Abstract :
The paper discusses experimental findings which aid an understanding of the basic limitations of a human operator as an element in a servo system. It is shown how experiments on threshold effects can guide the choice of both display magnification and gear ratio. At least 2 sec appears to be required to develop the maximum perceptual acuity, but a time greater than this is usually needed in accurately setting a control for full confidence in accuracy to develop. The mechanisms of muscular movements are discussed. Reciprocatingrnovements can be performed at higher rates than circular movements, although the latter are less fatiguing. Measurements of the mechanical efficiency of man may be worth applying under conditions of heavy work. The conditions of tracking are shown to resemble multiple-choice reaction-time experiments, with a lag between stimulus and response that is strictly indeterminable, but which is generally of the order of 0.3 sec. Discontinuity in the operator is demonstrated by the inability to halt a reacting movement within 0.2 sec of its start. In practical design, the advantages are stressed of incorporating the maximum possible display magnification; of abolishing any striction in the control, possibly by using two hands; of using high rates of turning; and of introducing as much inertia as the operator can manage. The paper concludes with a note on the acquisition of skill in tracking, particularly stressing the need for simple training aids which isolate the individual stimulus-response elements, and which allow accurate scoring and control of difficulty in the task facing the operator.