We are usually unaware of the motion of an image across our retina that results from our own movement. For instance, during slow-tracking eye movements we do not mistake the shift of the image projected onto the retina for motion of the world around us, but instead perceive a stable world. Following early suggestions by von Helmoltz(1), it is commonly believed that this spatial stability is achieved by subtracting the retinal motion signal from an internal reference signal, such as a copy of the movement command (efference copy)(2-4). Object motion is perceived only if the two differ. Although this concept is widely accepted, its anatomical underpinning remains unknown. Here we describe the case of a patient with bilateral extrastriate cortex lesions, suffering from false perception of motion due to an inability to take eye movements into account when faced with self-induced retinal image slip. This is indicated by the fact that during smooth-pursuit eye movements, he perceives motion of the stationary world at a velocity that corresponds to the velocity of his eye movement; that is, he perceives the raw retinal image slip uncorrected for his own eye movements. We suspect that this deficiency reflects damage of a distinct parieto-occipital region that disentangles self-induced and externally induced visual motion by comparing retinal signals with a reference signal encoding eye movements and possibly ego-motion in general.