Rehabilitation of the argument structure deficits in aphasia

Background: The syntactic impairment which frequently emerges in aphasia is mostly due to damage to the ability to map thematic information to the corresponding verb argument structure (VAS). In agrammatic patients this disorder is usually associated with initiation difficulties. Aims: The present study describes a new technique for the rehabilitation of VAS deficits in the matrix clause. Methods & Procedure: The treatment is organized in six levels, and is based on a set of sentence lists of increasing VAS complexity. The efficacy of the treatment was tested on a sample of five chronic (more than one year post-onset) agrammatic patients and three chronic severe non-classifiable aphasic patients suffering from predominant impairment in mapping thematic information to the appropriate VAS. Two sentence list sets were employed: one for the evaluation and treatment phases and the other for the evaluation phase only. Improvement in performance on the first list indicates efficacy of the training program on the treated material, while improvement on the second list indicates generalization to untreated material. The treatment efficacy and generalization to other communicative conditions was also tested with a picture-naming task of nouns and verbs, a picture description task and a spontaneous speech sample (mean sentence length). The patients’ syntactic abilities were tested at the beginning and at the end of the treatment, and again after a six-to-twelve month follow-up. Outcomes & Results: Patients improved significantly on both lists (i.e., there was significant training effect as well as generalization effect to the untreated material) and on the verb-naming task, the picture description task, and on mean sentence length in spontaneous speech. The treatment effect was seen to be stable at the six-to-twelve month follow-up. Conclusions: The VAS training described in the present study is an efficacious technique for the treatment of severe syntactic impairments of the matrix clause; furthermore, the procedure employed for testing its efficacy of the technique is valid for assessing the generalization of training effects to untreated material.