Latent Factors of Language Disturbance and Relationships to Quantitative Speech Features

Abstract

Background and Hypothesis Quantitative acoustic and textual measures derived from speech ("speech features") may provide valuable biomarkers for psychiatric disorders, particularly schizophrenia spectrum disorders (SSD). We sought to identify cross-diagnostic latent factors for speech disturbance with relevance for SSD and computational modeling. Study Design Clinical ratings for speech disturbance were generated across 14 items for a cross-diagnostic sample (N=343), including SSD (n=97). Speech features were quantified using an automated pipeline for brief recorded samples of free-speech. Factor models for the clinical ratings were generated using exploratory factor analysis, then tested with confirmatory factor analysis in the cross-diagnostic and SSD groups. Relationships among factor scores, speech features and other clinical characteristics were examined using network analysis. Study Results We found a 3-factor model with good fit in the cross-diagnostic group and acceptable fit for the SSD subsample. The model identifies an impaired expressivity factor and two interrelated disorganized factors for inefficient and incoherent speech. Incoherent speech was specific to psychosis groups, while inefficient speech and impaired expressivity showed intermediate effects in people with nonpsychotic disorders. Network analysis showed that the factors had distinct relationships with speech features, and that the patterns were different in the cross-diagnostic versus SSD groups. Conclusions We report a cross-diagnostic 3-factor model for speech disturbance which is supported by good statistical measures, intuitive, applicable to SSD, and relatable to linguistic theories. It provides a valuable framework for understanding speech disturbance and appropriate targets for modeling with quantitative speech features.

Voice patterns as markers of schizophrenia: building a cumulative generalizable approach via a cross-linguistic and meta-analysis based investigation

“…This multilingual corpus may represent an ideal benchmark dataset for testing the reliability and generalizability (e.g., out of sample predictability) of voice analysis results in schizophrenia 73 , and the necessary ground for assessing its clinical applicability 74 . Not least, future studies should focus more on cross-diagnostic comparisons aimed at capturing symptom dimensions which extend over a single disorder 75,76 , and implement longitudinal designs able to test more complex hypothesis on the interaction between antipsychotic medication type and dosage, clinical (e.g., illness severity and duration) and sociodemographic (e.g., sex differences 77 ) characteristics, and speech production.…”

Language Network Dysfunction and Formal Thought Disorder in Schizophrenia

“…Another challenge is that many studies seeking neural correlates of FTD have relied on either single item measurement of conceptual disorganisation or a composite score of FTD based on multiple items [18]. Following on the initial dichotomization of FTD by Fish (see [26] for an overview), factor analytic studies indicate that at least two (positive/disorganisation and negative/impoverishment), if not more [27], separable latent factors may account for these items [28]. Importantly, these factors appear to have distinct trajectories and treatment responsiveness [29,30], making it critical to parse these multiple factors and examine their neural correlates separately (the collapsed dimensions issue).…”

Automated, Objective Speech and Language Markers of Longitudinal Changes in Psychosis Symptoms

“…These methods are highly sensitive (8) and consistently predict schizophrenia diagnosis relative to healthy controls, as well as conversion to psychosis among individuals at clinical high risk (9). Different types of speech and language features are also sensitive to different dimensions of psychosis symptoms, cognition, and functioning (10)(11)(12)(13). One study used a range of lexical, coherence, and disfluency features to longitudinally estimate psychosis symptoms in 38 participants with psychotic disorders (99 total sessions) and found promising between-and withinparticipant relationships to positive and negative symptoms (14).…”