Abstract
Multiple object tracking measures the ability to covertly attend to multiple locations over time. Previously we demonstrated that participants tracking multiple targets often look at the center of the target group (Fehd & Seiffert, 2008), a strategy we label “center-looking”. This is intriguing because participants are not always looking directly at targets or saccading between them, a strategy we label “target-looking”. The current experiments investigated whether people engage in center-looking only when target-looking is too difficult. Participants tracked 4 targets moving randomly amidst 6 distractors while we measured the amount of time they viewed the targets or the center of the targets. To increase the demand for foveation and hence increase target-looking, Experiment 1 manipulated dot size from 0.3 to 0.06°, which pushed participants to their perceptual limits. On correct trials, however, no change in average viewing times was seen across dot sizes, suggesting that the need to foveate targets influences neither center-looking nor target-looking. To make target-looking more difficult, Experiment 2 increased the dot speeds from 3 to 24°/s. This manipulation examined whether center-looking reflects a reluctance to move gaze during tracking to avoid losing targets during saccades. While target-looking moderately decreased with increasing speeds (F(4,32)=2.49, p=.06), center-looking was consistent across speeds (F(4,32)[[lt]]1), indicating that center-looking was not a result of avoiding time-consuming eye movements. To make target-looking easier, Experiment 3 yoked targets to move in the same direction. Strengthening common target motion increased target-looking (F(3,27)=12.78 p[[lt]].05), but had no influence on center-looking (F(3,27)=1.60, p=.21). These results show that center-looking persists regardless of the ease with which targets can be foveated. Center-looking is not the default alternative to target-looking, but instead may reflect a different cognitive process. It seems that the time spent looking away from targets is not determined by the difficulty of looking at them.
This work was supported by the Vanderbilt Vision Research Center and NIH EY014984.