Abstract
Purpose: Research has shown that observers can track 4 or 5 out of 10 identical moving items. Pylyshyn proposed a spatial indexing theory in which a limited set of 4 or 5 spatial indices can be assigned to objects in the visual field. Yantis proposed that the tracked objects are grouped together and attended to as a single deforming object. These accounts of multiple object tracking involve spatially indexing or attending to visible items. Is it necessary to actually see the objects in order to track them? Surprisingly, our results show that observers can accurately track a set of 5 out of 10 identical moving objects even when all 10 are removed for one third of a second. Method: 10 identical white circles (radius = 1 deg) appeared on a rectangular gray background (30 × 23 deg). A subset of 5 circles blinked off and on several times to identify them for tracking, then all 10 circles began to move in random directions at a constant rate of 6 deg/s. Circles changed direction following a collision with the display boundaries or with other circles. After two seconds of tracking, all circles vanished for a brief period, then reappeared with their positions updated as if they had continued to move according to the same rules during the blank interval. After 5 total seconds, all 10 circles stopped and observers ‘clicked’ on the tracked items. The duration of the blank interval was staircased to estimate the blank interval over which observers could accurately track all 5 balls on 70% of trials. Results: The average of 6 participants shows that observers can track all 5 out of 10 balls accurately 70% of the time when all 10 objects vanished for 370 ms (sd 200 ms). Conclusion: The present results show that it is possible to track 5 out of 10 objects even when those objects are not visible for one third of a second. Apparently, the processes subserving multielement tracking are able to extrapolate the future course of at least five independent objects.