Abstract
We report a single case study of a 22-year-old, brain-damaged patient, Jennifer, who showed a semantic deficit affecting living entities (animals and fruit and vegetables) to a greater extent than non-living ones (implements and means of transport). We first show that this category effect was reliable both across time and naming conditions and that it was not an artefact of uncontrolled stimulus factors. We then show that Jennifer had no impairment at the visual or structural processing level and that her deficit was probably located at a semantic processing level. Specific semantic deficits for living entities have usually been explained by damage to the visual semantic system. However, when Jennifer's access to visual and non-visual semantics was assessed through an attribute-verification task, no evidence of an attribute-specific impairment was found: Jennifer was equally impaired in retrieving visual and non-visual attributes of living entities and she was not at all impaired in retrieving visual attributes of non-living entities. Thus, the hypothesis of damage to visual semantics cannot account for the pattern of living things impairment found in this patient. Rather, this pattern seems to require the assumption that the semantic system is organized according to the living/non-living dimension.