Abstract
A procedure for conducting a meaningful simulation of a man-machine system is presented and illustrated by various specific examples. The relationship of the various types of simulators to their use is outlined and desirable detailed characteristics are delineated. The tradeoffs between simulator complexity, realism and the interrelation of various feedback sensing cues (motion, visual or tactile) are discussed and the necessity of validating the simulation by use of a variable stability and variable control system aircraft is noted. It is shown that as the problem to be studied becomes more complicated or the questions asked of the simulation become more quantitative, the simulator characteristics must become more flight-like, since in the last analysis the best place to ask the question is when the pilot and the vehicle are immersed in the true environment (i.e., flight).