Abstract
The human oculomotor system makes saccades to bring objects of interest to the fovea. Losing foveal vision often leads to the adoption of a preferred retinal locus (PRL) for fixations, with saccades re-referenced to it. Factors underlying the development of a PRL are not well understood. We test the hypothesis that gaze control for saccades and fixation without a fovea is achieved through a minimum modification to the existing oculomotor plan. Using a gaze-contingent display, we asked normally sighted subjects to move a green ring, 6° in radius and centered at the fovea, to make contact with a small disc that appeared at a random location on the display in each trial. The subject had to maintain contact between the ring and the disc for 500 ms before the disc was dismissed and a new trial began. The inside of the ring was opaque to simulate a central scotoma. Four of the five subjects consistently used only a small region of the ring to make contact, even when the target disc was closer to another point on the ring. This development of a PRL was spontaneous and fast (<3 hrs, over a course of few days). Furthermore, the same PRL was retained several days after the experiment when the subjects were asked to "look at" a jumping target with a simulated central scotoma. In contrast, when the same subjects performed the task using a mouse to control the ring, they generally used the point on the ring closest to the target disc to make contact. Hence, while hand movements followed the shortest-distance strategy, eye movements opted for a fixed point and developed a PRL. It appears that the oculomotor system is highly constrained and can only adopt a constant offset to its existing motor plans for foveation.
Meeting abstract presented at VSS 2014