'Clock' and 'motor' components in absolute coordination of rhythmic movements.

Human subjects swung, through motions at the two wrists, hand-held pendulums of variable mass and length. Within a pair, the two pendulums could be of the same or different magnitude. The subjects were required to produce a coordinated state in which the two rhythmic units oscillated at a single common period. Fifty-four conditions of absolute coordination, that is, 54 different pairs of wrist-pendulum systems, were investigated for each of three subjects in the course of six sessions. Each condition of absolute coordination was conducted in the out of phase mode and at the single most comfortable period. The period variances of the right and left systems in the 162 instances of absolute coordination were analysed according to a method that assumes that a timekeeper function and a motor implementation function contribute independently to the variance in the periodic timing of a rhythmic movement. The major findings were that in absolute coordination: (a) a system's 'motor' variance, but not its 'clock' variance, depended on the deviation of the period of absolute coordination tau from the system's characteristic period; (b) right and left 'clock' variances were related and (c) neither the 'motor' variances nor the 'clock' variances were affected by deviations in the mean phase relation from 180 degrees. The results were discussed in terms of their implications for interpreting von Holst's notions of maintenance tendency and magnetic effect and, more generally, the neural and dynamical basis of absolute coordination.