Transcript Chapter 2. The Measurement of Motor Performance

Chapter 2
The Measurement of
Motor Performance
Concept: The measurement of motor performance
is critical to understanding motor learning
Why Study the Measurement of
Motor Performance?
Performance measurement essential for:
Performance assessment / evaluation
Motor learning and control research
Reaction Time
Common measure indicating how long it
takes a person to prepare and initiate a
movement
A stimulus or go signal is the indication to
act
Some type of warning signal also given
What are some motor skill performance
examples in which RT is important for achieving
the goal of the skill?
Reaction Time
Warning
signal
“Go”
signal
Initiation of the
response
Termination of the
response
Foreperiod
Reaction Time (RT) Movement Time (MT)
Response Time
Reaction Time, cont’d
Types of RT situations
– Simple RT: One signal - One response
– Choice RT: More than one signal - Each signal has a
specific response
– Discrimination RT: More than one signal - only one
response
Simple RT
Choice RT
Discrimination RT
Stimulus
lights
Response
key(s)
Index
finger
Index
Middle
Ring
Index
finger
RT Interval Components
EMG enables us to fractionate RT to
obtain more specific information about
movement preparation
Fractionated RT has two components
– Pre-motor time: Quiet interval of time between
the onset of stimulus and beginning of activity
– Motor time: Interval of time from the initial
increase in muscle activity until the actual limb
movement
Fractionated RT
EMG Recording
Pre-motor Time
Go Signal
Motor Time
Initiation of
muscle activity
Observable
Movement
Reaction Time
What do you think occurs in each RT component?
Use of RT in Research
RT has a long history as an “index” to
assess specific aspects of human
performance.
RT is used as a means to
– Infer what a performer does to prepare to
perform an action
– Identify the environmental context information
a person uses to prepare to perform an action
– Assess the capabilities of a person to
anticipate a required action and determine
when to initiate it
Error Measures
Error measures allow us to evaluate
performance for skills that have spatial or
temporal accuracy action goals
What are some examples of skills for which
spatial or temporal accuracy determines
performance success?
Assessing Error for Skills with OneDimension Accuracy Goals
Three error measures
1. Absolute error (AE): Absolute value of
difference between the actual performance
on each trial and the criterion for each trial
– AE = |(performance – criterion)| / no. of trials
– Provides a general index of performance
accuracy
+16, +4,
-10,
+11, -9
= 50/5 = 10
Assessing Error for One-Dimension
Accuracy Goals, cont’d
2. Constant error: Algebraic value of difference
between the actual performance on each trial
and the criterion for each trial
– CE = (performance – criterion) / no. of trials
– Provides an index of a tendency for the
performance error to be directionally biased
+16, +4,
-10,
+11, -9
= 12/5 = 2.4
3. Variable error: The standard deviation of the
CE scores; an index of performance
consistency (i.e. variability)
Assessing Error for One-Dimension
Accuracy Goals, cont’d
See “A Closer Look” on p. 31 for an example
of calculating AE, CE, and VE to
determine the accuracy characteristics of
stride lengths for walking
Assessing Error for Two-Dimension
Accuracy Goals
X2 + Y2 = h2
RE = √ h2
h
5 cm y
When the outcome of
performing a skill
requires accuracy in
the vertical and
horizontal directions
– e.g. Golf putt
x
X-axis distance = 102 = 100
Y-axis distance = 52 = 25
Sum = 125
RE = √125 = 11.2 cm
Radial error = General
accuracy measure for
two-dimensions
– See Figure 2.3
Assessing Error for Two-Dimension
Accuracy Goals, cont’d
Performance bias and
consistency are difficult
to quantitatively assess,
although can do
qualitative assessment
easily
Assessing Errors for Continuous
Skills
Many continuous skills
require spatial
accuracy over a
period of time
– e.g. Driving a car on a
highway
Root-Mean Squared
Error (RMSE)
Common accuracy
measure for
continuous skills
Kinematic Measures
Kinematics: description of motion
without regard to force or mass
Includes the following measures
[see Fig. 2.6]:
– Displacement =Spatial position of a
limb or joint over a period of time
– Velocity = Rate of change in an
object position with respect to time
(i.e. speed)
= Displacement / Time
– Acceleration = Change in velocity
during movement
= Velocity / Time
Kinetics
Kinetics: Force as a cause of motion
Human movements involve both external
and internal sources of force
Importance of force as a movement
measure: All three Newton’s laws of
motion refer to force
EMG Measures
Movement involves electrical
activity in the muscles
Electrodes detect electrical
activity
Electromyography (EMG) =
Recording of muscle electrical
activity
– Common use is to determine
when a muscle begins and ends
activation [see Figure 2.9]
– Also – Recall our earlier
discussion about use of EMG for
fractionated RT as an index of
movement preparation
Brain Activity Measures
Researchers have adopted brain activity
measures commonly used in hospitals and
clinics for diagnostic purposes
Three measures commonly reported in
motor learning and control research
– EEG
– PET
– fMRI
Brain Activity Measures, cont’d
Electroencephalography (EEG): Measures
electrical activity in brain
– Active brain regions produce electrical activity
Positron Emission
Topography (PET):
Neuroimaging (i.e., brain
scanning) technique that
measures blood flow in
the brain
– Active brain regions involve
increased amounts of
blood flow
Brain Activity Measures, cont’d
Functional Magnetic
Resonance Imaging
(fMRI): Neuroimaging
(i.e., brain scanning)
technique that measures
blood flow changes in
the brain by detecting
blood oxygenation
characteristics
Measuring Coordination
Assessment of the relationship of
movement of limb-segments and joints
Quantitative measurement of angle-angle
diagrams
– Cross-correlation technique
– NoRMS
Relative phase
– Relative phase [see Figure 2.10]
Angle-Angle Diagram
Measuring Coordination
Assessment of the relationship of
movement of limb-segments and joints
Quantitative measurement of angle-angle
diagrams
– Cross-correlation technique
– NoRMS
Relative phase
– Relative phase [see Figure 2.10]
Normalized Root Mean-squared
Error
Relative Phase
Index of the coordination between two limb
segments or limbs during a cyclic
movement.
Relative phase ranges from 0 to 180
degrees
– Relative phase near 0 indicates an in-phase
relationship.
– Relative phase near 180 degrees indicates an
out-of –phase relationship.