Lecture15 Psychophysics

Download as pdf or txt
Download as pdf or txt
You are on page 1of 22

ME 327: Design and Control of Haptic Systems

Autumn 2014

Lecture 15:
Human haptics:
Psychophysics
Allison M. Okamura
Stanford University

psychophysics
the scientific study of the relation between
stimulus and sensation
fundamental to psychology
has become fundamental to understanding
haptic devices and virtual environments
More information & sources of figures in this section: Gescheider,
Psychophysics: Method, Theory and Application, 1984
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

two principal functions


Descriptive: Involves the specification of
sensory capacities
Analytical: Testing of hypothesis about the
underlying biological mechanisms that determine
human sensory capacity

Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

history
1879 Wundt (British empiricist) articulated the
idea of senses as key to human understanding
Simultaneous advances in sensory physiology
Facilitated transition of psychology from a
philosophical to scientific discipline
1860: Fechner published Elements of
Psychophysics: techniques for measuring
mental events
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

measurement thresholds
sensory threshold is a central idea
absolute threshold
sensitivity
smallest amount of stimulus energy required to
produce a sensation
difference threshold
resolving power
amount of change in the stimulus required to
produce a just noticeable difference (JND)
in the sensation
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

Just Noticeable Difference


(JND)
the amount of change in a stimulus that creates a
perceptible increment in sensation
example:
stimulus intensity = 10 units
goes up to 12 units before observer notices a
change
therefore, JND = 2 units at that stimulus level
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

sensory dimensions
intensity or magnitude
e.g., amplitude, frequency
quality or sensory modality
e.g., visual or auditory stimulus
haptic: vibration, force, movement
extension
e.g., size, location, separation
haptic: bump width, space between bumps
duration
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

what its good for: description


Characterizing a
sensory systems
sensitivity has practical
purposes:
Why doesnt delivering
voice signals to a deaf
person through the skin
work very well?
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

psychophysical laws
Empirically derived
Hold true across all senses in many situations
Many such laws exist
Oldest example still has experimental relevance
(after 180 years!)

Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

Webers Fraction, 1834


German Physiologist E. H. Weber

Linear relationship between


differential threshold and
stimulus intensity
For example: to feel
different, 2 heavy weights
must differ more than two
light weights
Gescheider, 1984
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

Weber Fraction
for weights placed on the skin, the Weber
fraction is approximately 1/30
this provides a useful index of sensory
discrimination that can be compared across
different conditions and modalities
however, the WF law is not always perfect,
especially near the absolute threshold (=0)
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

revised Webers Law


when is very close to the absolute threshold,
Webers fraction increases

Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

Gustav Fechner, 1801-1887


there must be
quantitative laws
governing the relations
between mental
sensations and body
stimuli such that changes
in one correspond to
proportional changes in
the other
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

Fechners Law, 1860


Assumption: All JNDs are equal psychological
increments in sensation magnitude, regardless of
the size of

Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

Fitts Law
Fitts Law states that the time to acquire a target
(T) is a function of the distance to (D) and size
(W) of the target
W

index of difficulty
D

For a haptic virtual environment or teleoperation


system, you often want to show that you can minimize
difficulty via haptic feedback
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

psychophysical methods
methods for determining sensory thresholds:
method of limits (and staircase method)
method of constant stimuli
method of adjustment

Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

method of limits
measures absolute
and difference
threshold location
Present subjects with
ascending and
descending
stimulus series, and
ask: is a comparison
>, =, or < a
reference?
Stanford University

absolute difference: average transition


JND: size of equal band

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

staircase method
(Modified Method of Limits)

Begin with high-intensity


stimulus
l Intensity is reduced until
observer makes error
l Stimulus intensity reverses
until subject detects stimulus
l Reversal values averaged
l Multiple staircase methods:
step size
up/down increments
l

Stanford University

H. Levitt, Transformed Up-Down


Methods in Psychoacoustics, The
Journal of the Acoustical Society of
America, vol. 49, 1971, pp. 467-477.

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

example study
Yes

No

Finger fixed in place


with dental gum
l
l
l
l

X, Indentation depth (mm)

Model held at target


depth for 500 ms.

Staircase Tracking Algorithm


4

dB = 2

subjects trained with 5 models


initial indentation is random, below threshold
trial ends when last 10 indentations are within 2 dB
indentation depth required for detection is mean of
last five indentation values

Stanford University

ME 327: Design and Control of Haptic Systems

dB = 1
Reversal
1

dB = 0.5

Reversal
2

indentation depth
defined by:

Allison M. Okamura, 2014

method of constant stimuli


Absolute threshold
location at 50%
JND between 25%
and 75%

repeat same 5-9 stimuli


randomly present each ~100x
detect stimulus? (Y/N)
percent of positive responses calculated for each stimulus intensity
fit curve to get psychometric function (usually s-shaped)
advantage: subject can not predict level of next stimulus intensity
(removes errors of habituation and expectation)

Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

method of adjustment
gives absolute and
difference threshold
locations

l
l

l
l

set stimulus intensity far above or below threshold


subject tunes stimulus intensity to:
l be perceptible (absolute threshold)
l match a reference stimulus (difference threshold)
mean = subjective equality
difference threshold = standard deviation

Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

perceptual and
performance experiments
not all haptic experiments are
psychophysical experiments...
some are perceptual (i.e., they ask different
questions about perception)
some are related to user performance
Stanford University

ME 327: Design and Control of Haptic Systems

Allison M. Okamura, 2014

You might also like