Figure 2
From: The spatio-temporal architecture of everyday manual behavior
Synergies for canonical movements. (a) In the three canonical movements, PCA shows that a single PC explained 90% of the variance. (b) In Task 1, after PCA each time-window is characterized by a specific eigenvector. The within-group clustering divides these eigenvectors into two different groups. Hot colors show high correlation coefficients. (c) The mean correlation within clusters is significantly higher than across clusters (t-test). Asterisks denote significant differences (p < 0.05). (d) VRC analysis shows that for task 1 the optimal number of classes is 2 (grey line). For tasks 2–3, the clustering failed to separate two or more classes (red and blue line).
