Contribution of interaction torques during dart throwing: Differences between novices and experts

Human Movement Science xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Human Movement Science journal homepage: www.elsevier.com/locate/humov

Full Length Article

Contribution of interaction torques during dart throwing: Differences between novices and experts ⁎

Nasser Rezzouga, , Clint Hansenb, Philippe Gorcea, Brice Isableuc a b c

HandiBio EA 4322, Université de Toulon, Toulon, France Department of Neurology, University Hospital Schleswig-Holstein, Christian-Albrechts-University, Kiel, Germany Aix-Marseille Univ, PSYCLE, Aix en Provence, France

AR TI CLE I NF O

AB S T R A CT

Keywords: Interaction torque Impulses Coordination Dart throwing Inverse dynamics Expertise

We examined if experts and novices show different utilization of the torque components impulses during dart throwing. Participants threw darts continuously at a dartboard aiming for the centre (target bull’s eye). The upper-limb joint torque impulses were obtained through inverse dynamics with anthropometric and motion capture data as input. Depending on the joint degree of freedom (DOF) and movement phase (acceleration and follow-through), three main strategies of net torque (NET) impulse generation through joint muscle (MUS) and interaction (INT) torque impulses were highlighted. Firstly, our results showed that the elbow flexion–extension DOF leads the movement according to the joint leading hypothesis. Then, considering the acceleration phase, the analysis revealed differences in torque impulse decomposition between expert and novices. For the glenohumeral (GH) joint abduction–adduction and for wrist flexion, the INT torque impulse contributed positively to NET joint torque impulse in the group of experts unlike novices. This allowed to lower the necessary MUS torque impulse at these DOFs. Also, GH axial rotation was actively controlled by muscle torque impulse in the group of experts. During the follow-through, the experts used the INT torque impulse more proficiently than novices to break the elbow extension. The comparison between experts and novices through inverse dynamics document the link between the exploitation of interaction torques impulses and expertise in dart throwing for which the main objective is precision rather than velocity.

1. Introduction Depending on the sport discipline, the goal and the optimized biomechanical variables differ. In overarm throwing, the effective utilization of the non-muscular interaction torques (INT) which strongly depends on the throwing phase (Hansen, Rezzoug, Gorce, Venture, & Isableu, 2016) is a key factor in creating dynamic limb motion to maximize hand velocity (Hirashima, Kudo, Watarai, & Ohtsuki, 2007; Hirashima, Yamane, Nakamura, & Ohtsuki, 2008; Kinoshita et al., 2017; Naito & Maruyama, 2008; Neal, Snyder, & Kroonenberg, 1991; Timmann, Lee, Watts, & Hore, 2008). Studies considering 2D models have shown that INT at the elbow, due to proximal joint movements and torques, contributes to its acceleration while INT is counterbalanced at the wrist by the muscle torque (MUS) to precisely control the timing of ball release (Hirashima, Kudo, & Ohtsuki, 2003; Hirashima, Ohgane, Kudo, Hase, & Ohtsuki, 2003). Other studies considering 3D modeling and skilled baseball players revealed the presence of flexion INT at the wrist due to trunk and shoulder movements and muscles torques (Hirashima et al., 2007). Moreover, the braking of the shoulder

⁎

Corresponding author at: HandiBio, EA 4322, Université de Toulon, CS 60584, 83041 Toulon Cedex 9, France. E-mail addresses: [email protected] (N. Rezzoug), [email protected] (C. Hansen), [email protected] (P. Gorce), [email protected] (B. Isableu). http://dx.doi.org/10.1016/j.humov.2017.09.004 Received 30 May 2017; Received in revised form 6 September 2017; Accepted 6 September 2017 0167-9457/ © 2017 Elsevier B.V. All rights reserved.

Please cite this article as: Rezzoug, N., Human Movement Science (2017), http://dx.doi.org/10.1016/j.humov.2017.09.004

Human Movement Science xxx (xxxx) xxx–xxx

N. Rezzoug et al.

movement contributes to elbow acceleration and late INT at the elbow helps braking its movement (Debicki, Watts, Gribble, & Hore, 2010; Hore, Debicki, Gribble, & Watts, 2011). This contribution of INT is also important during movements such as reaching executed at self selected speed (Yamasaki, Tagami, Fujisawa, Hoshi, & Nagasaki, 2008). Therefore, the question arises if INT can be a marker of expertise in throwing activities such as dart throwing for which maximal velocity is not the primary objective. Previous work has shown that practice is associated with reduced movement variability and a decrease of cross-correlation between shoulder and elbow angular displacements (McDonald, van Emmerik, & Newell, 1989). The modification of joint angle variability over time (decrease of non-goal equivalent and goal equivalent variances) was also shown using the uncontrolled manifold hypothesis (Yang & Scholz, 2005). The experts minimize velocity errors (Smeets, Frens, & Brenner, 2002) and have either a longer release window to accurately throw a dart using the appropriate hand trajectories or very low timing errors compared to novices (Nasu, Matsuo, & Kadota, 2014). Also, no significant difference in performance nor consistency have been found according to the time of day for standard throws performed at a distance of 2.37 m from the target (Edwards, Waterhouse, Atkinson, & Reilly, 2007). Only few studies have investigated the dynamics of dart throwing and especially the possible link between INT exploitation and expertise. It has been shown that INT at the elbow is positively correlated with performance highlighting the importance of dynamic interaction (Tamei, Obayashi, & Shibata, 2011). However, this study included only four subjects (2 skilled and 2 novices) and further investigation is needed to study this possible relationship more thoroughly. In this framework, we examined the variations of the underlying dynamics with novice and expert dart players under regular conditions using optical motion capture. We hypothesized that experts, in contrast to novices, make use of INT impulse generated by the elbow and shoulder joints, to contribute to 1) the NET torque impulse at the wrist to control its flexion and 2) to decrease MUS impulse at the shoulder degrees of freedom (DOF). 2. Methods 2.1. Participants Eight male novice participants (mean = 24, sd = 2 years; mean = 1.80, sd = 0.05 m; mean = 73, sd = 8.8 kg) and thirteen expert male dart players (mean = 34, sd = 10 years; mean = 1.80, sd = 0.08 m; mean = 90.5, sd = 18.2 kg) were recruited to participate in the experiment. All participants were right handed according to the Edinburgh handedness questionnaire (Oldfield, 1971) and voluntarily took part in the experiment after signing a statement of informed consent pertaining to the experimental procedure which was approved by the EA 4532 local Ethics Committee as required by the Helsinki declaration (World Medical Association, 2013). Participants were free of sensory, perceptual, and motor (shoulder, elbow, and wrist) disorders and naïve to the purpose of the experiment. The novice subjects had no specific experience of dart throwing and had only played darts a few times before while the experts were competitors in the French dart throwing championship. Their experience playing darts ranged from four to twenty years. The data of one novice subject were withdrawn from the analysis due to technical problems during the recording session. 2.2. Instructions According to the World Darts Federation rules, the horizontal distance between the front of the board and any part of the shoes was at least 2.37 and the centre of the board (the target bull’s eye) was 1.73 m above the floor. The novice dart players performed the task with the same competition standard darts (18 g) and no manipulation or instruction was given on how the task should be performed except that the general throwing posture and technique should not vary throughout the trials. The experts on the other hand played with their personal darts to avoid perturbations in their performance. Following a 10-min warm-up, each subject was instructed to perform 10 dart throws repeatedly aiming for the target bull’s eye. Before the recording session, novices performed ten rounds of three darts to the board resulting in 30 throws, while the experts judged their readiness within the given time limit. For the experts, the data collection took part during the French Open and participants were free to “walk in” between their games. No specific time constraints were put on the players but data collection was between 10.00 am and 5.00 pm. The novices (students) were asked to perform the experiment during university day and data collection was between 09.00 am and 5.00 pm. 2.3. Movement phases The dart throwing motion consists of four phases: the aiming, the backward move, the acceleration and the follow through. The aiming consists in focusing on the target. It is followed by the backward move during which the elbow is slightly flexed. At the end of this phase, the upper limb joint velocities are equal to zero. The beginning of the acceleration phase coincides with the reversal from flexion to extension at the elbow joint and it ends with the instant of dart release (ToR) (Hansen, Rezzoug, Gorce, & Isableu, 2012). The follow-through corresponds to the end of the arm movement during which the articular movements of the joints are decelerated and then stopped. In this paper, the analysis was focused on the acceleration and the follow-through phases. 2.4. Performance measure The position of the dart on the target was ranked from 1 to 10 according to its vicinity with the target bull’s eye. The scores of the ten throws were averaged. 2

Human Movement Science xxx (xxxx) xxx–xxx

N. Rezzoug et al.

2.5. Joint angles computation Movements were recorded with a V8i VICON eight camera (Mcam2) motion capture system at a rate of 250 Hz (Vicon motion systems Inc., Oxford, UK). Participants wore seventeen markers on the right upper-limb and the torso located at the following anatomical points: xiphoid process, suprasternal notch, acromion, lateral and medial humeral epicondyles, radial and ulnar styloid processes, middle of the third metacarpi, distal extremity of the second and fifth metacarpi, seventh cervical vertebra, tenth thoracic vertebra, right and left anterior superior iliac spine, right and left posterior superior iliac spine. In addition, three technical markers were placed on the forearm and arm. The same person placed the markers in order to avoid bias. The resulting data were low pass filtered at 20 Hz using a 2nd-order Butterworth filter. The convention adopted to define the joint angles (glenohumeral flexion–extension, abduction–adduction, internal-external rotation, elbow flexion, forearm pronation-supination, wrist radial-ulnar deviation and flexion–extension) and joint centre positions followed the recommendations of the International Society of Biomechanics (ISB) (Wu et al., 2005) modified by Senk and Cheze (2006). The glenohumeral joint centre position was evaluated from the position of the right most dorsal point of the acromioclavicular joint marker according to (Schmidt, Disselhorst-Klug, Silny, & Rau, 1999). The joint angles were determined with global optimization (Lu & O'Connor, 1999; Roux, Bouilland, Godillon-Maquinghen, & Bouttens, 2002). 2.6. Inverse dynamics The dart throwing movement kinetics was assessed through the determination of the time series of NET, INT, MUS, and GRAV. For one degree of freedom (DOF), NET represents the torque due to the DOFs angular acceleration; INT reflects the torque effect of the movement of adjacent DOFs and can be decomposed in two different parts: 1. The torque due to the acceleration of adjacent DOFs, 2. The torque due to the DOFs angular velocity (centrifugal and Coriolis). Furthermore, GRAV reflects the torque due to gravity. Finally, the residual torque MUS corresponds to the active contribution of muscles and the passive action of connective tissues such as tendons, ligaments, articular capsules, and other connective tissues (Hollerbach & Flash, 1982; Sainburg, Ghilardi, Poizner, & Ghez, 1995): MUS = NET − INT − GRAV. The seven DOFs of the upper-limb model (glenohumeral (GH) flexion–extension (FE), abduction–adduction (AA) and axial rotation (AR), elbow (EL) flexion–extension, forearm (FO) pronation-supination (PS), wrist (WR) radioulnar deviation (RD) and flexion–extension) were defined based on the Denavit-Hartenberg convention (Denavit & Hartenberg, 1955). Then, the joint torque timeseries were computed by using a recursive Newton-Euler method (Luh, Walker, & Paul, 1980) as described in (Isableu et al., 2009). The details of the computations are given in the data-in-brief companion article. The participants’ mass and segments lengths were obtained from the optoelectronic system and the anthropometric data (i.e. segments mass, position of centre of mass and inertia matrices) set according to Dumas, Cheze, and Verriest (2007). The joint angular velocities and accelerations were obtained by numerically differentiating the joint angles with respect to time and the NET, INT, MUS, and GRAV impulses were evaluated by numerically integrating the torque time series during the acceleration and followthrough phases. This procedure allowed to sum up the contribution of the various torque components during the complete movement phases instead of considering torque values at isolated instants. This allowed to globally consider the contribution of each torque component similar to previous studies (Hore, Debicki, & Watts, 2005; Sainburg, 2002). The joint torque impulses obtained from the inverse dynamics computations were calculated for each trial and then averaged for each subject. 2.7. Statistical analysis Student t tests for independent samples were conducted between novices and experts with Statistica (Statsoft, Tulsa, OK, USA). The normality was checked with the Shapiro Wilk’s test and the homogeneity of variances with the Levene’s Test. The dependent variables were: movement time, joint movement amplitudes, NET, INT, MUS, and GRAV impulses at each DOF during the acceleration (ACC) and follow-through (FTH) phases. The level of significance was set to 0.05. The effect size was evaluated with the Hodge’s g due to the fact that the samples sizes were different. 3. Results The experts were significantly older than novices (mean = 34, sd = 10 vs mean = 24, sd = 2 years, t = 2.77, P < 0.02). 3.1. Performance measure All experts executed throws that attained either the target centre (score 10) or at least the first circle (score 9). No throw was missed. For the novices, two subjects missed one throw. The novices average score was 5.60 and the standard deviation was 2.39. Therefore, it was concluded that experts had better performances than novices. 3

Human Movement Science xxx (xxxx) xxx–xxx

N. Rezzoug et al.

Fig. 1. Comparison between NOVICES and EXPERTS according to the upper-limb joint angle amplitudes. A: acceleration phase, B: follow-through (GH: glenohumeral joint, EL: Elbow joint, WR: wrist joint, FE: flexion–extension, AA: abduction–adduction, AR: axial rotation, PS: pronation-supination, RD: radial-ulnar deviation). * P < 0.05.

3.2. Movement kinematics The duration of the acceleration phase was not significantly different between experts and novices (mean = 67 ms, sd = 18 ms for experts vs mean = 62 ms, sd = 11 ms for novices, t = −0.69, P = 0.50) and the follow-through was longer for experts compared to novices (mean = 63 ms, sd = 12 ms vs mean = 51 ms, sd = 6 ms, t = −2.41, P < 0.03). During the acceleration phase the experts exhibited a lower amplitude of flexion at the glenohumeral joint (mean = 4.4°, sd = 3.8 vs mean = 8.7°, sd = 4.5°, t = 2.33, P < 0.05) and an increased wrist ulnar deviation (mean = 15.8°, sd = 4.6° vs mean = 5.6°, sd = 1.5°, t = −5.96, P < 0.05). No significant difference was highlighted for the other DOFs (Fig. 1A). Concerning the follow through, the glenohumeral axial rotation (mean = 15.6°, sd = 8.2° vs mean = 6.6°, sd = 3.8°, t = −2.90, P < 0.05) and wrist ulnar deviation (mean = 18.0°, sd = 8.3° vs mean = 4.0°, sd = 1.7° vs t = −4.65, P < 0.05) were greater for the experts in comparison to the novices. The difference between experts and novices did not reach the significance level for the glenohumeral flexion (mean = 38.0°, sd = 16.0° vs mean = 25.5°, sd = 7.3°, t = −2.06, P = 0.052). The amplitude of joint movement was not significantly different for the other DOFs (Fig. 1B).

4

Human Movement Science xxx (xxxx) xxx–xxx

N. Rezzoug et al.

Table 1 Comparison between NOVICES and EXPERTS mean (M) and standard deviation (S) MUS, NET, INT, and GRAV torque impulses (N.m.s) during the acceleration (ACC) and follow through (FTH) phases. Each panel corresponds to an upper-limb DOF (GH: glenohumeral joint, EL: Elbow joint, WR: wrist joint, FE: flexion–extension, AA: abduction–adduction, AR: axial rotation, PS: pronation-supination, RD: radial-ulnar deviation, SD: standard deviation). Significant differences are highlighted in grey. t, P, and Hodge’s g effect size values are provided. ACC Shoulder

MUS

Nov Exp

NET

Nov Exp

INT

Nov Exp

GRV

Nov Exp

ACC Elbow

ACC Wrist

FTH Shoulder

FTH Elbow

FTH Wrist

FE

AA

AR

FE

PS

RD

FE

FE

AA

AR

FE

PS

RD

FE

M S M S t P g

−0,05 1,00 0,66 0,80 −1,73 0,10 0,81

0,60 0,23 −0,01 0,59 2,62 0,02 1,23

0,19 0,28 −0,35 0,35 3,57 0,002 1,67

−1,51 0,37 −1,48 0,48 −0,16 0,88 0,07

−0,02 0,02 −0,04 0,02 2,82 0,01 1,32

0,021 0,008 0,030 0,030 −0,77 0,45 0,36

0,03 0,02 0,01 0,02 2,19 0,04 1,03

0,42 0,77 0,23 0,49 0,67 0,51 0,32

−0,41 0,22 0,35 0,50 −3,80 0,001 1,78

−0,05 0,16 0,32 0,12 −5,92 0,00001 2,78

0,96 0,18 0,96 0,34 −0,01 0,99 0,002

0,02 0,01 0,04 0,01 −3,43 0,003 1,61

−0,02 0,01 −0,03 0,01 1,19 0,25 0,56

−0,07 0,02 −0,07 0,02 −0,004 1,00 0,002

M S M S t P g

0,82 0,39 0,89 0,55 −0,32 0,75 0,15

0,39 0,32 0,72 0,39 −1,88 0,08 0,88

0,05 0,15 −0,32 0,17 4,67 0,0002 2,19

−1,59 0,33 −1,65 0,54 0,26 0,79 0,12

−0,00005 0,00121 −0,00042 0,00194 0,46 0,65 0,22

0,021 0,008 0,030 0,030 −0,77 0,45 0,36

0,002 0,002 0,01 0,002 −3,63 0,002 1,70

−1,71 0,81 −1,63 0,83 −0,22 0,83 0,10

−0,63 0,53 −1,28 0,59 2,41 0,03 1,13

−0,01 0,14 0,24 0,15 −3,54 0,002 1,66

1,34 0,24 1,59 0,45 −1,39 0,18 0,65

−0,001 0,001 −0,001 0,001 −0,38 0,71 0,18

−0,001 0,0004 −0,003 0,001 5,76 0,00002 2,70

−0,001 0,002 −0,01 0,002 4,32 0,0004 2,02

M S M S t P g

1,28 0,92 0,83 0,55 1,37 0,19 0,64

−0,10 0,38 0,86 0,58 −3,90 0,001 1,83

−0,18 0,15 −0,03 0,32 −1,14 0,27 0,53

−0,13 0,04 −0,19 0,18 0,88 0,39 0,41

0,02 0,01 0,04 0,02 −2,66 0,02 1,25

−0,020 0,007 −0,028 0,029 0,72 0,48 0,34

−0,02 0,02 0,002 0,02 −2,75 0,01 1,29

−1,66 0,66 −1,12 0,73 −1,63 0,12 0,76

−0,11 0,60 −1,48 0,65 4,58 0,0002 2,15

0,02 0,09 −0,10 0,15 1,85 0,08 0,87

0,47 0,28 0,79 0,34 −2,15 0,04 1,01

−0,02 0,01 −0,04 0,01 3,34 0,004 1,57

0,02 0,01 0,02 0,01 −0,39 0,70 0,18

0,07 0,02 0,06 0,02 0,81 0,43 0,38

M S M S t P g

−0,41 0,12 −0,60 0,27 1,71 0,10 0,80

−0,10 0,06 −0,13 0,07 0,86 0,40 0,40

0,04 0,03 0,07 0,05 −1,38 0,19 0,65

0,05 0,03 0,02 0,03 2,10 0,05 0,99

0,001 0,001 0,002 0,002 −1,52 0,15 0,71

−0,001 0,001 −0,001 0,001 0,22 0,83 0,10

−0,005 0,002 −0,006 0,002 1,25 0,23 0,59

−0,48 0,10 −0,74 0,25 2,65 0,02 1,24

−0,11 0,06 −0,16 0,09 1,32 0,20 0,62

0,02 0,02 0,01 0,03 0,49 0,63 0,23

−0,09 0,02 −0,16 0,05 3,27 0,004 1,53

0,0008 0,0002 0,001 0,001 0,32 0,75 0,15

0,001 0,001 0,004 0,002 −3,10 0,01 1,45

0,001 0,001 0,003 0,002 −3,57 0,002 1,67

3.3. Movement dynamics 3.3.1. Torque impulses The torque impulse of MUS, NET, INT and GRAV were assessed separately for the acceleration and follow through phases. The differences between novices and experts for the examined parameters are outlined in Table 1. 3.3.2. Torque exploitation strategies Three main strategies of MUS and INT exploitation to generate the appropriate NET impulse were highlighted (Fig. 2 and Table 2). The first one (S1) involved an important MUS and a low INT impulse. In this case, the main contribution to the acceleration of one DOF (NET) was due to the active participation of the muscles and passive connective tissues. Two sub strategies were noticed: S1a for which MUS and INT had the same sign and acted in synergy and S1b for which MUS acted against INT with an opposite sign. The elbow flexion–extension kinetics followed S1a during the acceleration phase while experts’ glenohumeral axial rotation during FTH followed S1b. The second strategy (S2) was observed in the case of a low NET impulse compared to MUS and INT. Mean MUS and

Fig. 2. Identified strategies of torque impulse exploitation at each DOF.

5

Human Movement Science xxx (xxxx) xxx–xxx

N. Rezzoug et al.

Table 2 Torque impulse exploitation strategies according to DOF, movement phase and group. Degree of freedom

Glenohumeral flexion–extension Glenohumeral abduction–adduction Glenohumeral axial rotation Elbow flexion–extension Forearm pronation-supination Wrist radial-ulnar deviation Wrist flexion–extension

Acceleration phase

Follow through

Novices

Experts

Novices

Experts

S3a S1a S2 S1a S2 S2 S2

S3a S3a S1a S1a S2 S2 S3a

S3b S1a S1a S1a S2 S2 S2

S3b S3b S1b S1a S2 S2 S2

S2

NET had the same sign while MUS and INT had opposite signs because MUS counteracted the effect of interaction torques due to the movement of adjacent DOFs (forearm pronation-supination and wrist radial-ulnar deviation). The third observed strategy (S3) was found when NET and INT impulses had the same sign. Two sub-strategies have been identified: S3a if the INT impulse was lower than NET, then MUS acted in synergy with INT. On the contrary, if INT had greater amplitude compared to NET, then MUS had a different sign than INT to counteract its action (S3b). During the acceleration phase, GRAV impulse was not significantly different between groups except for the elbow flexion–extension but with very low values. The results are summarized in Table 1. S1a corresponded to a strategy for which muscular action mainly controlled the joint movement. S2 corresponded to the intervention of muscle action to compensate for an INT to generate a comparatively low NET impulse. Finally, S3b and S3a illustrated the exploitation of the chain kinetics to produce an appropriate NET impulse with lower muscular effort. In the present study, it was considered that S3a was representative of the skill level and the capacity to exploit the interaction torques in synergy with MUS. During the acceleration phase, experts tended to exploit the kinetics of the chain at both the glenohumeral joint (flexion–extension and abduction–adduction) and the wrist (flexion–extension) while such strategy was found only for the glenohumeral flexion–extension for novices. For both groups the elbow extension was the DOF with the greatest movement amplitude obtained mainly by muscular action (S1a). For novices, this seemed also to be the case for the glenohumeral adduction (S1a), because they relied on MUS while experts seemed to exploit the interaction torques (S3a) to generate the DOF NET impulse. For the glenohumeral axial rotation which remained low in amplitude, Novices used the S2 strategy while experts relied more on MUS (S1a). For the forearm pronation and wrist radial-ulnar deviation both groups used the S2 strategy whatever the movement phase. During the follow through, for both groups, the elbow was decelerated with a flexion impulse using strategy S1a. Also, for both groups, the forearm pronation-supination, wrist radial-ulnar deviation and flexion–extension INT were compensated for by MUS following the strategy S2. At the glenohumeral joint, the extension NET impulse was controlled via INT modulated by MUS impulse following strategies S3b. Similarly, to the acceleration phase, the novices used S1a to control the adduction NET impulse while experts relied on strategy S3b. The external rotation NET impulse was obtained via strategy S1b, i.e. via an important MUS contribution modulated by an INT of opposite sign while no clear strategy was found for novices due to the variability of the torque impulses (S1a or S2). 4. Discussion The aim of this paper was to study differences of dynamic torque exploitation between experts and novices during dart throwing. The main finding of the present study was consistent with our hypothesis stating that experts, in contrast to novices, use the interaction torque impulses more proficiently. During the acceleration phase, experts tended to exploit the kinetics of the chain at both the glenohumeral joint (flexion–extension and abduction–adduction) and the wrist (flexion–extension) while such strategy was found only for the glenohumeral flexion–extension for novices. 4.1. Joint leading hypothesis The leading joint hypothesis (Dounskaia, 2005) suggests that one degree of freedom in the kinematic chain dominates the others and lead the movement by the interaction torques arising from it. In this case, the leading DOF is the one for which the NET is due predominantly to the action of MUS that generates its movement (e.g. the S1 profile). For fast movements for which the maximum velocity is needed, the proximal DOF is often leading due to the relatively high inertia and increased musculature (Hirashima, Kudo, et al., 2003). In the current study, it appears that the elbow flexion–extension is the leading DOF since during the acceleration phase the MUS torque impulse contributes significantly to the NET torque impulse while INT impulse is low in comparison. The same behaviour was observed for the two groups. However, during the acceleration phase, the glenohumeral axial rotation is dominated by MUS in the group of experts. By considering the starting time of muscle activity from an EMG analysis of eight upper-limb muscles, a previous study has suggested that the Proximal-to-Distal segmental sequencing is not observed during dart throwing therefore corroborating the hypothesis of the elbow as the leading joint (Obayashi, Tamei, Imai, & Shibata, 2009). Studies analysing 2D over arm throwing which considered only the elbow and wrist joints found similar results (Hirashima, Kudo, et al., 2003) with an elbow extension NET due predominantly to muscle torque. Although, joint amplitudes were relatively low at the glenohumeral joint, the present study considered 3D unconstrained movements involving glenohumeral, elbow and wrist joints. Therefore, the fact that the 6

Human Movement Science xxx (xxxx) xxx–xxx

N. Rezzoug et al.

elbow is the leading joint may be considered as a specific characteristic of dart throwing. 4.2. Expertise and interaction torque The exact contribution of INT to the movement dynamics is still under debate because some studies argue that INT may be either resistive or assistive depending on the task (Debicki, Gribble, Watts, & Hore, 2011; Dounskaia, Wisleder, & Johnson, 2005) or even the limb’s postures (Gritsenko, Kalaska, & Cisek, 2011). Considering the movement decomposition in phases, the definition of several strategies of torque impulse exploitation is proposed that encompasses both aspects in a unified view. Our results show that the chosen strategies and consequently the usage of INT differ depending on the DOF and the level of expertise. Past research has shown that the suitable exploitation or counteraction of interaction torque can be considered as a marker of expertise in sport science for throwing movements involving a high level of velocity and acceleration e.g. during baseball pitching or overarm throwing (Debicki, Gribble, Watts, & Hore, 2004; Hirashima, Kudo, et al., 2003; Hirashima et al., 2007; Naito, Takagi, Yamada, Hashimoto, & Maruyama, 2014). These studies have shown that increased muscle torque at proximal joints (trunk and shoulder but not at elbow and wrist) are used to generate the appropriate INT to produce maximal hand velocity. Our results extend these conclusions to the case where maximal velocity is not the primary objective. In the present study, experts tend to use strategies corresponding to an efficient exploitation of INT which contributes positively to NET torques at the glenohumeral joint but also at the wrist during the acceleration phase while INT contributes more proficiently to the elbow braking during the follow-through in the group of experts (Hore et al., 2011). At the glenohumeral joint, the contribution of INT impulse tends to lower the required MUS torque impulse to execute the abduction–adduction movement. However, the contribution of MUS to control the humeral axial rotation is predominant and coincides with a low INT. The counteraction of INT torque impulse by MUS impulse at the wrist i.e. via strategy S2 has been proposed in previous studies considering 2D overarm throwing as a mean to control the wrist velocity in order to ensure an accurate release (Hirashima, Kudo, et al., 2003; Hirashima, Ohgane, et al., 2003). It has been suggested that the mechanical properties of the human wrist are the main determinant of the strategy S2 at this joint, which is advantageous for keeping the state of the wrist joint stable in multi-joint upper limb movements and would lead to avoidance of excessive wrist extension or flexion and simplification of extrinsic finger control (Hirashima, Ohgane, et al., 2003). The same behaviour has also been observed during reaching tasks for which INT always counteracts MUS (Galloway & Koshland, 2002). On the contrary, when 3D unconstrained movements executed by skilled baseball players are considered, it has been observed that INT arising from proximal joints at the trunk and shoulder contributes to wrist flexion (Hirashima et al., 2007). The S2 strategy has been highlighted in both groups for forearm pronation supination and wrist radial ulnar deviation while experts tend to exploit INT at the wrist to generate a flexion through strategy S3a. A close examination of the interindividual differences revealed that 8 out of 13 experts, presented a flexion INT impulse that acted in synergy with MUS to produce NET while none of the novices could generate such a flexion INT. Moreover, a lower MUS and higher NET impulse was observed for experts compared to novices. The dart throwing movement has a fundamental difference with overarm throwing, since its leading joint is the elbow and, therefore, the expertise in dart throwing seems to combine characteristics observed during 2D movements with fixed shoulder, i.e. an elbow leading joint, with characteristics of skilled throwers by the presence of a flexion INT at the wrist acting in synergy with MUS (Hirashima et al., 2007). On the contrary, during the follow-through, the same strategy (S2) was used by both groups who relied on muscle torque to decelerate the wrist flexion. A possible explanation may be that the glenohumeral joint also contributed to INT at the wrist as observed during 3D throws using the MUS generated at the glenohumeral axial rotation DOF which differentiates novices and experts. However, this hypothesis needs to be further verified by using e.g. an induced acceleration analysis (Hirashima, 2011). Some authors have suggested that the squared integral of INT at the elbow was correlated positively with performance (Tamei et al., 2011). However, no between groups difference of INT impulse was highlighted at the elbow during the acceleration phase in contrast to the follow-through during which a greater INT impulse at the elbow for experts compared to novices contributed to decelerate its flexion. In this study, the time of release was identified as the instant at which the vertical hand position was the highest. In contrast, some studies found the dart time of release to occur slightly before hand maximal vertical position (Hansen et al., 2012; Smeets et al., 2002). Therefore, the correlation observed between performance and INT at the elbow may correspond partly to the increase of INT during the follow through phase. This is the first study focussing on the influence of INT impulse during a precision throwing task using a seven DOF 3D model of the upper-limb while comparing the torque exploitation strategies between two independent cohorts. To our knowledge, a more proficient use of interaction torque impulse in experts compared to novices has not been shown before in darts. As a first experiment this study has some limitations. The sample size for each group was relatively small and, due to participant’s inter-individual variability, a larger cohort may be needed to strengthen the study conclusion. 5. Conclusion The results of our experiment showed that novices and experts present differences in the exploitation of torque impulse components (INT, MUS, NET and GRAV) of the upper-limb during dart throwing. The analysis of torque impulses revealed three main strategies depending on the DOF and movement phase. Our results showed that the classical proximal to distal pattern does not necessarily apply since the elbow flexion–extension DOF leads the movement according to the leading joint hypothesis. In more details, it appeared that the contribution of INT to the NET impulse for the glenohumeral joint abduction–adduction (lowering the necessary MUS impulse) and wrist flexion is prominent in the group of experts unlike novices while the humeral axial rotation NET 7

Human Movement Science xxx (xxxx) xxx–xxx

N. Rezzoug et al.

impulse is predominantly generated by the MUS impulse in the group of experts. Also, during the follow-through the experts exploited the INT torque impulse more proficiently than novices to break the elbow extension. In addition to activities requiring high velocities such as overarm throwing, the present results extend the observation that proficient INT exploitation is an important characteristic of skilled performers to the case of dart throwing for which precision is the most critical feature compared to velocity. Acknowledgements This research was granted by the French Ministère de l’Enseignement Supérieur et de la Recherche to the second author and by the Centre National de la Recherche Scientifique, Ile de France to the fourth author. References Debicki, D. B., Gribble, P. L., Watts, S., & Hore, J. (2004). Kinematics of wrist joint flexion in overarm throws made by skilled subjects. Experimental Brain Research, 154, 382–394. Debicki, D. B., Gribble, P. L., Watts, S., & Hore, J. (2011). Wrist muscle activation, interaction torque and mechanical properties in unskilled throws of different speeds. Experimental Brain Research, 208, 115–125. Debicki, D. B., Watts, S., Gribble, P. L., & Hore, J. (2010). A novel shoulder-elbow mechanism for increasing speed in a multijoint arm movement. Experimental Brain Research, 203, 601–613. Denavit, J., & Hartenberg, R. S. (1955). A kinematic notation for lower-pair mechanisms based on matrices. Transactions ASME Journal of Applied Mechanics, 23, 215–221. Dounskaia, N. (2005). The internal model and the leading joint hypothesis: implications for control of multi-joint movements. Experimental Brain Research, 166, 1–16. Dounskaia, N., Wisleder, D., & Johnson, T. (2005). Influence of biomechanical factors on substructure of pointing movements. Experimental Brain Research, 164, 505–516. Dumas, R., Cheze, L., & Verriest, J. P. (2007). Adjustments to McConville et al. and Young et al. body segment inertial parameters. Journal of Biomechanics, 40, 543–553. Edwards, B., Waterhouse, J., Atkinson, G., & Reilly, T. (2007). Effects of time of day and distance upon accuracy and consistency of throwing darts. Journal of Sports Sciences, 25, 1531–1538. Galloway, J. C., & Koshland, G. F. (2002). General coordination of shoulder, elbow and wrist dynamics during multijoint arm movements. Experimental Brain Research, 142, 163–180. Gritsenko, V., Kalaska, J. F., & Cisek, P. (2011). Descending corticospinal control of intersegmental dynamics. Journal of Neuroscience, 31, 11968–11979. Hansen, C., Rezzoug, N., Gorce, P., & Isableu, B. (2012). Is the time of release during a precision throwing task, predictable? Computer Methods in Biomechanics and Biomedical Engineering, 15(Suppl 1), 250–252. Hansen, C., Rezzoug, N., Gorce, P., Venture, G., & Isableu, B. (2016). Sequence-dependent rotation axis changes and interaction torque use in overarm throwing. Journal of Sports Sciences, 34, 878–885. Hirashima, M. (2011). Induced Acceleration Analysis of Three-Dimensional Multi-Joint Movements and Its Application to Sports Movements. In D. V. Klika (Ed.): InTech. Hirashima, M., Kudo, K., & Ohtsuki, T. (2003). Utilization and compensation of interaction torques during ball-throwing movements. Journal of Neurophysiology, 89, 1784–1796. Hirashima, M., Kudo, K., Watarai, K., & Ohtsuki, T. (2007). Control of 3D limb dynamics in unconstrained overarm throws of different speeds performed by skilled baseball players. Journal of Neurophysiology, 97, 680–691. Hirashima, M., Ohgane, K., Kudo, K., Hase, K., & Ohtsuki, T. (2003). Counteractive relationship between the interaction torque and muscle torque at the wrist is predestined in ball-throwing. Journal of Neurophysiology, 90, 1449–1463. Hirashima, M., Yamane, K., Nakamura, Y., & Ohtsuki, T. (2008). Kinetic chain of overarm throwing in terms of joint rotations revealed by induced acceleration analysis. Journal of Biomechanics, 41, 2874–2883. Hollerbach, M. J., & Flash, T. (1982). Dynamic interactions between limb segments during planar arm movement. Biological Cybernetics, 44, 67–77. Hore, J., Debicki, D. B., Gribble, P. L., & Watts, S. (2011). Deliberate utilization of interaction torques brakes elbow extension in a fast throwing motion. Experimental Brain Research, 211, 63–72. Hore, J., Debicki, D. B., & Watts, S. (2005). Braking of elbow extension in fast overarm throws made by skilled and unskilled subjects. Experimental Brain Research, 164, 365–375. Isableu, B., Rezzoug, N., Mallet, G., Bernardin, D., Gorce, P., & Pagano, C. C. (2009). Velocity-dependent changes of rotational axes in the non-visual control of unconstrained 3D arm motions. Neuroscience, 164, 1632–1647. Kinoshita, H., Obata, S., Nasu, D., Kadota, K., Matsuo, T., & Fleisig, G. S. (2017). Finger forces in fastball baseball pitching. Human Movement Science, 54, 172–181. Lu, T. W., & O'Connor, J. J. (1999). Bone position estimation from skin marker co-ordinates using global optimisation with joint constraints. Journal of Biomechanics, 32, 129–134. Luh, J. Y. S., Walker, M. W., & Paul, R. P. C. (1980). On-line computational scheme for mechanical manipulators. Journal of Dynamic Systems, Measurement, and Control, 102, 69–76. McDonald, P. V., van Emmerik, R. E. A., & Newell, K. M. (1989). The effects of practice on limb kinematics in a throwing task. Journal of Motor Behavior, 21, 245–264. Naito, K., & Maruyama, T. (2008). Contributions of the muscular torques and motion-dependent torques to generate rapid elbow extension during overhand baseball pitching. Sports Engineering, 11, 47–56. Naito, K., Takagi, H., Yamada, N., Hashimoto, S., & Maruyama, T. (2014). Intersegmental dynamics of 3D upper arm and forearm longitudinal axis rotations during baseball pitching. Human Movement Science, 38, 116–132. Nasu, D., Matsuo, T., & Kadota, K. (2014). Two types of motor strategy for accurate dart throwing. PLoS One, 9, e88536. Neal, R. J., Snyder, C. W., Jr, & Kroonenberg, P. M. (1991). Individual differences and segment interactions in throwing. Human Movement Science, 10, 653–676. Obayashi, C., Tamei, T., Imai, A., & Shibata, T. (2009). Comparison of experts and non-experts in throwing darts based on optimization criteria. In 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society (pp. 2647–2650). Oldfield, R. C. (1971). The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia, 9, 97–113. Roux, E., Bouilland, S., Godillon-Maquinghen, A. P., & Bouttens, D. (2002). Evaluation of the global optimisation method within the upper limb kinematics analysis. Journal of Biomechanics, 35, 1279–1283. Sainburg, R. L. (2002). Evidence for a dynamic-dominance hypothesis of handedness. Experimental Brain Research, 142, 241–258. Sainburg, R. L., Ghilardi, M. F., Poizner, H., & Ghez, C. (1995). Control of limb dynamics in normal subjects and patients without proprioception. Journal of Neurophysiology, 73, 820–835. Schmidt, R., Disselhorst-Klug, C., Silny, J., & Rau, G. (1999). A marker-based measurement procedure for unconstrained wrist and elbow motions. Journal of Biomechanics, 32, 615–621. Senk, M., & Cheze, L. (2006). Rotation sequence as an important factor in shoulder kinematics. Clinical Biomechanics (Bristol, Avon), 21 Suppl 1, S3–8. Smeets, J. B. J., Frens, M. A., & Brenner, E. (2002). Throwing darts: timing is not the limiting factor. Experimental Brain Research, 144, 268–274. Tamei, T., Obayashi, C., & Shibata, T. (2011). Throwing darts utilizes the interaction torque of the elbow joint. Conference Proceedings of IEEE Engineering in Medical and

8

Human Movement Science xxx (xxxx) xxx–xxx

N. Rezzoug et al.

Biology Society, 2011, 1283–1286. Timmann, D., Lee, P., Watts, S., & Hore, J. (2008). Kinematics of arm joint rotations in cerebellar and unskilled subjects associated with the inability to throw fast. The Cerebellum, 7, 366–378. World Medical Association (2013). World medical association declaration of helsinki: Ethical principles for medical research involving human subjects. JAMA, 310, 2191–2194. Wu, G., van der Helm, F. C. T., Veeger, H. E. J., Makhsous, M., Van Roy, P., Anglin, C., ... Buchholz, B. (2005). ISB recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion—Part II: shoulder, elbow, wrist and hand. Journal of Biomechanics, 38, 981–992. Yamasaki, H., Tagami, Y., Fujisawa, H., Hoshi, F., & Nagasaki, H. (2008). Interaction torque contributes to planar reaching at slow speed. BioMedical Engineering Online, 7, 27. Yang, J. F., & Scholz, J. P. (2005). Learning a throwing task is associated with differential changes in the use of motor abundance. Experimental Brain Research, 163, 137–158.

9