Speech Prosody as a Bridge Between Psychopathology and Linguistics: The Case of the Schizophrenia Spectrum

The contribution of prosody to machine classification of schizophrenia

We show how acoustic prosodic features, such as pitch and gaps, can be used computationally for detecting symptoms of schizophrenia from a single spoken response. We compare the individual contributions of acoustic and previously-employed text modalities to the algorithmic determination whether the speaker has schizophrenia. Our classification results clearly show that we can extract relevant acoustic features better than those textual ones. We find that, when combined with those acoustic features, textual features improve classification only slightly.

Sex related differences in the perception and production of emotional prosody in adults

This study aimed to investigate the features of sex-related emotional prosody production patterns and perception abilities in adult speakers. The study involved 42 native Turkish speakers (27 females and 15 males). Sex-related perception and production of the emotions "anger," "joy," "sadness," and "neutral" were examined. Participants were first asked to indicate the actor's emotional state by selecting one of the given emotion alternatives provided. They were then instructed to produce the same stimuli with varying emotions. We analyzed the change in voice characteristics employed in different emotions in terms of F0 (Hz), speaking rate (seconds), and intensity (dB) using pairwise emotion comparison. The findings showed no sex differences in emotional prosody perceptions (p = 0.725). However, differences in the production of emotional prosody between sex have been documented in pitch variation of speech. Within-group analyses revealed that women tended to use a higher pitch when expressing joy versus sadness and a neutral state of feeling. Both men and women exhibited varying loudness levels for different emotional states in the speech loudness analysis. When expressing sadness, both men and women speak slower than when expressing as contrasted to anger, joy, or neutral states of feeling. Although Turkish speakers' ability to perceive emotional prosody is similar to that of other languages, they favor speech loudness fluctuation in the production of emotional prosody.

Prosody and schizophrenia. Objective acoustic measurements of monotonous and flat intonation in young Danish people with a schizophrenia diagnosis. A pilot study

PURPOSE

Patients with schizophrenia have a flat and monotonous intonation. The purpose of the study was to find the variables of flat speech that differed in patients from those in healthy controls in Danish.

MATERIALS AND METHODS

We compared drug-naïve schizophrenic patients 5 men, 13 women and 18 controls, aged 18-35 years, which had all grown up in Copenhagen speaking modern Danish standard (rigsdansk). We used two different tasks that lay different demands on the speaker to elicit spontaneous speech: a retelling of a film clip and telling a story from pictures in a book. A linguist used the computer program Praat to extract the phonetic linguistic parameters.

RESULTS

We found different results for the two elicitation tasks (Task 1: a retelling of a film clip, task 2: telling a story from pictures in a book). There was higher intensity variation in task one in controls and higher pitch variation in task two in controls. We found a difference in intensity with higher intensity variation in the stresses in the controls in task one and fewer syllables between each stress in the controls. We also found higher F1 variation in task one and two in the patient group and higher F2 variation in the control group in both tasks.

CONCLUSIONS

The results varied between patients and controls, but the demands also made a difference. Further research is needed to elucidate the possibilities of acoustic measures in diagnostics or linguistic treatment related to schizophrenia.

Neurophysiological explorations across the spectrum of psychosis, autism, and depression, during wakefulness and sleep: protocol of a prospective case-control transdiagnostic multimodal study (DEMETER)

BACKGROUND

Quantitative electroencephalography (EEG) analysis offers the opportunity to study high-level cognitive processes across psychiatric disorders. In particular, EEG microstates translate the temporal dynamics of neuronal networks throughout the brain. Their alteration may reflect transdiagnostic anomalies in neurophysiological functions that are impaired in mood, psychosis, and autism spectrum disorders, such as sensorimotor integration, speech, sleep, and sense of self. The main questions this study aims to answer are as follows: 1) Are EEG microstate anomalies associated with clinical and functional prognosis, both in resting conditions and during sleep, across psychiatric disorders? 2) Are EEG microstate anomalies associated with differences in sensorimotor integration, speech, sense of self, and sleep? 3) Can the dynamic of EEG microstates be modulated by a non-drug intervention such as light hypnosis?

