Using explainable artificial intelligence to identify linguistic biomarkers of amyloid pathology in primary progressive aphasia.

Introduction: Recent success has been achieved in Alzheimer's disease (AD) clinical trials targeting amyloid beta (ß), demonstrating a reduction in the rate of cognitive decline. However, testing methods for amyloid-ß positivity are currently costly or invasive, motivating the development of accessible screening approaches to steer patients toward appropriate diagnostic tests. Here, we employ a pre-trained language model (Distil-RoBERTa) to identify amyloid-ß positivity from a short, connected speech sample. We further use explainable AI (XAI) methods to extract interpretable linguistic features that can be employed in clinical practice. Methods: We obtained language samples from 74 patients with primary progressive aphasia (PPA) across its three variants. Amyloid-ß positivity was established through the analysis of cerebrospinal fluid, amyloid PET, or autopsy. 51% of the sample was amyloid-positive. We trained Distil-RoBERTa for 16 epochs with a batch size of 6 and a learning rate of 5e-5, and used the LIME algorithm to train interpretation models to interpret the trained classifier's inference conditions. Results: Over ten runs of 10-fold cross-validation, the classifier achieved a mean accuracy of 92%, SD = 0.01. Interpretation models were able to capture the classifier's behavior well, achieving an accuracy of 97% against classifier predictions, and uncovering several novel speech patterns that may characterize amyloid-ß positivity. Discussion: Our work improves previous research which indicates connected speech is a useful diagnostic input for prediction of the presence of amyloid-ß in patients with PPA. Further, we leverage XAI techniques to reveal novel linguistic features that can be tested in clinical practice in the appropriate subspecialty setting. Computational linguistic analysis of connected speech shows great promise as a novel assessment method in patients with AD and related disorders.

Language Abnormalities in Alzheimer's Disease Arise from Reduced Informativeness: A Cross-Linguistic Study in English and Persian.

INTRODUCTION: This research investigates the psycholinguistic origins of language impairments in Alzheimer's Disease (AD), questioning if these impairments result from language-specific structural disruptions or from a universal deficit in generating meaningful content. METHODS: Cross-linguistic analysis was conducted on language samples from 184 English and 52 Persian speakers, comprising both AD patients and healthy controls, to extract various language features. Furthermore, we introduced a machine learning-based metric, Language Informativeness Index (LII), to quantify informativeness. RESULTS: Indicators of AD in English were found to be highly predictive of AD in Persian, with a 92.3% classification accuracy. Additionally, we found robust correlations between the typical linguistic abnormalities of AD and language emptiness (low LII) across both languages. DISCUSSION: Findings suggest AD linguistics impairments are attributed to a core universal difficulty in generating informative messages. Our approach underscores the importance of incorporating biocultural diversity into research, fostering the development of inclusive diagnostic tools.

Impaired language in Alzheimer's disease: A comparison between English and Persian implicates content-word frequency rather than the noun-verb distinction.

This study challenges the conventional psycholinguistic view that the distinction between nouns and verbs is pivotal in understanding language impairments in neurological disorders. Traditional views link frontal brain region damage with verb processing deficits and posterior temporoparietal damage with noun difficulties. However, this perspective is contested by findings from patients with Alzheimer's disease (pwAD), who show impairments in both word classes despite their typical temporoparietal atrophy. Notably, pwAD tend to use semantically lighter verbs in their speech than healthy individuals. By examining English-speaking pwAD and comparing them with Persian-speaking pwAD, this research aims to demonstrate that language impairments in Alzheimer's disease (AD) stem from the distributional properties of words within a language rather than distinct neural processing networks for nouns and verbs. We propose that the primary deficit in AD language production is an overreliance on high-frequency words. English has a set of particularly high-frequency verbs that surpass most nouns in usage frequency. Since pwAD tend to use high-frequency words, the byproduct of this word distribution in the English language would be an over-usage of high-frequency verbs. In contrast, Persian features complex verbs with an overall distribution lacking extremely high-frequency verbs like those found in English. As a result, we hypothesize that Persian-speaking pwAD would not have a bias toward the overuse of high-frequency verbs. We analyzed language samples from 95 English-speaking pwAD and 91 healthy controls, along with 27 Persian-speaking pwAD and 27 healthy controls. Employing uniform automated natural language processing methods, we measured the usage rates of nouns, verbs, and word frequencies across both cohorts. Our findings showed that English-speaking pwAD use higher-frequency verbs than healthy individuals, a pattern not mirrored by Persian-speaking pwAD. Crucially, we found a significant interaction between the frequencies of verbs used by English and Persian speakers with and without AD. Moreover, regression models that treated noun and verb frequencies as separate predictors did not outperform models that considered overall word frequency alone in classifying AD. In conclusion, this study suggests that language abnormalities among English-speaking pwAD reflect the unique distributional properties of words in English rather than a universal noun-verb class distinction. Beyond offering a new understanding of language abnormalities in AD, the study highlights the critical need for further investigation across diverse languages to deepen our insight into the mechanisms of language impairments in neurological disorders.

Language uncovers visuospatial dysfunction in posterior cortical atrophy: a natural language processing approach.

Introduction: Posterior Cortical Atrophy (PCA) is a syndrome characterized by a progressive decline in higher-order visuospatial processing, leading to symptoms such as space perception deficit, simultanagnosia, and object perception impairment. While PCA is primarily known for its impact on visuospatial abilities, recent studies have documented language abnormalities in PCA patients. This study aims to delineate the nature and origin of language impairments in PCA, hypothesizing that language deficits reflect the visuospatial processing impairments of the disease. Methods: We compared the language samples of 25 patients with PCA with age-matched cognitively normal (CN) individuals across two distinct tasks: a visually-dependent picture description and a visually-independent job description task. We extracted word frequency, word utterance latency, and spatial relational words for this comparison. We then conducted an in-depth analysis of the language used in the picture description task to identify specific linguistic indicators that reflect the visuospatial processing deficits of PCA. Results: Patients with PCA showed significant language deficits in the visually-dependent task, characterized by higher word frequency, prolonged utterance latency, and fewer spatial relational words, but not in the visually-independent task. An in-depth analysis of the picture description task further showed that PCA patients struggled to identify certain visual elements as well as the overall theme of the picture. A predictive model based on these language features distinguished PCA patients from CN individuals with high classification accuracy. Discussion: The findings indicate that language is a sensitive behavioral construct to detect visuospatial processing abnormalities of PCA. These insights offer theoretical and clinical avenues for understanding and managing PCA, underscoring language as a crucial marker for the visuospatial deficits of this atypical variant of Alzheimer's disease.

Language Uncovers Visuospatial Dysfunction in Posterior Cortical Atrophy: A Natural Language Processing Approach.

Introduction: Posterior Cortical Atrophy (PCA) is a syndrome characterized by a progressive decline in higher-order visuospatial processing, leading to symptoms such as space perception deficit, simultanagnosia, and object perception impairment. While PCA is primarily known for its impact on visuospatial abilities, recent studies have documented language abnormalities in PCA patients. This study aims to delineate the nature and origin of language impairments in PCA, hypothesizing that language deficits reflect the visuospatial processing impairments of the disease. Methods: We compared the language samples of 25 patients with PCA with age-matched cognitively normal (CN) individuals across two distinct tasks: a visually-dependent picture description and a visually-independent job description task. We extracted word frequency, word utterance latency, and spatial relational words for this comparison. We then conducted an in-depth analysis of the language used in the picture description task to identify specific linguistic indicators that reflect the visuospatial processing deficits of PCA. Results: Patients with PCA showed significant language deficits in the visually-dependent task, characterized by higher word frequency, prolonged utterance latency, and fewer spatial relational words, but not in the visually-independent task. An in-depth analysis of the picture description task further showed that PCA patients struggled to identify certain visual elements as well as the overall theme of the picture. A predictive model based on these language features distinguished PCA patients from CN individuals with high classification accuracy. Discussion: The findings indicate that language is a sensitive behavioral construct to detect visuospatial processing abnormalities of PCA. These insights offer theoretical and clinical avenues for understanding and managing PCA, underscoring language as a crucial marker for the visuospatial deficits of this atypical variant of Alzheimer's disease.

