Perplexity of utterances in untreated first-episode psychosis: an ultra-high field MRI dynamic causal modelling study of the semantic network.

BACKGROUND: Psychosis involves a distortion of thought content, which is partly reflected in anomalous ways in which words are semantically connected into utterances in speech. We sought to explore how these linguistic anomalies are realized through putative circuit-level abnormalities in the brain's semantic network. METHODS: Using a computational large-language model, Bidirectional Encoder Representations from Transformers (BERT), we quantified the contextual expectedness of a given word sequence (perplexity) across 180 samples obtained from descriptions of 3 pictures by patients with first-episode schizophrenia (FES) and controls matched for age, parental social status, and sex, scanned with 7 T ultra-high field functional magnetic resonance imaging (fMRI). Subsequently, perplexity was used to parametrize a spectral dynamic causal model (DCM) of the effective connectivity within (intrinsic) and between (extrinsic) 4 key regions of the semantic network at rest, namely the anterior temporal lobe, the inferior frontal gyrus (IFG), the posterior middle temporal gyrus (MTG), and the angular gyrus. RESULTS: We included 60 participants, including 30 patients with FES and 30 controls. We observed higher perplexity in the FES group, indicating that speech was less predictable by the preceding context among patients. Results of Bayesian model comparisons showed that a DCM including the group by perplexity interaction best explained the underlying patterns of neural activity. We observed an increase of self-inhibitory effective connectivity within the IFG, as well as reduced self-inhibitory tone within the pMTG, in the FES group. An increase in self-inhibitory tone in the IFG correlated strongly and positively with inter-regional excitation between the IFG and posterior MTG, while self-inhibition of the posterior MTG was negatively correlated with this interregional excitation. LIMITATION: Our design did not address connectivity in the semantic network during tasks that selectively activated the semantic network, which could corroborate findings from this resting-state fMRI study. Furthermore, we do not present a replication study, which would ideally use speech in a different language. CONCLUSION: As an explanation for peculiar speech in psychosis, these results index a shift in the excitatory-inhibitory balance regulating information flow across the semantic network, confined to 2 regions that were previously linked specifically to the executive control of meaning. Based on our approach of combining a large language model with causal connectivity estimates, we propose loss in semantic control as a potential neurocognitive mechanism contributing to disorganization in psychosis.

Morphometry and functional connectivity of auditory cortex in school-age children with profound language disabilities: Five comparative case studies.

Many neurodevelopmental conditions imply absent or severely reduced language capacities at school age. Evidence from functional magnetic resonance imaging is highly limited. We selected a series of five cases scanned with the same fMRI paradigm and the aim of relating individual language profiles onto underlying patterns of functional connectivity (FC) across auditory language cortex: three with neurogenetic syndromes (Coffin-Siris, Landau-Kleffner, and Fragile-X), one with idiopathic intellectual disability, one with autism spectrum disorder (ASD). Compared to both a group with typical development (TD) and a verbal ASD group (total N = 110), they all showed interhemispheric FC below two standard deviations of the TD mean. Children with higher language scores showed higher intrahemispheric FC between Heschl's gyrus and other auditory language regions, as well as an increase of FC during language stimulation compared to rest. An increase of FC in forward vs. reversed speech in the posterior and middle temporal gyri was seen across all cases. The Coffin-Siris case, the most severe, also had the most anomalous FC patterns and showed reduced myelin content, while the Landau-Kleffner case showed reduced cortical thickness. These results suggest potential for neural markers and mechanisms of severe language processing deficits under highly heterogeneous etiological conditions.

Comprehension of Embedded Clauses in Schizophrenia With and Without Formal Thought Disorder.

Formal thought disorder (FTD) in schizophrenia (SZ) is clinically manifested primarily through language production, where linguistic studies have reported numerous anomalies including lesser use of embedded clauses. Here, we explored whether problems of language may extend to comprehension and clause embedding in particular. A sentence-picture matching task was designed with two conditions in which embedded clauses were presupposed as either true (factive) or not. Performance across these two conditions was compared in people with SZ and moderate-to-severe FTD (SZ + FTD), SZ with minimal FTD (SZ-FTD), first-degree relatives of people with SZ, and neurotypical controls. The SZ + FTD group performed significantly worse than all others in both conditions, and worse in the nonfactive than in the factive one. These results demonstrate language dysfunction in comprehension specific to FTD is a critical aspect of grammatical complexity and its associated meaning, which has been independently known to be cognitively significant as well.