METHODS

This prospective cohort will include a population of adolescents and young adults, aged 15 to 30 years old, with ultra-high-risk of psychosis (UHR), first-episode psychosis (FEP), schizophrenia (SCZ), autism spectrum disorder (ASD), and major depressive disorder (MDD), as well as healthy controls (CTRL) (N = 21 × 6), who will be assessed at baseline and after one year of follow-up. Participants will undergo deep phenotyping based on psychopathology, neuropsychological assessments, 64-channel EEG recordings, and biological sampling at the two timepoints. At baseline, the EEG recording will also be coupled to a sensorimotor task and a recording of the characteristics of their speech (prosody and turn-taking), a one-night polysomnography, a self-reference effect task in virtual reality (only in UHR, FEP, and CTRL). An interventional ancillary study will involve only healthy controls, in order to assess whether light hypnosis can modify the EEG microstate architecture in a direction opposite to what is seen in disease.

DISCUSSION

This transdiagnostic longitudinal case-control study will provide a multimodal neurophysiological assessment of clinical dimensions (sensorimotor integration, speech, sleep, and sense of self) that are disrupted across mood, psychosis, and autism spectrum disorders. It will further test the relevance of EEG microstates as dimensional functional biomarkers.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier NCT06045897.

Markers of schizophrenia at the prosody/pragmatics interface. Evidence from corpora of spontaneous speech interactions

The speech of individuals with schizophrenia exhibits atypical prosody and pragmatic dysfunctions, producing monotony. The paper presents the outcomes of corpus-based research on the prosodic features of the pathology as they manifest in real-life spontaneous interactions. The research relies on a corpus of schizophrenic speech recorded during psychiatric interviews (CIPPS) compared to a sampling of non-pathological speech derived from the LABLITA corpus of spoken Italian, which has been selected according to comparability requirements. Corpora has been intensively analyzed in the Language into Act Theory (L-AcT) frame, which links prosodic cues and pragmatic values. A cluster of linguistic parameters marked by prosody has been considered: utterance boundaries, information structure, speech disfluency, and prosodic prominence. The speech flow of patients turns out to be organized into small chunks of information that are shorter and scarcely structured, with an atypical proportion of post-nuclear information units (Appendix). It is pervasively scattered with silences, especially with long pauses between utterances and long silences at turn-taking. Fluency is hindered by retracing phenomena that characterize complex information structures. The acoustic parameters that give rise to prosodic prominence (f0 mean, f0 standard deviation, spectral emphasis, and intensity variation) have been measured considering the pragmatic roles of the prosodic units, distinguishing prominences within the illocutionary units (Comment) from those characterizing Topic units. Patients show a flattening of the Comment-prominence, reflecting impairments in performing the illocutionary activity. Reduced values of spectral emphasis and intensity variation also suggest a lack of engagement in communication. Conversely, Topic-prominence shows higher values for f0 standard deviation and spectral emphasis, suggesting effort when defining the domain of relevance of the illocutionary force. When comparing Topic and Comment-prominences of patients, the former consistently exhibit higher values across all parameters. In contrast, the non-pathological group displays the opposite pattern.

Syntactic and affective prosody recognition: Schizophrenia vs. Autism spectrum disorders

Patients with a recent diagnosis of schizophrenia and individuals receiving a diagnosis of autism spectrum disorder without accompanying intellectual impairment (ASD w/o intellectual impairment) during their adulthood share several clinical characteristics. Exploring under-investigated aspects of these two clinical conditions may shed light on their possible connection and facilitate differential diagnosis at very early stages. To this end, we explored the ability of 15 adults with a recent diagnosis of schizophrenia, 15 individuals diagnosed with ASD w/o intellectual impairment as adults, and 15 healthy adults to resolve sentence ambiguities with the use of syntactic prosody, and to decode happiness, anger, sadness, surprise, fear, and neutrality based on affective prosody. Results revealed intact perception of syntactic prosody in adults with schizophrenia, but impaired affective prosody recognition, which could be attributed, however, to emotion processing difficulties overall. On the other hand, individuals with ASD w/o intellectual impairment were impaired on prosody comprehension per se, as evidenced in the most challenging conditions, namely the subject-reading condition and the emotion of surprise. The differences in prosody comprehension ability between the two clinical conditions may serve as an indicator, among other signs, during the diagnostic evaluation.

