Article Summary 3 & 4

Description

Overview

The ability to understand and analyze scientific journals and papers will help you strengthen your critical thinking and expand your knowledge. Now, you have the opportunity to study scientific articles and summarize them. Answer the questions associated with the article. Be sure to explain your reasoning and answer all parts of each question. The average response for each question should be 70 words. Upload your answers in the summary format.

Summary Format

______________________

Title of The Article

Question 1

-Answer 1

Question 2

-Answer 2

Question 3

-Answer 3

Question 4

-Answer 4

Article 3— Barefoot Running

What stood out to you as the most significant difference between barefoot and shod running? Why is it significant?
If you were an orthopedic surgeon, would you recommend barefoot running? Why or why not?
If you were a competitive distance runner, would you recommend barefoot running? Why or why not?
What factors do you think might lead to different preferred foot strike positions between people?

Article 4— Alpine skiing and Back Injuries

How was data collected in the field to measure trunk movements and GRF? What are the benefits of collecting data in the field vs. in the lab?
What is the relationship between frontal bending, lateral bending, and spinal torso and the loads put on the back? At what part of the ski turn are these variables the greatest?
How can the findings of this study be applied actually prevent possible back injuries in alpine skiers?

Unformatted Attachment Preview

JOURNAL
OF
ELSEVIER
BIOMECHANICS
Journal of Biomechanics 33 (2000) 269}278
www .elsevier.com/locate/jbiomech
Biomechanical analysis of the stance phase during
barefoot and shod running
Brigit De Wit!, Dirk De Clercq!,*, Peter Aerts”
!Department of Movement and Sport Sciences, University of Ghent, Watersportlaan 2, B-9000 Ghent, Belgium
“Department of Biology, University of Antwerp (UIA), Belgium
Received 6 February 1998; accepted 13 October 1999
Abstract
This study investigated spatio-temporal variables, ground reaction forces and sagittal and frontal plane kinematics during the
stance phase of nine trained subjects running barefoot and shod at three di!erent velocities (3.5, 4.5, 5.5 m sv1). Di!erences between
conditions were detected with the general linear method (factorial model). Barefoot running is characterized by a signi”cantly larger
external loading rate than the shod condition. The #atter foot placement at touchdown is prepared in free #ight, implying an actively
induced adaptation strategy. In the barefoot condition, plantar pressure measurements reveal a #atter foot placement to correlate
with lower peak heel pressures. Therefore, it is assumed that runners adopt this di!erent touchdown geometry in barefoot running in
an attempt to limit the local pressure underneath the heel. A signi”cantly higher leg sti!ness during the stance phase was found for the
barefoot condition. The sagittal kinematic adaptations between conditions were found in the same way for all subjects and at the three
running velocities. However, large individual variations were observed between the runners for the rearfoot kinematics. ( 2000
Elsevier Science Ltd. All rights reserved.
Keywords: Barefoot running; Ground reaction forces; Sagittal plane kinematics; Frontal plane kinematics; Kinematic adaptation
1. Introduction
Nowadays running can be considered one of the most
important recreational activities. Since most people are
running shod, many scienti”c studies investigated the
in#uence of alterations in the properties of the shoe on
the running style. Ground reaction forces and kinematic
variables were found to vary with shoe hardness and shoe
geometry (Nigg, 1986; Nigg and Morlock, 1987; Lees,
1988; Pratt, 1989; Edington et al., 1990; Van Woensel and
Cavanagh, 1992; McNair and Marshall, 1994). The relationship between kinematic and external and internal
kinetic variables was also studied using dynamic simulation models (Denoth, 1986; Gerritsen et al., 1995; Wright
et al., 1998).
But there are still many aspects concerning the manner
in which athletes adapt to di!erent surfaces, shoes and
other boundary conditions which are not well understood. This insight could be enhanced by studying the
