Patients who had undergone a unilateral trans-sylvian selective amygdalohippocampectomy as treatment for chronic intractable epilepsy were tested in a virtual Morris Water Maze (MWM) task where they were required to locate a hidden platform as a measure of spatial learning. These individuals’ performance on spatial tasks was compared to age-matched healthy controls and drug-matched healthy controls. Training occurred in two different maze environments, one with conventional cues such as windows and doors, and another with abstract cues, such as colours and patterns. Participants searched for a hidden platform in the virtual pool, guided by either the conventional or abstract cues. There was a significant impairment in the surgery group compared to the control groups in all environments, however in the abstract environment only the patients with right-sided lesions were significantly worse than the controls. There was no difference between the groups on a control egocentric navigation task. These results suggest that people who have had right-sided surgery are impaired in spatial tasks, and that the level of impairment on the spatial task may be dependent on the characteristics of the cues such as how easily the cues are verbalised. These results support the notion of the functional lateralization of specific elements of spatial memory and functional lateralization, and may shed light on previous inconsistencies in this area of research.