PubMed1992
The timing of repetitive alternating 80 degrees flexion-extension movements of the right wrist was studied in 42 patients with Parkinson's disease, tested while not receiving dompaminergic medication and 20 age- and sex-matched controls. Five rates of movement (0.5 Hz, 1 Hz, 1.5 Hz, 2 Hz, 2.5 Hz) were examined. The interval between two successive flexion movements, as measured from the electromyography (EMG) records was taken as the unit of analysis or inter-response interval (IRI). At 0.5 Hz, 1 Hz and 1.5 Hz there were no differences between groups in mean IRIs. At higher rates of movement (2 Hz and 2.5 Hz), however, the controls were significantly more accurate in timing of repetitive movements than the patients. At all five frequencies, the patients with moderate or severe Parkinson's disease were less accurate in timing of repetitive movements than those with mild disease, although the differences were not significant. In nine patients tested in the 'on' and 'off' medication states, administration of 250 mg of levodopa/carbidopa resulted in significantly more accurate timing of repetitive movements. Wing and Kristofferson's (1973b) two-process model of repetitive movements was applied to the IRI data. At various rates of movement, the primary prediction of the model that lag 1 autocorrelations should be in the 0 to -0.5 range was violated in 40-70% of the patient and controls. For those subjects who had lag 1 autocorrelations in the expected range, IRI variability was to break down into a timekeeper and a motor delay variance. At all frequencies, the patients had significantly higher variance for IRI, timekeeper and motor delay than the controls. Although not significant, more severe Parkinson's disease was associated with greater IRI, timekeeper and motor delay variance, while administration of levodopa resulted in reduction of the three types of variance. The validity of the Wing and Kristofferson model for the analysis of this type of movement is discussed.