Language in schizophrenia and aphasia: the relationship with non-verbal cognition and thought disorder.

Objective: To determine the relationship between language abnormalities and broader cognitive impairment and thought disorder by examining language and cognition in schizophrenia and aphasia (a primary language disorder).Methods: Cognitive and linguistic profiles were measured with a battery of standardised tests, and compared in a clinical population of n = 50 (n = 30 with schizophrenia and n = 20 with aphasia) and n = 61 non-clinical comparisons (n = 45 healthy controls and n = 16 non-affected first-degree relatives of patients with schizophrenia).Results: Both clinical groups showed linguistic deficits. Verbal impairment was more severe in participants with aphasia, whereas non-verbal performance was more affected in participants with schizophrenia. In schizophrenia, but not in aphasia, verbal and non-verbal performance were associated. Formal thought disorder was associated with impairment in executive function and in grammatical, but not naming, tasks.Conclusion: While patients with schizophrenia and aphasia showed language impairments, the nature and cognitive basis of these impairments may be different; language performance disassociates from broader cognitive functioning in aphasia but may be an intrinsic expression of a broader cognitive impairment in schizophrenia. Thought disorder may represent a core malfunction of grammatical processing. Results suggests that communicative ability may be a valid target in cognitive remediation strategies in schizophrenia.

Toward true integration.

Whether in sign or speech, language is more integrative than the target article suggests. A more integrative view embraces not only sign/speech and co-sign/speech gesture, but also indicative gestures irrespective of modality, and locations along with movements in the signed modality, as suggested by both linguistic acquisition and pathologies. An extended integrative view also proves advantageous in terms of conceptual coherence.

Can delusions be understood linguistically?

Delusions are widely believed to reflect disturbed cognitive function, but the nature of this remains elusive. The "un-Cartesian" cognitive-linguistic hypothesis maintains (a) that there is no thought separate from language, that is, there is no distinct mental space removed from language where "thinking" takes place; and (b) that a somewhat broadened concept of grammar is responsible for bestowing meaning on propositions, and this among other things gives them their quality of being true or false. It is argued that a loss of propositional meaning explains why delusions are false, impossible and sometimes fantastic. A closely related abnormality, failure of linguistic embedding, can additionally account for why delusions are held with fixed conviction and are not adequately justified by the patient. The un-Cartesian linguistic approach to delusions has points of contact with Frith's theory that inability to form meta-representations underlies a range of schizophrenic symptoms. It may also be relevant to the nature of the "second factor" in monothematic delusions in neurological disease. Finally, it can inform the current debate about whether or not delusions really are beliefs.

The 'L-factor': Language as a transdiagnostic dimension in psychopathology.

Thoughts and moods constituting our mental life incessantly change. When the steady flow of this dynamics diverges in clinical directions, the possible pathways involved are captured through discrete diagnostic labels. Yet a single vulnerable neurocognitive system may be causally involved in psychopathological deviations transdiagnostically. We argue that language viewed as integrating cortical functions is the best current candidate, whose forms of breakdown along its different dimensions are then manifest as symptoms - from prosodic abnormalities and rumination in depression to distortions of speech perception in verbal hallucinations, distortions of meaning and content in delusions, or disorganized speech in formal thought disorder. Spontaneous connected speech provides continuous objective readouts generating a highly accessible bio-behavioral marker with the potential of revolutionizing neuropsychological measurement. This argument turns language into a transdiagnostic 'L-factor' providing an analytical and mechanistic substrate for previously proposed latent general factors of psychopathology ('p-factor') and cognitive functioning ('c-factor'). Together with immense practical opportunities afforded by rapidly advancing natural language processing (NLP) technologies and abundantly available data, this suggests a new era of translational clinical psychiatry, in which both psychopathology and language may be rethought together.

