The Posertm Method: A Biomechanical and Physiological Comparison With Heel-Toe Running

The PoseRTM method : A biomechanical and physiological
comparison with heel-toe running.

FLETCHER, Graham J.
Available from Sheffield Hallam University Research Archive (SHURA) at:
http://shura.shu.ac.uk/20204/
This document is the author deposited version. You are advised to consult the
publisher's version if you wish to cite from it.
Published version
FLETCHER, Graham J. (2006). The PoseRTM method : A biomechanical and
physiological comparison with heel-toe running. Doctoral, Sheffield Hallam University
(United Kingdom)..
Copyright and re-use policy

See http://shura.shu.ac.uk/information.html
Sheffield Hallam University Research Archive

http://shura.shu.ac.uk
Learning and !T Services
Coilegiat 3 Learning. Centre
Collegiate C rescent Cam pus
Sheffield S1028P
101 853 255 2
REFERENCE
ProQuest Number: 10700849
All rights reserved
INFORMATION TO ALL USERS

The quality of this reproduction is d e p e n d e n t upon the quality of the copy subm itted.
In the unlikely e v e n t that the a u thor did not send a c o m p le te m anuscript

and there are missing pages, these will be noted. Also, if m aterial had to be rem oved,
a n o te will ind ica te the deletion.
uest
ProQuest 10700849
Published by ProQuest LLC(2017). C opyright of the Dissertation is held by the Author.
All rights reserved.

This work is protected against unauthorized copying under Title 17, United States C o d e
Microform Edition © ProQuest LLC.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 - 1346
T h e Pose® M e th o d : a b i o m e c h a n i c a l a n d
p h y sio lo g ic a l c o m p a ris o n w ith h e e l-to e
ru n n in g
Graham J. Fletcher
A thesis submitted in partial fulfilment o f the

requirements of
Sheffield Hallam University
for the degree o f Doctor o f Philosophy
December 2006
TABLE OF CONTENTS
Page
Title Page i
Table of Contents ii
List of Tables viii
List of Figures x
Acknowledgements xvii
Abstract xviii
CHAPTER 1: INTRODUCTION 1
1.1 INTRODUCTION TO THE PROBLEM 1

1.2 STATEMENT OF PURPOSE 6
1.3 CHAPTER ORGANISATION 7
CHAPTER 2: LITERATURE REVIEW 9
2.1 INTRODUCTION 9
2.2 RUNNING ECONOMY 9
2.2.1 Introduction 9
2.2.2 Kinematics and running economy 11
Impact patterns 11
Stride length and stride frequency 11
Lower and upper limb kinematics 12
Vertical oscillation of the body’s centre of mass and trunk lean 14
2.2.3 Ground reaction force and running economy 15
2.2.4 Electromyography and running economy 15
The stretch-shortening cycle of muscle contraction 16
2.3 AN OPTIMAL BIOMECHANICAL RUNNING PROFILE 17
2.3.2 Rumiing kinematics 18
Introduction 18
Kinematics of the Centre of mass 19
Spring-mass model 21
Spinal Engine Theory 24
Converted heel-toe runners 24
Stride length and stride frequency 25
Lower and upper extremity limb kinematics 28
2.3.3 Ground reaction force and optimal running biomechanics 31
Introduction 31
The force plate and stride modification 31
Differences between heel-toe and forefoot ground reaction force patterns 32
Vertical ground reaction force 32
The posterior and anterior (horizontal) ground reaction force 33
Stance time 34
Inverse dynamics method 34
2.3.4 Electromyographical research and optimal running biomechanics 35
Introduction 35
Electromyography and muscle force prediction 35
Gait muscles and electromyography activity 36
The stretch-shortening cycle and electromyography 39
2.4 RUNNING INJURIES 42
Introduction 42
Injury Incidence 42
Biomechanical research into injuries 44
Injury and technique 50
2.5 SUMMARY 51
CHAPTER 3: THEORETICAL MODELS 54
3.1 INTRODUCTION 54
3.2 BACKGROUND 54
3.3 TWO HIERARCHICAL MODELS OF RUNNING 58
3.3.2 The current hierarchical model of running 59
Description of the current running model
(adapted from Hay and Reid, 1988) 59
iii
Internal force: Muscle activity 63
Potential strain energy 65
External forces related to running: Ground reaction force 65
External forces related to running: Gravity 67
3.3.3 A Gravitational hierarchical model of running 69
Description of the Gravitational hierarchical running model 71
Support for the Gravitational hierarchical model of running 74
Gravitational torque from mid-terminal stance 75
Leg extension during the propulsive of stance 82
Flight Phase 89
3.4 SUMMARY 90
CHAPTER 4: EXPERIMENTAL PROCEDURES 93
4.1 INTRODUCTION 93
4.2 STATEMENT OF THE PROBLEM 93
4.3 HYPOTHESES 94
(R) <
4.3.1 Study 2 (Pose method training study; Chapter 6) hypotheses 94
4.4 RESEARCH DESIGN 97
4.4.2 Sample size and selection criteria 97
Study 1: Experienced Pose® and heel-toe runner’s study 97
Study 2: Pose® method training study 98
4.4.3 Equipment and design 98
4.4.4 Identification of research variables 99
Study 1: Selection of research variables 99
Study 2: Selection of research variables 100
4.4.5 Treatment of collected data 101
Assumptions 102
iv
4.4.6 Limitations 106
CHAPTER 5: EXPERIENCED POSE® AND HEEL-TOE
RUNNER’S STUDY 107
5.1 INTRODUCTION 107

5.2 THEORETICAL BACKGROUND 107
5.3 METHODS 110
Participants 110
Experimental protocol 111
Collection and processing of ground reaction force data 111
Collection and processing of kinematic data 112
Statistical analysis 113
5.4 RESULTS 113
Ground reaction force variables 113
Motion analysis variables 116
Knee angular displacements 121
5.5 DISCUSSION 143
5.51 Comparison of the two running techniques using the Gravitational
hierarchical model 145
Distance and time 145
Air resistance 146
Relative height of take-off 146
Knee flexion and extension angular velocity duringstance 147
Knee flexion angular velocity from terminal stanceto maximum swing 149
Step time 149
Distance travelled (landing, flight and take-off distance) 149
Mean horizontal velocity of the centre of mass throughout
the gait cycle 150
Mean step length and step frequency 153
Muscle force 155
Ground reaction force 156
Acceleration of the body 159
v
Centre of mass location to the support limb 160
Gravitational force 161
5.6 SUMMARY 162
CHAPTER 6: POSE® METHOD TRAINING STUDY 163
6.1 INTRODUCTION 163

6.2 THEORETICAL BACKGROUND 163
6.3 METHODS 168
Participants 168
Experimental protocol 168
Track test 169
The laboratory tests 169
The intervention 170
Collection and processing of ground reaction force data 170
Collection and processing of kinematic data 171
6.4 RESULTS 173
6.5 DISCUSSION 182
Kinetic and kinematic (primary and secondary) variables 183
Economy and time-trial runs 186
The prospective injury study 189
6.6 SUMMARY 192
CHAPTER 7: CONCLUSIONS 193
7.1 THEORETICAL MODELS 193

7.2 EXPERIENCED POSE® AND HEEL-TOE RUNNER’S STUDY 195
7.3 POSE® METHOD TRAINING STUDY 198
7.4 FUTURE RESEARCH 200
REFERENCES 202
vi
APPENDIX 1: PARTICIPANT’S RECRUITMENT 1
RISK ASSESSMENT PROFORMA 1

PARTICIPANT’S STUDY INFORMATION SHEET 2
PARTICIPANT’S CONTACT SHEET 4
QUESTIONNAIRE 5
INFORMED CONSENT 6
PRE-TEST MEDICAL QUESTIONNAIRE 7
PARTICIPANT’S PERSONAL DETAILS SHEET 9
APPENDIX 2: RUNNING TRACK TESTS 10
TRACK DATA COLLECTION SHEET ECONOMY RUN 10

TRACK DATA COLLECTION SHEET TIME TRIAL RUN 11
TRACK SIGN UP SHEET 12
VIDEO DATA COLLECTION SHEETS 13
BORG SCALE ORIGNAL RATING SCALE (BORG, 1970) 14
APPENDIX 3: LABORATORY TESTS 15
ROUTE MAP TO SHEFFIELD HALL AM UNIVERSITY 15

KISTLER FORCE PLATE CALIBRATION CERTIFICATE 16
PARTICIPANT’S LABORATORY BOOKING SHEET 17
LABORATORY DATA COLLECTION SHEET 18
APPENDIX 4: POSE® METHOD INTERVENTION 19
POSE® RUNNING INFORMATION SHEET 19

POSE® INTERVENTION RUNNING DRILLS 21
CONTROL GROUP’S INTERVENTION DRILLS 25
EXAMPLE TESTIMONIAL 26
INJURY RECORD SHEET 27
APPENDIX 5: RAW DATA (SEE ATTACHED CD DISK)
vii
LIST OF TABLES
Number Title Page
CHAPTER 2
2.3.1 Variables found to be different between rear foot and forefoot strikers
(Williams et al., 2000) 25
CHAPTER 4
4.4.1 Ryan-Joiner’s (similar to Shapiro-Wilk) normality test. 102

4.4.2 Levene’s test for homogeneity of variance. 103
4.4.3 Interactions for the primary research variables
for the 2*2 mixed factorial ANOVA for P = 0.05. 103
4.4.4 Tukey post hoc tests for the 2*2 mixed factorial ANOVA on
the primary research variables (treatment pre test is 1 and post
is 2; the control heel-toe group is 3 and the Pose® treatment group
is 4). 104
CHAPTER 5
5.4.1 Maximum vertical and anterior ground reaction force are recorded as
a percentage of stance and compared with maximum horizontal
acceleration of the centre of mass. Each value is an average of three
trials (N = 4). 138
5.4.2 Two experienced heel-toe and Pose® runner’s mean scores across all
collected speeds. The research variables are in bold (N = 4). 139
5.4.3 Variables are grouped as similar, lower and higher for the Pose®
runners compared with the heel-toe runners (TV= 4). 141
5.4.4 Variables for the heel-toe runners (N = 2) are matched with the literature
at similar speeds. Any minor speed differences are noted after the
v ii i
reference. The thesis’s research variables fall within one standard
deviation of the literature. Over-ground running studies were selected
over treadmill running where possible owing to kinetic and
kinematic differences (Wank et al., 1998). 142
CHAPTER 6
6.3.1 Participant’s personal data. (N = 16). 168

6.4.1 Heel-toe and Pose® group’s (N = 16) mean scores and standard deviations
plus the experienced Pose® and heel-toe model runners at 3.35 m-s*1
in bold. 176
6.4.2 Economy run (2400 m) (N = 16). 180
6.4.3 Time trial run (2400 m) (N = 16). 180
6.4.4 Questionnaire (scale 1-5) How important do you believe running
technique is to overall performance? (N = 16). 180
6.4.5 Weather on time trial and economy run days from Sutton
Bonnington. 181
6.4.6 Weather on time trial and economy run days from Leicester
University. 181
6.4.7 All but two participants continued running their normal weekly
schedules. A self-report injury study was conducted for
three-months post-test (N = 16). 182
LIST OF FIGURES
Number Title
CHAPTER 2
2.2.1 Scatter-plot of running economy versus hamstring-gastrocnemius co

activation during swing (Heise et al., 1996).
2.3.1 Running at 3.8 m s'1, the kinetic energy fluctuations of the centre of
mass are approximately in phase with the gravitational potential
energy fluctuations. Stance phases are identified at the bottom of
the graph (Ferris et al., 1996; p. 260).
2.3.2 A spring-mass model and a stick figure representation of a single stance
phase (Ferris and Farley, 1997). The model consists of a linear spring
representing the leg and a point mass equivalent to body mass. This
figure depicts the model at initial contact (left position), at mid-stance
(middle position) and terminal stance (right position). AL = change in
length.
2.3.3 Variations in stride length and stride rate with increase running
velocity (Luhtanen and Komi, 1978).
2.3.4 Heel-toe (rear-foot) and forefoot ground reaction forces for two
740 N males at 3.35 m s ’1.
2.3.5 Electromyography activity expressed as a percent of gait cycle at
3.35 m-s'1(adapted from Mann et al., 1986). Note the lack of
muscle activity after mid-stance. 1= m. Quadriceps; 2
= m. Hamstrings; 3 = m. Tibialis anterior; 4 = m. Gastrocnemius
2.3.6 Range of motion of knee during sprinting, running and jogging
(Mann et al., 1986). Notice the absence of knee extension during
stance in the sprint condition.
2.3.7 A comparison of the knee eccentric work between heel-toe,
(mid) forefoot and Pose® running (N=20) (Arendse, et al., 2004).
* denotes significant difference from heel-toe and forefoot running.
CHAPTER 3
3.3.1 Flight, take-off and landing distance totalling the runner’s

stride for the centre of mass. Note the runner is falling during
flight. 60
3.3.2 Current hierarchical model of running (Hay and Reid, 1988; p. 282). 62
3.3.3 Results of phasic quadriceps electromyography and knee flexion
-extension for a typical endurance runner averaged over six foot
strikes (adapted from McClay et al., 1990). 64
3.3.4 A typical vertical and anterior and posterior ground reaction force
curve for a heel-toe runner. The first peak is associated with the
shock of contact with the ground (absorption). The second
peak (vertical dashed line) beginning at or about mid-stance
is associated with both body mass and muscle force generation. 67
3.3.5 The Gravitational hierarchical model of ruiming. A gait cycle is right foot
contact to right foot toe contact of the same foot. 70
3.3.6 The running Pose (Romanov, 2002) a) impact to maximal vertical

(R)
ground reaction force: shoulder-hip-ankle vertical alignment,
which potentially allows gravity to maximise the stretch-shortening
reflex and enables the whole body to work as a unit, b) A
gravitational torque accelerates the body while the hamstring muscle
group pulls the support foot from the ground. 73
3.3.7 A torque created around an axis (support limb; ball of foot) from
gravitational force. CM = centre of mass; Fg = gravitational force,
_Ld = moment arm and the axis of rotation is the ball of the foot. 75
3.3.8 a) The vector cross product of the centre of mass’s position vector
XI
and the force of gravity (Fg) create a gravitational torque around the
support limb in the clockwise direction, r = position vector of the
centre of mass from the support limb, b) Fg = gravity, and its
components during stance are Fr, which is the force applied to the ground
and causes no torque while Fp is the propulsive force that causes the
gravitational torque. 77
3.3.9 a) Ground reaction force torque about the centre of mass
b) The ground reaction force and the gravity vector taken at
the same point of stance as the vector product in figure 3.3.8. 78
3.3.10 Gravity and ground reaction force vectors with gravity's
components included, showing the net force is Fp. Notice Fr
is equal and opposite to the ground reaction force vector, leaving
Fp as the net force or the resultant of gravity and ground reaction force. 78
3.3.11 Two identical rectangles are given with their centre of masses marked
as O a) Rectangle upright representing the runner at mid-stance with
their centre of mass above the supporting limb (ball of the foot), b)
The rectangle (runner) tips forwards via momentum. A gravitational
torque around the runner’s centre of mass subsequently moves the
horizontally forwards. Note however, the centre of mass rises owing
to the rectangles geometry geometry and the runner rises via elastic
recoil and leg extension from the rising body. 79
3.3.12 Vector representation of the ground reaction force in running for
two participants (a) A heel-toe runner, (b) A forefoot runner. The
view is of the right shoe from the lateral aspect with the shoe tilted
30° upward (adapted from Cavanagh and Lafortune, 1980). 82
3.3.13 Vertical ground reaction force, during the stance phase of running
xii
(a) vertical velocity of the centre of mass (b) and the vertical position
of the centre of mass (c). The vertical dotted line illustrates the timing
of maximal vertical ground reaction force with zero velocity and the
lowest point of the centre of mass (adapted from Enoka, 2002; p. 192). 83
3.3.14 A gymnast performed a flic-flac (stick figure) showing vertical
(Fz) and posterior-anterior (Fy) ground reaction force. Note
the spike in ground reaction force at push-off with the feet (Yeadon
and Challis, 1994). 84
3.3.15 Force-time recording of a forefoot runner (mass 80 kg, velocity
9.2 m-s'1) (Payne 1983; p. 749). 86
3.3.16 Force-time recording of a heel-toe runner (mass 63 kg, velocity
9.5 m-s'1) (Payne 1983; p.748). 87
3.3.17 Vertical displacement of the centre of mass for a good runner

(14 min 52 s, 5 km) and a poor runner (16 min 29 s, 5 km)
(Miura et al., 1973). 88
CHAPTER 5
5.3.1 Measurement set-up for laboratory data collection. I ll

5.4.1 Vertical and posterior-anterior ground reaction force for
the Pose® runners at 2.5 m-s'1. 114
the heel-toe runners at 2.5 m-s'1. 114
the heel-toe runners at 3.35 m-s'1. 11 4
xiii
the Pose1^ runners at 4.0 m-s'1. 115
5.4.9 Hip flexion and extension for the Pose® runners at 2.5 m-s'1. 117
5.4.10 Hip flexion and extension for the heel-toe runners at 2.5 m-s'1. 117
5.4.17 Vertical oscillation for the Pose® runners at 2.5 m-s’1. 119
5.4.18 Vertical oscillation for the heel-toe runners at 2.5 m-s'1. 119
5.4.19 Vertical oscillation for the Pose® runners at 3.35 m-s'1. 119
5.4.20 Vertical oscillation for the heel-toe runners at 3.35 m-s'1. 120
5.4.21 Vertical oscillation for the Pose® runners at 4.0 m-s'1. 120
5.4.22 Vertical oscillation for the heel-toe runners at 4.0 m-s"1. 120
5.4.23 Vertical oscillation for the Pose® runners at 4.5 m-s’1. 120
5.4.24 Vertical oscillation for the heel-toe runners at 4.5 m-s’1. 121
5.4.25 Knee adduction and abduction for the Pose® runners at 2.5 m-s'1. 121
5.4.26 Knee adduction and abduction for the heel-toe runners at 2.5 m-s'1. 121
5.4.28 Knee adduction and abduction for the heel-toe runners at
3.35 m-s'1. 122
xiv
5.4.33 Knee flexion and extension for the Pose® runners at 2.5 m-s"1. 123
5.4.34 Knee flexion and extension for the heel-toe runners at 2.5 m-s’1. 124
5.4.35 Knee flexion and extension for the Pose® runners at 3.35 m-s'1. 124
5.4.36 Knee flexion and extension forthe heel-toe runners at 3.35 m-s'1. 124
5.4.37 Knee flexion and extension forthe Pose® runners at 4.0 m-s'1. 124
5.4.39 Knee flexion and extension forthe Pose® runners at 4.5 m-s'1. 125
5.4.41 Pose® runner at 2.5 m-s"1for one step. 126
5.4.42 Pose® runner at 3.35 m-s'1for one stride. 126
5.4.43 Pose® runner at 4.0 m-s'1for one stride. 126
5.4.44 Pose runner at 4.5 m-s'1for one stride. 127
5.4.45 Heel-toe runner at 2.5 m-s'1for one stride. 127
5.4.49 The second heel-toe runner at 3.35 m-s’1for one stride. 129
5.4.50 Pose® runner at 2.5 m-s'1. 130
5.4.51 Heel-toe runner at 2.5 m-s'1. 130
5.4.52 Pose® runner at 3.35 m-s'1. 131
5.4.53 Heel-toe runner at 3.35 m-s’1. 131
5.4.54 Pose® runner at 4.0 m-s'1. 132
5.4.56 Pose® runner at 4.5 m-s'1. 133
5.4.58 Ground reaction force, horizontal velocity and acceleration
of the centre of mass for a Pose® and heel-toe runner at 2.5 m-s'1. 134
xv
of the centre of mass for a Posew and heel-toe runner at 4.0 m-s'1. 136
5.5.1 The Gravitational hierarchical model. A gait cycle is right foot contact
to right foot toe contact of the same foot. 144
5.5.2 a) represents the vertical oscillation of a Pose® runner at 4.0 m-s'1.
b) represents a heel-toe runner at 4.0 m-s'1. 152
CHAPTER 6
6.3.1 Measurement set-up for laboratory data collection. 171

6.4.1 Pre-test Heel-toe runner 1 at 3.35 ms'1 taken every 5 frames
at 120 frames.s’1. 173
6.4.2 Post-test Heel-toe 1 runner at 3.35 m s'1taken every 5 frames
at 120 frames.s’1. 174
6.4.3 Pre-test Pose® runner 1 at 3.35 m s'1 taken every 5 frames at
120 frames.s’1. 174
6.4.4 Post-test Pose® runner 1 at 3.35 ms'1taken every 5 frames at
120 frames, s'1. 174
6.4.5 Pre-test Pose® runner 2 at 3.35 m s'1 taken every 5 frames at
120 frames.s'1. 175
6.4.6 Post-test Pose ® runner 2 at 3.35 ms' 1 taken
i
every 5 frames at
120 frames.s"1. 175
6.4.7 Pre (left) and post-test (right) control group (heel-toe) mean
scores for ground reaction force, hip, knee, ankle and vertical
oscillation of the centre of mass. 178
6.4.8 Pre (left) and post-test (right) Pose® group mean scores for
ground reaction force, hip, knee, ankle and vertical oscillation of
the centre of mass. 179
xvi
ACKNOWLEDGEMENTS
I wish to express my warmest gratitude to:
Professor Roger Bartlett, my supervisor at Sheffield Hallam University/Otago University

and Dr. Nicholas Romanov for their expert advice, knowledge and continued
encouragement, without which this thesis would not have been possible.
Rick Velati, High Performance coach for British Triathlon for his tireless pursuit of
suitable participants, for arranging use of the facilities at Loughborough University and
aiding with the video-data collection at the University’s ruiming track.
Dr. Ali Foutoir who offered his time to carefully review the data in order to give relevant
statistical advice.
Dr. Regan Arendse for his assistance during the laboratory data collection.
Dr. Steve Keen for his vigilant and reliable proof reading.
My gratitude to the participants for giving up their time and energy for the study during
their examination period.
My wife, for sustaining me as I worked for many years late at night plus long weekends
on tliis thesis, which took incredible patience on her part, I owe her so much.
To Jesus Christ my Lord and saviour.
xvii
Research into endurance running performance (economy, optimal running biomechanics,
and injury mechanisms) lacks a universal running technique; however, landing heel-toe
identified eighty percent of runners. A new Gravitational hierarchical model o f running was
developed based upon a novel technique Pose® running, owing to deficiencies in the current
hierarchical model (Hay and Reid, 1988) on the interaction of forces involved in running. A
major deficiency in the current hierarchical model is viewing gravity as only active during
flight. In contrast, the Gravitational hierarchical model defines a gravitational torque as the
motive force in running. Ground reaction force is not a motive force but operates according
to Newton’s third law. The ground can only propel a runner forward via muscle activity, but
leg and hip extensor muscles have consistently proven to be silent during leg extension.
High Achilles tendon forces at terminal stance suggest the elastic recoil creates a re-bound
effect in the vertical direction only, thus reducing work against gravity. Two experienced
Pose® runners were compared (study 1) with two experienced heel-toe runners using
primary and secondary research variables derived from the Gravitational hierarchical
model, which effectively distinguished the two techniques while establishing gravity as the
motive force. For example, maximum horizontal acceleration of the centre o f mass occurred
before maximum horizontal ground reaction force, supporting the Gravitational hierarchical
model that only a gravitational torque could create acceleration. Finally, sixteen male
recreational endurance heel-toe runners (study 2 using the same primary variables as study
1) were randomly assigned into two groups where one group received a 7-hour Pose®
intervention over 7-days while the other group remained as heel-toe runners. A 2 x 2 mixed
factorial ANOVA where group (control vs. treatment) and trial (pre to post changes)
assessed the primary research variables. Significant interactions were explored using Tukey
post hoc tests, which found significance (Pose® runners pre-post test) for stance time (P=
0.001), centre of mass to support limb at 25 ms (P= 0.042), centre of mass displacement
during stance (P = 0.001), knee flexion angular velocity during stance (P= 0.005) plus
swing (P= 0.043) and stride frequency (P= 0.002). After 12-days the Pose® group’s post
test time-trial (2400 m) improved by a mean of 24.7 s compared with a 3 s decrease in the
heel-toe group. No significant changes pre-post test, were found for an economy run (2400
m) at 3.35 m-s'1. A preliminary prospective 3-month injury report found no injury incidence
in the Pose® runners compared to six injuries in the heel-toe group. It was concluded, that
the Pose® technique is a valid biomechanical technique, which improved performance,
while reducing injuries. Future work should further quantify gravity’s role in accelerated
running by ascertaining whether the horizontal acceleration of the centre o f mass also
occurs before maximal horizontal ground reaction force.
CHAPTER 1
INTRODUCTION
1.1 INTRODUCTION TO THE PROBLEM
The problem of identifying and correcting errors in athletic technique is an important

part of sports coaching. Technique represents the actions of internal and external forces
on the human body, which determine how the parts of the human body move during a
sports performance (Hay and Reid, 1988). Theories relating to technique tend to develop
through coaches experimenting in a trial and error approach. In professional sport,
athletes push themselves to their body’s limit, which often results in injuries (van
Mechelen, 1992). Dapena (1984) suggested making a biomechanical template from elite
performers technique as a potential solution. Unfortunately, Dapena’s method is little
better than the trial and error approach, owing to a variety of elite performer’s
techniques. This precludes trial and error and elite-template methods of technique
development as a process for long-term success. A more systematic approach is therefore
advised.
Mathematical modelling is a systematic and theoretical approach to analysing sports

movement. It uses physical laws that link the model of the performer and their
movement (Bartlett, 1999). Mathematical equations represent certain characteristics of
the sports movement creating a simplified model of the performance. The mathematical
model undergoes simulation (computer modelling) and finally optimisation (finding
optimum values that minimise or maximise performance outcomes). However, the
complexity of this approach for running is beyond the scope of this thesis.
Hierarchical modelling is another systematic approach. Relationships between the

performance criterion (desired aim of the technique) and performance parameters (key
performance variables) are established using biomechanical laws (Bartlett, 1997).
Development of a hierarchical model precedes data collection (Bartlett, 1997).
1
Following the model’s development, a data collection and analysis for a large range of
performers is usually completed. Finally, results from the data are assessed with respect
to the hierarchical model (for example, Hay et al., 1981). A hierarchical model functions
as the ‘gold standard' enabling variations in performance to be quantitatively measured.
Hierarchical models can also evaluate individuals or small groups of athletes’ technique
as opposed to a large range of performers (Bartlett, 1999). This thesis aims to review an
established hierarchical model for running (Hay and Reid, 1988; p. 282) while also
developing a new Gravitational hierarchical model. The Gravitational hierarchical model
will attempt to develop the interaction of the forces involved in running based on gravity
as the motive and hierarchically leading force in running.
If force is applied as input to a biomechanical system, the system undergoes a

displacement. Ensuing motion creates opposing forces, which are proportional to the
displacement. An external force can be evaluated by the degree of energy needed for
internal forces used in the displacement. External forces that cause displacement, which
do not require large reciprocal internal forces will obviously decrease energy output
(Bernstein, 1947). There are two key external forces (ignoring air resistance) involved in
running: gravity and ground reaction force. Then* combined effects on internal muscle
force during the gait cycle need further examination (for example, Fenn, 1930 began
work in this area). However, gravity’s role as a constant external force in running during
stance as well as flight, and its subsequent affect on muscle output is not apparent from
Hay and Reid’s hierarchical model.
Endurance running instead, has attempted to explore biomechanical research in three

main areas. The first area is running economy, which refers to the energy cost of running
(Daniels, 1985). The next area is determining an optimal biomechanical profile
(Williams, 1985) and finally, it is training without injury (van Mechelen, 1995).
Endurance running, including track, road, cross-country and triathlon, has many
unanswered questions today from these three areas (Cavanagh, 1990; Novacheck, 1998).
For example, the degree to which running biomechanics can be modified to promote a
better understanding of the association between metabolic and biomechanical correlates
of economy, requires further investigation (Williams and Cavanagh, 1987). Rumiing is a
fundamental movement pattern, yet the literature has not identified an optimal
biomechanical profile that is universal for all runners. Injury incidence remains between
24-77% (van Mechelen, 1992; Taunton et al., 2002), and this incidence rate has not
changed since the 1970’s running boom (Krissoff and Ferris, 1979).
Cavanagh (1990; p. 31) stated in a review of the running research literature over the last
one hundred years; “that today we are basically still tiying to answer questions posed by
Fischer and Fenn over 90 years agoF For example, Fenn (1930) suggested more
economical runners might waste less energy on forward pressure on the ground.
The literature describes two main running techniques, heel-toe and forefoot ground
contact. Approximately eighty percent of endurance runners are heel-toe runners,
making them the predominant technique compared to twenty percent of forefoot runners
(Kerr et al., 1983). Therefore, heel-toe running will be used as the current norm for this
thesis. Elite endurance heel-toe runners have a few biomechanical variables in common,
such as increased forward trunk lean (Girardin and Roy, 1984), increased knee flexion
during stance (Kerdok, et al., 2002) and decreased vertical oscillation (Williams and
Cavanagh, 1987). However, there are exceptions to these findings (Cavanagh and
Williams, 1982), suggesting the heel-toe technique is identified predominantly by initial
ground impact via the heel.
Several descriptions o f ‘good’ running have also been suggested in the trade literature:
“Good runners run tall, they tend not to hunch or lean. There is an economy and
integrity in their form and movement. Their heads remain poised and balanced on top o f
their spine. They run smoothly, lightly with the brakes off, they don 't pound or ‘excavate ’
on every stride. Good runners run with awarenessF (Alexander, 1989; p. 12).
Percy Cerutty, coach of Olympic gold medal winner Herb Elliot states:
“The head rests loosely on the shoulders, that is, it is not held rigid. It should be capable
o f movement as the needs o f the athlete demand. In my techniques I often test this rigidity
o f an athlete. Many are quite incapable o f turning their heads freely on their neck and
shoulders. Any rigidity here spreads right through the whole musculature. Keep the head
and neck free and the rest o f the moving parts will tend to be freeF (Simms, 2003; p. 31)
The trade and scientific literature have only made limited suggestions on improving
running technique, because no one has developed a running method that explains how to
effectively combine all the forces involved in running. Hay and Reid (1988) provide the
only attempt at modelling running. However, there are deficiencies with their model, for
example, viewing the constant force of gravity as only active during flight and not
3
relating gravity to forces exerted (muscle and joint forces) and the time forces act despite
gravity being a constant force.
In 1977, in the Soviet Union, a track and field coach and university lecturer in
biomechanics was wrestling with the question of how to teach rumiing. Dr. Romanov
rejected popular wisdom that running is a simplistic movement skill, and that the best
runners were genetically predisposed to success combined with hard training. He sought
a comprehensive technique that would create a universal running method. He
subsequently viewed teaching of other track and field events, throwing, jumping, and
other movement skills such as ballet, dancing and the martial arts. Technique, he
concluded, was of paramount importance for these sports, because, he realised that
complex skills such as ballet and karate are broken down into simple movements that
can be learned separately (i.e. Poses). Complex movements therefore derive from
individual Poses. A complex skill is achieved, only after each Pose through numerous
repetitions has thoroughly trained the neuromuscular pathways.
Using videotape of the world’s best runners, Dr. Romanov identified the rumiing Pose®.
The Pose® method technique is based upon a running Pose®, which is a whole body
Pose , that vertically aligns the shoulder, hip and ankle of the support leg, while standing
with the weight on the ball of the foot. This creates a ‘S’ like body shape. The runner
then moves from this Pose® on one leg to the other by potentially falling forward via a
gravitational torque. The support foot is pulled vertically upwards from the ground using
the hamstring muscles as the body falls forward, while the ipsilateral leg is not driven
forwards during flight but allowed to fall to the ground via gravity to land in the next
running Pose®. Many technique and strengthening drills have been developed to teach
the runner to fall forwards while pulling their support foot from the ground using their
hamstring muscles, and at the same time not driving their swing leg forwards via the hip
flexor muscles (Romanov, 2002). Romanov (2002) states the running Pose® is a
theoretically optimal position for external forces to cause displacement, without
requiring large reciprocal internal forces.
Dallam et al. (2005) identified a profile of economical endurance runners from the
literature. This profile included a reduced vertical oscillation, stance time and peak
vertical ground reaction force, alongside increased trunk lean and knee flexion during
4
support. Eight heel-toe male endurance runners trained in the Pose® method by Dallam
et al. (2005) found the above variables described Pose® running remarkably well, yet
their economy measures showed poorer economy for their Pose^ runners. These
conflicting results may originate from running on a treadmill, which inhibits Pose®
running owing to the foot being pulled backwards. Hence, clarification is needed on
whether Pose® running is more economical using over-ground running.
Arendse et al. (2004) explored Pose running (3-years after Dallam et al., 2005) in
reference to its similar lower-limb geometry with backward running in the hope it may
aid in injury prevention. They identified a distinct biomechanical profile for Pose®
running (supporting Dallam et al., 2005) compared to heel-toe and forefoot running.
However, they administered no performance test nor did they collect injury data for their
runners. This thesis will rectify these omissions and will attempt to explain the distinct
biomechanical profile of Pose® running.
Running technique is currently dependant on whether a runner lands on the forefoot or

heel-toe (Kerr et al., 1983). A universal running technique (which applies to all runners)
that identifies the timing of all the forces with the displacement and timing of the body
segments during the gait cycle would be useful. The literature has compared elite with
recreational runners in the hope of distinguishing an elite runner’s profile (Miura et al.,
1973; Williams and Cavanagh, 1987). This method though has failed to establish a clear
universal biomechanical profile of an elite runner. This thesis will take a different
approach. Owing to the scale of unanswered problems in endurance running within the
literature, the author seeks to begin with the Pose® method (Romanov, 2002) as a
universal running technique. Then, evaluate the Pose® method theoretically to determine
whether it is a universal running technique using the new Gravitational hierarchical
model. A biomechanical comparison with the predominant technique of endurance
running (heel-toe) using experienced heel-toe and Pose® runners will follow. Next,
evaluate the Pose® method’s effectiveness with a group of male heel-toe endurance
runners by training half of them in Pose" running and measuring performance via a 2400
m time trial and a 2400 m economy run. Biomechanical variables will distinguish
between the two rumiing techniques using inferential statistics. Finally, using a 3-month
prospective injury study determine injury incidence in the Pose® and heel-toe ruiming
techniques.
1.2 STATEMENT OF PURPOSE
Several beliefs predominate in the running community that have hindered the
development of a universal technique of running. For example:
1. Running technique is a simplistic movement pattern;
2. Individual differences between people make it impossible to have a universal
technique for all runners (Nytro, 1987).
3. Various distances and speeds require a different running technique;
4. There are numerous points of view based on subjective and unsubstantiated
non-biomechanical models (Wallack, 2004); and
5. Technique being solely associated with ground contact, i.e. eighty percent of
all endurance runners land on their heels (heel-toe runners) and twenty
percent of all endurance runners land on the front part of their foot (mid and
forefoot strikers) (Kerr et al., 1983; Whittle, 1991), whereas sprinters land on
the forefoot (Mann, 1980).
6. Hay and Reid's (1988; pp. 287-288) hierarchical model highlights the current
thinking that runners push off the ground via leg extension. The forces
exerted box in their model represents shoulders, hips, knees, ankles and other
joint forces (Hay and Reid, 1988; pp. 264 and 287-288). This focus on leg
extension via muscle force has hindered development of other viewpoints on
propulsive forces in running such as gravity being motive.
The primary aim of this thesis is to compare the Pose® method technique (Romanov,
2002; Arendse et al., 2004; Dallam et al., 2005) with the heel-toe running technique.
Development of a new Gravitational hierarchical model based on the Pose® method
using the laws of physics will enable the Pose® technique’s systematic evaluation. The
Gravitational hierarchical model and the current hierarchical model (Hay and Reid,
1988; p. 282) are reviewed highlighting the different biomechanical theories of running.
There are deficiencies with Hay and Reid’s model, for example, viewing the constant
force of gravity as only active during flight and not relating gravity to forces exerted and
the time forces act. Therefore, Hay and Reid’s model does not systemise the forces
involved in running but acts as general descriptor, whereas, the Gravitational hierarchical
model will aim to structure these forces. A comparison of the Pose® and heel-toe running
techniques will follow using two experienced male Pose® and heel-toe runners utilising
6
the primary and secondary research variables (study 1; chapter 5) derived from the
Gravitational hierarchical model. Upon establishment of the Pose® method technique, a
group of male heel-toe recreational runners (study 2; chapter 6) will participate in a
repeated measures design. An economy and time trial run will give the participants pre
test baseline physiological measures while the primary research variables will provide a
biomechanical profile. The post-test will determine whether there is an improved
performance on the economy and 2400 time trial run resulting from the potential
biomechanical changes in the Pose® group from a 7-hour Pose® training intervention. A
2 x 2 mixed factorial ANOVA will assess the main effects of group (control vs.
treatment) and trial (pre to post changes) on the primary research variables. Tukey’s tests
of honestly significant differences will be used to assess individual cell differences post
hoc. A prospective 3-month injury study on all the participants post intervention will
seek to determine injury incidence between the two running techniques.
Currently unanswered research questions on running biomechanics and their effects on

performance, which this thesis aims to answer, include:
• Can the Gravitational hierarchical model based on the Pose® running technique
clarify the relationship and timing of the forces involved in running?
• Can the Pose® running technique improve performance (economy and speed)?
• Can running injuries be reduced by the application of the Pose® running
technique?
1 .3 CHAPTER ORGANISATION
Chapter 2 reviews the literature on running economy, optimal running biomechanics,

and injury mechanisms. These topics are reviewed using data collection tools, such as, a
force plate, electromyography and two or three-dimensional analysis. The findings are
then summarised to clarify whether a potential universal running technique is evident
from the literature.
Chapter 3 conducts a detailed analysis of the forces involved in running while

evaluating the Pose® method to determine if it is a universal running technique. A new
7
Gravitational hierarchical model is developed and compared with Hay and Reid’s (1988;
p. 282) hierarchical model of running.
Chapter 4 evaluates the experimental procedures. A statement of the problem is given.

Hypotheses and the primary and secondary research variables derived from the
Gravitational hierarchical model are stated. The research design and statistical model are
specified.
Chapter 5 compares data for study 1 from two experienced male participants in the
Pose® and heel-toe techniques of running. Kinetic and kinematic (primary and
secondary) variables established from the Gravitational hierarchical model, distinguish
the two running techniques. The Pose® method is discussed using the Gravitational
hierarchical model.
Chapter 6 presents results from study 2 using sixteen male heel-toe endurance runners,
where eight were trained for 7-hours in Pose® running (treatment/Pose^ group). Sixteen
2 x 2 mixed factorial ANOVAs will assess the main effects of group (control vs.
treatment) and trial (pre to post changes) on the primary research variables
(physiological and biomechanical) derived from the Gravitational hierarchical model.
Tukey’s tests of honestly significant differences will be used to assess individual cell
differences post hoc. A 2400 m time trial and economy run will record physiological
performance, optimal rumiing biomechanics will be documented via a laboratory
assessment and a 3-month prospective injury study will record injury incidence.
Chapter 7 summarises the findings of the previous chapters and suggests possible areas
for future research.
CHAPTER 2
LITERATURE REVIEW
2.1 INTRODUCTION
This chapter follows biomechanical outlines from several comprehensive reviews of

running (Williams, 1985; Novacheck, 1998): Economy, an optimal biomechanical
profile and sustained injuries. Each one of these areas aims to improve running
performance. Running economy research attempts to identify patterns of motion which
improve economy and hence performance. Identification of an optimal biomechanical
profile may reveal a universal running technique, which should improve performance in
athletes of all abilities. High injury prevalence in runners can negatively influence
performance.
2.2 RUNNING ECONOMY
2.2.1 Introduction
Economy and efficiency of running are often incorrectly applied.

“Efficiency o f movement refers to the mechanical energy/ produced relative to the
metabolic energy’ used to cause the movement'' (Robergs and Roberts, 2000; p. 69).
Mechanical work is a key component of the work-energy relationship for efficiency of

movement, which divides into two areas during constant speed running. Firstly, external
work is the sum of work required to accelerate the centre of mass of the whole body, plus
work completed against gravity. Secondly, are the work done by the limbs around the
centre of mass, which is internal work. External work is higher than internal work up to
speeds of 5.5 m-s'1, above this speed the work ratio is reversed (Cavagna and Kaneko,
1977). A major problem arises, however, estimating the storage and return of elastic
9
energy associated with high muscular efficiencies (Cavagna et al., 1964; Williams and
Cavanagh, 1983; Kubo et al., 2000). In eccentric exercise, mechanical energy can
increase linearly, however, this appears to be individual owing to increases in stretching
velocity of lower-limb muscles (Auro and Komi, 1986). Although initially this method
seemed plausible, mechanical efficiency measures have varied enormously (22.7-75%)
making subtle observations of mechanical changes in runners obsolete (Margaria, 1968;
Asmussen and Bonde-Petersen, 1974; Kaneko et al., 1983; Bosco et al., 1987). Running
economy presented itself as an appropriate and alternative measure of energy
expenditure associated with running. Economy and efficiency however, are not
synonymous (Robergs and Roberts, 2000).
Running economy refers to the energy cost of performing movement (Bransford and
Howley, 1977; Daniels, 1985). Research into running economy tries to identify patterns
of motion that improve economy and hence performance. Economy is defined as, the
oxygen cost per kilogram of body mass per kilometre run (Cavanagh and Williams,
1982; Morgan et al., 1989; Bergh et al., 1991; Noakes, 2002) with substantial individual
variation in running economy observed among those running at the same speed
(Mayhew, 1977; Daniels, 1985). Meaning, economical runners consume less oxygen
relative to body mass when running at a submaximal intensity. Changes in running
economy link to the biomechanics of running technique (Bransford and Howley, 1977;
Cavanagh and Williams, 1982; Daniels, 1985) with 54% of the variation in running
economy attributed to biomechanical variables (Williams and Cavanagh, 1987). Further
variance may have been unidentified owing to nonlinear relationships with the
biomechanical variables. Increased running training (volume) can change running
economy in individual runners (Bransford and Howley, 1977; Conley and Krahenbuhl,
1980). However, several studies have shown no improvement in running economy when
training was increased (Costill et al., 1973; Wilcox and Bulbulian, 1984). Furthermore,
intra-individual oxygen uptake variation exists between different test dates and running
speeds (Morgan et al., 1987). Nevertheless, the term economy is the accepted
physiological criterion for efficient running performance (Cavanagh and Kram, 1985).
The next section discusses biomechanical variables associated with economical ruiming.
10
2.2.2 Kinematics and running economy
Impact patterns
Research identifies two patterns of landing in runners: forefoot and rear foot or heel-toe
(rear foot and heel-toe are used synonymously for this thesis) runners (Whittle, 1991).
Nilsson and Thorstensson (1987) reported no significant difference in running economy
between a group of six forefoot and six rear foot strikers; even when the runners
voluntarily changed from fore to rear foot and vice versa. In contrast, Williams and
Cavanagh’s (1987) work indicated that heel-toe runners maybe more economical in their
running style, although this was not consistent across their participants. For example, the
fastest 10-km runner from their study was a heel-toe runner who had the third poorest
economy. However, impact patterns by virtue o f the kinetic chain are influenced by
stride-length, stride frequency, joint angles, and the position of the centre o f mass at
impact and terminal stance (Mero et al., 1992).
Stride length and stride frequency

Stride length is defined as successive contacts of the same foot, with step length (half a
stride) as successive contacts of opposite feet (Cavanagh and Williams, 1982). Stride or
step length is used depending upon the paper reviewed. However, step length can denote
the distance the centre of mass travels during flight (Cavagna et al., 1976). Stride length
is one of the few variables established by direct experimental evidence to affect running
economy (Anderson, 1996). Curvilinear increases in aerobic demand occur when stride
length increases or decreases at any given speed from one self-selected by the runner
(Hogberg, 1952; Knuttgen, 1961; Morgan et al., 1989; Kyrolainen et al., 2000a).
Cavanagh and Williams (1982) report running economy as the primary reason for
runners— 10 well-trained participants—making self-selected stride lengths and stride
frequencies. The optimal economical solution identified was a combination of self
selected stride lengths and stride frequencies. Cavanagh and Kram (1990) documented
that stride length increased preferentially over stride frequency with increasing speed up
to 7 m-s"1. Again, it appears that economy dominates the choice of stride length. Martin
et al. (2000) state the typical explanation for this strategy (i.e., change in stride length
rather than stride rate) is that it requires less energy to lengthen the stride within
reasonable limits than to increase stride rate.
11
A four-year study by Nelson and Gregor (1976) in contrast observed varsity distance
runners decrease their stride length at a given speed as their performance improved. This
perhaps suggests that freely chosen stride length is possibly a variable quantity over a
runner’s life or they self-selected over-striding patterns to begin with. Morgan et al.
(1994) monitored runners who self-selected an over striding gait pattern. They responded
positively to step length optimisation training, because economy improved as step length
decreased.
Rowland (1989) and Unnithan and Eston (1990) noted boys produced shorter stride
lengths and higher stride frequencies than men, but the oxygen demand per stride was
similar in both groups. It is possible that height and body mass in boys affect movement
differences with their lower-limbs, which constitutes a substantial part of metabolic
expenditure. Williams and Cavanagh (1987) however, found no differences in
anthropometric measures for segmental lengths and masses in relation to an adult
rumier’s economy.
In reference to step frequency, Kaneko et al. (1987) measured energy consumption

during running at varying step frequencies while speed was kept constant at 2.5 m-s'1, 3.5
m s'1 and 4.5 m-s'1. Energy was minimized for all three speeds at approximately the same
step length, at about 2.9 steps-s'1, corresponding closely to the freely chosen step
frequency of 2.8 steps-s'1. Kyrolainen et al. (2000b) used seven triathletes who ran a
marathon to induce fatigue; step frequency increased and step length decreased. The
authors suggested the triathletes compensated for impaired neuromuscular function with
increased muscle activity, causing weakened economy. Cavanagh and Kram (1985)
using twelve endurance runners recorded step frequency for running speeds of 3 m-s'1 to
4.2 m-s'1. Step frequency increased by 4%, in contrast to a 28% increase in stride length
over the speed range. However, two runners had similar stride frequency across the
whole speed range. They concluded that there might be an individual economical stride
frequency at different speeds in distance running.
Lower and upper limb kinematics

Maximum thigh extension at terminal stance had a positive correlation (r = 0.53) with
increased economy (increased oxygen consumption) from 13 elite males (Williams and
Cavanagh, 1986). In contrast, thigh extension in 31 recreational runners showed no
12
correlation with oxygen consumption (Williams and Cavanagh, 1987). Williams et al.
(1987) did not provide an explanation for these opposing results. Miura et al. (1973)
associated thigh extension with increased push-off at terminal stance, identifying this as
‘good' and associated with their elite runners (14 min 48-56 s for 5 km), but recorded no
economy data to support this. Further, elite runners’ biomechanical profiles—thigh
extension—may differ from recreational runners in this respect, owing to running speed
differences.
Cavanagh et al. (1977) using elite runners demonstrated that more acute knee angles
during the swing phase enhanced economy. Theoretically, greater knee flexion allows a
faster recovery of the lower-limb by decreasing the moment of inertia (Bailey and Pate,
1991). In support, Enomoto et al. (1999) recommended a rapid recovery of the thigh
required a flexed lower-limb, and this action was associated with efficient runners.
Martin (1985) and Meyers and Steudel (1985) added leg weights to the distal end of
runners’ lower-limbs, producing oxygen consumption submaximal increases. It was
evident an increased moment of inertia for the lower-limb reciprocally increased oxygen
consumption.
Maximum knee flexion during support has also been associated with improved running
economy (Williams and Cavanagh, 1987). Maximum knee flexion correlated
significantly with less contact time, shank angle at impact and the second vertical ground
reaction force peak. Improved economy was also associated with less extension of the
knee (in contrast to Miura et al., 1973) during terminal stance, and less rapid knee
flexion velocity during stance. Williams and Cavanagh (1987) speculated lower-limb
kinematics cause subtle changes in muscular activity patterns before and during support
that are related to differences in metabolic energy costs. An alternative explanation is
lower knee flexion, shank angle at impact, plus the second vertical ground reaction force
peak and the reduction in contact time, were a result of the centre of mass moving
anterior of the support foot more quickly allowing earlier forward progression.
Maximum ankle plantar flexion angles at terminal stance were a mean of 10° more for
good runners (mean marathon time 2:34:40) than elite runners (mean marathon time
2:15:52) (Cavanagh et al., 1977). Decreased plantar flexion would logically be
associated with less push-off and lower muscle force. In support of this idea, Williams
13
and Cavanagh (1987) found more economical runners had 6.4° less plantar flexion at
terminal stance than their least economical runners.
Williams and Cavanagh (1987) and Anderson and Tseh (1994) demonstrated a trend for
more economical runners to use less arm movement, measured by wrist excursion during
one stride. These economical runners have lower arm amplitudes, which may indicate a
skilled-balance mechanism rather than a method of forward drive. Arm action reduced
side-ways rotations in runners giving further support to this theory (Hinrichs, 1987).
Mann (1981) also concluded from extensive analysis of sprint running that the upper-
limb’s contribution is balance maintenance only. Therefore, arms appear not to
contribute to the drive of the body in the anterior and posterior direction (Hinrichs et al.,
1987), and were not studied in this thesis.
Vertical oscillation of the body’s centre of mass and trunk lean

Williams and Cavanagh (1987) found a consistent, but non-significant trend for oxygen
cost, associated with lower vertical oscillation of the body’s centre of mass in good and
elite runners. Logically, excessive vertical oscillation would be ineffective for horizontal
displacement, yet many individuals run economically despite high vertical oscillation.
There is however, some general support for improved ruiming economy with lower
vertical oscillation (Slocum and James, 1968; Cavanagh, et al., 1977). Several
possibilities exist for the conflicting results. Firstly, measurement of vertical oscillation
differed between using head-markers or displacement of the centre of mass. The use of
head-markers assumes that trunk and neck angles remain neutral for each runner.
Secondly, the variety of biomechanical variables associated with running economy, and
their various combinations, make identification of one variable as an absolute
economical variable in all runners unlikely. The source of vertical oscillation should be
an important economy inquisition. Vertical oscillation can transpire from leg extension
or the reciprocal flexion and extension of the stretch-shortening cycle. If certain runners
predominantly utilise the stretch-shortening cycle, which requires lower muscle activity
(Cavagna et al., 1964), this provides a possible explanation for economy differences.
Williams and Cavanagh (1987) found runners with a mean forward trunk lean, relative to
the vertical axis, more economical at 5.9° compared with 3.3° and 2.4°. However,
excessive forward trunk lean may create increases in economy (James and Brubaker,
14
1973). There appears to be an optimal position for the trunk and by association the centre
of mass position, to maintain postural alignment for improved running economy.
2.2.3 Ground reaction force and running economy
Increased stance time correlated positively (r = 0.49) with poorer economy (Williams
and Cavanagh, 1986; Williams and Cavanagh, 1987), yet not all studies registered this
relationship (Cavanagh et al., 1977). An explanation for variations in stance time in the
three studies is use of both fore and / or heel-toe runners. For example, Williams et al.
(1987) used elite females (with no heel-toe or forefoot classification), whereas Williams
and Cavanagh (1987) used recreational runners, who were both forefoot and heel-toe
runners. Stance time differs between forefoot and heel-toe runners (Arendse et al., 2004).
Williams and Cavanagh (1987) found significantly lower first peaks for vertical ground
reaction force, plus less anterior and posterior ground reaction force in economical
recreational runners. They suggested lower extremity kinematic demands just before
ground contact, might have a significant affect on muscles both before and during
support, which in turn can affect economy. Forefoot strikers rely more heavily on the
musculature to assist with cushioning than heel-toe runners, who depend more heavily
on footwear and skeletal structures to cushion and support impact force (Williams and
Cavanagh, 1987). Kyrolainen et al. (2000a) speculated lower performances from poorer
economy stem from limited action of the hamstrings. They linked poor economy with
limited hamstring activity and higher braking forces at foot contact. Heise and Martin
(2001) found total vertical impulse and net vertical impulse correlated with running
economy (r = 0.62, r = 0.60, respectively) indicating a moderate correlation for stance
time and the amount of force applied during stance. Inferred, was increased muscle
support during stance produced reciprocal increases in economy.
2.2.4 Electromyography and running economy
Relationships between running economy and the temporal electromyography

characteristics of biarticular leg muscles were quantified in a group of well-trained
runners (Heise et al., 1996). Nine runners completed three test sessions: a determination
of maximal aerobic demand, an accommodation session at the experimental speed of
15
4.13 m-s"1 and a session during which electromyography and running economy data were
collected at 4.13 m-s'1 (see figure 2.2.1). Measures of muscle onset, on-time durations
and on-time co-activation durations were calculated from: rectus femoris, medial
hamstrings, lateral hamstrings and gastrocnemius muscles. Earlier onset of rectus
femoris during swing phase and a shorter duration of hamstring-gastrocnemius co
activation during swing were associated with economical runners. Individuals who
exhibited more co-activation (improved coordination) between biarticular muscles
during the stance phase of the running cycle tended to be more economical. Economical
runners produced different activation patterns with their biarticular muscles, which cross
two joints— ankle-knee and knee-hip— indicating economical runners had improved
segmental coordination.
48_
46—
0 5 10 15 20 25 30 35 40
C o -a c t iv a t io n d u r in g s w in g (% s w in g ) .
Figure 2.2.1 S c a tte r -p lo t o f r u n n in g e c o n o m y v e r s u s h a m s tr in g - g a s t r o c n e m iu s c o - a c t iv a t io n d u r in g
s w in g ( H e is e et al., 1 9 9 6 ) . N o t e th e p o s it io n o f t h e t w o m o s t u n e c o n o m ic a l r u n n e r s.
Kyrolainen et al. (2000a) using seventeen well-trained endurance runners, found

electromyography for the biceps femoris— a biarticular muscle increased across
thirteen different speeds. As running speed increased, power in the lower-limb joints
increased in the ground push-off phase. This provided a weak correlation with muscle
activity and energy expenditure (r = 0.48 at p < 0.05). However, no economy data were
collected.
The stretch-shortening cycle of muscle contraction

The mechanical efficiency of running exceeds the efficiency of the conversion of
chemical energy to kinetic energy by muscles (Cavagna and Kaneko, 1977; Winter,
16
1978; Cavanagh and Kram, 1985). Elastic strain energy stored during eccentric
contractions of running is subsequently released during concentric contractions, which
makes a substantial contribution to propulsion (Cavagna et al., 1964; Cavagna and
Kaneko, 1977; Hinrichs, 1990; Jung, 2003). A reduction in support time aids utilization
of stored elastic energy in the muscle mechanisms, (Zatsiorsky, 1995; Paavolainen et al.,
1999a). Although there is no method of quantifying storage of elastic energy, there is a
consensus that this phenomenon contributes to both efficiency and economy of
movement (Cavagna and Citterio, 1974; Cavagna et al., 1977; Winter, 1978; Luhtanen
and Komi, 1978; Cavanagh and Kram, 1985; Komi, 2000). The runner’s body possesses
kinetic energy while the muscle-tendon unit contains potential strain energy. An optimal
and complimentary combination of potential strain energy and conserved kinetic energy
would theoretically improve a runners’ economy.
2.3 AN OPTIMAL BIOMECHANICAL RUNNING PROFILE
2.3.1 Introduction
Cavanagh et al. (1985; p. 404) elaborated on whether elite rumiers exhibit a

biomechanical profile that is substantially different from recreational runners:
“Many obsemers might suggest that elite athletes have practiced their event so
frequently as to have self-optimised their movement patterns. It is a mistake, to assume
elite athletes have perfect body structures and ideal movement patterns. The elite athlete
cannot succeed without excellent physiological attributes. However, he or she can
succeed in spite o f poor running mechanics and body structure
They highlighted that identifying ideal running biomechanics from elite runners can be
problematic. This section will use elite and recreational runners, to appraise kinematic
analysis, ground reaction force and electromyography. Several points are worth noting
when reviewing these areas of biomechanical research. Firstly, as stated earlier there are
two types of foot contact with the ground in running. Ken- et al. (1983) and Whittle
(1991) identified eighty percent of all runners land on the heel (heel-toe runners) and
twenty percent of all runners land on the front part of the foot (forefoot rumiers). As
speed increases, rumiers often change from a heel landing to a forefoot landing (Kerr et
al., 1983; van Mechelen, 1995). Secondly, attempts to quantify running biomechanics
fall within two methods (Williams and Cavanagh, 1987): measures sensitive to the entire
17
running cycle such as mechanical power, while other studies measure certain variables at
or between discrete points within the gait cycle, such as knee flexion at impact.
No definitive relationships exist between mechanical power and economy (Norman et

al., 1976; Kaneko et al., 1983). Calculated power varies widely between studies
depending on which assumptions were made (Norman et al., 1976; Williams and
Cavanagh, 1983; Morgan et al., 1989), for example, inaccuracies in modelling the runner
as a multi-segment rigid body or choice of filter and cut-off stride frequencies applied to
the kinematic data (Arampatzis et al., 2000). Mechanical power is, therefore, not a
sensitive enough measure for running performance evaluations. Williams (1990) stated,
“that a multitude o f variables may be used to describe running action and it is therefore
difficult ‘a priori ’ to know which variables to select.” Selecting too few may overlook
vital variables while collecting too many variables may not enable identification of
relationships between them. This problem confounds determination of an optimal
running profile. Collecting many variables and using a factor analysis procedure to
identify independent sets of variables is one method (Williams and Cavanagh, 1987).
From this large number, a smaller set of variables may then be analysed in more detail.
Williams (1990) recommended this method for each study even though this could be a
time consuming approach.
There is another option for identifying an optimal biomechanical profile: construction of

a hierarchical model via identifying relationships between various performance variables
and the performance criterion (Bartlett, 1997; Glazier et al., 2003). Models o f this type
eliminate any arbitrary selection of performance variables, and specify critical
relationships between these variables and the performance criterion based upon the laws
of physics. This thesis adopts this approach.
2.3.2 Running kinematics
Introduction
The kinematic review will only appraise biomechanical variables o f running
substantiated in the literature. Most studies are descriptive, with only a few finding
causative relationships between certain biomechanical variables and running
performance (Williams, 1985; McClay, 2000).
18
Kinematics of the Centre of mass
Vertical oscillation of the centre of mass decreased with increased running speed
(Luhtanen and Komi, 1980; Ito et al., 1983a). For example, Luhtanen and Komi reported
vertical oscillations of 10.9 cm, 8.6 cm, and 6.7 cm for speeds of 3.9 m-s'1, 6.4 m-s'1, and
9.3 m-s'1 respectively, for six runners. Ito et al. (1983b) obtained similar results for
speeds of 1.9 m-s"1 to 6.1 m-s"1, while also observing the vertical oscillation of the centre
of mass decreased during contact but increased slightly during flight as speed increased.
If vertical oscillation decreases and vertical ground reaction force increases (Cavanagh
and Lafortune, 1980; Roy, 1982; Hamill et al., 1983) at faster running speeds, logic
dictates there is less vertical work. The increased ground reaction force reflects increased
acceleration of the centre of mass, whereas decreased oscillation may reflect an
increased angle swept by the lower-limb (Ferris et al., 1999).
Cavanagh and LaFortune (1980) found a mean decrease in the horizontal velocity of the
centre of mass of 0.18 m-s'1 during the braking phase for seventeen participants running
at 4.47 m-s'1. The propulsive phase found a mean increase of 0.27 m-s'1. Wide variability
exists between rumiers, for changes in horizontal velocity of the centre of mass during
support (Fenn, 1930; Bates et al., 1979). Perhaps the variability found in the horizontal
velocity of the centre of mass during stance exists because impact patterns differed
between the runners measured. For example, if the foot lands further ahead of the centre
of mass at impact for one runner compared to others, a greater braking effect can occur
(Fenn, 1930; Deshon and Nelson, 1964; Cavanagh et al., 1977; Bates et al., 1979; Kunz
and Kaufman, 1981; Girardin and Roy, 1984; Hinrichs, 1990). Landing on the heel as
opposed to the forefoot may decrease horizontal velocity at impact or vice versa. Impact
patterns also affect stance-limb impact angles, at the hip, knee and ankle, via the kinetic
chain (Bartlett, 2000), which subsequently alter the centre of mass’s vertical
displacement (Lee and Farley, 1998). Lower-limb angles, centre of mass and foot
position at impact, therefore, have an important relationship to reductions in horizontal
velocity of the runner.
In an extensive evaluation of running, Fenn (1930) combined vertical and horizontal

movements of the centre of mass within the body in the sagittal plane. In general, the
centre of mass moves diagonally upwards and backwards (stance), then, reversing to go
forwards and downwards (swing) within the body during the gait cycle. Only the slowest
19
runner (6.91 m-s'1 compared to 7.22 m-s'1 and 8.5 m-s'1) had their centre of mass pass
through a negative position in relation to their centre of mass in the standing position.
Fenn (1930) linked decreases in horizontal velocity of the body at impact with position
of the centre of mass in relation to their anatomical standing position. His faster rumiers
produced positive scores showing a forward position in relation to their standing centre
of mass location. The data clearly indicated that the body is pushed forwards from the
hip at foot contact (hip velocity exceeded that of the neck), because the body’s centre of
mass is in a positive position, with respect to their anatomical standing position at foot
contact. Slocum and Bowerman (1961) state in reference to the cause of deceleration of
the centre of mass:
UA simple force diagram will reveal that the further ahead o f the body the fo o t strikes the
ground, the more acute the angle and the greater the deceleration from ground
resistance.”
Deshon and Nelson (1964) also found that runners that are more proficient kept then foot
as close as possible beneath their centre of mass. Mann and Herman (1985) using all
sprint finalists (100 m, 200 m and 400 m) from the 1984 Olympics, determined Gold
medallist’s centre of masses were 0.217 m behind the foot at impact compared to Silver
medallists at 0.284 m and eighth position with 0.327 m.
Potential and kinetic energy of the centre of mass in running are minimum at mid-stance.
The knee is maximally flexed at mid-stance— the centre of mass at its lowest point—
which coincides with the braking phase during the first half of stance when horizontal
velocity of the centre of mass is reduced (Enomoto et al., 1999). During the second half
of stance, kinetic energy of the centre of mass increases owing to the accelerating affect
of ground reaction force and gravity. However, kinetic and potential energy in running
are nearly in phase with each other (see figure 2.3.1). Exchange of kinetic and
gravitational potential energy conserves less than 5% of the mechanical work required to
lift and accelerate the centre of mass (Cavagna et al., 1964). A biomechanical
intervention that could reduce braking force would conversely preserve the centre of
mass’s kinetic energy. In fact, Enomoto et al. (1999) found an improvement in
mechanical energy conservation owing to lower deceleration of the centre of mass in the
first half of support.
20
■cn bm
O2
oq
v:
O n^
© 7s
© til
Right Stance Left Stance

