To estimate the decrement, the trials were first distributed into 12 time bins in the slow and the fast tapping tasks (the first 12 s). Decrement was then defined as the decrease of the completion ratios from the first to the twelfth time bin in a trial. We evaluated the decrement by estimating the effect of time bin on the completion ratios in a mixed model [completion ratios ∼ 1 + time bin + random(1 | subjectID)]. To compare the decrements in the slow and fast tapping tasks over identical time spans, we also calculated completion ratios for each 1-s time bin during the first 12 s of slow tapping.
