Abstract
The perception of image focus can be strongly affected by prior adaptation to blurry or sharpened images or by simultaneous contrast from blurred or sharpened surrounds (Webster et al, 2002). We compared these interactions in the fovea and periphery to examine how they influence spatial judgments across the visual field. Stimuli were grayscale images of natural scenes filtered by increasing or decreasing the slope of the original amplitude spectrum to form a series of images ranging from moderately blurred (slope=−.5) to sharpened (slope=+.5). The displayed image level was varied in a staircase to find the point of best subjective focus, after adapting to the +.5 or −.5 image or in the presence of these images arranged as a spatially contiguous surround. The images subtended 4 deg and were presented at eccentricities ranging from 0 to 16 deg. Adaptation aftereffects were strong but did not show a consistent effect of retinal location, while induction effects were maximal for images in the near periphery (4 or 8 deg). The falloff at larger eccentricities may reflect a loss of sensitivity to the spectral differences in the stimuli rather than a loss of induction strength. Pre-adapt focus settings were similar in the fovea and periphery, suggesting that peripheral losses in contrast sensitivity are compensated to allow perceptual constancy for focus (e.g. Galvin et al, 1997). We examined whether this compensation reflected a reweighting of spatial sensitivity at different loci, by testing for the neutral stimulus level for the adaptation and induction (i.e. the level that does not bias the focus settings). These remained close to the physically focused slope, suggesting that sensitivity at different loci is calibrated for the losses in spatial resolution with eccentricity, and thus that variations in the intrinsic sensitivity may underlie the spatial constancy.