Abstract
Attention during pursuit is often found to be asymmetrically allocated, with more attention ahead of the pursuit target than behind it (e.g., van Donkelaar & Drew, 2002; Kahn et al., 2010). However, these studies used transiently appearing targets that attract attention. Lovejoy et al. (2009) controlled for luminance transients, and found the attention span to be tightly (±1 deg) and symmetrically distributed around the pursuit target. However, their stimulus, a linear array of 15 characters (0.6° spacing) may have artificially constrained attention due to crowding. We tested this hypothesis with the same character identification task but arranged the characters in a cross configuration and varied element spacing (0.6°, 2.0°, 4.0°) and set size (5, 9). All elements started as digital '8's, and after a random fixation period began moving as a unit. Shortly after movement onset, all elements changed briefly to either a '2' or '5' except for one probe element that changed to either an 'E' or a '3'. All elements then returned to '8's for the remainder of the trial, and the task was to identify the probe element. Eye movements were measured under all conditions (EyeLink 1000 @ 1000 Hz). We first replicated the Lovejoy et al. experiment, and found attention tightly centered on the pursuit target. However, stimuli with fewer elements and larger inter-element spacing produced wider scopes of attention, extending more than ±4.0° horizontally and vertically. We also found the same span of attention for stationary stimuli suggesting that attention does not operate differently during pursuit and fixation. We conclude that attention is flexibly allocated during pursuit, and limited by crowding and set size.
Meeting abstract presented at VSS 2014