Visual Prediction of the Point of Coincidence of Two Moving Targets

Publisher Website
Google Scholar
Abstract
This study investigates accuracy of prediction in a task which requires the estimation of the point of coincidence of two pointers rotating at different rates, the trailing one closing on the leading one with a speed ratio of 3 : 2. Thirty undergraduates made predictions after watching filmed displays which varied both the viewing-span (VS) (the opportunity for viewing the pointers), and the prediction-span (PS) (the length of the prediction period). It was found that error in estimation increased linearly with increasing PS. For spans of 30, 60, 90 and 120° the mean error remained constant at about 18 per cent underestimation. Variable error was also found to be a linear function of the PS. Increasing VS had only a slight effect upon mean error, the longest VS alone producing significantly less error than the shorter spans. When the rate is held constant, predictions may be based upon the relative positions of the pointers at the end of the VS rather than upon rate perception alone. Generalization from the results is difficult. Comparison of these findings with those of other prediction studies indicates that the task itself is an important variable affecting prediction accuracy.
Keywords

This publication has 5 references indexed in Scilit: