Interaction and Language Assessment in Aphasia and Dementia : A Comparative Perspective

Sammanfattning: Language problems in dementia resemble the symptoms of aphasia in many respects. A growing body of research discusses the cognitive deficits associated with aphasia. Despite common denominators, very little is written with a comparative perspective on the two clinical groups. Although speech and language pathologists (SLPs) play a central role in aphasia care, they are not routinely involved in healthcare services for dementia. By tradition, language assessments tend to be test-oriented, even though there is an awareness of the advantages with informal assessment approaches. The overall aim of this thesis was to examine interaction in persons with aphasia (PWA) and persons with dementia (PWD) in test conversation and more informal conversations. The thesis has an interactional focus with a comparative perspective on the two clinical groups, on conversational contexts, and on test results with reference to SLP services.  Study I, involving ten PWA, and study II, involving ten PWD, had similar approaches, investigating the organization of interaction between the participants and SLPs in test conversations and in more informal conversations. Furthermore, the participants’ interactional abilities were related to their actual test results on expressive tasks on an aphasia test battery. Study III in-volved detailed analyses of the Mini-Mental State Examination (MMSE) test interaction for both PWA and PWD. Additionally, the study set out to explore the relationship between linguistic and cognitive difficulties, in relation to assessments. Study IV investigated instances of trouble in conversations involving PWA/PWD and SLPs, with a particular focus on “trouble domain” and interactional consequences. The results of study I and II demonstrated substantial differences be-tween the organization of interaction in test conversations and informal con-versations, regardless of whom they involved. The informal conversations pro-vided opportunities for the PWA/PWD to talk more and to initiate own topics and multimodal resources were used by the PWA. This was seen to a far lesser degree among the PWD. With a few distinct exceptions, the PWD came across as rather communicatively competent. This was not necessarily reflected by the aphasia test results, since several participants struggled with a couple of test assignments. In PWA, the demonstration of aspects of communication that could be related to being a competent speaker did not fully correlate with aphasia test scores. The analyses showed that instances of trouble occurred equally often in the two conversational contexts whereas trouble characteristic differed. Mutual trouble-solving was almost solely observed in the informal conversations. Study III revealed some particularly challenging aspects of the MMSE test interaction. The study shed light on the problematic issue of separating language and cognition, since PWA and PWD had similar test results on the cognitive screening and since the “language” test items did not seem to capture linguistic problems more than the other remaining test items. Qualitative analyses of the interactional aspects of test situations may reveal information about both cognitive and linguistic abilities that otherwise would have been over-looked. The in-depth analyses of conversational trouble in study IV revealed that most troubles involving PWA were connected to primarily linguistic is-sues. Conversational trouble in PWD, however, typically labelled “lexical problems”, were many times due to primarily cognitive issues. It was also observed that the SLPs took a more passive role trouble-solving in conversations involving PWD. Less severe linguistic problems in the PWD, lack of shared personal common ground, and preconceived notions about the medical conditions alongside with SLPs’ professional culture and experience are discussed as potential motives for this behaviour.  Altogether, the results of the present thesis demonstrate that formal tests that attempt to measure language or cognition do not take into account that it is problematic to separate these abilities. Within SLP services, it would be preferable to move away from a fault-finding perspective on assessment and intervention, towards an approach in which language and cognition as co-constructed acts is central.