Human Movement Science 27 (2008) 636–648

Contents lists available at ScienceDirect

Human Movement Science

journal homepage: www.elsevier.com/locate/humov

A comparison of the mechanical effect of arm swing and

countermovement on the lower extremities in vertical
jumping
Mikiko Hara a,*, Akira Shibayama b, Daisuke Takeshita c, Dean C. Hay d,
Senshi Fukashiro e
a
Department of Sports Sciences, Japan Institute of Sports Sciences, 3-15-1 Nishigaoka, Kita-ku, Tokyo 115-0056, Japan
b
5-28-3 Hongo, Bunkyo-ku, Tokyo, Japan
c
Center for Neurodynamics and Department of Physics and Astronomy, University of Missouri-St. Louis, USA
d
Graduate School of Interdisciplinary Information Studies, The University of Tokyo, Japan
e
Department of Life Sciences, Graduate School of Arts and Sciences, The University of Tokyo, Japan

a r t i c l e i n f o a b s t r a c t

Article history: The purposes of this study were to quantify and compare how arm
Available online 31 July 2008 swing and countermovement affect lower extremity torque and
work during vertical jumping and to gain insight into the mecha-
nisms that enable the arm swing and countermovement to
PsycInfo classification: increase jump height. Five participants maximally performed two
3720 types of vertical squat jumps with (SJA) and without (SJ) an arm
swing and two types of countermovement vertical jumps with
Keywords:
Sports
(CJA) and without (CJ) an arm swing. The participants jumped from
Biomechanical analysis a force platform and all performances were videotaped with a
Jump high-speed video camera (200 Hz). Jump heights, joint torques
Countermovement and work were calculated by combining kinematic and kinetic
Arm swing data. It was found that of the four jumping conditions, the partic-
ipants jumped highest when they used an arm swing with counter-
movement (i.e., CJA). The increase of the countermovement jump
height with an arm swing is the result of the increase of the lower
extremity work. In the hip joint, the increase in torque caused by
the countermovement predominantly occurred at the beginning
of the propulsion phase, while the increase in torque caused by
the arm swing occurred in the rest of the propulsion phase. A

* Corresponding author. Tel.: +81 3 5963 0224; fax: +81 3 5963 0232.
E-mail address: haramikiko@aol.com (M. Hara).

0167-9457/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.humov.2008.04.001
 M. Hara et al. / Human Movement Science 27 (2008) 636–648 637

key finding of our study is that arm swing and countermovement

have independent effects on lower extremity work, and their
effects are additive in CJA to produce greater jump height.
Ó 2008 Elsevier B.V. All rights reserved.

1. Introduction

Voluntary arm movements and countermovement in the lower extremities are strategies
frequently used in jumping to improve performance. The effects of countermovement in jump
performance have been investigated for at least the last three decades (Asmussen & Bonde-
Petersen, 1974; Bobbert & Casius, 2005; Bobbert, Gerritsen, Litjens, & van Soest, 1996; Bosco
& Komi, 1979; Bosco, Viitasalo, Komi, & Ito, 1981; Bosco, Viitasalo, Komi, & Luhtanen, 1982;
Fukashiro & Komi, 1987; Komi, 1984; Sagawa, Kamuro, & Matsumoto, 1989). From the view-
point of the work output during countermovement jump, it has been reported that the greater
jump height in countermovement jump (CJ) compared to squat jump (SJ) was due to a great-
er work output of the hip extensor muscles (Bobbert & Casius, 2005; Bobbert et al., 1996;
Fukashiro & Komi, 1987; Sagawa et al., 1989). For example, Bobbert and Casius (2005) used
a two-dimensional neuromusculoskeletal model to show that the hip extensor muscles could
produce more force and work over the first 30% of their shortening range. This was explained
by the fact that, in this period, the extensor muscles had a higher active state in CJ than
in SJ.
Countermovement is not the only strategy employed to improve jump height performance. Several
researchers (Harman, Rosenstein, Frykman, & Rosenstein, 1990; Payne, Slater, & Telford, 1968) found
that the ground reaction force (FGR) in the latter half of the push-off phase was increased by the use of
the arms. The higher FGR resulted in an increased net ground reaction impulse which was the direct
reason for the augmented jump height.
Luhtanen and Komi (1978) studied the separate segmental movements at maximum intensity to
quantify the contribution of the different body segments including the arms. Ae and Shibukawa
(1980) and Lees and Barton (1996) divided segment momentum into two components: transfer
momentum and relative momentum. In order to express the contribution of arms to vertical
momentum, the relative momentum was used. Although, in these studies, the arm swing effect
was addressed as the relative momentum of the arm to the shoulder, the effect of the arm swing
on lower extremities was not considered. By employing an inverse dynamics method, Feltner, Fras-
chetti, and Crisp (1999) discussed the changes of the resultant torques on countermovement verti-
cal jump with and without an arm swing (CJA, CJ), but they did not calculate the work output of
each joint.
The work–energy theorem shows that work done by muscles is converted to mechanical energy of
a participant. Therefore, in the vertical jump, achieving high jump height is almost equivalent to
achieving a large amount of mechanical energy by achieving large amount of work done by muscles.
From a work–energy theorem perspective, Lees, Vanrenterghem, and Clercq (2004) and Hara, Shi-
bayama, Takeshita, and Fukashiro (2006) examined the role of arm swing during vertical jumping. For
example, Hara et al. examined how arm swing affects the lower extremity work using vertical squat
jumping with and without arm swing (SJA, SJ). They concluded that the increase of jump height with
arm swing resulted mainly from the increase of the lower extremity work (the work by the hip and
ankle), which is considered to have been brought about by the additional load on the lower extremity
due to the arm swing.
Although the influences of both arm swing and countermovement on jump performance have been
examined by many researchers, it is still unknown how the combination of both strategies affect joint
work and jump performance. By using both arm swing and countermovement simultaneously, the ef-
fects could be additive, multiplicative, or perhaps negative. Therefore, the interaction between the two
strategies has yet to be fully understood with respect to joint work.
638 M. Hara et al. / Human Movement Science 27 (2008) 636–648

