Author(s): MURAKAMI, T., KUSAFUKA, A., OKEGAWA, T., SAKAKIBARA, Y., ARAKAWA, H., HAKARIYA, N., YAMASAKI, D., AOKI, R., KANEKO, N., SASAKI, A., NAKAZAWA, K., Institution: THE UNVERSITY OF TOKYO, Country: JAPAN, Abstract-ID: 1881

INTRODUCTION:
To execute movement, the central nervous system must coordinate hundreds of muscles. Such coordination is continuously reorganized across the lifespan particularly during development and expertise, yet the mechanisms underlying this reorganization remain unclear. Throwing behavior emerges in early childhood and becomes highly refined in sports such as baseball. Overarm throwing, therefore, serves as an ideal model for investigating how humans develop and refine whole-body coordination. This study aims to elucidate the characteristics of the reorganization of muscle coordination during overarm throwing across development and expertise.
METHODS:
11 preschoolers (PS; 3–6 years) and 9 primary school children (SA; 7–9 years), 6 adults with no baseball experience (NE; 23–29 years), and 8 experienced baseball players (EX; 19–25 years) participated in this study. Children (PS and SA) threw balls toward a net placed 2 meters away, while adults (NE and EX) threw toward a net placed 18.44 meters away. Electromyographic (EMG) were recorded from the 16 muscles across the dominant upper limb and bilateral trunk and lower limbs. Development-related differences were examined by comparing PS and SA, and expertise-related differences by comparing NE and EX. To evaluate muscle coordination, muscle synergies were extracted from each participants using non-negative matrix factorization. To identify representative muscle synergies of each group, synergies from each group participants were clustered using k-means. Cluster similarity was evaluated using cosine similarity, with values > 0.9 considered common synergies. For group-specific synergies, we examined whether muscle synergies underwent merging from PS to SA or from NE to EX.
RESULTS:
The number of muscle synergies did not differ significantly between PS (mean ± SD: 6.0 ± 0.4) and SA (mean ± SD: 6.2 ± 1.7)(independent t-test, p > 0.05), whereas NE exhibited a significantly higher number of synergies (median [IQR]: 5 [5–6]) than EX (median [IQR]: 4 [4–4.5]) (Mann–Whitney U test, p < 0.05). K-means clustering identified 8 clusters in PS and SA, 10 in NE and 5 in EX. In the development-related comparison, 4 clusters were common between the two groups, while 4 clusters were specific to PS and SA respectively. 2 merging patterns were identified in both the development and expertise-related comparisons.
CONCLUSION:
In both development and expertise-related comparisons, synergies were reorganized, as reflected by the merging. However, changes in the number of synergies differed across timescales. In the development-related comparison, the number of muscle synergies did not differ. In the expertise-related comparison, the number of muscle synergies decreased, and the number of clusters converged toward the individual synergy number, suggesting a compression of the coordination repertoire toward a more consistent structure. Together, these findings suggest that the characteristics of synergy reorganization differed between development and expertise.