Psychological Review VOLUME 86 NUMBER S SEPTEMBER 1979 Motor-Output Variability: A Theory for the Accuracy of Rapid Motor Acts Richard A. Schmidt, Howard Zelaznik, Brian Hawkins, James S. Frank, and John T. Quinn, Jr. Department of Physical Education University of Southern California Theoretical accounts of the speed-accuracy trade-off in rapid movement have usually focused on within-movement error detection and correction, and have consistently ignored the possibility that motor-output variability might be pre- dictably related to movement amplitude and movement time. This article pre- sents a theory of motor-output variability that accounts for the relationship among the movement amplitude, movement time, the mass to be moved, and the resulting movement error. Predictions are derived from physical principles; empirical evidence supporting the principles is presented for three movement paradigms (single-aiming responses, reciprocal movements, and rapid-timing tasks); and the theory and data are discussed in terms of past theoretical accounts and future research directions. Examining the current level of understand- ing about the production and control of motor responses, many would no doubt be tempted to say that we have not come very far since the early writings of Woodworth (1899) and Hollingworth (1909). These writers were concerned with the basic laws of limb movements (analogous, perhaps to the basic laws of motion that were the corner- stone of physics) that denned the relation- ship between the simplest aspects of motor This investigation was supported by Biomedical Research Support Grant 5-S07-RR07012 from the Division of Research Resources, Bureau of Health Professions Education, and Manpower Training, National Institutes of Health. R. A. Schmidt is also affiliated with the Depart- ment of Psychology; H. Zelaznik is now at the Department of Physical Education, Purdue Uni- versity; and J. S. Frank is now at the Depart- ment of Physical Education, McMaster University, Hamilton, Canada. Requests for reprints should be sent to R. A. Schmidt, Department of Physical Education, 107 PED, University of Southern California, Los An- geles, California 90007. acts: the relations among a movement's speed, the movement distance, and the ac- curacy with which the movement was exe- cuted. A major step forward came in 1954 when Fitts presented a now-famous relation- ship among the movement time (MT), the movement amplitude (A), and the required accuracy (target width, W) in a principle that has since become known as Fitts's law. In the so-called Fitts paradigm, the subject taps alternately between two targets, each of whose width is W, with their centers separated by a distance A, attempting to produce as many taps as possible in a short period of time (e.g., IS sec). Fitts (1954) showed that the relation between A, W, and MT is given by MT = a + b log z (2A/W), (1) where a and b are empirically determined constants. Essentially, this equation states that MT is constant for any proportional combination of A and W, and that increasing the movement distance or requiring greater accuracy increases MT. The law thus de- Copyright 1979 by the American Psychological Association, Inc. 0033-29SX/79/860S-041S$00.7S 415