* Corresponding author. Tel.: #32-92646322; fax: #32-92646484.
E-mail address: [email protected] (D. De Clercq)
di!erence in kinematics between barefoot and shod running since in these situations boundary conditions in#uencing running kinematics are manipulated. Barefoot
running can be seen as a running condition wherein
external protection and shock reduction is minimal. So,
alterations in running style are expected to be more
pronounced than when comparing di!erent shod conditions.
Until now, several authors studied barefoot running
but con#icting results were presented in literature, probably because of limited samples. However, all studies
agreed with the fact that the external loading rate is
signi”cantly larger in the barefoot condition (Dickinson
et al., 1985; Komi et al., 1987; Lees, 1988; De Clercq et al.,
1994). Concerning the sagittal plane kinematics a more
extended body position and a smaller touchdown velocity of the foot were found during barefoot running (De
Koning and Nigg, 1993).
The aim of the current study is to provide a comprehensive description of barefoot running using a statistical
representative data set, and to compare barefoot with
shod running. Therefore, spatio-temporal variables,
ground reaction forces and sagittal and frontal plane
0021-9290/00/$ – see front matter ( 2000 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 2 1 – 9 2 9 0 ( 9 9 ) 0 0 1 9 2 – X
270
B. De Wit et al. / Journal of Biomechanics 33 (2000) 269}278
kinematics of barefoot and shod running at three di!erent velocities are analyzed and compared. In that way,
systematic adaptations in the running style can be detected and hypotheses about the underlying mechanisms
will be formulated.
2. Method
2.1. Subjects and experimental protocol
Nine trained male long distance runners
(30}40 km week~1) were tested while running barefoot
and shod (neutral jogging shoe; Adidas 033153, T-response) at three di!erent velocities (3.5, 4.5 and
5.5 m s~1). All of them were free of injuries at the time of
the experiment. They were informed about the procedures and signed an informed consent. The average characteristics were: age: 27.3$9 yr; height: 1.78$0.07 m;
body mass: 70$9 kg; shoe size: UK 8.9$1.5.
The three orthogonal components of the ground reaction forces were measured with a Kistler force plate
(frequency of resonance’800 s~1, 12bit A-D conversion
at 1666 s~1) mounted in the center of a 30 m indoor
tartan runway and connected with a PC (Ariel Performance Analysis System Inc) to obtain online a graphical
representation of the analogue signals (Fig. 1). Running
velocity was computed from the time interval measured
by infrared photocells mounted at shoulder height in
both directions 2.25 m from the center of the force platform (a deviation of 5% of the intended velocity was
permitted).
Foot movements were video-taped in the sagittal and
frontal plane. Sagittal plane images of the right leg were
obtained from two high-speed video-cameras. One (Nac
500; 250 Hz) provided images of the whole body, while
another (Nac 1000; 500 Hz) showed a detailed view of the
shank and foot during the stance phase. A dorsal camera
30m
~- — — — — — —– —- – – — – — — — — – —- – – — – — — – – – ->
~- —-4.:~~- — >
–~>~
IRP
IRP
I
I
~C;:··j·-,,’;’:’~’:’~T:~:J”””c;::::J”””””
I
::
”
I
: : 4.5 m
IRP infrared photo cells
FP force platform
CM contact mat
7.9m; :
:a
”
cl
NAC 1000
v
NAC400
[J VHS 25
Fig. 1. Experimental setup.
(Nac 400) obtained frontal images at 200 Hz. Five LED
lights were placed in the visual plane of all videocameras. These LED lights were electronically activated
with a time di!erence of 0.001 s each. Using these light
signals, the video-cameras and the force plate were synchronized with a precision of 0.001 s.
Each runner was given enough time to warm up and
become familiar with the speci”c condition and velocity.
The runners contacted the force plate with the right
foot without altering their technique. This was checked
visually where the most important criterion was an
equal braking/propulsive impulse exerted during foot
contact (Nigg, 1986). Subjects performed the test until
10 good trials for each condition and velocity were
made.
For seven runners, additional tests were performed