Time (s)
Figure 2.3.1 R u n n in g at 3 .8 m - s '1. th e k in e t ic e n e r g y f lu c tu a tio n s o f th e c e n tr e o f m a s s a re a p p r o x im a te ly
in p h a s e w it h th e g r a v ita tio n a l p o te n tia l e n e r g y f lu c tu a tio n s . S ta n c e p h a s e s a re id e n t if ie d at th e b o tto m o f
th e g r a p h (F e r r is et al., 1 9 9 6 ; p . 2 6 0 ) .
Estimated potential strain energy for a 70 kg man's total energy turnover in each stance
phase was 100 J at 4.5 m-s'1 (Ker et al., 1987). They estimated 35 J are stored as strain
energy in the Achilles tendon and 17 J in the arch of the foot, with more stored in the
quadriceps and patella tendons. Strain energy can reduce half of the energy needed by
muscles, for instance, when acting as brakes at impact or producing work at terminal
stance (Ker et al., 1987). This essentially reduces the muscles shortening and
lengthening action owing to the tendon being stretched while the muscles holds tension
(Alexander, 1992; Novacheck, 1998).
Spring-mass model
Spring-like motion of a bouncing ball models oscillation of a runner’s centre of mass
(figure 2.3.2). The spring-mass model consists of a single linear spring (leg-spring) and a
point-mass that is equivalent to the centre of mass (Alexander, 1988; Blickhan, 1989;
McMahon and Cheng, 1990). Leg-spring stiffness represents the overall stiffness of the
integrated musculoskeletal system. During early stance the leg-spring compresses (leg
flexion), and then lengthens (leg extension) for the second half of stance. This model
(figure 2.3.2) describes and predicts the dynamics of running remarkably well (Farley
and Gonzalez, 1996; Fukunaga, et al., 1997).
21
Centre of
AL mass
Leg-spring
Figure 2.3.2 A spring-mass model and a stick figure representation of a single stance phase (Ferris and
Farley, 1997). The model consists of a linear spring representing the leg and a point mass equivalent to
body mass. This figure depicts the model at initial contact (left position), at mid-stance (middle position)
and terminal stance (right position). AL = change in length.
Dynamic interaction between the stance leg and the ground are important: leg stiffness
seeks to express this relationship. Components of leg stiffness include vertical excursion
of the body’s centre of mass, stance time and ground reaction force (McMahon and
Cheng, 1990; Farley and Gonzalez, 1996). Leg stiffness is defined as the ratio o f ground
reaction force to the compression of the leg-spring (AL) at the instance of mid-stance
when the leg is maximally compressed (figure 2.3.2). Increased ground reaction force
produces an increased leg-stiffness ratio. Conversely, an increased knee flexion reduces
leg stiffness. Running on five different surfaces o f varying stiffness ranging from 75 to
945 kN-m'1, a 12.5 fold change; knee angle increased by 2.5% as surface stiffness
decreased while time of ground contact did not significantly change (Kerdok et al.,
2002). Kerdok et al. (2002) credited changes in leg-spring to local joint stiffness
variations and overall limb posture adjustment. The more compliant running surface was
attributed to leg-spring change, because it acts as an elastic substrate capable of storing
and returning mechanical energy in this case, 1.8 W o f energy metabolism.
Leg stiffness remains virtually the same for running speeds between 2-6 m-s’1 (He et al.,
1991), although as stride length increases stiffness decreases by up to 51% (Derrick et
al., 2000). To maintain leg stiffness across the endurance speed range the angle swept by
the leg during stance increased enabling decreased stance times (figure 2.3.2; angle &).
Owing to the larger angle swept by the leg at faster speeds, the vertical displacement of
the centre of mass is less at faster speeds, which reduces stance time (Ferris et al., 1999).
Alterations in leg stiffness while running are possible. Leg stiffness occurs with a change
in stride frequency (Farley and Gonzalez, 1996), to reflect say, off-set changes in surface
22
stiffness (Ferris et a l, 1996). By adjusting leg stiffness, runners were able to have the
same peak ground reaction force, stance time and vertical displacement for the centre of
mass regardless of the surface stiffness (Ferns et al., 1996). Dixon et a l (2000) in
contrast, found individual variation in these variables. Individual variations are possible
revealing skill differences, but their model works on a premise that at impact the centre
of mass increases its distance behind the point of contact as running speed increases.
This is not evident however in faster runners (Fenn, 1930; Deshon and Nelson, 1964;
Cavanagh et al., 1977; Hinrichs, 1990).
Joint stiffness is defined as a change in joint moments divided by change in the joint
angle (Kuitunen et al., 2002). These two variables potentially affect vertical and
horizontal excursion of the centre of mass by their roles in limb displacement and force
production. Kuitunen et al. (2002) measured ten sprinters at different speeds, and
identified that ankle joint stiffness remained constant. Knee joint stiffness increased
across the speed range. However, only a minimal change in knee joint angle took place,
which suggested knee moment was a prominent factor, unless elastic mechanisms can be
utilised.
As endurance runners fatigue, stance time increases (Nicol et al., 1991). This suggests
that a time delay between the muscle stretch and the shortening action may have
widened. In combination, peak braking and vertical ground reaction forces decrease with
fatigue (Paavolainen et a l, 1999b). These data suggest that many repetitive stretch
loads— fatigue—decreased the capacity of the neuromuscular system to generate force
rapidly (Mero et a l, 1992). One possible explanation may be a reduction of muscle
stiffness in the braking phase (Nummela et a l, 1994). This scenario would lead to a
reduction in stored elastic energy during the braking phase, reducing the ability to store
and return elastic energy. Knee flexion and extension angular velocity increases with
rising running speed allowing the spring-mass model to function. Greater impact spring
stiffness arises from increased stride length of up to 20% (Dietz et al., 1979; Derrick et
a l, 2000). At sprint running speeds, high leg stiffness may produce a stiff rebound
(Chelly and Denis, 2001). However, research does not support an equal flexion and
extension relationship during stance for thigh or knee angles (Milliron and Cavanagh,
1990; De Wit and De Clercq, 2000; Li, 2000). Non-equal knee/hip flexion and extension
may explain variations in position of the centre of mass to the support limb (see vertical
23
oscillation of the body’s general centre of mass p. 19). For example, Lee and Farley
(1998) concluded that stance-limb compression and impact angle both play an important
role in determining the trajectory of the centre of mass. However, they did not deduce
from their model that the centre of mass is behind the point of contact at impact thus
negatively affecting forward progression. In addition, vertical oscillations of the centre
of mass may reveal changes in the reciprocal knee flexion and extension angular
velocities, which may affect forward progression.
Spinal Engine Theory

Based upon the spines capability for torsion and compression: Gracovetsky (1988)
formulated two hypotheses. Firstly, runners’ legs cause the pelvis to rotate. Secondly, the
spine can convert a lateral bending movement into an axial torque to drive pelvic
rotation. Essential to the theory is the need to axially rotate the pelvis. In running, the hip
extensors fire as the foot leaves the ground (see gait muscles and electromyography
activity p. 35). Gracovetsky (1988) stated muscle power transfers directly to the spine
and trunk through two distinct complementary pathways. The sacrotuberous ligament
extends the action of the biceps femoris, through to the rib cage. This ligament crosses
over the posterior superior iliac crest and continues as the lumbar intermuscular
aponeurosis (Bogduk and Twomey, 1987), which has a direct link with the lumbar and
longissimus lumbrum as well as the spinous processes via multifidus. Secondly, gluteus
maximus connects to the lumbodorsal fascia, itself linked with latissimus dorsi and the
upper extremities. Consequently, locomotion should be possible for a legless individual.
Gracovetsky (1988) recorded legless participants walking on their ischiums by keeping
normal spinal motion and electromyography patterns for the trunk musculature. These
data corresponded to normal gait, except for amplitude of the movements. Justification
for this phenomenon was the lordotic spine converts lateral bend into an axial torque
driving the pelvis, thus demonstrating the need for compression and torsion in the spine
during locomotion. Accordingly, spinal engine theory postulates that the legs follow
pelvic motion in a desire to minimise energy expenditure. In addition, spinal engine
theory promotes the central role of gravity as the motive force for legless locomotion.
Converted heel-toe runners

Williams et al. (2000) took six male and three female heel-toe runners and converted
them to forefoot strikers in one laboratory session. Their research established a list of
24
variables (table 2.3.1) that differentiated heel-toe and forefoot runners. Proceeding
sections concentrate on the kinematic analysis of rear and forefoot strikers; the kinetics
are discussed in section 2.3.3 p. 31.
Table 2.3.1 V a r ia b le s f o u n d to b e d iffe r e n t b e t w e e n rea r f o o t a n d f o r e f o o t str ik e r s ( W illia m s et al.. 2 0 0 0 )

Kinematics Kinetics
Plantar flexion at impact Vertical ground reaction force
Inversion at impact Anterior ground reaction force
Dorsiflexion excursion Peak plantar flexion moment
Eversion excursion Ankle power absorption
Dorsiflexion velocity Ankle negative work
Eversion velocity Ankle inversion moment
Knee flexion at impact Peak knee flexion moment
Knee flexion excursion Knee power absorption
Knee internal rotation velocity Knee negative work
Vertical loading rate________
Stride length and stride frequency

Over striding requires considerable power during propulsion, resulting in excessive
vertical oscillation of the centre of mass. In turn, this produces an impact position that
creates increased braking forces and requires reciprocal joint ranges of motion, which
invoke increased internal friction and stiffness (Hinrichs, 1990). Conversely, short
strides increase internal work through increased frequency of reciprocal movements.
Nilsson and Thorstensson (1987), within a range of running speeds of 1.5-8 m s'1,
identified stride rates of 33 to 214 strides-min'1. These rates were the absolute lowest and
highest the rumiers could perform. Runners can alter cadence and stride length, thus
creating many different combinations (figure 2.3.3). Humans transition from walking to
running at a critical speed of 2.2 m-s* , rather than at a critical stride rate or length
(Laurent and Pailhous, 1986; Hreljac, 1995), plausibly indicating that the speed of
locomotion is not the result of leg-work but the trunk and possibly the spinal motion via
gravity’s work (see spinal engine p. 24). In addition, the flexibility to alter stride rate and
stride length implies that the central nervous system must have mechanisms for actively
controlling these variables. In support, the Froude number illustrates gravity’s role in
running as a combination of leg length and running speed.
Froude number - (speed of locomotion) (2.3.1)

gravitational acceleration x leg length
25
SR (Hz) SL (m)
4.5 2.0
4.0
stride length (SL) 1.5
3.5
3.0 stride rate (SR) 1.0
2.5
4 6 8 10
Running velocity (m-s*1)
Figure 2.3.3 Variations in stride length and stride rate with increase running velocity (Luhtanen and Komi,
1978).
Dillman (1975) found better or more ‘skilled’— experienced college cross-country—

runners tend to have longer strides at any given velocity than ‘poorer or less skilled’
runners. In contrast, Cavanagh et al. (1977) found elite runners took shorter absolute and
relative strides than good distance runners did. Trained sprinters on a treadmill took
considerably longer strides than non-sprinters despite similar leg-lengths (Armstrong and
Cooksey, 1983). A helpful requirement would be a clear definition of skilled runners,
which perhaps explains the conflicting reports. Dillman’s group had only four college
runners. Cavanagh’s study compared good and elite runners at 5 m-s'1, equivalent to five
and half minute mile pace (3 min 26 s per km), which might have been faster than good
runners were used to, hence leading to altered stride lengths on the treadmill. Further,
stride length and stride frequency may differ because sprinters over endurance athletes
often spend more time completing running drills and strength training (Dillman, 1975).
These ‘skill’ factors may have contributed to the differences found. No mention was
made of how stride length was produced for example, were hip flexors more or less
active in athletes with longer stride lengths owing to their associated knee lift increases
(Armstrong and Cooksey, 1983).
26
Longer strides may be associated with higher ground contact forces. Weyand et al.
(2000) concluded maximal sprint speed occurs through an enhanced ability to generate
and transmit muscular force to the ground, in contrast to a rapid repositioning of lower-
limbs through stride length and stride frequency. Data collected on a treadmill rather
than by over-ground running, possibly altered the findings (van Ingen Schenau, 1980;
Nigg et al., 1995). For example, stance time for over-ground running was 15% higher for
world-class sprinters Weyand et al. analysed from live footage than for their own
treadmill data. The effective force applied to the ground and the time of foot contact
determined aerial times (Weyand et al., 2000). Ariel times at maximal sprint speed
between runners were similar, however, shorter stance times increased stride frequency
in the fastest athletes. By applying higher ground contact forces, faster runners were able
to increase both stride length and stride frequency. Although vertical force increased
across the speed range (2-10 m-s'1), impulse decreased after 6 m-s'1. Ground reaction
force reflects the acceleration of the centre of mass. This study showed a mean increase
of 1 BW from 2 m-s'1 to 10 m-s'1. It seems unlikely that this increase in force alone
reflects the considerable increase in speed. The top speed reached was 11.1 m-s'1 and the
slowest was 6.2 m-s'1 from similar increases in force. There was more vertical force in
the declined position (-6°) than the level, indicating body orientation to the ground
(gravity) played a role in the vertical force found in running (Weyand et al., 2000).
Speed increases at low running speeds occur primarily through amplified stride length,
for at high speeds, stride frequency is the primary factor (Weyand et al., 2000; Hay,
2002). Endurance runners’ stride frequency reached a theoretical maximum at 5 Hz
(Mann and Herman, 1985; Nilsson and Thorstensson, 1985). As noted earlier, Cavanagh
and Kram (1985) found stride frequencies were individual, because runners appeared to
maintain their chosen frequency for speeds of 3 m-s'1 to 4.2 m-s'1. With increased speeds
(above 7 m-s"1), stride length did not increase significantly, whereas stride frequency
continued to increase. Runners at high speeds, therefore, tend to enhance speed through
increased stride frequency rather than stride length (Luhtanen and Komi, 1980).
Increased frequency may cause shorter strides, casting reservation on the idea that this
increase would amplify internal work (Hinrichs, 1990).
The running stride, further subdivides into support (stance) versus non-support (flight)
(Bates et al., 1979). Saito et al. (1974) found as running speed increased, the time for a
27
complete running cycle decreased. Further, absolute and relative time spent in support
decreased as running speed increased (Saito et a l, 1974; Bates et a l, 1978; Chapman,
1982). Stance time changed from 68% at 3.35 m-s'1 to 54% at 6.4 m-s'1, although there is
substantial individual variation (Nelson et al., 1972). Mami et al. (1986) found enlarged
joint range of motion as speed of running increased, with subsequent reduction in
support time (stance). They concluded that rises in angular velocity at the joints would
create increased energy expenditure and force, thereby raising the possibility of injury.
For example, in sprinting, hip flexion during swing was 80° compared to 40° for jogging
(Mann et al., 1986). Increased hip motion during swing raises the foot higher from the
ground possibly elevating impact forces. Increased running speed, however, has not been
associated with increased running injuries (see p. 42; Yzerman and Van Galen, 1987).
Lower and upper extremity limb kinematics

Hip joint kinematics can differ depending upon the method used. Two commonly
practised methods are orientation of the thigh with respect to the vertical axis or,
obtaining hip joint angles using some markers on the trunk to define the upper body
segment. However, there is generally a forward lean of the trunk in running (Williams
and Cavanagh, 1987). Therefore, inclusion of trunk orientation will tend to increase the
measured flexion angles and decrease extension angles compared to the measurement of
the thigh orientation angles alone (Milliron and Cavanagh, 1990).
The hip joint has been observed to extend during support from about 35°, and flex during
the swing phase from 40° for jogging, 60° for running and 80° for sprinting (Mann et al.,
1986). Maximum hip extension occurs at or immediately after terminal stance, and
maximum hip flexion occurs at about two-thirds of the way through the swing phase as
contralateral terminal stance begins (Sinning and Forsythe 1970; Dillman 1975;
Cavanagh et al., 1977; Mami et al., 1986). Maim et al. (1986) suggest hip flexion is the
main instigator for increasing running speed with the rest of the lower extremity
following the lead of the hip's increased angular velocity. Maim et al. (1986) also found
that knee extension began at the end of maximum hip flexion during the last third of the
movement, and occurred through the concentric contraction of the quadriceps. The
degree of maximal hip extension while occurring at terminal stance is less at moderate
than faster running speeds (Mann and Hagy, 1980). In contrast, Mann and Herman
(1985) found elite sprinters had less hip extension at terminal stance, because extension
28
occurred via momentum of the flexing hip, owing to silent quadriceps muscles during
hip extension. It might be pertinent to decrease elevation of the knee via decreased hip
flexion during swing because it would lower work done against gravity. Possible
differences in hip flexion and extension may also reflect the relationship with knee and
ankle kinematics owing to the kinetic chain (Bartlett, 2000).
At impact, the knee angle has been generally reported to be in the range of 10° to 30°
(Bates et al., 1978; Elliot and Blanksby, 1979; Clarke et al., 1983). A variety of angles
for knee flexion have been found for support, ranging from 38 to 50° for speeds of 3.4
m-s'1 to 7.5 m-s'1 (Cavanagh et al., 1977; Bates et al., 1978; Bates et al., 1979).
Although, only small changes of 1.5° to 4° in knee flexion have been identified during
support across speeds ranges of 3.33 m-s'1 to 6.67 m-s’1, despite absolute differences in
knee flexion among individuals (Sinning and Forsyth, 1970; Miller, 1978; Nilsson and
Thorstensson, 1985). After impact, flexion occurs in the cushioning phase and continues
through mid-stance, followed by a knee extension phase that lasts up to or slightly after
terminal stance.
At terminal stance, the knee does not fully extend although there appears to be a trend
towards greater extension with increased speed (Cavanagh et al., 1977; Bates et al.,
1978; Elliot and Blanksby, 1979). This finding is not universal (Sinning and Forsythe,
1970). Cavanagh et al. (1977) indicated that the knee did not fully extend throughout the
gait cycle and was almost the same angle at impact as terminal stance. At 2.5 m-s'1 to 5
m-s"1, knee extension was 23° (Elliot and Blanksby, 1979) and at 8 m-s'1, 18° (Mann et
al., 1986). Extension increases were found for sprinting at 8 m-s'1 but not for distance
running speeds of 2.5 m-s'1 to 5 m-s’1, contradicting a commonly held assumption, that
rumiers push off the ground via knee extension (Milliron and Cavanagh, 1990).
Bates et al. (1979) found no relationship between knee angle at impact and the amount of
decrease in horizontal velocity of the centre of mass during stance. Barefoot runners had
greater knee flexion angular velocity than shod, which began 0.02 s before impact (De
Wit et al., 2000). Running in shoes gave similar results, although comparison is not
straightforward owing to other altered kinematic variables in barefoot runners as
opposed to shod (Clarke et al., 1983). De Wit et al. (2000) concluded barefoot runners
make kinematic adaptations just before impact, including increased knee flexion velocity
29
0.02 s before, followed by a passive interaction between the ground and the contacting
leg. Further, the initial ground contact phase indicated the overall stiffness of the support
leg to be higher in barefoot running (see spring-mass model section p. 20).
At impact, the ankle is at approximately 90°, while at terminal stance, usually 20° of
plantar flexion occurs (Milliron and Cavanagh, 1990). Mann et a l (1986) observed for
jogging, running and sprinting dorsiflexion of the ankle joint occurred after foot contact.
Dorsiflexion continued through mid-support, which occurred at about 50% of the support
phase, followed by ankle plantar flexion until terminal stance and into swing. Magnitude
of dorsiflexion was 10° for jogging and running with 15° for sprinting (Nilsson et al.,
1985). The magnitude of plantar flexion, which began at mid-stance was 45° for jogging
and running with 35° for sprinting (Mann et a l, 1986). Plantar flexion is lower in
sprinting than running and jogging, again contradicting the assumption o f increased
push-off at faster speeds. At all speeds, plantar flexion reached its maximum at terminal
stance. For jogging and running progressive dorsiflexion occurred during the swing
phase. For sprinting, plantar flexion reached a maximum just before impact (Mann et a l ,
1986). Cavanagh et a l (1977) identified that elite runners had less plantar flexion at
terminal stance than good runners (67° versus 59°). If the runner is a heel-toe or forefoot
striker this will effect ankle dorsi and plantar flexion angles making comparisons
difficult (Williams et a l, 2000). For example, forefoot compared to heel-toe runners
plantar flexors, fatigue faster during exhaustive running (Pratt, 1989). Pratt suggested
this finding supported why eighty percent of distance runners are heel strikers.
Several studies found that the arms increased the vertical oscillation o f the body’s centre
of mass in running (Cavanagh, et al, 1977; Cavanagh and Williams, 1987; Hinrichs et
al., 1987). Hinrichs et a l (1987) concluded the arms tended to reduce fluctuations in the
forward velocity and in side-to-side movements o f the runner. Hinrichs (1987) measured
angular momentum during flight, finding total angular body momentum about the
vertical axis is constant in the absence of external torques. Therefore, the lower body
receives 100% of its angular impulse from the upper body. If the upper body where not
present (arms and upper trunk), the legs could not change direction while airborne, and a
recognisable swing would not be possible. The radius between the hip and ankle affects
angular momentum during flight, owing to an increased radius reducing the angular
velocity of the swing leg causing a potential increased energy cost.
30
Trunk lean has been observed to be in the range of 4° to 7° at impact at 7 m-s'1 with most
of the literature supporting a forward trunk lean (Fenn, 1930; Girardin and Roy, 1984;
Williams and Cananagh, 1987). Trunk lean in runners who ran against a head wind in a
wind tunnel, ‘leaned’ into the wind, indicating it was mechanically efficient to do so
(Davies, 1980). After impact, forward trunk lean has been reported to increase until mid
support, reaching 12 to 13° for speeds just over 5 m-s'1 (Elliot and Roberts, 1980; Elliot
and Ackland, 1981). An increased trunk lean would permit a more forward position of
the centre of mass during stance, subsequently allowing it to pass over the support foot
earlier, potentially generating faster running.
2.3.3 Ground reaction force and optimal running biomechanics
Introduction
Ground reaction force reflects acceleration of the total body centre of mass. The position
and direction of the runner’s centre of mass determine the direction and magnitude of the
ground reaction force. Hence, it is the sum of the acceleration of individual segments
contributing in proportion to their masses. The support limb transmits the force to the
ground, and theoretically need not create any contribution to the push (Miller, 1990).
Collection of three-five trials is normally used to establish reliable data (Williams, 1980;
Cavanagh and LaFortune, 1980; and Clarke et al., 1983). Diss (2001) found a high
reliability (/* = < 0.94) for four ground reaction force variables from a single trial when
compared to ten trials. Of the sixteen ground reaction force variables sampled, fifteen
needed less than four trials at r = < 0.90. Only the commonly used ground reaction force
variables that are relevant to thesis are reviewed.
The force plate and stride modification

Force plates positioned in a laboratory floor collect ground reaction forces. Data
collection assumes that a typical runner’s stride pattern is ‘normal’ (Hay, 1988; Miller,
1990). Related research has indicated that stride adjustments do occur when approaching
a target (Hay, 1988). A study that intentionally manipulated stride lengths, noted
differences in ground reaction forces, including support time, first maximum vertical
force, mean vertical force, and anterior-posterior maximum braking and propelling
forces (Martin and Marsh, 1992). Martin and Marsh (1992) recommend against
constraining stride length and stride frequency for evaluating representative gait
31
kinematics and kinetics, requiring gait laboratories to design adequate approach runways.
The laboratory runway was 15.66 m for this thesis.
Differences between heel-toe and forefoot ground reaction force patterns

Oakely and Pratt (1988) identified essential differences for vertical, posterior and
anterior ground reaction forces between heel-toe and forefoot runners. Forefoot runners
typically have lower vertical ground reaction force peaks than heel-toe runners (Harrison
et al., 1988). The initial vertical peak in heel-toe runners is absent or attenuated in
forefoot runners (see figure 2.3.4) (Cavangh et al., 1977; Cavanagh and Lafortune, 1980;
Harrison et al., 1988). Posterior and anterior ground reaction force for forefoot runners
produced a biphasic shape during the braking phase. Heel-toe runners typically showed a
single retarding peak (Cavanagh and Lafortune, 1980; Clarke et al., 1983). Many studies
have repeated these findings over the last twenty-five years.
Payne (1983) using a sprinter running at 9.5 m-s’1 who landed on their heel found
increased ground reaction forces compared to forefoot sprinters. Normally sprinters land
on the ball of the foot (Mero and Komi, 1986). Mero et al. (1992) recommended that the
braking force and contact time be very short to avoid loss of velocity during impact with
the ground, and associated this with forefoot landing.
Williams et al. (2000) using six male and three female trained heel-toe runners, asked
them to land on their forefoot instead of their heels while running across a force platform
(table 2.3.1; p. 24). All runners successfully changed from heel-toe to forefoot contact,
eliciting loading rates lower than their heel-toe scores. Comparison of the converted
forefoot runners with experienced forefoot runners showed similar movement patterns,
with one significant exception, decreased peak vertical ground reaction force. The
authors suggested it was owing to decreased plantar flexion force at push-off.
Vertical ground reaction force

Figure 2.3.4 illustrates a spike (known as an impact peak) on the vertical ground reaction
force for the heel-toe runner, which is absent on the forefoot runner. The impact peak for
running velocities of 3 m-s’1 to 6 m-s’1 usually falls within 2-3 body weights (Denoth et
al., 1983; Nigg, 1983). The onset-time of the vertical impact peak has been associated
with overuse injuries (Nigg, 1986).
32
F o re fo o t V e r t ic a l g r o u n d
r e a c tio n fo r c e
H e e l- t o e
z
P o ste r io r a n d
a n te r io r g r o u n d
o8 r e a c tio n fo r c e
H e e l- t o e F o r e fo o t
Figure 2.3.4 H e e l- t o e (r e a r -fo o t) a n d f o r e f o o t g r o u n d r e a c tio n fo r c e s fo r tw o 7 4 0 N m a le s at 3 .3 5 m -s' .
Twenty heel-toe runners in one laboratory session were asked to run normally—heel-
toe— then on their forefoot (Arendse et a l 2004). A Pose* running intervention
followed for an accumulative period of 7.5 h (1.5 h for 5-days). The heel-toe and
forefoot styles were accompanied by greater magnitudes (F = 28.2, p = <0.001) and
loading rates (F = 35 A, p = c.001) of the vertical impact force at 25 ms of stance and at
peak magnitude compared with Pose® method running. Increased vertical impact force
has been associated with injury mechanisms (see p. 44; Radin and Paul, 1971; Verbitsky
e ta l, 1998).
The posterior and anterior (horizontal) ground reaction force

Posterior and anterior ground reaction force is biphasic in running (Miller, 1990). The
initial horizontal ground reaction force is termed braking (posterior), and the latter phase
propulsion (anterior) (Cavanagh and Lafortune, 1980). Mason (1980) reported mean
braking and propulsive maxima of 38% and 50% body weight at 4.7 m -s1, 70% and 75%
body weight at 7.6 m-s'1 respectively. Forefoot strikers running at 4.5 m-s'1 had mean
braking maxima of 45% body weight and 50% for propulsive maxima at 4.9 m-s'1
(Cavanagh and Lafortune, 1980). Kunz (1978) recorded high-jumpers approach runs
across speed ranges of 2.5 m-s'1 up to 6.5 m-s'1. The posterior component of the ground
reaction force did not increase across the whole speed range yet the speed of running
increased by 4 m-s'1. This research used high jumpers (forefoot strikers) and not distance
runners, nonetheless the speed of running increased with no reciprocal increase in
posterior ground reaction force. Hoshikawa et al. (1973) found as stance time decreased
33
with running speed, the posterior component decreased in time in relation to the anterior
component, despite an increase in running speed. It appears in certain runners braking
force reduces as speed increases. Horizontal braking {F = 18.0, p = <.001) and
propulsive forces ( F = 3 . 6 , p = .035) were less in the Pose® than the heel-toe and forefoot
running styles (Arendse et al., 2004).
Stance time
Stance time is the period when the foot is in contact with the ground. Stance time and
velocity are negatively related (Munro et a l , 1987). Time spent on the ground can be
logically linked to the position and timing o f lower-limb in connection with the centre of
mass’s flight path. The more extended the knee at ground contact, and the further the
foot is in front of the centre of mass, the longer the foot remains on the ground (Kunz
and Kaufman, 1981; Mann and Herman, 1985). Barefoot runners compared with shod,
elicited shorter stance time when both landed on the heel at 3.5 m-s’1, 4.5 m-s'1 and 5.5
m-s'1 (De Wit et al., 2000). Stance time may denote global biomechanical changes,
because stride length, stride frequency, vertical loading rate and knee angular velocity
were significantly different between the two conditions (De Wit et al, 2000).
Inverse dynamics method

During the braking and push-off phases of running, the ankle, knee and hip joints have
negative and positive angular velocities (Ito et al., 1983a). Calculating external
mechanical work during running is the combination of kinematics with ground reaction
force (Fenn, 1930; Cavagna, 1975; Zatsiorsky and Fortney, 1993; Yeadon and Challis,
1994). The inverse dynamic method therefore, aims to account for all the forces
necessary to accelerate the body mass, such as, positive and negative work absorbed and
produced at each joint. Instantaneous values for joint kinetics and muscle function during
stance are included (van Ingen Schenau et al., 1983; Wells, 1988). However potentially
effective inverse dynamics is (and will be), certain difficulties exist with this method.
The inverse dynamics model does not account for the computation of multi-joint
movements or measurement of biarticular muscles (Yeadon and Challis, 1994). Precise
estimation of the mechanical work completed by extensor and flexor muscles o f the
lower-limb is also problematic (Yeadon and Challis, 1994). Employment o f joint angular
velocity to characterize eccentric and concentric actions of leg extensor muscles is
questionable (Zatsiorsky and Fortney, 1993). Co-contractions of lower-limb muscles are
34
difficult to account for and identify between runners (Belli et al., 2002). Assumptions of
the method usually are as follows: body segments are rigid; the line of action for ground
reaction force is at the head of the fifth metatarsal bone and elastic energy is also not
accounted for. The number of assumptions and limitations preclude this method of
analysis for this thesis.
2.3.4 Electromyographical research and optimal running biomechanics
Introduction
A simple joint system (rigid link, synovial joint, muscle, neuron and sensory receptor)
has five elements that interact to produce movement (Marieb, 1992). In this system, the
muscle is able to exert a force that can vary in magnitude, a variation that is profoundly
influenced by the nervous system. The most effective measurement of the neuromuscular
response to movement is electromyography (Basmajian and Deluca, 1985).
Physiologically muscles and motor units can be compared to one another based upon
discharge characteristics owing to biomechanical responses of the motor units to
different inputs: speed of contraction, magnitude of the force exerted and resistance to
fatigue (Burke et al., 1973). It is difficult to interpret the relationship between
estimations of muscle moments and measurements for electromyography activity,
because torque about a joint is affected by other anatomical and environmental factors.
At present, electromyography cannot therefore determine net muscle moments
(Pierrynowski, 1995). This section covers, muscle force prediction, electromyography of
gait muscles and the stretch-shortening cycle in relation to optimal running
biomechanics.
Electromyography and muscle force prediction

Using electromyographical signals to directly estimate individual muscles activity during
motion is based upon the assumption that quantitative relationships between the
electromyographical signal and muscle forces are known, or can be determined. The
assumption of a linear relationship between the rectified integrated electromyographical
signal and force is questionable (Hof and van den Berg, 1981). Muscles produce an
electrical signal when activated. Magnitude and intensity of the electromyographical
signal related to the force produced by a muscle under given conditions has been
35
reported at least qualitatively (Liu et al., 1999). Difficulties of relating
electromyography to corresponding force signals were associated with highly non-linear
and a time-variable relationship between these signals. Measuring electromyography
consistently and repeatedly compounds this relationship. Currently, the
electromyographical-force relationship in human skeletal muscle is still at a
developmental stage and actual force predictions plus validations across muscles are not
yet available. Liu et al. (1999) carried out promising work in this field with an artificial
neural network approach. They predicted cat gastrocnemius muscle force from
electromyographical data. Force predictions for walking cat’s gastrocnemius muscle
correlation coefficients were r = 0.96 at 0.4 m-s'1 and r = 0.93 for 1.2 m-s'1. This is
promising work but is not a feasible option for this thesis.
Another perspective involves force production, electromyography and fatigue. Studies

on fatigue using repetitive stretch-shortening cycles for performance, demonstrated an
acute decrease in force generating capacity in both prolonged (Nicol et al., 1991), and
short-term intensive exercises (Moritani et al., 1990; Nummela et a l, 1994) plus a long
bout of running (Hausswirth et al., 2000). Significant reductions in the integrated
electromyographical signals were identified from the fatigued states.
Gait muscles and electromyography activity

Altered muscle activity patterns should correlate to changes in running style. Patla et al.
(1989) have shown that transient increases in stride length produced increases in activity
of some muscles, and decreases in the activity of others. Unfortunately, there is no
research to the author’s knowledge on muscle activity differences between forefoot and
heel-toe runners. Unless stated, the electromyographical studies may either represent
heel-toe or forefoot runners, who are usually endurance athletes, or sprinters,
respectively.
36
0 10 20 30 40 50 60 70 80 90 100
Percentage of gait cycle

Figure 2.3.5 E le c t r o m y o g r a p h y a c t iv it y e x p r e s s e d a s a p e r c e n t o f g a it c y c le at 3 .3 5 m -s"1 (a d a p te d fr o m
M a n n et al., 1 9 8 6 ) . N o t e t h e la c k o f m u s c le a c tiv ity a fte r m id -s ta n c e . 1 = m . Q u a d r ic e p s ; 2 = m .
H a m s tr in g s ; 3 = m . T ib ia lis a n te r io r ; 4 = m . G a s tr o c n e m iu s
Semimembranosis is active just before maximum hip flexion and shortly after the onset
of knee extension. This activity continues throughout most of the support phase, where it
may act as a hip extensor during support (Mann et al., 1986). Kyrolainen et al. (2000a)
found as running speed increased hamstring activity for hip extension amplified in
magnitude and time duration during stance. Smidt (1973) using 26 runners reported
moment arm lengths for their hamstrings increased as the knee flexed, reaching a peak at
45°, then, decreasing again as flexion reached 90°. Relative positions of the knee and hip
affect the hamstrings role, which may explain this biarticular muscles varied, but
important activity.
The quadriceps femoris group includes: rectus femoris, vastus medialis, vastus lateralis
and vastus intermedius. Rectus femoris is active during early swing, associated with hip
flexion (MacIntyre and Robertson, 1987). Both rectus femoris and vastus lateralis
recorded peaks during mid-stance (Elliot and Blanksby, 1979). Vastus lateralis’s main
activity was eccentric muscle action during early stance while the knee was undergoing
flexion. An interesting observation is that quadriceps muscles were quiet during the
extension phase of stance, suggesting knee extension may occur without their assistance
(see figure 2.3.5 and Figure 2.3.6) (Branded, 1973; Schwab et al., 1983; Nilsson et al.,
1985; MacIntyre and Robertson, 1987). During submaximal constant-speed running, the
electromyographic activity of the quadriceps muscles increased owing to marathon type
fatigue (Komi et al., 1986). Increased quadriceps muscles activity could raise internal
work, which may cause a detriment in running technique.
37
120
Sprint
Terminal
80 Stance(TS)
40
/—
o
^120
s—✓
con Run
<D
D£ 80 TS
<D
Q
40
80
Jog
60
TS
40
20 40 60 80 100
Percentage of gait cycle
Figure 2.3.6 Range of motion of knee during sprinting, running and jogging (Mann et al., 1986). Notice
the absence of knee extension during stance in the sprint condition.
The tibialis anterior represents muscle function for the anterior compartment o f the lower
leg. It is active just after terminal stance, and remains active throughout most o f the
swing through mid-support, when maximum dorsiflexion occurs (Mann et al., 1986).
Elliot and Blanksby (1979) recorded tibialis anterior acting concentrically to bring the
shank forward, and eccentrically to control the lowering o f the foot during the first part
of ground contact (for heel-toe runners). This muscle identifies foot contact for heel-toe
runners.
Gastrocnemius represents the posterior compartment’s muscular activity, during foot

descent for the swing phase, foot contact and mid-support. Gastrocnemius is mainly
active at the middle of stance with a heel-toe runner (Winter, 1987). There is some
controversy over the gastrocnemius’s role during support for different running speeds.
38
Faster speeds showed more activity (Elliot and Blanksby, 1979; Mami et al, 1986) but
with slower speeds, the gastrocnemius muscle persisted for only 30-50% of plantar
flexion (see figure 2.3.5) (Mann and Hagy, 1980). Komi et al. (1987) found a higher
pre-activation before impact for the gastrocnemius muscle with barefoot compared to
shod runners, possibly identifying different body-foot landing positions. The activity of
the vastus lateralis, rectus femoris, anterior tibialis, gastrocnemius and the hamstring
muscle group are, therefore, important in gaining understanding of lower-limb motion in
running.
The stretch-shortening cycle and electromyography

Running involves lower-limb eccentric and concentric muscle actions. For example, at
the ankle, the gastrocnemius muscle group is active during initial lengthening through
early stance before shortening during plantar flexion. This movement is called ‘length-
shortening’, and is usually referred to as the stretch-shortening cycle (Kreighbaum and
Barthels, 1996). The cycle allows for preloading of the muscle, which enhances force
output (Pugh 1971; Cavagna and Kaneko, 1977; Komi and Bosco, 1978) for a given
neural innovation. Concentric muscle action occurs for 5-10% of the gait cycle and is
both preceded and followed by eccentric action. It appears that eccentric loading may be
owing to alternative central nervous system recruitment of motor units (Enoka, 1996;
Christou and Carlton, 1999; Jung, 2003). For instance, time to peak force was less for
eccentric contractions than concentric contractions. Auro and Komi (1986) using
integrated electromyography for vastus lateralis and gastrocnemius muscles found a
positive correlation in men and women (r = 0.45 at p <0.001) in the eccentric-concentric
ratio. The ratio rose linearly with increasing pre-stretch loading. They hypothesised
maximum elastic capability of the muscle was 40% to 50% of its normal contractile
capacity. Improved pre-stretch loading potentially enhances force output in runners. It
would seem pertinent, therefore, to train this component. Kubo et al. (2000) found
endurance runner’s vastus lateralis muscle were less compliant than their untrained
counterparts. Lack of compliance was associated with a lower potential for elastic energy
storage measured by less lengthening of the muscle-tendon complex. Fatigue can also
decrease the stretch-shortening cycle. Nicol et al. (1991) studying a 10-km run observed
increases in contact time as the event progressed owing to fatigue. Fatigued runners
increased their quadriceps electromyographical output (Komi et al., 1986), suggesting
possible reductions from the stretch-shortening cycle, owing to changes in stride length
39
and frequency (see p. 10 Kyrolainen et al., 2000a). Long-term running training may
therefore decrease the stretch-shortening cycle possibly increasing muscle work and
injury.
Electromyographical activity of the lower extremity muscles increased with faster

running (Elliott and Blanksby, 1979). In contrast, Ito et al. (1983b) using four runners,
found muscle activity remained constant for the support phase during speed increases of
3.7 m-s"1 to 9.3 m-s'1, but increased for the non-support phase. Mann et al. (1986) tested
fifteen runners at 3.35 m-s"1, 4.4 m-s'1 and 9-10 m-s'1 and found very little differences in
electromyographical activity. Ito et al. (1983b) suggested that during support, lower
recorded muscle activity was owing to increased elastic energy contributions. Elastic
energy is less when jogging owing to the shortening phase of muscle gaining little from
reduced muscle lengthening (Jacobs et al., 1993). Both studies only tested elite forefoot
runners. These findings warrant further investigation into whether heel-toe runners are
less able to utilise their elastic mechanisms owing to different foot contact placement,
and lower-limb-centre of mass positioning.
Electromyographical studies have shown that the knee extensor (vastus lateralis) and
knee flexor (gastrocnemius) muscles are simultaneously active. There is a higher force in
the extensors if the flexors are in use (Ferris et al., 1996). Further, length-shortening of
the gastrocnemius muscle allows strain energy to be utilised. The vastus lateralis muscle
is not active as leg extension begins (MacIntyre and Robertson, 1987), possibly owing to
the potential use of elastic energy (figure 2.3.5). This may explain the decreased
eccentric loading at the knee recorded for Pose® compared to heel-toe and forefoot
runners (figure 2.3.7).
40
o 0.50-
£
o
<L>
U
O
0.25-
ffl
0.0 Heel-toe (mid) Forefoot Pose

Running technique
Figure 2.3.7 A c o m p a r is o n o f th e k n e e e c c e n t r ic w o r k b e t w e e n h e e l- t o e , ( m id ) f o r e f o o t a n d P o s e K r u n n in g
( N = 2 0 ) ( A r e n d s e , et al., 2 0 0 4 ) . ♦ d e n o t e s s ig n if ic a n t d if f e r e n c e fr o m h e e l- t o e a n d f o r e f o o t r u n n in g .
Another method, known as muscle-tendon length change, estimates the instantaneous

muscle-tendon length from approximate origin and insertion sites for a given muscle-
tendon unit, and from joint kinematic data (Morrison, 1970; Grieve et al., 1978).
Unfortunately, this approach does not account for muscle-tendon length changes owing
to muscle fibre displacement, muscle fibre pennation angle change, or tendon strain.
Each of these factors can affect the total muscle-tendon length (Narici, et ah, 1996;
Fukanaga et al., 1997).
However, a new approach using sonomicricrometry measures muscle fibre displacement

and velocities in vivo (Caputi et al., 1992), by suturing piezoelectric crystals into a
muscle bundle. Recorded was the time required for ultrasound pulses to travel from one
crystal to another. Transit time is used to calculate instantaneous fibre length.
Interestingly, the medial gastrocnemius muscle fibres in walking cats shorten at the
beginning of stance, even though the muscle tendon unit lengthens during this phase
(Griffiths, 1991). This means tendons stretch further than has been currently predicted,
with a subsequent elastic energy increase. Applying this theory to muscle work in
running, a theoretically lower muscle force may be found in those runners able to utilise
the increased elastic components.
41
2 .4 RUNNING INJURIES
Introduction
Injury incidence and biomechanical research associated with injury are appraised.
Reference to injury and running technique are included.
Injury incidence
Running is one of the most popular leisure activities in the United Kingdom. Next to its
beneficial health effects, there are negative side effects in terms of sports injuries. A
runner has been defined as a person running a minimum of three days a week covering a
distance of at least 3 km during each session (Taunton et al., 1987). An injury—
specifically an overuse injury— is an overload or repetitive impact to the musculoskeletal
system (Renstrom, 1994). Since overuse symptoms usually occur over time, it may be
difficult to determine the onset and specific cause. Rate of exposure and intensity are
also difficult to define. For instance, is a two-hour run on concrete similar to a forest trail
run? Are hill repetitions the same intensity as track repeats? These areas require more
detailed scientific-work (Rolf, 1995). Reported injury incident rates differ between
studies. Person-incidence rates are the number of injured runners per 100 runners, injury -
incidence rates record the number of injuries per 100 runners and injury incidence per
exposure, are the number of injuries per 1000 hours of running (van Mechelen, 1992).
It appears that for the average recreational runner, who is steadily training for long
distance running, there is a large range in yearly incidence for running injuries. For
example, 37% and 56% injure using sample sizes > 500 runners (Yzerman and Van
Galen, 1987), 25% to 75% (van Mechelen, 1995), and 33-85% (Macera, 1992) for 583
runners with 24% for 482 runners (Blair et al., 1987). These rates are dependent on the
specificity of the group of runners concerned, such as, competitive athletes rather than
recreational joggers, or children as opposed to adults. If incidence is calculated in
relation to time, it varies from 2.5 to 12.1 injuries per 1000 hours of running (Lysholm
and Wiklander, 1987; van Mechelen, 1992). In comparison, skiers are six times more
likely to injure per 1000 hours of activity (Marti et al., 1988). Most running injuries are
lower extremity injuries, with the knee joint being the main site of injury (van Mechelen,
1992; Stanish and Wood, 1994). About 50% to 75% of all running injuries appear to be
overuse-injuries owing to constant repetition of the same movement (Lehman, 1984). A
42
recurrence of running injuries, results in 20% to 70% of recorded cases (van Mechelen,
1992; Brill and Macera, 1995). From epidemiological studies, it can be concluded, that
running injuries lead to a reduction or cessation of training for 30% to 90% of all
documented injuries (Walter et al., 1989). Medical consultation or medical treatment is
required for 20% to 70% of all injuries, with up to 5% resulting in absence from work
(Gudas, 1980; Macera et al., 1989).
Aetiological factors associated with running injuries include previous injury (Powell,
1986), lack of running experience (Powell, 1986; Blair et al., 1987), competitive rumiing
and excessive weekly running distance (James et al., 1978; Andrews, 1983; Harvey,
1983; Stanish, 1984; McKenzie et al., 1985). The association between running injuries
and factors such as warm-up, stretching exercises (Koplan et al., 1982; Jacobs and
Berson, 1986), stature (Walter et al., 1989), misalignment (Clement et al., 1981;
Lysholm and Wiklander, 1987), muscular imbalance (van Mechelen et al., 1987),
restricted range of motion (Subotnick, 1985), running frequency (Jacobs and Berson,
1986), ability (Clement et al., 1981) stability of running pattern (Stanish, 1984), and
shoes and in-shoe orthoses (Kaelin et al., 1985; Frederick, 1986; Winter and Bishop,
1992) remain unclear. Age (Koplan et al., 1982; Blair et al., 1987; Holmich et al., 1989;
Walter et al., 1989), gender (Johansson, 1986), body mass index (Koplan et al., 1982),
running hills (Yzerman and Van Galen, 1987; Macera et al., 1989), rumiing on hard
surfaces (Clement et al., 1981), participation in other sports (Jacobs and Berson, 1986;
Yzerman and Van Galen, 1987), time of the year and time of the day (Yzerman and Van
Galen, 1987) are not significantly associated with running injuries. Yet, Lysholm and
Wiklander (1987) registered a significantly higher injury incident rate during spring and
summer, a time typically associated with increased volume and intensity of running.
Running volume was an important risk factor for increased injury incidence, whereas,
running velocity was not associated with injury risk (Yzerman and Van Galen, 1987).
Powell (1986) concludes that sudden increases in volume and intensity are the cause of
running injuries, because the body has not had time to adapt to the new regimen.
Holmich et al. (1989) recommended as an injury prevention method that runners reduce
their volume. One solution was to run less than 30 km per week, which substantially
reduced injuries (Holmich et al., 1989), although it is not a realistic option for elite
athletes.
43
Eccentric and concentric muscle contractions in a stretch-shortening cycle are potential
components for overuse injury (Johansson, 1992). Repetitive impact loads may decrease
the ability of leg extensor muscles to sustain impact loads and, subsequently, the muscles
would lose part of their recoil abilities (Komi et al., 1987). In support, muscle strains
usually occur during eccentric contractions (Glick, 1980). For instance, hamstrings
muscles act in the running stride to decelerate knee extension, and the quadriceps act to
decelerate knee flexion, therefore, acting eccentrically (Cavagna et al., 1977). No
mention of different running styles or technique was made in reference to eccentric
contractions and injury incidence.
Running injury epidemiology should include denominator-based incidence rates, in

which the number of new injuries observed during one year is related to the population
of runners at risk (Hoeberigs, 1992). In ten studies with denominator-based incidences
selected from the literature, the annual incidence rates of injured runners vary from 24 to
65% (Hoeberigs, 1992), which is in fact similar to the incidence rates. Comparison of
denominator-based incidence rates from different studies requires clarity of what the
denominator and the numerator are, for example, the study population and the definition
of a running injury (Blair et al., 1987). Injury definitions differ from one study to the
next (Koplan et al., 1982). Subgroups may differ in origin: volunteers, runners from a
mailing list or entrants to a road race. Incidence rates are higher among supervised
volunteers than among listed runners, and higher in both these groups, compared with
race-entrants (Hoeberigs, 1992).
Biomechanical research into injuries

Current biomechanical injury research has primarily focused on impact forces and
different ruiming surfaces. Lower-limb biomechanics are discussed in reference to
potential impact forces. Anatomical mechanisms related to impact will be reviewed.
Vertical impact force has been strongly associated with the transmission of shock waves
upward through the musculoskeletal system (Radin and Paul, 1971; Verbitsky et al.,
1998). It has been suggested that a positive relationship exists between the magnitude of
impact force and the numbers of overuse injuries (Voloshin and Wosk, 1982; Wosk and
Voloshin, 1982; Cavanagh, 1990). Impact force with insufficient attenuation by natural
shock absorbers of the locomotion system, may cause over-loading, resulting in
44
subchondral bone micro-fractures, and subsequent articular cartilage degeneration and
osteoarthritis (Wosk and Volishin, 1982). There is speculation that impact-force peaks
play a role in the development of pain and injuries in runners (James et al., 1978;
Clement et al., 1981), indicating potential differences in injury rates for forefoot and
heel-toe runners. Furthermore, increased ground reaction forces might result in the need
for more intense muscular contributions to control segmental movements and stabilise
body position during the support phase of running (Williams, 1990). This would result
in an increased metabolic demand from the muscles involved. Williams tested this
hypothesis using twenty-two runners by inducing fatigue (determined when end tidal C0 2
pressure decreased). A lightweight accelerometer was placed on the tibial tuberosity and
whenever fatigue ensued, peak acceleration during stance increased. He concluded, that
fatigue and heel-strike induced shock waves are correlated. Fatigue-induced changes in
shock wave attenuation can therefore have important implications for the etiology of
injuries (Verbitsky et al., 1998).
Laughton et al. (2003) found a direct linear relationship between vertical ground reaction
force and peak tibial acceleration. However, Hennig and Lafortune (1991) were only
able to find a moderate correlation. Clarke et al. (1985) clarified the issue through use of
an accelerometer attached to the lower leg. He identified tibial accelerations increased by
taking longer strides, giving circumstantial evidence that increased stride length results
in increased impact force. The estimated muscle moments during sprinting correlated
significantly (r = 0.7 at p = 0.05) with the magnitude of the knee flexor moment at
impact for athletes with a history of hip extensor and knee flexor injuries (Mann and
Sprague, 1980). Findings from Stergiou et al. (1999) support this study, in that they
identified coordination patterns between the ankle and knee joint change with vertical
impact forces. Overall, these studies suggest lower-limb activity at impact affects the
vertical loading of the body (increase risk of injury), but again no mention is made of
how technique modifications can affect these changes.
Crossley et al. (1999) using forty-six males, half with a history of tibial stress fractures
and half with no record were measured for ground reaction forces and tibial bone
geometry. They discovered the tibial stress fracture group had significantly less tibial
cross sectional area (p = 0.02) than the non-injured runners. No ground reaction force
differences were found, underlying the difficulty in early identification of chronic
45
injuries. Kawamoto et a l (2002a) measured torsional moments at the tibia for running on
curved paths. Six male participants running unshod at 3.5 m-s'1 navigated a track
curvature of radii 15 m and 5 m. For comparison, a straight section of track was used.
Torsion moments increased from the~straight to the most sharply curved track (11 Nm,
12.2 Nm and 28.5 Nm respectively). For comparison, a typical 400 m running track has a
radius of 35 m and the static fracture load o f human tibia is 100 Nm. However, track-
running speeds are usually several times faster than the study speed. Forwood and Parker
(1989) reported less than 16.3% of the ultimate loading force could cause observable
micro-damage in rat tibia, therefore, lending some risk of tibial torsion in running on
tightly curved paths. Kawamoto et al. (2002b) took eight male runners and ran them on a
straight running track at 5 m-s'1. The moment for tibial torsion showed considerable
inter-individual variations during stance, and was correlated with vertical ground
reaction force (r = 0.7 at/? = 0.05) and outward tilt angle of the shank in the frontal plane
(r = 0.78 at p = 0.05). A suggestion from the authors was reductions in ground reaction
force, which reflects the acceleration of the centre of mass, might lead to a reduction in
running injuries, in athletes’ tibiae. Perhaps, a more specific strategy should seek an
understanding of how to modify the centre o f mass’s movement patterns that
subsequently produced the different tibial torsion.
Loading characteristics of the foot that are associated with fatigue were investigated in
nineteen non-injured heel-toe runners (Willson and Kemozek, 1999). A Borg rate of
perceived exertion scale quantified fatigue. Increased cadence and decreased loading of
the heel at landing through increased medial forefoot loading occurred when fatigued.
Christina et al. (2001) measured eleven female recreational runners who produced
significant changes for loading rates, impact and push-off peak magnitudes for vertical
ground reaction force and ankle joint motion. Further, a less inverted foot at heel-strike
at 2.9 m-s'1 resulted from induced fatigue of the ankle invertors and dorsiflexors. These
changes may play a role in lower extremity running injuries (Christina et al., 2001).
During a 10-km race, eight elite runners were filmed at different stages o f the race (Elliot
and Ackland, 1981). Fatigue produced a more extended lower-limb throughout the gait
cycle during the later stages of the race. Clearly, different foot contacts and possibly
limb-body geometry affect loading and muscle activity at impact. Therefore, technique
should play an important role in injury reduction interventions.
46
Surface stiffness has been associated with overuse injuries, although mainly from
circumstantial evidence. Dixon et al. (2000) used six heel-toe runners who performed
shod running trials over three rumiing surfaces: conventional asphalt, a rubber-modified
asphalt surface and an acrylic sports surface. Surfaces were categorised according to
impact absorbing ability (BS 7044). Mechanical impact absoiption was greatest for
rubber-modified asphalt, whereas conventional asphalt was the least. The rubber-
modified asphalt compared with the conventional asphalt exhibited lower participant
loading rates for peak impact force (p = 0.1). Individual and not group differences were
found for kinematic variables. For example, peak joint angles, peak joint angular
velocities and initial joint angles. Explanation of the different peak impact forces on
conventional asphalt for some runners was via increased initial knee flexion, while
others did not show compliance. This suggests that the lower extremity exhibited
compliance on an individual basis to the harder surface stiffness, perhaps alluding to
certain runners’ skill in reducing load through attenuated movement patterns.
Stride smoothness is a quantitative measurement based on jerk-cost (JC) function

