Abstract
Introduction. Eye movement latency to targets can be shortened by advanced preparation of saccadic programs. Specifically, advanced saccade preparation can be enhanced through training to attend to a specific location in space (Paré & Munoz, 1996). However, people often know the area where a target will appear rather than its precise location. Here, we investigated how saccades were affected by training to attend to a circular area within which a target appeared at random locations. Additionally, we looked at how training to attend to an area of one size influenced saccades to targets presented in a larger circular area.
Methods. Each trial began with a central fixation point, followed by simultaneously presenting a circular area-cue (6° or 10° diameter), for 400ms. These disappeared, and following a gap period (170ms or 220ms), the target was flashed for 68ms. Participants were required to quickly and accurately saccade to the target once it appeared. Saccade reaction time and position were recorded. To prevent anticipatory saccades, catch trials were included in pre- and post-training sessions where some targets were presented outside the area-cues. During the training sessions, the target was always presented within a 6° area at random locations.
Results. Post-training goal-directed saccades were mostly anticipatory. Participants each developed a preferred region inside the trained area, where post-training anticipatory saccades were directed. This preferred region scaled with cued-area size, i.e. post-training distributions of saccade-end points greatly overlapped once normalized for cued-area position and size. From the preferred region subjects often generated visually driven corrective saccades to the target inside the area cue. Over all, there was no speed-accuracy trade off.
Discussion. Present findings show that oculomotor preparations extend to areas, not just to a single target location. Compared to pre-training, the learned strategy assures that targets are acquired more quickly without loss of accuracy.
This work was funded by the Canadian Institutes for Health Research