Wielopoziomowe badania dyskursu użytkowników języka polskiego z diagnozą afazji mieszanej - dane badawcze

1. Project objectives and research questions
The scope of the present project is to investigate the discourse produced by Polish-speaking individuals affected by aphasia, an impairment of language due to injury to the brain, most commonly from stroke. In studies of aphasia, discourse is defined as a unit of language larger than a single sentence or as a type of message that is used to fulfill specific goals (e.g., a story, an opinion or a description of a procedure) (Halliday & Matthieson, 2004). A recent systematic scoping review of empirical studies of the Polish language in aphasia (Dębski, Wójcik-Topór & Knapek, in press) concludes that the discourse of Polish-speaking individuals diagnosed with aphasia “requires more attention from researchers […] to reflect the strong interest in the discourse of aphasic individuals in world literature”. The project therefore aims to study the discourse of Polish-speakers with mixed aphasia to: 1) describe and measure the linguistic
devices and patterns of cohesion; 2) develop a Polish adaptation of van Leer and Turkstra’s (1999) local/global coherence measuring scales; 3) investigate the relationship between the micro- and macro-linguistic aspects of discourse; 4) explore the impact of mixed aphasia on discourse cohesion and coherence, and 5) investigate the relationship between discourse genre and linguistic cohesion and coherence.
The study will be driven by the following research questions:
1. What grammatical and lexical discourse cohesion devices and patterns are used by the aphasic study participants? How do they compare with the grammatical and lexical
discourse cohesion devices used by the brain-healthy controls? How are they affected
by aphasia type, severity and discourse genre?
2. What is the local coherence of the discourse produced by the aphasic participants? How does it compare with the measured local coherence of the discourse produced the brainhealthy controls? How does it change depending on aphasia type, severity and discourse genre?
3. What is the global coherence of the discourse produced by the aphasic participants?
How does it compare with the measured global coherence of the discourse produced
the brain-healthy controls? How does it change depending on aphasia type, severity and discourse genre?
4. What relationships can be ascertained between breakdowns in the expression of
meaning at the different discourse levels: lexical, grammatical, cohesion, and local and
global coherence? How are those relationships mediated by aphasia type, aphasia
severity and discourse genre?
We hypothesise 1) the existence of a threshold at which micro-level breakdowns become so significant as to cancel the contributory clues of context and text structure, make aphasic texts functionally incoherent, leading to a macro-level breakdown (Anderson, 2000) and 2) that the discourse genres will impose different cognitive demands on the aphasic individuals leading to different patterns of discourse impairment (cf. Ulatowska, Allard & Bond Chapman, 1990).
2. Project significance
Research problem and project rationale: Aphasia is a burning issue in our society due to its negative impact on quality of life. Linguistically informed studies on aphasia have aimed to explain language impairment in aphasia by drawing on generative grammar theory (e.g., Bastiannse & van Zonneveld, 2006) or functional usage-based linguistic theories (e.g., Martínez-Ferreiro, Bastiaanse & Boye, 2020). Recently, an increase in linguistic analyses of aphasic discourse can be observed (e.g., Bryant, Ferguson & Spencer, 2016). These analyses have addressed such topics as: discourse treatment (Dipper et al., 2020), description of aphasic discourse (Bryant, Ferguson & Spencer, 2016; Panasiuk, 2019) and discourse measures/analysis methods (Pritchard, 2017).
A micro- or a macro-linguistic approach has been used to investigate the characteristics
of aphasic discourse (Armstrong, 2000). Research conducted from the micro-linguistic
perspective has discovered, for example, the use of paraphasias and non-specific lexical items by individuals with aphasia (Wagenaar, Snow & Prins, 1975) and described the syntactic features of aphasic discourse, such as syntactically less complex sentences and use of verbs with simpler argument structures (Zhang, Geng, Yang & Ding, 2020). Research undertaken from the macro-linguistic perspective has addressed the ways meaning is organised in discourse and the overall meaning of texts representing different discourse types. A range of issues related to discourse macrostructure have been explored, e.g., the use of specific interactive phenomena (conversational repair, adjacency pairs, turn-taking) (Lindsay & Wilkinson 1999) or speech acts (e.g., Prutting & Kirschner, 1987). There is insufficient research devoted to the potential connections between the micro- and macro-linguistic approaches to discourse analysis (Anderson, 2000) and disagreement on the impact of lexicalgrammatical impairment on the delivery of meaning in aphasia. On the one hand, studies reveal a dissociation between the micro- and macro-linguistic skills in aphasic individuals, explained by the hypothesised association of global coherence skills with the right hemisphere rather than the left hemisphere, responsible for micro-linguistic skills (Glosser & Deser 1990). On the other hand, there is research tracing relationships between disorders in using specific language subsystems and the ability to plan utterances and discourse (Żydek-Bednarczuk, 2005). Anderson (2000) hypothesises that the micro- versus macro-linguistic perspective might be a false dichotomy altogether and coherence should be regarded as a continuum, closely linked to text microstructure, with a point at which the quantity and/or quality of lexical-grammatical
breakdowns may lead to difficulties in the interpretation of meaning at the functional level (cf. Żydek-Bednarczuk, 2005). Also, cohesion analysis, as defined by Halliday and Hasan (1976), is proposed as a research methodology possibly allowing to overcome the traps of the dichotomy and link the micro- and macro-aspects of discourse (Zhang, Geng, Yang & Ding,2020).
To date, most studies of aphasic discourse have been conducted with the participation
of individuals with fluent or non-fluent aphasia (e.g., Zhang, Geng, Yang & Ding, 2020).
Researchers have suggested that cohesion and coherence research needs to include other aphasia types (Martínez-Ferreiro et al., 2017). Research of aphasic discourse is further complicated by evidence the discourse genres impose different demands on individuals with aphasia (Ulatowska, Allard & Bond Chapman, 1990; van Leer & Turkstra, 1999). Summary of project rationale, innovation and impact: The proposed project is both innovative and timely. Firstly, it will contribute much-needed evidence to verify the current hypotheses on the relationship between the micro- and macro-linguistic aspects of aphasic discourse. Secondly, the research will involve participants diagnosed with mixed aphasia, an aphasia type rarely investigated in this context. Thirdly, it will provide new evidence on aspects of cohesion and coherence of Polish aphasic discourse, so far absent in the literature. These project innovations will allow to advance linguistic knowledge and make significant contributions to the applied discipline of aphasiology. Also, popular local and global coherence measures will be translated into Polish, standardised and validated, equipping Polish aphasiologists with a new research/clinical tool. The source data will be made available internationally through the AphasiaBank database (https://aphasia.talkbank.org).
3. Research plan, research aims and risk analysis patients with the diagnosis of mixed aphasia with motor component, patients with mixed aphasia with sensory component and brain-healthy patients, serving as control participants, will be recruited at the Post-stroke and General Rehabilitation Ward and the Speech-Language Pathology Clinic of the Ludwik Rydygier Hospital and at the Polish Centre for Functional Rehabilitation VOTUM in Kraków. The following inclusion and exclusion criteria will be used: 1) mixed aphasia resulting from a stroke verified through neuroimaging or a clear medical diagnosis; 2) between 3 weeks and 36 months post-injury at the time of
entry; 3) co-existing apraxia and/or dysarthria acceptable; 4) Polish as first language as per self-report; 5) no dementia or co-morbidities associated with serious cognitive consequences;
6) vision and hearing (unaided or aided) adequate for testing. The aphasic patients will be matched one-to-one on age, gender and education level with brain-healthy individuals to form three samples across which comparisons can be attempted. Patients (or their proxies if necessary) will be asked to give informed consent to taking part in the proposed project. The recruitment process will be regulated by the clearance obtained from the Research Ethics Committee at the Institute of Polish Glottodidactics at the Jagiellonian University and the Institutional Review Board regulations of the AphasiaBank project. There will be two sources of data. The main data source will be the audio recordings elicited with the AphasiaBank discourse tasks, which will be translated, culturally adapted and piloted. The second data source will be the study participants’ medical records from which relevant demographic/medical information will be extracted. The audio recordings will be transcribed orthographically before the analysis. The analysis of the discourse data will involve several tasks leading to well-defined aims. Firstly, a lexical analysis of the corpus will be performed to identify the occurrence of paraphasias and non-specific nouns and pronouns, such