Episodic Thinking in Alzheimer's Disease Through the Lens of Language: Linguistic Analysis and Transformer-Based Classification.

PURPOSE: Episodic memory decline is a hallmark of Alzheimer's disease (AD) and linked to deficits in episodic thinking directed to the future. We addressed the question whether a deficit in episodic thinking can be picked up directly from connected speech and its detection can be automatized. METHOD: We linguistically classified 2,809 utterances (including embedded clauses in the utterances) from picture descriptions from 70 healthy older controls, 82 people with mild probable AD (pAD), and 46 people with moderate pAD for whether they were episodic, nonepisodic, or "other" (e.g., off-task). Generalized linear regression models were used to investigate how ratios of these categories change in AD, controlling for age, gender, and education. Finally, we applied deep learning technique to explore the feasibility of automating the episodicity analysis. RESULTS: Decline in episodicity significantly distinguished controls from both mild pAD and moderate pAD. Correlation analysis suggested this decline not to be an effect of age, gender, and education but of cognitive ability. The decline was not compensated by an increase of nonepisodic utterances but mainly of off-task expressions. A transformer-based classifier to explore the possibility of automatizing the classification of episodicity achieved a macro F1 score of 0.913 in the ternary classification. CONCLUSION: These results show that a loss of episodicity is an early effect in AD that is manifested in spontaneous speech and can be reliably measured by both humans and machines.

Navigating the semantic space: Unraveling the structure of meaning in psychosis using different computational language models.

Speech in psychosis has long been ascribed as involving 'loosening of associations'. We pursued the aim to elucidate its underlying cognitive mechanisms by analysing picture descriptions from 94 subjects (29 healthy controls, 18 participants at clinical high risk, 29 with first-episode psychosis, and 18 with chronic schizophrenia), using five language models with different computational architectures: FastText, which represents meaning non-contextually/statically; BERT, which represents contextual meaning sensitive to grammar and context; Infersent and SBERT, which provide sentential representations; and CLIP, which evaluates speech relative to a visual stimulus. These models were used to quantify semantic distances crossed between successive tokens/sentences, and semantic perplexity indicating unexpectedness in continuations. Results showed that, among patients, semantic similarity increased when measured with FastText, Infersent, and SBERT, while it decreased with CLIP and BERT. Higher perplexity was observed in first-episode psychosis. Static semantic measures were associated with clinically measured impoverishment of thought and referential semantic measures with disorganization. These patterns indicate a shrinking conceptual semantic space as represented by static language models, which co-occurs with a widening in the referential semantic space as represented by contextual models. This duality underlines the need to separate these two forms of meaning for understanding mechanisms involved in semantic change in psychosis.

Functional dysregulation of the auditory cortex in bilateral perisylvian polymicrogyria: Multiparametric case analysis of the absent speech phenotype.

The absence of speech is a clinical phenotype seen across neurodevelopmental syndromes, offering insights for neural language models. We present a case of bilateral perisylvian polymicrogyria (BPP) and complete absence of speech with considerable language comprehension and production difficulties. We extensively characterized the auditory speech perception and production circuitry by employing a multimodal neuroimaging approach. Results showed extensive cortical thickening in motor and auditory-language regions. The auditory cortex lacked sensitivity to speech stimuli despite relatively preserved thalamic projections yet had no intrinsic functional organization. Subcortical structures implicated in early stages of processing exhibited heightened sensitivity to speech. The arcuate fasciculus, a suggested marker of language in BPP, showed similar volume and integrity to a healthy control. The frontal aslant tract, linked to oromotor function, was partially reconstructed. These findings highlight the importance of assessing the auditory cortex beyond speech production structures to understand absent speech in BPP. Despite profound cortical alterations, the intrinsic motor network and motor-speech pathways remained largely intact. This case underscores the need for comprehensive phenotyping using multiple MRI modalities to uncover causes of severe disruption in language development.

Atypical cortical hierarchy in Aß-positive older adults and its reflection in spontaneous speech.

