Aiming accuracy of the line of gaze and redesign of the gaze-pointing system

The current experiment tested the aiming accuracy of the eye for targets at nine different positions and of two different sizes. The left eye was tracked to record the line of gaze with an infrared eye-monitoring device. For each subject, horizontal shift, vertical shift, standard deviations of the X and Y coordinates, and 95th percentile of the horizontal and vertical shifts from the nine target centers were calculated to indicate the precision, accuracy, and possible design limits of the gaze points. Analysis showed that the target's position had a strong effect on the vertical shift, standard deviations of the Y coordinate, and 95th percentile of the vertical shift. The effect of target size was not significant on any of the dependent measures. The research findings have important implications for the redesign of the gaze-pointing device. Visual targets appearing below the horizontal line of sight should be larger than those above. based on the 95th percentile of the horizontal and vertical shifts, targets should be approximately 2.0 [degrees] wide and 2.4 [degrees] long above or coinciding with the horizontal line of gaze and 2.6 [degrees] wide and 3.9 [degrees] (cm) long below the horizontal line of gaze.