upper -limb kinematics and coordination of short grip … · 51 nourrit et al. , 2003; vereijken et...
TRANSCRIPT
1
UPPER-LIMB KINEMATICS AND COORDINATION OF SHORT GRIP
AND CLASSIC DRIVES IN FIELD HOCKEY
P. BRETIGNY*, L. SEIFERT, D. LEROY, D. CHOLLET
C.E.T.A.P.S. UPRES EA 3832: University of Rouen, Faculty of Sports Sciences,
France
*Authors to whom all correspondence and reprint requests should be addressed
Perrine Brétigny
University of Rouen
Faculty of Sports Sciences
CETAPS
Bld Siegfried
76821 Mont Saint Aignan Cedex
France
e-mail: [email protected]
Tel: (33) 232 10 77 92
Fax: (33) 232 10 77 93
Key words
biomechanics, motor control, field hockey drive, expertise.
Date of resubmission
March 2008
Page 1 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
2
Abstract
The aim of this study was to compare the upper-limb kinematics and coordination of
the short grip and classic drives in field hockey. Ten elite female players participated
in the experiment. The VICONTM system was used to record the displacement of
markers placed on the stick and the players’ joints during five short grip and five
classic drives. Kinematic and coordination parameters were analysed. The ball’s
velocity was recorded by a radar device that also served as the drive target.
Kinematic differences were noted between the two drive conditions, with shorter
duration and smaller overall amplitude in the short grip drive, explained by the
shorter lever arm and the specific context in which it is used. No differences were
noted for upper-limb coordination. In both types of stick holding, an inter-limb
dissociation was noted on the left side, whereas the right inter-limb coordination was
in-phase. Moreover, the time lag increased in the disto-proximal direction,
suggesting wrist uncocking before impact and the initiation of descent motion by the
left shoulder. Medio-lateral analysis confirmed these results: coordination of left-
right limbs converged at the wrist but dissociated with more proximal joints (elbows
and shoulders).
Page 2 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
3
Introduction1
Field hockey games are often characterised by a high density of players in certain 2
parts of the field, which puts the players under intense temporal pressure. Long 3
propulsions usually serve to give velocity to the game and to pass the ball to players 4
in free spaces. The two most common drives used to send the ball long distances are: 5
the classic field hockey drive, with the hands joined together on the top of the stick, 6
and the short grip drive, with the hands also joined but held lower on the stick hose7
(the highest part of the stick that players grip). Experts can perform these two drives, 8
which serve two main purposes: shooting for goal and passing to players several 9
meters away. Although cultural factors seem to influence the choice of one drive 10
rather than the other, the principal factor is a tactical consideration in response to 11
temporal pressure. With a shorter lever arm, the short grip drive is often chosen for 12
rapid actions to counter-balance the opposing defence, while the classic drive is more 13
often used for placed attack schemes. Thus, the mastery of these drives also depends 14
on the player’s field position. Mid and forward positions impose high pressure from 15
opponents, because the density of players is greatest in these parts of the field. 16
Moreover, the passes in this field section are shorter. Players in these positions use 17
the short grip drive more frequently than players in the back field, who benefit from 18
more time to play and thus often choose the classic drive to give more velocity to the 19
ball. The advantage of the classic drive seems to be in power, whereas the interest of 20
the short grip drive lies in its short duration, especially from the beginning of the 21
drive to the impact with the ball.22
There is little information in the scientific literature on these drives and their uses. 23
The few studies on the biomechanics of field hockey have essentially been kinematic 24
analyses of the classic drive (Burgess-Limericket al., 1991; Faque, 1987; Francks et 25
Page 3 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
4
al., 1985; Kusuhara, 1993; Whitaker, 1992). Little is known about limb organisation. 26
Notably, Francks et al. (1985) reported that the drive is performed in a manner 27
similar to that of a golf or baseball swing (i.e. a biphasic movement consisting of an 28
initial backswing, followed by a downswing). Some authors have emphasised a third 29
phase just after ball impact, the follow-through or finish, even though the stick is no 30
longer in contact with the ball (Faque, 1987; Kusuhara, 1993; Whitaker, 1992). 31
These authors assume that this last part of the drive has an impact on ball direction 32
and precision, and plays a role in reducing injuries. The results concerning phase 33
duration and variability in the stick’s movement through space have been conflicting. 34
Burgess-Limerick et al. (1991) and Francks et al. (1985) observed that experts show 35
great spatial and temporal variability in their attempts to produce a consistent and 36
accurate downswing. Conversely, Kusuhara (1993) showed greater spatial and 37
temporal drive reproducibility for intermediate and trained players than for 38
beginners. But, first, the methods used were different and the cycle was not clearly 39
