Dynamic Cancellation of Perceived Rotation from the Venetian Blind Effect.

Joshua Dobias,Wm Stine

doi:10.3390/vision3020014

Abstract

Geometric differences between the images seen by each eye enable the perception of depth. Additionally, depth is produced in the absence of geometric disparities with binocular disparities in either the average luminance or contrast, which is known as the Venetian blind effect. The temporal dynamics of the Venetian blind effect are much slower (1.3 Hz) than those for geometric binocular disparities (4–5 Hz). Sine-wave modulations of luminance and contrast disparity, however, can be discriminated from square-wave modulations at 1 Hz, which suggests a non-linearity. To measure this non-linearity, a luminance or contrast disparity modulation was presented at a particular frequency and paired with a geometric disparity modulation that cancelled the perceived rotation induced by the luminance or contrast modulation. Phases between the luminance or contrast and the geometric modulation varied in 50 ms increments from −200 and 200 ms. When phases were aligned, observers perceived little or no rotation. When not aligned, a perceived rotation was induced by a contrast or luminance disparity that was then cancelled by the geometric disparity. This causes the perception of a slight jump. The Generalized Difference Model, which is linear in time, predicted a minimal probability in cases when luminance or contrast disparities occurred before the geometric disparities due to the slower dynamics of the Venetian blind effect. The Gated Generalized Difference Model, which is non-linear in time, predicted a minimal probability for offsets of 0 ms. Results followed the Gated model, which further suggests a non-linearity in time for the Venetian blind effect.

Highlights

Slight geometric differences between the images on each retina are often used to perceive depth within the environment
Surfaces can appear to rotate around a vertical axis when the image shown to one eye has lower average luminance or a Michelson contrast (Michelson contrast is defined as the difference between the maximum and minimum luminance values divided by the sum of the maximum and minimum luminance values, or (Lmax − Lmin )/(Lmax + Lmin ) [1].) [2,3,4,5,6,7,8,9] than a geometrically-identical image shown to the other eye (Venetian blind effect)
Since perceived depth from a geometric disparity is visible at higher frequencies than depth from a contrast disparity, it is likely that both result from separate neural mechanisms at some level

Summary

Introduction

Slight geometric differences between the images on each retina are often used to perceive depth within the environment. Surfaces can appear to rotate around a vertical axis when the image shown to one eye has lower average luminance or a Michelson contrast (Michelson contrast is defined as the difference between the maximum and minimum luminance values divided by the sum of the maximum and minimum luminance values, or (Lmax − Lmin )/(Lmax + Lmin ) [1].) [2,3,4,5,6,7,8,9] than a geometrically-identical image shown to the other eye (Venetian blind effect). The Venetian blind effect has been explained as the creation of a geometric retinal disparity that results from the imperfect optics of the eye [2,3,4]. Cibis and Haber [4] state that a decrease in retinal illuminance causes a shift in the location at which the retinal illuminance of, for example, the bars of a 100%

Methods

Results

Conclusion