| Abstract: | Background: The verbal communication of persons with aphasia may be disturbed by semantic, phonological, and/or syntactic processing deficits. For those with prominent linguistic-level disorders at least part of aphasia therapy is spent on the main linguistic skills, aimed at improvement of verbal communicative abilities. However, the relationship between these linguistic levels and verbal communication is not straightforward. This is especially true for deficits at the word level: semantic and phonological disorders. Aims: Exploration of the relative impact of semantic and phonological deficits on verbal communicative ability was the aim of the study. The hypothesis was that a lexical semantic deficit has a larger impact on the verbal communicative ability of persons with aphasia than a phonological deficit. Methods & Procedures: A total of 29 persons with aphasia who had both a semantic and phonological deficit were assessed by means of semantic and phonological tasks, and a test of verbal communication (ANELT-A). Semantic measures: Semantic Association Test, Synonym Judgement (receptive tasks), Wordfluency Categories, AAT spontaneous speech: semantics (expressive tasks). Phonological measures: Repetition of words, Repetition of non-words, Wordfluency Letters, AAT spontaneous speech: phonology. Linear regression analyses were performed with the ANELT-A as dependent variable and the semantic and phonological measures as independent variables. Multivariate regression analyses were performed to directly compare the independent contribution of the semantic version of a measure with the phonological version of the same measure. Outcomes & Results: Univariate regression analysis showed that the expressive semantic measures, Wordfluency Categories and AAT semantics, contributed significantly to the prediction of ANELT-A. One phonological measure, Wordfluency Letters, was selected as a significant predictor. However, in the multivariate regression analysis, only the semantic measures appeared to contribute independently to the prediction of the ANELT-A. The contribution of Wordfluency Letters, found in the univariate regression analysis, can be explained by a correlation with its semantic counterpart. Conclusions: These results support the hypothesis that semantic measures contribute more to the prediction of the ANELT-A (Understandability in verbal communication) than phonological measures. The ability to generate semantically correct content words, both in connected speech and in isolation, appears to be decisive for the aphasics' verbal communicative skills. Taking into account the severity of the semantic disorder in relation to other deficits, perhaps the best way to improve verbal communication in persons with both a semantic and a phonological deficit is to address the deficit with the greatest impact: the lexical semantic deficit. |
