The effect of a language therapy in a group of eight anomic mild patients (the Lexical Therapy group) was assessed by using a 5-month long Lexical Therapy in comparison with an occupational program used in a matched control group (AD; n = 8). The Lexical Therapy group benefited significantly from a language therapy as shown by the naming improvement postintervention. The improvement reached significance only for items that were included in the language therapy protocol and no significant generalization to untreated items was observed. In mild AD patients with anomia and no severe semantic impairment, a reinforcement of the relationship between the form of the object and the corresponding lexical label in episodic long term memory during language therapy may account for the observed lexical improvement.