Objectives: Anomia is one of the most common and persistent symptoms of aphasia. Although treatments of anomia usually focus on semantic and/or phonological levels, which both have been demonstrated to be effective, the relationship between the underlying functional deficit in naming and response to a particular treatment approach remains unclear. The aim of this study was to determine the relationship between the effects of specific treatments (Semantic feature Analysis and Phonological Components Analysis) and their underlying functional deficit patterns within the framework of a cognitive processing model. Methods:In an ABCB reversal control task design, four participants with aphasia were selected according to the criteria based on using a cognitive model of lexical processing. Each patient received two types of treatment. In SFA, features semantically associated to the target words were elicited from the patient, whereas in PCA treatment, the phonological components of the target words were the focus of treatment. Naming accuracy scores obtained in pretreatment baseline phase were compared to post-treatment accuracy scores. Here, both itemspecific effects and generalization of untrained items were analyzed.Results: Both SFA and PCA treatments have the potential to improve naming ability in participants; however, the treatment approach that corresponds exactly to the underlying deficit causing failure in word retrieval is more effective.Discussion: While PCA is more effective for participants with phonological impairments, SFA is more effective for participants with semantic impairments. Therefore, a direct relationship between underlying functional deficit and response to specific treatment was established for all participants.