while running barefoot at 4.5 m s~1. A pressure mat was
placed on top of the Kistler force plate in order to
measure dynamic local pressures (Footscant system,
sampling frequency of 200 Hz, 4 sensors cm~2). The local
pressure underneath the heel was obtained by averaging
the pressures measured by the sensors located underneath the tuber calcaneum.
2.2. Collection of data
Step length was calculated with an accuracy of 0.01 m.
The inverse of the step time between heel contact of right
and left foot as seen on the frontal plane video-images
(accuracy 0.005 s) gave the step frequency.
The three orthogonal components of the ground reaction forces were measured. The variables of the vertical
forces that were analyzed are described in Table 1.
Markers in the sagittal plane were placed on the
skin according to Bobbert et al. (1992) (Fig. 2a). The
marker on the hip was placed 0.02 m proximal to
the greater trochanter, at the knee joint in the center
0.02 m above the tibial plateau and at the ankle joint
at the lateral malleolus, 0.005 m anterior to its tip.
The shoulder marker was put at the height of the acromion.
In the barefoot condition, foot markers were placed at
the tuber calcaneum and at the “fth metatarsal joint. In
the shod condition, those markers were placed on the
shoe at the height of those landmarks. Relative and
absolute joint- and segment-angles were measured as
indicated in Fig. 2a.
Statistics were applied on numeric values collected at
distinct points: at 30 ms before touchdown; at touchdown (t”t ); at the time of the “rst vertical impact force
0
(t”t ); at the end of midstance (i.e. the moment the heel
*
leaves the ground) and at push-o! (i.e. the moment the
forefoot leaves the ground; t”t ).
#0/5
To analyze the foot and ankle movements in the
frontal plane, markers were placed according to Clarke
et al. (1983); Fig. 2b): (a) in the middle of the heel; (b) on
B. De Wit et al. / Journal of Biomechanics 33 (2000) 269}278
Table 1
Description of the calculated variables of the ground reaction forces,
frontal plane kinematics and pressure measurements
Variable
Description
Unit
F
z*
amplitude of the “rst vertical
impact force
time of occurrence of F
z*
maximal vertical loading rate
between touchdown and t
*
amplitude of the minimal
vertical force
time of occurrence of F
zmin
amplitude of the vertical active
force peak
time of occurrence of F
z!
total contact time of the
stance phase
change of the calcaneal eversion
from touchdown to the occurrence of the impact peak
the maximal pressure measured
by sensors located underneath
the tuber calcaneum
N; body weight (BW)
t
*
G
z*
F
zmin
t
.*/
F
z!
t
!
t
#0/5
*c
*.1
local
pressure
s
BW s~1
BW
s
BW
s
s
deg
N cm~2
B
A
b
a
Fig. 2. Placement of the markers in the sagittal (A) and frontal plane
(B;) with the calculated angles.
the upper part of the calcaneus; (c) on the Achilles tendon
at the height of the malleoli; (d) 15 cm above c in the
middle of the leg.
The temporal evolution of rearfoot angle (c) and
Achilles tendon angle (b) were determined. The rearfoot
angle (in deg) was measured from AB to the vertical
plane, where a negative value points at calcaneal eversion. The Achilles tendon angle was measured from AB
to CD, where a negative value points at eversion. According to the Cavanagh/Clarke convention (Edington et al.,
1990), the latter relative angle is used to describe subtalar
inversion and eversion.
Video-images were digitized using the Ariel Performance Analysis System for two-dimensional calculation.
The frontal plane kinematics were “ltered with a low-
271
pass digital “lter with a cuto! frequency of 18 Hz, a
procedure recommended in literature about rearfoot
kinematics (Hamill et al., 1994). For sagittal plane kinematics, no digital “ltering was used. Previous studies
showed incorrect values of the second derivative at the
endpoints of the data set when using a digital “lter or
cubic spline, whereas better results were obtained with
the quintic spline routines (Woltring, 1985; Wood,
1982). A quintic spline routine was used in the current
study to smooth the x- and y-coordinates of the sagittal plane variables. The standard error of the smoothing
procedure was individually determined and varied
between 0.12 and 0.2 cm. The frame before, during
and after touchdown of the heel was not smoothed.
