Although levels of verbal and pictorial performance are known to depend on the degree of left versus right atrophy in the early stages of semantic dementia, the nature of these differences remains controversial. It has been proposed that there is a unitary, bilaterally represented, abstract semantic system and that differential task performance reflects the impact of greater connectivity between the left anterior temporal lobe (ATL) and the left dominant language systems. This interpretation explains the greater involvement of the left ATL in verbally coded semantic knowledge, but not the prevalence of the right hemisphere in pictorial representations. An alternative account is provided by the sensory–motor model of conceptual knowledge, which assumes that each conceptual representation results from the convergence of different perceptual, motor, and verbally coded sources of knowledge in a given brain area. According to this model, the weight of verbal information should prevail in left ATL conceptual representations, because of the dominance of the left hemisphere for language, whereas the weight of sensory–motor sources of knowledge should be greater in the right ATL representations, because the right hemisphere plays a greater role in processing sensory–motor information. If the difference between right and left conceptual representations is quantitative and due to the different weight of sensory–motor and verbal sources of knowledge in their composition, we should observe an elementary, but selective representation of semantic–lexical knowledge in the intact right hemisphere and a mild but selective semantic–lexical impairment in right-brain-damaged patients. Results of the present survey support this hypothesis.