Previous research has demonstrated that adults can adapt to novel sensorimotor perturbations, a process thought to be achieved by the gradual update of an adaptive internal representation. However, few research studies have investigated the persistence of a newly acquired representation, as assessed by the reduction of performance errors after the perturbation has been removed (i.e., de-adaptation). The primary objective of this thesis was to determine if the central nervous system (CNS) could flexibly utilize visual and proprioceptive afference to de-adapt to novel sensorimotor perturbations. It has been previously demonstrated that the CNS relies more heavily on visual information for hand localization in the azimuthal direction whereas proprioception is more heavily weighted for hand localization in the radial direction. Seventy-two right-handed adults executed reaching movements during exposure to either an incremental visuomotor rotation or gain distortion. Visual feedback provided during post-exposure was manipulated. Results indicate that the CNS predominantly utilized visual afference to de-adapt to both perturbations, despite the fact that rotation adaptation resulted in movement errors in the azimuthal direction whereas gain adaptation resulted in movement extent errors. These data suggest that the CNS did not flexibly re-weight proprioceptive afference in the absence of visual feedback during a center-out drawing task.