Abstract
Everyday activities such as copying, drawing, and imitative gestures require allocentric representations of space for successful movement completion. Notably, the top-down nature of allocentric spatial representations is thought to render motor output via a slow and offline mode of cognitive control mediated via visuoperceptual networks. The present investigation sought to test this hypothesis by providing detailed trajectory analyses of allocentric and target-directed reaching tasks performed with and without concomitant limb vision. Allocentric tasks required reaches to a location defined by the distance between a target and reference stimulus, whereas target-directed tasks required reaches to a target's veridical location. To examine the extent to which tasks were controlled via feedback-based trajectory amendments (i.e., online) or central planning mechanisms (i.e., offline), we computed the proportion of variance explained (i.e., R2) by the spatial position of the limb at 75% of movement time relative to each response's ultimate movement endpoint for distance and direction axes. Results showed that target-directed limb visible trials produced smaller R2 values and decreased endpoint variability compared to their limb occluded counterparts. In turn, the latter trial-type exhibited R2 values and endpoint variability commensurate with allocentric limb visible and occluded trials (which did not differ). Accordingly, we propose that the presence of limb vision in a target-directed task affords an online mode of control supported via 'fast' visuomotor networks. In contrast, the absence of limb vision or presence of allocentrically defined endpoints is proposed to render a primarily slow and offline mode of cognitive control mediated via visuoperceptual networks.Acknowledgments: Natural Science and Engineering Council of Canada