Replication as Pedagogy: The Social and Cultural Dimensions of Transmitting Dance Knowledge

Dance is an important intangible cultural heritage, the means by which cultural identity, artistic traditions, and embodied knowledge are transmitted from one generation to another. Movement replication has traditionally been a key technique for teaching dance in terms of which movement skills can be acquired through observation, imitation, and practice. However, the objective evaluation of the effectiveness of movement replication while ensuring cultural authenticity is still a difficult task. In this study, the authors propose the Spatio-Temporal Graph Convolutional Network integrated with Light Gradient Boosting Machine (ST-GCN-LGBM) to provide support for replication-based dance pedagogy and to assess skill acquisition based on modeling of complex movements. The effectiveness of the proposed model was tested by using the ImperialDance dataset including 3D motion data of beginners, intermediates, and experts who performed repeated choreographies. The results of experiments showed that the accuracy of the proposed model reached 95.4%, and precision, recall, and F1-score was 0.95, exceeding the performance indicators of the CNN, LSTM, and standalone STGCN models. Moreover, the proposed framework received a movement similarity score of 0.92, with relatively low prediction errors (MAE = 0.015, RMSE = 0.018, and MAPE = 1.6%). Thus, movement patterns were successfully replicated with the help of the proposed framework, with a high degree of movement consistency achieved by learners, especially beginners. In addition, the statistical verification demonstrated the existence of significant differences between skill levels (p