Abstract
The ability to locate objects in space not only depends on accurate depth perception but also the ability to localize objects across the visual field. The present study investigated the accuracy of perceived eccentricity for target dots briefly flashed on a computer screen. Thirty-six normal-vision participants completed one of three conditions. The first two groups completed the experiment with a visible boundary created by a circular aperture, 30deg in radius, over the computer screen. Participants in one group made magnitude estimates of the targets' distances from fixation to the edge of the aperture while participants in another group made magnitude estimates relative to the edge of their visual fields. A third group of participants wore welder's goggles, eliminating the visibility of the aperture's edge, and made magnitude estimates of the targets' distances relative to the edge of their visual fields. In all conditions, seven equally spaced eccentricities were randomly tested along each of the four radii from fixation. For each participant, power functions, J = λ(Sγ), were then fit to the estimated magnitudes as a function of eccentricity for the horizontal and vertical meridians. 3(Experimental Group) × 2(Meridian) ANOVA's were calculated on the constants and exponents. For the constants (λ), no main effects or interaction were found. However for the exponents (γ), a main effect of Group was found with exponents increasing when participants judged eccentricities relative to the edge of their visual fields and no visual boundary was available (1.2 vs. 0.98 and 1.0). No effect of Meridian or Meridian × Group interaction was observed. This suggests an expansion of the estimated locations with increasing eccentricity when no visual boundary was available that is independent of the instructions given. The results also contrast with findings of expansion in perceived eccentricity by individuals with actual peripheral field loss.