We then investigated the relationship between various abnormal performance metrics in the tapping tasks and the clinical severity of motor impairment in patients as indexed by the hemi-body bradykinesia and rigidity scores from the MDS-UPDRS III. We tested the correlations between clinical scores and four performance parameters: EMG slope, decrement for slow tapping task and the mean tapping rate, and decrement for fast tapping task. We expected these measures to be similar to the motor impairment as assessed in the qualitative clinical examination. Of all the parameters, the correlation between EMG slope at tapping onset in slow tapping and the hemibody bradykinesia and rigidity scores was marginally significant (ρ = −0.55, P = 0.056, FDR corrected). This supports the assumption of the clinical validity of this parameter. There were no correlations between the clinical score and the decrement computed as the difference of the completion ratios between the 1st and 12th time bins in either the slow or the fast tapping task.
