Vishnu Narayan Kucheria
Assistant Professor,
Department of Psychology
Central University of Haryana, Mahendergarh
Objective
⚫To learn the concept of Signal Detection Theory and
its applications.
Signal Detection Theory (SDT)
⚫ Signal detection theory states that detecting a stimulus
requires making a judgment about its presence or absence,
based on a subjective interpretation of ambiguous
information.
Example: A radiologist is looking at a CAT scan for the
kind of faint shadow that signals an early stage cancer.
Her judgment will likely be affected by her knowledge of
the patient, training, experience, motivation , attention
and knowledge of the consequences of being wrong.
⚫Predicts how and when we detect a signal amid
background noise.
⚫ Assumes no absolute threshold.
HISTORY
⚫(SDT) that was developed for radar applications in the early 1950s.
In World War II radar waves were used to detect enemy aircraft.
⚫The soldiers had to determine if the little spots of light are
enemies, or simple noise (I.e. birds).
⚫ There was no clearly defined criteria for making these kinds of
decisions.
⚫Then applied to research in audition (hearing) in the late 1950s. Some
suggestions were put forward in the early 1970s of how SDT might be
applied in radiology.
⚫SDT was taken into psychophysics, then a century-old branch of
psychology, when the human observer's detection of weak signals, or
discrimination between similar signals, was seen by psychologists as a
problem of inference. In psychology, SDT is a model for a theory of
how organisms make fine discriminations and it specifies model-
based methods of data collection and analysis.
ASSUMPTIONS
⚫ Observers are both sensors and decision makers.
⚫ To evaluate the occurrence of an event, observers adopt a
decision criterion.
⚫ Sensitivity and response bias are independent.
⚫ Noise is always present and is normally distributed.
⚫ When an observer reports that he detects a stimulus, he
is simply making a decision as to whether his sensation
level has exceeded some internal criterion that he has set.
⚫ Signal detection research involves a series of trials in
which the signal is presented only in some trials. In each
trial, the participant must state whether he or she sensed
the stimulus.
⚫Four possible stimulus/response situations in signal
detection theory:
⚫Hit: Stimulus is present and observer responds “Yes.”
⚫Miss: Stimulus is present and observer responds “No.”
⚫False alarm: Stimulus is not present and observer
responds “Yes.”
⚫Correct rejection: Stimulus is not present and observer
responds “No.”
THE CLASSIC RADAR OPERATOR
With this example it is easier to see some of
the effects of different decisions.
In this example if the operator makes a:
⚫ hit. The enemy are engaged perhaps before they even reach land
and might be turned away before they can even attack their
target.
⚫miss. The enemy attach their target and perhaps catch our
planes on the ground which makes them very easy to destroy.
⚫false alarm. Our aircraft take off to no effect. This wastes fuel
which was in short supply at this time. In addition, you can
fatigue both the pilots and their aircraft making them less ready
for the next attack.
⚫correct rejection. The crews and planes get needed rest and fuel
is not wasted.
As you can see, the correct decision is very important. Errors have
negative consequences or costs associated with them.
EXAMPLES
⚫ It is easier to hear a friend talking in a quiet room than
in a room where other people are laughing loudly. And
when your nose is stuffy from a cold, your dinner may
seem to have little flavor. In first case the setting had
made a difference in your sensation and perception. In
second case your physical condition has made the
difference.
⚫ People who smell perfumes for a living have learned
through years of experience, how to detect subtle
differences others would not be able to smell.
APPLICATIONS
⚫Sensitivity or discriminability
Sensitivity refers to how hard or easy it is to detect that a target
stimulus is present from background events. For example, in a
recognition memory paradigm, having longer to study to-be-
remembered words makes it easier to recognize previously seen or
heard words. In contrast, having to remember 30 words rather than 5
makes the discrimination harder.
⚫ Bias
Bias is the extent to which one response is more probable than
another. That is, a receiver may be more likely to respond that a
stimulus is present or more likely to respond that a stimulus is not
present. Bias is independent of sensitivity. For example, if there is a
penalty for either false alarms or misses, this may influence bias. If the
stimulus is a bomber, then a miss (failing to detect the plane) may
increase deaths, so a liberal bias is likely. In contrast, crying wolf (a
false alarm) too often may make people less likely to respond, grounds
for a conservative bias.
APPLICATIONS
⚫Compressed sensing
Another field which is closely related to signal detection theory is
called compressed sensing (or compressive sensing). The objective of
compressed sensing is to recover high dimensional but with low
complexity entities from only a few measurements. Thus, one of the
most important applications of compressed sensing is in the recovery
of high dimensional signals which are known to be sparse (or nearly
sparse) with only a few linear measurements.