Abnormal deposition of Aß amyloid is an early neuropathological marker of Alzheimer's disease (AD), arising long ahead of clinical symptoms. Non-invasive measures of associated early neurofunctional changes, together with easily accessible behavioral readouts of these changes, could be of great clinical benefit. We pursued this aim by investigating large-scale cortical gradients of functional connectivity with functional MRI, which capture the hierarchical integration of cortical functions, together with acoustic-prosodic features from spontaneous speech, in cognitively unimpaired older adults with and without Aß positivity (total N = 188). We predicted distortions of the cortical hierarchy associated with prosodic changes in the Aß + group. Results confirmed substantially altered cortical hierarchies and less variability in these in the Aß + group, together with an increase in quantitative prosodic measures, which correlated with gradient variability as well as digit span test scores. Overall, these findings confirm that long before the clinical stage and objective cognitive impairment, increased risk of cognitive decline as indexed by Aß accumulation is marked by neurofunctional changes in the cortical hierarchy, which are related to automatically extractable speech patterns and alterations in working memory functions.

Your theory of the evolution of morality depends upon your theory of morality.

Baumard et al. attribute to humans a sense of fairness. However, the properties of this sense are so underspecified that the evolutionary account offered is not well-motivated. We contrast this with the framework of Universal Moral Grammar, which has sought a descriptively adequate account of the structure of the moral domain as a precondition for understanding the evolution of morality.

Coreference Delays in Psychotic Discourse: Widening the Temporal Window.

BACKGROUND AND HYPOTHESIS: Any form of coherent discourse depends on saying different things about the same entities at different times. Such recurrent references to the same entity need to predictably happen within certain temporal windows. We hypothesized that a failure of control over reference in speakers with schizophrenia (Sz) would become manifest through dynamic temporal measures. STUDY DESIGN: Conversational speech with a mean of 909.2 words (SD: 178.4) from 20 Chilean Spanish speakers with chronic Sz, 20 speakers at clinical high risk (CHR), and 20 controls were collected. Using directed speech graphs with referential noun phrases (NPs) as nodes, we studied deviances in the topology and temporal distribution of such NPs and of the entities they denote over narrative time. STUDY RESULTS: The Sz group had a larger density of NPs (number of NPs divided by total words) relative to both controls and CHR. This related to topological measures of distance between recurrent entities, which revealed that the Sz group produced more recurrences, as well as greater topological distances between them, relative to controls. A logistic regression using five topological measures showed that Sz and controls can be distinguished with 84.2% accuracy. CONCLUSIONS: This pattern indicates a widening of the temporal window in which entities are maintained in discourse and co-referenced in it. It substantiates and extends earlier evidence for deficits in the cognitive control over linguistic reference in psychotic discourse and informs both neurocognitive models of language in Sz and machine learning-based linguistic classifiers of psychotic speech.

Linguistic markers of autism spectrum conditions in narratives: A comprehensive analysis.

Background & Aims: Narratives are regularly elicited as part of standardized assessments for autism spectrum conditions (ASC) such as the ADOS, but have rarely been utilized as linguistic data in their own right. We here aimed for a specific and comprehensive quantitative linguistic profile of such narratives across nominal, verbal, and clausal domains of grammatical organization, and error patterns. Methods: We manually transcribed and annotated narratives elicited from the ADOS from a sample of bilingual autistic Spanish-Catalan children (n = 18), matched with typically developing controls (n = 18) on vocabulary-based verbal IQ. Results: Results revealed fewer relative clauses and more frequent errors in referential specificity and non-relational content-word choice in ASC. Frequent error types are also discussed qualitatively. Conclusions & Implications: These findings, based on more finegrained linguistically defined variables, help to disentangle previous inconsistencies in the literature, and to better situate language changes in the spectrum of neurocognitive changes in this population.

Linguistic markers of psychosis in Mandarin Chinese: Relations to theory of mind.

