Improvement of Pointing Time by Predicting Targets in Pointing With a PC Mouse

Abstract

A method was proposed for the prediction of a target to which a user is to point with a mouse on the basis of the trajectory of the mouse cursor. An empirical study was carried out in order to evaluate the validity of the proposed prediction algorithm to reduce the pointing time with the prediction accuracy remaining high. The effects of the distance between the edges of two adjacent targets and the position of the indicated target on the prediction accuracy were investigated. Pointing with no prediction mode was also conducted. In the prediction method, the angle between the cursor movement vector and the vector that connected the current cursor position and the center of each target was calculated respectively at various sampling frequencies and intervals, and the minimum cumulative value was determined as the prediction target. The pointing time of the prediction method was shown to be less than that of the control condition for various combinations of the sampling interval and frequency. The trade-off between the pointing time and the prediction accuracy was also clarified. The prediction accuracy was found to be affected by the position of the target. The prediction accuracy at Positions 2 and 4, which have neighbors on both sides, was especially low. The distance between the edges of two adjacent targets affected the prediction accuracy, especially at Positions 2 and 4. Implications for application of the algorithm are given.