(Hogan et al., 1987).
JC = /To (d3r)2 dt / dr’ (2.4.1)
Where T is total movement time, and r is the position vector of a limb segment. When
applying this function to evaluate smoothness of complex multi-joint movements, it has
been demonstrated that the position vector, r, is best represented in terms of end-point
Cartesian coordinates (Hogan et al., 1987). The theory of maximum smoothness
(minimum jerk) suggests that the trajectory of a movement endpoint is planned in a
manner that would minimise the jerk-cost function. In running, the endpoint is the foot.
Winter (1987) suggested the major motor function during the gait cycle is control of the
foot. Control of the foot trajectory achieves safe ground clearance and a gentle heel or
forefoot landing. In support, Dixon and Kerwin (1999) found significant correlation with
heel-lift height and reduced ankle dorsiflexion at heel contact, signifying an increased
surface area for the foot at impact, suggesting heel contact is undesirable.
Using a 1-metre depth jumping exercise as an exaggerated motion for landing while
running, ten healthy males landed on their heels or forefeet (Kovacs et al., 1999). The
47
first vertical ground reaction force impact peak was 3.4 times greater for heel-toe landing
compared to forefoot landing. The second peak was 1.4 times lower. During flexion, the
hip and knee joints contributed 40-45% of the total torque (heel-toe group), whereas
forefoot landing found the knee and ankle joints contributed 37% each. Total torque
contributions were highest in the heel-toe landing (during the extension phase) with 41%
and 45% for knee and ankle joints and 34% and 55% with the forefoot.
Electromyography for vastus lateralis muscle activity differed during re-contact for the
heel-toe landing, whereas gastrocnemius muscle activity in the forefoot group increased.
Exaggerated motion (depth jumps) highlights greater loading characteristics between
heel and forefoot contact, which in light of previous comments, suggests limb and body
positions relationships play important roles in force attenuation and injury incidence.
Hreljac (2000) procured twenty-four well-trained athletes of whom twelve were

experienced runners and twelve were team players (non-runners). All were heel strikers.
The heel ascended and descended (lower jerk-cost) more gradually for the experienced
runners than non-runners despite slight differences in vertical displacement. Bergmami et
al. (1995) hypothesised that smooth gait patterns reduce joint loading and thus decreased
the risk of injury. Hreljac et al. (2000) in a second study, took two groups of runners,
those who had at least one overuse running injury and others who had been injury free
for their entire running career. Using a sit-reach test, the injury-free group demonstrated
increased hamstring flexibility. The injury-free group had a lower vertical force impact
peak, maximal vertical loading rate and the maximal rate of rear foot pronation. This
suggests that the injury free runners had developed stride patterns that incorporate
relatively low impact forces, and a lower maximum rate of rear foot pronation. Muscle
induced fatigue for the hamstring group of twelve sprinters, running 40 m maximal run
repeats, identified differences between the fatigued and non-fatigued state (Pinniger et
al., 2000). These variables were significantly different from the fatigued maximal runs:
decreased knee and hip flexion at maximum knee extension in the swing phase,
decreased leg angular velocity immediately before impact and a decreased angular swing
phase. Electromyography indicated a significant increase in the duration of hamstring
activity and earlier cessation of rectus femoris activity during the swing phase. They
concluded the changes were a protective mechanism to reduce stress on the hamstring
muscles at critical stages of the gait cycle. Effective hamstring muscle function
48
(including increased flexibility), therefore, plays an important role in running
biomechanics, although its injury prevention remains unclear.
Overpronation at the subtalar joint, which is external rotation, dorsiflexion and calcaneal
eversion has been associated during the stance phase with injuries to the hip, knee,
Achilles tendon and the foot (James et al, 1978; Taunton et al. 1982). As body weight
moves forwards over the support foot after impact with the ground, the subtalar joint
everts, the ankle dorsiflexes, while the knee and hip joints also flex. Stress fractures have
been associated with poor alignment and overpronation (Slocum and James, 1968;
Lysholm and Wilander, 1987). However, Stanish (1984) contested that overpronation
was a causative factor in stress fractures. He cited lack of scientific support. A more
recent review concludes that lower extremities injuries are still not well understood in
reference to pronation in runners (Hintermann and Nigg, 1998). Risk of injury owing to
increased and excessive pronation is unclear at this time in heel-toe runners.
Iliotibial band friction syndrome (lateral knee pain) and patellofemoral pain syndrome
(anterior knee pain) are two of the most common complaints in runners (Clement et al.,
1981; Johansson, 1986). Downhill and fast running aggravated anterior knee pain. These
activities increase the negative impact from retarding knee flexion and extension motions
(Bennet and Stauber, 1986). To decrease load, increased initial knee flexion is used
(Dixon et al., 2000). Bennet and Stauber (1986) administered eccentric training to an
injured group of runners with good results, concluding motor control imbalance was a
contributing factor to anterior knee pain. Lateral knee pain results from an increased
compression of the iliotibial band against the lateral femoral epicondyle. Sutker et al.
(1981) reported iliotibial band friction syndrome was an overuse injury most commonly
affecting males running about 30 km a week for at least three years. Several etiological
factors given were excessive pronation of the subtalar joint, a tight iliotibial band and
excessive genu varum. Nutig (1981) observed this injury occurred from uphill running
owing to increased loading. Not mentioned was the runner’s uphill technique may have
altered their loading compared to level running.
A recent two-year study provides the latest published database for specific running
related injuries (Taunton et al., 2002). Risk factors were associated with being under
thirty-four years old for patellofemoral pain syndrome, iliotibial band friction syndrome,
49
patella tendinopathy and tibial stress syndrome. Interestingly, McKenzie et al. (1985)
recorded similar results nearly 20 years ago. Being active for less than eight and half
years was positively associated with tibial stress syndrome. Patellofemoral pain
syndrome was the most common injury, followed by iliotibial band friction syndrome,
plantar fasciitis, meniscal injuries of the knee and tibial stress syndrome (Taunton et a l ,
2002). These data suggest a chronic injury accumulation, possibly owing to poor running
biomechanics. Supporting this line of enquiry, Lewis et al. (2000) in their review on
back injuries identified runners lifetime prevalence is as high as 47%. Specific
predisposing factors to injury included: poor stability, muscle imbalances, weakness and
fatigue, including an increase in load imposed on the spine while running. Again, these
studies are descriptive without reference to prevention. However, a theoretical model
that follows the laws of physics could establish a standard (ideal technique). Deviation
from the standard technique would then quantifiably address injuries mechanisms. Those
runners that deviate from the standard in a certain dimension would possibly develop
certain types of injuries. Potentially, reduction in injuries may then occur through
changes in technique.
Injury and technique

The correlation between running technique and injury has little direct research. Nytro
(1987; p. 3196) states the current view,
“there is no scientifically founded technique, that suits everyone ... therefore there are
no possibilities o f evaluating individual dispositions fo r a certain event. Absolute
postulations that ‘this is wrong’ and ‘this is right’ are only revealing the coach’s lack o f
insight into the technical evaluation...”
Only one injury and technique study was found: McClay et al. (1999), possibly owing to
the viewpoint above. In a single participant design, they used a 40-year old female with
right plantar fasciitis and a left tibial stress fracture (recently healed) who underwent a
gait analysis (VICON 6 cameras at 120 Hz; BERTEC Corp, OH.). An 8-week, 3-times
per week training intervention was administered. The training intervention took place on
a treadmill running at 3.35 m-s*1. A mirror provided feedback for the first three weeks.
The participant began running for 10 min and progressed to 32 min by the end o f the
eight weeks. Hip internal rotation and adduction plus knee abduction for the right leg
were altered (improved). Peak braking force and loading rates on the left foot decreased.
50
These results suggest it is possible to modify gait biomechanics and potentially reduce
the risk of injury, because previous symptoms of pain abated for their participant.
2.0 SUMMARY
Many of the reported findings present conflicting observations for economy, optimal
biomechanics and potential injury mechanisms found in runners. The aim of the review,
therefore, has been threefold: To highlight confusion that exists within the research
literature, to review various assumptions, while presenting relevant findings. The
summary’s purpose is to draw together the various findings, in an attempt to clarify
whether a universal running technique currently exists.
Although there is an intuitive link between running biomechanics and energy cost, the
belief persists that individuals are unique both anatomically and biomechanically, and so
arrive at their own most economical running motion. However, biomechanical variables
account for at least 54% of the variation in running economy (Williams and Cavanagh,
1987). An alternative viewpoint suggests runners become economical through
continuous repetition of their gait pattern at a familiar speed, regardless of its
biomechanical effectiveness. For example, intra-individual oxygen consumption
variation exists between different running speeds— 2.83 m-s'1 to 4.88 m-s'1—highlighting
experienced runners become economical at their own well-trained pace (Morgan et al.,
1989), which might not always be associated with ‘good’ biomechanics. The optimal
economical solution identified was a combination of self-selected stride lengths and
stride frequencies.
Immediately a contention arises because, certain ‘skilled’ runners tend to have longer
strides at any given velocity than ‘poorer or less skilled' runners, while in contrast it was
found elite runners took shorter absolute and relative strides than good distance runners.
There appears to be an optimum stride length, however, its cause is complex. Driving the
leg forwards increases stride length, which is ineffective owing to increased braking
forces from longer strides. Foot placement in front of the body at impact decelerates the
body, which may affect the next stride. Further, driving the knee forwards increases
work against gravity. Increasing running speed through increased ground reaction force
51
appears to be ineffective, because amplified push-off from the ground was absent in elite
endurance runners, who recorded lower vertical ground reaction force for the second
vertical peak than good runners. Over-striding and short strides increase internal work.
Over-striding requires an increased thigh extension and leg drive shown to be
uneconomical. Further, tibial acceleration increases (from longer strides), raise impact
forces and potentially increase injuries. Shorter strides can increase internal work. An
alternative external force to ground reaction force that enabled forward progression in a
legless man was gravity. This novel idea of gravity being the prime motive force in
running requires exploration. If the length of stride were as long as the horizontal
displacement of the body during flight, and the foot landed under the body, theoretically
lower braking force would occur. Increased trunk lean from the vertical axis improved
economy and increased running speed, which suggests stride length from increased trunk
lean maybe effective. Hip horizontal velocity was greater than neck velocity at terminal
stance, indicating the drive forwards emanates from the hip. Increased forward
progression correlated with hip flexion. Less plantar flexion, and higher hamstring
activity after terminal stance were associated with runners that are more economical.
Further knee extension did not increase with rising speed while electromyography of all
gait muscles ceased well before terminal stance. It is therefore possible that running
speed originates from the body (gravity’s work on the runner’s mass) and not from the
leg-drive usually associated with stride length. Hence, differences in stride length maybe
attributed to the combination of gravity and ground reaction force, together with body-
foot position while reducing muscle activity at terminal stance.
Continuing to evaluate external forces and rumiing, an interesting observation is that

vertical oscillation of the centre of mass decreases with increases in running speed. In
addition, increases in ground reaction force (associated with increased running speed)
should raise the centre of mass. However, increased trunk lean and knee flexion during
stance (and flight), plus shorter contact time also occur with increases in running speed.
Running faster increases momentum of the centre of mass. Shorter time on the ground
and less braking force conserves momentum allowing the centre of mass to move more
quickly past the point of support (support foot). Increased momentum may explain the
lack of push-off (less knee extension) and muscle activity at terminal stance seen in elite
runners. Lower leg extension would also potentially reduce vertical oscillation at
terminal stance.
52
The stretch-shortening cycle is maximised with a forefoot landing with increased knee
flexion (greater eccentric stretch) and shorter contact time. Forefoot landing, increased
knee flexion and shorter contact time may allow the lower-limb to facilitate improved
elastic energy, theoretically enhancing economy owing to reduced internal muscle force.
In support, economical runners have effective co-activation of their biarticular muscles,
which they potentially maximise via a certain lower-limb geometry.
Stride frequency increases by only 4% at speeds of 3 m s'1 to 4.2 m-s'1. With increased
speeds (sprinting above 7 m-s'1), stride length did not increase significantly, whereas
stride frequency continued to rise. At sprint speeds keeping the foot under the body
differentiated Gold from Silver medallists at the 1984 Olympics. Applying this concept,
stride frequency should correspond with the speed of the body to enable the foot to land
under the body to allow the next stride to begin. Lack of higher frequency at endurance
speeds may be associated with mainly heel-toe runners landing ahead of their bodies thus
creating a slower turn over. Possible support for this idea is more economical runners
had an increased hamstring activity during swing, which allowed for a faster leg
recovery.
The review highlights the current lack of a clear universal running technique. However,
certain variables were associated with improved performance. Therefore, the summary
points out elite runners do not extend the knee, which requires less push-off and lower
vertical oscillation. They plantar flex the ankle less, have an increased trunk lean, land
with the foot more vertically aligned to their centre of mass and potentially utilise elastic
mechanisms more effectively. Further, leg and hip extensor muscles were not active
during leg extension. Finally, gravity, a constant force, may provide momentum
conservation through increased trunk lean. As the body leans past its point of support
(support foot), a gravitational torque overbalances the runner driving them forwards,
seen in the legless walkers. Gravity may provide an alternative external force for
locomotion. This thesis, therefore, aims to research a universal running technique that
explores gravity as the motive force. This universal running technique should potentially
reflect the variables associated with improved performance.
53
CHAPTER 3
THEORETICAL MODELS
3.1 INTRODUCTION
A brief background of the historical development of running technique is given. Key

discoveries are highlighted. Possible reasons for heel-toe running becoming the most
popular style of running in recent years conclude the background. Then, two hierarchical
models of running are discussed: the current hierarchical model of running (Hay and
Reid, 1988) and a new Gravitational hierarchical model based on the Pose® method
technique.
3 .2 BACKGROUND
Running has interested people throughout history from a theoretical and practical
standpoint. Many earlier observations and measurements were attempts to develop or
understand a clear biomechanical model of efficient running.
Several Greek vase paintings depict sprinters (circa 470 B.C.) and endurance runners
(circa 525 B.C.), which clearly illustrate differences in limb range of motion (Gardner,
1967; p. 134 and p. 138 respectively). Both runners land on the ball o f the foot and no
one straightens their legs at terminal stance. Whether this is artistic license or astute
observation may never be known. During this time, Aristotle (384-322 B.C.) wrote the
earliest known accounts of locomotion:
Concerning vertical oscillation of the body:

“If a man were to walk parallel to a wall in sunshine, the line described (by the shadow
o f his head) would not be straight but zig zag, becoming lower as he bends and higher
when he stands and lifts himself up” (Encyclopaedia Britannica, 1952; p. 247).
54
Concerning joint motion:
“For if one o f the parts o f an animal be moved, another must be at rest, and this is the
purpose o f their joints; an imals use joints like a centre, and the whole member, in which
the joint is, becomes both one and two, both straight and bent, changing potentially and
actually by reason o f the joint” (Nussbaum, 1978; p. 32).
Early resemblance to Newton’s third law:

“But the point o f rest in the animal is still quite ineffectual unless there be something
without which is absolutely at rest and immovable. And further the force o f that which
initiates movement must be made equal to the force o f that which remains at rest. For as
the pusher pushes so is the pushed pushed, and with equal fo rc e ” (Nussbaum, 1978; p.
44).
Another eighteen hundred years passed before records of running research are available.
Leonardo Da Vinci (1452-1519) wrote the following concerning motion, balance and
foot contact.
“Motion is created by the destruction o f balance, that is, o f equality o f weight, fo r
nothing can move by itself which does not leave its state o f balance, and that thing moves
most rapidly which is furthest from its balance” (McMahon 1956; pp. 133-134).
Again, this is re-stated. “That figure will appear swiftest in its course, which is about to
fa ll foi'wards” (McMahon 1956; p. 135); “A man running on level ground has it first on
his heels and then on the toes o f his fe e d (Keele, 1983; p. 175).
Giovanni Borelli (1608-1679) was the first to systematically study locomotion in his
classic “De Motu Animalum” published in 1682. He measured the centre o f mass of the
body during walking and described how balance is maintained through constant forward
movement of the supporting area provided by the feet. Wilhelm and Eduard Weber one
hundred and fifty years later published in 1836 a detailed work on human gait. Basic
definitions established distinguishing differences between step length in sprinters and
walkers; running involved smaller vertical oscillation than walking and the lower-limb
appeared to act as a pendulum.
Etianne Jules Marey (1830-1904) was the first to record displacement and force
simultaneously.
He noted: “All muscular actions which alter the centre o f gravity o f the body in such a
manner as to raise it augment the foot pressure on the ground’ (Marey 1972; p. 150).
55
Vierordt, (1881) was the first to record quantifiable evidence for the variability in step
parameters, and Eadweard Muybridge (1830-1894) produced excellent photographs of
running gait showing foot contact to be forefoot in barefoot running.
In the early twentieth century, Graham-Brown (1912; p. 875) wrote.

‘'The cycle ofprogression may be supposed to commence at a point at which one o f the
limbs is perpendicular to the ground. The ‘initial velocity’ then carries the body past this
point, and it then falls forwards along the circumference o f a circle the radius o f which
is the limb in contact with the ground. ”
A.V. H ilf s (1927) Nobel Prize winning work, which included the development of a
velocity curve for sprinting and the efficiency of running, showed that a combined
physiological and biomechanical approach could provide considerably more insight than
either approach in isolation. Fenn (1930) building on Hill’s work used film records, and
a force platform, where he estimated the work completed against gravity through total
body segment analysis. The migration of the centre of mass with respect to a fixed point
on the body was studied showing it moves upwards and backwards, then downwards and
forwards twice in each running cycle.
The modern growth and interest in running research has been spurred by the vast
increase in participation in the late 1960's and early 1970’s. This period of research was
essentially descriptive. Early work focused on the rear foot using high-speed
cinematography in two-dimensions digitized by hand (for example, Inman, 1976; Bates
et al., 1978). Alongside this were technical advances in shoe design. Interest in
differences between barefoot and shod rumiing were identified during this period. Shoes
tend to cause greater pronation producing an increased impact angle (Bates et al., 1978;
Nigg and Luethi, 1980). They identified barefoot impact angles ranged from 0.8°-1.9°,
whereas shod runners ranged from 3.7°-11.8°.
Although rear foot motion was the focus of most running studies, other studies focused
on sagittal plane motion of the ankle, knee and hip joints (Nigg, 1986). Nilsson and
Thorstensson (1985) found as running speed increased, joints moved through increased
ranges of motion in less time, resulting in higher angular velocities. Cavanagh and
LaFortune (1980) were the first to identify differences between rear foot and forefoot
56
strikers: the impact peak of the vertical ground reaction force curve was attenuated or
missing in forefoot strikers.
A new generation of research emerged in the 1990’s with the advent of automated
motion-analysis systems capable of obtaining three-dimensional data. The ability to
assess secondary planes of motion must still be viewed with caution owing to errors
caused by marker placement and soft-tissue movement (Allard et al., 1995). However,
coordination and timing of lower extremity joints were topics that could now be
examined. Further, more hypothesis driven research was being reported in the literature
(Bartlett, 1997; McClay, 2000). Prior research focused on heel-toe runners, and it was
not until recently kinematic differences between rear foot and forefoot runners were
investigated. McClay and Manal (1997) compared lower extremity biomechanics of ten
rear and forefoot strikers. Forefoot strikers landed with greater flexion and external
rotation at the knee.
Recent research has focused on the interaction of lower-limb joints (Zajac, 1993), rather
than previously, on individual joint activity. For example, coordination of movement
between the rear foot and knee may result in injury (Bates et al., 1983). The rear foot and
knee are linked mechanically through the subtalar joint, thus motions at the foot will
influence movements of the knee known as a coupling effect (McClay, 2000). Joint
coupling can also be explored by assessing the relative timing of motion between two
joints (De Wit and De Clercq, 2000).
It is not possible, therefore, to conclude that before modern running shoes most runners
landed on the forefoot as opposed to eighty percent of runners today, landing on the heel
(Kerr et al., 1983). Leonardo da Vinci concluded many years before running shoes that a
man running lands first on their heel and then on their toes. Therefore, it is not certain
whether modern running shoes really changed the biomechanics of runners to create
predominantly heel-toe running. Individual technique differences seem a more plausible
explanation for choosing heel-toe or forefoot running.
Gravity’s role as a constant force in running has been the subject of little research in
modern times (1931-present). Observations and measurements from the past (Leonardo
Da Vinci, 1452-1519; Graham-Brown, 1912; Fenn, 1930) have still not been harmonised
57
into a universal running technique. Instead, the current prevailing scientific belief is that
each runner optimises their own ideal running biomechanics (Nytro, 1987). This is in
contrast to swimming which has attempted to teach universal techniques of freestyle to
children as the ‘only way to swim’ (for example, Counsilman, 1977; Laughlin and
Delves, 1996).
3 .3 TWO HIERARCHICAL MODELS OF RUNNING
3.3.1 Introduction
From a biomechanical perspective, running is simply the horizontal displacement of a

runner’s centre of mass. However, scientists, coaches and athletes have a disparity of
views on this activity. Some scientists insist, “there is no commonly accepted running
model which will suit everyone” (Nytro, 1987). Similarly, the common views o f coaches
and athletes are: “No, there is no correct running form, and you couldn’t learn it. Form
is God-given. If you systematise it, you destroy z7” (Wallack, 2004). This thesis takes an
opposing view that a clear and comprehensively applied universal technique o f running
will be helpful to the running community.
Movement is the changing position of a material body in space, and the reason for this
change is the action of a force. Classical mechanics uses the term ‘force’ as a
quantitative measure of the interaction between material bodies. The action o f force on a
free material body (not interacting with a third party) is the acceleration of that body.
Therefore, a force alters, or has a propensity to alter, the state of motion of a runner’s
body. Whether at rest, or moving with constant velocity in a straight line, the runner will
continue in that state unless compelled to change by an external force exerted upon them.
An applied external force enables accelerated forward motion (horizontal displacement
of the centre of mass) for the runner. Their acceleration is directly proportional to the net
force acting on their body and inversely proportional to their mass. Hence, the body will
accelerate in the direction of the net unbalanced force. Increased horizontal acceleration
only occurs during stance, while during flight gravity accelerates the runner towards the
earth, with air resistance minimally reducing horizontal acceleration.
58
A certain number of known forces exist in running: external forces include gravity
(Graham-Brown, 1912; Femi, 1930) and ground reaction force (Cavanagh, and
Lafortune, 1980) plus an internal force, such as, muscle force (Heise et al., 1996); and
muscle elasticity (Cavagna et al., 1964). Air resistance is neglected in-line with most
other gait research (Anderson, 1996). These forces only increase horizontal acceleration
during stance but not flight. The centre of mass’s horizontal displacement during flight is
therefore, a consequence of the action of these forces throughout stance. The hierarchical
order of these forces involved in running and their relationship to each other is not
currently lucid. This thesis seeks theoretical (hypothetical) and numerical answers on the
role of gravity as the main force defining the interrelations between the applied external
and the internal forces involved in running. Numerical work on gravity’s role in running
will follow in later chapters. Section 3.3.2 briefly reviews the heel-toe technique of
running using Hay and Reid’s (1988) hierarchical model, while section 3.3.3 presents a
new Gravitational hierarchical model of running based upon the PoseR technique
(Romanov, 2002).
3.3.2 The current hierarchical model of running

The literature review provides background regarding current running research. This
section aims to review the only current attempt to hierarchically model running.
Therefore, examination of Hay and Reid’s (1988) model in figure 3.3.2 summarises the
current lack of a clear integration of the forces involved in running. For example, in the
current model, gravity is not in affect during stance nor does it relate to forces exerted
(muscle and joint forces; Hay and Reid, 1988; p. 264) and time forces act despite gravity
being a constant force. The basis of this section is to briefly show the current hierarchical
model of running fails to identify gravity’s key role.
Description of the current running model (adapted from Hay and Reid, 1988)
In running (figure 3.3.2), the result is the time taken by the runner to cover a specified
distance. Two factors determine time: The distance involved and the mean speed of the
runner over that distance. The distance that the runner travels is generally subject to race
conditions with minor variations, and is therefore not developed any further in this
model. The mean speed of the runner is equal to the product of the runner’s mean step
length and mean step frequency.
59
Step length may be considered the sum of three separate distances:
1) The horizontal distance between the runner’s centre of mass and the toe of the
take-off foot at the instant of take-off (take-off distance);
2) The horizontal distance that the centre of mass travels while the runner is in the
air {flight distance); and
3) The horizontal distance between the centre of mass and the toe of the leading foot
at the instant the runner lands {landing distance).
Take-off distance (figure 3.3.1) for a given runner is determined by their physique and
body position at the instant of take-off. Landing distance is similarly governed by
physique and body distance at the instant of landing. Flight distance of a runner is
determined by the speed and angle of projection into the air, relative height at take-off
air resistance and acceleration o f gravity. These are expressed as velocity at take-off the
relative height at take-off and air resistance.
«----------- ►N—
T a k e - o f f d is t a n c e F lig h t d is ta n c e L a n d in g d is t a n c e
Figure 3.3.1 F lig h t, t a k e - o f f a n d la n d in g d is ta n c e to ta llin g th e r u n n e r ’s ste p fo r th e c e n tr e o f m a s s . N o t e
th e r u n n e r is f a llin g d u r in g m o s t o f f lig h t.
Change in velocity is equal to velocity at take-off and velocity at impact. The change in
velocity during the support phase is determined by the forces exerted (muscle and joint
forces) on the runner, the times for which these forces act {time forces act), and the mass
of the runner.
Mean step frequency of a runner is determined by the mean length of time taken to
complete one step {mean step time). Step time is the sum of:
1) Time of ground contact {time o f support).
60
2) Time in the air (time o f non-support).
Time of non-support is determined by velocity at take-off relative height at take-off air

resistance and acceleration o f gravity.
Hay and Reid (1988) suggest that the sub-division of the performance criterion be
broken into a series of distinct, consecutive parts such as take-off distance and flight
distance etc. Following this procedure, the identification of performance parameters that
affect the performance criterion are selected. These performance parameters should be
mechanical quantities that are measurable (Bartlett, 1999). For practical application in
their running model, the forces exerted need clear specification similar to their high jump
model (Hay et al., 1981). For example, ground reaction force only acts during stance,
whereas gravity is a constant force. These differences should be distinct in the model and
their affects on performance measurable. In fact, forces exerted in their model, represents
shoulders, hips, knees ankles and other joint forces (Hay and Reid, 1988; p. 264).
Despite Hay and Reid (1988) focusing on muscle force, each force in running (gravity,
ground reaction force and muscle elasticity) will be reviewed. Potential inconsistencies
will be highlighted.
61
Time
Distance Mean speed
Mean Step length Mean Step frequency
Take-off Flight Landing Step time

distance distance distance
Body Body
position / position /
physique physique
Time of non Time of

support support
Velocity at Relative height Air Gravitational

take-off at take-off resistance acceleration
Velocity at Change in
touchdown velocity
Forces Times
exerted forces act
Figure 3.3.2 Current hierarchical model of running (Hay and Reid, 1988; p. 282).
62
Internal force: Muscle activity
Co-ordinated muscle activity plays a significant role in running (Elliot and Blanksby,
1979). Muscles can change the motion of body segments so that their relative
accelerations contribute to the total ground reaction force (Cole et al., 1996). Hay et al.
(1981) include joint torques broken down into several time intervals in their updated
high jump hierarchical model, whereas their running model remains unspecified.
Beginning with muscle activation before impact, there are accelerations of body
segments from muscle contractions via the swing leg (Montgomery et al., 1994). From
impact until maximum vertical ground reaction force body weight lowers, which requires
the muscle system to support the runner at the knee and hip as the quadriceps and
hamstring muscle groups co-contract during early support during knee flexion (Elliot and
Blanksby, 1979; Mann et al., 1986; Montgomery et al., 1994; Heise et al., 1996). These
muscles are therefore, resisting the work of gravity, as the body lowers from impact to
mid-stance (maximum vertical ground reaction force and maximum knee flexion).
Maximum quadriceps muscle activity occurs at the transition between knee flexion and
extension coinciding with maximum vertical ground reaction force (Branded, 1973;
Nilsson and Thorstensson, 1985). After this, hip and knee extensor muscle activity
begins to decrease and ends just as leg extension begins (Branded, 1973; Mann and
Hagy, 1980; Schwab et al., 1983; Nilsson and Thorstensson, 1985; Montgomery et al.,
1994; Wank et al., 1998). Therefore, as the leg is rapidly extending, the leg extensor
muscles are silent; this has become known as the ‘extensor paradox’ (McClay et al.,
1990). Maclntrye and Robertson (1987) noted the hamstring group recorded the greatest
variability in phasic activity of ad the knee muscles they examined. Elliot and Blanksby
(1979) who reported maximal activity for one female for biceps femoris just before
terminal stance, in contrast to others who found no activity during terminal stance and
early swing (Branded, 1973; Mann and Hagy, 1980; Schwab et al., 1983; Nilsson and
Thorstensson, 1985). Kyrolainen et al. (2000a) identified poorer running performances
from limited action of their hamstrings.
Only two exceptions of muscle activity at terminal stance where identified. At terminal
stance the hip flexors are active, but show peak activity at the end of early swing
(Montgomery et al., 1994). Findings for the gastrocnemius muscle activity are
conflicting. Some studies have shown maximal activity at terminal stance suggesting a
63
push-off motion (Elliot and Blanksby, 1979, Mann, 1982; Jonhagen et al., 1996), while
others found activity persisted for only 30-50% of plantar flexion (Branded, 1973; Mann
and Hagy, 1980), or produced no push-off at all (Reber et a l, 1993).
However, it is commonly accepted that the foot pushes off the ground in running. Figure
3.3.3 shows muscle activity of the leg extensors for a single runner at 4 m s'1 taken from
a group of six participants. Muscle activity is synchronised with his knee flexion and
extension. Interestingly, all three leg-extensor muscles cease activity simultaneously
approximately 20 ms (mean for six participants 29.2 ±10.4 ms) after peak knee flexion,
while knee extension continues in this participant for another 150 ms (mean for six
participants 133.7 ± 16.5 ms). Extension time is longer than electromechanical delay
(range 30-100 ms) between electromyography activity and force production, meaning the
leg is extending without muscle force (Cavanagh and Komi, 1979). The mean knee
flexion and extension time for the six participants was 162.8 ± 19.5 ms.
P ea k k n e e fle x io n
4 u n n g sta n ce
P ea k k n e e
e x te n s io n after
stance^-
-120 -100-80-60-40 -20 100 140 180 220 260

Time (ms)
Rectus femoris
Vastus Lateralis
Vastus Medialis
Figure 3.3.3 R e s u lt s o f p h a s ic q u a d r ic e p s e le c t r o m y o g r a p h y a n d k n e e f le x io n - e x t e n s io n fo r a t y p ic a l
e n d u r a n c e ru n n e r a v e r a g e d o v e r s ix f o o t s tr ik e s (a d a p te d fr o m M c C la y et at., 1 9 9 0 ) .
These data (figure 3.3.3) were obtained from a treadmill, which may have influenced the
results. However, Mami et al. (1986) showed extensor muscle activity ceased as leg
extension begins using over-ground running. This lack of muscle activity is supported by
many studies using both treadmill and over-ground running (Branded, 1973; Mann and
Hagy, 1980; Pare et al., 1981; Schwab et al., 1983; Nilsson and Thorstensson, 1985;
64
MacIntyre and Robertson, 1987; Kyrolainen et a l , 2003). Gluteus maximus also extend
the hip, however their activity is in late swing and the first third of stance (Mann and
Hagy, 1980; Nilsson et al., 1985; Montgomery et al., 1994), showing this muscle is also
inactive during the propulsive phase of stance. No comprehensive explanation for this
‘extensor paradox’ presently exists.
Potential strain energy

Cavagna et al. (1964) estimated oxygen consumption during running would be 30%
higher without contributions from elastic storage and return of strain energy. Mechanical
efficiency of running exceeds the efficiency of the conversion of chemical energy to
kinetic energy by muscles suggesting there is another source of energy (Cavagna and
Kaneko, 1977; Cavanagh and Kram, 1985). Elastic strain energy stored during eccentric
contractions during early stance subsequently releases during concentric contractions
(Cavagna et al., 1964; Cavagna and Kaneko, 1977; Hinrichs, 1987; Alexander, 1992;
Taylor, 1994; Jung, 2003) and is the other source of energy.
To effectively use stored elastic energy in the muscle mechanisms, support time must be
short (Zatsiorsky, 1995; Paavolainen et al., 1999a). Although there is no exact method of
quantifying storage of elastic energy—although real-time B-mode ultrasound apparatus
looks promising (Fukunaga et al., 1997)—there is a consensus that this phenomenon
contributes to both efficiency and economy of movement (Cavagna and Citterio, 1974;
Winter, 1978; Luhtanen and Komi, 1980; Taylor, 1994; Cavanagh and Kram, 1985;
Komi, 2000). The runner’s body possesses minimum kinetic energy at maximum vertical
ground reaction force, while the stretch-shortening cycle of the muscular-tendon unit
contains maximal potential strain energy at this point. This potential strain energy results
from gravity’s work during impact. However, ‘an ideal’ body geometry and the ability to
reduce stance time have not yet been clearly documented.
External forces related to running: Ground reaction force

Ground reaction force is not a propulsive force as of itself in running, but operates
according to Newton’s third law (Zatsiorsky, 2002). The ground reaction force
experienced by the runner is equal and opposite to the force exerted by the runner on the
ground. Muscle force is an active internal force and can displace one body part with
respect to another, but it cannot displace the centre of mass without an external force (i.e.
65
ground reaction force). However, without the ground it is not possible to move using the
muscle system alone. During the propulsive phase of running, as noted earlier, the
muscle system is inactive (‘extensor paradox’), casting doubt that the ground in
combination with the muscle system can propel the runner into the air.
Ground reaction force increases as it resists body weight from impact to maximum
vertical ground reaction force and reduces after this point reflecting a rising body. The
body is rising, but the ground is not pushing the runner upwards (Zatsiorsky, 2002).
Presently, conventional running theory postulates applying increased ground reaction
force to increase acceleration of the centre of mass (Munro et al., 1987; Hay and Reid,
1988; Weyand et al., 2000). Further, increased running speed is directly associated with
a greater force application to the ground enabling the runner/sprinter to drive themselves
forward (Hay and Reid, 1988; Weyand et al., 2000). Hunter et al. (2005) however, found
it was not advantageous to have a large vertical impulse during the acceleration phase of
a sprint. In fact, their fastest runners only produced relatively moderate vertical impulses.
Further, as running speed increases the decay rate during the propulsive phase from
maximal vertical ground reaction force until terminal stance increases (Miller, 1990).
Hunter et al. (2005) also found vertical impulse non advantageous, because maximal
vertical ground reaction force occurs at mid-stance while the body is at its lowest
position with the centre of mass poised to move almost vertically upwards. A large
vertically directed impulse cannot produce a horizontal displacement for example, a step
length of 1 m, but rather step length requires a horizontal force. The increased ground
reaction force in sprinters may simply reflect their increased acceleration from their
bodies at impact. However, sprinters also exhibit the ‘extensor paradox’ suggesting they
are also not using the ground to accelerate themselves (Mann and Hagy, 1980;
Novacheck, 1998).
Viewing ground reaction force in heel-toe runners, two peaks are usually identified,
while forefoot runners elicit an attenuated first peak (Cavanagh and LaFortune, 1980).
The first vertical ground reaction force impact peak (figure 3.3.4) is associated with the
shock of contact with the ground. The first impact peak is related to body mass, stature
and step length, vertical hip excursion, running speed, running technique, area of the
foot-ground contact, as well as the properties of the shoe and surface involved (Hamill et
al., 1983; Frederick and Hagy, 1986; Nigg et al., 1987; Bobbert et al., 1991; Nigg and
66
Wakeling, 2001). Frederick and Hagy (1986) found body mass correlated highly with the
second peak of the vertical ground reaction force (r = 0.95 at P = 0.05) coinciding with
the lowest position of the centre of mass. The leg and hip extensors are inactive after this
point coupled with decreasing ground reaction force, which strongly suggests another
force is causing propulsion.
Second
peak
Impact peak
.Vertical force
3
Posterior-anterior
1 force
0 0.2
Time (ms)
Absorption Generation
Figure 3.3.4 A ty p ic a l v e r tic a l a n d p o s te r io r -a n te r io r g r o u n d r e a c tio n fo r c e c u r v e fo r a h e e l- t o e r u n n e r .
T h e fir st p e a k is a s s o c ia t e d w it h th e s h o c k o f c o n ta c t w it h th e g r o u n d (a b s o r p tio n ). T h e s e c o n d p e a k
(v e r tic a l d a sh e d lin e ) b e g in n in g at o r a b o u t m id - s t a n c e is a s s o c ia t e d w it h b o th b o d y m a s s a n d m u s c le f o r c e
g e n e r a tio n .
External forces related to running: Gravity

On Earth, gravity is by far the strongest mechanical force among all the forces of nature,
where nothing can cancel its attractive pull owing to its constant manifestation. As a
result, gravity should be considered as the dominant force on Earth. Anokhin (1978; p.
29) stated, “all biological systems, the most essential characteristics o f it, are defined by
the Universal Law o f Gravity. ” Despite this knowledge, the role of gravity is not
regarded as a dominant or leading force in the current hierarchical running model. The
current hierarchical model considers gravity only acts during flight, not stance. However,
this contradicts the dynamic similarity hypothesis that predicts that animals and humans
travelling with equal Froude numbers will use the same gait, because in equation 3.3.1
gravity is significant in determining the biomechanics of locomotion.
Froude number = (speed of locomotion)2 (3.3.1)

Gravitational acceleration x leg length
67
The Froude number is the square root of the ratio of inertial (mass x acceleration) to
gravity forces (mass x gravity), while Dynamic similarity is an application of the
pendulum principle, being the interaction between gravity and inertia. Motions are
dynamically similar if they could be made identical by uniform changes of the scales of
length and time. Thus, two pendulums of different lengths but identical angular
amplitude are dynamically similar. Likewise, two running animals with different leg
lengths but the same ratio of step-length-to-leg-length are also dynamically similar.
Mathematically, pendulums (and running animals) are dynamically similar if they have
equal Froude numbers. When Froude numbers are applied to animal locomotion, speed
of locomotion refers to forward velocity.
Animals exhibit a preferred walking speed above which they normally switch to a
running gait. The change from a walking to a running gait in humans occurs abruptly
rather than as a gradual transition (Thorstensson and Roberthson, 1987). Individual
humans switch from walking to running at different absolute speeds, but at
biomechanically equivalent speeds (Kram et al., 1997). That is the transition occurs at a
similar Froude number. Alexander and Jayes (1983) further state, that animals and
humans running have step lengths proportional to their leg lengths so:
A /1 = F (speed of locomotion)2 (3.3.2)
Gravitational acceleration x leg length
Where 1 is leg length, A is step length and F is the same function for animals of different
sizes and species. This clearly illustrates again that the current model of running, which
postulates step length is from leg extensor muscle activity is questionable. Animals and
humans have consistently shown step length is based on running speed, gravity and leg
length without reference to strength or muscle activity. Maxwell-Donelan et al. (1998)
though, state the Froude-based hypothesis ignores the role of muscle-tendon spring
forces in determining running mechanics. However, their calculations must include the
vertical Froude number (instead of the horizontal Froude number) indicating the muscle-
tendon spring is predominantly vertical in direction not horizontal. Alexander (1989)
using the Strouhal number support the muscle-tendon spring motion is in the vertical
direction.
68
In summarising this critique of the current running model, there are several important
questions. Firstly, if the runner pushes off the ground why is ground reaction force
decreasing from mid-stance and why are the extensor muscles silent at this time?
Secondly, gravity is a constant force, yet its work has not been discussed during the
propulsive phase of stance nor hierarchcially linked to forces exerted. The Froude
number clearly shows the importance of gravity in running, which is illustrated in the
Gravitational hierarchical model.
3.3.3 A Gravitational hierarchical model of running

The model in figure 3.3.5 attempts to hierarchically model all the forces involved in
running. A description with supporting evidence is given for the Gravitational
hierarchical model. It should be pointed out the evidence used here from the literature
was not discussed or identified by the authors in their papers. Therefore, the
Gravitational hierarchical model presented is the first to fully integrate and explain all
the forces in running and how to potentially maximise them.
69
Time 15
Distance
Relative Knee angular

height at velocity for stance Air
11- resistance
14 take-off and swing lB
Body position Time of Time of

/physique support non-support
Landing Flight Take-off

distance distance distance Step time
Mean horizontal velocity of the centre o f mass

throughout the gait cycle
Mean step Mean step

length 5 frequency 6
Ground reaction Muscle force/strain

force 2-4 energy
Centre of mass Acceleration

location to support of runner’s
limb 1 body
Gravitational
force
Figure 3.3.5 The Gravitational hierarchical model of running. A gait cycle is right foot contact to right
foot toe contact of the same foot.
70
Description of the Gravitational hierarchical running model
In running, the result is the time taken by the runner to cover a specified distance. The
distance is covered in less time by running faster. Angular velocity fo r knee flexion and
extension during stance reduce time o f support thus directly affecting running speed via
their relationship with reduced landing and take-off distance from a more aligned body
position/physique. Increased horizontal velocity o f the centre o f mass results from a
shorter step time (reduced time o f support/non support) because of a reduced landing
distance where the centre of mass is vertically above the support limb. Non-support time
decreases via increases in angular knee flexion velocity during swing leading to the
swing foot returning earlier to the ground. The relative vertical position of the centre o f
mass to the toe of the support foot determines the relative height at take-off Lower
angular knee flexion and extension velocity during stance affects the relative height at
take-off by increasing time o f support. Increased time o f support increases the relative
height at take-off through increased leg extension that raises the centre of mass, which
subsequently reduces the gravitational torque via work against gravity {gravitational
force).
The time o f support and non-support affect the horizontal velocity o f the centre o f mass
(less time for a given distance). Air resistance can reduce the horizontal velocity o f the
centre o f mass, which consequently affects the time o f non-support. The horizontal
velocity o f the centre o f mass at terminal stance (with body position and height at take
off) determines time o f non-support and step length. Driving the swing leg forwards via
hip flexors causes the foot to land in front of the body increasing landing distance, thus
decreasing horizontal velocity o f the centre o f mass and flight distance (Cavanagh and
LaFortune, 1980). The body {acceleration o f the runner’s body) causes the runner’s
horizontal velocity o f the centre o f mass via a gravitational torque {gravitational force)
rather than using the leg extensors to push off the ground during leg extension. Muscle
force is the result of gravity (via impact) and the action of the swing leg is to keep up
with the body {acceleration o f the runner’s body) from gravity during flight.
Mean step frequency is determined by the mean length of time taken to complete one
step {mean step time). Step time is the sum of:
1) The time of ground contact {time o f support).
2) The time in the air {time o f non-support).
71
Step length may be considered as the sum of three separate distances:
1) The horizontal distance is from the runner’s centre of mass and the toe of the support
foot at the instant of take-off (take-off distance). Take-off distance creates the moment
arm for the gravitational torque. The relative position of the centre of mass to the toe of
the support foot determines the relative height at take-off The faster the foot leaves the
ground via reduced leg extension less vertical oscillation ensues, which increases the
gravitational torque.
2) The horizontal distance is the distance the centre of mass travels while the runner is in
the air {flight distance). This is affected by horizontal velocity o f the centre o f mass,
relative height at take-off and the take-off angle {body position).
3) Landing distance is the horizontal distance between the centre of mass and the toe of
the leading foot at the instant the runner lands. The more vertically aligned the centre of
mass is above the support limb upon landing, the shorter time required before the centre
of mass moves past the support point (ball of the foot). A more vertically aligned body
causes less reduction in the horizontal velocity o f the centre o f mass (Fenn, 1930;
Deshon and Nelson, 1964; Cavanagh et al., 1977; Bates et al., 1979; Kunz and Kaufman,
1981; Girardin and Roy, 1984; Hinrichs, 1987) and shorter stance time. A shorter stance
time improves elastic recoil (Zatsiorsky, 2002).
Step frequency and step length are a consequence of the forces of gravity, ground
reaction, muscle action and elastic strain energy/. The timing and combination of these
forces produce the mean velocity o f the centre o f mass throughout the gait cycle. These
forces cause the angular (limbs) and linear (displacement of centre of mass) motion of
the runner resulting in step length and step frequency, which is measured as mean
horizontal velocity o f the centre o f mass.
Ground reaction force reflects the acceleration o f the centre o f mass. During the first
half of stance, the gastrocnemius and quadriceps muscle groups {muscle force) are being
eccentrically stretched by gravitational force (absorption, figure 3.3.4). After maximum
vertical ground reaction force following muscle-tendon lengthening, the support leg
extends while the muscles shorten, known as the stretch-shortening cycle (Cavagna et
al., 1964). After mid-stance as the muscles shorten in elastic recoil, the hamstrings
{muscle force) begin to pull the support foot from the ground and continue their activity
into swing. Elastic recoil reduces work against gravity.
72
The mass is the body of the runner. Acceleration (horizontal and vertical) of the runner’s
mass is downwards by gravitational force during impact to mid-stance and during flight
(vertically downwards) but from mid-stance until terminal stance it is via a gravitational
torque (causing a horizontal acceleration). Centre o f mass position as it passes anterior to
the support limb via momentum from impact to maximum vertical ground reaction force
allows gravitational force to then create an external gravitational torque around the axis
of the support foot (ball of the foot) until terminal stance (figure 3.3.6).
Gravity (gravitational torque) is the motive external force from mid-stance until terminal
stance (see figure 3.3.6) because muscle activity is absent, but during flight, gravity
causes no horizontal acceleration.
f
i
T orque
CM
b)
Id
Figure 3 .3 .6 T h e r u n n in g P ose® ( R o m a n o v , 2 0 0 2 ) a ) im p a c t to m a x im a l v e r t ic a l g r o u n d r e a c tio n fo r c e :
s h o u ld e r - h ip - a n k le v e r tic a l a lig n m e n t, w h ic h p o te n tia lly a llo w s g r a v ity to m a x im is e th e s t r e tc h -s h o r te n in g
r e fle x a n d e n a b le s th e w h o le b o d y to w o r k a s a u n it, b ) A g r a v ita tio n a l to r q u e a c c e le r a t e s t h e b o d y w h i l e
th e h a m s tr in g m u s c le g r o u p p u lls t h e su p p o r t f o o t fr o m th e g r o u n d .
The research variables that were readily quantifiable, identified as important from the
literature and that adequately describe the Gravitational hierarchical model are presented:
1. C e n tr e o f m a s s d is t a n c e b e h in d t o e m a r k e r (h o r iz o n t a l d if f e r e n c e at 2 5 m s ) ( c m )
2. P o s te r io r g r o u n d r e a c tio n fo r c e (b r a k in g ) ( B W )
3. M a x im u m v e r tic a l g r o u n d r e a c tio n f o r c e ( B W )
4. A n te r io r g r o u n d r e a c tio n fo r c e ( p r o p u ls io n ) ( B W )
5. S te p le n g th fo r t h e r ig h t f o o t (c m )
6. S te p f r e q u e n c y ( H z )
7. C e n tr e o f m a s s h o r iz o n t a l v e lo c it y la r g e s t d e c r e a s e a fte r im p a c t to m a x im u m v e r t ic a l g r o u n d
r e a c tio n fo r c e ( c m .s - 1)
8. C e n tr e o f m a s s h o r iz o n t a l v e lo c it y la r g e st in c r e a s e fr o m m a x im u m v e r t ic a l g r o u n d r e a c tio n f o r c e
to te r m in a l s ta n c e ( c m .s ’1)
9. C e n tr e o f m a s s h o r iz o n t a l d is p la c e m e n t d u r in g s ta n c e
10. S ta n c e t im e ( m s )
11. M e a n k n e e f le x io n a n g u la r v e lo c it y fo r sta n c e ( r a d .s '1)
12. M e a n k n e e e x t e n s io n a n g u la r v e lo c it y fo r sta n c e (r a d .s '1)
13. M e a n k n e e f le x io n a n g u la r v e lo c ity 7 fr o m te r m in a l s ta n c e to m a x im u m s w in g a n g le ( r a d .s '1)
14. V e r tic a l o s c illa t io n o f t h e c e n tr e m a s s ( c m ).
15. T im e ( s ) ( t im e tr ia l)
73
Support for the Gravitational hierarchical model of running
Clearly, if gravity, ground reaction force, muscle activity and muscle elasticity acted as a
precisely coordinated system, efficient running would ensue. The main requirement for
the interaction of forces in running is an efficient translation of body weight, regardless
of the speed or distance of the event. Bernstein (1947) defined efficiency of movement
as a ratio between external forces, such as gravity, which does not require or extensively
utilise adenosine triphosphate (muscle contraction), as opposed to those that do, such as,
ground reaction force. Efficient movement, he concluded, emphasised external forces
that do not heavily engage muscles. Applying Bernstein’s concept to running,
hierarchical relationships of these forces when structured reveal their specific functions
during the running step. Gravity is constant throughout the gait cycle, whereas the other
forces are restricted to a certain point in time and space. This section seeks to establish
gravity as the motive and hierarchically leading force, over the other forces. Following
this order, all forces will act efficiently in time and space.
Gravity was historically regarded as a key motive force: Leonardo Da Vinci was the first
to recognise gravity as a propulsive force, when he stated that “motion is created by the
destruction o f balance, that is, o f equality o f weight fo r nothing can move by itself which
does not leave its state o f balance and that thing moves most rapidly which is furthest
from its balance” (Keele, 1983: p. 173).
Graham-Brown (1912, p. 786) 400 years later wrote in support of gravity as a motive
force: “It seems to me that the act ofprogression itself- whether it be by flight through
the air or by such movements as running over the surface o f the ground-consists
essentially in a movement in which the centre o f gravity o f the body is allowed to fa ll
forwards and downwards under the action o f gravity, and in which the momentum thus
gained is used in driving the centre o f gravity again upwards and forwards; so that, from
one point in the cycle to the corresponding point in the next, no work is done
(theoretically), but the mass o f the individual is, in effect, moved horizontally through the
environment. ”
Fenn (1930) following extensive work on running found his fastest runner in comparison
to a slower runner had his centre of mass further forward during stance (an increased
body lean). He went on further to say: “that a runner does not leap into the air through
rises in their centre o f mass because they are maximal after take-off ’ (Fenn, 1930).
Instead, the centre of mass rises while the foot is still on the floor and can continue
briefly after terminal stance. The runner then falls while they are in the air for about 3 cm
(Fenn, 1930). However, this research was not pursued and it is still unclear today how
gravity relates hierarchically to the other forces in running. For example, no one has
combined gravity’s work on the runner with the action of the lower-limb until Romanov
(2002) identified as the body falls via a gravitational torque, the support foot needs to be
pulled from the ground rather than pushed, in order to catch up with the body.
Gravitational torque from mid-terminal stance

In an attempt to model gravity’s role in running, it is necessary to find the point in the
gait cycle where gravity could be propulsive, or rather, produces a gravitational torque.
In fact, there is only one phase in the gait cycle: from maximal vertical ground reaction
force (mid-stance) until terminal stance (Romanov, 2002). At this point (at peak vertical
ground reaction force) in constant speed running, the centre of mass begins to move over
the support foot (ball of the foot). The body (centre of mass) rotates forward and upward,
while the foot acts as a pivot. Rotational movement of the body created by gravity
appears as a gravitational torque, so the body could fall forwards (figure 3.3.7). The
gravitational torque may therefore explain the ‘extensor paradox.’ If gravity is the
motive force, the leg extensors do not need to thrust the body forwards and upwards,
hence their silence as knee extension begins.
Centre o f ^ ^ i
mass Gravitational Torque
r = position vector
from the support
foot to the centre
of mass -------
_Ld
Figure 3.3.7 A to r q u e c r e a te d a r o u n d an a x is (su p p o r t lim b ; b a ll o f f o o t ) fr o m g r a v ita tio n a l fo r c e . C M =

c e n tr e o f m a s s ; F g = g r a v ita tio n a l f o r c e , I d = m o m e n t a rm a n d th e a x is o f r o ta tio n is th e b a ll o f th e fo o t .
75
By taking the position vector and gravity vector from figure 3.3.7, it is possible to
represent the gravitational torque from its vector components (figure 3.3.8). The vector
product of any two vectors A x B is defined as a third vector C, the magnitude of which
is AB sin 0, where the angle 0 is the angle between A and B. That is, if C is given by
C = |A | X I BI (3.3.3)
Then its magnitude is
| C | = \ A \ | B | sind (3.3.4)
Following the right hand rule the gravitational torque vector is in the clockwise direction
about the axis perpendicularly out of the paper. Therefore the gravitational torque is:
t = Fgr sinQ 0 < 0 < 180° (3.3.5)
However, the gravitational force is comprised of two components. The Fr component

creates no torque on the centre of mass because it acts through the axis of rotation,
however the propulsive component (Fp) creates the torque about the support foot.
Fr + Fp = £F (3.3.6)
76
- G r a v ita tio n a l to r q u e
a) b)
Figure 3.3.8 a ) T h e v e c t o r c r o s s p r o d u c t o f th e c e n tr e o f m a s s ’s p o s it io n v e c t o r a n d th e f o r c e o f g r a v ity
( F g ) c r e a te a g r a v ita tio n a l to r q u e a r o u n d th e su p p o r t lim b in th e c lo c k w i s e d ir e c tio n , r = p o s it io n v e c t o r o f
th e c e n tr e o f m a s s fr o m th e su p p o r t lim b , b ) F g = g r a v it y , a n d its c o m p o n e n t s d u r in g s t a n c e a r e F r. w h ic h
is th e fo r c e a p p lie d to th e g r o u n d a n d c a u s e s n o to r q u e w h il e F p is th e p r o p u ls iv e f o r c e th a t c a u s e s th e
g r a v ita tio n a l to r q u e .
Ground reaction force is absent until two bodies contact one another and gravity has no
components (Fr and Fp) until the runner touches the ground. As the runner falls forwards
after mid-stance, ground reaction force is decreasing reflecting the runner’s rising centre
of mass. There is also another torque that occurs via the ground reaction force vector and
its moment arm about the centre of mass (figure 3.3.9), which reflects the acceleration of
the centre of mass, as it moves in a non inertial frame during stance because of its
acceleration and deceleration. A better option for an inertial frame is the gravitational
torque around the support foot making it more suitable for analysis. More importantly,
the point of application for the ground reaction force vector is the support foot and
therefore the ground reaction force is in fact reflecting Fr. Fr is equal to the ground
reaction force but in the opposite direction. Further, the ground reaction force is not
propulsive because it is a reactive force (reacting to Fr). There is also no ground reaction
from the leg and hip muscle extensors because they are not active during the propulsive
portion of stance. The gravitational torque is therefore the motive torque and not ground
reaction force, which is reflecting gravity’s work. The resultant (net) force (figure 3.3.9
b) is the resultant of the ground reaction force and gravity vectors. However, the
resultant vector of gravity and ground reaction force is the same as Fp (figure 3.3.10) the
propulsive component of gravity.
77
Ground
reaction force
ce ^
Centre of ^
vector / Resultant
mass y i / 1
'N ^êctor (Fp)
Ground
Moment arm reaction force Gravity
vector vector
a) by
Figure 3.3.9 a) Ground reaction force torque about the centre of mass b) The ground reaction force and the
gravity vector taken at the same point of stance as the vector product in figure 3.3.8.
Figure 3.3.10 uses figure 3.3.9 b but includes the components of gravity. Fp creates the
gravitational torque representing the resultant of the two external forces (gravity and
ground reaction force) acting on the runner and is the net force causing the gravitational
torque that accelerates the runner in the direction of the net force. The net force (Fp) is
perpendicular to the position vector representing the radius for the centre o f mass around
the support foot. Fp causes a tangential acceleration, which changes the speed of the
centre of mass. Fr causes a radial acceleration, which changes the direction o f the centre
of mass. This motion is similar to an inverted pendulum where the centre o f mass swings
around the support foot. The torque (tangential component of gravity) for a pendulum is
the same (if I = r) equation as the gravitational torque for running giving further support
the gravitational torque (inverted pendulum) is the motive torque in running:
x = -mgl sin 6 (3.3.7)

Where -m g is the weight of the pendulum, 1 is the length of the cable and sine theta is
the angle between the cable and gravity.
Gro J
reac
vecl
Resultant
vector or (Fp)
Gravity
\T vector
Figure 3.3.10 Gravity and ground reaction force vectors with gravity’s components included, showing the
net force is Fp. Notice Fr is equal and opposite to the ground reaction force vector, leaving Fp as the net
force or the resultant of gravity and ground reaction force.
The body is also moving upwards from mid-stance until terminal stance so the runner is
working against gravity, but this does not negate the gravitational torque seen from
figure 3.3.11.
R is in g c e n tr e o f m a s s
o w in g to th e g e o m e t r ic a l
s h a p e o f th e o b je c t
Gravitational
rtorque
b) Id
Figure 3.3.11 T w o id e n tic a l r e c t a n g le s a re g iv e n w it h th e ir c e n tr e o f m a s s e s m a r k e d a s O F g = g r a v ity ,
a ) T h e u p r ig h t r e c t a n g le r e p r e s e n ts t h e r u n n e r a t m id - s t a n c e w ith t h e n c e n tr e o f m a s s a b o v e th e s u p p o r tin g
lim b (b a ll o f th e f o o t ) , b ) T h e r e c t a n g le (r u n n e r ) tip s fo r w a r d s v ia m o m e n t u m . A g r a v ita tio n a l to r q u e
a r o u n d th e r u n n e r 's c e n tr e o f m a s s s u b s e q u e n tly m o v e s th e m h o r iz o n t a lly fo r w a r d s . N o t e h o w e v e r , th e
c e n tr e o f m a s s r is e s o w in g to th e r e c t a n g le s g e o m e tr y a n d th e r u n n e r r is e s v ia e la s t ic r e c o il a n d le g
e x t e n s io n fr o m t h e r is in g b o d y .
The centre of mass rises (does work against gravity), because of the lower-limb
geometry (support/stance leg extends), whilst the gravitational torque horizontally
displaces the runner. The stance leg is extending (passively) at this time owing to
geometry of the body and elastic recoil, which raises the centre of mass. Obviously, less
leg extension will result in lower work against gravity. Cavanagh, et al. (1977) found
general support for improved running economy with lower vertical oscillation (less work
against gravity). The momentum of the runner in the vertical direction would slow at this
time owing to the work against gravity, except for the rebound effect of the muscle-
elastic system. The leg extensors are maximally active (eccentrically) at mid-stance as
the body absorbs landing and the centre of mass is at its lowest vertical position (Maim,
1980; Wank et al., 1998). The angle of the centre of mass at or just after this time is just
past the vertical (in the clockwise direction) where the leg extensor muscles become
inactive. Any muscle force at this time would be directed more vertically than
horizontally (Farley and Ferris, 1998). The elastic recoil of the musculotendinous
junctions therefore, contribute to the vertical oscillation of the centre of mass thus
79
reducing the work against gravity (Alexander, 1989). However, elastic recoil cannot
create a typical one-metre step length. Rather, the constant force of gravity after
maximum vertical ground reaction until terminal stance creates a gravitational torque,
which produces step length from horizontally accelerating the runner’s body.
Returning to the Froude number, comprised of leg length, running speed and gravity.
Gravity reflects the centre of mass’s height because it is the point about which gravity
acts. The height (centre of mass position) of falling is also dependent upon the length of
the leg. Alexander and Jayes (1983) proved that the longer the leg the longer the step. A
longer leg increases the height of release, which will increase step length. However,
running speed comprises of both step length and step frequency. When step length is
maximal the only way to increase speed is to increase step frequency. The increase in
step frequency at faster speeds is simply reflecting the ability to fall has been maximised
(angle of inclination is maximal), therefore, to increase running speed, higher step
frequency is required (more opportunities to fall), which is self-evident in sprinting. Step
length reaches a maximum value at a running speed of around 7.0 m-s'1 (Luhtanen and
Komi, 1978). To compensate, step frequency increases at faster rumiing speeds—than
7.0 m-s'1—and up to maximum speeds (Luhtanen and Komi, 1978). If step length
reflects these three components, it is possible to state, running speed increases via a
gravitational torque up to a point where the maximal angle of inclination (from the centre
of mass to the support foot) is reached, thus producing a maximal step length.
In support that gravity is creating forward progression, Chang et al. (2000) using three
conditions: increased gravity and inertia, increased inertia and reduced gravity, took
eight participants and ran them at 3 m-s'1 for all conditions. By increasing gravity and
inertia, the magnitude of the horizontal ground reaction force impulse increased. By
increasing inertia only, the horizontal impulse did not increase. As gravity was reduced,
the magnitude of the horizontal impulse reduced proportionally. This suggests gravity is
affecting the horizontal impulse. Perhaps this explains why the horizontal propulsive
ground reaction force does not coincide with maximal muscle activity, but in fact, may
reflect the gravitational torque. The vertical impulse had similar results to the horizontal
impulse, indicating gravity strongly influences the vertical ground reaction force. Step
length increased in reduced gravity and decreased in increased gravity and inertia but
remained stable with increased inertia. This clearly suggests gravity is affecting step
80
length and by implication, step length is created via a gravitational torque. The resultant
ground reaction force peak angles of braking, propulsive and peak force were constant
for all conditions. As running speed was constant for all conditions this would suggest
the angle of inclination of the body would be constant, which is exactly what their results
showed. A constant body angle would not change the gravitational torque (owing to the
same moment arm) and hence changes in the vertical and horizontal impulses would be
proportional to changes in gravity for the same leg-length.
Figure 3.3.12 illustrates the length of the force vectors are proportional to the vertical
and horizontal components of the ground reaction force. Regardless o f which technique
of running—heel-toe and forefoot—both show maximum ground reaction force occurs
on the ball of the foot when the body is in its lowest position reflecting maximum weight
illustrating gravity’s work (figure 3.3.13). It is important to also note that the body
cannot move forward until the centre of mass passes the ball of the foot (pivotal point of
support) which only begins after maximal vertical ground reaction force in constant
speed running.
81
Figure 3.3.12 Vector representation of the ground reaction force in running for two participants (a) A heel-
toe runner, (b) A forefoot runner. The view is of the right shoe from the lateral aspect with the shoe tilted
30° upward (adapted from Cavanagh and Lafortune, 1980).
Leg extension during the propulsive of stance

