A Test of Fitts' Law with Moving Targets

Abstract

Subjects used either a position or a velocity control system to capture stationary and moving targets. Fitts' Index of Difficulty was found to be a good predictor of capture time for the rate system. However, for the position control system, the Fitts measure failed to predict accurately the capture time for moving targets. An alternate Index of Difficulty measure is proposed which explicitly incorporates a velocity factor and which predicts the overall pattern of capture times for both systems with greater accuracy than Fitts' Index of Difficulty.

Towards a Model for Predicting Intention in 3D Moving-Target Selection Tasks

“…This may be due to the fact that the spheres get closer to the user throughout each trial, eventually annulling the z -component of the target's distance, this corroborates Jagacinski's findings on 1D selection times on moving targets [10]. The fact that absolute horizontal sphere positions (P i,x ) did not affect user choices may suggest that users prepared their hands horizontally, while waiting for the target, or that it was more comfortable for them to reach for the right sphere first, followed by the left sphere, which seems reasonable considering that most participants were right handed.…”

“…Beyond size, distance and movement, this approach could be extended to consider other factors such as target semantics, if any, as well as user behaviors and gestures. The potential of using other "indices of difficulty," formulated specifically for moving-target selection [1,9,10], to predict user intention should also be explored. Finally, it should also be possible to refine decision trees in real time, to adapt the generated models to each user.…”

“…To do so, however, the influence of target distance (D) and width (W ) on such predictions must be evaluated; as opposed to distance only, the common factor in Christenfeld's item selection studies. More importantly, it is possible that the correlation between ID and perceived effort will decrease with the addition of target motion, since the correlation between ID and selection time is reduced in moving-target selection relative to static selection [10]. Regardless, we hypothesize that ID, or another function of target size and initial distance may be predictive of user intention in moving-target selection tasks.…”

Hitting a moving target: Perception and action in the timing of rapid interceptions

“…Although capture can be executed in the pursuit (panel A) and receding (panel C) configurations, they are ill suited for hitting. Jagacinski, Repperger, Ward, and Moran (1980) reported the results of an experiment involving the pure pursuit configuration ( Figure 1A). Participants manipulated a joystick (rotated it to the left or the right about a fixed axis) to move a viewing window from rest across a display screen in order to bring a moving target within the viewing window (capture task).…”

Improving the Performance of Input Interfaces Through Scaling and Human Motor Models

“…The Joystick and Joystick 3D relate the input position to the output velocity through scaling (velocity control). Although some studies demonstrate that position control has a better performance than velocity control for stationary target acquisition (Jagacinski, Repperger, Ward & Moran, 1980), joysticks are velocity-controlled because the cursor must remain still when they are at their central position. Figure 14 shows the parameter C that conditions each interface.…”