Micro-pursuit and strategies of allocating attention during visual tasks with moving targets

Eye movements are often studied in the context of demanding visual tasks in which goals are achieved on the basis of the visual information gathered from sequence of fixations (Hayhoe, 2017). Much less attention has been devoted to how smooth pursuit is used during visual or visuomotor tasks. Prior studies investigated the link between smooth pursuit eye movements and manual interception. Results showed that pursuing a moving target was beneficial to estimate the time of the interception and larger tracking errors resulted in larger interception errors (Brenner and Smeets, 2011; Fooken et. al., 2016). The present study examined eye movement strategies and perceptual performance in a demanding visual task involving moving targets (discs), modeled after the judgments at traffic circles. The task required judgments of relative motions of two targets heading to a common meeting point, specifically, how much faster is one target relative to the other. The questions intended to be resolved are, what is the eye strategy used spontaneously during the course of the motion, and is the chosen eye strategy indeed the better eye strategy to use while judging the relative motions? Two motion discrimination tasks were assigned by blocks to all four subjects: (1) the which-first task requiring judging which disc would arrive the meeting point first and (2) the more challenging collision task requiring judging which disc would arrive first and whether the two discs would collide at the meeting point. Collision was defined as any overlapping of two discs. Subjects performed both tasks under a free-viewing condition, in which they received no instructions about where to look or whether to pursue either moving disc (Experiment 1). Perceptual results show that people can precisely judge the relative motions, with Weber fractions ranging from 3-6%. Three-dimensional representations were created showing the distributions of both averaged horizontal eye position (x-axis) and horizontal eye velocity (y-axis) at different epochs of time. This novel analysis allows comparisons between eye velocities and positions over time so that the sources of position change (saccades or pursuit) can be distinguished. Eye movement results show that the preferred eye strategy was to fixate near the meeting point and pursue the pair of the discs after the decision was made, anticipating the motion of the paired discs on the last lag of the display. The influences of the standard disc velocity and the comparison disc velocity on the averaged eye velocity at different representative times suggest that fixate was achieved at least in part by dividing attention between discs moving in opposite directions during the time the decision was being made. To better capture the effectiveness of the preferred eye strategy. All subjects performed the same tasks with the strategies determined by instructions (Experiment 2): (1) fixating near the meeting point or (2) switching between discs while making decisions. Performance of all four subjects showed precise discrimination for both tasks (which-first and collision) and both eye strategies (fixate and switch), with Weber fractions ranging from 2-5%. Better discrimination was obtained with the fixate strategy (2.94%) than the switch eye strategy (4.14%). Results were consistent in the collision task, indicating the fixate strategy led to better discrimination when judging the relative motion of the two discs heading toward the same meeting point.M.S.Includes bibliographical reference