Introduction: Although individuality has been discussed in movement science for a long time, the phenomenon itself mostly has been considered as a side effect of extensive training in top athletes. Structurally, individuality has been neglected thru the dominance of group studies and average oriented designs without being able to draw specific consequences on training or therapy. Meanwhile individuality of single types of movement has been shown (Schöllhorn et al., 2002) and this has led to the rethinking of the traditional training approaches (Schöllhorn, 2000). Our every day experience tells us that individuality is observable in the same person for different types of movements. In a first approach to the problem individual characteristics should be recognized in several types of movements in order to transfer the individual characteristics across these movement patterns. The aim of this study was to identify intra-individual movement patterns of decathletes in shot put, discus, and javelin. Methods: Within national qualification for European Athletics Junior Championships 7 right handed decathletes (mean age: 18.8 years; sd: 0.4) were recorded on two digital video cameras (200 fps; 1280 x 1024 px) located perpendicular to each other, one facing towards the flight direction of the throwing object. On the basis of a 13 segment body model the final throwing phases of 19 shot put, 19 discus, and 19 javelin trials were described by means of angular- and angular-velocity-timecourses of the main joints, except the variables of the throwing arm. Time and amplitude normalized data formed the input vectors for a classification procedure based on two Support Vector Machines (SVM) which were trained by cross-validation procedure, one for the identification of throwing disciplines and one for the identification of persons. Results: The classification of all 57 trials revealed a 100% recognition rate for the disciplines with the first SVM. The second SVM led to a recognition rate of 98.5% for the athletes. Discussion: The recognition rates provide strong evidence for the sensitivity of the applied movement recognition approach and for the dominance of individuality across disciplines. In expansion to the identification of individual movement characteristics in single movement trials the present investigation allows the transfer of individual information of one movement to another. The identification of transdisciplinary individual movement characteristics support the demands for more economic training by transdisciplinary training. In consequence, the results point to rethink group oriented training and suggest a more enhanced focus on individual characteristics in training even.
© Copyright 2014 19th Annual Congress of the European College of Sport Science (ECSS), Amsterdam, 2. - 5. July 2014. Veröffentlicht von VU University Amsterdam. Alle Rechte vorbehalten.