as “thing”, “stuff”, “that”(POL “tamto”, “rzeczy”, “tamto”), and compute the frequency of
occurrence of those lexical categories as proportions of total words produced by participants. Secondly, samples will be segmented into utterances in order to calculate the mean length of utterance in morphemes (mMLU) and percentage of grammatical utterances. Thirdly, a deductive-inductive content analysis framework (Hsieh & Shannon, 2005) will be applied. The analysis will start with a list of codes based on the work of Halliday and Hasan (1976) and Wojenka-Karasek (2016) and will be modified through inductive reading of the data to capture the current content. Fourthly, a quantitative cohesion analysis will be performed, focusing on several cohesive ties, and measure their quantity and diversity in the aphasia and neuronormative participants’ speech across the discourse genres. Finally, van Leer and Turkstra’s (1999) local and global coherence rating scales will be translated, adapted, piloted and used to
assess the levels of coherence in the discourse samples. The general lineal model will be used to compute relationships between the aphasia types and the micro- and macro-linguistic measures across the different discourse genres as well as correlations among the five discourse measures used in the study. The project results will be interpreted with reference to current international and Polish subject literature and disseminated through conference presentations and at least three articles published in high-impact international journals, such as International Journal of Speech-Language Pathology, Brain and Language and Aphasiology. Risk analysis: The projected number of participants is based on the number of poststroke patients undergoing treatment/rehabilitation at the collaborating institutions in any sixmonth period and it is feasible in the assessment of the collaborating speech pathologists.
However, given the unpredictable impact of the COVID-19 pandemic, the participant numbers or the inclusion/exclusion criteria may have to be adjusted in a way that does not diminish the project impact. The worst-case scenario is that some participants would have to be recruited and data collected via online consultations.
4. Research methods
A range of qualitative and quantitative methods and tools will be employed in an approach that can be described as a concurrent mixed methods research design (Creswell, 2014): Discourse data gathering script: The discourse data gathering script will consist of 7 discourse tasks: 1) Stroke story (free speech), 2) Important event (free speech), 3) Broken window (descriptive), 4) Refused umbrella (descriptive), 5) Cat rescue (descriptive), 6) Cinderella (narrative), 7) Ham and cheese sandwich (procedural). Lexical and grammatical analyses: A lexical analysis of the collected data will be performed in order to identify the occurrence of paraphasias and non-specific nouns and pronouns. Frequency of occurrence of those lexical categories will be computed as proportions of total words produced by participants. Second, the samples will be segmented into utterances and the mean length of utterance in morphemes (mMLU) and percentages of grammatical utterances will be calculated.
Directed content analysis (Hsieh & Shannon, 2005): The discourse transcripts will be coded for the presence of predetermined cohesion ties in the following categories: grammatical cohesion (reference, substitution and ellipsis), lexical cohesion (reiteration and collocation) and conjunction (Halliday & Hasan, 1976; cf. Zhang, Geng, Yang & Ding, 2020). The code definitions will be based on Wojenka-Karasek’s (2016) analysis of the cohesive devices in Polish texts written by foreigners. Any text unit that cannot be categorised according to the initial categories will be given a new code. The new codes will be analysed to see if they are a subcategory within an existing code or constitute a new category (Hsieh & Shannon, 2005).
The result of this analysis will be a classification of the cohesive ties used in Polish by
individuals with mixed aphasia, so far absent in the subject literature.
Quantitative cohesion analysis: A quantitative cohesion analysis will be conducted to
establish: 1) the quantity of the cohesive devices and 2) the diversity of the cohesive devices in the discourse samples collected in the project (cf. Zhang, Geng, Yang & Ding, 2020). Ratios of the total number of cohesive ties by the total number of T-units will be calculated and used as an indicator of the quantity of cohesive devices in the discourse of participants with different aphasia types and across the different discourse genres. The diversity of the cohesive devices will be established in the three general categories of cohesive devices (grammatical devices, lexical devices and conjunction). Proportions of each category of cohesive ties will be calculated by dividing the number of occurrences of a single category by the total number of occurrences of cohesive ties in the three categories. Coherence rating scales adaptation and coherence analysis: A Polish adaptation of van Leer and Turkstra’s (1999) scales for the assessment of local and global coherence will be developed and coherence analysis of the collected discourse data performed. Analysis of relationships: Correlation analysis will be used to trace relationships between the quantitative results of the lexical, grammatical, cohesion, local coherence and global coherence analyses. The study will also use the general linear model to compute relationships between 1) the aphasia presentations (mixed aphasia with motor component and mixed aphasia with sensory component), 2) aphasia severity; and 3) the discourse types elicited in the study (free
speech, description, narrative and procedural discourse) and the scores of the lexical, grammatical, cohesion, local coherence and global coherence analyses’ results.
