Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

How do saccades aid visual pattern recognition?

Open Access Open Access

Abstract

It has been shown that saccades play an important role in visual tasks, but they do not simply improve resolution of the display. To understand their role, I studied not only the way in which texture perception benefits from saccades, but also the patterns of saccades that are most beneficial. The displays (55) consisted of 1515 randomly oriented T's and L's. The T's formed a random polygonal shape (target) in the central 24 of the display. Eye movements were recorded while subjects estimated both the size and the shape of the target and were under instructions to (1) SCAN the display with saccades or (2) maintain the line of sight in the center without saccades (STAY). The results are as follows: (1) SCAN performance improved with increasing display duration; (2) STAY performance did not improve with increasing display duration longer than 2 s; (3) scanning improved performance for 5 s durations; (4) scanning improved the ability to detect texture patterns; (5) retinal image transients, created by flickering (2 Hz) or jumping (0.5 Hz, 15 ft. [4.57 m] peak-to-peak amplitude) the display, improved performance somewhat for size discrimination but not for shape discrimination; (6) boundary scanning was more beneficial for discrimination than other scanning patterns.

© 1990 Optical Society of America

PDF Article
More Like This
Phase-extraction pattern recognition

Tuvia Kotzer, Joseph Rosen, and Joseph Shamir
ThTT5 OSA Annual Meeting (FIO) 1990

Hypothesis testing in the planning of saccadic eye movements

Greg J. Zelinsky and Heinrich H. Bülthoff
MQQ10 OSA Annual Meeting (FIO) 1990

N-dimensional correlation filters for pattern recognition

Gregory Gheen
ThTT2 OSA Annual Meeting (FIO) 1990

Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.