defined by key points. Second, several types of drive are used to hit the ball. These 40
conflicting results thus suggest the complexity of the field hockey drive and the need 41
for not only a clarification of its kinematic features, but also a coordination analysis, 42
by a comparison of the two types of drive. 43
In fact, coordination analysis would provide further insight into the study of these 44
drives, because it is an explanatory feature of complex skills and expertise analysis. 45
Haken et al. (1985) pointed out the interest of inter-limb coordination analysis and 46
modelled coordination dynamics to better understand motor control when the control 47
parameter was changed. The dynamic approach was extended from bimanual 48
coordination to various complex skills in order to analyse the motor control of 49
subjects as they used a ski simulator (Delignières et al., 1999; Nourrit et al., 2000; 50
Page 4 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
5
Nourrit et al., 2003; Vereijken et al., 1992) or performed swings under the parallel 51
bars in gymnastics (Delignières et al., 1998) and serves in volleyball (Temprado et 52
al., 1997). In these activities, coordination differences were noted between novices 53
and experts, with a release of degrees of freedom for experts. For the volleyball 54
serve, experts used anti-phase shoulder-wrist coupling, suggesting limb dissociation 55
(Temprado et al., 1997). In gymnastics, experts developed partial decoupling in order 56
to overcome spontaneous tendencies highly constrained by the intrinsic dynamics of 57
the system (Delignières et al., 1998). The coordination of the upper-body limbs in 58
field hockey has been described as a succession of segmental accelerations, 59
proceeding from the most proximal to the most distal and serving to maximize the 60
stick speed at ball impact (Faque, 1987), but no published study has analysed the 61
upper inter-limb coordination during drive performance. 62
Therefore, the aim of this study was to compare the upper-limb kinematics and 63
coordination of the two expert field hockey drives (classic and short grip).64
We hypothesised that the two drives would show kinematic differences but similar 65
upper-limb coordination. More specifically, we expected that 1) the short grip drive 66
would require less time to perform than the classic drive, thus being more effective in 67
the context of temporal urgency, and 2) the expert players would adopt upper-limb 68
dissociation in both drives to maximize the final velocity given to the ball at impact.69
70
Methods 71
Subjects72
Ten female field hockey players participated in the study (age: 23.9±4.8 years, 73
height: 168.2±6.3 cm, and mass: 60.1±7.2 kg). All played at the elite national level 74
and either had been or still were on the national team. All had been training for 75
Page 5 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
6
12.3±5 years. The players provided informed written consent to participate in the 76
study, which was approved by the University ethics committee.77
78
Protocol79
A radar speed detection device (Speed ChekTM, Tribar Industries Inc., Ontario, 80
Canada) was used to obtain ball velocity after impact.81
The VICONTM optoelectronic system (Oxford Metrics, Oxford, UK) measured the 82
kinematic parameters during the drives, with five cameras operating at 50 frames per 83
second placed around the player (Atha, 1984). According to Shannon’s theorem 84
(movement frequency must be equal to or greater than twice the signal frequency) 85
(Shannon & Weaver, 1963), 50 Hz was sufficient to analyse the field hockey drive, 86
which is a 1-Hz frequency movement. The calibrated volume resolution was 2 m (Z 87
or vertical axis) / 2 m (X or medio lateral axis) / 3 m (Y or anterio-posterior axis), in 88
order to record all phases of the drive. The experimental space was statically and 89
dynamically calibrated and passive reflective markers were fixed with double-face 90
adhesive tape on the following anatomical sites: the left and right acromio-clavicular 91
joints (Lsho and Rsho), the epicondyles of each arm (Lelb and Relb), and the left and 92
right apophysis styloïd radius (Lwri and Rwri). With the VICON system, at a 93
sampling frequency of 50 Hz, the maximal error was 1° for the two large angle 94
cameras and 0.5° for the three other cameras. Errors due to skin movements were 95
reduced by placing the markers where the skin is thin and therefore primarily moves 96
with the underlying bone structure (Dujardin et al., 1997). Two other markers were 97
placed on the stick, one at the head (croB), and one 45 cm away from the first 98
(croH). The hockey players were dressed in shorts and bras, wore their own hockey 99
shoes, and used their own sticks, in order not to disturb their drives. However, all the 100
Page 6 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
7
sticks were similar in height, weight, and composition to ensure equal shooting 101
conditions. Each subject performed ten drives while standing with her left foot next 102
to the ball, which had been placed on a mark on the floor. Five drives were classic 103
(hands joined at the top of the stick) and five were short grip (hands joined lower on 104
the stick hose). Practice trials were allowed to ensure that the players were 105
comfortable with the experimental procedure. Each subject was asked to hit the ball 106
as if she were passing the ball to a player situated about 20 m away (distance 107
representative of field hockey match passes). They were asked to aim at the radar 108
device, which was on the floor, just behind the nylon indoor protection net (1.2 m 109
high and placed 1.6 m away from the player to prevent accidental damage to 110
equipment), and served as the target. Trials in which the ball landed 0.05 m or more 111
from this target were not analysed because these were considered not to respect the 112
precision conditions. In this case, subjects performed additional trials to record ten 113
accurate drives. Each drive was also assessed in terms of the ball velocity measured 114
by the radar. Synthetic carpet was placed on the ground to reproduce match 115
conditions.116
The drive was divided into the following three phases:117
1. Backswing, corresponding to the stick motion away from the ground. The 118
players were allowed to begin their swing wherever they wanted as long as 119
the stick motion was away from the ground. 120
2. Downswing, corresponding to stick motion in the opposite direction, from the 121
highest point to impact with the ball.122
3. Follow-throw or finish, from impact to the end of the drive, which 123
corresponded to the highest point reached by the stick at the end of the 124
movement just before the player became relaxed.125
Page 7 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
8
For the two drives, the left hand is placed higher than the right one on the stick hose126
(the highest part of the stick that players grip). The left arm then guides the stick 127
motion (lead arm), whereas the right one serves as a support to the movement (trail 128
arm).129
Insert figure 1130
131
The kinematic analyses were based on the right wrist displacements in the frontal 132
plane during the entire stroke, and the ball velocity. Analyses in the three planes 133
were performed before the experiment and the greatest variations in amplitude were 134
observed in the vertical plane. The choice was thus made to concentrate on this 135
macroscopic variable of the movement, which best typifies the coordination, in 136
accordance with the principles of dynamic approaches (Kelso, 1984 for bimanual 137
coordination). The absolute duration of each drive and each phase was then 138
established, using the right wrist height peaks as the reference to delimit the phase of 139
the movement (Gatt et al., 1998). In fact, at 50 Hz, the data losses for the two 140
markers placed on the stick were too numerous to permit the use of the stick as the 141
reference for determining the beginning and end of each drive phase. To compare the 142
drives, the relative duration of each phase was then calculated, in percentage of 143
cycle.144
Upper-limb coordination was analysed by punctual (measures made at discrete 145
instants) estimations of the relative phase (RP) calculated at ball impact, according to 146
Kusuhara (1993) who showed that this point was a determinant factor of expertise. 147
These punctual estimations of RP were used to analyse the proximo-distal and 148
medio-lateral couplings between the different joints: 1) proximo-distal coupling: 149
RP=(tdistal/tproximal)*360, where tdistal and tproximal are the time for the distal and the 150
Page 8 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
9
proximal joints to reach their minimal height, and 2) medio-lateral coupling: 151
RP=(tright/tleft)*360, where tright and tleft are the time for the right and the left joints to 152
reach their minimal height (Hamill et al., 2000). The data are circular so a 360°angle 153
corresponds to a 0°angle. For all values superior to 360, the formula 360-value was 154
used in order to have all angles between 0° to 360°.155
Concerning the right proximo-distal analysis (trail arm), five levels of coupling were 156
examined using the following markers: PD1 for croB and croH coupling, PD2 for 157
Rwri and croH coupling, PD3 for Relb and Rwri coupling, PD4 for Rsho and Relb 158
coupling, and PD5 for Rsho and Rwri coupling. For the left proximo-distal analysis 159
(lead arm), these five levels of coupling were also examined using the left markers.160
For medio-lateral coordination, three levels of coupling were compared: L1 for the 161
Rwri and Lwri relationship, L2 for the Relb and Lelb relationship, and L3 for the 162
Rsho and Lsho relationship.163
According to Dietrich and Warren (1995), the in-phase mode of coordination 164
corresponds to 0±30° values of RP and the anti-phase mode corresponds to 180±30° 165
values of RP. For field hockey analysis, values between 30° and 330° correspond to 166
an out-of-phase mode, whereas 0±30° values correspond to an in-phase mode.167
168
Statistical analysis169
For kinematic parameters, normal distribution (Ryan Joiner test) and homogeneity of 170
variance (Bartlett test) allowed parametric statistics. 171
A one-way ANOVA [condition (2 levels: short grip drive, classic drive)] was 172
performed on the values of ball velocity after impact, the absolute total duration, the 173
relative duration of each phase (backswing, downswing, and finish) and the right 174
Page 9 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
10
wrist/subject heights ratio at each key point (beginning of the drive, highest point, 175
impact, and end of the drive). 176
Mean and SD were calculated for each variables. RMS (root Mean Square) was 177
established for the displacement measures, in order to estimate the magnitude of the 178
varying quantity, using the formula: rms=√((x1²+x2²+...+xn²)/N); with x1, x2...xn =the 179
n values of the variable studied, and N = the number of values.180
Linear data processing was performed with Minitab 14.10 (Minitab Inc., 2004). 181
182
For upper-limb coordination parameters, Watson Williams F-tests [proximo-distal 183
(five levels: croB/croH, croH/wrist, wrist/elbow, elbow/shoulder, wrist/shoulder) and 184
medio-lateral (three levels: wrists, elbows, shoulders)] were used to compare RP, in 185
order to determine:186
• The differences between the short grip drive and the classic drive for each 187
level of coupling,188
• The differences between the proximal and distal joint coordination, and 189
between the medial and lateral joint coordination, for each condition.190
Circular data processing (Baschelet, 1981) was performed with Oriana 2.0 (W. 191
Kovach Computing Services, 1994-2003). 192
For all tests, the level of significance was set at 0.05.193
194
Results195
First, kinematic parameters will be presented, followed by coordination estimates.196
The way the stick was held did not change the ball velocity after impact (71.98±1.33 197
km.h-1 for the short grip drive and 71.09±1.75 km.h-1 for the classic drive) (F1,88=0.2, 198
p=0.7). 199
Page 10 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
11
As well, the short grip drive took significantly less time to perform than the classic 200
drive (0.96±0.03s vs. 1.04±0.02s) (F1,82=7.2, p<0.05).201
Differences between short grip and classic drives downswing relative duration was 202
not significant. As a result, the backswing of the short grip drive had a shorter 203
relative duration than with the classic drive (F1,82=0.8, p<0.05) , conversely the finish 204
of the short grip had a longer relative duration (F1,82=1.8, p<0.05). 205
206
Insert Table 1207
208
Wrist/subject height values were significantly lower with the short grip drive at each 209
key point: for the beginning of the drive (short grip: mean of 0.46±0.09cm, 210
RMS=0.47, vs. classic drive: mean of 0.48±0.09cm, RMS=0.49; F1,82=1.4, p<0.05); 211
the highest point (0.6±0.06cm, RMS=0.6 vs. 0.63±0.07cm, RMS=0.63; F1,82=4, 212
p<0.05); the impact (0.34±0.02cm, RMS=0.34 vs. 0.38±0.02cm, RMS=0.38; 213
F1,82=49, p<0.05); and the end of the drive (0.58±0.08cm, RMS=0.6 vs.214
0.63±0.08cm, RMS=0.6; F1,82=5.5, p<0.05).215
216
Few differences were observed concerning the proximo-distal and medio-lateral 217
coordination parameters.218
For the five levels of coupling, the right (trail) proximo-distal coordination was in-219
phase (RP values between 0° and 30° or between 330° and 360°). Significant 220
differences were observed only between the stick and the body parts (F4,39=9.4 and 221
F4,39=7.6, respectively, for the short grip drive and the classic drive, p<0.05). There 222
were no significant difference between the two conditions (F1,12=0.383; F1,12=1.144; 223
Page 11 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
12
F1,18=0.22; F1,18=0.594; and F1,18=0.587, respectively, for PD1, PD2, PD3, PD4, and 224
PD5; all p>0.05) (see Table 2 and Fig. 1).225
226
Insert Table 2 and Figure 2227
Concerning left (lead) proximo-distal coordination, for the two conditions, 228
significant differences were noted between the stick and joints, and between the 229
different joints (F4,39=32.18 and F4,39=16.5, respectively, for the short grip drive and 230
the classic drive; all p<0.05) (see Table 2 and Fig. 2). Here the coordination was out-231
of-phase (RP values between 30° and 330°).232
There were no significant difference between the two conditions (F1,12=0.383; 233
F1,12=1.202; F1,18=0.262; F1,18=0.006; and F1,18=0.133, respectively, for PD1, PD2, 234
PD3, PD4, and PD5; all p>0.05)235
236
Insert Figure 3237
238
Finally, no significant difference in the medio-lateral organisation was noted between 239
the two conditions. Therefore, Figure 4 illustrates the couplings of the right and left 240
joints for the mean of the two conditions. The coordination between the left and right 241
wrists was in-phase whereas the coordination between the left and right elbows, and 242
between the right and left shoulders, was out-of-phase (see Table 3). This time lag in 243
the coupling was greater at the proximal levels (F2,27=62.9 and F2,27=32.5, 244
respectively, for the short grip drive and the classic drive; all p<0.05). 245
246
Insert Table 3 and Figure 4 247
248
Page 12 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
13
Discussion249
Results showed no influence of stick holding on the ball velocity. Nethertheless, with 250
a shorter duration and a smaller overall amplitude, the short grip drive should have 251
presented a smaller ball velocity at release. But in the present experiment, in order to 252
prevent accidental damage to equipment, the subjects were asked to hit the ball as if 253
they were passing it to a player situated about 20 m away. Without this instruction, 254
players may have hit the ball stronger with the classic drive. For passing the ball, the 255
short grip drive seems to have produced a similar final ball velocity, with a smaller 256
absolute total duration, which offers strategic advantages.257
The shorter total time to perform the short grip drive (0.99s vs. 1.04s for classic 258
drive) can be explained by the position of the hands on the stick. Kusuhara (1993), in 259
his study comparing the classic drive in experts, intermediates and novices, reported 260
a longer total time than that of this experiment (1.12s vs. 1.04s). But these results 261
should be interpreted with caution because time decomposition is difficult, especially 262
for the beginning and end of the drive. 263
Concerning the relative temporal parameters, the general organisation of the stroke 264
was similar to that described by Kusuhara (1993), with the shortest phase being the 265
downswing, followed by the finish and then the backswing, and this for the three 266
levels of practice: beginner, intermediate and trained. This suggests a higher velocity 267
and / or shorter displacement during the downswing. 268
The comparison of the short grip and classic drives showed significant differences 269
only for the relative durations of the backswing and finish. These results agree with 270
the literature regarding the influences of uncertainty and material constraints on a 271
field hockey drive or a golf swing. Burgess-Limerick et al. (1991) and Francks et al. 272
(1985) reported that experts maintained the temporal consistency of the downswing 273
Page 13 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
14
until impact with the ball. In golf, Nagao and Sawada (1973) and, more recently, 274
Egret et al. (2003) reported similar results, with temporal consistency in the 275
downswing but spatial variability in the backswing, in relation with the use of 276
different clubs. Delay et al. (1997) also showed that for long target distances, players 277
increase the downswing amplitude (spatial parameter) but maintain the same 278
movement time in order to increase club velocity at ball contact. In this study, the 279
subjects were constrained by a technical parameter: holding the stick. This had an 280
influence on the kinematic parameters, but, as in the studies presented above, 281
temporal consistency of the downswing was preserved. The backswing phase was 282
shorter in the short grip drive than in the classic drive, and, as a consequence, the 283
short grip finish phase was longer. These results can be explained by the different 284
ways the two types of drives are used. The short grip drive is often used for rapid 285
actions in a context of temporal pressure, i.e. the offense / defence challenge. In 286
response to this pressure, players reduce their preparation time, which is represented 287
by the backswing, in order to hit the ball as quickly as possible and thus reduce 288
telegraphing their intentions to the opposing players. The time saved by reducing the 289
relative duration of the backswing is then used for the finish to adjust ball direction 290
and accuracy.291
To complete the kinematic analysis, a spatial parameter, the right wrist/subject height 292
ratio was used. The results showed that values were significantly lower with the short 293
grip drive than with the classic drive. The hands were placed lower on the stick hose 294
in the short grip drive, so they were closer to the ground, giving a smaller overall 295
amplitude. 296
Page 14 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
15
In summary, the kinematic parameters revealed differences in the short grip and 297
classic drives that justify their use in different contexts in relation with tactical 298
intentions. 299
In order to better understand the complexity of the movement, a coordination 300
analysis was performed on the upper-body limbs: no significant difference was noted 301
in the proximo-distal and medio-lateral organisation of the upper limbs between the 302
two conditions of drive. 303
In many throwing sports, final velocity is maximised by an inter-limb dissociation. 304
Fradet et al. (2004) reported several studies in which the fastest throwing action is 305
accomplished with a proximal-to-distal progression. This proximo-distal sequence 306
can be illustrated by the sudy of linear or angular velocities with maxima of the 307
proximal segments appearing before those of distal segments. This movement 308
organisation is considered to be the most efficient for mechanical and muscular 309
reasons. Gonzalez and Dietrich (2003) introduced a biomechanical model for 310
analysing javelin throw movement to determine how the maximal velocity at release 311
time is reached. The linear joint velocity study showed a successive rise in the speed 312
peaks, from the hip to the javelin, passing by the shoulder, the elbow and the wrist of 313
the throwing arm. Mero et al. (1994) found similar results that they attributed to a 314
transfer of kinetic energy, allowing progressive and successive summation of one 315
body limb to another. This concept of energy flow was also advanced for the 316
handball overarm throw (Jöris et al., 1985). The consecutive actions of the body 317
segments, from the proximal to the distal segments, were demonstrated to be 318
important for the optimal flow of energy to the ball during the last phase of the 319
throw. In our study, vertical joint displacements were used to obtain RP estimation at 320
ball impact. The results showed that, for the right side, joint displacements were in-321
Page 15 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
16
phase, meaning that the minimal heights of the joints were reached simultaneously 322
(RP=0° or 360° ±30°). The proximo-distal differences were significant only between 323
the stick and body parts, which can be explained by stick rotations in the hands. 324
For the left side, the same stick/body differences were observed, but the inter-limb 325
coordination was out-of-phase (i.e. the minimal heights of the joints were reached at 326
different moments in the movement, RP values between 30° and 330°). This 327
indicated a time lag between the left shoulder and elbow, shoulder and wrist, and 328
between elbow and wrist. This time lag increased significantly in the disto-proximal329
direction, which means that the stick descent motion until ball contact was initiated 330
by the left shoulder (lead arm), right arm (trail one) serving to support the movement. 331
This inter-limb dissociation agrees with the findings of Temprado et al. (1997), who 332
studied the volleyball serve by comparing the upper inter-limb coordination of 333
novices and experts. The novices used their arms as a single rigid limb, whereas the 334
experts displayed spatial and temporal inter-limb dissociation. This was revealed by 335
in-phase couplings for the novices and an anti-phase coordination (RP=180°) 336
between the shoulder and wrist for the experts. This shoulder-wrist relationship 337
implies wrist uncocking (rapid movement from full wrist extension to flexion) in 338
order to ensure a mechanically effective movement. In our study, the relationship 339
between the shoulder and wrist was not in anti-phase. But, statistical analysis showed 340
a significant difference between the coupling of these two joints and the other pairs, 341
indicating a longer time lag. The left shoulder initiated the movement of the stick and 342
arrived earlier at ball contact. Thus, wrist uncocking can also be assumed for the 343
field hockey drive, which would agree with Whitaker’s description (1992). He 344
observed wrist uncocking just before ball impact, which served to accelerate the 345
movement. Similar results were advanced by Faque (1987), who reported a time lag 346
Page 16 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
17
effect for the arms. This was shown by successive accelerations to give a greater 347
final velocity to the ball. In golf, Egret et al. (2004) showed the same phenomenon, 348
indicating that the swing should start with movements of the more proximal 349
segments and progress with faster movements of the more distal segments to 350
optimise final speed. For the field hockey drive, studies on left wrist velocity and 351
acceleration are needed to confirm wrist uncocking just before the impact with the 352
ball.353
The results for medio-lateral coordination were similar to those for proximo-distal354
coordination. Indeed, whereas the left and right wrists were in-phase, the shoulders 355
and elbows were out-of-phase. Significant differences were noted between the levels 356
of coupling, revealing a greater time lag between the two shoulders than between the 357
two elbows. Given the focal boundary condition of the hands on 358
the stick, coordination of the left-right segment converged at the wrist but dissociated 359
with more proximal joints (elbows and shoulders).360
361
In summary, the short grip and classic drives in field hockey differ kinematically. 362
With a shorter lever arm, the short grip drive requires less time and has a narrower 363
range of movement than the classic drive. These characteristics explain why field 364
hockey players more often use this drive for rapid actions, in the context of intense 365
temporal pressure. One consequence of its use in this type of context is that less time 366
is given to the backswing than is given in the classic drive. Despite these kinematic 367
differences, the upper inter-limb coordination was similar in the two drive 368
conditions. A proximo-distal dissociation was noted on the lead side, with a 369
relatively long time lag between the shoulder and wrist, suggesting medio-lateral 370
dissociation. This left wrist delay suggests wrist uncocking before impact with the 371
Page 17 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
18
ball but further research is needed to confirm this. To confirm these results, it would 372
be interesting to perform the same experiment on a field hockey pitch. The protection 373
net would no longer be necessary as risks of damage to equipment would be 374
eliminated. This would make the task more realistic but would also require portable 375
equipment for the motion analysis.376
4318 words.377
Page 18 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
19
References
Atha, J. (1984). Current techniques for mesuring motion. Applied Ergonomics, 15,
245-257.
Baschelet, E. (1981). Circular statistics in biology. San Diego, CA: Academic Press.
Burgess-Limerick, R., Abernethy, B., & Neal, R.J. (1991). Experience and
backswing movement time variability. A short note concerning a serenditipous
observation. Human Movement Science, 10, 621-627.
Delay, D., Nougier, V., Orliaguet, J.P., & Coello, Y. (1997). Movement control in
golf putting. Human Movement Science, 16, 597-619.
Delignières, D., Nourrit, D., Deschamps, T., Lauriot, B., & Caillou, N. (1999).
Effects of practice and task constraints on stiffness and friction functions in
biological movements. Human Movement Science, 18, 769-793.
Delignières, D., Nourrit, D., Sioud, R., Leroyer, P., Zattara, M., & Micaleff, J.P.
(1998). Preferred coordination modes in the first steps of the learning of a complex
gymnastics skill. Human Mouvement Science, 17, 221-241.
Dietrich, F.J., & Warren, W.H. (1995). Why change gaits? Dynamics of the walk-run
transition. Journal of Experimental Psychology: Human Perception and
Performance, 21, 183-202.
Dujardin, F.H., Ertaud, J.Y., Aucouturier, T., Nguen, J., & Thomine, J.M. (1997).
Smoothing technique using Fourier transforms applies to stereometric data obtained
from optoelectronic recordings of human gait. Human Movement Science, 529, 1-8.
Egret, C., Dujardin, F., Weber, J., & Chollet D. (2004). 3D kinematic analysis of the
golf swings of expert and experienced golfers. Journal of Human Movement Studies,
47,193-204.
Page 19 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
20
Egret, C.I., Vincent, O., Weber, J., Dujardin, F.H., & Chollet, D. (2003). Analysis of
3D kinematics concerning three different clubs in golf swing. International Journal
of Sports Medicine, 24, 465-469.
Faque, V. (1987). Approche biomécanique du shoot. Monographie BE2, INSEP,
France.
Fradet, L., Botcazou, M., Durocher, C., Cretual, A., Multon, F., Prioux, J., &
Delamarche, P. (2004). Do handball throws always exhibit a proximal-to-distal
segmental sequence? Journal of Sports Sciences, 22, 439-447.
Francks Ian, M., Weicker, D., & Robertson, D.G.E. (1985). The kinematics,
movement phasing and timing of a skilled action in response to varying conditions of
uncertainty. Human Movement Science, 4, 91-105.
Gatt, C.J., Pavol, M.J., Parker, M.D., & Grabiner, M.D. (1998). Three dimensional
knee joint kinematics during a golf swing. The American Journal of Sports Medicine,
26, 285-294.
Gonzalez, D., & Dietrich, G. (2003). Three-dimensional kinematics analysis of
javelin throw: from real situation to biomechanical model. Sciences & Sport, 18,
216-218.
Hamill, J., Haddad, J.M., & Mc Dermott, W.J. (2000). Issues in quantifying
variability from a dynamic system perspective. Journal of Applied Biomechanics, 16,
407-418.
Haken, H., Kelso, J.A., & Bunz, H. (1985). A theorical model of phase transitions in
human hand movements. Biological Cybernetics, 51, 347-356.
Page 20 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
21
Jöris, H.J.J., Edwards van Muyen, A.J., Van Ingen Schenau, G.J., & Kemper, H.C.G.
(1985). Force, velocity and energy flow during the overarm throw in female handball
players. Journal of Biomechanics, 18, 409-414.
Kelso, J.A.S. (1984). Phase transitions and critical behavior in human bimanual
coordination. American Journal of Physiology, Regulatory, Integrative and
Comparative Physiology, 15, R1000-R1004.
Kusuhara, K. (1993). Kinematic analysis of field hockey stroke. Proceeding of
International Society of Biomechanics XIVth Congress, 734-735.
Mero, A., Komi, P.V., Korjus, T., Navarro, E., & Gregor, R.J. (1994). Body segment
contribution to javelin throwing during final thrust phases. Journal of Applied
Biomechanics, 10, 166-177.
Nagao, N., & Sawada, Y. (1973). A kinematic analysis in golf swing concerning
driver shot and No. 9 iron shot. Journal of Sports Medicine, 13, 4-16.
Nourrit, D., Delignières, D., Caillou, N., Deschamps, T., & Lauriot, B. (2003). On
discontinuities in motor learning: a longitudinal study of complex skill acquisition on
a ski-simulator. Journal of Motor Behavior, 35, 151-170.
Nourrit, D., Deschamps, T., Lauriot, B., Caillou, N., & Delignières, D. (2000). The
effects of required amplitude and practice on frequency stability and efficiency in a
cyclical task. Journal of Sports Science, 18, 201-212.
Shannon, C.E., & Weaver, W. (1963). The mathematical theory of communication.
University of Illinois Press, Champaign, IL, USA.
Temprado, J., Della Grasta, M., Farrell, M., & Laurent, M. (1997). A novice-expert
comparison of (intra-limb) coordination subserving the volleyball serve. Human
Movement Science, 16, 653-676.
Page 21 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
22
Vereijken, B., Van Emmerick, R.E.A., Whiting, H.T.A., & Newell, K.M. (1992).
Freezing degrees of freedom in skill acquisition. Journal of Motor Behavior, 24, 133-
142.
Whitaker, D. (1992). The hockey workshop, a complete game guide. The Crowood
Press. Marlborough (England).
Page 22 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
23
Tables legends
Table 1. Means and SD of the relative durations of the three phases with the short grip and
classic drives.
Table 2. Means and SD of the right and left proximo-distal RP for the two conditions of drive
(short grip and classic)
Table 3. Means and SD of the medio-lateral RP for the two conditions of drive (short grip and
classic)
Page 23 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
24
Table 1
Phase Relative duration
(in percentage of cycle) of
short grip drive
Relative duration
(in percentage of cycle) of
classic drive
Backswing
Downswing
Finish
36.1±9.4
28.2±6.2
35.6±7.8
37.9±8.7 a
28.5±6
33.6±6.5 a
a: significant difference between drive types (p<0.05).
Table 2
Condition Right PD
(proximo-distal)
RP (°) Left PD
(proximo-distal)
RP (°)
Short grip drive PD1 (croB/croH) 8.4±7.9 PD1 (croB/croH) 8.4±7.9
PD2 (Rwri/croH) 8.1±7.3 PD2 (Lwri/croH) 18.3±8.6
PD3 (Relb/Rwri) 351.9±7.4 a d PD3 (Lelb/Lwri) 33.3±14.1 a d
PD4 (Rsho/Relb) 344.0±16.6 c d PD4 (Lsho/Lelb) 41.8±11.8 c d
PD5 (Rsho/Rwri) 336.0±17.6 b c d PD5 (Lsho/Lwri) 79.0±18.1 a b c d
Classic drive PD1 (croB/croH) 5.3±9.4 PD1 (croB/croH) 5.3±9.4
PD2 (Rwri/croH) 12.3±6.0 PD2 (Lwri/croH) 32.7±31.8
PD3 (Relb/Rwri) 352.3±5.4 a d PD3 (Lelb/Lwri) 36.3±10.2 d
PD4 (Rsho/Relb) 349.8±15.2 c d PD4 (Lsho/Lelb) 42.2±14.2 d
PD5 (Rsho/Rwri) 342.1±16.6 c d PD5 (Lsho/Lwri) 82.5±22.2 a b c d
a: significant difference with previous level of PD, b: significant difference with PD3, c:
significant difference with PD2, d: significant difference with PD1
Page 24 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
25
Table 3
Condition L (medio-lateral) RP (°)
Short grip drive L1 (Rwri/Lwri) 8.8±6.0
L2 (Relb/Lelb) 52.6±15.6 a
L3 (Rsho/Lsho) 122.5±32.1 a b
Classic drive L1 (Rwri/Lwri) 18.0±23.2
L2 (Relb/Lelb) 64.1±28.2 a
L3 (Rsho/Lsho) 133.0±35.2 a b
a: significant difference with previous level of L, b: significant difference with L1
Page 25 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
26
Figures legends
Figure 1. Schematic description of field hockey drive movement.
Figure 2. Vertical displacement of the stick and right upper limb during the field hockey drive,
showing that the inter-limb coordination is in-phase.
Figure 3. Vertical displacement of the stick and the left upper limb during the field hockey drive.
The time lags between the minimal heights of the joints indicate that the coordination mode is
out-of-phase.
Figure 4. Vertical displacement of the right and left wrists (a), elbows (b), and shoulders (c)
during the field hockey drive.
Page 26 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
27
Figure 1.
DOWNSWING
Highest point Impact
End of the drive
Beginning of the drive
BACKSWING FINISH
Page 27 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
28
0
20
40
60
80
100
120
140
160
10 20 30 40 50 60 70 80 90 100
% of drive
Hei
gh
t (c
m)
stick head
stick hose
right wrist
right elbow
rightshoulder
Figure 2.
IMPACT
Minimal height
Page 28 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
29
0
20
40
60
80
100
120
140
160
10 20 30 40 50 60 70 80 90 100
% of drive
Hei
gh
t (c
m)
stick head
stick hose
left wrist
left elbow
left shoulder
Figure 3.
Minimal height
IMPACT
Page 29 of 30 Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.
30
40
50
60
70
80
90
100
110
8 16 24 32 40 48 56 64 72 80 88 96
% of drive
Hei
gh
t (c
m)
right wrist
left wrist
60
70
80
90
100
110
120
130
10 20 30 40 50 60 70 80 90 100
% of drive
Hei
gh
t (c
m)
right elbow
left elbow
80
90
100
110
120
130
140
10 20 30 40 50 60 70 80 90 100
% of drive
Hei
gh
t (c
m)
right shoulder
left shoulder
Figure 4.
Out-of-phase coordinationmode
IMPACT
In-phase inter-limbcoordination
Out-of-phase coordination mode
a
c
IMPACT
IMPACT
Minimal height
Minimal height
Minimal height
b
Page 30 of 30Journal of Applied Biomechanics © 2008 Human Kinetics, Inc.