To give you some idea of what subjects experience when participating in
a psychophysical threshold experiment.
to gain experience in psychophysical methods and calculation of threshold
under the various methods.
To begin to understand how it is possible to learn about the operation
of human sensory systems.
Theory Of Psychophysical Methods
General Goal and Philosophical Underpinning
Originated by Gustav Fechner - founder, note German, immense capacity for boredom.
Goal - determine lawful relationship(s) between dimensions of the
physical world and resulting aspect of sensation and/or perception.
In other words, how does changing something in a stimulus change what we
experience in our perception.
What do each of these different types of curves mean about how experiences is
related to perception?
Practical Aspect of the Approach - simplify the task and judgment of the
control to within an inch of it's life the physical properties of the stimulus.
That is, know exactly what the stimulus is that you are presenting to the
participant. Have their response be something very simple so you know what
it means, such as: I see something, or no, I don't see something.
Measures (Threshold) or what are we trying to measure about
There are many possible measures. Threshold is only
one. Others will be discussed later.
What is a threshold? (Limen is Latin for threshold) - use room analogy.
Absolute Threshold - Definition: the lowest sensory level that can
be detected reliably.
Difference Threshold (differential) - Definition: minimum stimulus change or
difference from a given level that is reliably detectable. also called Just
Noticeable Difference or JND.
Classical Psychophysical Methods
Method of Limits
A set of stimulus intensities are chosen by the experimeter, say 7.
It is important that the lowest intensity is below the participants threshold
and the highest is above the participants threshold.
The stimuli are presented from lowest to highest intensity or reverse.
I usually like to start from the highest intensity first since it gives
participants some idea of the stimulus they are trying to detect. These
ordered presentations, either from lowest to highest or reversed, are called
After each stimulus, the participant is to report whether they heard the
stimulus or not. If starting with the greatest intensity, the
participant should hear the stimulus.
The staircase continues until the participant no longer responds (there
are variations on this rule). This point where the response changes is
recorded. It is sometimes called a turnaround.
Then the next staircase is begun. It is usually a staircase going in
the opposite direction.
The threshold is calculated as average all of the turnarounds.
Method of Constant Stimuli
This is similar in many ways to the Method of Limits
A set of stimuli are developed going from below the threshold to above the
threshold. Again, say 7 stimuli are chosen.
Then the stimuli are presented in a random order.
Again, the participants response is whether the stimulus is detected or
Threshold is the stimulus intensity that was detected 50% of the time.
It is often the case that no stimulus is detected exactly 50% of the time.
Then a procedure known as linear interpolation is used to estimate this
Method of Adjustment
Usually a range of stimulus intensities, or values, are available, often
continuous or near continuous.
The stimulus intensity is controlled by the participant (either directly or via
The participants task is to move the stimulus to their threshold.
The participant is often instructed to the stimulus until just can or just
cannot detect the stimulus in an absolute threshold experiment.
The starting point is chosen randomly and the procedure is repeated
The threshold is the average of the final settings.
Limitations of Classical Methods
Can you think of Any?
consider ways that the data may be not accurate or
ways that the subjects behavior might lead to an incorrect measure of
Forced Choice Methods
There are several versions. In this class, we will use forced choice
with only two alternatives.
During a single trial, there are two stimuli presented. In an
absolute threshold experiment, one of the stimuli is a stimulus intensity of 0
(or no stimulus) and the other is the stimulus at some intensity.
The two stimuli can be presented at the same time but in different
locations or at different times, one after the other.
If presenting the stimuli at the same time, but in different locations,
which stimulus is in each location is randomly decided.
If presenting the stimuli at different times, which stimulus is presented
first is chosen randomly.
The participant's task is different than in the classic methods.
If the both stimuli are presented at the same time, the task is to
indicate where the target (more intense) stimulus is.
If the stimuli are presented one after the other, the task is to indicate
if the target stimulus is presented first or second.
In both cases, if the participant cannot be sure when or where the target
stimulus is, the participant must guess.
This task is considered to be a pure measure of a participants sensitivity
without any input from a cognitive bias.
Threshold is the stimulus intensity that can be correctly identified 75%
of the time.