In this way, the sudden decrease of heel velocity caused
by collision of the foot with the ground was not attenuated.
A possible limitation of the current study is the use of
a two-dimensional technique. Areblad et al. (1990) reported that most two-dimensional angular values measured from a posterior view were very sensitive to the
alignment angle between the foot and the camera axis.
However, small errors were observed for relative pronation angles during midstance. A comparative statistical
analysis of Van Gheluwe et al. (1995) between a two- and
a three-dimensional approach of calculating rearfoot
kinematics showed comparable results for the variables
between touchdown and midstance. Additionally, a
two-dimensional method is a standard technique in
literature for evaluating rearfoot kinematics. If properly
applied, reliable data can be obtained (Edington et al.,
1990; Hamill et al., 1994). Nevertheless, the lack of
information about longitudinal rotation of the leg segments is a limitation of the two by two-dimensional
method.
2.3. Statistical design
The ground reaction forces were calculated for all 10
trials, from which 5 trials for each condition and velocity
were analyzed for the kinematic results. This yields to
a dataset of 540 trials for the ground reaction forces and
270 for the kinematic results. Mean and standard deviation were used to describe the individual and general
results. Statistical e!ects of velocity and condition were
tested with the statistical package SPSS (SPSS inc.) using
GLM (general linear method), factorial model, with a
signi”cance level p40.05. Both interactions between factors and the e!ects of individual factors were investigated. All trials were included and treated as repeated
measurements. The Pearson product moment correlation was used to provide single correlations, while a linear regression (stepwise method) gave the coe$cients of
a linear equation, involving several independent variables that best predict the value of one dependent variable.
272
B. De Wit et al. / Journal of Biomechanics 33 (2000) 269}278
Vertical Ground Reaction Forces
shod) was observed for the amplitude of both the impact
force peak (F ) and the active force peak (F ). The
;*
;!
results are presented in Table 2. The maximal local pressure underneath the heel was for the barefoot condition
97$35 N cm~2 (while running at 4.5 m s~1).
4
~
~
0
50
100
150
Time (ms)
200
3.3. Sagittal and frontal plane kinematics
Mean sagittal plane joint and segment angles for all
velocities are presented in Table 3. Fig. 4 shows the mean
stick “gure for all subjects at 4.5 m s~1 at four discrete
time intervals: (A) touchdown; (B) at the time of the
vertical impact force peak; (C) at the end of midstance
(the moment the heel leaves the ground); (D) at toe-o!.
Both the occurrence of the impact peak and the end of
midstance are reached signi”cantly faster for barefoot
running than for shod running (Tables 2 and 3). Most
statistical di!erences are found at the level of the distal
segments during the initial contact of the foot.
Concerning the frontal plane kinematics, a signi”cantly smaller initial eversion at impact (*c ) was ob*.1
served for the barefoot condition. All the other variables
describing rearfoot kinematics (c) and subtalar eversion/inversion (b) displayed signi”cant interaction e!ects
of subjects with condition.
250
Fig. 3. Vertical ground reaction curves of 1 representative person (1
trial barefoot and 1 trial shod) at a velocity of 4.5 m s~1.
3. Results
3.1. Spatio-temporal variables
The results are presented in Table 2. For all the tested
velocities, runners take signi”cantly smaller steps at
a higher frequency for the barefoot condition and a shorter contact time was found.
3.2. Kinetic variables
4. Discussion
Fig. 3 shows the representative curves of one runner
for the vertical ground reaction forces at 4.5 m s~1. Barefoot running is characterized by a signi”cantly larger
loading rate than in shod running and, in general, more
than one impact peak was found for the barefoot condition. No signi”cant main e!ect of condition (barefoot-
In the current study di!erences in kinematics and
in ground reaction forces, between running with and
without running shoes, were studied in order to gain
