To date the anatomical substrate(s) of prism adaptation remain(s) particularly debated, with two main candidates emerging from the literature: the posterior parietal cortex (PPC) and the cerebellum. The functional processes involved in the acquisition of the adaptive aftereffects also remain largely unknown. The main result shown here is that a patient with a bilateral optic ataxia can adapt to an optical deviation, which allows us to make a step forward on these two issues. First, it demonstrates that the corresponding part of the PPC is not a necessary substrate for prism adaptation. Second, since this patient exhibits deficit for fast visuo-motor guidance, it provides direct evidence for a dissociation between on-line visuo-motor control and visuo-motor plasticity. Since the intermanual transfer rate of adaptation is larger in this patient than in control subjects, the PPC may still have an influence on adaptation under normal conditions. We propose a model of the relative contribution of the PPC and the cerebellum during prism exposure, associating these two structures with the two interacting behavioural components of prism adaptation described by previous psychophysical experiments: the strategic component would be linked to the PPC and the adaptive component to the cerebellum. In this model, the strategic component enters in conflict with the development and the generalisation of the adaptive aftereffects. This idea is compatible with the fact that a lesion of the PPC increases the transfer rate and the generalisation of the adaptation, as is also observed in unilateral neglect.