Abstract
Recent studies (Hafed & Clark, VR 2002; Engbert & Kliegl, VR 2002) have shown that microsaccades can be taken as an overt measurement of covert attention shifts during fixation. It is unknown, however, if microsaccades also occur during pursuit and, if so, whether these microsaccades are related to covert attention shifts. We carried out an experiment to investigate this issue. In our study, subjects were asked to maintain pursuit of a horizontally moving cross. At the onset of every trial, a square cue appeared on the left or right side of the display and remained visible for a variable period between 700 and 1100 msec in duration. The pursuit target appeared at the location of the cue and began to move horizontally as soon as the cue disappeared. The total duration for each trial after cue onset was 3.5 sec. At a random time after pursuit initiation, a square shape briefly appeared at a distance of 11 degrees either to the left or right of the pursuit target. Subjects were instructed to report as soon as possible whether it had the same color as the pursuit target. The pursuit target changed its color randomly during the trial. We used four different pursuit velocities (1.6, 3.4, 5.0, 7.2 deg/sec)in the experiments, blocked into sets of 100 trials. Our results show that microsaccades did indeed occur during ocular pursuit and that they had similar characteristics to those occurring during fixation. An analysis of the directions and latencies of microsaccades found that they were correlated with both the flash directions and the pursuit velocities. In particular, it was observed that microsaccade directions were biased in the direction of pursuit and that this bias increased with increases in pursuit velocities. This result is in agreement with the finding of Van Donkelaar & Drew (PBR 2002), who found that covert attention leads pursuit targets, with a lead amount that increases with the pursuit velocity.
This research was funded by grants from IRIS, Precarn, and NSERC.