Oral language and alzheimer's disease: A reduction in syntactic complexity

A curious case of retrogenesis in language: Automated analysis of language patterns observed in dementia patients and young children

Introduction

While linguistic retrogenesis has been extensively investigated in the neuroscientific and behavioral literature, there has been little work on retrogenesis using computerized approaches to language analysis.

Methods

We bridge this gap by introducing a method based on comparing output of a pre-trained neural language model (NLM) with an artificially degraded version of itself to examine the transcripts of speech produced by seniors with and without dementia and healthy children during spontaneous language tasks. We compare a range of linguistic characteristics including language model perplexity, syntactic complexity, lexical frequency and part-of-speech use across these groups.

Results

Our results indicate that healthy seniors and children older than 8 years share similar linguistic characteristics, as do dementia patients and children who are younger than 8 years.

Discussion

Our study aligns with the growing evidence that language deterioration in dementia mirrors language acquisition in development using computational linguistic methods based on NLMs. This insight underscores the importance of further research to refine its application in guiding developmentally appropriate patient care, particularly in early stages.

Methodologies for assessing morphosyntactic ability in people with Alzheimer's disease

BACKGROUND

The detection and description of language impairments in neurodegenerative diseases like Alzheimer's Disease (AD) play an important role in research, clinical diagnosis and intervention. Various methodological protocols have been implemented for the assessment of morphosyntactic abilities in AD; narrative discourse elicitation tasks and structured experimental tasks for production, offline and online structured experimental tasks for comprehension. Very few studies implement and compare different methodological protocols; thus, little is known about the advantages and disadvantages of each methodology.

AIMS

To discuss and compare the main behavioral methodological approaches and tasks that have been used in psycholinguistic research to assess different aspects of morphosyntactic production and comprehension in individuals with AD at the word and sentence levels.

METHODS

A narrative review was conducted through searches in the scientific databases Google Scholar, Scopus, Science Direct, MITCogNet, PubMed. Only studies written in English, that reported quantitative data and were published in peer-reviewed journals were considered with respect to their methodological protocol. Moreover, we considered studies that reported research on all stages of the disease and we included only studies that also reported results of a healthy control group. Studies that implemented standardized assessment tools were not considered in this review.

OUTCOMES & RESULTS

The main narrative discourse elicitation tasks implemented for the assessment of morphosyntactic production include interviews, picture-description and story narration, whereas the main structured experimental tasks include sentence completion, constrained sentence production, sentence repetition and naming. Morphosyntactic comprehension in AD has been assessed with the use of structured experimental tasks, both offline (sentence-picture matching, grammaticality judgment) and online (cross-modal naming,speeded sentence acceptability judgment, auditory moving window, word detection, reading). For each task we considered studies that reported results from different morphosyntactic structures and phenomena in as many different languages as possible.

CONCLUSIONS & IMPLICATIONS

Our review revealed strengths and weaknesses of these methods but also directions for future research. Narrative discourse elicitation tasks as well as structured experimental tasks have been used in a variety of languages, and have uncovered preserved morphosyntactic production but also deficits in people with AD. A combination of narrative discourse elicitation and structured production tasks for the assessment of the same morphosyntactic structure has been rarely used. Regarding comprehension, offline tasks have been implemented in various languages, whereas online tasks have been mainly used in English. Offline and online experimental paradigms have often produced contradictory results even within the same study. The discrepancy between the two paradigms has been attributed to the different working memory demands they impose to the comprehender or to the different parsing processes they tap. Strengths and shortcomings of each methodology are summarized in the paper, and comparisons between different tasks are attempted when this is possible. Thus, the paper may serve as a methodological guide for the study of morphosyntax in AD and possibly in other neurodegenerative diseases.

WHAT THIS PAPER ADDS

What is already known on this subject For the assessment of morphosyntactic abilities in AD, various methodological paradigms have been implemented: narrative discourse elicitation tasks and structured experimental tasks for production, and offline and online structured experimental tasks for comprehension. Very few studies implement and compare different methodological protocols; thus, little is known about the advantages and disadvantages of each methodology. What this paper adds to existing knowledge The paper presents an overview of methodologies that have been used to assess morphosyntactic production and comprehension of people with AD at the word and sentence levels. The paper summarizes the strengths and shortcomings of each methodology, providing both the researcher and the clinician with some directions in their endeavour of investigating language in AD. Also, the paper highlights the need for further research that will implement carefully scrutinized tasks from various experimental paradigms and will explore distinct aspects of the AD patients' morphosyntactic abilities in typologically different languages. What are the potential or actual clinical implications of this work? The paper may serve as a reference point for (psycho-)linguists who wish to study morphosyntactic abilities in AD, and for speech and language therapists who might need to apply morphosyntactic protocols to their patients in order to assess them or design appropriate therapeutic interventions for production and comprehension deficits.