The purposes of this study were to quantify and compare how arm swing and countermovement
affect lower extremity torque and work during vertical jumping and to gain insight into the mecha-
nisms that enable the arm swing and countermovement to increase jump height.

2. Methods

2.1. Experimental procedure

Five healthy male participants (age: 27.6 ± 3.8 years; height: 172.2 ± 8.9 cm, body mass:
69.9 ± 5.8 kg; mean ± SD) participated in this study. All participants gave their informed consent.
The study was approved by the Ethics Committee of the University of Tokyo.
For the link segment analysis, the anatomical landmarks (the joint centers of the wrist, elbow,
shoulder, hip, knee, ankle, and the metatarsal-phalangeal joint) were marked on the right side of
the body of the participant.
Before the trials, participants performed light aerobic movements to warm up, followed by sev-
eral practice jumps. Once the participants were comfortable with the jumping tasks, they were
asked to perform randomized maximal jumps with 2 min of rest from a force platform (9281B,
Kistler, Switzerland). The four different jumping tasks were: squat jump without arm swing
(SJ), squat jump with an arm swing (SJA), countermovement jump without arm swing (CJ), coun-
termovement jump with an arm swing (CJA) (Fig. 1). Each of the jumping tasks was performed
four times.
A high-speed camera (MEMRECAMc2s, NAC, Japan) was used to record each of the jumps. Partici-
pants were first asked to perform a maximal CJ without any other instructions. The video of that jump
was viewed on a monitor to determine the frame at which the countermovement was at its lowest

Fig. 1. Stick pictures of the jumping motions used in this study. SJ: squat jump with no arm swing, SJA: squat jump with arm
swing, CJ: countermovement jump with no arm swing, CJA: countermovement jump with arm swing. The jump height was
defined as the maximum displacement of the center of mass of the body from the standing position.
 M. Hara et al. / Human Movement Science 27 (2008) 636–648 639

point. The joint centers and segment angles at that point were marked on the monitor for subsequent
calibration of SJ and SJA initial postures. For each of the SJ and SJA trials, each participant was given
verbal instructions to adjust his posture so that it aligned with the frozen video image of the prere-
corded CJ posture before executing the jump. In cases where the alignment was not quickly reached,
the participant was allowed to stand and have a brief rest period to avoid fatigue, after which posture
alignment was retried.
In the arm swing jumps (SJA, CJA), each participant extended his straightened arms backward and
held them in that position prior to the jump. Each participant was instructed to swing his arms at a
self-preferred time once the jumping motion had been initiated. For the no arm swing jumps (SJ,
CJ), participants were instructed to keep their hands on their iliac crests throughout the jumping
movement.

2.2. Measurement

The camera was positioned securely on a tripod 7.2 m from the force plate with its optical axis ori-
ented perpendicular to the sagittal plane of the participants. The camera provided a field of view
(approximately 1.5 m  2.5 m) large enough to record the trials. Before the trials, a calibration board
(1 m  2 m) with 30 reflective markers (diameter: 1 cm) was placed on the force platform and video-
taped. Following that, the kinematics of the right side of the body and the vertical and horizontal com-
ponents of the ground reaction forces (FGR Þ were simultaneously collected at 200 Hz for all trials
(Fig. 2).

Fig. 2. Joint angle definitions and the experimental setting (PC: personal computer, AMP: amplifier, HC: high-speed camera, FP:
force plate).
640 M. Hara et al. / Human Movement Science 27 (2008) 636–648

2.3. Analyses

2.3.1. Center of pressure and displacement of the CM

From the force platform, the center of pressure (COP) and the vertical and horizontal components
of FGR were obtained (Nigg & Herzog, 1995). The displacement of the CM of the body at time t was
calculated as follows:
Z t
1
VCM ðtÞ ¼ ðFGR þ MgÞ dt;
M t0
Z t
DRCM ðtÞ ¼ VCM ðtÞ dt;
t0

where M is the body mass, g is the acceleration of gravity, VCM ðtÞ is the velocity of the CM, and DRCM ðtÞ
is the displacement of the CM, t0 is the time slightly before FGR began to rise or fall. The jump height
was defined as the maximum displacement of the CM from the standing position. The jump height of
the four trials for each jump condition was calculated. For the subsequent analyses of each condition,
we chose the jump that had achieved the greatest height.

2.3.2. Calibration and calculation of the kinematics

First, we digitized the centers of the calibration points and used it together with their known loca-
tions to calculate the direct linear transformation (DLT) parameters for the camera (Abdel-Aziz & Kar-
ara, 1971). Using the DLT method, the two-dimensional locations of the joint centers for the jump
performances were reconstructed. All digitizing and the DLT calculations were performed with the
Frame-DIAS Digitizing and Calibration System (DKH Co., Japan). The obtained time course coordinate
data were smoothed with a low-pass filter (fourth-order zero-lag Butterworth filter, cut off frequency
6 Hz; Winter, 1990). We used a link segment model to calculate the kinematic variables (angle, angu-
lar velocity, and angular acceleration) of each joint (ankle, knee, hip, shoulder, and elbow) from the
beginning of the motion to take-off. Fig. 2 displays the joint angle definitions.

2.3.3. Torque
The equations of translational and rotational motion of each segment were used to calculate joint
reaction forces and torques (inverse dynamics method: Feltner et al., 1999; Winter, 1990). The equa-
tions are as follows:
RJFPROXIMAL ¼ mðaCM  gÞ  FDISTAL ;
RJT PROXIMAL ¼ IG x
_  RJT DISTAL  T F PROXIMAL  T F DISTAL ;

where IG is the moment of inertia created about the segment CM, x _ is the angular acceleration of the
segment, aCM is the acceleration of the segment CM, g is the acceleration of gravity, RJFPROXIMAL is the
proximal resultant joint force exerted on the segment, FDISTAL is the distal force applied to the segment,
RJT PROXIMAL is the proximal resultant joint torque, IG x
_ is the rate of change of the angular momentum
with respect to time about the segment CM (=the resultant torque about the segment CM), RJT DISTAL is
the distal resultant joint torque applied to the segment, T F PROXIMAL is the torque created about the seg-
ment CM by the proximal resultant joint forces applied to the segment, and T F DISTAL is the torque cre-
ated about the segment CM by the distal resultant joint forces applied to the segment. The equations
for the feet, legs, thighs, forearms, and arms were used to calculate the joint reaction force and torque
at the ankle, knee, hip, elbow, and shoulder joints, respectively. The mass of each segment and the mo-
ment of inertia values about the transverse axis of each segment reported by Ae (1996) were used and
personalized for each participant.

2.3.4. Power and work

The power of each joint at time t was given by
Pi ðtÞ ¼ T i ðtÞ  xi ðtÞ;
where Ti is the resultant joint torque and xi is the joint angular velocity. Work done by each joint
throughout the push-off phase of the CM was calculated by integrating power with respect to time:
 M. Hara et al. / Human Movement Science 27 (2008) 636–648 641

Z 0
Wi ¼ Pi ðtÞ dt:
tl

In these equations, Wi is the positive work. The subscript i is the index of each joint. The instant of
take-off was chosen as t = 0. t l is the time when the CM was at its lowest position.
The lower extremity work (W L Þ, the upper extremity work (WU), and the total body work (WT) were
calculated as follows:
WL ¼ WA þ WK þ WH;
WU ¼ WS þ WE;
WT ¼ WL þ WU;
where each subscript is defined as A: ankle joint, K: knee joint, H: hip joint, S: shoulder joint and E:
elbow joint.