Evaluating the clinical utility of speech analysis and machine learning in schizophrenia: A pilot study

BACKGROUND

Schizophrenia is a serious mental disorder that significantly impacts social functioning and quality of life. However, current diagnostic methods lack objective biomarker support. While some studies have indicated differences in audio features between patients with schizophrenia and healthy controls, these findings are influenced by demographic information and variations in experimental paradigms. Therefore, it is crucial to explore stable and reliable audio biomarkers for an auxiliary diagnosis and disease severity prediction of schizophrenia.

METHOD

A total of 130 individuals (65 patients with schizophrenia and 65 healthy controls) read three fixed texts containing positive, neutral, and negative emotions, and recorded them. All audio signals were preprocessed and acoustic features were extracted by a librosa-0.9.2 toolkit. Independent sample t-tests were performed on two sets of acoustic features, and Pearson correlation on the acoustic features and Positive and Negative Syndrome Scale (PANSS) scores of the schizophrenia group. Classification algorithms in scikit-learn were used to diagnose schizophrenia and predict the level of negative symptoms.

RESULTS

Significant differences were observed between the two groups in the mfcc_8, mfcc_11, and mfcc_33 of mel-frequency cepstral coefficient (MFCC). Furthermore, a significant correlation was found between mfcc_7 and the negative PANSS scores. Through acoustic features, we could not only differentiate patients with schizophrenia from healthy controls with an accuracy of 0.815 but also predict the grade of the negative symptoms in schizophrenia with an average accuracy of 0.691.

CONCLUSIONS

The results demonstrated the considerable potential of acoustic characteristics as reliable biomarkers for diagnosing schizophrenia and predicting clinical symptoms.

Disembodiment and Language in Schizophrenia: An Integrated Psychopathological and Evolutionary Perspective

Different hypotheses have flourished to explain the evolutionary paradox of schizophrenia. In this contribution, we sought to illustrate how, in the schizophrenia spectrum, the concept of embodiment may underpin the phylogenetic and developmental pathways linking sensorimotor processes, the origin of human language, and the construction of a basic sense of the self. In particular, according to an embodied model of language, we suggest that the reuse of basic sensorimotor loops for language, while enabling the development of fully symbolic thought, has pushed the human brain close to the threshold of a severe disruption of self-embodiment processes, which are at the core of schizophrenia psychopathology. We adopted an inter-disciplinary approach (psychopathology, neuroscience, developmental biology) within an evolutionary framework, to gain an integrated, multi-perspectival model on the origin of schizophrenia vulnerability. A maladaptive over-expression of evolutionary-developmental trajectories toward language at the expense of embodiment processes would have led to the evolutionary "trade-off" of a hyper-symbolic activity to the detriment of a disembodied self. Therefore, schizophrenia psychopathology might be the cost of long-term co-evolutive interactions between brain and language.

Conversational metrics, psychopathological dimensions and self-disturbances in patients with schizophrenia

Difficulties in interpersonal communication, including conversational skill impairments, are core features of schizophrenia. However, very few studies have performed conversation analyses in a clinical population of schizophrenia patients. Here we investigate the conversational patterns of dialogues in schizophrenia patients to assess possible associations with symptom dimensions, subjective self-disturbances and social functioning. Thirty-five schizophrenia patients were administered the Positive and Negative Syndrome Scale (PANSS), the Clinical Language Disorder Rating Scale (CLANG), the Scale for the Assessment of Thought, Language and Communication (TLC), the Examination of Anomalous Self-Experience Scale (EASE), and the Social and Occupational Functioning Assessment Scale (SOFAS). Moreover, participants underwent a recorded semi-structured interview, to extract conversational variables. Conversational data were associated with negative symptoms and social functioning, but not with positive or disorganization symptoms. A significant positive correlation was found between "pause duration" and the EASE item "Spatialization of thought". The present study suggests an association between conversational patterns and negative symptom dimension of schizophrenia. Moreover, our findings evoke a relationship between the natural fluidity of conversation and of the natural unraveling of thoughts.

