Abstract
Introduction: Previous studies argue that saccade adaptation, traditionally induced by intrasaccadic step (ISS), remaps visual perception. Here, we ask whether saccadic vector alteration is cause or consequence of the perceptual remapping. For this reason, we applied ISS on a landmark near an unmoved saccadic target, and then tested subjects’ localization performance and their saccadic gains towards the landmark, in two separate experiments. Methods: In Experiment 1, six subjects made delayed saccades (cued by offset of fixation point) to a target with a nearby landmark. Subjects used a mouse cursor to localize the post-saccadic position of the landmark, which in some trials disappeared on saccade onset (Landmark-Off) and in the others stayed on until the appearance of the cursor (Landmark-On). After a baseline period, the landmark underwent ISS. Experiment 2 was the same as Experiment 1, except subjects localized landmarks via saccades (with screen blanking on saccade onset to eliminate any feedback). When the saccadic goal was the target, the landmark stayed on for 400ms after saccade onset. Again, after a baseline period the landmark underwent an ISS. Results: In Experiment 1, the landmark was significantly mislocalized in both Landmark-On and, more importantly, Landmark-Off trials—despite no corresponding change in saccade gain to the target. In Experiment 2, saccades to the target were also unchanged, but saccades to the landmark were adapted, despite the lack of visual feedback for saccades to the landmark. Discussion: We conclude that the visual system includes non-target (landmark) error signals between pre-saccadic and post-saccadic visual information to maintain its spatial accuracy. These findings also suggest that saccade adaptation, induced by ISS, is not just a motor recalibration based on saccadic landing error: Part of the saccadic gain change is derived by a prior change in visual perception.
Meeting abstract presented at VSS 2015