Purchase this article with an account.
Chuang Lyu, Xuemin Zhang; How long can you keep tracking?: a flying mission game. Journal of Vision 2014;14(10):349. doi: 10.1167/14.10.349.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Previous research had demonstrated that observers can track up to 4 or 5 items in multiple object tracking (MOT) task. Here we proposed a new question: how long can observers maintain tracking. We used an adapted MOT task. Several (3 levels: four items, five items, six items) different blue rectangles, which represented barriers, and a red square, which represented airplane, were presented on an initial background randomly without overlapping. When participants clicked the red square, the blue ones began a rectilinear motion respectively (initial velocity had 3 levels: 7 pixel/frame, 6 pixel/frame, 5 pixel/ frame. We refreshed a frame every 16 milliseconds.) with changing speed (by ±5% each frame and ranging from (initial velocity+2) pixel to (initial velocity-2) pixel). The blue targets passed through each other, but bounced back when approaching to the boundary of the moving background, which was larger than the initial background. The red square was controlled by moving mouse. The main task was to avoid crashing with blue barriers and boundaries of initial background. We provided more rewards every ten extend minutes. Significant interactive effect and main effect of the two factors were found through repeated ANOVA. Pairwise comparisons indicated that discrepancy between different levels of two factors were all significant. In addition, the difference between four items and five items was much larger than that between five and six items. Similarly, the difference between slow speed and middle speed was much larger than middle speed and high speed. Thus, our results show that tracking time for four items is much longer than five or six items. When training for tracking, tracking more than six items or more quickly than 7 pixels/ frame might be less helpful in improving performance. This task is useful and ecological in investigating multiple object tracking.
Meeting abstract presented at VSS 2014
This PDF is available to Subscribers Only