more insight in the adaptation of athletes to changes
in the mechanical characteristics of the foot}ground
interface.
Table 2
Spatio-temporal and kinetic variables (means and standard deviations of nine persons, 10 trials; BW”body weight; v”signi”cant main e!ect of
velocity (p(0.05); c”signi”cant main e!ect of condition (p(0.05))
N”9
3.5 m s~1
10 trials
Bare
Step freq
Step length
t
#0/5
t
&-*’)5
F
z*
t
*
G
z*
F
zmin
t
.*/
F
z!
t
!
(s~1)
(m)
(s)
(s)
(BW)
(s)
(BW s~1)
(BW)
(s)
(BW)
(s)
4.5 m s~1
Shod
Bare
5.5 m s~1
Shod
Bare
Shod
M
SD
M
SD
M
SD
M
SD
M
SD
M
SD
2.74
1.28
0.239
0.127
1.8
0.014
409
1.2
0.030
2.6
0.094
0.17
0.08
0.008
0.023
0.3
0.005
139
0.2
0.005
0.2
0.008
2.64
1.33
0.251
0.129
1.9
0.038
91
1.7
0.048
2.8
0.104
0.18
0.09
0.011
0.026
0.3
0.006
35
0.2
0.004
0.1
0.007
2.87
1.57
0.200
0.151
2.4
0.011
575
1.5
0.026
2.9
0.081
0.20
0.13
0.008
0.023
0.5
0.005
203
0.2
0.006
0.2
0.009
2.73
1.61
0.219
0.151
2.3
0.033
123
1.9
0.045
2.9
0.092
0.21
0.12
0.014
0.025
0.4
0.005
48
0.3
0.004
0.2
0.008
3.03
1.85
0.175
0.156
2.8
0.008
731
1.6
0.023
3.0
0.071
0.19
0.14
0.011
0.020
0.1
0.003
307
0.4
0.005
0.2
0.008
2.85
1.92
0.193
0.156
2.8
0.030
186
2.1
0.043
3.1
0.082
0.14
0.11
0.012
0.019
0.6
0.004
87
0.3
0.004
0.2
0.005
v
v
v
v
v
v
v
v
v
v
v
c
c
c
c
c
c
c
c
B. De Wit et al. / Journal of Biomechanics 33 (2000) 269}278
PUS H-OFF
MID STANCE
INIT IAL FOOT CONTACT
touchdown
impact peak
end of
midstance
A
B
C
toe-off
D
BARE
SHOD
Fig. 4. Mean stick “gure at a velocity of 4.5 m s~1. H”signi”cant main
e!ect of condition (p(0.05). The barefoot condition is shown as the
black line and the shod condition as the grey line.
In barefoot running, a signi”cantly larger loading rate
during impact (G ) was found, agreeing with results of
;*
previous studies (Dickinson et al., 1985; De Koning and
Nigg, 1993; De Clercq et al., 1994). Two studies demonstrated this G to be a prime correlating variable with the
;*
subjects perception of impact severity (running: Hennig
et al., 1996; simulated impact with human pendulum:
Lake and Lafortune, 1998). Therefore, based on this
literature, one could assume that the runners will adapt
their running style in an `impact-reducinga way when
running barefoot. However, results of the current study
do not con”rm this, as will be discussed in the impact
section.
The characteristics of the vertical ground reaction
force during the impact phase (at the most the “rst 0.05 s
of stance) depend upon the initial conditions at touchdown and, subsequently, upon the way the segments of
the body are decelerated during the impact phase (Bobbert et al., 1992). The results will be discussed in this
order. After this impact-related discussion, attention will
be paid to more general kinematic changes during the
entire foot contact phase.
4.1. Impact
Regarding the sagittal plane kinematics at touchdown,
most statistical di!erences are found at the level of the
distal segments (see Fig. 4A). In barefoot running, placement of the foot is signi”cantly more horizontal than in
the shod condition: the absolute di!erence for the sole
angle between the two conditions is 143 at 4.5 m s~1 (123
at 3.5 m s~1;153 at 5.5 m s~1). This #atter foot placement
results from a signi”cantly larger plantar #exion of the
ankle and a signi”cantly more vertical position of the
shank in the barefoot condition. The latter is caused by
a larger knee #exion because there is no di!erence in
thigh orientation at touchdown between barefoot and
273
shod running. This larger knee #exion was also found by
De Koning and Nigg (1993).
Some studies adopted a modelling approach to investigate the relationship between the initial kinematic conditions at touchdown and the vertical impact force peak.
These dynamic simulations showed that impact loading
can be reduced by decreasing the vertical momentum of
the caudal body parts (Gerritsen et al., 1995) and by
adopting a touchdown geometry that favours deceleration (Denoth, 1986; Gerritsen et al., 1995). Surprisingly,
in barefoot running, the subjects of the present study ran
with a non-di!erent vertical touchdown velocity and
with a #atter foot placement (more plantar #exion),
which does not concur at all with a strategy to reduce the
severity of impact.
It is interesting to note that this more horizontal foot
placement is prepared well before touchdown. In the
barefoot condition, the ankle is already signi”cantly
more plantar #exed at 0.03 s before touchdown and the
knee becomes signi”cantly more #exed from 0.02 s before
touchdown. A linear regression was calculated between
the sole angle at touchdown as dependent variable and
shank segment angle and ankle angle, both at 0.03 s
before touchdown, as independent variables (see Table 4:
r”0.92 barefoot; r”0.93 shod; both p(0.05). So, the
joint con”guration of the leg at touchdown is prepared in
free #ight, implicating an actively induced adaptation
strategy to barefoot running.
The #atter foot placement in barefoot running could
be explained by another functional demand. In previous
research it was shown that at “rst ground contact the
heel pad is suddenly deformed to a physiological maximum when running barefooted (De Clercq et al., 1994).
The deformation of the fatty heel tissue is proportional to
the local stress acting on the plantar side of the bare heel.
This means that for a given vertical impact force * the
F being non-di!erent in barefoot/shod*, the local
;*
pressure on the heel can be reduced by adopting a #atter
foot placement, through which initial ground contact
covers a larger plantar area. In this way overloading of
the heel could be prevented. Indeed, in the barefoot
running condition the maximal local pressure underneath the heel correlates negatively with the sole angle at
touchdown (r”!0.7, p(0.05). The more horizontal
the foot, the smaller the maximal pressure acting on the
heel.
It is also remarkable that the horizontal component of
the touchdown velocity is signi”cantly smaller in barefoot running. This could give rise to a reduction in shear
forces acting on the heel, but in the current experiment
we could not obtain reliable fore-after ground reaction
forces during the impact phase. Nevertheless, as sensation of mechanical inputs and pain is well established in
the foot sole (Bojsen-Moller and Jorgensen, 1991) it is
assumed that runners adopt a #atter foot placement in
barefoot running in an attempt to limit the local pressure
N”9
3.5 m s~1
5 trials
Bare
Shod
M
Sole angle
(deg)
Shank
segm angle
(deg)
Thigh
segm angle
(deg)
Knee angle
(deg)
Knee angle
velocity
(deg s~1)
Max knee #exion
(deg)
Time of max knee #exion (%)
contact time
Heel vel
(m s~1)
Hor heel vel
(m s~1)
Vert heel vel
(m s~1)
SD
5.5 m s~1
Bare
M
SD
Shod
M
SD
Bare
M
SD
Shod
M
SD
M
SD
1.7
0.102
79.1
48.9
19.9
1.9
0.005
4.1
9.4
3.5
5.2
0.123
84.2
48.8
22.9
2.5
0.005
4.5
10.5
3.7
1.7
0.088
85.5
72.4
23.4
2.4
0.004
4.7
12.1
3.0
5.0
0.113
91.9
68.0
27.3
2.0
0.010
5.5
12.2
3.4
1.2
0.073
91.6
93.4
25.5
1.5
0.009
6.3
11.0
2.8
5.2
0.096
98.1
93.9
29.9
2.8
c
0.012 v c
6.7
v c
14.0
v
4.6
v c
1.1
0.4
1.9
0.3
1.1
0.3
2.1
0.4
1.2
0.4
2.2
0.4
c
36.7
9.2
30.1
7.6
42.2
21.0
31.4
6.3
42.2
10.6
36.8
13.4
c
90.0
90.4
91.0
12.9
6.4
2.0
102.9
96.8
92.9
64.2
113.3
113.7
105.3
168.8
163.5
159.1
139.0
!172.3
!306.1
!371.1
137.6
4.7
4.6
3.4
5.6
4.5
3.3
3.4
2.1
3.6
2.1
2.5
2.4
2.6
4.5
3.5
4.5
4.3
142.8
121.2
103.4
4
81.9
82.2
90.0
21.9
18.0
!1.4
103.3
99.8
87.9
61.7
112.2
114.1
104.3
171.1
167.6
153.8
137.4
!81.1
!251.6
!447.7
135.8
5.5
5.0
3.6
4.9
5.0
1.7
2.9
1.6
2.4
2.9
2.1
3.0
3.1
3.6
2.1
4.6
4.1
144.7
113.0
86.6
5
91.3
91.8
92.2
15.2
6.6
2.1
106.6
98.6
94.4
64.1
116.4
116.6
106.2
168.5
162.2
157.8
137.9
!225.8
!353.4
!413.1
136.9
5.9
5.7
3.2
7.2
5.6
3.8
3.0
1.5
3.3
2.0
2.3
2.3
2.6
3.3
2.3
3.4
4.3
133.1
136.5
125.2
4
81.6
81.7
91.5
26.6
20.8
!0.6
107.6
102.1
90.3
60.5
115.6
116.3
104.4
170.7
166.5
154.0
136.2
!127.0
!265.3
!474.0
134.9
5.2
4.5
3.2
4.9
4.9
2.2
2.8
1.4
2.7
2.8
1.8
2.7
2.9
3.4
2.0
4.8
4.3
161.7
128.3
111.1
4.6
92.0
93.5
94.3
16.8
5.5
1.5
108.9
98.9
95.4
64.2
119.3
119.0
107.9
168.5
159.5
156.4
136.3
!263.2
!359.1
!413.9
135.6
7.5
6.6
3.9
7.6
6.1
4.0
3.4
1.6
1.7
3.1
1.5
1.6
2.6
3.8
2.0
2.0
3.7
149.8
139.4
121.1
3.9
82.2
82.8
91.9
28.3
20.3
!0.8
109.9
102.7
90.4
61.0
119.3
118.4
106.2
168.1
163.4
151.9
134.8
!139.6
!288.8
!472.2
133.7
5.8
5.9
4.0
6.3
6.9
2.6
3.1
2.1
2.8
4.0
2.2
3.1
3.3
3.5
1.9
4.4
4.2
141.7
136.6
139.9
4.6
c
v c
v
v c
c
2.7
v c
touchdown
!30 ms
touchdown
impact
!30 ms
touchdown
impact
!30 ms
touchdown
impact
end midst
touchdown
impact
end midst
!20 ms
touchdown
impact
