Abstract
Light enters the brain via the retina, which transmits a retinotopic map of visual space to higher centers. Because the eyes move constantly, an object in a given spatial location can appear at many retinal locations. The brain must establish a spatially accurate map by transforming this retinotopic map into a stable map of space independent of eye/head position. There are two classical solutions to this problem: Helmholtz postulated that the brain utilizes corollary discharge arising from the motor commands for eye and head movements to transform the retinotopic map. Sherrington postulated that the brain measures eye and head position, and uses this to calculate the location of the object in space. Poletti et al. suggested both were used: corollary discharge soon after the appearance of a stimulus in the environment, which is replaced by a proprioceptive mechanism for long-term visuospatial memory. There is a proprioceptive representation of eye position in Area 3a of monkey somatosensory cortex (Wang et al). We asked if the cortical representation of proprioception were necessary for establishing spatially accurate visual memory. We trained C57Bl/6 mice to use the visual environment to learn the location of a hidden platform in a Morris water maze. We made sham or aspiration lesions in the dysgranular zone of somatosensory cortex (S1DZ), the mouse equivalent of Area 3a. We found that lesioning S1DZ impairs both the continued learning of a partially-learned platform location, and the learning of a new platform location. Both groups could learn the location of a visible platform after their lesions. Air-righting and rotarod tests showed that both groups of mice had normal motor and vestibular function. These results suggest that proprioception is necessary for the establishment of long-term visual memory of a vanished target and that the retinal-spatial transformation for on-line vision utilizes corollary discharge.