Figure 3.3.13 illustrates the centre of mass reaches its lowest position and velocity at
maximum vertical ground reaction force in running and hence its lowest gravitational
potential and kinetic energy (McMahon et al., 1987; Farley and Ferris, 1998).
Combining figure 3.3.12 with figure 3.3.13, it is possible to see that the centre o f mass is
lowest at maximal vertical ground reaction, and this point, is on the ball o f the foot.
Further, maximal vertical velocity occurs 0.1 ms (40% of stance time later) after
maximal vertical ground reaction force. In fact, vertical ground reaction force is
decaying too below body weight at maximal vertical velocity, thus it must be reflecting
the vertical elastic re-bound affect, which subsequently causes the change in vertical
velocity and not ground reaction force.
82
3000
2000
1000
BW
150
100
-25
-50
0.00 0.05 0.10 0.15 0.20 0.25
Time (ms)
Figure 3.3.13 Vertical ground reaction force, during the stance phase of running (a) vertical velocity of the
centre of mass (b) and the vertical position of the centre of mass (c). The vertical dotted line illustrates the
timing of maximal vertical ground reaction force with zero velocity and the lowest point of the centre of
mass (adapted from Enoka, 2002; p. 192).
83
In support of the muscle system not pushing the runner forwards, the flic-flac (figure
3.3.14) illustrates there is a distinct thrust (active leg extensors) just before leaving the
ground reflected by sudden increased vertical and horizontal ground reaction force
traces. This trace is not evident in running, because the leg extensors are not active. In
fact, in rumiing, there is a rapid decay in the vertical ground reaction force reflecting the
body leaving the ground coupled with silent extensor muscle activity (‘extensor
paradox’; McClay et al., 1990).
Tim#!*)
Figure 3.3.14 A g y m n a s t p e r fo r m e d a f l ic - f l a c (s tic k f ig u r e ) s h o w in g v e r tic a l ( F z ) a n d p o s te r io r -a n te r io r

( F y ) g r o u n d r e a c tio n fo r c e . N o t e th e s p ik e in g r o u n d r e a c tio n fo r c e at p u s h - o f f w ith th e f e e t ( Y e a d o n a n d
C h a llis , 1 9 9 4 ) .
Fenn (1930) identified that the body pushes forwards from the hip at foot contact
because hip velocity exceeded that of the neck. In fact, the hips begin extension before
the knee and ankle during stance (Mann et al., 1986; Montgomery et al., 1994), which
supports Fenn’s observation. However, hip extension that produces forward movement,
84
which leads the knee and ankle joints, is not universally accepted (Winter 1983).
Montgomery et al. (1994) suggest the ankle musculature stabilizes the ankle while the
trunk moves forward rather than these muscles producing an active push-off (Reber et
al., 1993). Whereas, Winter (1983) suggested the calf musculature pushed off at terminal
stance. However, Cavanagh et al. (1977) found elite rumiers did not plantar flex the
ankle at terminal stance compared to less able runners, suggesting it is not effective. In
figures 3.3.15 and 3.3.16, at the third position for each runner, which is just past
maximum ground reaction force the hip is just about to begin extension. While positions,
4 and 5 show both runners extending their hips as vertical ground reaction force is
decreasing, noting leg extensor muscle activity ceases just after leg extension begins
(Montgomery et al., 1994; Wank et al., 1998). Hunter et al. (2005) also dismissed leg
extension at terminal stance as a cause of increased propulsion in sprinters, but found the
mean of hip extension velocity was associated with greater propulsion. Therefore, owing
to the body’s horizontal velocity, the body moves forward of the foot (having almost
zero horizontal velocity) and leg extension will occur primarily because of the body’s
inertia (reflected by hip motion). The body continues past the support point (ball of the
foot), where the ankle musculature stabilises the ankle, and then the body accelerates via
a gravitational torque. Leg extension is also aided via the stretch-shortening cycle.
Changes in the length of the muscles belly are minimal, because most of the change in
length comes from the stretch and recoil of the respective tendons (McMahon, 1990).
Consequently, muscle tendons lengthen during the eccentric phase (impact to maximum
vertical ground reaction force) and recoil elastic energy during the propulsive phase of
stance. The muscles still exert tension but elastic energy allows the muscles to shorten
less, thus reducing muscle work (Ker et al., 1987). Short stance time and correct body
geometry (landing with the centre of mass over the ball of the foot to increase the pre
stretch via a more vertical weight vector) all potentially increases the stretch-shortening
cycle. The release of elastic energy aids leg action through tension in the Achilles tendon
at terminal stance (Kyrolainen et al., 2003) rather than an active push-off. Shorter stance
time will result from a more vertically aligned body at impact, because the body travels a
shorter horizontal distance during stance. A shorter stance time also enhances the elastic
mechanisms (Zatsiorsky, 1995; Paavolainen et al., 1999a).
85
1 2 3 4 5
2000 T
1000
0.1 SEC
200
Figure 3.3.15 Force-time recording of a forefoot runner (mass 80 kg, velocity 9.2 m-s'1) (Payne, 1983; p.
749).
86
1 2 3 4 5
\M
2 0 0 0 -T-
n s e c . ---------
200
200-
Figure 3.3.16 F o r c e - t im e r e c o r d in g o f a h e e l- t o e r u n n e r (m a s s 6 3 k g , v e lo c it y 9 .5 m - s '1) ( P a y n e , 1 9 8 3 ;

p .7 4 8 ) .
Further evidence for the support leg being passively extended is seen in figure 3.3.17.
Both runners have similar vertical oscillations until position 3, possibly illustrating
comparable knee flexion angles. The vertical displacement of the centre of mass rises
from mid-stance (position 3) until terminal stance (or just after) in both runners.
However, the good runner does not extend his leg as much as the poor runner—position
6—possibly leading to lower vertical oscillation from position 4 to 7. Note the term
‘poor’ is relative in relation to his 5-km time making him still a very competitive athlete.
Miura et al. (1973) found that the good runner’s centre of mass deviation from the
vertical axis was greater on average at 34.7° compared to 31.8° for the poorer runner. A
larger angle from the vertical axis increases the moment arm for the gravitational torque,
87
which potentially increases forward progression (increases running speed) while
passively extending the leg. The lower leg extension in the faster runner may be related
to his increased angle of deviation.
Poor runner
Good runner
.’I
«J *
A! «w*
f vs "j
i W * 11
4 2
Figure 3.3.17 V e r tic a l d is p la c e m e n t o f th e c e n tr e o f m a s s fo r a g o o d r u n n e r ( 1 4 m in 5 2 s , 5 k m ) a n d a

p o o r r u n n e r ( 1 6 m in 2 9 s . 5 k m ) (M iu r a et al., 1 9 7 3 ).
Milliron and Cavanagh (1990) also support a passive leg extension. Using four
participants on a treadmill at a negative grade of -20% , level and a positive grade of
20%, they found that leg extension decreased from level to uphill (20%) and increased
for downhill running (-20%). These data suggest that downhill running causes greater
extension (32.1°) owing to a faster body speed passively extending the leg further than
uphill (13.0°) or level running (15.1°). Paradisis and Cooke (2001) also found similar
results. In addition, the most variability in phasic activity of all the knee muscles is the
hamstrings (Heise et al., 1996). Hamstring variance might better explain runners who
flex their legs (via hamstring activity) earlier and faster at terminal stance than poorer
runners do. Earlier and faster flexion at terminal stance would possibly reduce leg
extension, maintain momentum—through reduced inertia of a less extended leg— and
hence lower vertical oscillation and work done against gravity. Potentially a runner then
leaves the ground as the body falls forwards. If the body is accelerating from the
gravitational torque and the next step is beginning, there is no need to push-off with the
stance leg. In fact, Romanov (2002) is the first to mention pulling the support foot from
the ground. If the foot is pulled from the ground, it is also moving in the same direction
(upwards) as the elastic recoil of the lower-limb muscles. Elliot and Blanksby (1979)
reported maximal biceps femoris activity at terminal stance suggesting rapid knee
flexion supporting a potential pulling of the foot from the ground. The concept of pulling
the foot from the ground among endurance runners is not common. However, in most
sprinters, the knee does not fully extend during stance and rapidly flexes at terminal
stance, suggesting the foot is indeed pulled from the ground (Mann and Hagy, 1980;
Mann et al., 1986). If the hamstring muscle group flexes the knee at the same time as the
muscles experience elastic recoil a coordinated lower-limb/body motion should ensue. If
this action is coordinated with the body falling forwards, at this point, the body is rising,
which un-weights the foot allowing it to be pulled from the ground. The foot will then
essentially be catching up to the body as the runner enters flight.
Flight Phase
During flight the runner falls to the ground under gravity’s work (Fenn, 1930) and
providing the opposite leg— swing leg— is not driven forwards (via hip flexion) gravity
should allow the foot to drop under the centre of mass for the next step cycle to begin
again. If the swing leg is driven forwards into the flight phase the swing foot will land in
front of the centre of mass at impact reducing horizontal velocity (Bates et al., 1979;
Hinrichs et al., 1987). A reasonable assumption is runners are thinking about driving
their swing leg forwards at terminal stance, rather than pulling the support foot from the
ground. This difference perhaps divides Pose® running from heel-toe rumiing (Romanov,
2002). If the runner is not thinking about pulling the foot from the ground—using the
hamstring muscles— as the leg extends, their centre of mass continues to rise via leg
extension completing more work against gravity, thus reducing the gravitational torque.
Reducing the gravitational torque reduces the take-off velocity, which subsequently
reduces running speed.
In summary of the Gravitational hierarchical model that describes the Pose® technique of
running, two key concepts emerge: falling under a gravitational torque and pulling the
support foot (knee flexion) from the ground via the hamstring muscle group. The
89
Gravitational hierarchical model has attempted to reflect these concepts while ordering
the timing of the forces involved in running. Support presented for gravity being the
motive force was the runner could only move forwards once the centre o f mass passes
the support point (ball of the foot) at mid-stance. Inertia brings the runner’s centre of
mass after impact to this point (mid-stance), where ground reaction force is maximal,
while vertical displacement and horizontal velocity of the centre of mass are minimal. In
a flic-flac, there is a clear push-off the ground, but in running the ground reaction force
decays from maximal vertical ground reaction force until terminal stance. The decay rate
increases with running speed, suggesting increases in the acceleration of the centre of
mass (faster angle of deviation) as it leaves the ground. A gravitational torque creates
acceleration of the runner from mid-stance, potentially explaining the lack of leg and hip
extensor muscle activity. In support, maximum propulsive horizontal ground reaction
force occurs after extensor muscle activity ceases. Pulling the foot from the ground as
the body is falling, reduces vertical oscillation by potentially decreasing leg extension
(Wank et al., 1998), which allows the foot to catch-up with the body more quickly as a
result of reduced lower-limb inertia. High Achilles tendon forces at terminal stance,
suggest the elastic recoil plays a significant role in reducing work against gravity from
mid-terminal stance. During swing, if the ipsilateral leg drops under gravity’s influence
during flight, and is not driven forwards via the hip flexors, the foot will land under the
centre of mass for the cycle to begin again.
3 .4 SUMMARY
There are three known forces involved in running (gravity, ground reaction force, muscle
force) and potential strain energy. Several deficiencies identified in the current running
model include: viewing the constant force of gravity as only active during flight and not
relating gravity to forces exerted and the time forces act despite gravity being a constant
force. The Gravitational model stated the forces involved in running could hierarchically
function when gravity orders their timing and work. A gravitational torque was shown to
be the main motive force in running as the body falls forwards. Ground reaction force is
not a motive force but operates according to Newton’s third law, which simply reflects
the forces it receives. The ground can only propel a runner forward via muscle activity.
Leg and hip extensor muscles have consistently proven to be silent during the support
90
leg’s extension, in contrast to Hay and Reid (1988; pp. 264 and 287-288), who suggest
leg extension via ankle, knee and hip musculature is the motive force in running. High
Achilles tendon forces at terminal stance suggest the elastic recoil is creating a re-bound
effect in the vertical direction thus reducing work against gravity. Gravity is the only
external force that can displace the centre of mass from maximal vertical ground reaction
force until terminal stance if muscle extensor activity is silent. The Gravitational
hierarchical model therefore, identifies gravity’s important role in running. Forward
movement (falling via a gravitational torque) begins once the centre of mass has passed
over the ball of the foot. The flic-flac shows a clear spike in the ground reaction force
trace reflecting a push-off, which is completely absent in runners. In fact, the ground
reaction force decay rate increases as running speed increases. The decay rate is
reflecting the body leaving the ground and not the support limb’s muscle activity
pushing off the ground because the lower-limb’s muscles are silent. The Froude number
supports gravity’s work in creating running speed, because step length differences from
short or long legged runners are proportional to leg length. Muscle force is not a
component of the Froude number because step length occurs via a gravitational torque
not muscle force. Further, the hip’s extension velocity was significantly associated with
the support leg’s passive extension. This again suggests the body is moving forward of
the support limb (via a gravitational torque) and not being pushed forwards via the ankle
and knee joints. The hip passively extends because the body moves past the stationary
dR)
foot extending the lower-limb. In Pose running the hamstring muscles pull the support
foot rapidly from the ground when the foot un-weights with the aid of the stretch-
shortening cycle optimised from a short contact time. By pulling the support foot and not
driving the swing leg forward of the body via the hip flexors, gravity can drop the swing
leg under the body at impact enabling the cycle to begin again. The swing leg’s centre of
mass must be vertically below the body’s centre of mass during flight for this to occur.
Driving the swing leg forwards causes the foot to land ahead of the body slowing
horizontal velocity of the centre of mass owing to increasing the distance of the centre of
mass from the support foot. The Gravitational model, therefore, hierarchically explains
the forces involved in running. Gravity causes the muscle system to absorb body weight
on landing (maximum ground reaction force), which then produces elastic strain energy
in the musculotendinous junctions. As the centre of mass passes over the support limb (at
maximum vertical ground reaction force), a gravitational torque is produced as extensor
muscle activity ceases. The runner falls forwards while ground reaction force decreases
91
(the body is leaving the ground) and vertical work against gravity is reduced via elastic
mechanisms. At terminal stance, as the foot is un-weighted, it is rapidly pulled from the
ground (hamstring muscle activity) in order to reduce lower-limb inertia and to catch up
with the body during flight. Finally, the runner returns to the ground via gravity during
flight. The ability to fall via a gravitational torque and to pull the foot from the ground
effectively, potentially divides the two techniques: Pose® and heel-toe running. The
Gravitational hierarchical model, therefore, reflects these differences between the two
running techniques and potentially offers a quantifiable method to differentiate between
the two.
92
CHAPTER 4
EXPERIMENTAL PROCEDURES
4.1 INTRODUCTION
The research aims are presented with the hypotheses for the main study. A research
design for each of the two studies is provided covering sample size, equipment, research
variables and treatment of data.
4 .2 STATEMENT OF THE PROBLEM
The primary aim of this thesis is to compare the Pose® method technique (Romanov,
2002; Arendse et al., 2004; Dallam et al., 2005) with the heel-toe rumiing technique.
Development of a new Gravitational hierarchical model based on the Pose® method
using the laws of physics will enable the Pose® technique’s systematic evaluation. The
Gravitational hierarchical model and the current hierarchical model (Hay and Reid,
1988; p. 282) are reviewed highlighting the different biomechanical theories of running.
There are deficiencies with Hay and Reid’s model, for example, viewing the constant
force of gravity as only active during flight and not relating gravity to forces exerted and
the time forces act. Therefore, Hay and Reid’s model does not systemise the forces
involved in running but acts as general descriptor, whereas, the Gravitational hierarchical
model will aim to structure these forces. A comparison of the Pose® and heel-toe running
techniques will follow using two experienced male Pose® and heel-toe runners utilising
the primary and secondary research variables (study 1; chapter 5) derived from the
Gravitational hierarchical model. Upon establishment of the Pose® method technique, a
group of male heel-toe recreational runners (study 2; chapter 6) will participate in a
repeated measures design. An economy and time trial run will give the participants pre
test baseline physiological measures while the primary research variables will provide a
biomechanical profile. The post-test will determine whether there is an improved
93
performance on the economy and 2400 time trial run resulting from the potential
biomechanical changes in the Pose® group from a 7-hour Pose® training intervention. A
2 x 2 mixed factorial ANOVA will assess the main effects of group (control vs.
of honestly significant differences will be used to assess individual cell differences post
hoc. A prospective three-month injury study on all the participants post intervention will
seek to determine injury incidence between the two running techniques.
4 .3 THE HYPOTHESES
The thesis will reject Ho and accept Hi at P < 0.05. Each research and null hypothesis for
study 2 are given including a brief explanation. Study 1 was a descriptive study.
4.3.1 Study 2 (Pose® method training study; Chapter 6) hypotheses
The treatment group were heel-toe runners who had 7-hours o f Pose® instruction.
Fourteen null and research hypotheses derived from the Gravitational hierarchical
model’s primary research variables are given:
1. Reduced stance time correlates with elite runners and improved economy (Williams
and Cavanagh, 1987; Nicol et al., 1991).
• Hi: The treatment group will show a decrease stance time.
• Hq: The treatment group will show no change in stance time.
2. There will be a decrease in vertical ground reaction force betweenthe two groups
(Arendse et al., 2004).
• Hi: The treatment group will show a decrease in vertical ground reaction forces.
• Hq: The treatment group will have no change in vertical ground reaction forces.
3. There will be no difference in posterior and anterior groundreactionforce between

the two groups (Chang et al., 2000).
• Hi: The treatment group will show no change in posterior and anterior ground
reaction forces.
• Hq: The treatment group will increase in posterior and anterior ground reaction
forces.
4. A decrease in horizontal velocity of the centre of mass at impact to maximal vertical

ground reaction force has generally been found in ‘poorer’ runners (Fenn, 1930;
Slocum and Bowerman, 1961; Deshon and Nelson, 1964; Enomoto et al., 1999).
• Hi: At impact, there will be less decrease in the horizontal velocity of the body’s
centre of mass for the treatment group.
• H0: At impact, there will be no change in the horizontal velocity of the body’s centre
of mass for the treatment group.
5. An increase in horizontal velocity of the centre of mass from maximal vertical

ground reaction force until terminal stance for the Pose® runners will be used to
support gravity’s role in forward progression (Keller et al., 1996).
• Hi: From maximal vertical ground reaction force until terminal stance, the treatment
group will increase horizontal velocity of the centre of mass.
• H0: From maximal vertical ground reaction force until terminal stance, the treatment
group will have no change in horizontal velocity of the centre of mass.
6. Vertical alignment of the body at impact is associated with faster runners (Deshon
and Nelson, 1964; Mann and Herman, 1985).
• Hi: The treatment group will show increased vertical alignment of the body at
impact.
• Ho: The treatment group will show no change in vertical alignment at impact.
7. Less horizontal displacement of the centre of mass during stance will reduce stance
time (Ferris et al., 1999).
• Hi: The treatment group will show a decrease in horizontal displacement of the
centre of mass during stance.
• FL: The treatment group will show no change in horizontal displacement of the
centre of mass during stance.
8. Reduced vertical oscillation of the centre of mass has been associated with ‘good’
runners (Miura et al., 1973; Williams and Cavanagh, 1986).
95
• Hi: The treatment group will show a decrease in vertical oscillation of the centre of
mass.
• H0: The treatment group will show no change in vertical oscillation of the centre of
mass.
9. Mean knee angular velocity for flexion and extension during stance were
significantly correlated with stride length, stride frequency, vertical loading rate and
stance time (De Wit et al., 2000). A decrease in knee angular velocity for flexion and
extension may reflect lower muscle activity.
• Hi: The treatment group will show a decrease in mean angular knee flexion and
extension velocity during stance.
• H0: The treatment group will show no change in angular knee flexion and extension
velocity during stance.
10. Increased mean knee flexion angular velocity from terminal stance until maximum
swing shows a faster leg recovery, which correlates with faster running (Heise et al.,
1996).
• Hi: The treatment group will show an increase in mean knee flexion angular
velocity terminal stance until maximum swing.
• H0: The treatment group will show no change in mean knee flexion angular velocity
terminal stance until maximum swing.
11. Increased step length can cause braking of the runner’s horizontal velocity and
potentially increase injuries (Volishin and Wosk, 1982; Hunter et al., 2005).
• Hi: The treatment group will show a decrease in step length.
• Hq! The treatment group will show no change step length.
12. Increased step frequency is associated with Pose® running (Romanov, 2002; Arendse
et al., 2004).
• Hi: The treatment group will show an increase in step frequency.
• H0: The treatment group will show no change step frequency.
13. An increase in running speed for Pose® runners in the post-test 2400 m time trial will
establish the Pose® method enables faster rumiing.
96
• H i: The Pose® group will show an increase in speed for the post-test time trial.
(S)
• H0: The Pose group will show no change in speed for the post-test time trial.
(K)
14. An improved economy for Pose runners in the post-test 2400 m economy run will
establish the Pose® method enables more economical running.
• Hi: The Pose® group will show an increase (improvement) in economy for the post
test economy run.
• H0: The Pose® group will show no change in economy for the post-test economy run.
4 .4 RESEARCH DESIGN
4.4.1 Introduction
On completion of the new Gravitational hierarchical model, two studies will assess the
Pose® technique. Comparison of two experienced male Pose® and heel-toe runners will
complete study 1 using the primary and secondary research variables from the
Gravitational hierarchical model. A pre and post-test design using sixteen male
recreational heel-toe runners, where eight will be trained in Pose® running, from a 7-hour
intervention over one-week, will be compared using inferential statistics on the primary
research variables.
4.4.2 Sample size and selection criteria
Study 1: Experienced Pose® and heel-toe runner’s study

Only two suitably experienced male Pose® runners were prepared to travel to the UK for
testing. No experienced Pose® runners lived in the UK at the time of data collection.
Two experienced heel-toe runners who matched the Pose® runners in age, stature, weight
and experience were recruited for the study. Descriptive statistics compared and
evaluated the two running techniques using the primary and secondary research
variables.
97
Study 2: Pose® method training study
A population of local runners and triathletes recruited from Loughborough University
served as volunteers for study 2. The main criteria for participant selection included that:
(a) all subjects were males; (b) were of a competitive age of 18-30 years old; (c) were
sub-elite and had been running for at least two-years (10-km personal best time 37-45
min equating to 6 to 7 min per mile pace) who were distance runners (5-km-marathons).
Sub-elite runners were selected owing to their availability over elite athletes, (d) They
also had not read or heard of the Pose® method of running; (e) landed on their heel first
when running; (f) were willing to change their running technique for the testing period;
(g) had been without injury in the past 3-months and finally (h) were able to make the
study research dates.
These criteria were selected to give the study internal and external validity, which would
allow the results to be generalised to a population o f recreational male endurance
runners/triathletes. Selection of this population was based on them having the largest
number of participants relative to the total running population. The emphasis was on a
minimum and maximum age and ability, which reduced inter-participant variance in
scores. All participants were heel-toe runners, which represented eighty percent of the
endurance running population (Kerr et al., 1983). Using pilot data and data from the
literature a power test was calculated giving n = 16 for the desired sample size for 99%
confidence of rejecting the null hypothesis when it is true.
4.4.3 Equipment and design
Study 1: Experienced Pose® and heel-toe runner’s study

A force plate and a three-dimensional analysis system collected the research variables
derived from the Gravitational hierarchical model for the two running techniques.
Descriptive statistics described the two techniques.

A force plate, portable gas analyser, heart rate monitor and a three-dimensional analysis
system were used to quantify the changes between the heel-toe and Pose® groups for the
laboratory experiment and the field-tests. Both heel-toe and Pose® groups completed a
pre-test economy run and time trial of 2400 m involving a portable gas analyzer
98
(Cosmed, K4 b2) and a heart rate monitor. They also completed a laboratory pre and
post-test using a force plate and a three-dimensional analysis system for five trials run at
3.35 m-s'1. The Pose® group received 60 min of instruction in the Pose® method from a
qualified Pose® coach for 7-days. Immediately following the intervention, both groups
were administered the post-tests using the same instrumentation.
4.4.4 Identification of the research variables
Study 1: Selection of research variables

Variables derived form the Gravitational hierarchical model (primary research variables)
were used to assess differences between two experienced Pose® and heel-toe runners.
1. Stance time (ms).
2. Posterior and anterior ground reaction force (BW).
3. Maximum vertical ground reaction force (BW).
4. Centre of mass horizontal velocity from impact to maximum vertical ground
reaction force (cm-s-1).
5. Centre of mass horizontal velocity from maximum vertical ground reaction force
until terminal stance (cm-s"1).
6. Centre of mass behind the toe marker (horizontal difference at 25 ms) (cm).
7. Centre of mass horizontal displacement during stance (cm).
8. Vertical oscillation of centre mass (cm).
9. Mean knee flexion angular velocity for stance (rad-s'1).
10. Mean knee extension angular velocity for stance (rad-s"1).
11. Mean knee flexion angular velocity from terminal stance to maximum swing
angle (rad-s'1).
12. Step length for right foot (cm).
13. Step frequency (Hz).
Secondary research variables aided the analysis of the two running techniques.
I. Time to maximum braking and propulsive ground reaction force (percentage of
stance).
II. Time to maximum vertical ground reaction force (percentage of stance).
III. Centre of mass horizontal velocity from terminal stance until maximum swing
(cm-s'1).
IV. Centre of mass at terminal stance forward of the support foot (horizontal
difference) (cm).
V. Centre of mass horizontal velocity at terminal stance (cm-s"1).
VI. Horizontal velocity of the centre of mass from maximum vertical ground reaction
force until terminal stance minus horizontal velocity of the centre of mass from
impact until maximum vertical ground reaction (cm-s"1).
VII. Centre of mass and toe marker angle at terminal stance (rad).
VTII. Knee flexion from impact until mid-stance (rad).
IX. Knee extension from mid-stance until terminal stance (rad).
X. Knee flexion from terminal stance until maximum swing (rad).
99
XI. Knee angle at impact (rad).
XII. Knee angle at maximum vertical ground reaction force (rad).
XIII. Flight time (ms).
XIV. Time of maximum horizontal acceleration of the centre of mass (cm-s'2).
Study 2: Selection of research variables

The research variables derived form the Gravitational hierarchical model (primary
research variables) were used to assess differences between a group of sixteen heel-toe
runners where eight participants were trained in Pose® running technique from a 7-hour
intervention.
1. Stance time (ms).
2. Posterior and anterior ground reaction force (BW).
3. Maximum vertical ground reaction force (BW).
4. Centre of mass horizontal velocity from impact to maximum vertical ground
reaction force (cm-s-1).
5. Centre of mass horizontal velocity from maximum vertical ground reaction force
until terminal stance (cm-s'1).
6. Centre of mass behind the toe marker (horizontal difference at 25 ms) (cm).
7. Centre of mass horizontal displacement during stance (cm).
8. Vertical oscillation of centre mass (cm).
9. Mean knee flexion angular velocity for stance (rad-s'1).
10. Mean knee extension angular velocity for stance (rad-s'1).
11. Mean knee flexion angular velocity from terminal stance to maximum swing
angle (rad-s’1).
12. Step length for right foot (cm).
13. Step frequency (Hz).
14. Time for 2400 m time trial (s).
15. Oxygen consumption (ml).
Secondary research variables aided the analysis of the two running techniques.
I. Time to maximum braking and propulsive ground reaction force (percentage of
stance).
II. Time to maximum vertical ground reaction force (percentage of stance).
III. Centre of mass horizontal velocity from terminal stance until maximum swing
(cm.s'1).
IV. Centre of mass at terminal stance forward of the support foot (horizontal
difference) (cm).
V. Centre of mass and toe marker angle at terminal stance (rad).
VI. Knee flexion from impact until mid-stance (rad).
VII. Knee extension from mid-stance until terminal stance (rad).
VIII. Knee flexion from terminal stance until maximum swing (rad).
IX. Knee angle at impact (rad).
X. Knee angle at maximum vertical ground reaction force (rad).
XI. Flight time (ms)
100
4.4.5 Treatment of collected data
Study 1: Experienced Pose' and heel-toe runner’s study

Descriptive statistical procedures compared the two running techniques. Discussion
focused on differences supported by the primary and secondary research variables and
running research.

A power test for a repeated measures ANOVA statistical procedure was conducted to
determine the probability of detecting a ‘true’ effect when it exists. Using proc power in
SAS statistical software (version 9.1) sample size was calculated. Selected power was
0.8 at P < 0.05 (Cohen, 1988). Step length was selected as the variable since it
generically describes the gait cycle because it is the direct result of the kinetic and
kinematic motions of the runner. Using heel-toe runner’s pilot data for the sample mean
(mean step length 97 cm), the literature was used to determine the population mean
(mean step length 119 cm; Cavanagh and Kram, 1985) and standard deviation (14 cm;
Cavanagh and Kram, 1985) at the test running speed of 3.35 m-s'1enabling sample size to
be calculated. Proc power requires the following information in order to do the power
analysis: a) the number of levels (or groups = 2), b) the means for each level (119, 97), c)
the common group standard deviation (14), d) the alpha level (P < 0.05) and e) the power
(0.8). For n = 16 the calculated power was 0.99.
A 2 x 2 mixed factorial ANOVA assessed the main effects of group (control vs.
of honestly significant differences were used to assess individual cell differences post
hoc. Data analysis was completed using the Minitab (version 14) statistical analysis
package. An alpha level of P < 0.05 was used to establish statistical significance for the
inferential procedures.
101
Assumptions
Assumptions were satisfied for the inferential statistical analysis and are given below.
Significance is taken at P = 0.05.
1. Independence: In this study, the two groups contain separate individuals so the
data are independent.
2. Scale of measurement: The scale of measurement was ratio.
3. Normality:
Table 4.4.1 Ryan-Joiner’s (similar to Shapiro-Wilk) normality test (N = 16).
Variable Rvan-Joiner normalitv test

Stance time 0.1
Anterior-posterior GRF (brak) 0.1
Vertical GRF 0.1
Anterior-posterior GRF (prop) 0.1
Centre of mass horizontal velocity (brak) 0.1
Centre of mass horizontal velocity (prop) 0.1
Centre of mass to toe-marker at 25 ms 0.1
Centre of mass displacement during stance 0.1
Vertical oscillation 0.1
Knee flexion angular velocity 0.08
Knee extension angular velocity 0.1
Knee flexion angular velocity swing 0.1
Step length 0.1
Step frequency 0.07
Time trial 0.08
Oxveen consumntion 0.15
4. Homogeneity:
The variances between each group was examined using Levene's test for homogeneity o f
variances.
102
Table 4.4.2 L e v e n e 's test fo r h o m o g en eity o f v aria n c e ( N = 16).
V a r ia b le s Levene Statistic Significance

Stance time Based on Mean 2.25 0.16
Anterior-posterior GRF (brak) Based on Mean 2.42 0.14
Vertical GRF Based on Mean 4.29 0.60
Anterior-posterior GRF (prop) Based on Mean 0.14 0.70
Centr e o f mass horizontal velocity (brak) Based on Mean 0.03 0.09
Centre o f mass horizontal velocity (prop) Based on Mean 0.30 0.60
Centre o f mass to toe-marker at 25 ms Based on Mean 7.91 0.07
Centre o f mass displacement during stance Based on Mean 2.59 0.13
Vertical oscillation Based on Mean 1.45 0.24
Knee flexion angular velocity Based on Mean 2.03 0.18
Knee extension angular velocity Based on Mean 0.93 0.35
Knee flexion angular velocity swing Based on Mean 1.21 0.29
Step length Based on Mean 0.36 0.56
Step frequency Based on Mean 1.01 0.33
Time trial Based on Mean 1.46 0.25
Oxygen Consumption Based on Mean 0.05 0.80
Table 4 .4 .3 In te r a c tio n s fo r th e p r im a r y r e s e a r c h v a r ia b le s fo r th e 2 * 2 m ix e d fa c to r ia l A N O V A fo r P =
0.05 (A =16).
Variables F value P value

Stance time
T r e a tm e n t * G r o u p 7.28 0 .0 1 2
Anterior-Posterior GRF (Brak)
T r e a tm e n t * G r o u p 0.0 0.960
Vertical GRF
T r e a tm e n t * G r o u p 0.22 0.642
Anterior-Posterior GRF (Prop)
T r e a tm e n t * G r o u p 0.53 0.472
Centre of mass velocity (brak)
T r e a tm e n t * G r o u p 1.20 0.282
Centre of mass velocity (Prop)
T r e a tm e n t * G r o u p 0.00 0.983
Centre of mass at 2 5 ms
T r e a tm e n t * G r o u p 6.88 0 .0 1 4
Centre of mass displacement
T r e a tm e n t * G r o u p 6.88 0 .0 3 0
Vertical Oscillation
T r e a tm e n t * G r o u p 1.62 0.213
Knee flexion angular velocity7
T r e a tm e n t * G r o u p 1.27 0.269
Knee extension angular velocity
T r e a tm e n t * G r o u p 2.49 0.126
Knee flexion angular velocity
(swing)
T r e a tm e n t * G r o u p 0.51 0.481
Step length
T r e a tm e n t * G r o u p 1.50 0.231
Step frequency
T r e a tm e n t * G r o u p 9.37 0 .0 0 5
Time trial
103
T r e a tm e n t * G r o u p 0 .3 1 0 .5 8 2
Oxygen consumption
T r e a tm e n t * G r o u p 0 .6 5 2 0 .6 5 2
Table 4.4.4 T u k e y post hoc te s t s fo r th e 2 * 2 m ix e d fa c to r ia l A N O V A o n t h e p r im a r y r e s e a r c h v a r ia b le s

(tr e a tm e n t p r e te s t is 1 a n d p o s t is 2 ; th e c o n tr o l h e e l- t o e g r o u p is 3 a n d th e P o s e 5 tr e a tm e n t g r o u p is 4 ) (N
= 1 6 ).
Variables Treatment Group P-value

Stance Time I n te r a c tio n s a m o n g L e v e ls o f T r e a tm e n t* G r o u p 1 4 0.9991
T r e a tm e n t = 1 G r o u p = 3 su b tr a c te d fr o m :
2 3 0.8457
2 4 0.0006
Stance Time T r e a tm e n t = 1 G r o u p = 4 su b tr a c te d fr o m : 2 3 0.7763
2 4 0.0004
Anterior-Posterior GRF (Brak) I n te r a c tio n s a m o n g L e v e ls o f 1 4 0.9999
T r e a tm e n t* G r o u p T r e a tm e n t = 1 G r o u p = 3 s u b tr a c te d fr o m :
2 3 0.4904
2 4 0.4904
Anterior-Posterior GRF (Brak) T r e a tm e n t = 1 G r o u p = 4 su b tr a c te d 2 3 0.4486
fr o m :
2 4 0.4486
Vertical GRF (Brak) I n te r a c tio n s a m o n g L e v e ls o f T r e a tm e n t* G r o u p 1 4 0.9975
T r e a tm e n t = 1 G r o u p = 3 s u b tr a c te d fr o m :
2 3 0.8187
2 4 0.5417
Vertical GRF (Brak) T r e a tm e n t = 1 G r o u p = 4 su b tr a c te d fr o m : 2 3 0.7145
2 4 0.4298
Anterior-Posterior GRF (Prop) I n te r a c tio n s a m o n g L e v e ls o f 1 4 0.8421
T r e a tm e n t* G r o u p T r e a tm e n t = 1 G r o u p = 3 su b tr a c te d fr o m :
2 3 0.9968
2 4 1.0000
Anterior-Posterior GRF (Prop) T r e a tm e n t = 1 G r o u p = 4 su b tr a c te d 2 3 0.9252
fro m :
2 4 0.8421
Centre of mass horizontal velocity (Brak) I n te r a c tio n s a m o n g L e v e ls 1 4 0.9966
o f T r e a tm e n t* G r o u p T r e a tm e n t = 1 G r o u p = 3 su b tr a c te d fr o m :
2 3 0.9217
2 4 0.8924
Centre of mass horizontal velocity (Brak) T r e a tm e n t = 1 G r o u p = 4 2 3 0.9749
s u b tr a c te d fr o m :
2 4 0.7940
Centre of mass horizontal velocity (Prop) I n te r a c tio n s a m o n g L e v e ls 1 4 0.7612
2 3 1.00000
2 4 0.7187
Centre of mass horizontal velocity (Prop) T r e a tm e n t = 1 G r o u p = 4 2 3 0.7855
104
s u b tr a c te d fr o m :
2 4 0 .9 9 9 8
Centre of mass at 25 ms I n te r a c tio n s a m o n g L e v e ls o f 1 4 0 .9 6 8 9
2 3 0 .8 0 4 7
2 4 0 .1 0 9 2
Centre of mass at 25 ms T r e a tm e n t = 1 G r o u p = 4 su b tr a c te d fr o m : 2 3 0 .9 6 9 2
2 4 0.0420
Centre of mass horizontal displacement I n te r a c tio n s a m o n g L e v e ls o f 1 4 0 .9 8 2 0
2 3 0 .5 3 7 4
2 4 0.0002
Centre of mass horizontal displacement T r e a tm e n t = 1 G r o u p = 4 2 3 0 .7 6 0 7
s u b tr a c te d fro m :
2 4 0.0005
Vertical oscillation In te r a c tio n s a m o n g L e v e ls o f T r e a tm e n t* G r o u p 1 4 0 .9 9 8 6
2 3 0 .9 9 9 4
2 4 0 .2 7 6 6
Vertical oscillation T r e a tm e n t = 1 G r o u p = 4 su b tr a c te d fr o m : 2 3 0 .9 9 2 5
2 4 0 .3 5 1 6
Knee flexion angular velocity (Stance) I n te r a c tio n s a m o n g L e v e ls o f 1 4 0 .9 9 2 8
T r e a tm e n t * G r o u p T r e a tm e n t = 1 G r o u p = 3 su b tr a c te d fr o m :
2 3 0 .2 6 7 9
2 4 0.0046
Knee flexion angular velocity (Stance) T r e a tm e n t = 1 G r o u p = 4 2 3 0 .4 0 1 7
s u b tr a c te d fro m :
2 4 0.0091
Knee extension angular velocity (Stance) I n te r a c tio n s a m o n g L e v e ls 1 4 0 .9 8 1 7
2 3 0 .8 6 6 9
2 4 0 .7 0 1 7
Knee extension angular velocity (Stance) T r e a tm e n t = 1 G r o u p = 4 2 3 0 .9 7 8 5
su b tr a c te d fro m :
2 4 0 .4 7 4 6
Knee flexion angular velocity (Swing) I n te r a c tio n s a m o n g L e v e ls o f 1 4 0 .9 8 1 5
2 3 0 .2 9 9 8
2 4 0.0180
Knee flexion angular velocity (S w ing) T r e a tm e n t = 1 G r o u p = 4 2 3 0 .5 0 2 3
su b tr a c te d fro m :
2 4 0.0430
Step Length I n te r a c tio n s a m o n g L e v e l s o f T r e a tm e n t* G r o u p 1 4 0 .5 9 7 8
'J
2 0 .9 1 5 4
2 4 0.0058
105
Step Length Treatment = 1 Group = 4 subtracted from: 2 3 0.9287
2 4 0.1046
Step Frequency Interactions among Levels of Treatment*Group 1 4 0.9999
Treatment = 1 Group = 3 subtracted from:
2 3 0.9937
2 4 0.0016
Step Frequency Treatment = 1 Group = 4 subtracted from: 2 3 0.9887
2 4 0.0019
Time Trial Interactions among Levels of Treatment*Group Treatment 1 4 0.9997
= 1 Group = 3 subtracted from:
2 3 0.9999
2 4 0.8760
Time Trial Treatment = 1 Group = 4 subtracted from: 2 3 0.9989
2 4 0.8344
Oxygen Consumption Interactions among Levels of Treatment*Group 1 4 0.8473
Treatment = 1 Group = 3 subtracted from:
2 3 0.6391
2 4 0.5372
Oxygen Consumption Treatment = 1 Group = 4 subtracted from: 2 3 0.9815
2 4 09475
4.4.6 Limitations
The limitations of this thesis are given below in no particular order.

• Only two experienced Pose® runners were available, making definitive
statements on Pose® running preliminary.
• The runway length of 15.66 m possibly restricted the participants on the
force plate because they were thinking of slowing down owing to a brick-
wall 7 m away.
• Predominantly, the heel-toe intervention was unsuccessfully applied
owing to participant non-compliance, making the heel-toe group a control
group without an intervention.
106
CHAPTER 5
EXPERIENCED POSE® AND HEEL-TOE RUNNER’S
STUDY
5.1 INTRODUCTION
A clear method of running would be advantageous in the identification of an ‘ideal’

running technique. A universal standard of what is ‘good’ technique needs identifying to
make authoritative recommendations. The Pose® method (Romanov, 2002) has made
theoretical claims of being a universal technique that applies the laws of physics to
running, whereas the heel-toe technique is the most popular technique with endurance
runners. This study therefore seeks to quantify the differences between experienced
Pose and heel-toe runners using the Gravitational hierarchical model. Firstly, a review
(R)
of how the two techniques are presently distinguished is given. Secondly, using the
primary and secondary variables from the Gravitational hierarchical model, the two
techniques of running are compared.
5 .2 THEORETCIAL BACKGROUND
Skill, technique and style are terms used to describe human movement patterns
(Kreighbaum and Barthels, 1996). Skill is defined as a general movement pattern that is
adapted to the constraints of a particular sport, such as high jumping, which is a skill
within the general movement pattern of jumping (Kreighbaum and Barthels, 1996).
Technique represents the actions of internal and external forces on the human body,
which determine how the parts of the human body move during the performance of the
movement pattern (Hay and Reid, 1988). Following the high jump example, the Fosbury
flop and straddle are two high jump techniques. Style refers to individual adaptations of
a technique, such as a curved or straight run-up towards the high jump bar. Skills,
techniques and styles develop because of the constraints or limitations associated with a
107
particular sport. In running, the environment (forces of nature, hills, opponents and
visibility) and the human body (size, strength, power and endurance) are constraints
(Kreighbaum and Barthels, 1996). Heel-toe running is the predominant technique in
endurance runners (Ken* et al., 1983). However, a new technique (Pose®) has made
claims that it is an effective way to run. To compare the two running techniques, the
Gravitational hierarchical model (see chapter 3) will be employed.
Pose* runners have shown a distinct biomechanical profile in comparison to heel-toe

runners. Two studies took heel-toe runners and trained them in Pose® with a seven and
half-hour and twelve-hour intervention respectively (Arendse et al., 2004; Dallam et al.,
2005). A consistent finding is stride length is lower in Pose® runners compared to heel-
toe runners yet stride frequency increases in Pose® runners (Arendse et al., 2004; Dallam
et al., 2005). As running speed is a combination of stride length multiplied by stride
frequency, an endurance runner with a 2 m stride holding two strides-s'1 has a speed of 4
m-s"1. If stride length increased to 2.5 m while holding the same cadence, running speed
will be 5 m-s"1. Alternatively, if stride length remains at 2 m and stride frequency
increased to three strides-s"1, running speed will be 6 m-s"1. If the increase to three
strides-s"1 were accompanied by a decrease in stride length to 1.5 m, the running speed
would be only 4.5 m-s"1 (Hay and Reid, 1988). These measurements are obviously
correct, however, clarity on how to reduce internal muscle force in relation to producing
stride length and stride frequency in combination with the external forces of nature is not
currently lucid.
Cavanagh and Kram (1985) in a descriptive study found at speeds of 3 m-s"1 to 4.2 m-s"1,
stride length increased as opposed to stride frequency. An increase in stride frequency
potentially requires greater muscle activity during a shorter time in stance (Kyrolainen et
ah, 2000). This possibly explains why endurance runners increase stride length over
stride frequency up to 7 m-s"1, but around 7 m-s"1 stride length maximises (Cavanagh and
Kram, 1990). An increase in stride length can involve a reciprocal increase in leg
extension (Williams, 1985), although this finding is not universal (Mami et al., 1986).
Clarke et al. (1985) identified tibial accelerations increased by taking longer strides,
giving circumstantial evidence that increased stride length results in an increased impact
force. Stride frequency increases over stride length above 7 m-s'1 lacks a clear
biomechanical explanation (Luhtanen and Komi, 1978). The Gravitational hierarchical
108
model seeks to offer a clear explanation of why stride length maximises at rumiing
speeds above 7 m-s'1.
A second finding is vertical oscillation of the centre of mass decreased in Pose® runners
(Arendse et al., 2004; Dallam et al., 2005). Vertical oscillation of the centre of mass
decreases in runners generally as running speed increases (Luhtanen and Komi, 1978;
Mero et al., 1992). Further, knee extension of the support leg correlates with increased
vertical oscillation of the centre of mass after mid-stance (Wank et al., 1998). However,
knee extension decreases as running speed increases (Mann et al., 1986). Logically,
excessive vertical oscillation would be ineffective for horizontal displacement (Williams,
1985) yet, many individuals run economically despite high vertical oscillation (Williams
and Cavanagh, 1987). Vertical oscillation is maximal at terminal stance or just after
coinciding with maximal leg extension of the support limb (Fenn, 1930). Dallam et al.
(2005) found that the hip to ankle distance at impact reduced as well as stance time in
dR)
Pose runners. It would be reasonable to assume that joint angles, impact patterns and
the position of the centre of mass at impact and terminal stance influence vertical
oscillation by virtue of the kinematic chain (Mero et al., 1992). The increased body
angles at take-off and impact found at faster speeds may be associated with potential
skill differences. Changes in the body angle at take-off affects performance owing to a
change in the gravitational torque on the runner’s centre of mass in relation to their
support limb (see chapter 3). In addition, at impact, a more vertical alignment of the
centre of mass to the support limb reduces losses in horizontal velocity (Cavanagh et al.,
1977; Bates et al., 1979; Girardin and Roy, 1984). Further study of these body positions
using these variables from the Gravitational hierarchical model will aim to clarify
biomechanical differences between heel-toe and Pose® runners.
Williams et al. (2000) identified that heel-toe and forefoot runners differed in knee
flexion at impact and knee flexion excursion during stance. Arendse et al. (2004) also
found an increased knee flexion in preparation for and at initial contact in Pose® runners
as opposed to when they were heel-toe runners. Williams et al. (2000) found peak knee
flexion moment, knee power absorption and negative work at the knee differed between
heel-toe and forefoot runners. Arendse et al. (2004) also measured less eccentric work at
the knee and more eccentric work at the ankle in Pose® runners compared with heel-toe
running. The position of the torso during the gait cycle may explain the reduced knee
eccentric work and increased ankle eccentric work in Pose® running compared with their
109
heel-toe running. Arendse et al. (2004) recommended that the position of the torso and
the centre of mass should be included in future studies o f running technique
modification.
Transition from walking to running occurs at a critical speed of 2.2 m-s*1 or a similar
Froude number, rather than at a critical stride rate or length (Alexander and Jayes, 1983;
Hreljac, 1995). This suggests the centre of mass’s horizontal velocity determines the
walk-run transition rather than a particular stride length or stride rate. A change in the
centre of mass’s horizontal velocity would obviously reflect work by an external force.
Pose® running technique suggests the external force is a gravitational torque. Therefore,
falling forwards as the centre of mass passes over the support limb’s foot causes a
gravitational torque while the support foot is pulled rapidly from the ground represents
the Pose® technique. The Pose® and heel-toe running techniques differ theoretically, in
which external force is motive. Pose® technique emphasises a gravitational torque as the
motive force in running, whereas heel-toe running suggest leg extension via increased
ground reaction force increases running speed. The aim therefore, using the primary and
secondary variables is to compare the two techniques of running using the new
Gravitational hierarchical model with two skilled heel-toe and Pose® runners.
5 .3 METHODS
Participants
Two experienced male Pose® runners were recruited for this study (aged: 28 yrs and 51
yrs, stature: 1.74 m and 1.86 m, mass: 75.5 kg and 71.4 kg and 10-km personal best
times of 42 min and 29 min) and were matched (in mass, stature and age) with two
experienced male heel-toe runners (aged: 24 yrs and 47 yrs, stature: 1.76 m and 1.95 m,
mass: 72.5 kg and 71.4 kg and 10-km personal best times of 34 min). Written informed
consent and ethical approval from Sheffield Hallam University were obtained. All
participants had no history of surgical intervention, chronic pain, orthotic use or current
pathology of the lower extremity. Each participant wore his normal running shoes. All
shoes were qualitatively assessed for rigidity o f the heel counter and compliance of the
cushioning element of the rear section of the shoe.
110
Experimental protocol
All participants were pre-selected before the testing date through personal contact. All
respondents were selected based on the following criteria: 1) they had been without
injury in the past 12 months, 2) were male, 3) competitive athletes and 4) were
experienced in their respective styles of running and were similar in age, experience,
stature and weight. After their individual warm up, each participant run across a force
plate at 2.5 m-s'1, 3.35 m-s'1, 4 m-s'1 and 4.5 m-s'1 while simultaneously recording
kinematic data.
Lab door
Falcon HR 240
8 camera
system
Host
.66 m Computer
processor
Photoelectric and SVGA
cells / s monitor
Video
processor
(MIDAS)
Kistler
4.19m
Force 60 cm
3.0 m Plate
40 cm
7.0 m
Crash mat covering the brick wall
Figure 5.3.1 Measurement set-up for laboratory data collection
Collection and processing of ground reaction force data

Biomechanical data were collected using a piezoelectric transducer force plate with eight
output signals, which were sampled at 1200 Hz (9281-CA Kistler Instruments
Corporation, Winterthur, Switzerland). The plate was mounted flush with the running
surface on a 15.66 m runway. The force plate was 8.66 m from the start position and
111
measured 40 by 60 cm (figure 5.3.1). Participants were instructed to run normally while
aligning their run-up to land with their right foot on the force plate. Each participant was
allowed as many practice trials as needed to achieve ‘normal’ foot contact with the plate
in their regular running shoes, with five trials collected. After each trial, the data was
visually inspected to ensure that the footfall was ‘normal’ and centred on the force plate.
After data collection the ground reaction force analogue data was processed through
Matlab 6 software to convert volts into Newtons. In accordance with Munro et al.
(1987), ground reaction force data were normalised to the participant’s body weight.
Initial contact of the foot with the ground was identified from the ground reaction force
data. Foot contact began once the vertical ground reaction force data exceeded 20 N and
ended as it went below 20 N (Bobbert et al., 1991; Wright et al., 1998; Tirosh and
Sparrow, 2003).
Collection and processing of kinematic data

A modified Helen Hayes marker set (22-markers) was used to collect their three-
dimensional trajectories for computing kinematic data. Retroflective spheres of a
diameter of 2 cm (4.4 g) were applied to their bodies in order to define a model
comprised of twelve body segments: two segments for feet, lower legs, upper legs,
forearm, upper arm and two single segments comprising the head/neck and trunk. These
markers were applied directly onto the skin using athletic tape on the acromion
processes, elbow and radiocarpal joints, sacrum, anterior superior iliac spines, lateral
femoral condyles, lateral malleolus, and the dorsim of the feet between the second and
third metatarsals. Two lateral wands were used to calculate the medial knee and ankle
markers during the dynamic trials. The two lateral wands were 10 cm long and attached
using athletic tape to the thigh, midway between the hip and knee joints, and for the
shank, midway between the knee and ankle joints. Both wands were vertically aligned
with the relevant joints. The attachment and length of the wands were such that
oscillations of the marker at the tip of the wands were insignificant. An off-set marker
placed on the right scapula precipitated software marker recognition through body
asymmetry. Video data were collected using a VP AT 310 video recorder and an eight-
camera (Falcon HR 240) motion analysis system recording at 120 Hz (Motion Analysis
Corporation, Santa Rosa, California) which collected data for one step of the right leg
while running in the positive X direction. The cameras were zoomed in as far as possible
to a volume of 3.6 m in x (fore-aft axis) by 1.4 m in y (medio-lateral axis) and 2 m in z
(vertical axis) with the centre of the base corresponding to the centre of the force plate.
112
The laboratory (global) orthogonal coordinate system followed the right hand rule and
had the positive x-direction orientated in the direction of forward progression, the
positive y-direction orientated to the left and the positive z-direction orientated vertically
upward. The volume was calibrated before data collection using a cube and a 500 mm
wand associated with EvaRT 3.2 data collection software (Motion Analysis Corporation,
Santa Rosa, California). During data collection, the experienced heel-toe and Pose®
runners ran at 2.5 m-s'1, 3.35 m-s'1, 4 m-s'1 and 4.5 m-s'1. Running speeds were measured
by two photoelectric cells 4.19 m apart and 3 m from the centre of the force plate, and
mounted so the participant’s waist triggered the photoelectric cells. Up to five trials were
recorded at each speed, and only trials in which ‘good’ foot contact with a steady stride
and speeds within 5% of the measured speed were analysed. The three dimensional
coordinate data were filtered using a 2nd order low-pass Butterworth filter; a cut-off
frequency of 6-8 Hz was used and was selected through visual inspection of the fit
(Winter, 1987). Video and analogue data were time synchronised using impact, recorded
when vertical ground reaction force exceeded 20 N. Following the filtering process in
EvaRT 3.2, the filtered video data were extracted using OrthoTrak 5 software.
Coordinate and angular data were calculated according to the method used by Winter
(1990) from the filtered raw data.
Statistical analysis
Descriptive statistical comparisons between Pose® and heel-toe running were
undertaken.
5 .4 RESULTS
Ground reaction force variables

Vertical and posterior-anterior ground reaction forces for both techniques of running are
compared at each speed (figures 5.4.1-5.4.8). The Pose® runners elicit similar ground
reaction force traces to forefoot runners, whereas the heel-toe runners match their
traditional heel-toe counterparts. The first and second graph will always be the same
rumier in both techniques for all speeds.
113
Figure 5.4.1 Vertical and posterior-anterior ground reaction force for the Pose® runners at 2.5 nrs'1.
Figure 5.4.2 Vertical and posterior-anterior ground reaction force for the heel-toe runners at 2.5 nrs'1.
Tim. (ms) Tim. (ms)
Figure 5.4.3 Vertical and posterior-anterior ground reaction force for the Pose® runners at 3.35 m-s'1.
114
Figure 5.4.4 Vertical and posterior-anterior ground reaction force for the heel-toe runners at 3.35 m-s'1.
Figure 5.4.5 Vertical and posterior-anterior ground reaction force for the Pose® runners at 4.0 nrs'
Figure 5.4.7 Vertical and posterior-anterior ground reaction force for the Pose® runners at 4.5 nrs"1.
Motion analysis variables

Flexion-extension, abduction-adduction for the right (support) knee and hip plus vertical
oscillation of the centre of mass (figures 5.4.9-5.4.40) are graphed in the sagittal plane at
each running speed. Each graph begins at impact with the vertical line representing
terminal stance. The graph finishes at maximum swing angle for the right leg.
Hip angles represent the thigh relative to the pelvis (figures 5.4.9-16). Larger hip flexion
angles occur for all speeds in the Pose® runners at impact. At terminal stance, the heel-
toe runners exhibit hip extension, whereas the Pose® runners do not extend the hip or do
so much less.
116
\
\ .
_
\
\
\
\
\ \
\ \
\
\
\
\
\ \
\
\ \ / |
\
__ :
.... .............. _ ........... \ ,y-x ‘>3*' v‘- 055 :
---------
Figure 5.4.9 Hip flexion and extension for the Pose® runners at 2.5 m-s'1.
Timm ( ms ) T1mm{m*«)
Figure 5.4.10 Hip flexion and extension for the heel-toe runners at 2.5 m-s'1.
Figure 5.4.11 Hip flexion and extension for the Pose® runners at 3.35 m-s'1.
<U5«
Tim* (m i) Timm (me)
Figure 5.4.12 Hip flexion and extension for the heel-toe runners at 3.35 m-s'1.
117
X
\\
\\
\\
\
\ /
\\ y
.• .........
Figure 5.4.13 Hip flexion and extension for the Pose® runners at 4.0 nrs'1.
Figure 5.4.14 Hip flexion and extension for the heel-toe runners at 4.0 nrs'1.
X
X
<»
\\
\
\ \
K
\\
\ y/
«, ....
V
... . . ... ... .» »» v. .a,
Figure 5.4.15 Hip flexion and extension for the Pose runners at 4.5 nrs'1.
Figure 5.4.16 Hip flexion and extension for the heel-toe runners at 4.5 nrs'1.
118
Vertical oscillation (figures 5.4.17-5.4.24) is lower in Pose runners for all speeds. The
vertical path of the centre of mass follows a similar pattern for both techniques of
running. However, from maximum vertical ground reaction force until terminal stance
there is an increase in vertical oscillation of the centre o f mass in the heel-toe runners
compared to the Pose® participants.
Figure 5.4.17 Vertical oscillation for the Pose® runners at 2.5 nrs'1.
Figure 5.4.18 Vertical oscillation for the heel-toe runners at 2.5 nrs'1.
119
i.ii >.25
120
Knee angular displacements

Right knee (support leg) abduction and adduction angles are recorded in figures 5.4.25-
5.4.32. They verify accuracy of the marker and motion analysis system. Acceptable
accuracy was determined by abduction and adduction knee angles under 14° (Kadaba et
al. 1989). All runners for all speeds remained under 14° except the first Pose® runner
from terminal stance until maximum swing at 2.5 m-s'1, 3.5 m-s'1 and 4 m-s'1.
/
//
Figure 5.4.25 Knee adduction and abduction for the Pose® runners at 2.5 m-s'1.
Figure 5.4.26 Knee adduction and abduction for the heel-toe runners at 2.5 m-s'1.
/ \
i
I / \
I"!
* i
/ \
I f ‘i I
I I ! I
I *i I
1 ! I /
■\1 /
Figure 5.4.27 Knee adduction and abduction for the Pose® runners at 3.35 nrs'
A
/
i \
/“\ / \ !
/ \ / I 'i!'A\
/ \ / J
rJ !i \
i V - / \ J
/ \
\ /'
\J
CCKJ 002 0 03 ew 007 OOJ O.jy 012 U13 0.15 017 tli 020 0.32 0-23 0-25 027 02* U.W A — ---------------------------------------------
-
Tim# (ms) Tire* (ms)
Figure 5.4.29 Knee adduction and abduction for the Pose® runners at 4.0 m-s'1.
Tims (ms)
122
Figure 5.4.31 Knee adduction and abduction for the Pose® runners at 4.5 nrs'1.
Tim* (ms) Timt(nw)
Figure 5.4.32 Knee adduction and abduction for the heel-toe runners at 4.5 nrs’1.
Knee flexion and extension follow similar patterns for all four runners. Differences in the
knee during impact, maximum vertical ground reaction force and swing are recorded in
table 5.4.2. Increased knee extension from maximum vertical ground reaction force until
terminal stance in the heel-toe runners, correlates with their increased vertical oscillation
of their centre of mass (Wank et al., 1998).
Figure 5.4.33 Knee flexion and extension for the Pose® runners at 2.5 m-s'1.
123
Tima (ma) Tima (ma)
Figure 5.4.34 Knee flexion and extension for the heel-toe runners at 2.5 nrs'1.
Tima (ma) Tima (ma)
Figure 5.4.35 Knee flexion and extension for the Pose® runners at 3.35 nrs'1.
0.U7 *.JJJ
Tima (ma) Tima (ms)
Figure 5.4.36 Knee flexion and extension for the heel-toe runners at 3.35 nrs'1.
Thfna(aw) Tima(m a)
Figure 5.4.37 Knee flexion and extension for the Pose® runners at 4.0 nrs'1.
124
Figure 5.4.38 Knee flexion and extension for the heel-toe runners at 4.0 nrs"1.
/
/
//
/ //
//
/
//
i
\
/
// //
\ \.
/
/
//
2 /
/ /
\\
I 30 / / /
/
Figure 5.4.39 Knee flexion and extension for the Pose® runners at 4.5 m-s'1.
|8 so
| 40
I 30
Figure 5.4.40 Knee flexion and extension for the heel-toe runners at 4.5 m-s'1.
The running gait for one Pose® and heel-toe runner at each test speed are given (figures
5.5.41-5.5.49). Each still frame was taken for every five frames at a frame rate of 120
frames-s'1. At impact, the ball of the foot is the first point o f contact for the Pose® runner,
in contrast to the heel for the heel-toe runners. The Pose® runner has a more vertically
aligned body at impact plus he leaves the ground without fully extending his leg in
contrast to the heel-toe runners. Heel-toe runners had greater range o f motion for the
lower-limb—knee flexion and extension—during stance than the Pose® runners. Knee
flexion was similar for both techniques from terminal stance to maximum swing (Table
125
5.4.2). Knee flexion and extension angular velocity during stance was higher in the heel-
toe runners. However, swing knee flexion angular velocity was much higher in the Pose®
runners. The horizontal displacement of the centre of mass during stance was less for the
Pose® runners. The horizontal displacement of the centre of mass for flight was still
lower in Pose® runners but not as marked as actual step length. For example, Pose®
runners displaced their centre of masses during flight 18.7 cm, 41.6 cm, 49 cm and 55.7
cm for each test speed beginning at 2.5 m-s’1. The heel-toe runners displaced their centre
of masses during flight for 26.5 cm, 42.1 cm, 57.3 cm and 69.3 cm beginning at 2.5 m-s'
. Figures 5.4.41-5.4.49 allow visual inspection of these biomechanical differences for
the Pose® and heel-toe runners.
>
* Vi tijU
v 4 ‘n
% '4 - '4 4
Figure 5.4.41 P o s e R r u n n e r at 2 .5 m - s '1 fo r o n e ste p .
rr~. CL
A X
4
............ , , ......... 1...............
i ...
4
, '& > *
‘ft! %
^ * % *
t \ . \
I . / *
.» . /
s
i \ i \ ■. i
Figure 5.4.42 P o s e * r u n n e r at 3 .3 5 m - s '1 fo r o n e str id e .
Jr*
! «V J;.
*
*
r. \ f . ^ ‘
2 ......... ; ....... _2J1_. . •■ ;_ l.
JBi§g§s jfSSSIf - j iis s r *
r-1
A ii
ft ! J. 1
k
. . . .
7 v i ; __
Figure 5.4.43 P o se £ ru n n er at 4.0 m -s '1 fo r o n e stride.

