Most studies examining category specificity are single-case studies of patients with living or nonliving deficits. Nevertheless, no explicit or agreed criteria exist for establishing category-specific deficits in single cases regarding the type of analyses, whether to compare with healthy controls, the number of tasks, or the type of tasks. We examined two groups of patients with neurological pathology frequently accompanied by impaired semantic memory (19 patients with Alzheimer's disease and 15 with Herpes Simplex Encephalitis). Category knowledge was examined using three tasks (picture naming, naming-to-description, and feature verification). Both patient groups were compared with age-and education-matched healthy controls. The profile in each patient was examined for consistency across tasks and across different analyses; however, both proved to be inconsistent. One striking finding was the presence of paradoxical dissociations (i.e., patients who were impaired for living things on one task and nonliving things on another task). The findings have significant implications for how we determine category effects and, more generally, for the methods used to document double dissociations across individual cases in this literature.