Inter-rater reliability tests: All discourse samples will be coded and scored by two linguists.Inter-rater reliability will be calculated for T-units, lexical, grammatical, cohesive tie, local and global coherence analyses. The minimum acceptable inter-rater reliability will be defined as k> 0.80 (Cohen’s kappa). Differences will be discussed until consensus is reached.
5. References
Armstrong, E. (2000). Aphasic discourse analysis: The story so far. Aphasiology, 14(9), 875- 892.
Bastiaanse, R. & van Zonneveld, R. (2006). Comprehension of passives in Broca’s aphasia.Brain and Language, 96(2), 135–142.
Bryant, L., Ferguson, A. & E. Spencer (2016) Linguistic analysis of discourse in aphasia: Areview of the literature. Clinical Linguistics & Phonetics, 30(7), 489-518.
Cahana-Amitay, D. & Jenkins, T. (2018). Working memory and discourse production in
people with aphasia. Journal of Neurolinguistics, 48, 90–103Creswell, 2014
Dębski, R., Wójcik-Topór, P. & Knapek, M. (in press). Polish language of aphasia: A scoping review in the era of the ICF. Linguistica Silesiana, 42.
Dipper, L., Marshall, J., Boyle, M., Botting, N., Hersh, D., Pritchard, M. & Cruice, M.
(2020): Treatment for improving discourse in aphasia: a systematic review and synthesis of the evidence base, Aphasiology.
Glosser, G. & Deser, T. (1990). Patterns of discourse production among neurological patients with fluent language disorders. Brain and Language, 40, 67-88.
Halliday, M.A.K. & Hasan, R. (1976). Cohesion in English. London: Longman.
Halliday M.A.K. & Matthieson, M.I.M. (2004). Halliday’s introduction to functional
grammar. Hachette UK: Hodder Education.
Hsieh, H-F. (2005). Three approaches to qualitative content analysis. Qualitative Health
Research, 15(9), 1277-1288.
Lindsay, J. & Wilkinson, R. (1999). Repair sequences in aphasic talk: a comparison of
aphasic-speech and language therapist and aphasic-spouse conversations. Aphasiology, 13(4-5), 305-325.
Martínez-Ferreiro, S., Bastiaanse, R. & Boye, K. (2020). Functional and use-based
approaches to aphasia: the grammatical-lexical distinction and the role of frequency.
Aphasiology, 34.
Pritchard, M., Hilari, K., Cocks, N. & Dipper, L.T. (2017). Reviewing the quality of
discourse information measures in aphasia. Journal of Lang and Comm Disorders, 52(6), 689-732.
Prutting, C.A. & Kirschner, D.M. (1987). A clinical appraisal of the pragmatic aspects of
language. Journal of Speech & Hearing Disorders, 52(2), 105–119.
Ulatowska H.K., Allard L., Chapman S.B. (1990) Narrative and Procedural Discourse in
Aphasia. In Y. Joanette, H.H Brownellm (Eds). Discourse Ability and Brain Damage.
New York, NY: Springer.
van Leer, E. & Turkstra, L. (1999). The effect of elicitation task on discourse coherence and cohesion in adolescence with brain injury. Journal of Comm Disorders, 32, 327-349.
Wagenaar, E., Snow, C. & Prins, R. (1975). Spontaneous speech of aphasic patients: a
psycholinguistic analysis. Brain and Language, 2, 281-303.
Wojenka-Karasek, M. (2016). Wykładniki spójności w tekstach pisanych przez obcokrajowców (na materiale egzaminów certyfikatowych z języka polskiego jako obcego). Rozprawa doktorska. Uniwersytet Łódzki.
Zhang, M., Geng, L., Yang, Y. & Ding, H. (2020): Cohesion in the discourse of people with post-stroke aphasia. Clinical Linguistics & Phonetics.
Żydek-Bednarczuk U. (2005). Wprowadzenie do lingwistycznej analizy tekstu, Kraków:
Universitas.