Automated analysis of written language in the three variants of primary progressive aphasia.

Despite the important role of written language in everyday life, abnormalities in functional written communication have been sparsely investigated in primary progressive aphasia. Prior studies have analysed written language separately in each of the three variants of primary progressive aphasia-but have rarely compared them to each other or to spoken language. Manual analysis of written language can be a time-consuming process. We therefore developed a program that quantifies content units and total units in written or transcribed language samples. We analysed written and spoken descriptions of the Western Aphasia Battery picnic scene, based on a predefined content unit corpus. We calculated the ratio of content units to units as a measure of content density. Our cohort included 115 participants (20 controls for written, 20 controls for spoken, 28 participants with nonfluent variant primary progressive aphasia, 30 for logopenic variant and 17 for semantic variant). Our program identified content units with a validity of 99.7% (95%CI 99.5-99.8). All patients wrote fewer units than controls (P < 0.001). Patients with the logopenic variant (P = 0.013) and the semantic variant (0.004) wrote fewer content units than controls. The content unit-to-unit ratio was higher in the nonfluent and semantic variants than controls (P = 0.019), but no difference in the logopenic variant (P = 0.962). Participants with the logopenic (P < 0.001) and semantic (P = 0.04) variants produced fewer content units in written compared to spoken descriptions. All variants produced fewer units in written samples compared to spoken (P < 0.001). However, due to a relatively smaller decrease in written content units, we observed a larger content unit-to-unit ratio in writing over speech (P < 0.001). Written and spoken content units (r = 0.5, P = 0.009) and total units (r = 0.64, P < 0.001) were significantly correlated in patients with nonfluent variant, but this was not the case for logopenic or semantic. Considering all patients with primary progressive aphasia, fewer content units were produced in those with greater aphasia severity (Progressive Aphasia Severity Scale Sum of Boxes, r = -0.24, P = 0.04) and dementia severity (Clinical Dementia Rating scale Sum of Boxes, r = -0.34, P = 0.004). In conclusion, we observed reduced written content in patients with primary progressive aphasia compared to controls, with a preference for content over non-content units in patients with the nonfluent and semantic variants. We observed a similar 'telegraphic' style in both language modalities in patients with the nonfluent variant. Lastly, we show how our program provides a time-efficient tool, which could enable feedback and tracking of writing as an important feature of language and cognition.

Measuring Sentence Information via Surprisal: Theoretical and Clinical Implications in Nonfluent Aphasia.

OBJECTIVE: Nonfluent aphasia is characterized by simplified sentence structures and word-level abnormalities, including reduced use of verbs and function words. The predominant belief about the disease mechanism is that a core deficit in syntax processing causes both structural and word-level abnormalities. Here, we propose an alternative view based on information theory to explain the symptoms of nonfluent aphasia. We hypothesize that the word-level features of nonfluency constitute a distinct compensatory process to augment the information content of sentences to the level of healthy speakers. We refer to this process as lexical condensation. METHODS: We use a computational approach based on language models to measure sentence information through surprisal, a metric calculated by the average probability of occurrence of words in a sentence, given their preceding context. We apply this method to the language of patients with nonfluent primary progressive aphasia (nfvPPA; n = 36) and healthy controls (n = 133) as they describe a picture. RESULTS: We found that nfvPPA patients produced sentences with the same sentence surprisal as healthy controls by using richer words in their structurally impoverished sentences. Furthermore, higher surprisal in nfvPPA sentences correlated with the canonical features of agrammatism: a lower function-to-all-word ratio, a lower verb-to-noun ratio, a higher heavy-to-all-verb ratio, and a higher ratio of verbs in -ing forms. INTERPRETATION: Using surprisal enables testing an alternative account of nonfluent aphasia that regards its word-level features as adaptive, rather than defective, symptoms, a finding that would call for revisions in the therapeutic approach to nonfluent language production. ANN NEUROL 2023;94:647-657.

Less is more in language production: an information-theoretic analysis of agrammatism in primary progressive aphasia.

