Monologues and dialogues in aphasia: Some initial comparisons

Abstract

Background: In recent years, there has been a shift in aphasia research interest from analysis of monologues to conversational dialogues. However, where monologic research has focused primarily on aspects such as language structure and quantity, conversational research has mainly focused on pragmatic aspects of discourse such as turn taking and repair. While there is some acknowledgement of the differences between language use in monologues and dialogues in aphasia, few comparisons have been made that cover a broad range of both syntactic and semantic aspects. Aims: This study aims to compare the discourse of aphasic speakers and non-brain-damaged control speakers in monologues and dialogues by using a range of syntactic and semantic analyses. In so doing, it aims to explore whether or not it is necessary for researchers and clinicians to sample both contexts in order to gain a full picture of a person's linguistic skills. Methods & Procedures: Two individuals with aphasia of varying severity provided monologues about their stroke experience, then participated in dialogues with a “significant other” on the same topic. Two control speakers, matched as closely as possible with the speakers with aphasia, undertook the same tasks, however they discussed a serious illness they had experienced. The SALT program (Miller & Iglesias, 2008) analysed the resulting data for syntactic productivity and complexity, and semantic analyses from a Systemic Functional Linguistic framework (Halliday & Matthiessen, 2004) were used to explore ideational, interpersonal, and textual meanings. Outcomes & Results: The syntactic measures demonstrated greater consistency across contexts, with the participants with aphasia using more and longer C-units in the monologue, although demonstrating more mazes in this context. The control speakers demonstrated slightly more variation but, like the speakers with aphasia, tended to use more words per C-unit in the monologue. Semantic measures demonstrated more variable patterns although some similarities also existed. Conclusions: The variation noted in the study supports the notion that clinicians and researchers should sample a variety of contexts in order to gain an overall picture of a person's language skills. Factors such as personality, topic, and conversation partner influence all seem to play a role in discourse production, particularly from the perspective of meanings conveyed. This study raises questions as to the actual effects of these factors on language structure and function and provides direction for future research in the area.