Fig. 3

Histogram of average outcomes of the Continuous (top) and Segmented (bottom) Coffee Task for both control approaches Pre and Post-training. Data for movements with an intact limb (i.e. no emulator) are shown for the Continuous Coffee Task only as segmental errors are not possible to quantify with intact limbs. Neither version of the Coffee Task demonstrated ceiling or floor effects. Both Coffee Task outcomes were skewed toward lower values