Figure 4.
Characterization of C-timing variability when W was fixed across the trials of a session. Variability of C timing was quantified by the variance of time interval between the Cmax's of every synergy pair. A,B, In elderlies, score gain correlated significantly with the initial (session 1) time-interval variability of 2 synergy pairs (B) involving 3 different W's (A). C, D, In elderlies, score gain also correlated significantly with the change (from session 1 to 3) of time-interval variability of 1 synergy pair (D) involving 2 W's (C). Again, note consistency of the W's from session 1 to 3. Interestingly, the pair of W's corresponding to the C's in panel C included activation components in the same set of elbow/wrist flexors we noted earlier (FlexCarUln, PronTer, Biceps) in Fig. 3(C) as well as medial deltoid (DeltM). E, F, In young adults, time-interval variability correlated significantly, and negatively, only with the change of ball release speed (F) but not score gain. Three significant correlations involving 4 W's (E) were identified.