126
Figure 5.4.44 Pose® runner at 4.5 nrs'1for one stride.
frT r:'1 | jj k K
!4-\ini iKi k1
* I*—'iM| 4*—-& • > 1$;- Ll"'
^ M hi
%\ iii ;i.!
k^ 1 :=-f■:
: : 7 : / . iy :: .
1: 7-
'v!V-' iV-V*
;. <.7 777; ^ ’§ ;i
,
.i%
Lr '•* b r 'l
. 'i\ i r l - R l
J \- u.i\i; ''rsl t |-
U4 , iJf 1 3-
N -.
**T*i / ;
/ , j; i v]
i \
•••'•' ; 7 . 7
:;:.3;4pK
. ■* -;■ kw77"E •;-v ... ,- \ ..
Figure 5.4.45 Heel-toe runner at 2.5 nrs'11for one stride.
Figure 5.4.46 Heel-toe runner at 3.35 nrs"1for one stride.
127
Figure 5.4.47 Heel-toe runner at 4.0 nrs'1for one stride.
Figure 5.4.48 Heel-toe runner at 4.5 m-s'1for one stride.

Figure 5.4.49 The second heel-toe runner at 3.35 nrs'1for one stride.
Figures 5.4.50-5.4.57 time-align the support leg’s hip, knee and ankle joint flexion and
extension with maximum vertical ground reaction force. Vertical oscillation is also
included to highlight the centre of mass displacement. The graphs illustrate at maximum
vertical ground reaction force for the Pose® runner the hip begins to extend or has been
extending depending upon running speed. The knee and ankle follow in sequence each
extending after the hip, proceeding maximum vertical ground reaction force. The heel-
toe runner at all speeds has begun to extend the hip at maximum vertical ground reaction
force, but the knee has also begun its extension. Vertical oscillation o f the centre of mass
changes direction from downwards to upwards at maximum vertical ground reaction
force. In the heel-toe runner the centre of mass has just began to rise at maximum
vertical ground reaction force. This rise of the centre of mass probably correlates with
the knee extension (Wank et al., 1998).
/
./
Figure 5.4.50 Pose® runner at 2.5 m-s'1. Figure 5.4.51 Heel-toe runner at 2.5 m-s'1.
130
QUO 0.117 OI<3 01.IP 6167 0217 Oil* 0 2SO 026? 0.2M 0*10 IMP 0 HI 0.1» 0.16?
T lm ( n i)
S(i<0 *(Ki7
OWII 0 117 0.131 0150 0167 « 1*3 I20U 0250 0 267 02* '
Figure 5.4.52 Pose® runner at 3.35 m-s'1. Figure 5.4.53 Heel-toe runner at 3.35 m-s‘
131
Tim* ( m )
Figure 5.4.54 Pose® runner at 4.0 m-s'1. Figure 5.4.55 Heel-toe runner at 4.0 m-s*1.
132
/
I10 II? ■IS
Figure 5.4.56 Pose® runner at 4.5 m-s'1. Figure 5.4.57 Heel-toe runner at 4.5 m-s'1.
133
Figures 5.4.58-61 have vertical lines running from maximum vertical and anterior
(propulsive) ground reaction force through the horizontal velocity and acceleration
graphs. Maximum horizontal acceleration of the centre of mass occurs before maximum
horizontal propulsive ground reaction force. Only one heel-toe and Pose® runner at 2.5
m-s'1 had a horizontal acceleration after the propulsive horizontal ground reaction force
(figure 5.4.58). Table 5.4.1 gives the maximum values for vertical and anterior ground
reaction force and maximum horizontal acceleration of the centre of mass and
percentages of stance time. Pose® runners consistently had higher horizontal acceleration
of the centre of mass during the propulsive phase of stance.
G r o u n d r e a c tio n f o r c e fo r a P o sd * r u n n e r G r o u n d r e a c tio n fo r c e fo r a h e e l- t o e ru n n e r
H o r iz o n ta l v e lo c it y o f th e cen tre: o f m a s s H o r iz o n ta l v e lo c it y o f th e c e n tr e o f m a s s fo r a h e e l- t o e
fo r a P o s e * r u n n e r . ru n n er.
H o r iz o n ta l a c c e le r a tio n o f th e c e n tr e o f m a s s H o r iz o n ta l a c c e le r a t io n o f th e c e n tr e
fo r a P o s e * r u n n e r . o f m a s s fo r a h e e l- t o e r u n n e r .
Figure 5.4.58 G r o u n d r e a c tio n fo r c e , h o r iz o n t a l v e lo c it y a n d a c c e le r a t io n o f th e c e n tr e o f m a s s f o r a P o s e *

a n d h e e l- t o e r u n n e r at 2 .5 m -s ’1.
134
Ground reactionIforce for a Pose runner Ground reaction orce for a heelitoe runner
Horizontal velocity of the centrejof mass Horizontal velocity of the centre; of mass for a heel-toe
for a Pose® runner. I runner.
Horizontal acceleration of the centre of mass Horizontal acceleration of the centre

for a Pose® runner. of mass for a heel-toe runner.
Figure 5.4.59 Ground reaction force, horizontal velocity and acceleration of the centre of mass for a Pose®
and heel-toe runner at 3.35 m-s'1.
Ground reaction force for a Pofce® runner Ground reaction force for a heeljtoe runner
Horizontal velocity bf the centfe of mass Horizontal velocity of the centre! of mass for a heel-toe
for a Pose® runner. I runner.

136
Ground reaction force for a Pos£ runner Ground reaction force for a heelttoe runner
Horizontal velocity of the centra of mass Horizontal velocity of the centre! of mass for a heel-toe
for a Pose® runner. j runner. I j

137
Table 5.4.1 M a x im u m v e r t ic a l a n d a n te r io r g r o u n d r e a c tio n fo r c e a re r e c o r d e d a s a p e r c e n t a g e o f s ta n c e
a n d c o m p a r e d w it h m a x im u m h o r iz o n t a l a c c e le r a tio n o f th e c e n tr e o f m a s s . E a c h v a lu e is a m e a n o f th r e e
tr ia ls (N = 4).
Participant run Time of maximum Time of maximum Time of maximal

speeds vertical ground anterior ground horizontal
reaction force (%) reaction force (%) acceleration of the
centre of mass (%)
1 P o se® 2 .5 m s '1 36 71 69
1 P o se® 3 .3 5 m -s 1 37 72 65
1 P o se® 4 . 0 m -s"1 40 74 72
1 P o se® 4 .5 m s 1 43 74 65
2 P o se® 2 .5 m - s '1 38 70 71
2 P o se® 3 .3 5 m - s '1 41 65 50
2 P o se® 4 .0 m -s"1 43 71 66
2 P o se® 4 .5 m s '1 40 71 70
1 H e e l- t o e 2 .5 m -s 1 42 75 88
1 H e e l- t o e 3 .3 5 m -s 1 48 73 70
1 H e e l- t o e 4 .0 m s 1 40 74 63
1 H e e l- t o e 4 .5 m -s 1 43 74 63
2 H e e l- t o e 2 .5 m - s '1 41 72 67
2 H e e l- t o e 3 .3 5 m -s"1 44 74 66
2 H e e l- t o e 4 .0 m s '1 45 74 59
2 H e e l- t o e 4 .5 m s'1 46 74 58
Primary and secondary variables for both the heel-toe and Pose® runners at all the test
speeds are given in Table 5.4.2. Similar to other studies, stance time for both styles of
running decreases with increases in rumiing speed (Williams, 1985). All the ground
reaction force variables increase with running speed (Miller, 1990). Stride length also
increased as running speed increased. However, stride frequency increased as running
speed increased for the Pose® runners in contrast with all other research on endurance
running speeds, whereas the heel-toe runners followed conventional patterns of no
increase (Patla et al., 1989). Vertical oscillation reduced with running speed but more
significantly in Pose® runners. Horizontal velocities for the centre of mass—braking and
propulsion—were consistent across the speed range in the heel-toe runners but at 4 m-s'1
and 4.5 m-s'1, the Pose® runners had significant reductions in braking.
138
Table 5.4.2 Two experienced heel-toe and Pose* runner's mean scores across all collected speeds. The
primary research variables are in bold ( N = 4).
Variables Model 2.5 nrs'1 3.35 nrs'1 4.0 firs’1 4.5 nrs'1
ST (ms) HT 272 260 230 209
P 262 216 195 174
M A-P GRF (brak) (BW)HT -0.16 -0.20 -0.25 -0.30
P -0.14 -0.23 -0.25 -0.28
M V GRF (BW) HT 2.46 2.58 2.84 2.93
P 2.32 2.63 2.77 2.84
M A-P GRF (prop) (BW)HT 0.108 0.15 0.19 0.25
P 0.084 0.16 0.2 0.24
T A-P GRF (brak) HT 23.7 23.8 23.4 25.1
(% o f stance) P 12.5 8.4 13.6 7.7
T V GRF HT 41.3 44.1 42.5 44.8

(% o f stance) P 37.8 37.9 39.9 40.6
T A-P GRF (prop) HT 73.7 73.3 74.4 74.7
(% o f stance) P 69.9 71.2 72.7 72.9
CM (brak) (cm-s'1) HT -18.3 -22.7 -16.9 -18.3
P -12.1 -15.3 -2.2 -3.0
CM (prop) (cm-s'1) HT 12.2 20.4 18.5 22.2
P 9.7 17.7 8.3 12.9
CM diff (cm-s'1) HT -6.1 -2.3 1.6 3.9
P -2.4 2.4 6.1 9.9
HVTO (m-s'1) HT 2.64 3.50 4.13 4.6
P 2.51 3.6 4.12 4.8
CM S (cm-s'1) HT -16.0 -18.9 -33.3 -36.3
P -10.8 -27.3 -14.8 -13.7

CM 25 (cm) HT 35.5 37.2 38.4 37.8
P 29.8 32.7 33.2 32.1
CM TO (cm) HT 32.9 40.5 46.0 50.1
P 34.3 36.8 43.1 45.5
CM 0 (rad) HT 0.30 0.35 0.40 0.44
P 0.27 0.34 0.37 0.42

CM dis (cm) HT 75.1 89.3 91.4 94.5
P 64.4 75.6 78.2 82.9
SHA (cm) HT 15.6 16.6 12.9 11.3
P 7.95 8.0 5.2 2.8
VO (cm) HT 12.0 11.9 11.95 11.1

P 8.5 9.3 8.5 7.2
KF (rad) HT 0.57 0.55 0.51 0.45
P 0.42 0.44 0.39 0.33
KE (rad) HT 0.56 0.64 0.69 0.69
P 0.39 0.48 0.45 0.46
KSW (rad) HT 1.2 1.5 1.8 1.9
P 1.2 1.6 1.7 1.8
KTD (rad) HT 0.25 0.34 0.36 0.41
P 0.26 0.32 0.36 0.43
KVGRF (rad) HT 0.83 0.87 0.84 0.83
P 0.67 0.74 0.75 0.75
KFV (rad-s'1) HT 5.0 5.4 6.1 5.5
P 4.4 5.3 6.1 4.2
KEV (rad-s'1) HT 3.1 4.1 4.9 5.2
P 2.7 3.5 4.1 4.8
KFVS (rad-s'1) HT 5.4 6.4 7.5 7.95
P 6.2 7.7 9.1 10.1
SL (cm) HT 103.0 130.8 153.0 170.3
P 84.8 119.2 130.3 141.8
FT (ms) HT 110.0 120.0 150.0 150.0
P 90.0 120.0 120.0 110.0
SF (Hz) HT 2.5 2.7 2.6 2.8
P 2.9 3.0 3.2 3.5
Technique: P = Pose® and HT = heel-toe, ST = stance time (ms) right foot, M A-P GRF (brak) = maximum
posterior ground reaction force (braking) (BW), M V GRF = maximum vertical ground reaction force
(BW), M A-P GRF (prop) = maximum anterior ground reaction force (propulsive) (BW), T A-P GRF
(brak) = time of posterior ground reaction force (braking) (% of stance), T V GRF = time of maximum
vertical ground reaction force (% of stance), T A-P GRF (prop) = time of anterior ground reaction force
(propulsion) (% of stance), CM (brak) = centre of mass horizontal velocity largest decrease after impact
until maximum vertical ground reaction force (cm-s'1), CM (prop) = centre of mass horizontal velocity
largest increase from maximum vertical ground reaction force until terminal stance (cm-s'1), CM diff =
horizontal velocity of the centre of mass from maximum vertical ground reaction force until terminal
stance minus horizontal velocity of the centre of mass from impact until maximum vertical ground reaction
(cm-s'1), HVTO = horizontal velocity of the centre of mass at terminal stance (m-s'1), CM S = centre of
mass horizontal velocity decrease between terminal stance and maximum swing (flexion) for the right leg
during flight (cm-s'1), CM 25 - centre of mass behind the to e-marker (horizontal difference at 25 ms) (cm),
CM TO = centre of mass forward of the toe-marker (horizontal difference at terminal stance) (cm), CM 0
= centre of mass and toe-marker angle at terminal stance from the vertical axis going clockwise (rad), CM
dis = centre of mass displacement during stance (cm), SHA = shoulder, hip, and ankle marker difference in
vertical alignment at 25 ms (cm), VO = vertical oscillation of the centre of mass (cm), KF = knee flexion
range during stance (rad), KE = knee extension during stance (rad), KSW = knee flexion from terminal
stance to maximum swing (rad), KTD = knee angle at impact (rad), KV GRF = knee angle at maximum
vertical ground reaction force (rad), KFV = mean knee flexion angular velocity for stance (rad-s'1), KEV =
140
mean knee extension angular velocity for stance (rad-s'1), KFVS = mean knee flexion angular velocity
from terminal stance to maximum swing angle (rad.s'1), SL = step length for right foot (cm), FT = flight
time (ms), SF= step frequency (Hz).
The primary and secondary variables are grouped in table 5.4.3 as similar, lower or
higher for the Pose® runners compared with the heel-toe runners.
Table 5.4.3 Variables are grouped as similar, lower and higher for the Pose® runners compared with the
heel-toe runners (N = 4).
Similar Lower Higher

M V GRF T A-P GRF (brak) KFVS
M A-P GRF (brak) TV GRF SF
M A-P GRF (prop) CM (brak) CM diff
T A-P GRF (prop) CM (prop) H accel. CM
CM 0 CMS Hip flexion
KSW CM 25
KTD CM TO
HVTO CM disp (stance)
SHA
VO
KF
KE
KVGRF
KEV
KFV
ST
SL
FT
Hip ext.
Ankle flexion at impact
CM disp (flight)_______
The experienced heel-toe runners were compared with the literature where possible to
ensure they accurately represented the recreational endurance running population.
141
Table 5.4.4 Variables for the heel-toe runners (N = 2) are matched with the literature at similar speeds.
Any minor speed differences are noted after the reference. The thesis research variables fall within one
standard deviation of the literature. Over-ground running studies were selected over treadmill running
where possible owing to some kinetic and kinematic differences (Wank et al., 1998).
RV (m-s'1! HT Literature Sources

ST (ms) 2.5 272 283 (22) Martin (1985)
3.35 260 258(18) Munro et al. (1987)
4.0 230 229(14) Munro et al. (1987)
4.5 209 214(13) Munro et al. (1987)
M A-P GRF (brak) (BW)
2.5 -0.16
3.35 -0.20 -0.17(0.03) Munro et al. (1987)
4.0 -0.25 -0.21 (0.02) Munro et al. (1987)
4.5 -0.30 -0.24 (0.02) Munro et al. (1987)
M V GRF (BW)
2.5 2.46
3.35 2.58 2.52 (0.19) Williams and Cavanagh (1987) at 3.57 n
4.0 2.84 2.72 (.17) Munro et al. (1987)
4.5 2.93 2.79 (.18) Munro et al. (1987)
M A-P GRF (prop) (BW)
2.5 0.108
3.35 0.15 0.16(0.01) Munro et al. (1987)
4.0 0.19 0.20 (0.01) Munro et al. (1987)
4.5 0.25 0.23 (0.02) Munro et al. (1987)
T A-P GRF (brak) (% of stance)
2.5 23.7
3.35 23.8
4.0 23.4 21.48 Hamill et al. ((1983)
4.5 25.1
TV GRF (% of stance)
2.5 41.3
3.35 44.1 40.1 (3.7) Frederick and Hagy (1986)
4.0 42.5 42.3 (3.5) Frederick and Hagy (1986) at 3.83 m-s'1
4.5 44.8 42.1 (4.1) Frederick and Hagy (1986) at 4.47 m-s'1
T A-P GRF (prop) (% of stance)
2.5 73.7
3.35 73.3
4.0 74.4 73.05 Hamill et al. ((1983)
4.5 74.7
CM (brak) (cm-s-1)
2.5 -18.3
3.35 -22.7
4.0 -16.9
4.5 -18.3 -18.0 Cavanagh and LaFortune (1980)
CM (prop) (cm-s'1)
2.5 12.2
3.35 20.4
4.0 18.5
4.5 22.2 27.0 Cavanagh and LaFortune (1980)
VO (cm)
2.5 12.0
3.35 11.9 11.6(1.2) Collins et al. (2000)
4.0 11.95
4.5 11.1 11.8(1.1) Collins et al. (2000)
KF (rad)
2.5 0.57 0.57 (0.09) Christina et al. (2001) at 2.9 m-s'1
3.35 0.55
4.0 0.51
142
4.5 0.45
KSW (rad)
2.5 1.2
3.35 1.5
4.0 1.8 1.85 Nilsson and Thorstensson (1985) at 3.83 m-s'
4.5 1.9
KTD (rad)
2.5 0.25
3.35 0.34
4.0 0.36 0.24 (0.09) Diss (2001)
4.5 0.41
KVGRF (rad)
2.5 0.83
3.35 0.87 0.78 (0.06) Morgan et al. (1990)
4.0 0.84
4.5 0.83
KFV (rad-s'1)
2.5 5.0 5.2(1.03) Martin (1985)
3.35 5.4
4.0 6.1
4.5 5.5
SL (cm)
2.5 103.0
3.35 130.8 119.5 (0.14) Cavanagh and Kram (1985)
4.0 153.0 152.0 (.07) Wank et al. (1998)
4.5 170.3
FT (ms)
2.5 0.11
3.35 0.12 0.093 (0.025) Martin (1985)
4.0 0.15
4.5 0.15
SF (Hz)
2.5 2.5 2.6 (0.07) Kyolainen et al. (1995)
3.35 2.7 2.7 (0.09) Kyolainen et al. (1995)
4.0 2.6 2.72 (0.10) Kyolainen et al. (1995)
4.5 2.8
5 .5 DISCUSSION
Discemable differences between the two running techniques were identified. Both
techniques utilise the same external (gravity and ground reaction force) and internal
force (muscle force), however differences occur from the timing o f these forces resulting
in different displacements of the runner’s body and lower-limbs. Discussion o f these
differences follows the Gravitational hierarchical model. The Gravitational hierarchical
model is included for ease of reference from chapter 3.
143
Time
Distance
Relative Knee angular

height at velocity for stance Air
take-off and swing resistance
Body position Time of Time of

/physique support non-support
Landing Flight Take-off

distance distance distance Step time
Mean horizontal velocity of the centre of mass

throughout the gait cycle
Mean step Mean step

length 5 frequency
Ground reaction Muscle force/strain

force 2-4 energy
Centre of mass Acceleration

location to support of runner’s
limb 1 body
Gravitational
force
Figure 5.5.1 T h e G r a v ita tio n a l h ie r a r c h ic a l m o d e l. A g a it c y c l e is r ig h t f o o t c o n ta c t to r ig h t f o o t t o e

c o n ta c t o f th e s a m e f o o t.
144
5.5.1 Comparison of the two running techniques using the Gravitational
hierarchical model
The heel-toe runners exhibited similar profiles to those already established (see table.
5.4.4). For example, Munro et a V s (1987) ground reaction force data were similar to this
study at the same test running speeds. All the variables in table 5.4.4 for the experienced
heel-toe runners were within one standard deviation of those found within the literature.
Owing to the similarities between the experienced heel-toe runners and the literature, it
seems reasonable to suggest the two experienced heel-toe participants were a good
representation of the male recreational heel-toe endurance running population. The two
experienced Pose® runners are also a good example of Pose® running. Arendse et al.
(2004) recorded similar characteristics from their Pose® runners as this study’s
experienced Pose® runners. For example, the experienced Pose® runners were similar to
Arendse et al. (2004) for vertical oscillation, stance time and step length.
Comparison of running techniques across the four speeds illustrates clear differences
between the heel-toe and Pose® runners (table 5.4.2). Running is a full body activity so
thirty-four kinetic and kinematic variables (fourteen primary research and twenty
secondary research variables) were used to describe both running techniques. Table 5.4.3
depicts these variables for the Pose® runners as similar, lower or higher than the heel-toe
technique. Vertical and posterior-anterior ground reaction forces and take-off velocity at
terminal stance have similar maximums. The time of the propulsive ground reaction
force is also similar as well as the angle o f inclination o f the centre of mass from the
vertical axis at terminal stance. Finally, knee angle at impact and maximum swing were
similar. Only five variables are higher in the Pose® runners: knee flexion angular
velocity from terminal stance to maximum swing, the difference between braking and
propulsive horizontal velocity of the centre of mass, horizontal acceleration o f the centre
of mass, hip flexion at impact plus step frequency. Twenty-one kinematic variables were
lower for the Pose® runners. The proceeding discussion using the Gravitational
hierarchical model highlights these differences between heel-toe and Pose® running.
Distance and time

Time measures how long it takes to run the distance for a particular event and is the
performance criterion. The performance criterion is the measure used to evaluate the
level of success in the performance of a sports skill (Bartlett, 1999). Notice that the
145
hierarchical model is relevant for any running event from sprinting to marathons. The
Pose® method claims to be a technique that both sprinters and marathon runners can use
because both still experience a gravitational torque, whereas endurance runners
predominantly use the heel-toe technique with sprinters landing on the forefoot.
Air resistance
Air resistance was not measured in-line with most other gait research (Anderson, 1996).
However, strong winds change runner’s technique by causing them to lean into the wind
(Davies, 1980). Interestingly, Pose® running suggests forward movement is the result of
a gravitational torque; by leaning forward in the wind, the torque increases via an
increased moment arm.
Relative height at take-off

Vertical oscillation was 3.4 cm lower on average in Pose® runners for the mean of all the
test speeds. As running speed increased vertical oscillation decreases for both techniques
in-line with other research (Luhtanen and Komi, 1980; Mann and Hagy, 1980; Ito et al.,
1983a). Knee flexion from impact to maximum vertical ground reaction force was also
lower in Pose® runners, which suggests decreased knee flexion accounts for reduced
vertical oscillation. Less knee extension also correlates with lower vertical oscillation
(Wank et al., 1998). As the body passively extends the support knee (Mann et al., 1986),
having the support foot remain longer on the ground potentially increases vertical
oscillation in heel-toe runners. Stance time was less in Pose® runners and knee extension
was lower, possibly supporting this concept. As running speed increased, vertical
oscillation of the centre of mass decreased, again suggesting leg extension is not related
to running speed increases (Hunter et al., 2005). In fact, the Pose® technique recorded no
increases in leg extension across the endurance run speeds of 3.35-4.5 m-s'1. The heel-toe
runners extend their knee and increase their vertical oscillation after maximal vertical
ground reaction force until terminal stance, but they do not increase knee extension as
running speed increases (4.0-4.5 m-s'1). Pose® runners have increased mean knee angular
velocity from terminal stance until maximum swing as they rapidly pull the foot from the
ground yet they have lower mean knee extension angular velocity during stance. This
may explain why heel-toe runner’s knees extend more, because they are not thinking
about pulling then foot from the ground, and hence their knee passively extends further,
increasing vertical oscillation. The Gravitational hierarchical model illustrates the
146
relationship of vertical oscillation with knee flexion and extension angular velocity
owing to it being on the same level in the model.
Knee flexion and extension angular velocity during stance

Enomoto et al. (1999) using an effective index for mechanical energy tested thirty-two
participants. They found less knee flexion in the first half of stance, and lower vertical
oscillation correlated with smaller reductions in horizontal velocity of the centre of mass
(P < 0.01). Their fastest runners exhibited these traits (less knee flexion, lower vertical
oscillation and less braking of the centre of mass horizontal velocity), which matched
this study’s Pose® runners. In support, Lee and Farley (1998) stated stance-limb impact
angle and virtual stance-limb compression both play important roles in determining the
trajectory of the centre of mass. In agreement, figure 5.5.2 clearly shows the different
centre of mass’s vertical displacements of the two techniques, highlighting the affect
gravity has on certain body geometry. Arendse et al. (2004) identified ankle position also
distinguished between the two techniques at impact, with the ankle dorsiflexed (-18.3°
±9.8°) in heel-toe and neutral (-0.4° ±4.9°) in Pose® running. Low work values were also
found at the knee and high eccentric work values for the ankle 6.91 J/kg (±1.93)
compared to 5.85 J/kg (±1.39) as heel-toe runners. This suggests the magnitude of the
moment arm at the ankle in Pose® running is greater than at the knee during eccentric
loading. Arendse et al. (2004) proposed this might be owing to a combination of the
lower-limb geometry and the position of the torso at impact. Positioning of the torso over
the supporting limb at impact may influence the direction of the ground reaction force in
relation to the lower-limb joint centres thus increasing eccentric work at the ankle in
Pose® runners. Whether running barefoot in Arendse et a V s (2004) study caused
increased eccentric work at the ankle in the Pose® technique or the technique itself,
requires further investigation, which is outside the scope of this thesis.
Increasing running speed via knee extension lacks support (Mann et al., 1986). Instead,
the propulsive phase of stance is attributed to hip motion. Belli et al. (2002) measured
peak joint ankle, knee and hip power for nine male endurance runners at speeds of 4.0
m-s 1 to 6.0 m-s'1 and at their maximal running speed. Ankle and knee joint power
increased with running speed, but the highest changes were for the hip (327 ±203 W at
4.0 m-s'1and 1642 ±729 W at maximal speed). They concluded that hip extensors are the
prime forward movers as running speed increases in contrast to the knee and ankle. In
support, power generation at the knee was significantly lower in Pose^ runners (5.01
147
W/kg ±2.49) than heel-toe runners (6.27 W/kg ±1.44) (Arendse et al., 2004). Arendse et
al. (2004) found no significant increase in ankle power generation and concentric work
in Pose runners. They suggested the hip is the source of power for forward propulsion,
(D •
which was likely increased in Pose® running. Novacheck (1996) noted that the
contribution of knee and ankle power is minimised as running speed increases. The
concept that the hip extends leading the knee and ankle joints from maximum ground
reaction force to initiate or enable forward movement is actually well documented
(figures 5.4.50-57; McClay et al., 1990; Montgomery et al., 1994; Pink et al., 1994; Belli
et al., 2002; Hunter et al., 2005). Leg extension by the forward progression of the hip
passively extending the knee seems a plausible explanation. In confirmation, the hip,
knee and ankle reach their maximum moments just after maximum vertical ground
reaction force in sprinters and these moments were associated with impact (Mann, 1981;
Biewener et al., 2004). After maximum vertical ground reaction force, they decreased.
This suggests joint moments are associated with impact and are not propulsive per se.
The Pose® technique suggests the body is falling via a gravitational torque from
maximum vertical ground reaction force, because leg and hip muscle extensors are silent,
and ground reaction force is decaying (see chapter 3). Knee extension angular velocity is
lower in Pose® runners while the hip is leading the lower-limb joints extension seen
clearly in figures 5.4.50-57 which time aligned the lower-limb. The Pose® technique
proposes it is body weight (via a gravitational torque) leading the propulsive movement
as the runner falls forwards after mid-stance, reflected by hip extension, hence the lower
knee extension angular velocity in the Pose® runners.
Figures 5.5.50-57 show the two techniques time aligned. A consistent observation is
knee extension has just begun at maximum vertical ground reaction force in the heel-toe
runners, whereas the knee is at maximum flexion in Pose® runners. In both heel-toe and
Pose runners, the hip has consistently begun to extend before the knee, but in heel-toe
runners the knee extends earlier. Perhaps the early knee extension in the heel-toe runners
increases work against gravity as stated earlier through increased vertical oscillation. Leg
extension of the support limb however, should be passive by using the elastic recoil and
forward movement of the body. Mann et al. (1986) in support states: “Knee extension
which occurs during the toe-offphase o f support during jogging and running is the result
o f the forward movement o f the body over the fixed foot and tibia.” Passive leg extension
148
would complement the Pose^ method’s concept of pulling the foot from the ground
rather than trying to extend the leg via the ‘push-off foot-ground contact.
Knee flexion angular velocity from terminal stance to maximum swing

Knee flexion angular velocity from just before or at terminal stance until maximum knee
swing angle was higher in Pose runners for all speeds, but particularly at the faster
(R)
speeds. At the two faster running speeds, Pose® runners had 121% and 127% greater
mean angular knee flexion velocity than the heel-toe runners. Pose® runners significantly
increased their mean angular knee flexion velocity from terminal stance to maximum
swing potentially via pulling their foot from the ground. This would bring the foot closer
to a faster moving body during flight, potentially reducing angular inertia of the lower-
limb (Enomoto et al., 1999). Knee flexion angles are similar for both techniques at
maximum swing, yet the Pose® runners reach this position with greater angular velocity,
which supports a lower rotational inertia of the leg from terminal stance. The
Gravitational hierarchical model therefore, attempts to reflect the lower-limb’s key
relationship to the centre of mass’s horizontal velocity and the forces that cause their
motion.
Step time
Step time consists of flight and stance time and both were less in the Pose® runners
compared to the heel-toe runners. Increased stance time has correlated positively (r =
0.4; P = 0.05) with poorer running economy (Williams and Cavanagh, 1986; Williams
and Cavanagh, 1987). As endurance runners become fatigued, stance times increases
(Nicol et al., 1991), possibly reflecting reduced elastic energy in the muscle mechanisms
(Zatsiorsky, 1995; Paavolainen et al., 1999a). Nicol et al. (1991) suggest that a time
delay between the muscle stretch and the shortening action may widen when fatigued
possibly reducing elastic energy. The Pose1 runner’s body geometry and lower-limb
range of motion appear to reduce stance time, which may potentially enhance elastic
energy.
Distance travelled (landing, flight and take-off distance)

Landing distance, flight distance and take-off distance represent one-step length in the
hierarchical model. Several primary variables represent these distances, landing distance
matches the centre of mass displacement in the horizontal direction during stance, flight
distance is step length and the centre of mass horizontal displacement at terminal stance
149
to the support limb is take-off distance. All distances were less in Pose runners. The
centre of mass horizontal displacement for Pose® runners during stance was 12.3 cm less
on average across the speed range. Possibly the increased horizontal displacement
travelled by the heel-toe runners decreases their horizontal velocity for their centre of
mass owing to different body geometry at impact and terminal stance than the Pose®
runners. Step length is lower in Pose® runners, however horizontal velocity of the centre
of mass at terminal stance were similar, with a mean for all test speeds of 3.75 m-s 1and
3.73 m-s'1 for Pose® and heel-toe runners respectively. In a legless projectile a similar
horizontal take-off velocity, would produce a similar horizontal displacement. It appears
heel-toe runners are extending their leg at impact to increase step length. Further,
increases in the distance between the support limb and the centre of mass at impact
decrease horizontal velocity, which impedes the runner’s progress (Mann et al., 1986).
Importantly, increased stride length correlates with increased ground reaction force
(Clarke et al., 1985) cited as a potential injury risk (Volishin and Wosk, 1982) although
not evident in this study. The Gravitational hierarchical model attempts to connect the
horizontal velocity of the centre of mass and the lower-limb with the forces that cause
stride length, potentially leading to a better understanding of running injury mechanisms.
Mean horizontal velocity of the centre of mass throughout the gait cycle
The mean horizontal velocity of the centre of mass for the gait cycle is divided into three
phases: impact to maximum vertical ground reaction force (braking) and from this point
until terminal stance (propulsion), then from terminal stance to maximum swing knee
flexion angle. The heel-toe runners had consistent braking in their progression across all
speeds. The Pose® runners brake less at 2.5 m-s’1 and 3.35 m-s’1 by -0.2 cm-s’1 and -7.2
cm-s'1. At 4.0 m-s’1 and 4.5 m-s'1, the differences were more marked at -14.7 cm-s'1 and -
15.3 cm-s’1 respectively. Pose® runners at 4.0 m-s’1 and 4.5 m-s’1 were only braking their
forward progression of their centre of mass by -2.2 cm-s'1 and -3.0 cm-s'1 respectively,
compared with -12.1 cm-s'1 and -15.3 cm-s'1 at the lower speeds of 2.5 m-s'1 and 3.35
m-s"1. The more aligned position of the support limb and the centre of mass at impact,
especially at the faster speeds, may explain these differences, along with the vertical
alignment of the shoulder, hip and ankle in the Pose® runners. In support, Bates et al.
(1979) showed their 400 m runner with the largest deceleration during stance also had
the largest distance from the ankle to their centre of mass at impact of 0.27 m compared
with 0.14 m for their runner with the least deceleration. The timing of the maximum
150
braking force is also earlier in Pose runners suggesting accelerations of the body in the
horizontal direction are different between the two techniques at impact. Bobbert et al.
(1992) found a backward acceleration of the lower-limb segments relative to the rest of
the body just before impact. The knee marker’s velocity became horizontal in the
forward direction at impact, as it rotated around the now stationary ankle, because the
ankle can no longer move forwards while the foot is on the ground. Further, the hip’s
tangential acceleration remained positive, indicating it lead the lower-limb through
stance owing to the body’s inertia. A lower braking effect on the horizontal velocity of
the Pose® runners may reflect their distinct body geometry at impact (shoulder, hip and
ankle vertical alignment).
The propulsive horizontal velocity from maximum vertical ground reaction force until
terminal stance was higher at all speeds for the heel-toe runners. The differences were
much larger again at the faster speeds of 4 m s 1 and 4.5 m s'1 with heel-toe runners
having 10.2 cm-s'1and of 9.3 cm-s 1higher propulsive velocity respectively. An increased
propulsive horizontal velocity in the heel-toe runners may reflect their swing leg being
driven forwards, discussed later. Driving the swing leg forwards owing to its large mass
will have a significant affect on the centre of mass’s velocity. Interestingly, horizontal
velocity at terminal stance was similar between both techniques. A possible explanation
is the propulsive horizontal velocity— change between maximum vertical ground
reaction force and terminal stance—masks a rapid acceleration in the Pose® runners as
they fall utilising the gravitational torque more effectively via less vertical oscillation. In
support, figures 5.4.58-5.4.61 consistently show higher horizontal accelerations of the
centre of mass in the Pose® runners during the propulsive phase of stance. Further, when
the differences between braking and propulsion of the centre of mass’s horizontal
velocity were calculated, Pose® runners show better overall propulsion, suggesting
movement that is more efficient. The mean propulsive horizontal velocity (difference
between braking and propulsion) across the four speeds was 16 cm-s'1 and -2.9 cm-s'1 for
the Pose® and heel-toe runners respectively. By braking less, the overall horizontal
velocity was positive in Pose® runners. It would seem reasonable to suggest this is more
efficient, because the heel-toe runners needed to generate higher increases in horizontal
velocity after maximum vertical ground reaction force potentially via driving their swing
legs forwards.
151
j M ax im u m a n te rio r T erm in a l stan ce
in M axim um j gro u n d re actio n
horizontal I force
a c c e le ra tio n
o f th e c e n tre o f Fg ▼
Impact
P ropulsive co m p o n e n t
o f gravity
M ax im u m vertical ground
in
o re a c tio n force —
T e rm in a l stanc
Im pact
M ax im u m vertical ground
re a c tio n fo rce
P ro p u lsiv e com p o n e n t
o f gravity
P osition v ec to rs
in
CO
0.00
P o in t o f support Time (m s )
Figure 5.5.2 a ) r e p r e s e n ts th e v e r tic a l o s c illa t io n o f a P o se® r u n n e r at 4.0 m -s '1. b ) r e p r e s e n ts a h e e l- t o e

r u n n e r at 4.0 m -s '1.
Figure 5.5.2 illustrates gravity’s acceleration on the two runners using the vertical
oscillation of their centre of masses. Gravity has no affect on the horizontal velocity of
the centre of mass during flight. However, while the support foot is on the ground a
propulsive gravitational torque occurs from maximum vertical ground reaction force
until terminal stance. The runner’s vertical oscillation is from impact until maximum
vertical height reached during flight. The support foot’s position to the centre of mass is
drawn beneath maximum vertical ground reaction force to certain points on the path of
the centre of mass (light grey arrow). Both runners have their actual time of maximum
horizontal acceleration of their centre of mass (dotted line) and their maximum anterior
ground reaction force (dashed line) marked. Gravity’s work (black arrow) on the centre
of mass as it rises from maximum vertical ground reaction force via elastic mechanisms
and body geometry (see chapter 3) until terminal stance is constant. However, gravity’s
propulsive component (dark grey arrow) changes depending upon the position of the
centre of mass to the support foot (vector product: sine of the angle between gravity and
the position vector). It is evident that the Pose® runner is able to utilise gravity more
effectively for horizontal progression via less vertical oscillation compared to the heel-
toe runner, owing to the larger propulsive component of gravity (dark grey arrow).
152
The maximal horizontal acceleration of the centre o f mass at all speeds for the Pose®
runners was higher, possibly reflecting a more effective use of the gravitational torque.
Table 5.4.1 consistently shows the maximal horizontal acceleration of the centre of mass
occurs before the maximum horizontal ground reaction force. This suggests the
gravitational torque is the only other possible force (only two external forces in running)
that could cause this maximum horizontal acceleration o f the centre of mass. Further
research should determine the horizontal acceleration of the centre of mass during stance
using accelerated running, at a frame rate of at least 240 frames-s'1. This should verify
the Pose® technique and whether the Gravitational hierarchical model effectively
explains running. However, the Gravitational hierarchical model does reflect the
complex inter-relationship of the effects of force and body kinematics on the horizontal
velocity of the centre of mass.
Mean step length and step frequency

Stride length and stride frequency commonly evaluate individual runners and their
differences at various running speeds (Hogberg, 1952; Cavanagh and Williams, 1982;
Martin, 1985; Patla et al., 1989). Both running techniques increase their step length as
running speed increases in-line with other studies (for example, Patla et al., 1989). The
Pose® runners have shorter step lengths than the heel-toe runners for all speeds, with the
faster running speeds— 4.0 and 4.5 m-s'1—differences being 22.7 cm and 28.5 cm
respectively. At running speeds of 2.5 m-s'1 and 3.35 m-s'1, the differences were 19.0 cm
and 11.6 cm respectively, whereas step frequency was higher at all speeds for the Pose®
runners. Heel-toe runner’s step frequency is similar to other gait studies (for example,
Wank et al., 1998; table 5.4.4). Pose® runners increased step length by 167% from 2.5-
4.5 m-s'1 compared to 165% in the heel-toe runners. However, Pose® runners increased
their step frequency from 2.5-4.5 m-s'1 by 121% compared to 112% in the heel-toe
runners. Similar step length increases took place despite different absolute lengths, but
Pose® runners increased their frequency more over the speed range.
Vertical alignment of the shoulder, hip and ankle in the Pose® runners at impact, may
suggest reaching or driving the swing leg forwards in front of the body, is absent in
Pose® runners. Heel-toe running emphasises driving the swing leg forwards (Williams
and Cavanagh, 1987). However, Williams and Cavanagh (1987) identified less knee
drive in their most economical runners. Driving the swing leg will increase step length,
and may possibly affect the timing of the braking ground reaction force (Pose® and heel-
153
toe runners braking force was 10.6% and 24% of stance respectively). Support for
driving the swing leg forwards in heel-toe runners is exhibited by lower differences for
horizontal displacement during flight between the centre of mass and actual step length
for the Poses runners. For example, the differences in step length and centre of mass
horizontal displacement during flight between the two techniques for the four test speeds
beginning at 2.5 m-s'1 were 10.4 cm; 11.1 cm; 14.4 cm; 14.9 cm respectively. As the
differences during flight for step length and centre of mass horizontal displacement were
lower in the Pose® runners, it suggests the actual step length differences result from the
heel-toe runners driving their swing leg forwards. In addition, the horizontal
displacement of the centre of mass was lower in the Pose® runners for all test speeds,
suggesting their swing leg returned to the ground earlier.
Kaneko et al. (1987) calculated optimal step frequency for 2.5 m-s'1, 3.5 m-s'1 and 4.5
m-s'1 using the work-energy relationship. Highest efficiency occurred at step frequencies
between 2.8-3.0 Hz. The heel-toe runners fall below this range (2.5-2.8 Hz) while the
Pose® runners are within or above this range (2.9-3.5 Hz). Kaneko et al. (1987)
concluded that at lower frequencies, runners’ muscles must develop relatively high
external power (external work) to obtain longer strides, and at higher frequencies, the
power to move limbs increases internal work. It is not certain yet, which is more
beneficial for performance. Increasing stride frequency for a given speed consistently
increases oxygen consumption and raises heart rate (Knuttgen, 1961; Cavanagh and
Williams, 1982). Dallam et al. (2005) found a mean increase in sub-maximal absolute
oxygen cost of 3.28 (±0.36) Lmiin"1 to 3.53 (±0.43) Lnnin'1, P < 0.01 in their Pose®
trained group, whereas this thesis found no change (see chapter 6). This studies control
group (see chapter 6) and Dallam et aV s (2005) exhibited no significant changes in
either running kinematics or oxygen cost. The Gravitational hierarchical model suggests
step length and step frequency are a consequence of the forces involved in running. From
the Pose® perspective, the body falls forwards creating step length without the need to
drive the swing leg forwards. In support, Novacheck (1998) recorded the pelvis and
trunk tilt further forwards as running speed increases, suggesting body position is
creating the increased speed. Increased body tilt, increases the gravitational torque,
which potentially results in faster running (changes in stride length and stride frequency)
up to a point where the angle of inclination maximises. After this theoretical point,
running speed could only increase via increased step frequency.
154
Muscle force
Reviewing electromyographical studies on running, each study consistently found that
leg and hip extensor muscle activity ceased just after maximal vertical ground reaction
force as leg extension begins supporting the ‘extensor paradox’ (Brandell, 1973; Mann
and Hagy, 1980; Pare et al., 1981; Schwab et al., 1983; Nilsson and Thorstensson, 1985;
Montgomery et al., 1994; Heise et al., 1996). Montgomery et al. (1994) recorded eleven-
gait muscles for jogging, training and race pace. They found similar activity patterns at
each speed with a general trend for increased activity as running speed increased. They
determined the loading phase began at impact and continued until the centre of mass
passed in front of the knee. The quadriceps muscle group were all maximally active up to
this point to stabilise the knee. The biceps femoris muscle demonstrated an increase in
activity as the centre of mass passed the knee, which they suggested initiated hip
extension. Another possible explanation is the lower-limb extends because the body is
moving forwards. Therefore, leg extension does not cause the centre of mass to pass
forwards of the knee, rather the hip passively extends via inertia of the body activating
the biceps femoris muscle (muscle lengthens via hip extension). Of the eleven gait
muscles, only one muscle was significantly active after the centre of mass passed the
knee until terminal stance: the iliacus muscle. Their conclusion was hip flexion during
early and middle swing and knee extension during late swing provided forward
propulsion, rather than knee or ankle extension at terminal stance. The Pose® technique,
however, regards hip extension during stance’s propulsive phase as a consequence of a
gravitational torque, because the limbs during flight cannot horizontally (or vertically)
displace the centre of mass. Montgomery et al. (1994) also noted eccentrically
contracting muscles at impact (i.e. the rectus femoris, vastus intermedius and short and
long head of biceps, semimembranosus) may have increased risk of injury at faster
running paces. Pose® runners had lower knee eccentric work of 17 J/kg (0.08) than when
they ran heel-toe 46 J/kg (0.15) (Arendse et al., 2004), whereas they had an increase in
eccentric ankle work 36 J/kg (0.9) and 43 J/kg (1.5) respectively.
Mero et al. (1992) recorded electromyography for the rear leg at the sprint start. The
biceps femoris recorded peak activity at terminal stance while other muscles were
virtually off, except for minimal activity from the rectus femoris. This potentially
indicates that the rear leg is pulled-out (knee flexion) of the block because the leg
extensors are not active and knee flexion is occurring via the hamstring group. It is
plausible if a sprinter can pull his leg out of the blocks, then Pose® runners could be
155
pulling their feet from the ground. In support of this concept, knee angular flexion
velocity from terminal stance until maximum swing is 17% higher in Pose® runners for
the mean across all test speeds. Further, leg extension angular velocity during stance is
13% lower in the Pose® runners. It appears the Pose® runners do not forcefully extend
their legs, yet they powerfully flex their knees from terminal stance until maximum
swing, which strongly suggests a pulling action.
The stretch-shortening cycle enhances muscle force output; however, muscle

electromyography for leg extensors after maximum vertical ground reaction force
reduces rapidly (Komi, 2000). A possible explanation for this dichotomy, is the muscles
that extend on landing which enable a pre-stretch such as ankle and knee extensors, both
interestingly have long tendons (Achilles and patella tendons) that are excellent storage
sites for elastic strain energy (Alexander and Bennet-Clark, 1977). Efficiency of
movement is possible because elastic mechanisms do not require adenosine triphosphate
breakdown. During eccentric contraction, the cross-bridges forcibly detach and reattach
without adenosine triphosphate splitting (Curtin and Davies, 1974). The stored elastic
strain energy subsequently releases in the concentric muscle phase beginning at or just
after maximum vertical ground reaction force (Farley and Ferris, 1998). The leg then
acts as a spring allowing extension, but not from muscle contraction (owing to a lack of
muscle activity reflected by electromyography at this time), although tension remains in
the musculotendinous junctions (Edman, 1979). Further, vertical ground reaction force is
decaying from its maximum point until terminal stance (body is rising off the ground),
while elastic recoil acts in a predominantly vertical direction (angle of inclination at
terminal stance is 70° to the horizontal), reducing work against gravity. Whether Pose®
runners are more effective in utilising strain energy is unknown at this time. Muscle
force and muscle elasticity are hierarchically above gravity in the Gravitational model,
because gravity (body weight) causes the timing of their activity at impact as they absorb
landing and as the gravitational torque accelerates the runner after mid-stance muscle
activity ceases.
Ground reaction force

Mean braking ground reaction force was 0.23 body weights for Pose8 and heel-toe
runners across all four speeds. The mean time for the braking posterior ground reaction
force across all speeds, as a percentage of stance was 10.6% for Pose® and 24% for heel-
toe runners. Horizontal distance of the centre of mass from the support limb and running
156
speed at impact affects impact force (Mero et al., 1992; Hay, 1994). The time of braking
force is earlier in Pose® runners, possibly reflecting the shorter distance the centre of
mass was behind the support limb at impact via a more vertically aligned posture (Hamill
et al., 1983). Unshod Pose® runners had a lower braking force than their heel-toe running
trials of 113.7 N (15% body weight) and 148.7 N (20% body weight) respectively
(Arendse et al., 2004). Munro (1984) recorded 25% body weight at a similar running
speed of 3 m-s'1 in shod runners. Presently it is uncertain whether shoes or running
technique, cause differences in the posterior ground reaction force.
Maximum vertical ground reaction also revealed almost similar bodyweights—2%

difference—for both techniques across all test speeds. Frederick and Hagy (1986) found
body mass accounted for 90% of maximum vertical ground reaction force. Chang et al.
(2001) using reduced gravity with a specially designed harness for treadmill running
reduced gravity to 0.5 g, 0.38 g and 0.25 g. Maximum vertical ground reaction force
reduced from 1495 (100) N at Earth’s gravity to 848 (58) N, 664 (40) N and 463 (39) N
at the reduced gravity respectively. Clearly, gravity significantly affects maximum
vertical ground reaction force. Chang and Kram (1999) applied a horizontal force o f
10%, 15% and 20% of gravity specific body-weight to his runners. Interestingly,
maximum vertical ground reaction force barely changed with the applied horizontal
force, suggesting maximum vertical ground reaction force is weight related and thus
caused by gravity.
Maximum vertical ground reaction force and vertical oscillation of the centre o f mass
were time-aligned (figures 5.4.50-57). In both techniques, maximum vertical ground
reaction force aligns with the lowest point o f the centre o f mass (Cavanagh and
LaFortune, 1980; Bobbert et al., 1991; Derrick et al., 2000). The ground reaction force
curves decays from this point until terminal stance as the knee extends. The ‘extensor
paradox’ (McClay et al., 1990) identifies a lack o f muscle activity as the knee extends
from just after maximum vertical ground reaction force until terminal stance. If the
lower-limb extensor muscles were active, a spike or plateau would be evident in the
vertical ground reaction force curve (see Yeadon and Challis, 1994; figure 3.3.14, p. 84).
However, ground reaction force curves for all techniques of running have a smooth
decay rate. Similar to gravity’s work on landing as weight is absorbed (force curve
increases) until maximum vertical ground reaction force, the decay rate from this point
until terminal stance reflects the body leaving the ground (centre of mass is rising).
157
Therefore, as the runner’s body leaves the ground (rises by muscle elasticity/body
geometry) the runner falls forwards via a gravitational torque, while ground reaction
force reduces. In fact, at faster running speeds, the decay rate increases (Miller 1990),
reflecting faster falling.
Propulsive ground reaction force may reflect the body pivoting around the foot as the
runner leaves the ground without the foot actually pushing backwards (Mann et al.,
1986; Miller, 1990). If the foot pushed backwards, leg extension would begin at the
ankle and not the hips (figures 5.4.50-57; Mann et al., 1986; Montgomery et al., 1994;
Riley et al., 2001; Belli et al., 2002). The propulsive anterior ground reaction force
(R)
occurred at 72% and 74% of stance in Pose and heel-toe runners respectively, with a
mean of 0.17 body weights recorded for both heel-toe and Pose® runners. The angle of
inclination (producing the moment arm) for the centre of mass at terminal stance were
also similar for both techniques possibly adding support that the gravitational torque is
the motive force owing to the similar running speeds. Hunter et al. (2005) using 28
sprinters at 8 nvs"1 grouped them into high and low propulsion trials. They found the
high propulsion trials increased propulsive impulse (0.37 m s’1 to 0.33 m-s'1 for low
propulsion), expressed relative to body mass, thus reflecting the change in velocity. It is
possible that the faster running speeds produce increased friction from a faster body
motion, rather than an increased pushing off the ground. Further, Hunter et al. (2005)
using multiple linear regression to predict sprint velocity, found relative propulsive
impulse explained 57% and relative braking impulse 7% of the variance in sprint
velocity. Ground reaction force is a reflective (reactive) force with leg extensor muscles
silent during the propulsive phase of stance; ground reaction force must be decreasing.
This suggests a gravitational torque causes the ground reaction force trace, owing to a
lack of any other force at this time (see chapter 3 for more a more detailed explanation).
Chang et al. (2000) reduced gravity by 0.75%, 0.50% and 0.25%, which reduced braking
and propulsive ground reaction force as well as the horizontal impulse significantly.
Conversely, when they increased gravity and inertia by 130% of body weight, propulsive
ground reaction force increased and the horizontal impulse increased by 28%. However,
increases in inertia only of 130% of body mass increased horizontal impulse by only
10%. This clearly suggests gravity (in relation to a gravitational torque) has a strong
relationship to propulsive ground reaction force (18% change in horizontal impulse).
This potentially explains the ‘extensor paradox’ (silent leg extensor muscles during the
158
propulsive phase of stance) owing to gravity's strong relationship to propulsive ground
reaction force rather than muscle activity. Further, Chang et al. (2000) used the same test
speed for all gravitational conditions, and found the angle of inclination at impact and
terminal stance were the same. This suggests the gravitational torque and the propulsive
ground reaction force are related, because similar changes in gravity cause reciprocal
reductions in the propulsive force. These reductions however, do not change the angle of
inclination, because this appears directly related to maintaining their constant running
test speed across all conditions of 3.0 m-s"1. Chang et al. (2000) concluded that gravity
and not inertia has a major influence over both vertical and horizontal ground reaction
force generation during running. In agreement with Chang et al. (2000), the
Gravitational hierarchical model shows gravity affecting ground reaction force via body
position.
Acceleration of the body

