An investigation of fitts' law using a wide range of movement amplitudes.

The relationship between Fitts' Index of Difficulty (ID = log2 2A/W) and movement time was investigated for finger, wrist, and whole arm motions over a wide range of movement distances (0.25 to 30.5 cm). Results supported Fitts' original speculation that various limb segments may show different maximum information processing rates. Short-distance finger and wrist motions showed much higher rates (38 and 23 bits/sec, respectively) than longer-distance arm motions (10 bits/sec). Examination of motion trajectories qualitatively supported a descriptive model whereby a visually mediated discrete-correction control process is used, as proposed by Crossman and Goodeve (Note 1) and Keele (1968). However, evidence of severe nonlinearities in the measured human movement responses did not support the use of linear control models in explaining the empirical validity of Fitts' law in predicting human motor performance.