Two mental disorders as disease models of language: linguistic (dis)organisation in schizophrenia and Huntington's disease
Language is a core aspect of human cognition and richly interacts with aspects of nonverbal cognition, as evidence from normal development (Perszyk & Waxman, 2018), abnormal development (Schroeder et al., 2020), and aphasia following brain damage (Fonseca et al., 2019; González et al., 2020) has...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis doctoral |
| Estado: | Versión publicada |
| Fecha de publicación: | 2022 |
| País: | España |
| Institución: | CBUC, CESCA |
| Repositorio: | TDR. Tesis Doctorales en Red |
| OAI Identifier: | oai:www.tdx.cat:10803/687810 |
| Acceso en línea: | http://hdl.handle.net/10803/687810 |
| Access Level: | acceso abierto |
| Palabra clave: | Mental disorders Trastornos mentales Schizophrenia Huntington’s disease Enfermedad de Huntington Esquizofrenia 81 |
| Sumario: | Language is a core aspect of human cognition and richly interacts with aspects of nonverbal cognition, as evidence from normal development (Perszyk & Waxman, 2018), abnormal development (Schroeder et al., 2020), and aphasia following brain damage (Fonseca et al., 2019; González et al., 2020) has come to suggest. Yet, how cognitive dysfunction relates to language dysfunction in adult-onset mental disorders still remains an open area of research. Profiling language disturbances in this clinical context could be practically significant in helping to discover representative biomarkers of disease progression and early detection, apart from contributing theoretical insights into the relation between language and cognition. In line with this, the main objective of this thesis is to profile deviant linguistic patterns in people with Huntington’s disease (HD) and Schizophrenia (SZ), both of which have only rarely been analysed from a linguistic point of view. Five linguistic studies were conducted. Study 1 (chapter 3) develops the linguistic profile of HD through the analysis of spontaneous speech samples from 20 individuals compared to 20 neurotypical subjects (NTs). Its aims were to replicate and expand a previous study in a different sample (Hinzen et al., 2018), using a total of 56 variables related to grammatical organisation. The second and third studies are two experimental neuropsychological studies, one using a grammaticality judgment task (Study 2, chapter 4), the other a sentence-picture matching task (Study 3, chapter 5). For these two studies, a new sample of 31 NTs and 31 patients with HD classified into pre-symptomatic, early and advanced stages was collected. Motivated by results from the general linguistic profiling of the HD population in the previous study, the focus here was experimental and the purpose was to specifically target the processing of illicit syntactic movement and the processing of Binding Theory principles respectively (Chomsky, 1984). Studies 4 and 5 focus on two unique corpora of speech in SZ. Study 4 (chapter 6) focuses on a rare corpus of spontaneous speech from 38 patients with an unusually high severe form of formal thought disorder (FDT), a core symptom of SZ according to the DSM-5. This corpus was collected and first analysed in Moya (1989). Our study captured patterns of linguistic disintegration across different linguistic strata (referential anomalies, argument structure, lexicon and morphosyntax). Study 5 (chapter 7) is based on a corpus of hallucinated voice talk in patients with a high symptom load of auditory verbal hallucinations (AVHs). A linguistic profile of hallucinated voice talk based on literal transcriptions was built, through the analysis of transcripts from 19 patients with highly frequent voice talk. Overall, results from both HD and SZ populations provided evidence for distinctive and specific linguistic effects, which are not easily interpreted as secondary to primary motor dysfunction (in the case of HD) or to intellectual disability or nonverbal neurocognitive impairments (in the case of SZ). Specifically, pre-symptomatic and symptomatic HD patients exhibited distinct but complementary language patterns in certain grammatical domains: in the fluency domain, pre-symptomatic patients manifested patterns marked by prolongations, fill pauses, and repetitions, while symptomatic patients were prone to use more empty pauses, truncations and reformulations. In the domain of sentence connectivity, their speech was generally characterised by poor grammatical connections, since the use of parataxis and coordination was very common. The reduction of subordination could be interpreted as a weakness in building syntactic hierarchy. Following this line of thought, the evidence collected in Study 2 further experimentally confirmed a loss of cognitive control over the structural hierarchy as built through linguistic movement of syntactic constituents. Study 3 expanded this result to for syntactic principles involved in licensing (co-) reference. The latter study specifically demonstrates difficulties in processing syntactic locality constraints in the HD group, as captured by traditional principles of the Binding Theory (Chomsky, 1984). In the case of SZ, Study 4 captured patterns of linguistic disintegration comparatively across hierarchical layers of linguistic organization in patients with FTD. In terms of broad linguistic domains, it turned out that even in FTD at this severe end, the morphosyntax and the lexicon were relatively little affected, and much less so proportionally when comparing it with the total number of referential errors, while argument structure was placed in the middle. In turn, the linguistic analysis of hallucinated voice talk (Study 5) revealed a strong dominance of parataxis (isolated clauses without grammatical connection), use of non-anaphoric noun phrases (without connection with previous units), and the relative absence of the first person, grammatical errors and semantic errors. Overall, these results show the feasibility and richness of linguistic profiling outside of neurological disorders said to be language-specific (i.e. aphasia). We regard such profiles as necessary new baselines for integration into neurocognitive models of these diseases; and possibly as informing the development of clinical tools for assessing, monitoring, and detecting cognitive changes and related symptoms. They provide a new dimension for neuropsychological profiling as well, where current test batteries may not capture the relevant linguistic phenomena, thereby adding an extra layer of relevant data. In addition, the richer linguistic disorders turn out to be in mental disorders, the more they motivate new theoretical models that rethink the relationship between cognition and language and link them in systematic ways. |
|---|