Search

The application of Conversation Analysis (CA) to the investigation of agrammatic aphasia reveals that utterances produced by speakers with agrammatism engaged in everyday conversation differ significantly from utterances produced in response to decontextualised assessment and therapy tasks. Early studies have demonstrated that speakers with agrammatism construct turns from sequences of nouns, adjectives, discourse markers and conjunctions, packaged by a distinct pattern of prosody. This article presents examples of turn construction methods deployed by three people with agrammatism as they take an extended turn, in order to recount a past event, initiate a discussion or have a disagreement. This is followed by examples of sequences occurring in the talk of two of these speakers that result in different, and more limited, turn construction opportunities, namely "test" questions asked in order to initiate a new topic of talk, despite the conversation partner knowing the answer. The contrast between extended turns and test question sequences illustrates the effect of interactional context on aphasic turn construction practices, and the potential of less than optimal sequences to mask turn construction skills. It is suggested that the interactional motivation for test question sequences in these data are to invite people with aphasia to contribute to conversation, rather than to practise saying words in an attempt to improve language skills. The idea that test question sequences may have their origins in early attempts to deal with acute aphasia, and the potential for conversation partnerships to become "stuck" in such interactional patterns after they may have outlived their usefulness, are discussed with a view to clinical implications.

Historically, agrammatism, a symptom of Broca's aphasia, has been associated with dysprosody, on account of speakers' slow, halting, and effortful speech. Almost all investigations of this phenomenon use experimental methods (reading, repetition). Thus, little is known about how prosody is used by speakers with agrammatism and understood by their interlocutors in everyday conversations. This paper takes an interactional approach to prosody, using Conversation Analysis to explore everyday conversations between three speakers with agrammatism and their family members/friends, recorded in the home. A distinct prosodic pattern is revealed in their talk, whereby non-final words in an agrammatic utterance are produced with mid-level or minor rising pitch, and final words with a prominent pitch excursion. The analysis shows that conversation partners orient to terminal pitch movement as a signal of turn completion. Conversely, they do not take the floor when pitch signals continuation, despite significant pausing and severe grammatical disruption. Thus, prosody appears to function to regulate turn taking in the same way as it does in typical (non-language disordered) conversation. For these three speakers, intact prosodic skills appear to compensate for impaired grammatical ability, by packaging a series of haltingly produced words into an utterance, the meaning of which is responded to by the conversation partner as the sum of its agrammatic parts.

Current approaches to assessing agrammatism use data from restricted contexts, such as picture description and story telling tasks. There is evidence in the conversation analysis literature to suggest that conversational grammar may differ markedly from the grammar of such elicited language samples. The disparity between conversational and test grammar suggests that it is possible for the form of an agrammatic utterance to be motivated by the context in which it occurs. Thus, behaviours previously considered to be symptoms of agrammatism may be adaptations to talking in different environments. This study analyses two distinctive test response forms made by two agrammatic speakers, and discusses whether they might be strategic adaptations to testing rather than symptoms of impairment.

This paper uses the methodology and analytical findings of conversation analysis to investigate the notion that aphasic grammar may be understood at least partly in the context of the demands of turns at talk in conversation. An investigation of the conversation of an English-speaking person with aphasia reveals two distinct grammatical phenomena, and it is suggested that their use may be interactionally motivated by the need to take a relatively unproblematic turn at talk despite the constraints of non-fluent aphasia. The grammatical patterns that are revealed by this analysis look considerably different from those elicited by standard methods of data sampling. The possibility that interactional grammatical phenomena are not visible in the language data elicited by clinical assessments is raised. This question is addressed via an in-depth comparison of the same speaker's interactional and elicited grammar in a linked article. It is suggested that the tool of conversation analysis provides researchers with a new and fruitful approach to the study of grammatical abilities in aphasia.

This paper investigates the grammatical difficulties of an English-speaking person with non-fluent aphasia using clinical assessments based on picture description and story telling. The same individual's conversation grammar, which was investigated in detail in a linked article is reviewed here, and the notion that interactional grammatical phenomena may not necessarily be visible in elicited language data is explored. Data analysis shows that the aphasic speaker's grammar looks considerably different in the context of clinical assessment than it does in conversation. Analysis of elicited grammar reveals that the majority of utterances produced are subject-verb-object (SVO) type sentences, whereas prior analysis of conversation data indicated patterns of interactional grammatical phenomena that differ from such sentence types. It is suggested that language tests and conversation provide complementary but essentially different information about grammatical abilities. The clinical implications of this finding are discussed.