end midst
!20 ms
touchdown
impact
touchdown
impact
touchdown
impact
touchdown
impact
36.3
3
41.3
2
38.1
2.6
44.4
2.2
40.1
3.6
44.6
1.23
0.57
1.11
0.52
!0.51
!0.19
0.28
0.22
0.26
0.20
0.13
0.15
1.67
0.24
1.58
0.22
!0.52
0.08
0.26
0.08
0.29
0.07
0.13
0.07
1.39
0.64
1.16
0.58
!0.74
!0.21
0.27
0.29
0.29
0.25
0.12
0.21
1.86
0.37
1.64
0.31
!0.84
0.17
0.29
0.11
0.35
0.10
0.12
0.10
1.54
0.85
1.21
0.75
!0.90
!0.32
0.33
0.22
0.38
0.21
0.16
0.19
1.99
0.45
1.67
0.37
!1.04
0.20
v c
v c
v c
c
v
v
v c
v c
v c
v c
v c
c
c
c
c
0.32 v c
0.10 v c
0.38
v
0.09 v c
0.13 v
0.14
c
B. De Wit et al. / Journal of Biomechanics 33 (2000) 269}278
*c
(deg)
*.1
End of midstance (s)
Displ hip (stance) (cm)
Displ hip (#ight) (cm)
Hor displ hip-heel
(cm)
Vert. decel. distance ankle
(cm)
Kleg
KN m~1
Ankle angle
(deg)
4.5 m s~1
274
Table 3
Frontal and sagittal plane kinematics (means and standard deviations of nine persons, 5 trials; v”signi”cant main e!ect of velocity (p(0.05); c”signi”cant main e!ect of condition (p(0.05))
B. De Wit et al. / Journal of Biomechanics 33 (2000) 269}278
275
Table 4
Regression equations between the sole angle at touchdown and the ankle and shank angle at 30 ms before touchdown (a and b) and at touchdown (c
and d)
Equations for both conditions are:
(a) Barefoot:
(b) shod:
(c) Barefoot:
(d) Shod:
sole-angle(0) “34.5!1.0 ankle angle(!0.030 s)#0.6 shank angle(!0.030 s)
r: 0.92
r2: 0.85
p(0.05
sole-angle(0)”39.1!0.9 ankle angle(!0.030 s)#0.5 shank angle(!0.030 s)
r: 0.93
r2: 0.87
p(0.05
sole-angle(0)”!4.8!1.0 ankle angle(0)#1.0 shank angle(0)
r: 0.98
r2: 0.96
p(0.05
sole-angle(0) “7.4!1.0 ankle angle(0) #0.9 shank angle(0)
r: 0.98
r2: 0.97
p(0.05
underneath the heel. This assumption accords with the
“ndings of Hennig et al. (1996), who measured a substantial reduction in heel loading, with a shift towards more
weight bearing in the forefoot, when running with shoes
with harder soles.
Concerning the initial ground contact phase (from
touchdown till the time of the vertical impact peak force)
the vertical deceleration distance of the ankle is signi”cantly reduced in barefoot running (di!erence between
barefoot and shod at 3.5 m s~1: 0.8 cm; at 4.5 m s~1:
1.0 cm; at 5.5 m s~1: 1.0 cm; p(0.05). This can be attributed to the absence of a deformable shoe sole and to
a smaller movement range for plantar #exion through the
#atter foot placement in barefoot running. When observing the frontal plane kinematics, the initial eversion between contact and the occurrence of the impact peak
(*c ) is also signi”cantly smaller for barefoot than for
*.1
shod running. According to the biomechanical model of
Staco! et al. (1988), a larger initial eversion o!ers an
additional deceleration mechanism during initial foot
contact. As a consequence, following the extensive
mechanical analysis of the landing phase in running by
Bobbert and co-workers (1992), the momentum of the
support leg will be less adequately decelerated in barefoot running. In the current study, a compensation is
found in the higher knee #exion velocity during the “rst
30 ms in barefoot running (see Fig. 5), matching with
a strategy to reduce impact loading by reducing the
e!ective mass of the contacting leg (Wright et al., 1998).
However, these authors showed with a forward dynamic
simulation model that this higher knee #exion velocity
occurred immediately after touchdown as a passive result
from running with shoes with harder soles. In the current
study, it should be stressed that in barefoot running this
faster knee #exion originated 0.02 s before touchdown
(see Table 3; Fig. 5), again pointing at an actively induced
adaptation in running style when comparing barefoot to
shod running. The actively induced kinematic adaptations in the current study are also in line with the EMG
data of Komi et al. (1987) which showed a higher preactivation level of the gastrocnemius muscle for the barefoot condition than for the shod condition.
800
Knee Angle Velocity
3.5m.s·1
400
-800
‘,n
gi
4.5 m.s·1
400
I!!
g>
~I
.i!’
~
O
-400
I
~ – ;-=·
g
–c:._~::.-~~-:;l;,:–
-800
400
Time(ms)
Fig. 5. Knee angle velocity curves of a representative subject at the
three running velocities. The barefoot condition is shown as the black
line and the shod condition as the grey line. The “gure shows a time