A Theoretical Framework for Human and Nonhuman Vocal Interaction

Vocal communication is a critical feature of social interaction across species; however, the relation between such behavior in humans and nonhumans remains unclear. To enable comparative investigation of this topic, we review the literature pertinent to interactive language use and identify the superset of cognitive operations involved in generating communicative action. We posit these functions comprise three intersecting multistep pathways: (a) the Content Pathway, which selects the movements constituting a response; (b) the Timing Pathway, which temporally structures responses; and (c) the Affect Pathway, which modulates response parameters according to internal state. These processing streams form the basis of the Convergent Pathways for Interaction framework, which provides a conceptual model for investigating the cognitive and neural computations underlying vocal communication across species.

Reading Warm-Up, Reading Skill, and Reading Prosody When Reading the My Grandfather Passage: An Exploratory Study Born Out of the Motor Planning Theory of Prosody and Reading Prosody Research

PURPOSE

The Motor Planning Theory of Prosody and reading prosody research indicate that "out of the blue" oral reading, as practiced in clinical and research settings, invokes surface rather than covert prosody, particularly when readers are recorded, less skilled, and/or speech impaired. Warm-up is not considered in passage reading for motor-speech assessment. We report on a preliminary study aimed to investigate the effect of warm-up on reading prosody in two conditions: silent reading alone and reading "out of the blue" followed by silent reading. A secondary aim of the study was to examine the effect of reading skill on reading prosody.

METHOD

Twenty-one monolingual, English-speaking volunteers were recorded reading the My Grandfather Passage (GP) while their eye movements were tracked. Participants were randomly assigned to one of two reading conditions: (a) silent-oral (SO) and (b) oral-silent-oral (OSO). In the SO condition, participants read the GP silently as a warm-up for the subsequent oral reading. In the OSO condition, participants first read the GP aloud ("out of the blue") and then read the same passage silently with the instruction to do this in preparation for a second oral reading. Reading skill was quantified using eye-voice span and Wide Range Achievement Test-Fourth Edition testing. Reading prosody was evaluated using pause indexes, the Acoustic Multidimensional Prosody Index, and speech rate.

CONCLUSIONS

One oral reading before a silent reading but not a silent reading alone before oral reading was shown to affect reading prosody. In terms of reading skill, results indicate that predictive associations patterned differently in the reading conditions explored, suggesting different underlying skill sets.

It is time to address language disorders in schizophrenia: A RCT on the efficacy of a novel training targeting the pragmatics of communication (PragmaCom)

INTRODUCTION

Language and communication disruptions in schizophrenia are at the center of a large body of investigation. Yet, the remediation of such disruptions is still in its infancy. Here we targeted what is known to be one of the most damaged language domains in schizophrenia, namely pragmatics, by conducting a pragmatics-centered intervention with a randomized controlled trial design and assessing also durability and generalization. To the best of our knowledge, this is the first study with these characteristics.

METHODS

Inspired by the Gricean account of natural language use, we tailored a novel treatment addressing the pragmatics of communication (PragmaCom) and we tested its efficacy in a sample of individuals with schizophrenia randomized to the experimental group or to an active control group. The primary outcome with respect to the efficacy of the PragmaCom was measured by changes in pragmatic abilities (as evaluated with the global score of the Assessment of Pragmatic Abilities and Cognitive Substrates test) from baseline to 12 weeks and at 3-month follow-up. The secondary outcome was measured by changes in metaphor comprehension, abstract thinking, and global functioning from baseline to 12 weeks and at 3-month follow-up.

RESULTS

Relative to the control group, at post-test the PragmaCom group showed greater and enduring improvement in global pragmatic skills and in metaphor comprehension. At follow-up, these improvements persisted and the PragmaCom exerted beneficial effects also on functioning.

CONCLUSIONS

Despite the limited sample size, we believe that these findings offer initial yet encouraging evidence of the possibility to improve pragmatic skills with a theoretically grounded approach and to obtain durable and clinically relevant benefits. We argue that it is time that therapeutic efforts embrace communicative dysfunctions in order to improve illness outcome.