Links between vision and somatosensation. Vision can improve the felt position of the unseen hand.

During reaching movements, sensory signals must be transformed into appropriate motor commands. Anatomical, electrophysiological, and neuropsychological evidence suggest that there is no single, supramodal map of space that is used to guide reaching. Instead, movements appear to be planned and controlled within multiple coordinate systems, each one attached to a different body part. Recent neuropsychological investigations demonstrating that somatosensory impairments can be ameliorated by visual cues, and visual impairments by proprioceptive cues, have been interpreted as evidence that arm-centered representations may exist in humans. A critical difference between the findings obtained in the monkey and in humans, however, is that in the latter case, vision of the limb appears be critical for visual somatosensory binding. Here, we report a case study of a patient (C.T.) recovering from unilateral somatosensory impairment, including tactile extinction, who executed reaches toward visually defined or proprioceptively defined locations. We demonstrate that when the target location of a reach was defined proprioceptively, by passively positioning our patient's impaired hand beneath the table surface, vision of the workspace immediately adjacent to the unseen hand dramatically increased the endpoint accuracy of her reaching movements, even though such cues could not possibly signal the position of the target directly.