Fig. 3

Distribution of generated muscle activations (a1 to a3) and prosthesis forces (b2 to b3) for each target level. For the EMG task (a1, light grey) and the Prosthesis task without (a2, b2, dark grey) and with (a3, b3, black) feedback, the distribution of the (per-subject) median of muscle activations (a) / forces (b) is visualized using boxplots, depicting the overall median (circle), interquartile range (box), maximal/minimal values (lines) and outliers (pluses). Black continuous lines denote the target levels, black dashed lines the level of saturation. For each task, significant differences between the muscle activations / forces generated while aiming at the neighboring target levels are marked with an asterisk (p < 0.05, Bonferroni-Holmes corrected). For levels 4 to 6, the median percentage of saturations per subject is given