Heading Perception of Tunnel-Vision Patients

Li Li

Schepens Eye Research Institute,
Harvard Medical School

Previous studies showed that normal-vision observers could accurately perceive heading from a radial random-dot flow pattern even when the central field of view was restricted to 50~100. While tunnel-vision patients have lost almost all their peripheral vision, their central visual field is restricted but functions normally. Here, we investigate whether tunnel-vision patients can perceive heading from optic flow. Two groups of retinitis pigmentosa (RP) patients (with ~50 and ~100 visual field respectively) and age-matched controls viewed displays simulating observer translation (2m/s) over a random-dot or a textured-mapped ground. Subjects adjusted a probe to indicate perceived heading. In the Full-field Viewing condition, the subject viewed the full display (1120H x 950V) with free fixation; in the Aperture Viewing condition, the subject viewed the display through an aperture with a fixation point at the center. The size of the aperture was either 50 or 100 depending on the size of the patient’s visual field. In the Full-field condition, two RP patients (one from each group) could accurately perceive heading and so could all age-matched controls. The other two RP patients (with ~100 visual field) displayed center bias. In the Aperture condition, RP patients displayed a much stronger center-bias than the age-matched controls. We conclude that with full display and free fixation, scanning eye movements allowed some RP patients to judge heading accurately. With restricted display and controlled fixation, patients had very poor heading performance. Results suggest that scanning eye movements are important for accurate heading perception with a small visual field.

The lecture will take place:

On Wednesday, April 5, 2000
at 1:00 pm