Maximum horizontal acceleration of the participant’s centre of masses showed only one
trial at 2.5 m-s"1 for the Pose® and heel-toe runners occurring after the propulsive
horizontal ground reaction force. Despite difficulties in obtaining acceleration from
position data, it appears that maximum horizontal acceleration does not occur at
maximum horizontal ground reaction force or just after it, but rather before. Further
research should verify these findings, because a large horizontal acceleration without a
corresponding horizontal force is not possible. The Gravitational hierarchical model
suggests the horizontal acceleration is caused via a gravitational torque and the ground
reaction force reflects the work of gravity on the runner’s body.
The heel-toe runners increased vertical oscillation of the centre of mass from maximum
vertical ground reaction force until terminal stance increases work against gravity, which
potentially reduces their gravitational torque. The Pose® runners consistently produced
higher horizontal acceleration of the centre of mass coupled with less vertical oscillation.
The Gravitational hierarchical model links gravity’s work on the rumier by identifying
the centre of mass location in reference to the support limb. As the centre of mass moves
anterior to the support limb a gravitational torque ensues. Pose1 runners appear to utilise
the gravitational torque more effectively through lower vertical oscillation and a faster
leg recovery (less leg extension).
159
Centre of mass location to the support limb
At impact with the ground taken at 25 ms of stance to allow for potential differences in
vertical impact (Arendse et al., 2004), noticeable variations were found for the centre of
mass position in relation to the point of support. Vertical aligmnent— centre of mass
more vertically aligned to the support foot— in the Pose® runners was 7 cm less (more
aligned) at 25 ms for the mean of four speeds compared with the heel-toe runners. In
fact, as running speed increased this distance decreased in both Pose® runners and heel-
toe runners possibly indicating it is beneficial to do so. Hunter et al. (2005) in support,
found at impact the distance between support and the centre of mass was lower for
runners with a reduced braking impulse.
Shoulder, hip and ankle vertical alignment at 25 ms after impact also showed more
vertically aligned Pose® runners— 8 cm less for the mean of four speeds—with this
difference becoming more marked as running speed increased. Shoulder, hip and ankle
vertical aligmnent resembles a coiled spring because of the body’s ‘S’ like shape, which
Pose® runners are trained to land in. Lee and Farley (1998) referred to the centre of mass
progression in running as similar to a bouncing ball. Perhaps this landing body geometry
aids in enhancing elastic mechanisms. Further, the body cannot move forward until the
centre of mass passes over the support foot (ball of the foot). Therefore, a more aligned
body position at impact enables the runner’s centre of mass to pass more quickly over
the support foot. This would suggest Pose® runners conserve momentum, because the
centre of mass’s difference between braking and propulsive horizontal velocity was
positive in the Pose® runners (table 5.4.2).
The horizontal distance from the centre of mass to the toe-marker at terminal stance was
less in Pose® runners for all speeds except 2.5 m-s'1. The angles of inclination of the
centre of mass from the vertical axis at terminal stance were similar in Pose® and heel-
toe runners (mean of four speeds 20.4°: 21.2° respectively). If the angle of inclination is
similar and the horizontal distance less in Pose® runners, the heel-toe runners must have
more extended legs at terminal stance, which is in fact the case. The maximum
acceleration of the centre of mass was higher in the Pose® runners. This suggests that
despite similar angles of inclination, the Pose01' runners accelerate more from maximal
vertical ground reaction force until terminal stance. Less knee extension in the Pose®
runners, therefore, may facilitate increased horizontal acceleration via a gravitational
160
torque. A clear link between the lower-hmb’s flexion and extension angular velocity and
the centre of mass’s displacement is therefore evident (Farley and Gonzalez, 1996).
Gravitational force
As an external force causes motion, the Gravitational hierarchical model identifies
gravitational force as the leading cause of movement because the ground and muscle
system do not cause propulsion (Zatsiorsky, 2002; p.51). Several body weights are
recorded at impact with the ground after flight in running, reflecting gravity’s work on a
runner’s body mass, with both techniques—heel-toe and Pose®—being equally affected
(Bobbert et al., 1991). Flight is predominately spent in descent, because maximum
vertical oscillation occurs at or just after terminal stance (Fenn, 1930; Cavanagh et al.,
1977). There are, however, differences in landing posture between the two techniques, so
the impact effects of gravity’s work on the runners will vary. The body continues to
lower after impact under gravity’s work until maximum vertical ground reaction force as
the centre of mass reaches its lowest position (Cavanagh and Lafortune, 1980; Bobbert et
al., 1991). Immediately after this as the centre of mass passes anterior to the support limb
(ball of the foot), the two techniques again show significant variations (figures 5.4.50-
57) in movement as a gravitational torque potentially accelerates the runner’s body to
terminal stance and into swing as the cycle is repeated again. Interestingly, the Pose®
runners had lower propulsive velocity, which is more marked at the faster speeds.
However, the Pose® runner’s angle of inclination and horizontal velocity at terminal
stance was similar to the heel-toe runners. A possible explanation for these apparently
conflicting results is that Pose® runners have lower vertical oscillation coupled with less
leg extension. Therefore, the Pose® runners do less work against gravity and hence are
potentially more able to utilise the gravitational torque to their advantage (see chapter 3;
figure 3.3.11) gaining an increased horizontal acceleration of their centre of mass. In
support, Gracovetsky (1988) recorded legless participants walking on their ischiums by
keeping normal spinal motion (spinal engine theory) and electromyography patterns for
the trunk musculature. Spinal engine theory promotes the central role of gravity as the
motive force for legless locomotion, illustrating gravity’s potential for forward
progression.
161
5 .6 SUMMARY
Pose® and heel-toe runners had similar ground reaction forces. The time of braking and
the vertical ground reaction force were earlier in the Pose® runners. Propulsive ground
reaction force occurred at a similar percentage of stance time. Horizontal velocity of the
centre of mass was lower in Pose® runners for braking and propulsion, while the angle of
inclination of the centre of mass at terminal stance increased. Vertical displacement and
horizontal displacement of the centre of mass were lower in Pose® runners. Their body
was more aligned at impact; with knee flexion and extension during stance being lower
in the Pose® runners. Knee angle at impact was the same but at maximum vertical
ground reaction force it was lower in the Pose® runners. Stance time, flight time and step
length were lower, while stride frequency was higher in Pose® runners. Mean knee
flexion and extension angular velocity were lower during stance, whereas mean angular
knee flexion velocity during swing was higher. Pose® runners therefore, exhibit very
different gait characteristics to traditional heel-toe runners. It was argued, that leg
extension did not propel the runner forwards, while using hip motion, it was implied that
a gravitational torque creates forward progression. Support included silent leg and hip
extensor muscle activity, decaying ground reaction force, an increased propulsive
component of gravity in Pose® runners and the hip leading the lower-limb joints
extension during the propulsive phase of stance. Joint moments were maximal at
maximum vertical ground reaction force and therefore, not related with propulsion.
Ground reaction force reflects gravity’s work as the body leaves the ground because the
lack of muscle activity during propulsion coupled with reducing ground reaction force
suggests a gravitational torque is the motive force. Both heel-toe and Pose® runner’s
maximal horizontal acceleration occurred before their horizontal maximal ground
reaction force, suggesting a gravitational torque by default is the motive force in running.
It appears that both heel-toe and Pose® running techniques fall forwards via a
gravitational torque, suggesting that these runners do not push off the ground. It was
proposed that owing to lower vertical oscillation Pose® runners used the gravitational
torque more effectively. The research variables were therefore, able to clearly explain the
Pose® technique and the Gravitational hierarchical model. Whether Pose® running is
effective at improving performance or reducing injury is discussed in the next chapter.
162
CHAPTER 6
POSE® METHOD TRAINING STUDY
6.1 INTRODUCTION
A review on the importance of improved running economy an optimal biomechanical

profile and injury reduction is given. This study then seeks to apply the Pose® technique
(Romanov, 2002) of running to eight male endurance runners and compare it with eight
heel-toe runners using the primary and secondary research variables from chapter 3 and
5. The intervention of the Pose® method technique applied for 7-hours over seven days,
will seek to improve performance. This chapter therefore, aims to review the
biomechanical changes (optimal biomechanical profile) from a Pose® method
intervention, and their affect upon performance, running economy and injury incidence.
6.2 THEORETCIAL BACKGROUND
Three parameters affect running performance: economy, an optimal biomechanical

profile and sustaining an injury (Williams, 1985). Each parameter is reviewed in
reference to how it may improve running performance.
Endurance runners are an array of shapes and sizes, which influences their running
economy (Bergh et al., 1991). From a survey of 1,468 male endurance runners, the mean
stature was 176.5 cm and body mass was 65.8 kg, but stature ranged from 152-197 cm
and body mass from 40-100 kg (Frederick and Clarke, 1981). In animals, size also has a
dramatic affect on running economy; a mouse uses twenty times the energy of a pony to
run a mile (Taylor, 1994). There has been substantial research on the energy cost of
running and differing body mass (for example, Oyster and Wooten, 1971; Bale et a l,
1986). Their conclusions on endurance running attribute running economy to the effects
of stature and body mass. Running economy’s relationship with stature and body mass
163
relates to the Froude number. The Froude number links running speed to gravity (body
mass) and leg length (stature), suggesting gravity as an external force plays a significant
role in running performance. If speed of locomotion increases and leg extension at
terminal stance remains the same as speed increases (Hunter et al., 2005), then gravity is
the only mechanism that can increase running speed. However, gravity is a constant
force. In order for gravity to increase running speed, gravity must be able to work in such
a way as to increase its affect on the runner. Gravity’s increased affect on running speed
must be the result of the angle of inclination of the body (creating a gravitational torque)
during the propulsive phase of stance (see chapter 3, 5).
Williams and Cavanagh (1987) identified that 54% of the variance in ruiming economy
was attributed to biomechanical variables. Examining the affects of technique on running
economy identified several related biomechanical factors (Williams and Cavanagh,
1987). For example, running economy has been associated with a freely chosen stride
frequency and length (Cavanagh and Williams, 1982). It has also been related with lower
vertical oscillation (Cavanagh et al., 1977), earlier co-activation of hamstring and
gastrocnemius muscles during the support phase (Heise et al., 1996), and reduced plantar
flexion occurring at higher speeds during foot removal (Williams and Cavanagh, 1987).
Lower vertical oscillation and less plantar flexion would reduce work against gravity and
therefore, enhance the effects of the gravitational torque on the runner.
Miura et al. (1973) suggested, technique attributed to the differences in running speed—
5.6 m-s'1 to 5.2 m-s'1— for their participants running with the same maximum oxygen
uptake of 70 ml/kg-min. Their faster rumiers all had a greater angle of inclination (34°) of
the body at terminal stance compared with their poor runners (31.8°). An increased angle
of inclination produces an increased moment arm for gravity from the centre of mass to
the support foot, which increases the gravitational torque potentially increasing running
speed.
Several studies attempted to modify running economy by manipulating running

technique directly via stride length (Cavanagh and Williams, 1982, Petray and
Krahenbuhl, 1985; Kaneko, 1987, Morgan et al., 1994). However, these studies failed to
achieve a significant change in either running technique or economy. In contrast, Dallam
et al. (2005) was successful in producing significant changes in triathletes’
biomechanical profiles by employing the Pose® technique using twelve hours of
164
instruction. Elite triathletes in the British Olympic squad via anecdotal reporting (Chris
Jones: head coach) were able to improve their running times by employing the Pose®
technique. The mean improvement in their 10-km personal best times for two males and
two females during the 2003/04 season was 108.5 s, although the mechanisms for this
improvement remain unclear. This present study will aim to identify the potential
biomechanical changes and their effects on performance.
Cavanagh and LaFortune (1980) found a mean decrease in the horizontal velocity of the
centre of mass of 0.18 m-s'1 during the braking phase for seventeen participants running
at 4.47 m-s'1. If the foot lands further ahead o f the centre of mass at impact, an increased
braking effect can occur (Fenn, 1930; Deshon and Nelson, 1964; Cavanagh et al., 1977;
Bates et al., 1979; Kunz and Kaufman, 1981; Girardin and Roy, 1984; Hinrichs, 1990).
Impact patterns also affect stance-limb impact angles at the hip, knee and ankle, via the
kinetic chain (Bartlett, 2000), which subsequently alter the centre of mass vertical
displacement (Lee and Farley, 1998). Lower-limb angles, vertical oscillation o f the
centre of mass and foot position at impact, therefore, have an important relationship to
reductions in horizontal velocity of the runner and consequently performance.
Miura et al.’s (1973) good runners exhibited a longer stride length of 1.77 m compared
to their poor runners of 1.60 m over a 5-km race, while stride-time was similar. In
contrast, Cavanagh et a l (1977) found their good runners took shorter strides with a
higher frequency in comparison to poorer runners. Possible differences may reflect the
fact that Cavanagh et al.’s (1977) study was conducted on a treadmill where ten strides
were filmed after ten minutes of running at 4.47 m-s'1, whereas Miura et aV s (1973)
runners ran 5-km on a track at 5.2-5.6 m-s'1. It is currently not lucid whether runners
should have longer strides or higher frequency.
Dallem et al. (2005) found a significant reduction in their Pose® group in vertical
oscillation in comparison to the control group at a given treadmill velocity following the
Pose® method intervention. Although not statistically significant, the Pose® group
recorded reductions in mean hip to ankle distance at foot strike (more vertically aligned
body position) and lower support time. However, the modifications in running
biomechanics elicited by the Pose® method intervention were also associated with a
significantly increased submaximal oxygen cost (a 4.2 ml/kg/min increase in steady state
oxygen cost at 215 m-min following the instructional period). This may be owing to the
165
fact that the majority of the participants were experienced enough to have already
developed their most economical self-selected stride rate. Notably one of the eight Pose®
group participants did demonstrate an improvement in economy following the
instructional period. He was the least experienced triathlete/runner in the group with a
two year training history in running and triathlon, as well as having the lowest initial
stride rates (2.6 steps-s'1 at 215 m-min and 2.8 steps-s'1 at 250 m-min). This participant
may not have been experienced enough to have developed his most economical freely
chosen stride length before the study allowing him to possibly respond more favorably to
the instruction. The mean increase in running economy of approximately 8%, should
substantially affect running performance in their Pose® group. However, the Pose® group
in Dallem et al. (2005) also demonstrated a similar perceived exertion during the pre and
post steady state trial speeds (RPE = 11.3 + 2.1 versus 11.0 ± 1.8 and 14.5 + 1.6 versus
14.7 + 3.1). They suggested the increase in steady state oxygen cost on the participant’s
steady state rate of perceived exertion may have been negated by a reduction in the
muscular activation associated with each individual step (reduced stride length and
increased stride frequency).
Dallam et al.'s methodological design may have also attributed to the increase in running
economy. The running economy data was collected after the lactate threshold and
maximum oxygen consumption tests. Further, biomechanical data was acquired after the
physiological tests when all the participants were in an extremely fatigued state. The data
were also collected on a treadmill, which pulls the foot backwards. The pulling back of
the foot significantly affects Pose® running technique because the foot should be pulled
directly upwards in Pose® running not backwards. It is unlikely the Pose® group were
able to maintain their new technique when fatigued while on a treadmill, which could
have affected the results. Therefore, this study aims to collect oxygen consumption in a
non-fatigued state during over-ground running, to determine, if the Pose® technique does
in fact improve running economy. Dallam et al.'s (2005) study did not include a
performance test. This present study will include a 2400 m time trial to determine if
Pose® running improves performance from the potential biomechanical changes.
The Pose® technique has also made claims that it reduces running injuries (Romanov,
2002). Pose® runners land on the forefoot, however, forefoot running is not associated
with a lower risk of injury (Cavanagh and LaFortune, 1980). Arendse et al. (2004)
trained twenty heel-runners in Pose® running. They found Pose® running was
166
characterised by a shorter stride length, lower vertical impact force, a greater knee
flexion in preparation for and at initial contact, less eccentric work at the knee and more
eccentric work at the ankle compared with heel-toe and forefoot running. They proposed
the position of the torso during the gait cycle may explain the reduced knee eccentric
work and increased ankle eccentric work in Pose® running compared with the heel-toe
running technique. They suggested that the position of the torso and the centre of mass
should be included in future studies of running technique modification. This present
study will measure the centre of mass position and velocity including torso position
during the gait cycle.
Arendse et al. (2004) evaluated their runners in the barefoot condition, to which they
were not habituated. Although standard to all test conditions, the barefoot condition may
prevent the adoption of a conventional heel-toe running technique. Runners protect the
heel during barefoot running and reduce distortion of the heel fat pad (De Wit and De
Clercq, 2000). Subsequently, initial contact occurs predominantly with the anterior
portion of the foot. Their horizontal braking and propulsive ground reaction forces
differed between Pose® and heel-toe running in contrast to this thesis (chapter 5). Pose®
running also produced significantly smaller vertical displacement of the body, the feet
were kept close to the supporting surface with a shorter stride length compared with
heel-toe running. However, Arendse et al. (2004) collected no injury data post
intervention to determine if these Pose® running changes reduced injuries.
The aim of this present study is to measure via an economy and time trial run the
participant’s pre-test baseline physiological measures while the primary research
variables will provide a biomechanical profile. The post-test will determine whether
there is an improved performance on the economy and 2400 tune trial run resulting from
the potential biomechanical changes in the Pose® group from a 7-hour Pose® training
intervention. A 3-month post intervention survey for any injury incidence will follow for
all participants.
167
6 .3 M E TH O D S
Participants
Sixteen recreational male heel-toe runners and triathletes were recruited for this study.
The heel-toe group = 8) and the Pose group's = 8) personal information are
(K) •
(7 7 (7 7
recorded in table 6.3.1. Written informed consent and ethical approval from Sheffield
Hallam University were obtained before testing. All participants had no history of
surgical intervention, chronic pain, orthotic use or current pathology of the lower
extremity, except one participant from the heel-toe group post-test who had sore ankle
musculature. Each participant wore his normal running shoes. All shoes were
qualitatively assessed for rigidity of the heel counter and compliance of the cushioning
element of the rear section of the shoe.
Table 6.3.1 P a r tic ip a n t's p e r s o n a l d a ta (N = 1 6 )
Group Age (yrs) Stature (cm) Mass (kg) 10-km P.B. time (min)
P re P o st
Heel-toe Group
Range 1 8 -2 7 1 7 4 - 1 9 3 .5 6 5 .3 - 1 0 0 .5 6 5 .3 - 1 0 0 .5 3 8 -4 5
Mean 2 0 .8 1 8 0 .2 8 2 .3 8 2 .0 41
S.D. 3 .6 6 .2 11 1 0 .9 2 .1
Pose® Treatment Group
Range 1 9 -2 4 1 7 3 -1 8 9 7 1 .4 - 8 5 .7 6 9 .4 - 8 5 .7 3 8 -4 5
Mean 2 1 .1 1 8 0 .9 7 6 .9 7 6 .5 4 0 .6
S.D. 1 .7 5 .2 5 .4 6 .7 2 .8
Experimental protocol (see appendix 1 for participant recruitment)

All participants were pre-selected before testing via personal contact. All respondents
were selected based on the following criteria: 1) they had been without injury in the past
3-months. 2) They had been running for at least 2-years. 3) No-one in the study had read
or heard anything about the Pose® method of running. 4) They were heel strikers, 5) sub
elite runners (37-45 minute 10-km P.B.), 6) male, 7) and aged between 18-30 years old.
8) The pre-selection meeting finalised whether a participant was accepted into the study.
Two participants were rejected from the study based on the study criteria. Sixteen
participants were available for the study dates and divided into two groups of heel-toe
and Pose® treatment runners. The participants were randomly distributed through an
168
equal bias towards running technique from the pre-test questionnaire scores (table 6.4.4).
All participants were tested on two separate occasions in a pre-post test design.
Track test (see appendix 2 for track data collection sheets)

The track pre-tests were held between May 24th and 25th, 2003 where all the participants
ran a 2400 m economy run followed immediately by a 2400 m time trial on a running
track at Loughborough University. The post-test track tests took place on June 6th 2003.
After a self-selected warm-up a 2400 m economy run was collected at 8 min per mile
pace. Mean heart rate, oxygen consumption via a Cosmed/K4b2, stride rate, and time
were recorded. The Cosmed/K4b2 was found to have a 2% measurement error, and,
Kawakami et al. (1992) recommended it as a powerful measurement tool for the field
with strong correlation to values obtained from a metabolic cart (McLaughlin, et al.,
2001; Pinnington et al., 2001). To ensure participants ran constantly at 8 min per mile
orange pace cones were placed at each 100 m segment of the track. Each 100 m section
took them 30 s, which the participants checked using a hand-held stopwatch. Stride rates
were determined from a digital video recorder (DCR-TRV, Sony Corporation, Japan)
which filmed at 25 frames-s"1 20 m of the home straight for each lap of their economy
and time trial run. For the 2400 m time trial that followed immediately after the economy
run, mean and maximum heart rate, stride rate, Borg (Borg, 1970) scores and time were
collected. No verbal encouragement was given during the time trial. In trained, non-elite
runners, the coefficient of variation for economy ranges from ±2.64% in 95% of all trials
for field tests of 1600 to 2400 m (Morgan et al., 1991). They concluded a stable measure
of running economy can be obtained in a single data collection session involving trained
non-elite male runners if the testing environment is controlled to minimise non-
biological variability.
The laboratory tests (see appendix 3 for laboratory data collection sheets)
A printed itinerary was given to each participant before their test date covering
procedures and test times. Before the start of the test, each participant was given a verbal
explanation of the test procedures. Every participant warmed up for as long as needed.
All participants wore their own running flats for both pre and post-tests. Participants
were asked to maintain normal weekly training activities. The laboratory pre-tests were
held between May 26th-29th, 2003 and required each participant to run across a force
plate at 3.35 m-s"1 (8 min per mile pace) while simultaneously recording kinematic data.
The post-tests were held between June 9th- 11th, 2003.
169
The intervention (see appendix 4 for intervention sheets)
Both interventions were held between May 30th and June 5th. The Pose® group received
seven one-hour daily intervention sessions in the Pose® method of running technique by
a trained and experienced Pose® instructor. The heel-toe group were to complete seven
intervention sessions of traditional running drills and interval repeats on their own. Both
interventions required the participants to complete a series of specific drills each day.
The heel-toe group were trained to drive the leg forwards and to push-off the ground in
bounding movements using traditional running drills such as butt kicks, high knee lifts
and kick outs. The Pose® group using many specific drills were taught to use their body
weight to create speed through falling forwards while pulling their foot from the ground
using their hamstring muscles.
Collection and processing of ground reaction force data

Biomechanical data were collected using a piezoelectric transducer force plate with eight
output signals, which were sampled at 1200 Hz (9281-CA Kistler Instruments
Corporation, Winterthur, Switzerland). The force plate was mounted flush with the
running surface on a 15.66 m runway. The force plate was 8.66 m from the start position
and measured 40 by 60 cm (figure 6.3.1). Participants were instructed to run normally
while aligning their run-up to land with their right foot on the force plate. Each
participant was allowed as many practice trials as needed to achieve ‘normal' foot
contact with the plate in their regular running shoes, with five trials collected. After each
trial, the data was visually inspected to ensure that the footfall was ‘normal’ and centred
on the force plate. After data collection, the ground reaction force analogue data was
processed through Matlab 6 software to convert volts into newtons. In accordance with
Munro et al. (1987), ground reaction force data were normalised to the participant’s
body weight. Initial contact of the foot with the ground was identified from the ground
reaction force data. Foot contact began once the vertical ground reaction force data
exceeded 20 N and ended as it went below 20 N (Bobbert et a l, 1991; Wright et al.,
1998; Tirosh and Sparrow, 2003).
170
Lab door
F a lc o n H R 2 4 0
5 cam era
sy ste m
H ost
.66 m C o m p u te r
p ro cesso r
P h o t o e le c t r ic and S V G A
Ch
c e lls / s m o n it o r
V id e o
p rocessor
(MIDAS)
K is tle r
4 .1 9 m
F orce 60 cm
3 .0 m P la te
40 cm
7.0 m
Crash mat covering the brick wall
Figure 6.3.1 M e a s u r e m e n t se t-u p fo r la b o r a to r y d a ta c o lle c t io n
Collection and processing of kinematic data (see appendix 5 for raw data)
A modified Helen Hayes marker set (22-markers) was used to collect their three-
dimensional trajectories for computing kinematic data. Retroflective spheres of a
diameter of 2 cm (4.4 g) were applied to their bodies in order to define a model
comprised of twelve body segments: two segments for feet, lower legs, upper legs,
forearm, upper arm and two single segments comprising the head/neck and trunk. These
markers were applied directly onto the skin using athletic tape on the acromion
processes, elbow and radiocarpal joints, sacrum, anterior superior iliac spines, lateral
femoral condyles, lateral malleolus, and the dorsim of the feet between the second and
171
third metatarsals. Two lateral wands were used to calculate the medial knee and ankle
markers during the dynamic trials. The two lateral wands were 10 cm long and attached
using athletic tape to the thigh, midway between the hip and knee joints, and for the
shank, midway between the knee and ankle joints. Both wands were vertically aligned
with the relevant joints. The attachment and length of the wands were such that
oscillations of the marker at the tip of the wands were insignificant. An off-set marker
placed on the right scapula precipitated software marker recognition through body
asymmetry. Video data were collected using a VP AT 310 video recorder and an eight-
camera (Falcon HR 240) motion analysis system recording at 120 Hz (Motion Analysis
Corporation, Santa Rosa, California) which collected data for one step of the right leg
while running in the positive X direction. The cameras were zoomed in as far as possible
to a volume of 3.6 m in x (fore-aft axis) by 1.4 m in y (medio-lateral axis) and 2 m in z
(vertical axis) with the centre of the base corresponding to the centre of the force plate.
The laboratory (global) orthogonal coordinate system followed the right hand rule and
had the positive v-direction orientated in the direction of forward progression, the
positive y-direction orientated to the left and the positive z-direction orientated vertically
upward. The volume was calibrated before data collection using a cube and a 500 mm
wand associated with EvaRT 3.2 data collection software (Motion Analysis Corporation,
Santa Rosa, California). All participants ran a pre-post test at 3.35 m-s'1 in two separate
laboratory sessions eleven-days apart. Running speeds were measured by two
photoelectric cells 4.19 m apart and 3 m from the centre of the force plate, and mounted
so the participant’s waist triggered the photoelectric cells. Up to five trials were
recorded, and only trials in which ‘good’ foot contact with a steady stride and speeds
within 5% of the measured speed were analysed. The three dimensional coordinate data
were filtered using a 2nd order low-pass Butterworth filter (commonly known as the
Butterworth fourth order); a cut-off frequency of 6-8 Hz was used and was selected
through visual inspection of the fit (Winter, 1987). Video and analogue data were time
synchronised using impact recorded when vertical ground reaction force exceeded 20 N.
Following the filtering process in EvaRT 3.2, the filtered video data were extracted using
OrthoTrak 5 software. Coordinate and angular data were calculated according to the
method used by Winter (1990) from the filtered raw data.
Statistical analysis
Assumptions of normality were satisfied for the inferential statistical analysis using the
Ryan-Joiner’s normality test. Homogeneity of variance were satisfied using Levene's
172
test. Individual means and standard deviations were calculated for a variety of secondary
research variables both before and after the instructional period. Sixteen 2 x 2 mixed
factorial ANOVAs assessed the primary research variables for the main effects of group
(control vs. treatment), and treatment (pre to post changes). Tukey’s tests of honestly
significant differences were used to assess individual cell differences post hoc. Data
analysis was conducted using Minitab (version 14) statistical analysis software.
Statistical significance for the inferential statistical procedures was established at P =
0.05.
6 .4 RESULTS
The seven-hour intervention in Pose® running aimed to make consistent biomechanical

changes to eight of the heel-toe participants while the other eight participants (control
group) practiced traditional running drills. Figures 6.4.1-6.4.6 allow visual inspection of
the biomechanical changes for the Pose® and heel-toe group. Heel-toe runners (control
group) pre and post-test illustrate a lack of any biomechanical changes. Pose® runners
show distinct biomechanical changes pre and post-test.
Figure 6.4.1 Pre-test heel-toe runner 1 at 3.35 ms'1taken every 5 frames at 120 frames.s'1.
173
Figure 6.4.2 Post-test heel-toe 1 runner at 3.35 ms'1 taken every 5 frames at 120 frames.s'1.
-
tfTi\.
Iv . <■1
lY.L
iV
t”™
' ■l&
“ T"4 iV " __ _m u
~ 3 " -l T — -"-;rr:
mm
<k
Ilk. t
%
f
■“te:
n
t
M, ...&
Figure 6.4.3 Pre-test Pose® runner 1 at 3.35 ms’1 taken every 5 frames at 120 frames.s’1.
d .
A
■ te
ite
.........
I?
:V.
WmS^l
iteg
t e '“ -
iA te
te a te ::
llte te l
llifl
Slilte
IIS: ite
Figure 6.4.4 Post-test Pose® runner 1 at 3.35 ms'1 taken every 5 frames at 120 frames.s'1.
Figure 6.4.5 Pre-test Pose® runner 2 at 3.35 ms'1 taken every 5 frames at 120 frames.s'1.
Figure 6.4.6 Post-test Pose® runner 2 at 3.35 ms'1taken every 5 frames at 120 frames.s'1.
The primary and secondary variables in Table 6.4.1 give the pre and post-test scores for
the heel-toe and Pose® runners. Mean scores for comparison are given in bold from the
experienced Pose® and heel-toe runners from chapter 5 at the same running speed of 3.35
Table 6.4.1 H e e l- t o e a n d P o s e ® g r o u p 's (N=16) p r im a r y (in b o ld ) a n d s e c o n d a r y m e a n s c o r e s a n d sta n d a r d
d e v ia t io n s p lu s fo r c o m p a r is o n th e e x p e r ie n c e d P o s e R a n d h e e l- t o e te c h n iq u e r u n n e r s at 3 .3 5 m -s"1 in b o ld .
Pose® Heel-Toe
Variables Pre Post Pose® Pre Post Heel-Toe
S T (m s) 2 4 5 (3 ) 2 1 0 (1 )* 220 2 5 0 (2 ) 2 5 0 (1 ) 260
M A - P G R F ( B W ) - 0 .1 8 ( 0 .0 3 ) - 0 .1 8 (0 .0 3 ) -0.23 - 0 .2 1 ( 0 .0 3 ) - 0 .2 2 ( 0 .0 4 ) -0.20
M V G R F (B W ) 2 .6 3 ( 0 .4 ) 2 .6 9 ( 0 .2 4 ) 2.63 2 .5 9 ( 0 .2 2 ) 2 .5 7 (0 .1 6 ) 2.58
M A - P G R F ( p ) ( B W ) 0 .1 6 ( 0 .0 3 ) 0 .1 6 ( 0 .0 2 ) 0.16 0 .1 7 (0 .0 3 ) 0 .1 6 (0 .0 1 ) 0.15
CM (brak) - 2 1 .3 ( 8 .4 ) - 1 6 .2 ( 1 1 .6 ) -15.3 - 1 9 .7 ( 7 .6 ) - 2 0 .0 ( 6 .5 ) -22.6
(cm -s" 1)
CM (prop) 1 7 .1 ( 5 .9 ) 1 4 .1 ( 6 .7 ) 17.7 1 7 .4 (7 .3 ) 1 5 .7 (8 .1 ) 20.4
(cm -s" 1)
CM 25 (cm) 3 2 .7 (6 .3 ) 2 4 .9 (5 .8 )* * 24.9 3 1 .2 ( 3 .1 ) 3 2 .1 ( 2 .3 ) 31.9
V O (c m ) 1 1 .5 ( 2 .3 ) 9 .7 ( 2 .1 ) 9.3 1 1 .4 (1 .5 ) 1 1 .3 ( 1 .6 ) 11.9
K F V (rad-s"1) 6 .1 ( 1 .5 6 ) 5 .2 7 (1 .4 8 )* 5.25 7 .5 2 ( 0 .7 1 ) 6 .9 4 (1 .3 7 ) 5.41
K E V (rad-s"1) 4 .2 2 ( 0 .7 1 ) 3 .7 2 ( 1 .4 7 ) 3.48 4 .1 6 (0 .4 9 ) 4 .1 1 ( 0 .4 1 ) 4.1
K F V S (rad-s"1) 6 .8 9 ( 0 .8 5 ) 7 .9 1 ( 1 .1 4 ) * * 7.65 6 .2 9 (0 .9 7 ) 6 .1 4 ( 0 .8 5 ) 6.42
S L (c m ) 1 0 3 .5 ( 9 .4 ) 8 9 .1 ( 9 .3 ) * * 89.9 1 0 6 .4 (1 0 .6 ) 1 0 0 .7 (7 .9 6 ) 98.4
CM D is ( c m ) 8 7 .3 (1 0 .4 ) 7 2 .7 (5 .4 )* 75.6 8 9 .8 ( 5 .8 ) 8 7 .4 ( 6 .4 ) 89.3
S F (H z) 2 .6 (0 .0 9 ) 3 .0 (0 .2 5 ) * 3.04 2 .6 ( 0 .1 7 ) 2 .5 7 (0 .1 8 ) 2.65
Time trial ( s ) 5 6 8 .1 ( 1 0 .6 ) 5 4 3 .4 ( 7 .1 ) 5 7 0 ( 6 3 .5 ) 5 7 3 .3 ( 6 2 .2 )
02 Con ( m l) 2 9 3 6 .9 ( 2 5 2 .6 ) 2 9 1 8 .1 ( 1 7 4 .8 ) 3 1 6 8 .9 (3 9 9 .9 ) 3 0 1 8 .5 ( 4 4 4 .3 )
T A -P G R F (b ) 2 6 .2 ( 2 .2 ) 1 7 .2 (8 .5 5 ) 8.4 2 3 .8 ( 2 .3 ) 2 3 .8 ( 1 .8 5 ) 23.8
(% o f s ta n c e )
T V GRF 4 4 .8 (3 .9 8 ) 4 4 .0 ( 3 .8 4 ) 37.9 4 4 .1 ( 2 .4 5 ) 4 4 .2 ( 1 .7 3 ) 44.1
(% o f s t a n c e )
T A -P G R F (p ) 7 5 .8 ( 1 .9 9 ) 7 4 .9 ( 2 .2 8 ) 71.2 7 6 .1 ( 2 .1 5 ) 7 4 .8 ( 1 .8 8 ) 73.3
(% o f s t a n c e )
C M S (cm -s" 1) - 2 2 .0 ( 0 .7 ) - 1 5 .2 ( 0 .5 ) -27.3 - 2 8 .7 ( 6 .4 ) - 3 1 .7 ( 6 .4 ) -18.9
C M T O (c m ) 3 7 .0 (5 .4 ) 3 4 .1 ( 4 .6 ) 36.8 4 2 .8 ( 4 .3 ) 4 0 .7 ( 3 .9 ) 40.5
C M 6 (r a d ) 0 .3 3 ( 0 .0 5 ) 0 .3 ( 0 .0 3 ) 0.34 0 .3 8 ( 0 .0 2 ) 0 .3 7 ( 0 .0 3 ) 0.35
S H A (c m ) 1 3 .8 ( 7 .7 ) 8 .7 ( 4 .7 ) 8.0 1 2 .6 ( 3 .0 3 ) 1 3 .5 ( 2 .4 ) 16.1
K F (r a d ) 0 .5 4 ( 0 .0 7 ) 0 .4 3 ( 0 .1 1 ) 0.44 0 .5 8 ( 0 .0 3 ) 0 .5 8 ( 0 .0 4 ) 0.55
K E (r a d ) 0 .5 9 ( 0 .0 8 ) 0 .4 6 ( 0 .1 9 ) 0.48 0 .6 6 ( 0 .0 5 ) 0 .6 3 ( 0 .0 9 ) 0.64
K S W (r a d ) 1 .5 4 ( 0 .2 5 ) 1 .6 (0 .2 4 ) 1.57 1 .4 2 ( 0 .1 8 ) 1 .3 6 (0 .2 2 ) 1.49
K T D (r a d ) 0 .2 2 ( 0 .0 9 ) 0 .2 8 ( 0 .0 9 ) 0.32 0 .2 3 ( 0 .0 7 ) 0 .2 4 ( 0 .0 7 ) 0.34
K V G R F (r a d ) 0 .7 1 ( 0 .1 2 ) 0 .6 8 ( 0 .1 1 ) 0.74 0 .7 7 ( 0 .0 6 ) 0 .7 8 ( 0 .0 8 ) 0.87
F T (m s) 1 3 6 (4 ) 126 13) 120.0 1 3 4 631 1 3 0 (2) 120.0
S ig n if ic a n t d if f e r e n c e (P = 0 .0 1 ) fr o m * P o s e 16 p r e to p o s t - t e s t . S ig n if ic a n t d if f e r e n c e (P = 0 .0 5 ) fr o m * *
P o s e * p r e to p o s t - t e s t . N o s ig n if ic a n c e w a s f o u n d p r e -p o s t t e s t fo r th e c o n tr o l g r o u p o r b e t w e e n p r e - te s t s
fo r th e P o s e * a n d c o n tr o l g r o u p s .
176
Experienced Pose^ and heel-toe runner’s scores are in bold as well as the training study’s primary research
variables. Standard deviations are in parenthesis. ST = stance time (ms) right foot, M A-P GRF (brak) =
maximum posterior ground reaction force (braking) (BW), M V GRF = maximum vertical ground reaction
force (BW), M A-P GRF (prop) = maximum anterior ground reaction force (propulsive) (BW), CM (brak)
= centre of mass horizontal velocity largest decrease after impact until maximum vertical ground reaction
force (cm-s'1), CM (prop) = centre of mass horizontal velocity largest increase from maximum vertical
ground reaction force until terminal stance (cm-s*1), CM 25 = centre of mass behind the toe-marker
(horizontal difference at 25 ms), VO = vertical oscillation of the centre of mass (cm), KFV = mean knee
flexion angular velocity for stance (rad-s"1), KEV = mean knee extension angular velocity for stance (rad-s'
l), KFVS = mean knee flexion angular velocity from terminal stance to maximum swing angle (rad.s'1), SL
= step length for right foot (cm), CM dis = centre of mass displacement during stance (cm), SF= step
frequency (Hz), Time trial (s), 02 con = oxygen consumption (ml), T A-P GRF (brak) = time of posterior
ground reaction force (braking) (% of stance), T V GRF = time of maximum vertical ground reaction
force (% of stance), T A-P GRF (prop) = time of anterior ground reaction force (propulsion) (% of
stance), CM diff = horizontal velocity of the centre of mass from maximum vertical ground reaction force
until terminal stance minus horizontal velocity of the centre of mass from impact until maximum vertical
ground reaction (cm-s*1), CM S = centre of mass horizontal velocity decrease between terminal stance and
maximum swing (flexion) for the right leg during flight (cm-s'1), CM TO = centre of mass forward of the
toe-marker (horizontal difference at terminal stance) (cm), CM 0 = centre of mass and toe-marker angle at
terminal stance from the vertical axis going clockwise (rad), SHA = shoulder, hip, and ankle marker
difference in vertical alignment at 25 ms (cm), KF = knee flexion range during stance (rad), KE = knee
extension during stance (rad), KSW = knee flexion from terminal stance to maximum swing (rad), KTD =
knee angle at impact (rad), KV GRF = knee angle at maximum vertical ground reaction force (rad), FT =
flight time (ms).
Figures 6.4.7-8 give pre and post-test mean scores for the Pose® and heel-toe groups for
time-aligned hip, knee and ankle joint displacements with maximum vertical ground
reaction force and vertical oscillation of the centre of mass. Note, the horizontal
displacement of the centre of mass during flight for the Pose® group post-test is 44.9 cm
and the heel-toe group is 46.2 cm, whereas step length decreased in the Pose® group
post-test.
177
Figure 6.4.7 Pre (left) and post-test (right) control group (heel-toe) mean scores for ground reaction force,
hip, knee, ankle and vertical oscillation of the centre of mass.
178
/
J
Figure 6.4.8 Pre (left) and post-test (right) Pose® group mean scores for ground reaction force, hip, knee,
ankle and vertical oscillation of the centre of mass.
179
The following tables (tables 6.4.2-6.4.6) cover the economy run and the time trial. The
London Met. office had no weather stations at Loughborough University so the two
nearest stations were used to record the weather for the economy run and the time trial.
Table 6.4.2 Economy run (2400 m) (A = 16)

Pose® Heel-toe
Variables Pre-test Post-test Pre-test Post-test
AV HR 169.9 (10.5) 166.9 (7.1) 170.0 (22.0) 167.6 (20.5)
SF 15.7 (0.59) 16.6 (1.5) 15.4 (0.74) 15.6 (0.92)
o2 2936.9(252.6) 2918.1 (174.8) 3168.9(399.9) 3018.5(444.3)
AV HR = mean heart rate (bpm), SF = Step frequency (mean for the 6 laps for the steps taken from a 20 m
stretch in the home straight), 02 = Economy run (oxygen consumption) (ml)
Table 6.4.3 Time trial run (2400 m) ( N = 16)

Pose® Heel-toe
Variables Pre-test Post-test Pre-test Post-test
T 568.1 (10.6) 543.4 (7.1) 570.0 (63.5) 573.3 (62.2)
M HR 194.0 (9.9) 194.6 (6.9) 195.9(13.6) 193.1 (11.3)
AV HR 184.9 (9.58) 185.6 (7.2) 183.1 (15.4) 181.0(13.3)
Borg 17.5 (0.53) 18.1 (0.99) 16.8 (0.71) 17.0 (1.41)
SF 12.7(1.1) 13.4(1.5) 13.0 (1.34) 13.8 (1.51)
T = Time (s), M HR = Maximum heart rate (bpm), AV HR = mean heart rate (bpm), Borg = Borg Scale
(6-20), SF = Step frequency (mean for the 6 laps from the steps taken from a 20 m stretch in the home
straight).
Table 6.4.4 Questionnaire (scale 1-5) How important do you believe running technique is to overall
performance? ( N = 16)
Pose® Heel-toe
pre-test post-test pre-test post-test
5 5 4 4
3 4 5 5
4 4 5 5
4 5 5 4
5 5 5 4
4 5 4 3
5 5 4 4
5 4 4 4
Mean 4.4 4.6 4.5 4.1

S.D. 0.74 0.52 0.53 0.64
1 = not important at all; 2 = fairly unimportant; 3 = in between; 4 fairly important; 5 = extremely important
180
Table 6.4.5 Weather on time trial and economy run days from Sutton Bonnington.
Date WN ws W Gust (N) W Gust Temp Hum Rain

May-24 268 7 266 14 13.7 82.8 0.0
May-25 294 8 295 16 14.7 78.6 0.0
Jun-06 209 9 207 17 16.8 85.3 0.1
NGR = 4507E 3259N, Altitude = 48 metres, Latitude = 52:83 N Longitude = 01:25 W (London Met.
office ©, 2003) W N = Wind - Mean Direction Degrees True Clockwise from North, W S= Wind - Mean
Speed (kn), W Gust (N) = Wind - Max Gust Direction Degrees True Clockwise from North, W Gust =
Wind - Max Gust Speed (kn), Temp = Temperature - Dry Bulb (°C), Hum = Relative Humidity %, Rain =
Rain - Amount (mm).
Table 6.4.6 Weather on time trial and economy run days from Leicester University.
Date WN WS W Gust (N) W Gust Temp Hum Rain
May-24 pre-test 245 5 265 12 14.2 80.4 0.0
May-25 pre-test 305 7 295 15 14.9 76.7 0.0
Jun-06 post-test 192 6 195 14 16.4 88.3 0.1
NGR = 4595E 3032N, Altitude = 80 metres, Latitude = 52:62 N Longitude = 01:12 W (London Met office
©, 2003) W N = Wind - Mean Direction Degrees True Clockwise from North, W S= Wind - Mean Speed
(kn), W Gust (N) = Wind - Max Gust Direction Degrees True Clockwise from North, W Gust = Wind -
Max Gust Speed (kn), Temp = Temperature - Dry Bulb (°C), Hum = Relative Humidity %, Rain = Rain -
Amount (mm).
Statistical Analysis
Statistical tests were performed with a commercial statistical programme Minitab
(version 14). Significant differences between the means o f the primary research variables
were found using a 2 * 2 mixed factorial ANOVA, with the level o f significance posted
at P < 0.05. Significant differences between the means as indicated by significant F-
ratios were explored with the Tukey post hoc analysis test. Significant interaction effects
were found for stance time (P = 0.012), centre of mass to support limb at 25 ms (P =
0.014), centre of mass displacement during stance (P = 0.030) and step frequency (P =
0.005). Tukey post hoc tests revealed significant differences between the Pose® group
pre and post-test for stance time (P = 0.001), centre of mass to support limb at 25 ms (P
= 0.042), centre of mass displacement during stance (P = 0.001), knee flexion angular
velocity during stance (P = 0.005) plus swing (P = 0.043) and step frequency (P =
0 .002).
181
A three-month prospective injury report was undertaken post intervention for the sixteen
participants. Their self-report data is summarised. It was not possible to control the
participants rumiing training owing to an inability to maintain personal contact (the
researcher lived 6000 miles away). Each participant was asked to keep a diary of training
mileage and injury incidence. All the participants (except two who ceased rumiing)
maintained their normal weekly mileage (mean 25-km).
Table 6.4.7 A ll b u t t w o p a r tic ip a n ts c o n t in u e d r u n n in g th e ir n o r m a l w e e k ly s c h e d u le s . A s e lf-r e p o r t in ju r y

stu d y w a s c o n d u c t e d fo r t h r e e -m o n th s p o s t - t e s t (N = 1 6 )
Pose® group Heel-toe group

N o in ju r ie s in th e s ix r u n n e r s F o u r d e v e lo p e d p a t e llo f e m o r a l p a in
w h o c o n t in u e d to ru n P o se® . sy n d ro m e.
T w o r u n n e r s ( w h o le a s t le a r n e d P o se ® r u n n in g ) O n e d e v e lo p e d s u b c a lc a n e a l p a in
r e v e r te d b a c k to th e ir fo r m e r h e e l- t o e sy n d ro m e.
t e c h n iq u e a n d d e v e lo p e d ilio t ib ia l b a n d O n e d e v e lo p e d c h r o n ic c o m p a r tm e n t
fr ic tio n s y n d r o m e . sy n d r o m e in th e ir le g s .
T w o c e a s e d r u n n in g b u t w e r e n o t in ju r e d .
6 .5 DISCUSSION
The major findings of this present study on Pose® running were as follows: 1) There
were no significant differences in ground reaction forces in the Pose® group between pre
and post-test. 2) There were significant differences in body and limb activity in the
Pose® group between pre and post-test. 3) There was no significant change in economy
in the Pose® group’s economy run, yet they ran faster in the 2400 m time trial. 4) No
injuries occurred in the Pose® group who continued to run the Pose® technique, whereas
all of the heel-toe technique that continued to run developed injuries. However, owing to
the small sample size conclusive comment about injury incidence is not possible at this
time.
A'endse et al. (2004) administered a Pose® method intervention to twenty male and
female endurance runners using a repeated measures experimental design (within-
subjects). Biomechanical data were collected for their natural heel-toe running style.
During the same data collection, they ran on their forefoot while the same biomechanical
data was collected. Pose® running instruction was then administered, which consisted of
a maximum of seven and half hours and comprised of one and half-hour sessions daily
for five consecutive days. Seven and half hours of Pose® instruction from a trained Pose®
182
method teacher was adequate to change their heel-toe runners into Pose^ runners. This
study therefore employed a similar seven-hour Pose® instruction over seven-days.
Previous attempts to modify running technique have focused on one or two (stride
length/frequency) modifications (for example, Petray and Krahenbuhl, 1985; Messier
and Cirillo, 1989; Williams and Ziff, 1991). Although these studies recorded individual
kinematic variables— stride length, vertical oscillation, trunk inclination and arm
swing— their results were inconsistent across their sample groups, showing a resistance
to change from their individual running technique. In contrast, Pose® running produced
consistent biomechanical changes (see table 6.4.1; Arendse et al., 2004; Dallam et al.,
2005). This study seeks to evaluate these biomechanical changes in the Pose® group (no
changes were found in the heel-toe group) in relation to a performance test, economy run
and injury incidence.
Kinetic and kinematic (primary and secondary) variables

The 2 x 2 mixed factorial ANOVA found all the ground reaction force variables non
significant. Lack of change in the ground reaction force makes sense. Chang et al. (2000)
increased and decreased gravity on eight male endurance runners. They also increased
inertia without changing gravity. Horizontal impulses, maximum braking and propulsive
ground reaction force changed substantially with increased gravity compared to inertia.
With increased gravity, the braking and propulsive impulses increased proportionally,
whereas with inertia they did not. The vertical ground reaction force did not change with
increased inertia, but when gravity reduced, the vertical ground reaction force reduced
proportionally. This confirms gravity and maximum vertical ground reaction force (the
runner’s weight) strongly relates with impact and is not propulsive (discussed in detail in
chapter 5; Biewener et al., 2004).
Propulsive (horizontal) ground reaction force did not change in the Pose® runners, while
muscle activity is absent in all runners during the propulsive phase of stance
(Montgomery et al., 1994), indicating the runner is not pushing off the ground with their
muscle system (McClay et al., 1990). The lack of muscle activity, maximum horizontal
acceleration occurring before maximum horizontal ground reaction force (see chapter 5
table 5.4.1), and changes in ground reaction force that are proportional with changes in
gravity (Chang et al., 2000) lead to the deduction that a gravitational torque creates
propulsion (see chapter 3, 5) rather than the ground and the muscle system.
183
However, propulsive impulse increases with running speed (Hunter et al., 2005). Using
multiple linear regression to predict sprint velocity, relative propulsive impulse
explained 57% of then variance in sprint velocity . The impulse-momentum relationship
correctly explains motion changes in runners, but as running speed rises, the propulsive
impulse increases do not have to come from more muscle activity, because the leg and
hip extensors are silent during propulsion (McClay et al., 1990; Montgomery et al.,
1994). Rather an increase in the propulsive impulse may occur via a gravitational torque
increasing the acceleration of the runner’s centre of mass. At faster running speeds, the
increase in trunk inclination will subsequently increase the gravitational torque via an
increased moment arm. The propulsive impulse may simply be reflecting the increased
acceleration of the centre of mass via the increased gravitational torque at higher running
speeds.
The timing of the ground reaction forces only differed in the braking condition for the
Pose® runners. This may reflect their more aligned shoulder, hip and ankle at impact
after the intervention (5.1 cm) and the significant difference found in their centre of mass
being more vertically aligned (7.8 cm) with the support foot at impact. Although not
significantly different, the Pose® group had a lower braking effect (0.051 m-s'1) for the
horizontal velocity of their centre of mass during this time possibly reflecting a more
aligned body position. As the Pose® group were faster post-test for the time trial this
more aligned body position may reflect the cause of improved performance, because
faster runners brake less (Bates et al., 1979; Girardin and Roy, 1984; Hinrichs, 1990).
Horizontal velocity of the centre of mass during the propulsive phase of stance was also
not significantly different in the Pose® group although it did decrease. It is possible that
the Pose® group did not need to accelerate as much during this phase to maintain a
constant running speed because they braked less. Another possible explanation is the
ipsilateral leg is driven forwards during this part of stance in heel-toe runners, which
would increase the horizontal velocity of the centre of mass. Pose® runners are trained
not to drive the swing leg forwards. Finally, calculation of the change in horizontal
velocity of the centre of mass (difference between maximum vertical ground reaction
force and terminal stance), may have masked increased acceleration in the Pose® group,
because the experienced Pose® runners in chapter 5 had higher horizontal acceleration of
their centre of mass (chapter 5; figures 5.4.58-61).
184
The Pose® group reduced their vertical oscillation in-line with other Pose® studies
(Arendse et al., 2004; Dallam et al., 2005). Reduced vertical oscillation appears to be a
distinguishable characteristic of Pose® running. Less knee extension correlates with
lower vertical oscillation (Wank et al., 1998). The Pose® group reduced knee extension
(0.13 rad) and vertical oscillation (1.8 cm). Stance time also significantly decreased in
Pose® runners, suggesting having the support foot remain longer on the ground
(increased stance time) via increased leg extension potentially increases vertical
oscillation. Increased stance time correlated positively (r = 0.49) with poorer economy
reflecting decreased performance (Williams and Cavanagh, 1986; Williams and
Cavanagh, 1987). In addition, as endurance runners fatigue, stance time increases (Nicol
et al., 1991). Directly the body begins to fall forwards (after mid-stance), Pose® runners
are taught to pull their foot from the ground, which potentially decreases the knee
passively extending, thus reducing vertical oscillation. Pose® running is characterised by
decreased vertical oscillation, which may correlate with reduced stance time and lower
knee extension possibly affecting performance. Enomoto et al. (1999) using thirty-two
participants found their fastest runners produced lower vertical oscillation and less
braking of the centre of mass’s horizontal velocity. This study’s Pose® runners improved
their time trial performance eliciting these same traits.
The Pose® group significantly reduced knee flexion angular velocity at impact to
maximum vertical ground reaction force. This possibly reflects their more aligned body
at impact, because their knees flex less from impact to maximum vertical ground
reaction force (0.08 rad). Knee extension angular velocity from maximum vertical
ground reaction force until terminal stance decreased. The Pose® group flex and extend
the knee less during stance, which potentially translates into less work and improved
performance (Arendse et al., 2004). The knee flexion angular velocity for swing in the
Pose® group post-test significantly increased, whereas knee range of motion barely
changed (0.06 rad). The reduced moment of inertia at the knee (less extension) during
stance and the significantly increased knee flexion angular velocity during swing may
enhance performance based on lower energy cost (Enomoto et al., 1999).
The Pose® group reduced step length and significantly increased step frequency creating
another recognisable characteristic of Pose® running (Arendse et al., 2004; Dallam et al.,
2005). Increased vertical alignment of the shoulder, hip and ankle in the Pose® group at
185
impact, may suggest reaching or driving the swing leg forwards in front of the body, is
absent in Pose® runners. Driving the swing leg will increase step length, and may
possibly affect the timing of the braking ground reaction force. Support for driving the
swing leg forwards in heel-toe runners is exhibited by lower differences for horizontal
displacement during flight between the centre of mass and actual step length for the
Pose® runners.
Economy and time-trial runs

Dallam et al. (2005) found a mean increase in sub-maximal absolute oxygen cost of 3.28
(±0.36) Lmiin"1to 3.53 (±0.43) L»min-1, P < 0.01 in their Pose® trained group. However,
this study found no significant change in economy. Differences between the two studies
methodologies may explain the conflicting results. In contrast, to Dallam et al. (2005)
this study employed an over-ground economy run test of 2400 m at 3.35 m-s'1 when the
participants were in a non-fatigued state. Dallam et al. (2005) tested economy
immediately after a maximal oxygen consumption test on a treadmill, leaving the Pose®
runners in a very fatigued state. Use of a treadmill may also affect Pose^J running.
Consistent increases in submaximal oxygen cost occur with short term attempts to
manipulate running economy by altering the self-selected stride rate or length (Cavanagh
and Williams, 1982; Kaneko et al., 1987; Messier and Cirillo, 1989; Cavagna et al.,
1991). This is the first study in running to alter self-selected stride rate and length and
not subsequently increase oxygen consumption. A possible reason for the difference is
the Pose® method produces other biomechanical changes as well as changes in stride
length and frequency (Table 6.4.1). Previous studies do not address how stride length
and stride frequency occur, because they require participants to manipulate their own
changes. The Pose® method instruction attempts to teach runners that stride length is a
consequence of the body’s displacement (via a gravitational torque) not a cause of it.
Stride frequency results from the need to pull the foot from the ground to keep up with
the moving body while the swing leg is not driven forwards. Finally, the swing foot in
the Pose® runners lands under the body identified by a significant change in the vertical
alignment of the centre of mass over the support foot at impact.
In one exception to the above, Morgan et al. (1994) found that step length optimization
training among runners who had self selected a stride length that was longer than that
predicted as most economical resulted in improved economy. By testing these runners at
186
multiple stride lengths, they determined which was most economical. Approximately
20% of the runners tested demonstrated a self-selected stride length that was not the
same as their most ideal economical stride length. Consequently, the runners improved
their economy by training for three weeks (Monday-Friday for 30 min) with audiovisual
feedback that allowed them to reproduce the reduced stride length determined to be their
most economical. Morgan et al. (1994) suggested the longer stride lengths were less
economical owing to reduced elastic mechanisms and an increase in postural
maintenance. In contrast, Williams et al. (1991) reported lowered predicted economy
across a series of relative stride lengths (± 20% of leg-length) after 3-weeks of training at
a non-optimal stride length condition (+10% of leg-length from their freely chosen stride
length). They suggested improved economy results from favourable biomechanical
changes that were not quantified. This study quantified biomechanical changes in the
Pose® group and in agreement with Morgan et al. (1994) posture significantly altered.
For example, at impact the Pose® group became more vertically aligned at the ankle, hip
and shoulder, their centre of mass displaced less during stance and they had lower
vertical oscillation. They left the ground with a less extended knee and were faster to flex
the knee during early swing.
Evidence of a similar response may be found in cycling studies that examined pedalling
rate, oxygen cost and cycling efficiency. Chavarren and Calbet (1999) demonstrated the
most economical cycling cadences in experienced road cyclists (60 rpm) are lower than
cyclists commonly self-select for training and racing (90-105 rpm). They attributed this
phenomenon to a small but significant increase in delta efficiency with increased
cadence. The change in muscular activation patterns resulting from movements at a
higher frequency at the same power output, while cycling or running, may contribute
positively to reducing peripheral fatigue mechanisms by reducing local leg fatigue later
in the race. The participant feedback (Appendix 4) in this study indicated the Pose®
runners felt reduced fatigue in their legs, which they concluded was significant in
improving their performance.
Curvilinear increases in aerobic demand occur when stride length increases or decreases
at any given speed from one self-selected by the runner (Hogberg, 1952; Knuttgen, 1961;
Morgan et al., 1989; Kyrolainen et al., 2000a). The optimal economical solution
identified was a combination of self-selected stride lengths and stride frequencies. The
Pose® group self-selected their stride lengths and stride frequencies, which were
187
significantly altered from their original rumiing technique. Pose1*" training does not teach
or emphasise changing stride lengths and stride frequencies. The Pose® technique
teaches the runner to fall forwards from mid-stance and pull the support foot from the
ground without regard for changing stride length and stride frequency. As the runner
follows these commands their stride length and stride frequency are a consequence not
the cause of running gait. That is, the speed the body leaves the ground from falling
determines stride length. Stride frequency results from pulling the support foot faster
from the ground to keep up with a faster moving body as rumiing speed increases. The
Pose® method’s novel focus on body posture (rumiing Pose®) and a precise action of
pulling the support foot from the ground may explain economy differences between this
study and all other studies.
Other variables identified with economical running such as maximum thigh extension at
terminal stance had a positive correlation (r = 0.53) with increased economy (Williams
and Cavanagh, 1986) although this is not a consistent finding (Williams and Cavanagh,
1987). Miura et al. (1973) associated thigh extension with increased push-off at terminal
stance, identifying this as ‘good' and associated with their elite runners, however, this
increases the moment of inertia at terminal stance, which increases work (Fenn, 1930).
Cavanagh et al. (1977) found more acute knee angles during the swing phase enhanced
economy suggesting a reduced moment of inertia is beneficial (Bailey and Pate, 1991).
The Pose® group reduced knee extension at terminal stance and increased significantly
their knee flexion angular velocity from terminal stance to maximum swing. There is
general support for improved rumiing economy with lower vertical oscillation (Slocum
and James, 1968; Cavanagh, et al., 1977; Williams and Cavanagh (1987). The Pose®
group reduced vertical oscillation, which correlates with reduced knee extension (Wank
et al., 1998), possibly suggesting driving off the ground is not economical. Finally,
Nilsson and Thorstensson (1987) reported no significant difference in running economy
between a group of six forefoot and six rear foot strikers potentially negating changes in
foot contact in the Pose® group had any affects on economy.
Increased heart rate and ventilation correlated strongly with inter-individual variation in
running economy (Pate et al., 1989), indicating lower heart rate and ventilation
correlated with improved economy. Increased ventilation and heart rate are associated
with stress, often caused by change (Hahn and Payne, 1999). Despite the Pose® group
changing their biomechanical profiles they did not worsen economy, in fact, mean heart
188
rate reduced by three beats post-test. In Dallam et al. (2005), fatigue and running on a
treadmill, which they had not trained for in their Pose® intervention, increased minute
ventilation and heart rate and thus worsened economy.
Despite not improving running economy all the Pose® group ran faster in the time trial
with a mean improvement of 24.7 s for the post-test, with the heel-toe group differing by
3 s. Interestingly, maximum and mean heart rate were almost identical in the Pose® and
heel-toe groups pre and post-test. One Pose® runner was 56 s quicker for his post-test run
with a mean heart rate 3 bpm lower than his pre-test score while having the same
maximum heart rate. His Borg scale scores were 16 post-test and 17 pre-test. Borg scale
scores did not change in the Pose® and heel-toe groups, pre and post-test. It is evident
that improvements over one-week were not owing to physiological adaptation, but reflect
the significant biomechanical changes post-test in the Pose® group.
The prospective injury study

The questionnaire asked, “How important do you believe running technique is to overall
performance?” The Pose® group increased by 4% their belief that technique was
important, whereas the heel-toe group reduced their belief by 8%. These shifts are small
and probably insignificant because both groups still regarded technique as fairly too
extremely important. Regarding technique as important does not necessarily predispose a
runner to less injury. Interestingly, reviews on running injuries consistently neglect to
mention technique (and changes) as a possible method of injury reduction or even as a
cause (for example, Yzerman and Van Galen, 1987; van Mechelen, 1995; Taunton et al.,
2002). All participants recorded their running training plus injuries sustained over a
three-month period post intervention, while maintaining their normal running routine.
The prospective injury study revealed that those runners who continued to run Pose® did
not injure in the proceeding three-months after the intervention. However, the two Pose®
runners who returned to heel-toe running possibly because they found it hard to maintain
Pose® running through practice drills, both developed iliotibial band friction syndrome.
Six of the heel-toe runners developed injuries while the two without injuries had ceased
running altogether (Table 6.4.7). Therefore, the self-report data collected from the post
test three-month injury study identified six overuse injuries in the heel-toe group and two
overuse injuries in the Pose® group. Renstrom (1994) states overuse symptoms usually
occur over time, making it difficult to determine the onset and specific cause. If
incidence is calculated in relation to time, it varies from 2.5 to 12.1 injuries per 1000
189
hours of running (Lysholm and Wiklander, 1987; van Mechelen, 1992). This studies
injury rate was 12.5 injuries per 1000 hours for the heel toe group and 4.2 injuries per
1000 hours for the Pose® group. However, the two injured runners in the Pose® group
reported they reverted to heel-toe running owing to difficulties in maintaining the Pose®
technique, essentially leaving no injuries in those that remained running Pose®.
Most rumiing injuries are lower extremity injuries, with the knee joint being the main
site of injury (van Mechelen, 1992; Stanish and Wood, 1994). In support of this finding,
four heel-toe and two Pose^ runners who reverted to their heel-toe running had knee
injuries. Therefore, the knee joint was the main site of injury in this study. Eccentric and
concentric muscle contractions in a stretch-shortening cycle are potential components for
overuse injury at the knee (Johansson, 1992). He suggested knee extensor muscle
loading result from variations in running technique. Arendse et al. (2004) found knee
power absorption and knee eccentric work were significantly less in Pose® runners
compared to running heel-toe. Muscle strains usually occur during eccentric contractions
(Glick, 1980). Ankle power absorption and eccentric work were greater in Arendse et
n/.’s Pose® runners compared with their heel-toe running. Their knee power generation
and concentric work were less in Pose® rumiing compared to their heel-toe running,
however, there were no differences in ankle power generation and concentric work
between the running techniques. In-line with Arendse et al. (2005), this study’s Pose®
runners may have had less eccentric work and power absorption at the knee, however
larger injury studies will confirm whether the Pose® technique significantly reduces knee
injuries or possibly increases ankle injuries.
Vertical impact force has been strongly associated with the transmission of shock waves
upward through the musculoskeletal system (Radin and Paul, 1971; Verbitsky et a l,
1998). Several investigators identified a positive relationship between the magnitude of
impact force and the numbers of overuse injuries in runners (Voloshin and Wosk, 1982;
Wosk and Voloshin, 1982; Cavanagh, 1990). Vertical impact force was similar between
the two techniques in this thesis. However, the time of the maximum horizontal braking
force was earlier in the Pose® runners coupled with a more vertically aligned body
position (shoulder, hip and ankle joints). Pose® runners are trained to land in this body
position because the ‘S’ like position is meant to reduce the effects of impact forces. The
‘S’ like position did not reduce maximum vertical impact force but maximum horizontal
braking force occurs earlier suggesting a change in the ground reaction force vector.
190
Stergiou et al. (1999) support this study, in that they identified coordination patterns
between the ankle and knee joints change the ground reaction force vector. Overall, these
studies suggest lower-limb activity at impact affects the loading of the body.
Running volume is an important risk factor for increased injury incidence, whereas,
running velocity is not associated with injury risk (Yzerman and Van Galen, 1987). The
mean weekly mileage was 25-km for both groups (before and after the intervention),
however, running intensity was not recorded. During a 10-km race, eight elite runners
were filmed at different stages of the race (Elliot and Ackland, 1981). Fatigue produced a
more extended lower-limb throughout the gait cycle (negative impact on performance)
during the later stages of the race. Therefore, differences in landing posture between the
two techniques in this study may vary gravity’s work on the runners. For example, centre
of mass location to the support limb at impact may influence the impact force, owing to
Arendse et al. (2004) finding Pose® running reduced eccentric work at the knee, while
increasing eccentric work at the ankle. Hunter et al. (2005) found at impact that the
distance between the support foot and the centre of mass was lower for runners with a
reduced braking impulse. Impact force did not change in this study but the time it
occurred did for the Pose® group, while impact distance of the support limb to the centre
of mass reduced significantly after the PoseR intervention. The timing of braking force
was earlier in the Pose* group post intervention possibly reflecting the more aligned
body position at impact, which would affect the braking impulse and may be related to
reduced injury incidence.
Using a 1-metre depth jumping exercise as an exaggerated motion for landing while
rumiing, ten healthy males landed on their heels or forefeet (Kovacs et al., 1999). Total
torque contributions were highest for heel-toe landing (during the extension phase) with
41% and 45% attributed to the knee and ankle joints and 34% and 55% with the forefoot.
Electromyography for vastus lateralis muscle activity differed during re-contact for the
heel-toe landing, whereas gastrocnemius muscle activity in the forefoot group increased.
Exaggerated motion (depth jumps) highlights greater loading characteristics between
heel and forefoot contact, which in light of previous comments, suggests limb and body
positions relationships play important roles in force attenuation and injury incidence.
This study’s sample is obviously too small to draw any conclusions about Pose® running
reducing injury, except to mention the results are encouraging in that Pose® technique
191
changes may reduce injuries. Therefore, large retrospective and prospective injury
studies should be conducted.
6 .6 SUMMARY
A 2 x 2 mixed factorial ANOVA post hoc found significant differences between the
Pose® group pre and post-test for stance time (P = 0.001), centre of mass to support limb
at 25 ms (P = 0.042), centre of mass displacement during stance {P = 0.001), knee
flexion angular velocity during stance (P = 0.005) plus swing (P = 0.043) and stride
frequency (P = 0.002). It was suggested, posterior and anterior ground reaction force did
not change because of its strong correlation with gravity, a constant force. It appears
seven-hours of instruction in Pose® running is an adequate time to significantly make
consistent biomechanical changes in male recreational endurance heel-toe runners, which
matched other Pose® runners. Pose® runners ran faster after the intervention for a 2400 m
time trial, while heart rate, and perceived exertion did not change. In addition, economy
did not change for the Pose® group post-test in the 2400 m economy run. The improved
running speed for the Pose® group in the time trial was attributed to the significant
biomechanical changes in the Pose® technique rather than any physiological adaptations,
owing to the short intervention time. It was suggested that the running Pose® (vertical
alignment of the shoulder, hip and ankle at impact) improved perfonnance via reduced
braking of the horizontal velocity of the centre of mass. Potentially, after mid-stance, as
the Pose® runners fell forwards they made better use of the gravitational torque through
lower leg extension and vertical oscillation. A faster leg recovery during swing enabled
the leg to keep up with the body but without the need to drive the swing leg forwards of
it, which decreases horizontal velocity of the centre of mass at impact. Stride length and
stride frequency therefore, were a consequence of running Pose®, not the cause. A
preliminary prospective three-month injury report revealed lower injury incidence in the
Pose® runners as opposed to the heel-toe group.
192
CHAPTER 7
CONCLUSIONS
Research into endurance running performance has focused on three main areas:
economy, optimal running biomechanics, and injury mechanisms. However, this
research has not provided a universal running technique, except to identify 80% of
endurance runners land on their heels. Hay and Reid’s (1988) hierarchical model on
running reflects propulsion in running occurs via pushing off the ground through leg
extension. Alternatively, Pose® running claimed to be a universal technique that was
economical, having an optimal biomechanical profile while reducing injuries. Therefore
a new Gravitational hierarchical model of running based upon the Pose® technique was
developed, which clarified the interaction of the forces involved in running by
establishing gravity as the motive and hierarchically leading force in running.
Experienced Pose® runners were then compared with experienced heel-toe runners
(chapter 5; study 1). The study confirmed the Gravitational hierarchical model explains
the interactions of the forces involved in running while clearly distinguishing heel-toe
and Pose® running. The Pose® method study (chapter 6; study 2) replicated the same
biomechanical profile as the experienced Pose® runners by training eight heel-toe
recreational endurance runners in Pose®. The Pose® group improved performance
without changes in economy while reporting no injuries in those who continued running
Pose®.
7.1 THEORETCIAL MODELS
There are three known forces (neglecting air resistance) involved in running (gravity,
ground reaction force, and muscle force) plus potential strain energy. Theoretical support
for the Pose® method and the new Gravitational hierarchical model were identified from
the literature.
193
Ground reaction force is not a motive force but operates according to Newton's
third law (Zatsiorsky, 2002; p.51).
The ground can only propel a runner’s centre of mass forward in combination
with muscle activity (Zatsiorsky, 2002; p.51). Leg and hip extensor muscles have
consistently proven to be silent after knee extension begins (Branded, 1973;
Mann and Hagy, 1980; Pare et al., 1981; Schwab et al., 1983; Nilsson and
Thorstensson, 1985; MacIntyre and Robertson, 1987; Montgomery et al., 1994;
Kyrolainen et al., 2003) suggesting the ground in combination with the muscle
system are not propulsive.
High muscle-tendon forces at terminal stance suggest elastic recoil is doing work
against gravity rather than muscle force during leg extension (Cavagna et al.,
1964; Cavagna and Kaneko, 1977; Hinrichs, 1987; Alexander, 1992; Taylor,
1994; Jung, 2003).
Elastic recoil is predominantly in the vertical direction (Farley and Ferris, 1998).
A flic-flac movement shows a spike at terminal stance in the ground reaction
force from a push-off action reflecting leg extensor activity, which is completely
absent in runners (Yeadon and Challis, 1994). In fact, the ground reaction force
decay rate increases as ruiming speed increases, owing to the higher acceleration
of the centre of mass as it leaves support (Miller, 1990). Therefore, the decay rate
is reflecting the body rising from the ground and does not support a pushing off
from the lower-limb musculature.
Gravity (via a gravitational torque) is the only external force that can horizontally
displace the centre of mass if muscle activity is silent during leg extension.
Gravity causes a torque around the support foot as the centre of mass passes
maximal vertical ground reaction force until terminal stance, creating a horizontal
displacement of the centre of mass as the body falls forwards (Gracovetsky,
1988).
The horizontal ‘propulsive’ ground reaction force may reflect the gravitational
torque’s propulsive component, because in reduced (or increased) gravity,
horizontal propulsive ground reaction force changed proportionally (Chang et al.,
2000 ).
The hip’s extension coincided with reducing vertical ground reaction force
(Payne, 1983). The hip passively extends via a gravitational torque as the body
moves past the stationary foot, because hip extensors are only active at mid-
stance and late swing while hip flexors are mainly active in early swing
194
(Montgomery et al., 1994). This again suggests the body is moving forward of
the support limb (via a gravitational torque) and not being pushed forwards from
the foot/ground via the ankle and knee joints (Fenn, 1930; McClay et al., 1990;
Montgomery et al., 1994; Pink et al., 1994; Belli et al., 2002; Hunter et al.,
2005). This is in direct contrast to Hay and Reid (1988; pp. 264 and 287-288)
who suggest leg extension via ankle, knee and hip musculature is the motive
force in rumiing.
• Driving the swing leg forwards causes the foot to land ahead of the body. This
potentially reduces horizontal velocity of the centre of mass by increasing the
horizontal distance of the centre of mass from the support foot at impact (Fenn,
1930; Deshon and Nelson, 1964; Cavanagh et al., 1977; Bates et al., 1979; Kunz
and Kaufman, 1981; Girardin and Roy, 1984; Hinrichs, 1987).
7 .2 EXPERIENCED POSE® AND HEEL-TOE RUNNER’S STUDY
Two experienced male heel-toe and Pose® runners were matched for age, mass, stature
and experience. Comparison of these two running techniques was conducted using the
research variables (primary and secondary), identified from the Gravitational hierarchical
model. All data used here are the mean of the four test running speeds (2.5 m s'1, 3.35
m s'1, 4.0 m-s'1 and 4.5 m-s'1). The research variables distinguished the two techniques of
running, while giving strong support for the Gravitational hierarchical model.
• Pose and heel-toe runners had similar ground reaction forces, possibly reflecting
their similar body mass and running speed. It was suggested gravity causes
maximum vertical and anterior-posterior ground reaction force (Chang et al.,
2001). Running speed was associated with the angle of inclination of the centre
of mass at terminal stance and the subsequent gravitational torque (Chang et al.,
2001).
• Each rumiing speed (2.5 m-s'1, 3.35 m-s'1, 4.0 m-s'1 and 4.5 m-s'1) produced
increasing angles of inclination and ground reaction forces for the two
techniques. It was suggested, increased angles of inclination (increased moment
arm) increase the gravitational torque producing increased horizontal acceleration
of the centre of mass. Increased horizontal acceleration of the centre of mass will
195
subsequently increase ground reaction force. Further, maximal horizontal
acceleration of the centre of mass increased with running speed.
The time of braking ground reaction force was earlier as a percentage of stance
for the Pose® runners by 13.5% of stance time. Different body geometry at
impact explained the braking ground reaction force differences (Deshon and
Nelson, 1964; Bates et al., 1979). Propulsive and vertical ground reaction force
occurred at a similar percentage of stance time for both techniques. Similar
angles of inclination for each running speed explained the similar timing of
propulsive ground reaction force, owing to their potential relationship to the
gravitational torque (Chang et al., 2001).
Both heel-toe and Pose runner’s maximal horizontal acceleration of the centre
of mass occurred before their horizontal maximal ground reaction force,
suggesting by default that a gravitational torque is the motive force in running.
The propulsive component of gravity caused increased horizontal velocity for the
Pose runners. The increase was observed by graphing the vertical oscillation of
(R)
one Pose® and heel-toe runner’s centre of mass. This clearly illustrated gravity’s
increased propulsive component in the Pose® runner.
In addition, the Pose runners had consistently higher horizontal acceleration of
• (R)
their centre of mass at all test speeds supporting the larger observed propulsive
component of gravity in figure 5.5.2.
Lack of muscle activity during the propulsive phase of stance and maximum
horizontal acceleration occurring before the maximum horizontal ground reaction
force, suggest that both rumiing techniques do not push off the ground but fall
forwards via a gravitational torque.
Horizontal velocity of the centre of mass was lower in Pose® runners for braking
and propulsion by -11 cm s'1 and 6.2 cm-s"1 respectively. The angle of inclination
of the centre of mass at terminal stance and the horizontal velocity were similar,
suggesting the greatest change in the propulsive horizontal velocity masks a rapid
acceleration via the gravitational torque in the Pose® runners.
Lower vertical oscillation was associated with less work against gravity in the
Pose® runners. Vertical displacement and horizontal displacement during stance
for the centre of mass were lower in Pose runners by 3.4 cm and 12.3 cm
respectively. The Pose® runners utilise the gravitational torque more effectively
via lower vertical oscillation, while lower vertical oscillation is associated with
faster running (Miura et al., 1973).
196
• Less horizontal displacement during stance was associated with a more vertically
aligned body at impact. Pose® runners were more vertically aligned at impact, for
shoulder, hip and ankle (8 cm), and the centre of mass more vertically above the
support foot (7 cm). This is associated with less braking of the horizontal velocity
of the centre of mass (Hunter et al., 2005).
• Knee flexion and extension during stance was lower in the Pose® runners by 0.12
rad and 0.2 rad respectively. Less knee extension was associated with lower
vertical oscillation (Wank et al., 1998).
• Knee angle at impact was similar in both techniques but at maximum vertical
ground reaction force, it was lower by 0.11 rad in the Pose® runners. Less knee
flexion was associated with increased leg-spring and possibly increased elastic
recoil during stance (McMahon and Cheng, 1990).
• Stance time, flight time and step length were lower by 31 ms, 23 ms, and 20.2 cm
respectively for the Pose® runners. While step frequency was higher in Pose®
runners by 0.5 Hz. Conflicting research on whether shorter steps and increased
frequency is more efficient make conclusive comment untenable at this time.
• Knee flexion and extension angular velocity were lower in the Pose® runners
during stance by 0.5 rad-s'1 and 0.6 rad-s'1 respectively, whereas angular knee
flexion velocity during swing was higher in the Pose® runners by 1.5 rad-s'1. Less
knee angular flexion during stance may again be associated with a more aligned
body geometry. Less knee extension angular velocity was associated with Pose®
runners passively extending their legs. The higher knee flexion angular velocity
from terminal stance until maximum swing was associated with the Pose®
runners pulling their foot from the ground, in order to catch up with their body,
potentially reducing rotational inertia of the lower-limb via less leg extension
(Bailey and Pate, 1991).
• It was suggested in Pose® running that the hamstring muscles pull the support
foot rapidly from the ground with the aid of the stretch-shortening cycle
optimised from a short contact time (Zatsiorsky, 1995; Paavolainen et al., 1999a).
By pulling the support foot and not driving the swing leg forward o f the body,
gravity can drop the swing leg under the body at impact so the cycle can begin
again. If the swing leg’s centre of mass is vertically under the body’s centre of
mass during flight, they will both land vertically aligned if the hip flexors are not
active.
197
• Lack of electromyography on Posew runners inhibits specific comment on muscle
activity. There appears conflicting data on gastrocnemius activity at terminal
stance (for example, Winter, 1983; Reber et al., 1993), although the leg and hip
extensors are consistently silent during most of leg extension (Kyrolainen et al.,
2003).
• Pose runners therefore, exhibit very different gait characteristics to traditional
(R) •
heel-toe runners, such as: A more vertically aligned body position at impact
(rumiing Pose®), shorter stance time, less knee flexion and extension during
stance, less decrease in the centre of mass horizontal velocity at impact, increased
knee flexion angular velocity from terminal stance until maximum swing angle,
shorter step length, higher step frequency, and lower vertical oscillation.
Therefore, Pose® runners exhibit a distinct biomechanical profile. This is the first
attempt to create a universal biomechanical profile for running based upon the
Gravitational hierarchical model.
• The Gravitational hierarchical model explained the ‘extensor paradox.’ The fact,
maximum horizontal acceleration of the centre of mass occurred before
maximum horizontal ground reaction force supported a gravitational torque as the
motive force in running. Although further research is needed using accelerated
running to determine if horizontal acceleration of the centre of mass also occurs
before propulsive horizontal ground reaction force, similar to constant speed
running.
• Therefore, both heel-toe and Pose® runners fall forwards via a gravitational
torque refuting the theory that rumiers push off the ground. However, Pose®
runners were better able to utilise the gravitational torque. The Gravitational
hierarchical model is therefore a universal model of running, describing both
Pose and heel-toe running techniques and offers an attempt to create a universal
(R) •
profile for running.
7 .3 POSE® METHOD TRAINING STUDY
Sixteen male recreational endurance heel-toe runners were recruited and randomly
assigned into two groups for a repeated measures design. One group received a 7-hour
Pose intervention over 7-days while the other group remained as heel-toe runners.
Biomechanical and physiological (primary and secondary research variables) data
including injury incidence were collected. The Pose® trained group improved their
performance post-test on the 2400 m time trial run, significantly changed their
biomechanical profile and recorded reduced injury incidence. Sixteen 2 x 2 mixed
factorial ANOVAs with repeated measures were used to assess the main effects of group
(control vs. treatment), and trial (pre to post changes) on the dependent variables
(primary research variables). Tukey’s tests of honestly significant differences assessed
individual cell differences post hoc.
• The Ryan-Joiner’s normality test found the data to be normally distributed.

• Levene’s test for homogeneity of variance revealed no significant group
variability.
• A 2 x 2 mixed factorial ANOVA assessed the main effects of group (control vs.
treatment) and trial (pre to post changes) on the primary research variables.
• Significant interaction effects (Pose runners pre-post-test) were found for stance
(K)
time (P - 0.012), centre of mass to support limb at 25 ms {P = 0.014), centre of

mass displacement during stance (P = 0.030) and step frequency (P = 0.005).
• Tukey post hoc tests revealed significant differences between the Pose® group
pre and post-test for stance time (P = 0.001), centre of mass to support limb at 25
ms (P = 0.042), centre of mass displacement during stance (P = 0.001), knee
flexion angular velocity during stance (P = 0.005) plus swing (P = 0.043) and
step frequency {P = 0.002).
• It was suggested that the posterior and anterior ground reaction did not change
because of its potentially strong correlation with gravity (Chang et al., 2000). The
propulsive horizontal velocity variable may have masked the higher horizontal
acceleration of the centre of mass in the Pose runners. Lack of significance for
this variable may reflect the same ruiming speeds and similar angles of
inclination of the centre of mass.
• Mean heart rate in both groups for the 2400 m economy run reduced by 3 bpm.
• No significant statistical change was found in the Pose® and heel-toe group’s
economy data for the 2400 m economy run pre and post-test. Step frequency for
the post-test rose slightly (0.9 steps) in the Pose® group.
(K) r*
• Pose runners ran faster (although not statistically significantly) after the 7-hour
intervention over 7-days on the 2400 m time trial by 24.7 s, while heart rate, and
199
perceived exertion did not change, indicating improved performance was owing
to the Pose® intervention.
• The mixed 2 x 2 factorial ANOVA revealed the Pose® method significantly
changed the biomechanics (to match experienced Pose® runners; table 6.4.1) of
heel-toe runners after the 7-hour Pose® running intervention.
• It appears 7-hours of Pose® running instruction is an adequate time to
significantly make universal biomechanical changes in male recreational
endurance heel-toe runners.
• These biomechanical changes were clearly related to improved performance
(increased running speed) over a 2400 m run from a short intervention period
negating physiological adaptation as the reason for the performance change.
• Running economy did not change or improve in the Pose® group, however, this
was the first study to change technique and not subsequently increase oxygen
consumption (economy decreases).
• A preliminary prospective three-month injury report revealed lower injury
incidence in Pose® runners compared with the heel-toe runners.
7 .4 FUTURE RESEARCH
Gravity’s role in running should be quantified by measuring the horizontal acceleration

of the centre of mass from maximal vertical ground reaction force until terminal stance
in accelerated running. Maximum horizontal acceleration of the centre of mass and
propulsive horizontal ground reaction force should be time matched to determine if
acceleration occurs before the ground reaction force, similar to constant speed running.
The biomechanical technique of Pose® running is based upon a theoretical framework of
using body weight (gravity) to produce movement, which could then also be applied to
other sports upon its substantiation, creating future research opportunities.
1. To quantify gravity’s role in running by using accelerated running (data collected

August, 2005). Measurement of the horizontal acceleration of the centre o f mass
from maximum vertical ground reaction force until terminal stance at 240 frames-s'1
provides an accurate horizontal acceleration of the centre of mass. The maximal
horizontal acceleration of the centre of mass occurs before the maximum horizontal
200
ground reaction force. This supports the propulsive force in running is not ground
reaction force but a gravitational torque.
2. Collect electromyography on Pose runners and compare it with heel-toe runners.
3. Replicate the Pose® training study using elite runners and triathletes.
4. Replicate the Pose® training study using female runners.
5. Replicate the Pose® training study using sprinters.
6. Measure technique changes and differences between heel-toe/forefoot runners and
Pose® runners for downhill, uphill and running against strong winds.
7. Evaluate running technique in children by assessing changes through maturation and
their relation to Pose® method via performance and injury incidence.
8. Develop the theoretical framework of Pose® movement for other sports based upon
the Pose® technique of body weight (gravity) as the key motive force (current
research is being conducted in cycling, golf, rowing and climbing) in relation to
performance and injury.
9. Apply the Pose® running technique in team sports and evaluate running and
performance changes.
10. Specific research into injury mechanisms in Pose® and heel-toe runners should begin
with a large retrospective study involving several hundred runners (current data
collection in London, England and America).
11. Evaluate how Pose® running is learnt using dynamical system theory (pilot study
March, 2006).
201
REFERENCES
Prim ary R e fe re n ce s
Alexander, M.F. (1989) The Alexander Technique. The Essential Writings of F. Matthias
Alexander, p. 12. New York: Lyle Stuart.
Alexander, R. McN. (1988) Optimization and gaits in the locomotion of vertebrates.

Physiological Review, 69, 1199-1227.
Alexander, R. McN. (1989) Elastic mechanisms in animal movement. Cambridge:

Cambridge University Press.
Alexander, R. McN. (1992) The human machine, p. 74-78. London: Natural History
Museum Publications.
Alexander, R. McN. and Bennet-Clark, H.C. (1977) Storage of elastic strain energy in
muscle and other tissues. Nature, 265, 114-117.
Alexander, R. McN. and Jayes, A.S. (1983) A dynamic similarity hypothesis for the gaits
of quadrupedal mammals. Journal o f Zoology, 201, 135-152.
Allard, P., Stokes, I.A.F. and Blanchi, J.P. (1995,) Three-dimensional analysis o f human
movement. Champaign: Human Kinetics.
Anderson, T. (1996) Biomechanics and running economy. Sports Medicine, 22, 76-79.
Anderson, T. and Tseh, W. (1994) Running economy, anthropometric dimensions and

kinematic variables. Medicine and Science in Sports and Exercise, 26, SI 70.
202
Andrews, J.R. (1983) Overuse syndromes of the lower extremity. Clinics in Sports
Science, 2, 137-148.
Arampatzis, A., Knicker, A., Metzler, V. and Bruggemann, G. P. (2000) Mechanical

power in running: a comparison of different approaches. Journal o f Biomechanics, 33
457-463.
Arendse, R.E., Noakes, T.D., Azevedo, L.B., Romanov, N., Schwellnus, M.P. and
Fletcher, G.J. (2004) Reduced eccentric loading of the knee with the Pose running
method. Medicine Science in Sport and Exercise, 36, 272-277.
Armstrong, L.E. and Cooksey, S.M. (1983) Biomechanical changes in selected collegiate
sprinters due to increased velocity. Track and Field Quarterly Review, 3, 10-12.
Auro, O. and Komi, P.V. (1986) The effects of prestretch intensity on mechanical
efficiency of positive work in natural locomotion. International Journal o f Sports
Medicine, 7, 137-143.
Bailey, S.P. and Pate, R.R. (1991) Feasibility of improving rumiing economy. Sports
Medicine, 12, 228-236.
Bale, P., Bradbury, D. and Colley, E. (1986) Anthropometric and training variables
related to 10-km running performance. British Journal o f Sport Medicine, 20, 170-173.
Bartlett, R.M. (1997) Current issues in the mechanics of athletic activities a position
paper. Journal o f Biomechanics, 30, 477-486.
Bartlett, R.M. (1999) Sports Biomechanics. Reducing injuiy and improving performance.
Oxford: E. and F.N. Spon Press.
Bartlett, R.M. (2000) Principles of throwing. In IOC Encyclopaedia o f Sports Medicine:

Biomechanics in Sport, (edited by V.M. Zatsiorsky), 6, 365-380. Oxford: Blackwell
203
Science.
Basmajian, J.V. and Deluca, C.J. (1985) Muscles alive. Baltimore: Williams and
Wilkins.
Bates, B.T., James, S.L. and Ostering, L.R. (1978) Foot function during the support
phase of running. Running, 3, 24-31.
Bates, B.T., Ostering, L.R. and Mason, B.R. (1979) Variations of velocity within the
support phase of running. In Science and Athletics (edited by J. Terauds and G. Dales),
pp. 51-59. Del Mar: Academic Publishers.
Bates, B.T., Ostering, L.R., Sawhill, J.A. and James, S.L. (1983) An assessment of
subject variability, subject-shoe interaction, and the evaluation of rumiing shoes using
ground reaction force data. Journal o f Biomechanics, 16, 181-191.
Belli, A., Kyrolainen, H. and Komi, P.V. (2002) Moment and power of lower limb joints
in running. International Journal o f Sports Medicine, 23, 136-141.
Bennet, J. and Stauber, W. (1986) Evaluation and treatment of anterior using eccentric
exercise. Medicine Science in Sport and Exercise, 18, 526-530.
Bergh, U. Sjordin, B. and Forsberg, A. (1991) The relationship between body mass and
oxygen uptake during running in humans. British Journal o f Sports Medicine, 23, 205-
11.
Bergmann, G., Kniggendorf, H., Graichen, F. and Rohlmann, A. (1995) Influence of

shoes and heelstrike on the loading of the hip joint. Journal o f Biomechanics, 28, 817-27.
Bernstein, N. (1947) On constructing movement, p. 31. Medgiz: Moscow.
Biewener, A. A., Farley, C.T., Roberts, T. J. and Temaner, M. (2004) Muscle mechanical
204
advantage of human walking and rumiing: implications for energy cost. Journal o f
Applied Physiology/, 97, 2266-2274.
Blair, S.N., Kohl, H.W. and Goodyear N.N. (1987) Rates and risks for running and
exercise injuries: studies in three populations. Research Quarterly fo r Exercise and
Sport, 58, 221-228.
Blickhan, R. (1989) The spring-mass model for rumiing and hopping. Journal o f
Biomechanics, 22, 1217-1227.
Bobbert, M.F., Schamhardt, H.C. and Nigg, B.M. (1991) Calculation of ground reaction
force estimates during running position data. Journal o f Biomechanics, 24, 1095-1105.
Bobbert, M.F., Yeadon, M.R. and Nigg, B.M. (1992) Mechanical analysis of the landing
phase in heel-toe running. Journal o f Biomechanics, 25, 223-234.
Bogduk, N. and Twomey, L.T. (1987) Clinical anatomy o f the lumbar spine. New York:
Churchill Livingstone.
Borg, G.A. (1970) Perceived exertion as an indicator of somatic stress. Scandinavian

Journal Rehabilitation Medicine, 2, 69-77.
Bosco, C., Montanari, G. and Ribacchi, R. (1987) Relationship between efficiency of

muscular work during jumping and energetics of running. European Journal o f Applied
Physiology, 56, 138-143.
Branded, B.R. (1973) An analysis of muscle coordination in walking and running gaits.
Medicine and Sport: Biomechanics 111 (edited by S. Cerquiglini, A. Venerando and J.
Wartenweiler), pp. 278-287. Basel, Switzerland: Karger.
Bransford, D.R. and Howley, E.T. (1977) Oxygen cost of running in trained and
untrained men and women. Medicine Science in Sport and Exercise, 9, 41-44.
205
Brill, P.A. and Macera, C.A. (1995) The influence o f running patterns on running
injuries. Sports Medicine, 20, 365-368.
Burke, R.E., Levine, D.N., Tsairis, P. and Zajac, F.E. (1973) Physiological types and
histochemical profiles in motor units of the cat gastrocnemius. Journal o f Physiology,

234, 723-748.
Caputi, A.A., Hoffer, J.A. and Pose, I.E. (1992) Velocity of ultrasound in active and
passive cat medial gastrocnemius muscle. Journal o f Biomechanics 25, 1067-74.
Cavagna, G.A. (1975) Force platforms as ergometers. Journal o f Applied Physiology, 39,
174-179.
Cavagna, G.A. and Citterio, G. (1974) Effect o f stretching on the elastic characteristics
and the contractile component of frog striated muscle. Journal o f Physiology, 239, 1-14.
Cavagna, G.A. and Kaneko, M. (1977) Mechanical work and efficiency in level walking
and running. Journal o f Physiology, 268, 467-481.
Cavagna, G.A., Saibene, F.P. and Margaria, P. (1964) Mechanical work in running.
Journal o f Applied Physiology, 18, 1-9.
Cavagna, G.A. Thys, H. and Zamboni, A. (1976) The sources o f external work in level
walking and running. Journal o f Physiology, 262, 639-657.
Cavagna, G.A., Hegland, N. and Taylor, C. (1977) Mechanical work in terrestrial

locomotion: two basic mechanisms of minimising energy expenditure American Journal
o f Physiology, 233, 243-261.
Cavagna, G.A., Williems, P.A. and Franzetti, P. (1991) The two power limits
conditioning step frequency in human running. Journal o f Physiology, 437, 95-108.
206
Cavanagh, P.R. (1990) Biomechanics o f distance running, p. 31. Champaign: Human
Kinetics.
Cavanagh, P.R. and Komi, P.V. (1979) Electromechanical delay in human skeletal
muscle under concentric and eccentric contractions. European Journal o f Applied
Physiology>, 42, 159-163.
Cavanagh, P.R. and LaFortune, M.A. (1980) Ground reaction forces in distance running.
Journal o f Biomechanics, 11, 487-491.
Cavanagh, P.R. and Williams, K.R. (1982) The effect of stride length variation on
oxygen uptake during distance running. Medicine and Science in Sports and Exercise,
14, 30-35.
Cavanagh, P.R. and Kram, R. (1985) The efficiency of human movement- a statement of
the problem. Medicine and Science in Sports and Exercise, 17, 304-308.
Cavanagh, P.R. and Kram, R. (1990) Stride length in distance running: Velocity, body
dimensions and added mass effects. Medicine and Science in Sports and Exercise, 21,
467-479.
Cavanagh, P.R., Pollock, M.J. and Landa, J. (1977) A biomechanical comparison of elite
and good distance runners. Annals New York Academia Science, 301, 328-345.
Cavanagh, P.R., Kram, R., Rodgers, M.M., Sanderson, D.J. and Hennig, E.M. (1985) An
approach to biomechanical profiling of elite distance runners. International Journal o f
Sports Biomechanics, 13, 397-406.
Chang, Y.H. and Kram, R. (1999) Metabolic cost of generating horizontal forces during
human running. Journal o f Applied Physiology, 86, 1657-1662.
207
Chang, Y.H., Huang, H.W., Hamerski, C.M. and Kram, R. (2000) The independent
effects of gravity and inertia on running mechanics. The Journal o f Experimental
Biology', 203, 229-238.
Chang, Y.H., Hamerski, C.M. and Kram, R. (2001) Applied horizontal force increases
impact loading in reduced-gravity running. Journal o f Biomechanics. 34, 679-685.
Chapman, A.E. (1982) Hierarchy of changes induced by fatigue in sprinting. Canadian

Journal o f Applied Sports Sciences, 7, 116-122.
Chavarren, J. and Calbet, J.A. (1999) Cycling efficiency and pedalling frequency in road
cyclists. European Journal o f Applied Physiology> and Occupational Physiology, 80,
555-563.
Chelly, S.M. and Denis, C. (2001) Leg power and hopping stiffness: relationship with
sprint running performance. Medicine and Science in Sports and Exercise, 33, 326-33.
Christina, K.A., White, S.C. and Gilchrist, L.A. (2001) Effect of localised muscle fatigue
on vertical ground reaction forces and ankle joint motion during running. Human
Movement Science, 20, 257-276.
Christou, E.A. and Carlton, L.G. (1999) Motor output variability during concentric and
eccentric contractions of the quadriceps femoris muscle group. Presented at the 23rd
Annual Meeting o f the American Society o f Biomechanics, University of Pittsburgh.
Clarke, T.E., Frederick, E.C. and Hamill, C.L. (1983) The effect of shoe design upon
rearfoot control in running. Medicine and Science in Sports and Exercise, 15, 376-381.
Clarke, T.E., Cooper, L.B., Hamill, C.L. and Clarke, D.E. (1985) The effect of varied
stride rate upon shank deceleration in running. Journal o f Sports Sciences, 3, 41-49.
208
Clement, D.B., Taunton, J.E., Smart, G.W. and McNicol, K.L. (1981) A survey of
overuse running injuries. Physician and Sports Medicine, 9, 47-58
Cohen, J. (1988) Statistical power analysis fo r the behavioural sciences. Hillsdale:

Erlbaum.
Cole, G.K., Nigg, B.M., van den Bogert, A.J. and Gerritsen, K.G.M. (1996) Lower
extremity joint loading during impact in running. Clinical Biomechanics, 11, 181-193.
Collins, M.H., Pearsall, D.J., Zavorsky, G.S., Bateni, H., Turcotte, R. and Montgomery,
D.L. (2000) Acute effects of intense interval training on running mechanics. Journal o f
Sports Sciences, 18, 83-90.
Conley, D.L. and Krahenbuhl, G. (1980) Running economy and distance running
performance of highly trained athletes. Medicine and Science in Sports and Exercise, 12,
357-360.
Costill, D.L., Thomason, H. and Roberts, E. (1973) Fractional utilization of the aerobic
capacity during distance running. Medicine and Science in Sports and Exercise, 5, 248-
252.
Counsilman, J.E. (1977) Competitive Swimming Manual fo r Coaches and Swimmers.

Counsilman Co: Bloomington.
Crossley, K., Bennel, K.L., Wrigley, T. and Oakes, B.W. (1999) Ground reaction forces,
bone characteristics and tibial stress fractures in male runners. Medicine Science in Sport
and Exercise, 31, 1088-1093.
Curtin, N.A. and Davies, R.E. (1974) Very high tension with very little ATP breakdown
by active skeletal muscle. Biochemistiy Cell Motility, 3, 147-154.
209
Dallam, G.M. Wilber, R.L., Jadelis, K., Fletcher, G.J. and Romanov, N. (2005) Effect of
a global alteration of running technique on kinematics and economy. Journal o f Sports
Sciences, 23, 757-764.
Daniels, J. (1985) A physiologist’s view of running economy. Medicine Science in Sport

and Exercise, 17, 332-338.
Dapena, J. (1984) The pattern of hammer speed fluctuation during hammer throw and
influence of gravity on its fluctuations. Journal o f Biomechanics, 14, 135-142.
Davies, C.T.M. (1980) Effects of wind assistance and resistance on the forward motion
of a runner. Journal o f Applied Physiology, 48, 702-709.
De Wit, B., De Clercq, D. and Aerts, P. (2000) Biomechanical analysis of the stance
phase during barefoot and shod running. Journal o f Biomechanics, 33, 269-278.
Denoth, J., Nigg, B.M. and Neukomn, P.A. (1983) Quantifying load on the human body.
Problems and possible solutions. In Biomechanics VIII - B. (edited by H. Matsui and K.
Kobayashi), pp. 88-99. Champaign: Human Kinetics.
Derrick, T.R., Caldwell, G.E. and Hamill, J. (2000) Modelling the stiffness
characteristics of the human body while running with various stride lengths. Journal o f
Applied Biomechanics, 16, 36-51.
Deshon, D.E. and Nelson, R.C. (1964) A cinematographical analysis of sprint running.
Research Quarterly for Exercise and Sport, 35, 453-454.
Dietz, V., Schmidtbleicher, D. and Noth, J. (1979) Neuronal mechanisms o f human

locomotion. Journal o f Neurophysiology, 42, 1212-1222.
Dillman, C.J. (1975) Kinematic analysis of running. Exercise Sport Science Review, 1,
189-216.
210
Diss, C.E. (2001) The reliability of kinetic and kinematic variables used to analyse
normal running gait. Gait and Posture, 14, 98-103.
Dixon, S.J. and Kerwin, D.G. (1999) The influence of heel lift manipulation on sagittal
plane kinematics in running. Journal o f Applied Biomechanics, 15, 34-39.
Dixon, S.J., Collop, A.C. and Batt, M.E. (2000) Surface effects on ground reaction
forces and lower extremity kinematics in running. Medicine and Science in Sports, 32,
1919-1926.
Edman, K.A.P. (1979) The velocity of unloaded shortening and its relation to sarcomere
length and isometric force in vertebrate muscle fibres. Journal o f Physiology, 291, 143-
159.
Elliot, B.C. and Blanksby, B.A. (1979) The synchronisation of muscle activity and body
segment movements during a running cycle. Medicine and Science in Sports, 11, 322-
327.
Elliot, B.C. and Roberts, A.D. (1980) A biomechanical evaluation of the role of fatigue
in middle-distance running. Canadian Journal o f Applied Sport Science, 5, 203-207.
Elliot, B.C. and Ackland, T. (1981) Biomechanical effects of fatigue on 10,000 metre
running technique. Research Quarterly fo r Exercise and Sport, 52, 160-166.
Enoka, R. (1996) Eccentric contractions require unique activation strategies by the

nervous system. Journal o f Applied Physiology, 81, 2339-2346.
Enoka, R. (2002) Neuromechanics o f human movement, p. 192. Champaign: Human

Kinetics.
Farley, C.T. and Gonzalez, O. (1996) Leg stiffness and stride frequency in human
running. Journal o f Biomechanics, 29, 181-186.
211
Farley, C.T. and Ferris, D.P. (1998) Biomechanics of walking and running: Centre of
mass movements to muscle action. Exercise and Sports Science Reviews, 26, 253-285.
Fenn, W.O. (1930) Work against gravity and work due to velocity changes in running.
American Journal o f Physiology>, 92, 433-462.
Ferris, D.P. and Farley, C.T. (1997) Adjustments in running mechanics to accommodate
different surface stiffnesses. Journal o f Applied Physiology?, 82, 15-22.
Ferris, D.P., Louie, M. and Farley, C.T. (1996) Interaction of leg stiffness and surface
stiffness during human hopping. Physiologist, 39, A59.
Ferris, D.P., Liang, K. and Farley, C.T. (1999) Runners adjust leg stiffness for their first
step on a new running surface. Journal o f Biomechanics, 32, 787-794.
Forwood, M.R. and Parker, A.W. (1989) Microdamage in response to repetitive torsional
loading in the rat tibia. Calcification o f Tissue International, 45, 47-53.
Frederick, E.C. and Clarke, T.E. (1981) Consequences of scaling on impact loading in
running. Medicine and Science in Sports, 13, 96.
Frederick, E.C. and Hagy, J.L. (1986) Factors affecting peak vertical ground reaction
forces in rumiing. International Journal o f Sports Biomechanics, 2, 41 -49.
Fukunaga, T.Y., Kawakami, S., Funato, K. and Fukashiro, S. (1997) Muscle architecture
and function in humans. Journal o f Biomechanics, 30, 457-463.
Gardner, E.N. (1967) Athletics in the ancient world. Oxford: Clarendon.
Girardin, Y. and Roy, B. G. (1984) Effects of a nondirective type of training program on

the running patterns of male sedentary subjects. In Biomechanics IV, (edited by R.C.
Nelson and C.A. Moorhouse), pp. 119-125. Baltimore: University Park Press.
212
Glazier, P. S., Davids, K. and Bartlett, R.M. (2003) Dynamical systems theory: a
Relevant Framework for Performance-Oriented Sports Biomechanics Research
Sportscience, 7, 1-7.
Glick, J. (1980) Muscle strains. Prevention and treatment. Physician Sports Medicine, 8,
73-77.
Gracovetsky, S. (1988) The spinal Engine. Springer- Verlag: New York.
Graham-Brown, T. (1912) Note on some dynamic principles involved in progression.

British Medical Journal, 2, 785-787.
Grieve, D.W., Pheasant, S. and Cavanagh, P.R. (1978) Prediction of gastrocnemius

length from knee and ankle posture. In Biomechanics VI-A, (edited by E. Asmussen and
K. Jorgensen), pp. 405-412. Baltimore: University Park Press.
Griffiths, R.I. (1991) Shortening of muscle fibres during stretch of the active cat medial
gastrocnemius muscle: the role of tendon compliance. Journal o f Physiology, 436, 219-
236.
Gudas, C. (1980) Patterns of lower extremity injury in 224 runners. Comparative

Therapy, 6, 50-59.
Hahn, D.B. and Payne, W.A. (1999) Focus on health. WCB McGraw-Hill.
Hamill, J. Bates, B.T., Knutzen, K.M. and Sawhill, J.A. (1983) Variations in ground
reaction force parameters at different running speeds. Human Movement Science, 2, 47-
56.
Harrison, R.N., Less, A., McCullagh, P.J.J. and Rowe, W.B. (1988) The effects of foot-
strike on muscle and joint forces generated in the human lower limbs during middle
213
distance running of highly trained athletes. Biomechanics in Sport, pp. 81-88. London:
Mechanical Engineering Publications Limited.
Harvey, J. (1983) Overuse syndromes in young athletes. Clinical Sports Medicine, 6, 40-
50.
Hausswirth, C., Brisswalter, J., Vallier, J.M., Smith, D. and Lepers, R. (2000) Evolution
of electromyographic signal, running economy, and perceived exertion during different
prolonged exercises. International Journal o f Sports Medicine, 21, 429-436.
Hay, J.G. (1988) Approach strategies in the long jump. International Journal o f Sports
Hay, J.G. (2002) Cycle rate, length, and speed of progression in human locomotion.
Journal o f Applied Biomechanics, 18, 78-88.
Hay, J.G. and Reid, J.G. (1988) Anatomy, mechanics, and human motion, p. 282.
Prentice-Hall: Englewood Cliffs.
Hay, J.G., Vaughan, C.L. and Woodworth, C.G. (1981) Technique and performance:
identifying the limiting factors. In Biomechanics VII-B, (edited by A. Morecki, K.
Fidelus and A. Wit), pp. 511-520. Baltimore: University Park Press.
He, J.P., Kram, R. and McMahon, T.A. (1991) Mechanics of running under simulated
low gravity. Journal o f Applied Physiology, 71, 863-870.
Heise, G.D. and Martin, P.E. (2001) Are variations in running economy in humans
associated with ground reaction force characteristics? European Journal o f Applied
Physiology, 84, 2031-2035.
214
Heise, G.D., Morgan, D.W., Hough, H. and Craib, M. (1996) Relationships between
running economy and temporal EMG characteristics of bi-articular leg muscles.
International Journal o f Sports Medicine, 17, 128-133.
Hennig, E.M. and Lafortune, M.A. (1991) Contribution of angular motion and
gravitation to tibial acceleration. Medicine and Science and Sports and Exercise, 23,
360-363.
Hill, A.V. (1927) Living machinery, six lectures. New York: Harcourt.
Hinrichs, R.N. (1987) Upper extremity function in running. 11: Angular momentum
considerations. International Journal o f Sports Biomechanics, 3, 232-263.
Hinrichs, R.N. (1990) Upper extremity function in distance running biomechanics. In

Biomechanics o f distance running (edited by P. R. Cavanagh), pp. 107-133 Champaign:
Human Kinetics.
Hinrichs, R.N., Cavanagh, P.R. and Williams, K.R. (1987) Upper extremity function in
running: 1. Centre of mass and propulsive considerations. International Journal o f
Sports Biomechanics, 3, 222-241.
Hintermann, B. and Nigg, B.M. (1998) Pronation in Runners: Implications for Injuries.
Sports Medicine, 26, 169-176.
Hoeberigs, J. (1992) Factors related to the incidence of running injuries. Sports

Medicine, 13, 408-422.
Hof, A.L. and van den Berg, J. (1981) EMG to force processing: 1. An electrical
analogue of the Hill muscle model. Journal o f Biomechanics, 14, 747-758.
Hogan, N., Bizzi, E., Mussa-Ivaldi, F.A. and Flash, T. (1987) Controlling multi-joint
motor behaviour. Exercise Sport Science Reviews, 15, 153-189.
215
Holmich, P., Christensen, S.W., Darre, E., Jahnsen, F. and Hartvig, T. (1989) Non-elite
marathon runners: Health training and injuries. British Journal o f Sports Medicine, 23,
177-178.
Hoshikawa, T., Matsui, H. and Miyashita, M. (1973) Analysis of running pattern in

relation to speed. In Medicine and Sport, Vol 8: Biomechanics 111. pp. 342-348. Basael:
Karger.
Hreljac, A. (1995) Determinants of the gait transition speeds in human locomotion.

Journal o f Biomechanics, 28. 669-677.
Hreljac, A. (2000) Stride smoothness evaluation of runners and other athletes. Gait and
Posture, 11, 199-206.
Hunter, J.P., Marshall, R.N. and McNair, P.J. (2005) Relationships between ground
reaction force impulse and kinematics of sprint-running acceleration. Journal o f Applied
Inman, V.T., Ralston, H.J. and Todd, B. (1981) Human walking, pp. 62-77. Baltimore:
Williams and Wilkins.
Ito, A., Komi, P.V., Sjodin, B., Bosco, C. and Karlsson, J. (1983a) Mechanical efficiency
of positive work in running at different speeds. Medicine Science and Sport and
Exercise, 15, 299-308.
Ito, A., Fuchimoto, T. and Kaneko, M. (1983b) Quantitative analysis of EMG during
various speeds of running. Proceedings o f the Ninth International Congress o f
Biomechanics, Waterloo: Canada.
Jacobs, S.L. and Berson, B.L. (1986) Injuries to runners: A study of entrants to a 10-km
race. American Journal o f Sports Medicine, 14, 151-155.
216
Jacobs, R., Bobbert, M.F. and van Ingen Schenau, G.J. (1993) Function of mono and bi-
articular muscles in running. Medicine Science and Sport and Exercise, 25, 1163-1173.
James S.L. and Brubaker, C.E. (1973) Biomechanical and neuromuscular aspects of
rumiing. Exercise and Sport Science Review, 1, 189-216.
James S.L., Bates, B.T. and Osternig, L.R. (1978) Injuries to runners. American Journal
o f Sports Medicine, 6, 40-51.
Johansson, C. (1986) Injuries in elite orienteers. American Journal o f Sports Medicine,

14,410-415.
Johansson, C. (1992) Knee extensor performance in runners. Differences between

specific athletes and implications for injury prevention. Sports Medicine, 14, 75-78.
Jung, A.P. (2003) The impact of resistance training on distance running performance.
Kadaba, M.P., Ramakrishnan, Ft.K., Wootten, M.E., Gainey, J., Gorton, G. and Cochran,
G.V.B. (1989) Repeatability of kinematic, kinetic and electromyographic data in normal
adult gait. Journal o f Orthopaedic Research, 7, 849-860.
Kaelin, X., Denoth, J., Stacoff, A. and Stussi, E. (1985) Cushioning during running:
Material tests contra subject tests. In Biomechanics in Sport Current Interdisciplinary
Research, (edited by S.M. Perrin), pp. 651-656. Dordrecht: Martin Nijhoff.
Kaneko, M., Fuchimoto, T., Ito, A. and Toyooka, J. (1983) Mechanical efficiency of
sprinters and distance runners during constant speed running. In Biomechanics V l l l - B
pp. 754-761. Champaign: Human Kinetics.
217
Kaneko, M., Matsumoto, M., Ito, A. and Fuchimoto, T. (1987) Optimum step frequency
in constant speed running. In Biomechanics X-B, (edited by B. Jonsson), pp. 803-807.
Champaign: Human Kinetics.
Kawakami, Y., Nozaki, D., Matsuo, A. and Fukanaga, T. (1992) Reliability of

measurement of oxygen uptake by a portable telemetric system. European Journal o f
Applied Physiology, 65, 409-414.
Kawamoto, R., Ishige, Y., Wataria, K. and Fukashiro, S. (2002a) Influence of curve
sharpness on torsional loading of the tibia in running. Journal o f Applied Biomechanics,
18, 1065-78.
Kawamoto, R., Ishige, Y., Wataria, K. and Fukashiro, S. (2002b) Primary factors
affecting maximal torsional loading of the tibia in running. Sports Biomechanics, 1, 167-
86 .
Keele, K.D. (1983) Leonardo da Vinci’s elements o f the science o f man. New York:
Academic Press.
Keller, T.S., Weisberger, A.M., Ray, J.L., Hasan, S.S., Shiavi, R.G. and Spengler, D.M.
(1996) Relationship between vertical ground reaction force and speed walking, slow
jogging and running. Clinical Biomechanics, 11, 253-259.
Ker, R.F., Bennet, M.B., Bibby, S.R., Kester, R.C. and Alexander, R. McN. (1987) The
spring in the arch of the human foot. Nature, 325, 147-149.
Kerdok, A.E., Biewener, A.A., McMahon, T.A., Weyand, P.G. and Herr, H.M. (2002)
Energetics and mechanics of human running on surfaces of different stiffnesses. Journal
o f Applied Physiology, 92, 469-478.
Kerr, B.A., Beauchamp, L., Fisher, V. and Neil, R. (1983) Footstrike patterns in distance
running, (edited by B.M. Nigg and B.M., Kerr), Biomechanical Aspects o f Sports Shoes
218
and laying Surfaces, pp. 135-142. Calgary, Canada: University Printing.
Komi, P.V. (2000) Stretch-shortening cycle: a powerful model to study normal and
fatigued muscle. Journal o f Biomechanics, 33, 1197-1206.
Komi, P.V. and Bosco, C. (1978) Utilisation of stored elastic energy in leg extensor
muscle by men and women. Medicine Science and Sport and Exercise, 10, 261-265.
Komi, P.V., Kaneko, M. and Aura, O. (1987) EMG activity of the leg extensor muscles
with special reference to mechanical effects in concentric and eccentric exercise.
International Journal o f Sports Medicine, 8, 22-29.
Koplan, J.P., Powell, K.E., Sikes, R.K., Shirley, R.W. and Campbell, G.C. (1982) An
epidemiological study of the benefits and risks of running Journal o f the American
Medical Association, 248, 3118-3121.
Kovacs, I., Tihanyi, J., Devita, P., Racz, L., Barrier, J. and Hortobagyi, T. (1999) Foot
placement modifies kinematics and kinetics during drop jumping. Medicine Science and
Sport and Exercise, 31, 708-716.
Kram, R., Domingo, A. and Ferris, D. P. (1997) Effect of Reduced Gravity on the
Preferred Walk-Run Transition Speed. Journal o f Experimental Biology, 200, 821-826.
Kreighbaum, E. and Barthels, K. (1996) Biomechanics: A qualitative approach fo r

studying human movement. Boston: Allyn and Bacon.
Krissoff, W.B. and Ferris, W.D. (1979) Runners injuries. Physician and Sports
Medicine, 7, 55-64.
Kubo, K., Kanehisa. H., Kawakami, Y. and Fukunaga, T. (2000) Elastic properties of
muscle-tendon complex in long distance runners. European Journal o f Applied
Physiology, 81, 181-187.
219
Kuitunen, S., Komi., P.V. and Kyrolainen, H. (2002) Knee and ankle joint stiffness in
sprint running. Medicine Science and Sport and Exercise, 10, 261-265.
Kunz, H. and Kaufman, D.A. (1981) Annotation biomechanical analysis of sprinting:

decathletes versus champions. British Journal o f Sports Medicine, 15, 177-181.
Kyrolainen, H., Komi, P.V. and Belli, A. (2000a) Changes in muscle activity patterns
kinetics with increasing running speed. Journal o f Strength and Conditioning Research,
13, 404-406.
Kyrolainen, H., Finni, T., Avela, J, and Komi, P.V. (2003) Neuromuscular behaviour of
the triceps surae muscle-tendon complex during running and jumping. International
Journal o f Sports Medicine, 24, 153-5.
Kyrolainen, H., Pullinen, T., Candau, R., Avela, J., Huttumen, P. and Komi, P.V.
(2000b) Effects of marathon running on running economy and kinematics. European
Journal o f Applied Physiology, 82, 297-304.
Laughlin, T. and Devles, J. (1996) Total immersion. The revolutionary way toswim
better, faster, and easier. New York: Simon and Schuster.
Laughton, C.A., McClay Davis, I., and Hamill, J. (2003) Effect of strike pattern and
orthotic intervention on tibial shock during running. Journal o f Applied Biomechanics,
19, 153-168.
Laurent, M. and Pailhous, J. (1986) A note on modulation of gait in man: Effect of

constraining stride length and frequency. Human Movement Science, 5, 333-343.
Lee, C.R. and Farley, C.T. (1998) Determinants of the centre of mass trajectory in
human walking and running. Journal Experimental Biology>,201, 235-44.
220
Lehman, W.L. (1984) Overuse syndromes in runners. American Family Physician, 29,
179-181.
Lewis, G., Schwellnus, M.P. and Sole, G. (2000) The aetiology and clinical features of
flow back pain in distance runners: A review. International Journal o f Sports Medicine,
1, 1-25.
Li, L. (2000) The equilibrium points of a mass-spring model of running while using
different stride lengths. Journal o f Sport and Exercise Psychology, 22, 23-27.
Liu, M.M., Herzog, W., Hans, J. and Savelberg, H.C.M. (1999) Dynamic muscle force
predictions from EMG: an artificial neural network approach. Journal of
Electromyography and Kinesiologyy 9, 391-400.
Luhtanen, P. and Komi, P.V. (1978) Segmental contribution to forces in a vertical jump.
European Journal o f Applied Physiology;, 38, 181-188.
Luhtanen, P. and Komi, P.V. (1980) Force-power and elasticity-velocity relationships in

walking, running and jumping. European Journal o f Applied Physiology, 44, 270-289.
Lysholm, J. and Wilander, J. (1987) Injuries in running. American Journal o f Sports