interval between 20 ms before touchdown and 100 ms after touchdown.
All the above discussed sagittal plane kinematic adaptations to barefoot running were in the same way for
all subjects and for the three running speeds tested in the
current study indicating a pronounced adaptive strategy.
Considering barefoot running as an extreme hard
shoe-condition, results of the current study could be
interpreted in the light of kinematic adaptations to running with shoes with di!ering sti!ness. For instance, one
276
B. De Wit et al. / Journal of Biomechanics 33 (2000) 269}278
experimental study indeed showed a tendency towards
a higher knee #exion velocity immediately following
touchdown when running with harder shoes (Clarke
et al., 1983). However, as indicated in the introduction,
no general picture exists about how subjects modify their
running style as a function of shoe sti!ness. Additionally,
the barefoot condition di!ers also in other mechanical
aspects such as the geometry of the foot-ground interface.
Therefore, the interpretation of the barefoot running
results towards adaptations expected in running with
harder shoes is not straightforward.
Based on our data and on the “ndings of Wright et al.
(1998), the following working hypothesis can be put forward: in response to changing cushioning properties of
the foot-ground interface, the active initiation of the
kinematic adaptations just before foot contact, is followed by a more passive kinetic interaction between the
contacting leg and the ground, during the initial ground
contact phase. This will be the topic of future research.
4.2. General sagittal plane kinematics
During the initial ground contact phase (Figs. 4A and
B) the support leg changes from a more extended (at
touchdown) to a more #exed con”guration in shod running, compared to barefoot running. Attention must be
paid to the instant represented by Fig. 4B: the impact
peak occurs signi”cantly later for shod running (33 versus 11 ms in barefoot running). Anyway, the more #exed
knee position in shod running continues throughout
midstance (see Figs. 4B and C) resulting in a larger
maximal knee #exion in the shod condition (see Table 3).
Towards the end of the stance phase at push-o!, kinematic di!erences between both conditions disappear.
The external forces are non-di!erent during midstance
(absolute F ). This implies the overall sti!ness of the
;!
support leg to be higher during barefoot running. The
ratio of the maximal vertical ground reaction force (F )
;!
to the leg compression (from touchdown till the end of
midstance) can be taken as a measure for overall leg
sti!ness during stance phase in locomotion (Ferris et al.,
1998) and indeed displays signi”cantly higher values for
the barefoot condition (K : see Table 3).
-%’
Ferris and co-workers (1998) showed that runners adjust their leg sti!ness to accommodate for rather large
changes in surface sti!ness. In that way, their subjects
kept their overall vertical sti!ness * i.e. f divided by
;!
vertical displacement of the centre of gravity from touchdown to the end of midstance * constant, resulting in
similar general running mechanics. For instance, ground
contact time and stride frequency remain the same on
di!erent surfaces. A less compliant surface was compensated by a lower leg sti!ness en vice-versa. However, in
the current experiments the foot}ground interface is less
compliant in the barefoot running condition (see the
smaller ankle displacement), but the overall leg sti!ness
during stance phase is higher compared to shod running.
This presumes that there is no equivalent compensation
towards a constant vertical sti!ness, which can be inferred from the signi”cantly smaller foot contact time in
barefoot running (see Table 2). The higher step frequency
relates signi”cantly to the latter (correlation between step
frequency and contact time: bare: r”!0.6, p(0.05;
shod: r”!0.6, p(0.05). The horizontal distance
travelled through the stance phase is smaller in barefoot
running and explains to a large extent the reduction in
step length (correlation between step length and horizontal distance: bare: r”0.7, p(0.05; shod: r”0.67,
p(0.05). On the other hand, the #ight phase remains
una!ected: distance and duration of the airborne phase
do not di!er