2.4. Statistics

A one-way multiple comparison (repeated measures, Bonferroni) was used without using ANOVA
(Bernhardson, 1975; Byron & Hollander, 1977, chap. 10; Glantz, 2005; Hochberg & Tamhane, 1987) to
compare SJ, SJA, CJ, and CJA for the jump height (JH), the displacement of the CM from the standing
position to the lowest position (DepthCM Þ, the displacement of the CM from the standing position to
the take-off position (DispCM Þ, total body work (WT), upper extremity work (WU), the lower extremity
work (WL), and each joint work.
In order to confirm the within-subject reliability, we calculated the coefficients of variance for each
participant’s jump heights. The coefficients of variance in SJ, SJA, CJ, and CJA were 1.87 ± 0.94,
2.45 ± 0.77, 2.28 ± 0.41, and 2.68 ± 0.45%, respectively. Because of the small coefficients of variance,
the point can strongly be made that the within-subject reliability was high in this study.

3. Results

3.1. Comparison of SJ, SJA, CJ, and CJA

Table 1 compares SJ, SJA, CJ, and CJA for the jump height (JH), the displacement of the CM from the
standing position to the lowest position (DepthCM Þ, the displacement of the CM from the standing po-
sition to the take-off position (DispCM Þ, Vcm at takeoff, total body work (WT), upper extremity work
(WU), lower extremity work (WL), and each joint work. Listed values are ensemble averages
(n ¼ 5) ± SD.

Table 1
Comparison between SJ, SJA, CJ, and CJA for kinematic and kinetic values

SJ SJA CJ CJA
JH (101 m) 4.50 ± 0.47 5.31 ± 0.70 4.93 ± 0.40 5.80 ± 0.57
DepthCM (1O1 m) 3.12 ± 0.60 3.02 ± 0.40 3.29 ± 0.34 3.18 ± 0.24
DispCM at takeoff (101 m) 1.45 ± 0.31 1.83 ± 0.27 1.77 ± 0.23 2.11 ± 0.12
Vcm at takeoff (ms1) 2.46 ± 0.24 2.68 ± 0.26 2.62 ± 0.15 2.80 ± 0.17
WE (102 J) – 0.02 ± 0.02 – 0.01 ± 0.01
WS (102 J) – 0.10 ± 0.09 – 0.01 ± 0.12
WH (102 J) 1.90 ± 0.51 2.16 ± 0.43 2.13 ± 0.43 2.31 ± 0.46
WK (102 J) 1.20 ± 0.44 1.07 ± 0.37 1.09 ± 0.72 1.51 ± 0.36
WA (102 J) 1.41 ± 0.23 1.55 ± 0.30 1.38 ± 0.24 1.67 ± 0.38
WU (102 J) – 0.12 ± 0.10 – 0.02 ± 0.12
WL (102 J) 4.52 ± 0.67 4.77 ± 0.59 4.87 ± 0.85 5.48 ± 0.59
WT (102 J) 4.52 ± 0.67 4.89 ± 0.60 4.87 ± 0.85 5.50 ± 0.77

Listed values are ensemble averages ± SD (n = 5).

642 M. Hara et al. / Human Movement Science 27 (2008) 636–648

3.1.1. Body position at the start of push-off

For the displacement of the CM from the standing position to the lowest position (DepthCM Þ, no sig-
nificant difference was found between any of the conditions. Therefore, the body positions at the start
of the push-off phase can be regarded as the same in all jump conditions.

3.1.2. Height and velocity of the CM at take-off

For the displacement of the CM from the standing position to the take-off position (DispCM Þ, the
value of SJ, SJA, CJ, and CJA was, 1.45 ± 0.31, 1.83 ± 0.27, 1.77 ± 0.23, and 2.11 ± 0.12 dm, respectively.
Significant differences were observed in the following pairs of conditions: SJA > SJ, CJ > SJ, CJA > SJ, and
CJA > CJ.
For the velocity of the CM (V CM Þ, the value of SJ, SJA, CJ, and CJA was 2.46 ± 0.24, 2.68 ± 0.26,
2.62 ± 0.15, and 2.80 ± 0.17 m/s, respectively. Significant differences were observed in the following
pairs of conditions: SJA > SJ, CJ > SJ, CJA > SJ, and CJA > CJ.

3.1.3. Jump height

The jump heights of SJ, SJA, CJ, and CJA were 45.0 ± 4.7, 53.1 ± 7.0, 49.3 ± 4.0, and 58.0 ± 5.7 cm,
respectively. Significant differences were observed in the following pairs of conditions: CJ > SJ (9.6%
increase, ðCJSJÞ
SJ
 100Þ, SJA > SJ (18.0% increase, ðSJASJÞ
SJ
 100), CJA > CJ (17.6% increase, ðCJACJÞ
CJ
 100),
CJA > SJA (9.2% increase, ðCJASJAÞ
SJA
 100), CJA > SJ (28.9% increase, ðCJASJÞ
SJ
 100Þ.
These results show that the benefit from the arm swing is similar in SJ and CJ, and that the benefit
from the countermovement is similar in jumps with and without arm swing. The results also show
that the benefit from arm swing (approx. 18%) is almost double the size of the benefit from counter-
movement (approx. 10%), and that these two benefits are additive in CJA ððCJASJÞ SJ
 100 ¼ 28:9%Þ.

3.1.4. Work
Similar properties to jump height were found in total body work (WT): CJ > SJ (7.7% increase,
ðCJSJÞ
SJ
 100), SJA > SJ (8.2% increase, ðSJASJÞ
SJ
 100), CJA > CJ (12.9% increase, ðCJACJÞ
CJ
 100), CJA > SJA
ðCJASJAÞ
(12.5% increase, SJA  100).
For upper extremity work (WU), there was no significant difference between any of the paired jump
conditions. For lower extremity work (WL; Fig. 3), significant differences were observed in the follow-
ing pairs of conditions: CJ > SJ, SJA > SJ, CJA > SJ, CJA > CJ, and CJA > SJA.

3.2. Torque–angle graph

Figs. 4–6 show torque–angle ensemble averages (n = 5) for the hip, knee, and ankle. The horizontal
axes represent the joint angle from the lowest vertical position of the CM to take-off (the push-off
phase). The areas under the curves correspond to the work done by the joints.

Fig. 3. Comparison between SJ, SJA, CJ, and CJA for lower extremities’ work (J). p < .01**, p < .05*.
 M. Hara et al. / Human Movement Science 27 (2008) 636–648 643

Fig. 4. Torque–angle ensemble averages (n = 5) for the hip joint. The positive values indicate hip extension for SJ, SJA, CJ, and
CJA.

Fig. 4 shows that hip joint torque was larger in CJ than in SJ and larger in CJA than in SJA slightly
after the start of the push-off phase. On the other hand, the hip torque in SJA was larger than that in SJ
and the torque in CJA was larger than that in CJ during the middle part of the push-off phase. These
results suggest that the increase in torque caused by countermovement predominantly occurred at
the beginning of the push-off phase, while the increase in torque caused by arm swing occurred later
in the push-off phase. Moreover, arm swing and countermovement increase hip joint work at different
periods during the push-off phase, and their effects are additive, thereby resulting in increased CJA
jump height.
Fig. 5 shows that the knee joint torque in CJ was larger than that in SJ and that the knee joint
torque in CJA was larger than that in SJA at the beginning of the push-off phase. The knee joint
torque in CJ and CJA was less than that in SJ and SJA during the middle part of the push-off phase,
respectively. On the other hand, the torque in SJA was smaller than that in SJ during the last half of
the propulsive phase and the torque in CJA was smaller than in CJ during the first half of the push-
off phase. These results suggest that countermovement had the effect of lowering knee work output
during the first half of the knee extension phase, while arm swing appeared to have a negligible
effect.
Ankle joint torque (Fig. 6) was greater in SJA than in SJ, and greater in CJA than in CJ,
throughout the push-off phase. In comparisons between CJ and SJ, and between CJA and SJA,
a different time course was observed. While the countermovement jump torques (CJ & CJA)
were smaller during the first half of the push-off phase than in the squat jumps (SJ & SJA), they
became larger during the last half of the push-off phase. These results suggest that while coun-
termovement did not influence positive work, arm swing did increase ankle work in the push-
off phase.
644 M. Hara et al. / Human Movement Science 27 (2008) 636–648

Fig. 5. Torque–angle ensemble averages (n = 5) for the knee joint. The positive values indicate knee extension for SJ, SJA, CJ, and
CJA.

4. Discussion

For upper extremity work (WU), there was no significant difference between any of the paired jump
conditions. In a previous study (Lees et al., 2004), WU was considerably larger than in our experiment.
The reason why the upper extremity work was smaller in our study may be the following. In Lees
et al.’s paper, positive work during the push-off phase at the shoulder joint was reported, whereas
the shoulder joint work in our study was defined as the sum of the positive and negative work (net
work). As a result, the calculation range in our study was different.
For lower extremity work (WL; Fig. 3), significant differences were observed in the following pairs
of conditions: CJ > SJ, SJA > SJ, CJA > SJ, CJA > CJ, and CJA > SJA. It was confirmed that the use of arm
swing and/or countermovement increases the lower extremity work output, which is considered to
be the main factor of the increased jump height.
Work is the integral of joint torque with respect to joint angles, and this study also revealed that
countermovement and arm swing produces different torque–angle histories. This means that coun-
termovement and arm swing have different enhancing effects on lower extremity mechanical
output.
Higher joint torques at the same joint angle may essentially be attributed to two possibilities. One
possibility is that arm swing and/or countermovement increased the active state of the muscles. The
other possibility is related to the force–velocity relationship (Hill’s force–velocity relationship: Hill,
1938) and the torque-angular velocity relationship of the lower limb (Anderson, Madigan, & Nuss-
baum, 2007). It has been demonstrated that muscle force decreases as concentric shortening velocity
increases (Gülch, 1994). Furthermore, Anderson et al. reported an inverse relationship between angu-
lar velocity at the ankle, knee, and hip and respective joint moments in concentric movements. In or-
der to determine if the associated angular velocities of each joint changed due to arm swing and/or
countermovement in our study, angular velocity–angle displacement, and Torque–angle displacement
 M. Hara et al. / Human Movement Science 27 (2008) 636–648 645

Fig. 6. Torque–angle ensemble averages (n = 5) for the ankle joint. The positive values indicate plantar flexion for SJ, SJA, CJ, and
CJA.

graphs were plotted (Fig. 7). The joint angle (horizontal axis) was standardized from 0% (the beginning
of the push-off phase) to takeoff.
As mentioned in the results section, the hip joint torque became larger in CJ than in SJ and larger in
CJA than in SJA slightly after the start of the push-off phase. In Fig. 7, the angular velocity at the hip
joint in CJ and CJA was larger than that in SJ and SJA slightly after the start of the push-off phase,
respectively. This implies that the increase of the hip joint torque with countermovement in this per-
iod may have been caused by the higher active state in muscles, not by the force–velocity relationship.
This speculation is consistent with simulation studies of Bobbert et al. (1996; Bobbert & Casius, 2005).
They reported that the hip joint torque increased during the first 30% of its shortening range in CJ due
to a higher active state in CJ than in SJ.
Conversely, with the use of arm swing, the hip joint torque was augmented during the middle and
latter half of the push-off phase in this study. Fig. 7 shows that the hip angular velocity in SJA and CJA
was smaller than that in SJ and CJ during the later stages of the push-off phase, respectively, which
resulted in the increase of the hip joint torque. Some previous jump studies with arm swing examined
the torque and angular velocity time course curve and suggested that this phenomenon was due to the
slower extension of the hip joint caused by the additional load by the arm swing (Feltner et al., 1999;
Hara et al., 2006). Of course, with arm swing, a higher active state is also a possible factor during the
push-off phase, which may have exerted an increase in hip joint torque.
In the ankle joint, the angular velocity was lower in CJ and CJA than that in SJ and SJA through-
out the push-off phase, respectively (Fig. 7). Although countermovement lowers the angular veloc-
ity of the ankle joint, the ankle joint work was not augmented by the countermovement.
Conversely, the arm swing augmented ankle joint torque throughout the push-off phase, which re-
sulted in the increase of ankle joint work. A possible reason why torque increased is that the angu-
lar velocity of the ankle joint was respectively smaller in SJA and CJA than in SJ and CJ throughout
646 M. Hara et al. / Human Movement Science 27 (2008) 636–648

Fig. 7. Angular velocity–angle (left) and torque–angle (right) ensemble averages (n = 5) for the hip, knee, and ankle. The joint
angle (horizontal axis) was standardized from 0% (the begining of the push-off phase) to 100% (takeoff). The positive direction
was set for hip extension, knee extension, and planter flexion. SJ (gray line), SJA (gray line with circles), CJ (black line), and CJA
(black line with triangles).

the push-off phase. Another possibility for the enhanced torque might have been that the plantar
flexors were more activated against a larger downward load with arm swing to maintain the pos-
ture to jump vertically.
In the knee joint, although the knee angular velocity was smaller in CJA compared to CJ, and was
almost the same in SJA and SJ (Fig. 7), torque decreased throughout the push-off phase in arm swing
trials. These results suggest that arm swing makes it difficult for the knee joint to exert large force. In
 M. Hara et al. / Human Movement Science 27 (2008) 636–648 647

addition, we must also take into account the act of the biarticular muscles in the knee joint. The caput
longum m. biceps femoris (which acts as both hip extensor and knee flexor) and caput mediale m. gas-
trocnemius (which acts as both plantar flexor and knee flexor) are considered to exert larger force in
the arm swing condition and may decrease the knee extension torque.
There are some differences in the initial joint angles among each jump condition, but this is con-
sidered to have only a limited effect on the increase in the joint work. The reasons are the following.
First, the difference in hip joint initial angle between SJ and CJ (Fig. 4; top right) is only about 8%,
while the initial torque value in CJ is twice as large as that in SJ. As a result, it is the initial hip torque
value, and not the minor difference in joint angle, which is the major contributor to the increase in hip
work (Table 1). Second, Bobbert et al. (1996), Bobbert and Casius (2005) compared SJ to CJ in simula-
tion studies in which the initial angle and joint range of motion in CJ was the same as that in SJ. They
revealed that the enhancement of joint work in CJ was mainly attributed to the fact that joint mo-
ments were higher during the first part of joint extension.
At the knee joint, knee joint range of motion was greater in CJ, while work was greater in SJ (Table
1, Fig. 5; top right). On the other hand, CJA knee joint range of motion and work was greater than in SJA
(Table 1, Fig. 5; bottom right). In a previous study, the initial knee torque (slightly after the push-off)
was also greater in CJ than in SJ (Bobbert et al., 1996). While differences in initial joint angles may have
contributed to variations in work between jump conditions, these results indicate that is the initial
joint torques that are the key factors when discussing the effects of countermovement.
Regardless of the initial angle, the torque value after push-off was larger in CJ than in SJ, which was
the major contributor to the increase of the hip joint work in CJ. As a result, the conclusion of our study
– the increase in work with countermovement and arm swing was caused by the increase in the joint
torque – remains unchanged.
In this way, arm swing and countermovement seem to have independent effects on lower extrem-
ity work, and their effects are additive in CJA to produce greater jump height.

5. Summary

For the most common jumping motions, humans utilize arm swings and/or countermovements to
maximize height. The purposes of this study were to quantify and compare how arm swing and coun-
termovement affect lower extremity torque and work during vertical jumping and to gain insight into
the mechanisms that enable the arm swing and countermovement to increase jump height.
Jump heights, joint torques, and work were calculated by combining kinematic and kinetic data. It
was found that of the four jumping conditions, the participants jumped highest when they used an
arm swing with countermovement (i.e., CJA).
The increase of the jump height with the countermovement jump and/or an arm swing is the result
of the increase of the lower extremity work. In the hip joint, the increase in torque caused by coun-
termovement predominantly occurred at the beginning of the propulsion phase, which might be aug-
mented by the higher active state with countermovement than without countermovement. On the
other hand, the increase in torque caused by arm swing occurred in the latter half of the propulsion
phase. It was found that the angular velocity of the hip joint was smaller with arm swing than without
arm swing in this period. In the ankle joint, the torque was augmented throughout the push-off phase
only using the arm swing because the hip angular velocity was smaller with arm swing than without
arm swing.
A key finding of our study is that arm swing and countermovement have independent ef-
fects on lower extremity work, and their effects are additive in CJA to produce greater jump
height.

Acknowledgements

We thank Mr. Yasuhiko Takanashi for advice about the statistical analysis, Dr. Akinori Nagano for
suggestions about the data analysis, and Dr. Toshiharu Yokozawa for advice during revision of the
manuscript.
648 M. Hara et al. / Human Movement Science 27 (2008) 636–648