Medicine, 15, 168-171.
Macera, C.A. (1992) Lower extremity injuries in runners- advances in prediction. Sports
Medicine, 13, 50-57.
Macera, C.A., Pate, R.R., Powell, K.E., Jackson, K.L. and Kendrick, J.S. (1989)
Predicting lower extremity injuries among habitual runners. Archives o f internal
Medicine, 149, 2565-2568.
221
MacIntyre, D.L. and Robertson, D.G.E. (1987) EMG profiles of the knee muscles during
treadmill running. In Biomechanics X-A, (edited by Bengt Jonsson), pp. 289-294.
Mann, R.A. (1980) Biomechanics of walking, running and sprinting. American

Orthopaedic Society o f Sports Medicine, 8, 345-350.
Mann, R.A. (1981) Kinetic analysis of sprinting. Medicine Science Sports and Exercise,
13, 325-328.
Mann, R. A. (1982) Biomechanics of running. In Prevention and treatment o f running

injuries (edited by R. D ’Ambrosia and D. Drez), pp. 1-13. Thorofare: C.B. Slack.
Mann, R.A. and Hagy, J. (1980) Biomechanics of walking, running and sprinting.
American Journal Sports Medicine, 8, 345-9.
Mann, R.A. and Sprague, P. (1980) A kinetic analysis o f the ground leg during sprint
running. Research Quarterly for Exercise and Sport, 51, 334-348.
Mann, R.A. and Herman, J. (1985) Kinematic analysis of Olympic sprint performance:
in men’s 200 m. International Journal Sport Biomechanics, 1, 151-162.
Mann, R.A., Moran, G. T. and Dougherty, S. (1986) Comparative electromyography o f

the lower extremity in jogging, running and sprinting. American Journal Sports
Medicine, 14, 501-510.
Marey, EJ. (1972) Movement. New York: Amo. (original work published in 1895).
Marieb, E.N. (1992) Human anatomy and physiology. New York: Benjamin Cummings.
222
Marti, B., Vader, J.P., Minder, C.E. and Abelin, T. (1988) On the epidemiology of
running injuries: the 1984 Bern Grand-Prix study. American Journal o f Sports Medicine,
16, 285-294.
Martin, P.E. (1985) Mechanical and physiological responses to lower extremity loading
during running. Medicine Science and Sports and Exercise, 17, 427-433.
Martin, P.E. and Marsh, A.P. (1992) Step length and frequency effects on ground
reaction forces during walking. Journal o f Biomechanics, 25, 1237-1239.
Martin, P.E., Sanderson, DJ. and Umberger, B.R. (2000) Factors affecting preferred
rates of movement in cyclic activities. In Biomechanics o f Sport, (edited by V.M.
Zatsiorsky), pp. 143-160. Oxford: Blackwell Science.
Maxwell-Donelan, J., Park, J. Y. and Kram, R. (1998) A new dynamic similarity

hypothesis for running animals. Presented at the North American Congress on
Biomechanics, University of Waterloo: Canada.
Mayhew, J.L. (1977) Oxygen cost ad energy expenditure of running in trained runners.
British Journal o f Sports Medicine, 11, 116-121.
McClay, I.S. (2000) The evolution of the study of the mechanics of running. Journal o f
American Podiatry and Medical Association, 90, 133-148.
McClay, I.S. and Manal, K. (1997) Coupling parameters in runners with normal and
excessive pronation. Journal o f Applied Biomechanics, 13, 109-124.
McClay, I.S., Lake, M.J. and Cavanagh, P. R. (1990) Muscle activity in Running.
Biomechanics o f distance running (edited by P.R. Cavanagh), pp. 165-185. Champaign:
Human Kinetics.
223
McClay, I.S., Williams, D.S. and Laughton, C.A. (1999) Can gait be retrained to prevent
injury in runners? Presented at the 23ul Annual Meeting o f the American Society o f
Biomechanics, University of Pittsburgh.
McKenzie, D., Clement, D. B. and Taunton, J. E. (1985) Running shoes, orthotics and
injuries. Sports Medicine, 2, 324-327.
McLaughlin, J.E., King, G.A., Howley, E.T., Bassett, D.R. and Jr, Ainsworth, B.E.
(2001) Validation of the COSMED K4 b2 portable metabolic system. International
Journal Sport Medicine, 22, 280-284.
McMahon, P. (1956) Leonardo da Vinci’s treatise on painting. Princeton: Princeton

University Press.
McMahon, T.A. (1990) Spring-like properties of muscles and reflexes in running:

multiple muscle systems. Biomechanics o f Movement Organisation, 37, 578-590.
McMahon, T.A. and Cheng, G.C. (1990) The mechanics of rumiing: how does stiffness
couple with speed. Journal o f Biomechanics, 23, 65-78.
McMahon, T.A. Valient, G.A. and Frederick, E.C. (1987) Groucho running. Journal o f
Applied Physiology\ 62, 2326-2337.
Mero, A. and Komi, P.V. (1986) Force, EMG and elasticity-velocity relationships at
submaximal, maximal and supramaximal running speeds in sprinters. European Journal
o f Applied Physiology, 55, 553-561.
Mero, A., Komi, P.V. and Gregor, R.J. (1992) Biomechanics of sprint running. Sports
Medicine, 13, 376-392.
224
Messier, S.P. and Cirillo, K.J. (1989) Effects of a verbal and visual feedback system on
running technique, perceived exertion and running economy in female novice runners.
Journal o f Sports Science, 7, 113-126.
Meyers, M.J. and Steudel, K. (1985) Effect of limb mass and its distribution on the
energetic cost of running. Journal Experimental Biology>, 116, 363-373.
Miller, D. I. (1978) Biomechanics of running—what should the future hold? Canadian

Journal o f Applied Sports Sciences, 3, 229-236.
Miller, D. E (1990) Ground reaction forces in distance rumiing. In Biomechanics o f

distance running, (edited by P.R. Cavanagh), pp. 203-224. Champaign: Human Kinetics.
Milliron, M.J. and Cavanagh, P.R. (1990) Sagittal plane kinematics of the lower
extremity during distance running. In Biomechanics o f Distance Running, (edited by P.R.
Cavanagh), pp. 85-93. Champaign: Human Kinetics.
Miura, M., Kobayashi, K., Miyashita, M., Matsui, H. and Sodeyama, H. (1973)
Experimental studies on biomechanics on long distance runners. In review o f our
researches, 1970-1973, (edited by H. Matsui), pp. 45-46. University of Nagoya, Japan.
Montgomery, W.H., Pink, M. and Perry, J. (1994) Electromyographic analysis of hip and
knee musculature during running. American Journal o f Sports Medicine, 22, 272-278.
Morgan, D.W., Martin, P.E. and Craib, M. (1994) Effect of step length optimization on
the aerobic demand of rumiing. Journal o f applied Physiology, 77, 245-51.
Morgan, D.W., Baldini, F.D., Martin, P.E. and Kolirt, W.M. (1987) Ten-kilometre
performance and predicted velocity at VCb max among well-trained male runners.
Medicine and Science in Sports and Exercise, 21, 78-83.
225
Morgan, D.W., Martin, P.E., Krahenbuhl, G.S. and Baldini, F.D. (1991) Variability in
running economy and mechanics among trained male runners. Medicine and Science in
Sports and Exercise, 23, 378-383.
Moritani, T., Takaishi, T. and Matsumoto, T. (1993) Determination of maximal power

output at neuromuscular fatigue threshold. Journal o f Applied Physiology 74, 1729-1734.
Morrison, J.B. (1970) The mechanics of muscle function in locomotion. Journal o f

Munro, C.F., Miller, D.I. and Fuglevand, A.J. (1987) Ground reaction forces in running:
A re-examination. Journal o f Biomechanics, 20, 147-156.
Nelson, R.C. and Gregor, R.J. (1976) Biomechanics o f distance running: A longitudinal
study. Research Quarterly for Exercise and Sport, 47, 417-428.
Nelson, R.C., Dillman, C.J., Lagasse, P. and Bickett, P. (1972) Biomechanics o f over
ground running versus treadmill running. Medicine Science in Sports, 4, 223-240.
Nicol, C., Komi, P.V. and Marconnet, P. (1991) Fatigue effects of marathon running on
neuromuscular performance 11: changes in force, integrated electromyographic activity
and endurance capacity. Scandinavian Journal o f Medicine and Science in Sports, 1, 18-
24.
Nigg, B.M. (1983) External force measurements with sport shoe and playing surfaces. In
Symposium on Biomechanical Aspects o f Sports Shoes and Playing Surfaces, (edited by
B.M. Nigg and B.A. Kerr), pp. 11-32. Calgary: University of Calgary.
Nigg, B.M. (1986) Biomechanics o f running shoes. Champaign: Human Kinetics.
Nigg, B.M., de Boer, R.W. and Fisher, V. (1995) A kinematic comparison of overground
running versus treadmill running. Medicine Science in Sports, 27, 98-105.
226
Nigg, B. M., Bahlsen, H. A., Luethi, S.M. and Stokes, S. (1987) The influence of
running velocity and midsole hardness on external impact forces in heel-toe rumiing.
Journal o f Biomechanics, 20, 951-959.
Nilsson, J. and Thorstensson, A. (1987) Characterisation of different foot strike patterns

in running and their effects on ground reaction forces and rumiing economy. 10fh
International Congress o f Biomechanics, p. 230. Amsterdam: Free University.
Noakes, T. D. (2002) The Lore o f Running, p. 56. Oxford: Oxford Press.
Norman, R., Sharratt, J.J., Pezzack, J.J. and Noble, E. (1976) Re examination of he
mechanical efficiency of horizontal treadmill running. In Biomechanics V-B, (edited by
P.V. Komi), pp. 87-93. Baltimore: University Park.
Novacheck, T.F. (1996) Running and Sprinting. A dynamic analysis. St Paul, MN:
Gillette Children’s Hospital.
Novacheck, T.F. (1998) The biomechanics of running. Gait and Posture, 7, 77-95.
Nummela, A., Rusko, H. and Mero, A. (1994) Electromyographical activities and ground
reaction forces during fatigued and non-fatigued sprinting. Medicine Science in Sports
Exercise, 26, 605-609.
Nussbaum, M.C. (1978) Aristotle’s De Moto Animalium (Animal motion). Princeton:

Princeton University Press.
Nutig, M. (1981) Orthopedic injuries in runners. Orthopedic Review, 10, 97-100.
Nytro, A. (1987) What is Correct Technique? Track Technique, 100, 3195-3205.
Oakley, T. and Pratt, D.J. (1988) Skeletal transients during heel and toe strike running
and the effectiveness of some materials in their attenuation. Clinical Biomechanics, 3,
227
159- 165.
Oyster, N. and Wooten, E.P. (1971) The influence of selected anthropometric

measurements on the ability of college women to perform a 35-yard dash. Medicine
Science in Sports Exercise, 3, 130-134.
Paavolainen, L., Hakkinen, K., Hamalainen, A.N. and Rusko, H. (1999a) Explosive-
strength training improves 5-km run time by improving running economy and muscle
power. Journal o f Applied Physiology/, 86, 1527-1533.
Paavolainen, L., Nummela, A., Rusko, H. and Hakkinen, K. (1999b) Neuromuscular

characteristics and fatigue during 10-km running. International Journal o f Sports
Medicine, 20, 516-521.
Paradisis, G.P. and Cooke, C.B. (2001) Kinematic and postural characteristics of sprint
running on sloping surfaces. Journal o f Sports Sciences, 19, 149-159.
Patla, A.E., Armstrong, C.J. and Silverira, J.M. (1989) Adaptation of the muscle
activation patterns to transitory increases in stride length during treadmill locomotion in
humans. Human Movement Science, 8, 45-66.
Payne, A.H. (1983) Foot to ground contact forces in elite runners. In Biomechanics
V l l l - B (edited by H. Matsui and K. Kobayashi), pp. 19-41. Champaign: Human
Kinetics.
Petray, C.K. and Krahenbuhl, G.S. (1985) Running training, instruction on running
technique, and running economy in 10-year old males. Research Questions in Exercise
and Sports, 56, 251-55.
Pierrynowski, M.R. (1995)Analytic representation of muscle line of action and

geometry. In Three-Dimensional Analysis o f Human Movement (edited by P. Allard,
I.A.F. Stokes, and J.P Blanchi), pp. 215-256. Champaign: Human Kinetics
228
Pink, M., Perry, J., Houglum, P.A. and Devine, D.J. (1994) Lower extremity range of
motion in the recreational sport runner. American Journal o f Sports Medicine, 22, 541-
549.
Pinniger, G.J., Steele, J.R. and Groeller, H. (2000) Does fatigue induced by repeated
dynamic efforts affect hamstring muscle function? Medicine Science in Sports Exercise,
32, 647-653.
Pinnington, H.C., Wong, P., Tay, J., Green, D. and Dawson, B. (2001) The level of
accuracy and agreement in measures of FE02, FEC02 and VE between the Cosmed
K4b2 portable, respiratory gas analysis system and a metabolic cart. Journal o f Science
Medicine and Sport, 4, 324-35.
Powell, K.E. (1986) A physiological perspective on the causes of running injuries.

Physician and Sports Medicine, 14, 100-114.
Pratt, D.J. (1989) Mechanisms of shock attenuation via the lower extremity during
running. Clinical Biomechanics, 4, 51-57.
Pugh, L.G.C.E. (1971) The influence of wind resistance in running and walking and the
mechanical efficiency of work against horizontal and vertical forces. Journal o f
Physiology', 213, 255-276.
Radin, E.L. and Paul, I.L. (1971) Response of joints to impact loading. Arthritis and
Rheumatism, 14, 356-362.
Reber, L., Perry, J. and Pink, M. (1993) Muscular control of the ankle in running.
American Journal o f Sports Medicine, 21, 805-810.
Renstrom, P. (1994) An introduction to chronic overuse injuries. In Oxford Textbook o f

Sports Medicine, pp. 531-545. Oxford: Oxford Medical Publications.
229
Riley, P.O., Croce, U.D. and Kerrigan, D.C. (2001) Propulsive adaptation to changing
gait speed. Journal o f Biomechanics, 34, 197-202.
Robergs, R.A. and Roberts, S.O. (2000) Fundamental principles o f Exercise Physiology
for Fitness performance and health, p. 69. New York: McGraw Hill.
Rolf, C. (1995) Overuse injuries of the lower extremity in runners. Scandinavian Journal
o f Medicine and Science in Sports, 5, 181-190.
Romanov, N. (2002) Dr. Nicholas Romanov’s Pose® method o f running. PoseTech:

Miami.
Rowland, T. (1989) Oxygen uptake and endurance fitness in children: a developmental

perspective. Paediatric Exercise Science, 1, 313-328.
Roy, B. (1982) Biomechanical characteristics of endurance running. Canadian Journal

o f Applied Sciences, 7, 104-115.
Saito, M., Kobayashi, K., Miyashita, M. and Hoshikawa, T. (1974) In Biomechanics IV,
(edited by R.C. Nelson and C.A. Morehouse), pp. 106-111. Baltimore: University Park
Press.
Schwab, G.H., Moynes, D.R., Jobe, F.W. and Perry, J. (1983) Lower extremity
electromyographic analysis of running gait. Clinical Orthopaedics, 176, 166-170.
Simms, G. (2003) Why die? The extraordinary Percy Cerutty, maker o f champions, p.
31. Sydney: Lothian Books.
Sinning, W.E. and Forsythe, H.L. (1970) Lower limb actions while running at different
velocities. Medicine Science Sports and Exercise, 2, 28-34.
230
Slocum, D.B. and Bowerman, W. (1961) The biomechanics of running. Proceedings o f
the Second National Conference fo r Medical Aspects o f Sports, p. 53-58.
Slocum, D.B. and James, S.L. (1968) Biomechanics of running. Journal American
Medical Association, 205, 97-104.
Smidt, G.L. (1973) Biomechanical analysis of the knee flexion and extension. Journal o f
Stanish, W.D. (1984) Overuse injuries in athletics: a prospective study. Medicine Science
Sports and Exercise, 16, 1-7.
Stanish, W.D. and Wood, R. (1994) Overuse injuries of the knee. In Oxford Textbook o f
Sports Medicine, pp. 568-582. Oxford: Oxford Medical Publications.
Stergiou, N., Bates, B.T. and James, S.L. (1999) Asynchrony between subtalar and knee
joint function during running. Medicine Science in Sports Exercise, 31, 1645-55.
Subotnick, S.I. (1985) The biomechanics of running. Journal o f Sports Medicine, 2, 144-
153.
Sutker, A., Jackson, D. and Paglioano, J. (1981) Illiotibial band syndrome in distance
runners. Physician Sports Medicine, 9, 1.
Taunton, J.E., Clement, D.B., Smart, G.W. and McNicol, K. (1987) Non-surgical
management of overuse injuries in runners. Canadian Journal o f Sports Sciences, 12, 11-
18.
Taunton, J.E., Clement, D.B., Smart, G.W., Wiley, J.P. and McNicol, K. (1982) A
triplanar electrogoniometer investigation of running mechanics in runners with
compensatory overpronation. International Journal o f Sports Medicine, 2, 31.
231
Taunton, J.E., Ryan, M.B., Clement, D.B., McKenzie, D.C., Lloyd-Smith, D.R. and
Zumbo, B.D. (2002) British Journal o f Sports Medicine, 36, 95-101.
Taylor, C.R. (1994) Relating mechanics and energetic during exercise. Advanced
Veterinaiy Science, 38A, 181-215.
Thorstensson, A. and Roberthson, H. (1987) Adaptations to changing speed in human

locomotion: Speed of transition between walking and running. Acta Physiologica
Scandinavica, 131, 211-4.
Tirosh, O. and Sparrow, W.A. (2003) Identifying heel contact and toe-off using
forceplate thresholds with a range of digital-filter cutoff frequencies. Journal o f Applied
Unnithan, V.B. and Eston, R.G. (1990) Stride frequency and submaximal treadmill
running economy in adults and children. Paediatric Exercise Science, 2, 149-155.
van Ingen Schenau, G.J. (1980) Some fundamental aspects of the biomechanics of
overground versus treadmill locomotion. Medicine Science in Sports Exercise, 12, 257-
61.
van Ingen Schenau, G.J., Boots, P.J.M., Groot, G., de Snackers, R.J., Woensel, W. W.
and Van, L. M. (1993) The constrained control of force and position in multi-joint
movements. Neuroscience, 46, 197-202.
van Mechelen, W. V. (1992) Running injuries. A review of the epidemiological

literature. Sports Medicine, 14, 320-335.
van Mechelen, W. V. (1995) Can running injuries be prevented? Sports Medicine, 19,
161-165.
Verbitsky, O., Mizrahi, J., Voloshin, A., Treiger, J. and Isakov, E. (1998) Shock
232
transmission and fatigue in human running. Journal o f Applied Biomechanics, 14, 35-41.
Volishin, A. and Wosk, J. (1982) An in vivo study of low back pain and shock
absorption in the human locomotor system. Journal o f Biomechanics, 15, 21-27.
Wakeling, J. M. and Nigg, B.M. (2001) Modification of soft tissue vibrations in the leg
by muscular activity. Journal o f Applied Physiology>90, 412-420.
Wallack, T. (2004) Save Your Knees. Runner’s World, 109, 68-73.
Walter, S.D., Hart, L.E., McIntosh, J.M. and Sutton, J.R. (1989) Study of running related
injuries. Archives o f Internal Medicine, 149, 2561-2564.
Wank, V., Frick, U. and Schmidtbliecher, D. (1998) Kinematics and electromyography

of lower limb muscles in overground and treadmill running. International Journal o f
Weyand, P.G., Sternlight, D.B., Belizzi, M.J. and Wright, S. (2000) Faster top running
speeds are achieved with greater ground forces not more raid leg movements. Journal o f
Applied Physiology, 89, 1991-1999.
Whittle, M. (1991) Gait Analysis: An Introduction. Butter worth & Heinemann.
Wilcox, A.R. and Bulbulian, R. (1984) Changes in running economy relative to KCFmax
during a cross-country season. Journal o f Sports Medicine, 9, 41-44.
Williams, D., McClay, I. and Manal, K. (2000) Lower extremity mechanics in runners
with a converted forefoot strike pattern. Journal o f Applied Biomechanics, 16, 210-218.
Williams, K.R. (1980) A biomechanical evaluation o f distance running efficiency. Ph.D.

dissertation. Pennsylvania State University.
233
Williams, K.R. (1985) Biomechanics of running. Exercise Sport Science Review, 13,
389-441.
Williams, K.R. (1990) Relationship between distance running biomechanics and running
economy. In Biomechanics o f distance running (edited by P. R. Cavanagh), p. 277.
Williams, K.R. and Cavanagh, P.R. (1983) A model for the calculation of mechanical
power during distance running. Journal o f Biomechanics, 16, 115-128.
Williams, K.R. and Cavanagh, P.R. (1986) Biomechanical correlates with running
economy in elite distance runners. In Proceedings o f the North American Congress on
Biomechanics, pp. 287-288. Montreal: Canada.
Williams, K.R. and Cavanagh, P.R. (1987) Relationship between distance running
mechanics, running economy, and performance. Journal o f Applied Physiology, 63,
1236-1245.
Williams, K.R. and Ziff, J.L. (1991) Changes in distance running mechanics due to
systematic variations in running style. International Journal o f Sports Biomechanics, 7,
76-90.
Williams, K.R., Cavanagh, P.R. and Ziff, J.L. (1987) Biomechanical studies o f elite
female distance runners. International Journal o f Sports Medicine, Suppl, 107-118.
Willson, J.D. and Kemozek, T.W. (1999) Plantar loading and cadence alterations with
fatigue. Medicine Science in Sports Exercise, 31, 1828-1833.
Winter D.A. (1978) A new definition of mechanical work done in human movement.
Exercise and Sports Science Reviews, 6, 183-201.
Winter D.A. (1983) Moments of force and mechanical power in jogging. Journal o f
234
Winter, D.A. (1987) The biomechanics and motor control o f human gait. Waterloo:
University of Waterloo Press.
Winter, D.A. (1990) Biomechanics and motor control o f human movement. NY: John
Wiley and Sons Inc.
Winter, D.A. and Bishop, P.J. (1992) Lower extremity injury. Biomechanical factors
associated with chronic injury to the lower extremity. Sports Medicine, 14, 149- 156.
Wosk, J. and Volishin, A. (1982) Wave attenuation in skeletons o f young healthy

persons. Journal o f Biomechanics, 14, 261-267.
Wright, I.C., Neptune, R.R., van den Bogert, A.J. and Nigg, B.M. (1998) Passive
regulation of impact forces in heel-toe running. Clinical Biomechanics, 13, 521-531.
Yeadon, M.R. and Challis, J.H. (1994) The future o f performance related sports
biomechanics research. Journal o f Sports Science, 12, 3-32.
Zajac, F.E. (1993) Muscle coordination of movement: a perspective. Journal o f

Zatsiorsky, V. M. (1995) Science and Practice o f Strength Training. Champaign: Human

Kinetics.
Zatsiorsky, V. M. (2002) Kinetics o f human motion, p. 51. Champaign: Human Kinetics.
Zatsiorsky, V. M. and Fortney, V.L. (1993) Sports biomechanics 2000. Journal o f Sports
Science, 11, 279-283.
235
S eco n d ary re fe re n c e s
Anokhin, P. K. (1978) Selected works. Philosophical aspects o f the Functional System

Theory, pp. 27-48. Nauka. M. Publications (In Russian).
Asmussen, E. and Bonde-Petersen, F. (1974) Apparent efficiency and storage of elastic

energy in human muscles during exercise. Acta Phvsiologica Scandinavica, 92, 537-545.
Enomoto, Y., Ae, M., Fujji, M. and Okada, H. (1999) Running technique in long
distance running and the effectiveness of mechanical energy utilization. Japanese
Journal o f Biomechanics in Sports and Exercise, 3, 12-19.
Hogberg, P. (1952) How do stride length and stride frequency influence the energy
output during running? Arbeitsphysiologie, 14, 437-441
Knuttgen, H.G. (1961) Oxygen uptake and pulse rate while running with undetermined
stride lengths at different speeds. Acta Phvsiologica Scandinavica, 52, 366-371.
Komi, P.V., Hyvarinen, T., Gollhofer, A. and Mero. A. (1986) Man-shoe-surface: special
problems during marathon running. Acta Unix Oul A., 179, 69-72.
Kunz, H. (1978) The training of take-off forces. Schweizer Turnen und Leichtathletik,
15, 12-14.
Margaria, R. (1968) Positive and negative work performance and their efficiencies in
human locomotion. International Z Angew Physiol, 25, 339-351.
Mason, B.R. (1980) A kinematic and kinetic analysis o f selected parameters during the
support phase o f running. Unpublished doctoral dissertation, University of Oregan.
236
Narici, M. V., Hoppeler, H., Kayser, B., Landoni, L., Claassen, H., Gavardi, C., Conti,
M. and Cerretelli, M. (1996) Human quadriceps cross-sectional area, torque and neural
activation during 6 months strength training. Acta Physiologica Scandinavica, 157, 175.
Nigg, B.M. and Luethi, S. (1980) Movement analysis and running shoes.
Sportwissenshaft, 3, 309-320.
Nilsson, J. and Thorstensson, A. (1985) Adaptability in frequency and amplitude of leg

movements during locomotion at different speeds. lCfh International C o n fess o f
Biomechanics Abstract Book, 20: 194. Solna, Sweden: Arbetar-Skydd Sverket.
Nilsson, J., Thorstensson, A. and Halbertsma, J. (1985) Changes in leg movements and
muscle activity with speed of locomotion and mode of progression in humans. Acta
Physiologica Scandinavica, 123, 457-475.
van Mechelen, W.V., Hlobil, H. and Kember, H.C.G. (1987) How can sports injuries be
prevented? p. 43. Netherlands: NISGZ publication no. 25E.
Yzerman, J.C. and Van Galen, W.C.C. (1987) Blessures bij Lange Afstan. Report
K.N.A.U.
237
APPENDIX 1
PARTICIPANT RECRUITMENT
SHEFFIELD HALLAM UNIVERSITY

SCHOOL OF SPORT AND LEISURE MANAGEMENT
RISK ASSESSMENT PROFORMA
PROCEDURE:
ASSESSMENT No.
DATE ASSESSED:
ASSESSED BY:
SIGNED: POSITION:
HAZARDS RISKS AND SPECIFIC CONTROL

MEASURES
1.
RISK EVALUATION COVERALL)
GENERAL CONTROL MEASURES
EMERGENCY PROCEDURES
MONITORING PROCEDURES
REVIEW PERIOD
REVIEWED BY: DATE:

1
POSE® M E TH O D R U N N IN G S T U D Y IN FO R M A TIO N
Please read the following information regarding a study on the Pose® method of running
that will be conducted at Sheffield Hallam University’s centre for Sport and Exercise
Science, Collegiate Crescent Campus and Loughborough university by Graham Fletcher.
The aim: is to make a biomechanical and physiological analysis of the Pose® method of
running using male runners with a 10 km best time between 37-45 minutes.
Selection of participants: to take part in this study you must be:

• Male (18-30 years old);
• Able to run a 10 km personal best time between 37-45 minutes equating to 6-71/2
minute per mile pace.
• A runner who lands on the heel as the first point of ground contact.
• Able to make a time slot for data collection between May 24th-29th and June 6-
13th 2003.
This study will be conducted using two groups:
Group 1: Control group

The first group will continue to run and train as normal while once a day for a week on
your own, complete some drills, such as high knee lifts and butt kicks. This group will
then need to attend the laboratory test and track run (between May 24th-29th and June 6-
13th). A training log will then need to be kept for three months recording any injuries
sustained. This can be handed in or emailed to Rick Velati.
Group 2: Pose® group

The second group will be trained in the Pose® method for one week with training
sessions of 60-minutes each. This group will also need to attend the laboratory test and
track run (between May 24th-29th and June 6-13th). A training log will then need to be
kept for three months recording any injuries sustained. This can be handed in or emailed
to Rick Velati.
Laboratory and track dates

Please fill in a time and day for both the lab and track pre and post tests (see sign up
sheets).
Data collection dates:
Lab tests
• First lab test will be between May 26th and May 29th.
• Second lab tests will be between 9-11th June.
Procedure (what you need to do):

Please bring a pair of running flats (flattest pair of running shoes you have). A short tight
fitting pair of running shorts. Please wear the same pair of shoes for both tests. You will
need to sign the consent form and warm-up on arriving at the laboratory. Surface
electrodes will be placed on five leg muscles to record the electrical muscle activity
while you run (you will not feel any electrical impulses from this procedure). Reflective
markers will be taped to certain points on your body so the cameras will be able to
measure your running action. There will be a forceplate under the carpet that will record
the forces you make when landing on the ground. Next you will practice running on the
16 m runway (the length of the lab) to ensure the markers and electrode wires do not
2
interfere with your natural running gait. Once you are comfortable with the running
speed then five successful trials will be collected, at 3.35 m-s'1(8-minute mile pace).
Running track data collection (2400 m economy and time trial run)
• First test May 24-25th (Saturday and Sunday between 8.30 am - 8 pm)
• Second test June 6th(Friday between 8.30 am - 8 pm)
You will select a 30-minute time slot between those times that suits your schedule on
either day.
Procedure (what you need to do)

The warm up will consist of two laps at 8-minute mile pace. You will then be asked to
run an economy run of 2400 m at 8-minute mile pace while oxygen consumption is
collected by a portable gas analyser. Each 200 m section must be run evenly at a 1
minute per 200 m. Heart rate and time will be recorded from a polar heart rate monitor
for the run and at the finish for the time it takes for your heart rate to drop to 120 beats
per minute. Stride frequency will be recorded by a digital video camera for a 20 m
section each lap.
Immediately following the economy run after removing the portable gas analyser a time
trial will be run for 2400 m. Maximum, average heart rate and time will be recorded.
Stride frequency will be recorded by a digital video camera for a 20 m section each lap.
At the finish of the time trial run, the time heart takes to return to 120 beats per minute
will be recorded. You will be asked to rate how hard the run felt using a Borg scale.
Please note:
Regardless of which group I am assigned too I will maintain for the one-week study:
• To allow 48 hours of light training prior to the track testing days (this is essential
for valid data collection);
• To maintain a consistent diet and drinking procedure before testing and during
the testing period;
• To maintain a training log of my training during the one-week testing period.
• And then maintain a training log of any injuries for 3-months after the study.
Please contact Graham Fletcher at [email protected] or Rick Velati at

rick [email protected] before May 20th, 2003 if you are interested in taking part in
this study.
3
PARTICIPANT’S CONTACT SHEET
Name Contact phone # Time and day available for

training
4
Q U E S T IO N N A IR E
Name: Date:
Pre test:
Post test:
Please read carefully and circle he number that matches your opinion.
Please read the following item, and circle the number that best reflects your
agreement with the statement.
Running technique is defined as the technical details and skill (pattern

and sequence of movements) that determine the degree of excellence in
accomplishing running movement.
Question Not Fairly un In-between Fairly Extremely

important at important important important
all
How 1 2 3 4 5
important do
you believe
running
technique is
to overall
performance?
5
INFORMED CONSENT
INFORMED CONSENT
N am e:
A d d re ss:
Tel:
The full details of the tests have been explained to m e. I am d e a r about what will be
involved and I am aware of the purpose of the tests, the potential benefits and the
potential risks.
The test results are confidential and will only be communicated to others sue* a s my
coach if agreed in advance.
I have no injury or illness that will affect my ability to successfully complete the tests.
My p erm issio n to perform th is te s t is voluntary; I am free to s to p

th e te s t a t a n y point, if I s o desire.
I____________________________________________________________ a ck n o w led g e th a t
th e re a r e risks a n d d a n g e rs in h ere n t in physical e x e rc ise a n d d e c la re th a t I know
o f no re a s o n w hy I sh o u ld not c o m p le te th e te s ts d e sc rib e d . A ny q u e stio n s I
h a v e a b o u t th e te s t s e s s io n h a v e b e e n a n sw e red , a n d th e inform ation given o n
th e m edical questio n n aire is correct, a n d should b e tre a te d a s confidential.
S igned:____________________________________________ D ate:______________
S ig n a tu re of g u ard ian / p a ren t if u n d e r 18
S ig n a tu re o f te ste r:_____________________ D ate:

PR E-TEST M ED IC A L Q U E S T IO N N A IR E
PRE-TEST MEDICAL QUESTIONNAIRE

N am e:...................... j.................. ...................... ..............,........... ........... .......... .............
D ate of Birth:................ ............. A ge:................ S e x :......... ......................
Please circle the appropriate re sp o n se o r fill In the blank.
1. How would you describe your present level of activity?
Sedentary Moderately active Active Highly active
2. How would you describe your present (evet of fitness?
Unfit Moderately fit Trained Highly trained
3. How would you consider your present body weight?
Underweight Ideal Slightly over Very overweight
A. Smoking Habits Are you currently a smoker? Yes No

How many do you smoke ..perday
Are you a previous smoker? Yes No
How long Is it since you stopped? ....... ....... .......
Were you an occasional smoker? Yes No
perday
Were you a regular smoker Yes No
....perday
5. Do you drink alcohol? Yes No

If you answered Yes, do you have?
An occasional drink A drink every day More than one drink a day
6. Have you had to consult your doctor within the last six months? Y es No
If you answered Yes, please give details................. ........... ............ ..........
7, Are you presently taking any form of medication? Y es No

If you answered Yes, please give details .............................. ...............
0. As far a s you are aware, do you suffer or have you ever suffered from;
a Diabetes? Yes NO b Asthma? Yes No
c Epilepsy? Yes No d Bronchitis? Yes No
e ‘Any form of heart complaint? Yes No f Raynaud's Disease? Yes No
g ‘Marian's Syndrome? Yes No h ’Aneurysm/embolism? Yes No
1 Aneamia Yes No
9. Is there a history of heart disease in your family? Yes No

If Yes, please give details,,.,...,...
10. Do you currently have any form of muscle or joint Injury? Yes No
If Yes, please (jive details , ...........................................................................
11. Have you had lo suspend normal training In Ihe last two w eeks? Y es No
If the answ er Is Yes, please give details,................. ,......
12. As for os you ore aware, Is there anything that might prevent you from
successfully completing the tests that have been outlined to you? Y es No
If you answ ered Yes to any question pro coded by a * p lease ask your G P to
complote th e 'D octors C onsent Form* provided.
If blood sa m p les are to be taken, p lease a n sw e r th e following q u e stio n s;
a) Are you suffering from any known serious infection? Yes No

b) Have you had jaundice within the previous year? Yes NO
c) Have you ever had any form of hepatitis? Yes Ko
d) Are you HIV antibody positive? Yes No
e) Have you had unprotected sexual intercourse with any
person from an HIV high-risk population? Yes No
0 Have you ever been involved in intravenous drug use? Yes No
g) Axe you hemophiliac? Y es No
* If the answer to any of the above is Yes then please discuss the nature of th e problem
with your Sport and Exercise Scientist.
Signature:................... ........... ........................ ............... ..........
Signature of Parent or Guardian if the subject is under 10:
Date:.........................
PARTICIPANT’S PERSONAL DETAILS SHEET
Name 10 km time Age Current injuries
9
APPENDIX 2
RUNNING TRACK TESTS
TRACK DATA COLLECTION SHEET ECONOMY RUN
Date and tim e__________________ /__________________
Pre test: |
Post test:
Name/number Oxygen Stride Av HR Max HR Rec HR

frequency
10
TR A C K DATA C O LLECTIO N S H E E T T IM E TR IA L RUN
Date and tim e__________________ /
Pre test: |
Post test:
Name/number Time Stride Av HR Max HR Rec HR Borg

frequency scale
11
TRACK SIGN UP SHEET
The economy and time trial run will take pace at Loughborough university track. Please
select a convenient day and time for both pre and post tests.
Date Time Name Date Time Name

May 24th 8.30 am May 24th 2.30 pm
May 24th Mid-day May 24th 6.30 pm
May 24th 12.30 pm May 24th 7.00 pm

May 25th 8.30 am May 25lh 2.30 pm
May 25th Mid-day May 25th 6.30 pm

June 6th 8.30 am June 6th 2.30 pm
June 6th Mid-day June 6th 6.30 pm
June 6th 12.30 pm June 6th 7.00 pm
12
V ID E O DATA CO LLECTIO N SHEETS
Each participant had a test name/number for each test, which was filmed using a sheet of
paper before each economy and time trial run for the stride frequency data collection.
This enabled each participant to be clearly identified for each test. An example is given
below for participant 1 for each test.
Pre economy 1
Post economy 1
Pre time trial 1
Post time trial 1
13
Borg scale (Borg, 1970)
Original rating scale
6
7 Very, very light
8
9 Very light
10
11 Fairly light
12
13 Somewhat hard
14
15 Hard
16
17 Very hard
18
19 Very, very hard
20
14
APPENDIX 3
LABORATORY TESTS
ROUTE MAP
Sheffield Hallam University Online Visitors Guide * The Route to C... Page 1 of 2
Visitors Guide
►SHU h om epage The ro u te to Collegiate C rescent

kVisitors hom epage
kOur location W e can provide you with personalised driving .directions
kPlan your route to th e U niversity or u se th e general Information b elow .
kTravel information
i How to g e t to From Junction 33 of the Ml, follow signs for A630 Sheffield
■ City C #m pv* City Centre. Continue on the A630 until the dual carriageway
■ C re sc e n t ends at e roundabout with traffic signals (Park Square) © .
• P*alt*r L*nc
kCampus plans
■ City C am pu* Take the fourth exit from the roundabout signposted City
• Cre«c*M Centre, then the third left tu rn into Arundel Gate O . The
■ P u t U r U f t t __________
Crucible Theatre is on your right and the City Campus on your
left. Continue along the dual carriageway and go straight
ahead (second exit) at the first roundabout © .
At the second roundabout, take the third exit onto the ring
road © . Continue through th e traffic lights to the next
roundabout and take the first exit signposted A625
(Castlcton) © .
The Collegiate Crescent Campus is approximately 750 m etres

(1/2 mile) along on the right.
T ravelling b y b u s? take th e 8 1 ,8 2 , 83, 84, 85 or 86 from

Pmstone S treet (outside the Town Hall) If visiting the Ecdesall
Road end of the Campus. If visiting th e Clarkehouse Road
end of the campus, take the 50 (Dore) from the Transport
Interchange or Church S treet o r m e 60 (Ranmoor or Crimicar
Lane) from Church S treet or directly outside the train station.
For Clarkehouse Road Schools (Health and Social Care/Social
Studies and Law) you need to get off a t th e Haflamshfre
Hospital.
itlc C o u rt
rMPUSj
M
http://www.shu.ac.uk/services/marketing/visit/tocoll.html 23/05/2003
15
K1STLER FORCE PLATE CALIBRATION*.
f*■#*+&KPXJ LAIS LTT< P0M m
uA'fe c< py KJSTLER

ms *«-«»*MR
Mt jMKicI
K A ,U » t0 tm i4 m m m m tm n
m m m tm m m m
tit *m 9891CA
fcflo* s*dm
r, V
u s s z a ^ Utmtlmpi H 0 -B L s 9-.i3
W *B*W 4> Umitmpt 19 0_OJ5 0 -W 0~2
JNMH
iarii/aNrfJi I * 3 < I
n+T 0*4 HR 1W
leldl/MH* fJ4? t*79 1*K 1JM2 §JW &M4 0,*» « S i
•MKll/tefa!' 4493 &7SO 3,773 38*3' 1.433 9^27 1,618 t.804
m n& s m m itjm n ,m sn» syaw *bm t4*i
Mis*=1/AmrpI* *,* 37,056 37,569 37,683 3&J7S 19*90 19*05 tO,W 1*082
3*-T» '51*
f.-f, i,-.r, !.->. 1,-1, '.-J,
fhlpcta M i </ &a*l lap * 1. 0.3 0,7 -02 0.1 -0,1 *0t
J/&K*i topi * cgs : . *7 -0.1 0.1 -3,1 4&£
"MiCttlUKI •Odttwsv
K £3
% 43
7,
PKafccfttffc Mfl W mU iflt w ijfcflrIb p m sM jm im 0 — if m e m tmutk * >*mrt rtwfcr it

jg Vm>i\ fjMmtn TrdhiT(mSmA s I h m r lk ÔMrfâMirfiBMMp-irCaW1,
MimmWtii IO/4w M /W ft
. 5*»SR=*3ifitowirf t&**r *067*8 'iM W W W craS I
S®S27
fr—pgi IMr 4fefefc'9ia '3Tt8SS
... ...... . —
i’
16
PARTICIPANT’S LABORATORY BOOKING SHEET
Day/date Car driver Morning Afternoon Evening

Monday
May 26th
Tuesday
May 27th
Wednesday
May 28th
Thursday
May 29th
Monday
June 9th
Tuesday
June 10th
Wednesday
June 11th
17
LABORATORY DATA COLLECTION SHEET
Name: Date and time:
Weight: Height:
Filename:
Consent/medical: Pre-test:
Questionnaire: Post-test:
Trial Accept GRF hit Time EMG Clean
18
APPENDIX 4
POSE® METHOD INTERVENTION
POSE® RUNNING INFORMATION SHEET

(Dr. N. Romanov and Graham Fletcher M.Sc)
A general overview of the theory of the Pose® method can begin with the laws of
Thermodynamics. The law states that any system in nature goes from a higher order to
lower level of energy. Therefore, any system needs to be built by an outside force. A
plant needs the sun in order to build its cell structure. Running is the same, where the
structure (body) needs to be able to utilise higher order forces (gravity) and energy
(muscle contractivity and elasticity) in the most effective order. By making the most
effective use of gravity, the flow of energy loss is least costly. The continuous use of
muscle activity is limited by the body’s ability to breakdown adenosine triphosphate
(ATP). To accomplish an effective use of the available forces and energy in running, an
economical skilful technique is paramount. Specifically, this relates to where force
should be applied to continue movement, and the order in which these forces should be
applied.
A brief discussion of linear kinematics can highlight the Pose® method’s structure. The
force of gravity when released (potential energy) is transferred to kinetic energy. A
pencil held in the fingers above a table has potential energy and, when it is released it is
transferred to kinetic energy. Following the pencil example, a pencil balanced on its end
has potential energy, which becomes kinetic energy when it over-balances. The over
balancing of the human body from one support limb to the other, in running,
immediately has kinetic energy, and the more skilful application of this energy can result
in a more economical running gait. Kinetic energy can also be greatly increased by the
body’s velocity. The greater the velocity of the general center of mass (GCM) in flight
increases the body’s kinetic energy.
Focusing on the position of the support limb in relation to the GCM, the line of gravity is
restricted from exerting its pull due to the position of the support limb being forward of
the line of gravity at toe-off. The heel strike is also known as a breaking force. The term
breaking in running is illogical in itself. Momentum will continue to remain constant in
the absence of other forces, so breaking momentum each stride reduces forward velocity.
The Pose method teaches that the support limb must break contact with the ground
immediately:
1) to not hinder the gravitational line of pull
2) to maximise muscle elasticity, using the law of the coefficient of restitution- the
greater elastic value the higher energy return. This can only be achieved by a
rapid action, which is best achieved by a mid-foot strike.
The following, from a previous article, illustrate the actual body movements needed to
achieve this theoretical model.
19
1) S (spring-like) shape of the body (the “running pose”) - the element of running
technique used for utilizing muscular elasticity and integration of the body into a
single system.
2) Free falling - utilizes gravity as one of the main forces, driving the body forward
horizontally.
3) Changing support -integrates forces that remove the foot from the ground while
allowing gravity to pull the body horizontally forwards.
4) Pulling the support foot straight (vertically) up-wards using the hamstring
muscles - allowing the foot to break contact with the ground quickly.
Developing these concepts, further, beginning with point (1) it is essential to understand
that body position is integral to all movement. The body position determines the position
of the GCM and the GCM determines the direction of the whole body. Once the body
leaves the ground the path of the GCM cannot be altered, so the position the body is in
before take-off is essential. At foot-strike, the position of the body is paramount because
it affects how much the lower limb (particularly the knee) is loaded.
Point two, is interesting, because the need to apply an external force to initiate movement
is obvious, but the question is what force to utilise? Current thinking is to apply
Newton’s third law of motion and push off from the ground. Following this logic to its
natural conclusion, leads to wishing the runner, can push harder or faster, both requiring
greater strength. Many seeking greater strength have unfortunately turned to illegal
substances for the answer. However, there is another external force, which if utilised
correctly, can be more effective. The following quotation written in 1912 by Graham -
Brown in the British Medical Journal reveals the logic of this idea.
“7/ seems to me that the act o f progression itself- whether it be by flight through the air
or by such movements as running over the surface o f the ground-consists essentially in a
movement in which the centre o f gravity o f the body is allowed to fall forwards and
downwards under the action o f gravity, and in which the momentum thus gained is used
in driving the centre o f gravity again upwards and forwards; so that, from one point in
the cycle to the corresponding point in the next, no work is done (theoretically), but the
mass o f the individual is, in effect, moved horizontally through the environment. ”
Developing point three and four is again re-thinking our current running action ideas.
Muscle force is predominantly focused upon when coaching or discussing running, yet
another muscle property can greatly aid in limb movement. This is muscle elasticity. The
musculotendinous junctions in the feet, Achilles and quadriceps tendon allow for storage
and use of muscle elasticity, just as an elastic band can store elastic / strain energy when
stretched. The key to effectively utilsing this energy to move the legs, is the speed at
which the runner can change support. This means the speed at which you can remove
your foot from the ground before the elastic energy has dissipated from those
musculotendinous junctions.
This article has attempted to develop more fully some of the concepts of the Pose®
method of running and highlight their differences with current running theory.
20
POSE® IN T E R V E N T IO N R U N N IN G DRILLS
Summary of the week

Days 1-2 Developing the concept
Days 3-4 Reinforcing the concept
Days 5-7 Individual technique and skill development
Overall aims and objectives

Develop cognitive models, perception o f falling and pulling the foot from the ground and
finally auto correction of technique.
Practical daily outline

Introduce the concept: theoretically and practically. Utilise specific drills to gain a feel of
the concept of falling and pulling the foot from the ground. Video each participant too
aid learning. Give verbal and written feedback after each session.
Day 1
Short theoretical session in a classroom
There are four forces involved in running: gravity, ground reaction force, muscle force
and muscle elasticity. Gravity, ground reaction force and muscle elasticity are free in
reference to internal energy costs. Pose questions on how does gravity work in running
and which external force moves the body forwards?
Show body tipping and falling forwards and assess muscle forces involved. Clarify
gravity causes the tipping and no muscle forces were needed to fall. Explain then how to
continue moving forwards by pulling the foot from the ground. Explain the use of
muscle elasticity and its role in aiding pulling the foot from the ground. Emphasise the
timing of falling and pulling the foot from the ground through the key concept of
shoulder, hip and ankle vertical alignment.
Key to learning:
Increase participant’s perception of the movement. Ascertain how they felt after each
drill and running activity.
1) Use Pose® biomechanical model as standard to compare against.
2) Develop their perception but note their perceptions may be wrong so increase
the correct perception.
3) They have to perceive two things: to feel falling and to pull the foot from the
ground immediately they begin to fall forwards.
Drills
Warm-up: Video each participant running prior to intervention.
Falling position:
• Stand on heels and try and fall forwards. Try the same thing with the leg behind
and in front.
Interaction with support:
• Stop participants in a freeze frame. Ask them where their weight is on their feet.
• Lift heels to see if weight is on the ball of the foot.
• Unlock knees and rapidly lift heels
• Keep weight on the ball of the foot at all times.
Move the body as an integrated system:
21
• Push participant back and forth and side-to-side while maintaining an integrated
body position.
Weight position in relation to foot and centre of mass:
• Place a hand on their chest and take the participants weight as they fall forwards.
Feel weight move from foot to chest.
• Repeat but let go this time.
• Repeat but demonstrate how small a lean is needed to fall forwards.
• Repeat and show where participant’s foot lands in front of their body.
• Repeat, but ask them to pull their foot as they fall.
Feel pull of the foot from the ground:
• Hold their heel as they pull the foot from the ground
• Push foot down as they resist.
Video running and give feedback.
Summary:
Understand the model of gravity’s work on the body, body leads leg and the foot is
pulled from the ground as they fall forwards.
Day 2
Perception drills for falling
• Hand on belly button and feel vertical relationship to the ball of the foot.
• Repeat and fall forwards.
• Repeat and feel how small and angle is needed to fall.
• Run with fingers on belly button.
• Repeat with eyes closed; use partner.
• Arms stretched out behind the back and run.
• Hands pushing on hips and run.
• Hand on chest with partners and run.
• Partners fingers on back and run.
Range of motion of the lower-limb:
Emphasise a decreased range of motion. Show using running shoes the position o f the
foot, on landing and flight and impact again. Do not drive the leg forwards.
• Run and video for feedback
Rubber bands to illustrate a correct leg action
• Leg behind and in front as they run with the bands attached to their ankles.
• Run with partners in front and behind with hands on their shoulders.
Reduce effort needed by only using hamstrings to pull the foot.
Reduce effort needed by only using the minimal amount of hamstrings to pull the foot.
Technique problems to look for:
• Landing ahead of the centre of mass with the foot; do not drive with hip flexors.
• Landing on toes; feel ball of the foot on landing.
• Landing rigidly; relax and land with a neutral ankle.
• Landing on the sides of the foot; use pull up toes drill.
• Landing on the sides of the foot and rigidly; use Pony, tapping, slow light
running.
• Landing hard, decelerate foot with hamstrings.
Hips and muscle integration:
• Standard hip drill, front back, behind and sides and run after each one.
• Standing and push person from all sides while holding them. Ensure body
remains integrated.
• Run and video for feedback.
22
Body integration drills: check their perception:
• Partner running with eyes closed. Feel lightness and integration.
• Push from behind and resist with whole body and then run.
• Press back on partners hands hard and then run.
• Run while partner pressing down on their head to feel no vertical movement.
Summary:
Reinforce their perceptions. Can they feel falling and pulling of the foot from the
ground? Do they feel lightness and body integration?
Day 3
Reinforcing the concept
Increase their perception for falling and speed of pulling the foot from the ground.
• Run with arms in front and behind.
• Partners push the shoulder from the side intermittently to check rapid change of
support while running.
Pulling:
• Use rubber bands to increase feel of hamstring work.
• Standing band work and running with bands
• Short sharp downhill run to feel pulling action
• Individual holds bands and pulls vertically upwards from the shoulders and push
out to the side for increased tension. Pull foot vertically upwards.
Pattern of movement reinforcement:
• Pony drill
• Tapping drill
• Skipping drill
• Front lunge drill
• Run 200 m and video for feedback
Summary:
Can they feel falling as they run? Can they pull the foot from the ground as they run?
Day 4
Reinforce fall and pull
• Press-up position without bands face down and upwards. Pull for hamstring
work.
• Press-up position with bands face down and upwards. Pull for hamstring work.
Run 400 m and video for feedback:
• Use a metronome or count to develop stride frequency of over 180 per minute.
Observe mistakes:
• Keep knees flexed; do not extend leg at toe-off
• Maintain a vertical alignment on landing.
• Do not land ahead of the centre of mass.
• Do not leave the support leg behind.
• Fall.
• Do not fix the ankle on landing.
• Hips not integrated.
Look for:
• Lightness.
• Body integration.
• Pose position on landing
• Ease of running
• No pressure, tightness or pain.
• Fall and pull action.
23
Summary:
Give specific drills for each individual from feedback.
Days 5-7
Individual technique development
Day 5
Warm-up:
• Pony, tapping and skipping
• Cross steps.
Drills:
• Jumps
Run and video for feedback:
Run on gravel to emphasise pulling of the foot.
Summary:
Give individual drills and comments.
Day 6
Partner work with hips and hamstring and run.
Run and video give feedback individually.
Summary:
Give individual feedback.
Day 7
• Jumps on one leg; movement in the hips not knee.
• Feel hip, knee, ankle and ball of foot light and loose.
• Harder hip work with partners
• Run and video
• Show them day 1 and today for comparison.
Summary:
Final comments and feedback.
24
CONTROL GROUP’S DRILLS
Please read carefully the description of the drills and the daily format. Follow the daily
format on your own for each of the seven days.
Drill description
Marching drill
Raise the right knee to hip level with each stride
Butt kicks
Daily routine
Please complete each day on your own. Repeat drills for each leg for as long a distance
or time as you can accommodate comfortably while holding form.
Day 1 Marching drill, high knee lifts and butt kicks repeat four times for
each leg.
Day 2 Marching drill, high knee lifts and kicks outs repeat four times for
each leg
Day 3 Marching drill, kicks outs and bounding repeat four times for each
leg
Day 4 Butt kicks, kick outs repeat six times for each leg and bounding
four times 50 m.
Day 5 Butt kicks six times, high knee lifts repeat four times for each leg
and bounding four times 50 m.
Day 6 Marching drill repeat four times, high knee lifts four times and
bounding six times 50 m
Day 7 Butt kicks six times, kick outs four times for each leg and
bounding six times 50 m.
25
TESTIMONIAL
Testimonial
Name: ■.;. D*Uk: .'ii^s>i
PJG13 C wrile a statement about your c x f C K m o f teanuag; tl#c Po*c method o f funnbg.
A~b-,“ MM. kûtH <Acu^ e-| bcu^l»v^ »c Pr.se t

f e l t ,,/Ji C Q ^ jA tg . 1.i t WY ■..
■■" b w rn r n. k± c c ^ " c*sA . tVvv^ U">ftA tjvjs,
l .,,££x /^.. O cLp -^. c1L j ...
11 *v S
uftfwia
VXLCftfr.I
ijSV . .........1^ .11,III,jP f i l
kfL&.-Q---£..^:,,fl£i.Li^j ..£»lK
jys>. UcXA7-„cf C.| lU_*
ruLf^L’va. ...Â: ■ sJ1
-1 T ■
tr\ firaC nc&SL rJ- Ikk,^
' T— ■
..kCi, —I Q l lA.—la: ikk 1_

Jd L ... gCL. ^ 1 ^ . f u A rV lg W m E .: r i^ V .y i: _£u
.Ccx?JLdz'Li. . Ctt . :trryn3L*s£ «yvj -^vêieLj V»C"»*<

I cl>.S*i|y»..-A£A >fUg e«~ 1 ^
I Jhfc&L...,...*
■; , . . .. V~t'?wCI I ^.|'iChihin
ty • j . t z . ’p 4ug. p .:..v a e .^ . ^ 4
o W <)f^.p btf+c.... pc*ix in r, I %Kr^jA .ha,*.c.

f r X v .L .* =.•*,f r C V r l4 f t \ «.J g & lX r .it U f i . : t ■■ ^ a « o y t a . __
> < > ^ X 3 » k r. jg .— _ . e |- j a r i i J b t l. t .... f e t A m> Ik rri / , * < \Jfu J i\f. ^ ____
■ T -ii. I ■ i T X .1 . * - .r \f i. f y ^ C ? - £ .[A f____* * * ^ ) 11 n j A s - l - r j a i . _ £ ^ L
Jba&u C" rACclAt f,. ^ t Kg ^.:^- fi-xr ^.r .jCc/N ^
^ n/
S C Ci- O m - ; tti lT h ~ /
26
INJURY RECORD SHEET
Name
Date Cause of injury Type or Days off running

description of the
injury
27

The Posertm Method: A Biomechanical and Physiological Comparison With Heel-Toe Running

Uploaded by

Copyright:

Available Formats

The Posertm Method: A Biomechanical and Physiological Comparison With Heel-Toe Running

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Posertm Method: A Biomechanical and Physiological Comparison With Heel-Toe Running

Uploaded by

Copyright:

Available Formats

The PoseRTM method : A biomechanical and physiological

comparison with heel-toe running.

Copyright and re-use policy

Sheffield Hallam University Research Archive

All rights reserved

INFORMATION TO ALL USERS

In the unlikely e v e n t that the a u thor did not send a c o m p le te m anuscript

Published by ProQuest LLC(2017). C opyright of the Dissertation is held by the Author.

All rights reserved.

p h y sio lo g ic a l c o m p a ris o n w ith h e e l-to e

A thesis submitted in partial fulfilment o f the

1.1 INTRODUCTION TO THE PROBLEM 1

CHAPTER 2: LITERATURE REVIEW 9

CHAPTER 3: THEORETICAL MODELS 54

CHAPTER 4: EXPERIMENTAL PROCEDURES 93

CHAPTER 5: EXPERIENCED POSE® AND HEEL-TOE

RUNNER’S STUDY 107

5.1 INTRODUCTION 107

CHAPTER 6: POSE® METHOD TRAINING STUDY 163

6.1 INTRODUCTION 163

CHAPTER 7: CONCLUSIONS 193

7.1 THEORETICAL MODELS 193

RISK ASSESSMENT PROFORMA 1

APPENDIX 2: RUNNING TRACK TESTS 10

TRACK DATA COLLECTION SHEET ECONOMY RUN 10

APPENDIX 3: LABORATORY TESTS 15

ROUTE MAP TO SHEFFIELD HALL AM UNIVERSITY 15

APPENDIX 4: POSE® METHOD INTERVENTION 19

POSE® RUNNING INFORMATION SHEET 19

APPENDIX 5: RAW DATA (SEE ATTACHED CD DISK)

Number Title Page

4.4.1 Ryan-Joiner’s (similar to Shapiro-Wilk) normality test. 102

6.3.1 Participant’s personal data. (N = 16). 168

2.2.1 Scatter-plot of running economy versus hamstring-gastrocnemius co­

3.3.1 Flight, take-off and landing distance totalling the runner’s

3.3.3 Results of phasic quadriceps electromyography and knee flexion

-extension for a typical endurance runner averaged over six foot

strikes (adapted from McClay et al., 1990). 64

shock of contact with the ground (absorption). The second

peak (vertical dashed line) beginning at or about mid-stance

is associated with both body mass and muscle force generation. 67

contact to right foot toe contact of the same foot. 70

3.3.6 The running Pose (Romanov, 2002) a) impact to maximal vertical

ground reaction force: shoulder-hip-ankle vertical alignment,

which potentially allows gravity to maximise the stretch-shortening

reflex and enables the whole body to work as a unit, b) A

gravitational torque accelerates the body while the hamstring muscle

group pulls the support foot from the ground. 73

gravitational force. CM = centre of mass; Fg = gravitational force,

support limb in the clockwise direction, r = position vector of the

centre of mass from the support limb, b) Fg = gravity, and its

3.3.9 a) Ground reaction force torque about the centre of mass

b) The ground reaction force and the gravity vector taken at

the same point of stance as the vector product in figure 3.3.8. 78

3.3.10 Gravity and ground reaction force vectors with gravity's

components included, showing the net force is Fp. Notice Fr

is equal and opposite to the ground reaction force vector, leaving

as O a) Rectangle upright representing the runner at mid-stance with

The rectangle (runner) tips forwards via momentum. A gravitational

2.2.1 Scatter-plot of running economy versus hamstring-gastrocnemius co