Disorganized and impoverished language is a key feature of schizophrenia (Sz), but whether and which linguistic changes previously observed in Indo-European languages generalize to other languages remains unclear. Targeting Mandarin Chinese, we aimed to profile aspects of grammatical complexity that we hypothesized would be reduced in schizophrenia in a task of verbalizing social events. 51 individuals with Sz and 39 controls participated in the animated triangles task, a standardized measure of theory of mind (ToM), in which participants describe triangles moving in either a random or an 'intentional' condition. Results revealed that clauses embedded as arguments in other clauses were reduced in Sz, and that both groups produced such clauses and grammatical aspect more frequently in the intentional condition. ToM scores specifically correlated with production of embedded argument clauses. These results document grammatical impoverishment in Sz in Chinese across several structural domains, which in some of its specific aspects relate to mentalizing performance.

Automated Classification of Cognitive Decline and Probable Alzheimer's Dementia Across Multiple Speech and Language Domains.

BACKGROUND: Decline in language has emerged as a new potential biomarker for the early detection of Alzheimer's disease (AD). It remains unclear how sensitive language measures are across different tasks, language domains, and languages, and to what extent changes can be reliably detected in early stages such as subjective cognitive decline (SCD) and mild cognitive impairment (MCI). METHOD: Using a scene construction task for speech elicitation in a new Spanish/Catalan speaking cohort (N = 119), we automatically extracted features across seven domains, three acoustic (spectral, cepstral, and voice quality), one prosodic, and three from text (morpholexical, semantic, and syntactic). They were forwarded to a random forest classifier to evaluate the discriminability of participants with probable AD dementia, amnestic and nonamnestic MCI, SCD, and cognitively healthy controls. Repeated-measures analyses of variance and paired-samples Wilcoxon signed-ranks test were used to assess whether and how performance differs significantly across groups and linguistic domains. RESULTS: The performance scores of the machine learning classifier were generally satisfactorily high, with the highest scores over .9. Model performance was significantly different for linguistic domains (p < .001), and speech versus text (p = .043), with speech features outperforming textual features, and voice quality performing best. High diagnostic classification accuracies were seen even within both cognitively healthy (controls vs. SCD) and MCI (amnestic and nonamnestic) groups. CONCLUSION: Speech-based machine learning is powerful in detecting cognitive decline and probable AD dementia across a range of different feature domains, though important differences exist between these domains as well. SUPPLEMENTAL MATERIAL: https://doi.org/10.23641/asha.23699733.

Harnessing acoustic speech parameters to decipher amyloid status in individuals with mild cognitive impairment.

Alzheimer's disease (AD) is a neurodegenerative condition characterized by a gradual decline in cognitive functions. Currently, there are no effective treatments for AD, underscoring the importance of identifying individuals in the preclinical stages of mild cognitive impairment (MCI) to enable early interventions. Among the neuropathological events associated with the onset of the disease is the accumulation of amyloid protein in the brain, which correlates with decreased levels of Aß42 peptide in the cerebrospinal fluid (CSF). Consequently, the development of non-invasive, low-cost, and easy-to-administer proxies for detecting Aß42 positivity in CSF becomes particularly valuable. A promising approach to achieve this is spontaneous speech analysis, which combined with machine learning (ML) techniques, has proven highly useful in AD. In this study, we examined the relationship between amyloid status in CSF and acoustic features derived from the description of the Cookie Theft picture in MCI patients from a memory clinic. The cohort consisted of fifty-two patients with MCI (mean age 73 years, 65% female, and 57% positive amyloid status). Eighty-eight acoustic parameters were extracted from voice recordings using the extended Geneva Minimalistic Acoustic Parameter Set (eGeMAPS), and several ML models were used to classify the amyloid status. Furthermore, interpretability techniques were employed to examine the influence of input variables on the determination of amyloid-positive status. The best model, based on acoustic variables, achieved an accuracy of 75% with an area under the curve (AUC) of 0.79 in the prediction of amyloid status evaluated by bootstrapping and Leave-One-Out Cross Validation (LOOCV), outperforming conventional neuropsychological tests (AUC = 0.66). Our results showed that the automated analysis of voice recordings derived from spontaneous speech tests offers valuable insights into AD biomarkers during the preclinical stages. These findings introduce novel possibilities for the use of digital biomarkers to identify subjects at high risk of developing AD.

Natural Language Processing Markers for Psychosis and Other Psychiatric Disorders: Emerging Themes and Research Agenda From a Cross-Linguistic Workshop.

