| The sense of location is a fundamental aspect of the perception of objects. Spatially extended objects can be localized consistently and accurately in both perceptual and saccadic eye movement tasks. What determines the default reference position used for localization? In a series of studies, perceptual and saccadic localization was compared under different task conditions. In contrast to prior localization studies, target objects included non-convex 2D shapes, with centers-of-gravity lying outside the boundary of the shape and not coinciding with any prominent perceptual features, as well as computer-generated perspective images of 3D objects. For single saccades made to look at 2D targets as a whole, as well as in the selection of a preferred fixation locus, saccades landed near the COG, with local features, such as part boundaries, also playing a role in localization performance. Saccades made in a more natural sequential scanning task involving multiple objects and longer trial durations landed closer and more reliably at the COG and did not show effects of local features, suggesting that scanning may provide a better method to characterize the default landing position for saccades to extended targets. Perceptual localization tasks showed configural effects that did not influence eye-movements, indicating that the representation of object location for eye-movements may be less elaborate than for perception, depending primarily on the physical characteristics of the target itself, rather than the stimulus configuration as a whole. Unlike the perceptual system, eye-movements may be immune to reference frame effects, just as the motor system is immune to other visual illusions such as that of size. Results from the simulated 3D targets indicated that default saccadic landing positions are not based solely on primitive visual representations that replicate the configuration of the retinal image, but may instead have access to depth-scaled visual representations, of the sort known to be present as early as VI. The most surprising finding was observed shifts in saccadic landing position toward the 3D COG of the shape. Saccadic positions coincident with the center of the perceived 3D object may be critical for hand-eye coordination in object manipulation. |