Abstract
Previous studies on localization of stimuli in peripheral vision have predominantly reported underestimation of the distance of targets from fixation (foveal bias). However, a few studies have also found peripheral biases, and these studies have all employed manual responses. Here we show that peripheral biases in localization are also evident with verbal responses. Experiments were conducted using a Goldmann Perimeter with target dots presented at 10° intervals along the cardinal axes. Participants either marked perceived location of targets along a line on a sheet of paper that represented their visual field extent along each axis, or they made verbal magnitude estimates between 0 (point of fixation) and 100 (perceived edge of visual field). Both tasks showed peripheral biases when participants estimated target locations relative to their perceived visual field extent. However, the magnitude of responses as a function of eccentricity only deviated from linearity for those axes that are not restricted by an external visual boundary (i.e., the brow and the nose). The importance of the external visual boundary is supported by comparing scaling of localization estimates along the left axis (horizontal meridian) in binocular or monocular (right eye) viewing conditions. Scaling was linear for the monocular condition (where the left axis represents the nasal visual field) and deviated from linearity, with a significant peripheral bias, in the binocular viewing condition. This result cannot be accounted for by localization biases that are specific to the nasal visual field, because when verbal judgments were made relative to the edge of a physical aperture, all axes showed a linear scaling of responses, significant foveal biases, and no differences in scaling along the left axis between monocular and binocular conditions. We conclude that the type of boundaries used to define a space influences biases in peripheral localization that are independent of response type.