This workshop summary on natural language processing (NLP) markers for psychosis and other psychiatric disorders presents some of the clinical and research issues that NLP markers might address and some of the activities needed to move in that direction. We propose that the optimal development of NLP markers would occur in the context of research efforts to map out the underlying mechanisms of psychosis and other disorders. In this workshop, we identified some of the challenges to be addressed in developing and implementing NLP markers-based Clinical Decision Support Systems (CDSSs) in psychiatric practice, especially with respect to psychosis. Of note, a CDSS is meant to enhance decision-making by clinicians by providing additional relevant information primarily through software (although CDSSs are not without risks). In psychiatry, a field that relies on subjective clinical ratings that condense rich temporal behavioral information, the inclusion of computational quantitative NLP markers can plausibly lead to operationalized decision models in place of idiosyncratic ones, although ethical issues must always be paramount.

Grammar in 'agrammatical' aphasia: What's intact?

BACKGROUND: Aphasia following cerebro-vascular accidents has been a primary source of insight for models of language in the brain. However, deviant language patterns in aphasia may reflect processing limitations and cognitive impairment more than language impairment per se. AIMS: We sought to obtain new evidence from spontaneous speech in Broca's aphasia (BA) for the intactness of grammatical knowledge, operationalized as the preservation of the basic hierarchical structure of syntactic projections. METHODS & PROCEDURES: Speech obtained with the AphasiaBank protocol from 20 people with BA, which were independently rated as also being agrammatic, was analyzed and compared to 20 matched non-brain-damaged controls. We quantified (i) marking of Aspect, Tense, and Modality (A-T-M), which are located at specific (high) layers of the syntactic hierarchy and ordered in relation to one another ([M…[T…[A…]]]); (ii) hierarchies of clausal units ([C…[C]]); (iii) discourse markers embedding clauses, located at the highest layer of the hierarchy; and (iv) attachment of adjuncts at different heights of a given hierarchical syntactic structure. Supplementary evidence was obtained from a typology of errors and from pauses subcategorized according to their hierarchical syntactic position. OUTCOMES & RESULTS: Groups did not quantitatively differ on rates of either Aspect or Modality but underproduced T and embedded clauses. Evidence for compensatory effects was seen in both of the latter two cases. While all adjunct types were underproduced in BA, and pauses overproduced, both showed the same relative proportions within both groups. Errors were largely restricted to omissions, of a kind that would also be expected in condensed neurotypical speech. CONCLUSIONS: Overall, these patterns support the hypothesis of intactness of grammatical knowledge in BA clinically rated as agrammatic, questioning it as a disease model of language impairment.

Temporal Overlap Between Gestures and Speech in Poststroke Aphasia: Is There a Compensatory Effect?

PURPOSE: If language production is impaired, will gestures compensate? Evidence in favor of this prediction has often been argued to come from aphasia, but it remains contested. Here, we tested whether thought content not present in speech due to language impairment is manifested in gestures, in 20 people with dysfluent (Broca's) aphasia, 20 people with fluent (Wernicke's) aphasia, and 20 matched neurotypical controls. METHOD: A new annotation scheme was created distinguishing types of gestures and whether they co-occurred with fluent or dysfluent/absent speech and were temporally aligned in content with coproduced speech. RESULTS: Across both aphasia types, noncontent (beat) gestures, which by their nature cannot compensate for lost speech content, constituted the greatest proportion of all types of gestures produced. Content (i.e., descriptive, referential, and metaphorical) gestures were largely coproduced with fluent rather than dysfluent speech and tended to be aligned with the content conveyed in speech. They also did not differ in quantity depending on whether the dysfluencies were eventually resolved or not. Neither aphasia severity nor comprehension ability had an impact on the total amount of content gesture produced in people with aphasia, which was instead positively correlated with speech fluency. CONCLUSIONS: Together, these results suggest that gestures are unlikely to have a role in compensating for linguistic deficits and to serve as a representational system conveying thought content independent of language. Surprisingly, aphasia rather is a model of how gesture and language are inherently integrated and aligned: Even when language is impaired, it remains the essential provider of content.