Quantifying incoherence in speech: An automated methodology and novel application to schizophrenia

Abstract

Incoherent discourse, with a disjointed flow of ideas, is a cardinal symptom in several psychiatric and neurological conditions. However, measuring incoherence has often been complex and subjective. We sought to validate an objective, intrinsically reliable, computational approach to quantifying speech incoherence. Patients with schizophrenia and healthy control volunteers were administered a variety of language tasks. The speech generated was transcribed and the coherence computed using Latent Semantic Analysis (LSA). The discourse was also analyzed with a standard clinical measure of thought disorder. In word association and generation tasks LSA derived coherence scores were sensitive to differences between patients and controls, and correlated with clinical measures of thought disorder. In speech samples LSA could be used to localize where in sentence production incoherence occurs, predict levels of incoherence as well as whether discourse "belonged" to a patient or control. In conclusion, LSA can be used to assay disordered language production so as to both complement human clinical ratings as well as experimentally parse this incoherence in a theory-driven manner.

Lexical analysis in schizophrenia: How emotion and social word use informs our understanding of clinical presentation

“…A fourth limitation of this study is that participants tended to have been ill for long periods, and findings may not reflect word use in individuals with shorter durations of psychosis. To address these limitations, In the future, researchers should also seek to integrate lexical analysis with other types of speech instruments, such as latent semantic analysis and discourse incoherence measures, both of which have shown promise for investigating speech across the schizophrenia-spectrum (Docherty, 2012;Elvevag et al, 2007;Minor and Cohen, 2010;Minor and Cohen, 2012).…”

Quantitative approaches to content analysis: identifying conceptual drift across publication outlets

“…Computational approaches have been applied to phenomena in Management, Sociology, Marketing, Health, and other disciplines -for example, in literature indexing (Foltz, 1995), evaluation of medical interventions and diagnoses (Elvevåg et al, 2007;Al Qenaei, 2009), and language-based communication (Dong, 2005). Still, these applications were restricted to one analysis approach in isolation, leading to a gap in knowledge about the similarities and differences in the examination of data between different analytical techniques.…”

Detecting order–disorder transitions in discourse: Implications for schizophrenia

“…In order to establish an even better comparison and analyze further, we examined speech samples generated in response to the question "What activities do people generally do during the course of the day? ", (from Elvevåg et al, 2007) from a healthy participant (Sample 2), and three patients with schizophrenia (Samples 3 to 5). The responses were rated by two human raters for coherence (a score of 1 = very coherence versus 7 = very incoherent) and tangentiality (a score of 1 = very incisively related to question versus 7 = completely unrelated to question).…”

“…Indeed, such studies -using Latent Semantic Analysis (LSA) which models and matches discourse content (Landauer and Dumais, 1997;Landauer et al, 2007) -have demonstrated that it is possible to evaluate patients with schizophrenia based on open-ended verbalizations. These automatically derived language scores have distinguished patients from controls accurately (and patients from other patients, and also from their family members), using both large discourse samples as well as responses consisting of only a few words (Elvevåg et al, 2007(Elvevåg et al, , 2010.…”