits/ yFalse alarms

How can SDT be applied to memory experiments

Is presented stimulus old (i.e. presented in learning phase?) Hit = yes/old False alarm = yes/new Miss = no/old Correct rejection = no/new

How is the SDT term 'percent hits' related to the sensitivity of diagnostic questionnaires?

A correct Yes response to True Yes cases Example: someone has corona and the test is positive (Hit)

How is the SDT term 'false alarms' related to the specificity of diagnostic questionnaires

Incorrect yes response to true no cases

Why is sensitivity in SDT not the same as the sensitivity of diagnostic questionnaires?

SDT: sensitivity (d’) = zHits – zFalse alarm Diagnostics: sensitivity = %Hits

Three methods for culculating SDT measures

- tabular methods

- signal detection theory software

- general purpose software

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo