Catapult effect in pole vaulting: Is muscle coordination determinant? Julien Frère a,b,⇑ , Beat Göpfert c , François Hug a , Jean Slawinski d , Claire Tourny-Chollet b a Laboratoire ‘‘Motricité, Interactions, Performance’’ (MIP) (EA 4334), University of Nantes, F-44300 Nantes, France b Centre d’Etude des Transformations des Activités Physiques et Sportives (EA 3832), University of Rouen, France c Laboratory of Biomechanics and Biocalorimetry, Clinical Morphology & Biomedical Engineering, University of Basel, Switzerland d Centre de Recherches sur le Sport et le Mouvement (EA 2931), University of Paris Ouest Nanterre La Défense, France article info Article history: Received 5 January 2011 Received in revised form 18 September 2011 Accepted 2 October 2011 Keywords: Electromyography Motion analysis Non-negative matrix factorization Sport performance abstract This study focused on the phase between the time of straightened pole and the maximum height (HP) of vaulter and aimed at determining the catapult effect in pole vaulting on HP. Seven experienced vaulters performed 5–10 vaults recorded by two video cameras, while the surface electromyography (sEMG) activity of 10 upper limbs muscles was recorded. HP was compared with an estimated maximum height (HP est ) allowing the computation of a push-off index. Muscle synergies were extracted from the sEMG activity profiles using a non-negative matrix factorization algorithm. No significant difference (p > 0.47) was found between HP est (4.64 ± 0.21 m) and HP (4.69 ± 0.23 m). Despite a high inter- individual variability in sEMG profiles, two muscle synergies were extracted for all the subjects which accounted for 96.1 ± 2.9% of the total variance. While, the synergy activation coefficients were very sim- ilar across subjects, a higher variability was found in the muscle synergy vectors. Consequently, whatever the push-off index among the pole vaulters, the athletes used different muscle groupings (i.e., muscle synergy vectors) which were activated in a similar fashion (i.e., synergy activation coefficients). Overall, these results suggested that muscle coordination adopted between the time of straightened pole and the maximum height does not have a major influence on HP. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introduction Pole vault can be modelled in four phases: the run-up phase, the take-off phase, the pole bending phase, and the pole straightening phase. Several key times were determined within the pole straight- ening phase: the pole straightened (PS) and the pole release times were placed between the maximum pole bending position and the maximum height of centre of gravity of the pole vaulter (HP) (Frère et al., 2010). The energetic interaction between the pole and the athlete reveals that besides the mechanical energy of the vaulter coming from the run-up, the pole vaulter is also able to store an additional elastic energy in the pole by means of muscular work. This additional elastic energy allows the vaulter to have higher mechanical energy when crossing the bar than at take-off (Arampatzis et al., 2004). Also, it has been reported that elite pole vaulters may have different strategies to perform the vault and obtain this final mechanical energy gain, either favouring the pole bending phase or favouring the pole straightening phase. Accord- ingly, a recent review (Frère et al., 2010) highlighted that the actions of the shoulder muscles are of major importance in pole vaulting since they apply a bending moment on the pole and en- able the elevation of the lower limbs and pelvis over the shoulders before the bar clearance. Since previous studies have shown that the vertical velocity of the vaulter reaches its maximum around PS and decreases until HP (Gros and Kunkel, 1990; Angulo-Kinzler et al., 1994), the rela- tionship between the final actions of the vaulter (i.e., before pole release) and the performance may be questionable. Also, the cross- ing of a bar placed at 6 m for the victory in the fourth IAAF World Championships in Athletics in 1993 did likely not necessitate a push-off action of the arms on the pole when the pole was fully straightened (Hommel and Houvion, 1994). Thus, one would ex- pect that the vaulter benefits from the previous straightening of the pole to be catapulted over the bar in a relative ballistic fashion rather than performing a final push-off. The aim of this study was to determine the catapult effect on the height of the vaulter’s centre of gravity (CG) in pole vaulting. To investigate this question we first compared the maximum height of CG of the pole vaulter to an estimated value calculated from kine- matics recorded at PS. Then, we determined the coordination of the upper-limbs muscles during the vault between PS and HP. We hypothesized that the measured maximum height of the vaulter’s CG would not differ from the estimated one and that the final ac- tions of the pole vaulter would not be correlated to the maximum 1050-6411/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jelekin.2011.10.001 ⇑ Corresponding author at: Laboratory ‘‘Motricité Interactions Performance’’ (MIP) (EA 4334), Faculty of Sports Science, University of Nantes, 25 bis boulevard Guy Mollet, BP 72206, 44322 Nantes, Cedex 3, France. Tel.: +33 (0) 2 51 83 72 38; fax: +33 (0) 2 51 83 72 10. E-mail address: julien_frere@hotmail.com (J. Frère). Journal of Electromyography and Kinesiology 22 (2012) 145–152 Contents lists available at SciVerse ScienceDirect Journal of Electromyography and Kinesiology journal homepage: www.elsevier.com/locate/jelekin