Agrammatism is a disorder of language production characterized by short, simplified sentences, the omission of function words, an increased use of nouns over verbs and a higher use of heavy verbs. Despite observing these phenomena for decades, the accounts of agrammatism have not converged. Here, we propose and test the hypothesis that the lexical profile of agrammatism results from a process that opts for words with a lower frequency of occurrence to increase lexical information. Furthermore, we hypothesize that this process is a compensatory response to patients' core deficit in producing long, complex sentences. In this cross-sectional study, we analysed speech samples of patients with primary progressive aphasia (n = 100) and healthy speakers (n = 65) as they described a picture. The patient cohort included 34 individuals with the non-fluent variant, 41 with the logopenic variant and 25 with the semantic variant of primary progressive aphasia. We first analysed a large corpus of spoken language and found that the word types preferred by patients with agrammatism tend to have lower frequencies of occurrence than less preferred words. We then conducted a computational simulation to examine the impact of word frequency on lexical information as measured by entropy. We found that strings of words that exclude highly frequent words have a more uniform word distribution, thereby increasing lexical entropy. To test whether the lexical profile of agrammatism results from their inability to produce long sentences, we asked healthy speakers to produce short sentences during the picture description task. We found that, under this constrained condition, a similar lexical profile of agrammatism emerged in the short sentences of healthy individuals, including fewer function words, more nouns than verbs and more heavy verbs than light verbs. This lexical profile of short sentences resulted in their lower average word frequency than unconstrained sentences. We extended this finding by showing that, in general, shorter sentences get packaged with lower-frequency words as a basic property of efficient language production, evident in the language of healthy speakers and all primary progressive aphasia variants.

Using Generative Artificial Intelligence to Classify Primary Progressive Aphasia from Connected Speech.

Neurodegenerative dementia syndromes, such as Primary Progressive Aphasias (PPA), have traditionally been diagnosed based in part on verbal and nonverbal cognitive profiles. Debate continues about whether PPA is best subdivided into three variants and also regarding the most distinctive linguistic features for classifying PPA variants. In this study, we harnessed the capabilities of artificial intelligence (AI) and natural language processing (NLP) to first perform unsupervised classification of concise, connected speech samples from 78 PPA patients. Large Language Models discerned three distinct PPA clusters, with 88.5% agreement with independent clinical diagnoses. Patterns of cortical atrophy of three data-driven clusters corresponded to the localization in the clinical diagnostic criteria. We then used NLP to identify linguistic features that best dissociate the three PPA variants. Seventeen features emerged as most valuable for this purpose, including the observation that separating verbs into high and low-frequency types significantly improves classification accuracy. Using these linguistic features derived from the analysis of brief connected speech samples, we developed a classifier that achieved 97.9% accuracy in predicting PPA subtypes and healthy controls. Our findings provide pivotal insights for refining early-stage dementia diagnosis, deepening our understanding of the characteristics of these neurodegenerative phenotypes and the neurobiology of language processing, and enhancing diagnostic evaluation accuracy.

A syntax-lexicon trade-off in language production.

Spoken language production involves selecting and assembling words and syntactic structures to convey one's message. Here we probe this process by analyzing natural language productions of individuals with primary progressive aphasia (PPA) and healthy individuals. Based on prior neuropsychological observations, we hypothesize that patients who have difficulty producing complex syntax might choose semantically richer words to make their meaning clear, whereas patients with lexicosemantic deficits may choose more complex syntax. To evaluate this hypothesis, we first introduce a frequency-based method for characterizing the syntactic complexity of naturally produced utterances. We then show that lexical and syntactic complexity, as measured by their frequencies, are negatively correlated in a large (n = 79) PPA population. We then show that this syntax-lexicon trade-off is also present in the utterances of healthy speakers (n = 99) taking part in a picture description task, suggesting that it may be a general property of the process by which humans turn thoughts into speech.

Natural language processing in psychiatry: the promises and perils of a transformative approach.

A person's everyday language can indicate patterns of thought and emotion predictive of mental illness. Here, we discuss how natural language processing methods can be used to extract indicators of mental health from language to help address long-standing problems in psychiatry, along with the potential hazards of this new technology.

A machine learning approach to predicting psychosis using semantic density and latent content analysis.

Subtle features in people's everyday language may harbor the signs of future mental illness. Machine learning offers an approach for the rapid and accurate extraction of these signs. Here we investigate two potential linguistic indicators of psychosis in 40 participants of the North American Prodrome Longitudinal Study. We demonstrate how the linguistic marker of semantic density can be obtained using the mathematical method of vector unpacking, a technique that decomposes the meaning of a sentence into its core ideas. We also demonstrate how the latent semantic content of an individual's speech can be extracted by contrasting it with the contents of conversations generated on social media, here 30,000 contributors to Reddit. The results revealed that conversion to psychosis is signaled by low semantic density and talk about voices and sounds. When combined, these two variables were able to predict the conversion with 93% accuracy in the training and 90% accuracy in the holdout datasets. The results point to a larger project in which automated analyses of language are used to forecast a broad range of mental disorders well in advance of their emergence.

Emotional self-other voice processing in schizophrenia and its relationship with hallucinations: ERP evidence.

Abnormalities in self-other voice processing have been observed in schizophrenia, and may underlie the experience of hallucinations. More recent studies demonstrated that these impairments are enhanced for speech stimuli with negative content. Nonetheless, few studies probed the temporal dynamics of self versus nonself speech processing in schizophrenia and, particularly, the impact of semantic valence on self-other voice discrimination. In the current study, we examined these questions, and additionally probed whether impairments in these processes are associated with the experience of hallucinations. Fifteen schizophrenia patients and 16 healthy controls listened to 420 prerecorded adjectives differing in voice identity (self-generated [SGS] versus nonself speech [NSS]) and semantic valence (neutral, positive, and negative), while EEG data were recorded. The N1, P2, and late positive potential (LPP) ERP components were analyzed. ERP results revealed group differences in the interaction between voice identity and valence in the P2 and LPP components. Specifically, LPP amplitude was reduced in patients compared with healthy subjects for SGS and NSS with negative content. Further, auditory hallucinations severity was significantly predicted by LPP amplitude: the higher the SAPS "voices conversing" score, the larger the difference in LPP amplitude between negative and positive NSS. The absence of group differences in the N1 suggests that self-other voice processing abnormalities in schizophrenia are not primarily driven by disrupted sensory processing of voice acoustic information. The association between LPP amplitude and hallucination severity suggests that auditory hallucinations are associated with enhanced sustained attention to negative cues conveyed by a nonself voice.

Is this my voice or yours? The role of emotion and acoustic quality in self-other voice discrimination in schizophrenia.

INTRODUCTION: Impairments in self-other voice discrimination have been consistently reported in schizophrenia, and associated with the severity of auditory verbal hallucinations (AVHs). This study probed the interactions between voice identity, voice acoustic quality, and semantic valence in a self-other voice discrimination task in schizophrenia patients compared with healthy subjects. The relationship between voice identity discrimination and AVH severity was also explored. METHODS: Seventeen chronic schizophrenia patients and 19 healthy controls were asked to read aloud a list of adjectives characterised by emotional or neutral content. Participants' voice was recorded in the first session. In the behavioural task, 840 spoken words differing in identity (self/non-self), acoustic quality (undistorted/distorted), and semantic valence (negative/positive/neutral) were presented. Participants indicated if the words were spoken in their own voice, another person's voice, or were unsure. RESULTS: Patients were less accurate than controls in the recognition of self-generated speech with negative content only. Impaired recognition of negative self-generated speech was associated with AVH severity ("voices conversing"). CONCLUSIONS: These results suggest that abnormalities in higher order processes (evaluation of the salience of a speech stimulus) modulate impaired self-other voice discrimination in schizophrenia. Abnormal processing of negative self-generated speech may play a role in the experience of AVH.

Did you or I say pretty, rude or brief? An ERP study of the effects of speaker's identity on emotional word processing.