References

Abdel-Aziz, Y. I., & Karara, H. M. (1971). Direct linear transformation from comparator coordinates into object space coordinates
in close-range photogrammetry. In: ASP symposium on close-range photogrammetry (pp. 1–18).
Ae, M. (1996). Inertial parameters of Japanese athletes and children. Japanese Journal of Sports Sciences, 15, 13–15.
Ae, M., & Shibukawa, K. (1980). A biomechanical method for the analysis of the contribution of the body segments-with an
example of vertical jump take-off. Japanese Journal of Physical Education, 25, 233–243.
Anderson, D. E., Madigan, M. L., & Nussbaum, M. A. (2007). Maximum voluntary joint torque as a function of joint angle and
angular velocity: Model development and application to the lower limb. Journal of Biomechanics, 40, 3105–3113.
Asmussen, E., & Bonde-Petersen, F. (1974). Apparent efficiency and storage of elastic energy in human muscles during exercise.
Acta Physiologica Scandinavica, 92, 537–545.
Bernhardson, C. S. (1975). Type I error rates when multiple comparison procedures follow a significant F test of ANOVA.
Biometrics, 31, 229–232.
Bobbert, M. F., & Casius, L. J. (2005). Is the effect of a countermovement on jump height due to active state development?
Medicine and Science in Sports and Exercise, 37, 440–446.
Bobbert, M. F., Gerritsen, K. G. M., Litjens, M. C. A., & van Soest, A. J. (1996). Why is countermovement jump height greater than
squat jump height? Medicine and Science in Sports and Exercise, 28, 1402–1412.
Bosco, C., & Komi, P. V. (1979). Potentiation of the mechanical behavior of the human skeletal muscle through prestretching.
Acta Physiologica Scandinavica, 106, 467–472.
Bosco, C., Viitasalo, J. T., Komi, P. V., & Ito, A. (1981). Prestretch potentiation of human skeletal muscle during ballistic
movement. Acta Physiologica Scandinavica, 111, 35–140.
Bosco, C., Viitasalo, J. T., Komi, P. V., & Luhtanen, P. (1982). Combined effect of elastic energy and myoelectrical potentiation
during stretch-shortening cycle exercises. Acta Physiologica Scandinavica, 114, 557–565.
Byron, B. W., & Hollander, M. (1977). Analysis of k-sample problems. Statistics: A biomechanical introduction. New York: John
Wiley & Sons.
Feltner, M. E., Fraschetti, D. J., & Crisp, R. J. (1999). Upper extremity augmentation of lower extremity kinetics during
countermovement vertical jumps. Journal of Sports Sciences, 17, 449–466.
Fukashiro, S., & Komi, P. V. (1987). Joint moment and mechanical power flow of the lower limb during vertical jump.
International Journal of Sports Medicine, 8(Suppl. 1), 15–21.
Glantz, S. A. (2005). Primer of biostatistics (6th ed.). New York: McGraw-Hill.
Gülch, R. W. (1994). Force–velocity relations in human skeletal muscle. International Journal of Sports Medicine, 15(Suppl. 1),
S2–S10.
Hara, M., Shibayama, A., Takeshita, D., & Fukashiro, S. (2006). The effect of arm swing on lower extremities in vertical jumping.
Journal of Biomechanics, 39, 2503–2511.
Harman, E. A., Rosenstein, M. T., Frykman, P. N., & Rosenstein, R. M. (1990). The effect of arms and countermovement on vertical
jumping. Medicine and Science in Sports and Exercise, 22, 825–833.
Hill, A. V. (1938). The heat of shortening and the dynamic constants of muscle. Proceedings of the Royal Society of London B, 126,
136–195.
Hochberg, Y., & Tamhane, A. C. (1987). Multiple comparison procedures. John Wiley & Sons.
Komi, P. V. (1984). Physiological and biomechanical correlates of muscle function: Effects of muscle structure and stretch-
shortening cycle on force and speed. Medicine and Science in Sports and Exercise, 12, 81–117.
Lees, A., & Barton, G. (1996). The interpretation of relative momentum data to assess the contribution of the free limbs to the
generation of vertical velocity in sports activities. Journal of Sports Sciences, 14, 503–511.
Lees, A., Vanrenterghem, J., & Clercq, D. D. (2004). Understanding how an arm swing enhances performance in the vertical jump.
Journal of Biomechanics, 37, 1929–1940.
Luhtanen, P., & Komi, P. V. (1978). Segmental contribution to force in vertical jump. European Journal of Applied Physiology, 38,
181–188.
Nigg, B. M., & Herzog, W. (1995). Biomechanics of the musculo skeletal system. New York: John Wiley and Sons.
Payne, A. H., Slater, W. J., & Telford, T. (1968). The use of a force platform in the study of athletic activities. A preliminary
investigation. Ergonomics, 11, 123–143.
Sagawa, K., Kamuro, S., & Matsumoto, A. (1989). The effects of countermovements on positive work of three joints related to
lower extremities during vertical jump. Japanese Journal of Sports Sciences, 8, 635–640.
Winter, D. A. (1990). Biomechanics and motor control of human movement (2nd ed.). New York: John Wiley and Sons.