Abstract
In a previous report (VSS 2002), we found no differences in tracking performance on MOT between “asynchronous” conditions (where objects changed colors smoothly and at the same time so that no two objects were ever the same color) and “synchronous” conditions (where objects changed colors in the same way but in synchrony so that objects were always the same color) see also Klieger et al. (VSS 2004). However, a different result might be obtained if the object property were more “spatial”, or if it segregated targets and nontargets as a group. In this study, we manipulated object size in a way that produced a perception of looming and receding. Subjects tracked four independently-moving target circles among four identical distractors for 5 or 10 seconds in four conditions where objects changed size (1) synchronously, or (2) asynchronously. In addition, we added two “segregated” conditions where targets expanded or shrank smoothly to a fixed size: (3) 138% of the nontarget size, or (4) 62% of the nontarget size. In conditions (3) and (4) targets and nontargets could easily be distinguished during the midpart of the trial by their different size (but not at the beginning or end when they were the same size). Since conditions (3) and (4) constituted only 16.5% of the trials presented at random, we reasoned that subjects might not notice the correlation of object size with targethood. As with the earlier color study, tracking performance in both synchronous and asynchronous conditions was the same for 5-second trials (84% correct) and 10-second trials (77% correct). However, segregating targets and nontargets by size yielded consistently better tracking (93% correct), which was the same for both 5-second and 10-second trial durations. This suggests that spatially-perceived properties may be used in tracking to help discriminate between targets and nontargets.