Incompleteness features in the descriptive discourse of Chinese elders with and without Alzheimer's disease

Alzheimer's disease (AD) can manifest itself with prominent language dysfunction. Incompleteness in discourse refers to the lack of indispensable sentence-constructing elements that hinder communication fluency and accuracy. The current study investigates how the pattern of incompleteness is associated with the descriptive discourse produced by elders withoutAD and those with different stages ofAD. The Chinese discourse samples were collected from the picture description of 40 elders with mild probableAD (Mini-Mental State Examination (MMSE) 21-26, Montreal Cognitive Assessment Scale-Basic (MoCA-B) 15-19), 40 elders with moderate probableAD (MMSE 11-20, MoCA-B 10-14), and 40 controls (MMSE 26-29, MoCA-B 24-29). The total production of incomplete sentences and six incompleteness features were examined. The MildAD, ModerateAD, and Control groups differed in the total output of the incomplete sentence. Group differences also emerged in four incompleteness features: missing subject, missing predicate, missing object, and missing functional word. The ModerateAD group differed from the MildAD group with respect to most significant features, while MildAD and Control groups were very similar. The results suggested thatAD impairs the sentence construction ability of Chinese elders, especially at the later stage. These statistically significant differences between the groups might provide some references when diagnosing the risk and possibility of cognitive impairment of Chinese elders, facilitating the design of clinical evaluation or screening for probableAD.

Detecting Alzheimer's Disease Using Natural Language Processing of Referential Communication Task Transcripts

BACKGROUND

People with Alzheimer's disease (AD) often demonstrate difficulties in discourse production. Referential communication tasks (RCTs) are used to examine a speaker's capability to select and verbally code the characteristics of an object in interactive conversation.

OBJECTIVE

In this study, we used contextualized word representations from Natural language processing (NLP) to evaluate how well RCTs are able to distinguish between people with AD and cognitively healthy older adults.

METHODS

We adapted machine learning techniques to analyze manually transcribed speech transcripts in an RCT from 28 older adults, including 12 with AD and 16 cognitively healthy older adults. Two approaches were applied to classify these speech transcript samples: 1) using clinically relevant linguistic features, 2) using machine learned representations derived by a state-of-art pretrained NLP transfer learning model, Bidirectional Encoder Representation from Transformer (BERT) based classification model.

RESULTS

The results demonstrated the superior performance of AD detection using a designed transfer learning NLP algorithm. Moreover, the analysis showed that transcripts of a single image yielded high accuracies in AD detection.

CONCLUSION

The results indicated that RCT may be useful as a diagnostic tool for AD, and that the task can be simplified to a subset of images without significant sacrifice to diagnostic accuracy, which can make RCT an easier and more practical tool for AD diagnosis. The results also demonstrate the potential of RCT as a tool to better understand cognitive deficits from the perspective of discourse production in people with AD.

Automated analysis of propositional idea density in older adults

Previous research suggests oral and written language can act as barometers of an individual's cognitive function, potentially providing a screening tool for the earliest stages of Alzheimer's disease (AD) and other forms of dementia. Idea density is a measure of the rate at which ideas, or elementary predications, are expressed and may provide an ideal measure for early detection of deficits in language. Previous research has shown that when no restrictions are set on the topic of the idea, a decrease in propositional idea density (PID) is associated with an increased risk of developing AD. However, this has been limited by moderate sample sizes and manual transcribing. Technological advancement has enabled the automated calculation of PID from tools such as the Computerized Propositional Idea Density Rater (CPIDR). We delivered an online autobiographical writing task to older adult Australians from ISLAND (Island Study Linking Ageing and Neurodegenerative Disease). Linear regression models were fitted in R. We analysed text files (range 10-1180 words) using CPIDRv5 provided by 3316 (n = 853 males [25.7%], n = 2463 females [74.3%]) ISLAND participants. Over 358,957 words written in 3316 written autobiographical responses were analysed. Mean PID was higher in females (53.5 [±3.69]) than males (52.6 [±4.50]). Both advancing age and being male were significantly associated with a decrease in PID (p < .001). Automated methods of language analysis hold great promise for the early detection of subtle deficits in language capacity. Although our effect sizes were small, PID may be a sensitive measure of deficits in language in ageing individuals and is able to be collected at scale using online methods of data capture.

Language Decline Characterizes Amnestic Mild Cognitive Impairment Independent of Cognitive Decline

