Abstract
Purpose: When the velocity of a moving retinal image exceeds about 3deg/s discrimination declines. Recently we showed that a single moving letter is discriminated better than a moving letter surrounded by other letters. This experiment was to determine what role retinal smear plays in reducing discrimination with the surrounding letters. Methods: 10 subjects participated. Subjects sat 4m from a black target plane. The projected target was a single white Landolt C or a Landolt C in the center of 8 Cs. In the latter case the 9 Cs were arranged in 3 columns. In each trial subjects fixated a dim spot that was extinguished just prior to target presentation. Subjects identified the orientation (up, down, left, right) of the (central) Landolt C as it moved across fixation. The target was revealed by opening a shutter. Thus, when the Landolt C gap was horizontal, the gap was not smeared either at the beginning or end of the presentation. The orientation of the gap, target velocity (1–5deg/s), direction of target motion (left, right), and target appearance (single (SI), surrounded (SU)) were all randomized. Targets were presented for 113ms to prevent eye motion. Each subject completed 144 trials. Results: At all velocities the mean number of targets whose orientation was correctly identified was greater in the SI condition. The mean incorrect (out of 36) was 14±5 SU for horizontal gaps, 20±6 SU for vertical gaps, 4±3 SI for horizontal gaps, and 16±5 SI for vertical gaps. Discussion: Discrimination was better for horizontal gaps than for vertical gaps in SI and SU because the horizontal gaps could not be smeared at either the beginning or end of target motion. Thus, retinal smear reduced discrimination for vertical gaps. Performance for horizontal and vertical gaps was more similar for SU than SI, suggesting that poorer letter discrimination for SU is at least partially the result of retinal smear and is not explained entirely by a shift in attention away from the central letter.
AFOSR Grant #F49620-02-1-0050