During speech comprehension, multiple cues need to be integrated at a millisecond speed, including semantic information, as well as voice identity and affect cues. A processing advantage has been demonstrated for self-related stimuli when compared with non-self stimuli, and for emotional relative to neutral stimuli. However, very few studies investigated self-other speech discrimination and, in particular, how emotional valence and voice identity interactively modulate speech processing. In the present study we probed how the processing of words' semantic valence is modulated by speaker's identity (self vs. non-self voice). Sixteen healthy subjects listened to 420 prerecorded adjectives differing in voice identity (self vs. non-self) and semantic valence (neutral, positive and negative), while electroencephalographic data were recorded. Participants were instructed to decide whether the speech they heard was their own (self-speech condition), someone else's (non-self speech), or if they were unsure. The ERP results demonstrated interactive effects of speaker's identity and emotional valence on both early (N1, P2) and late (Late Positive Potential - LPP) processing stages: compared with non-self speech, self-speech with neutral valence elicited more negative N1 amplitude, self-speech with positive valence elicited more positive P2 amplitude, and self-speech with both positive and negative valence elicited more positive LPP. ERP differences between self and non-self speech occurred in spite of similar accuracy in the recognition of both types of stimuli. Together, these findings suggest that emotion and speaker's identity interact during speech processing, in line with observations of partially dependent processing of speech and speaker information.

Abnormal interactions between context, memory structure, and mood in schizophrenia: an ERP investigation.

This study used event-related potentials to examine interactions between mood, sentence context, and semantic memory structure in schizophrenia. Seventeen male chronic schizophrenia and 15 healthy control subjects read sentence pairs after positive, negative, or neutral mood induction. Sentences ended with expected words (EW), within-category violations (WCV), or between-category violations (BCV). Across all moods, patients showed sensitivity to context indexed by reduced N400 to EW relative to both WCV and BCV. However, they did not show sensitivity to the semantic memory structure. N400 abnormalities were particularly enhanced under a negative mood in schizophrenia. These findings suggest abnormal interactions between mood, context processing, and connections within semantic memory in schizophrenia, and a specific role of negative mood in modulating semantic processes in this disease.

Abnormalities in the processing of emotional prosody from single words in schizophrenia.

BACKGROUND: Abnormalities in emotional prosody processing have been consistently reported in schizophrenia and are related to poor social outcomes. However, the role of stimulus complexity in abnormal emotional prosody processing is still unclear. METHOD: We recorded event-related potentials in 16 patients with chronic schizophrenia and 16 healthy controls to investigate: 1) the temporal course of emotional prosody processing; and 2) the relative contribution of prosodic and semantic cues in emotional prosody processing. Stimuli were prosodic single words presented in two conditions: with intelligible (semantic content condition-SCC) and unintelligible semantic content (pure prosody condition-PPC). RESULTS: Relative to healthy controls, schizophrenia patients showed reduced P50 for happy PPC words, and reduced N100 for both neutral and emotional SCC words and for neutral PPC stimuli. Also, increased P200 was observed in schizophrenia for happy prosody in SCC only. Behavioral results revealed higher error rates in schizophrenia for angry prosody in SCC and for happy prosody in PPC. CONCLUSIONS: Together, these data further demonstrate the interactions between abnormal sensory processes and higher-order processes in bringing about emotional prosody processing dysfunction in schizophrenia. They further suggest that impaired emotional prosody processing is dependent on stimulus complexity.

Evaluation of chaperone-like activity of alginate: microcapsule and water-soluble forms.

To evaluate the chaperone-like activity of alginate stabilization and refolding of alkaline phosphatase (ALP) was investigated in the presence of alginate through two different approaches, the soluble form and microcapsule assisted methods. It was found that in the presence of microcapsules, ALP can be stabilized to a higher degree compared with the water-soluble form, whereas the denatured ALP is refolded with a higher yield through latter method. Lower refolding yields of alginate beads compared with its soluble form may be the result of lower refolding rate of ALP upon elution of the bound enzyme by dispersing the precipitate in NaCl which left the unfolded protein in an unsuitable environment, providing enough time for protein aggregation and leading finally to lower recovered activity compared with application of soluble form of alginate. In addition in the case of alginate capsules, the choice of suitable divalent ion is essential for stability and assistance in refolding.