Purpose This research investigated the nature of cognitive decline in prodromal Alzheimer's disease (AD), particularly in mild cognitive impairment, amnestic type (aMCI). We assessed language in aMCI as compared with healthy aging (HA) and healthy young (HY) with new psycholinguistic assessment of complex sentences, and we tested the degree to which deficits on this language measure relate to performance in other general cognitive domains such as memory. Method Sixty-one individuals with aMCI were compared with 24 HA and 10 HY adults on a psycholinguistic measure of complex sentence production (relative clauses). In addition, HA, HY, and a subset of the aMCI participants (n = 22) were also tested on a multidomain cognitive screen, the Addenbrooke's Cognitive Examination-Revised (ACE-R), and on a verbal working memory Brown-Peterson (BP) test. General and generalized linear mixed models were used to test psycholinguistic results and to test whether ACE-R and BP performance predicted performance on the psycholinguistic test similarly in the aMCI and HA groups. Results On the psycholinguistic measure, sentence imitation was significantly deficited in aMCI in comparison with that in HA and HY. Experimental factorial designs revealed that individuals with aMCI had particular difficulty repeating sentences that especially challenged syntax-semantics integration. As expected, the aMCI group also performed significantly below the HY and HA groups on the ACE-R. Neither the ACE-R Memory subtest nor the BP total scores predicted performance on the psycholinguistic task for either the aMCI or the HA group. However, the ACE-R total score significantly predicted psycholinguistic task performance, with increased ACE-R performance predicting increased psycholinguistic task performance only for the HA group, not for the aMCI group. Conclusions Results suggest a selective deterioration in language in aMCI, specifically a weakening of syntax-semantics integration in complex sentence processing, and a general independence of this language deficit and memory decline. Results cohere with previous assessments of the nature of difficulty in complex sentence formation in aMCI. We argue that clinical screening for prodromal AD can be strengthened by supplementary testing of language, as well as memory, and extended evaluation of strength of their relation.

Language Impairment in Alzheimer's Disease-Robust and Explainable Evidence for AD-Related Deterioration of Spontaneous Speech Through Multilingual Machine Learning

Alzheimer's disease (AD) is a pervasive neurodegenerative disease that affects millions worldwide and is most prominently associated with broad cognitive decline, including language impairment. Picture description tasks are routinely used to monitor language impairment in AD. Due to the high amount of manual resources needed for an in-depth analysis of thereby-produced spontaneous speech, advanced natural language processing (NLP) combined with machine learning (ML) represents a promising opportunity. In this applied research field though, NLP and ML methodology do not necessarily ensure robust clinically actionable insights into cognitive language impairment in AD and additional precautions must be taken to ensure clinical-validity and generalizability of results. In this study, we add generalizability through multilingual feature statistics to computational approaches for the detection of language impairment in AD. We include 154 participants (78 healthy subjects, 76 patients with AD) from two different languages (106 English speaking and 47 French speaking). Each participant completed a picture description task, in addition to a battery of neuropsychological tests. Each response was recorded and manually transcribed. From this, task-specific, semantic, syntactic and paralinguistic features are extracted using NLP resources. Using inferential statistics, we determined language features, excluding task specific features, that are significant in both languages and therefore represent "generalizable" signs for cognitive language impairment in AD. In a second step, we evaluated all features as well as the generalizable ones for English, French and both languages in a binary discrimination ML scenario (AD vs. healthy) using a variety of classifiers. The generalizable language feature set outperforms the all language feature set in English, French and the multilingual scenarios. Semantic features are the most generalizable while paralinguistic features show no overlap between languages. The multilingual model shows an equal distribution of error in both English and French. By leveraging multilingual statistics combined with a theory-driven approach, we identify AD-related language impairment that generalizes beyond a single corpus or language to model language impairment as a clinically-relevant cognitive symptom. We find a primary impairment in semantics in addition to mild syntactic impairment, possibly confounded by additional impaired cognitive functions.

Linguistic markers predict onset of Alzheimer's disease

BACKGROUND

The aim of this study is to use classification methods to predict future onset of Alzheimer's disease in cognitively normal subjects through automated linguistic analysis.

METHODS

To study linguistic performance as an early biomarker of AD, we performed predictive modeling of future diagnosis of AD from a cognitively normal baseline of Framingham Heart Study participants. The linguistic variables were derived from written responses to the cookie-theft picture-description task. We compared the predictive performance of linguistic variables with clinical and neuropsychological variables. The study included 703 samples from 270 participants out of which a dataset consisting of a single sample from 80 participants was held out for testing. Half of the participants in the test set developed AD symptoms before 85 years old, while the other half did not. All samples in the test set were collected during the cognitively normal period (before MCI). The mean time to diagnosis of mild AD was 7.59 years.

FINDINGS

Significant predictive power was obtained, with AUC of 0.74 and accuracy of 0.70 when using linguistic variables. The linguistic variables most relevant for predicting onset of AD have been identified in the literature as associated with cognitive decline in dementia.

INTERPRETATION

The results suggest that language performance in naturalistic probes expose subtle early signs of progression to AD in advance of clinical diagnosis of impairment.

FUNDING

Pfizer, Inc. provided funding to obtain data from the Framingham Heart Study Consortium, and to support the involvement of IBM Research in the initial phase of the study. The data used in this study was supported by Framingham Heart Study's National Heart, Lung, and Blood Institute contract (N01-HC-25195), and by grants from the National Institute on Aging grants (R01-AG016495, R01-AG008122) and the National Institute of Neurological Disorders and Stroke (R01-NS017950).

Communication behavior of cognitively impaired older inpatients : A new setting for validating the CODEM instrument

BACKGROUND AND OBJECTIVE

Acutely ill older patients with cognitive impairment represent a major subgroup in acute care hospitals. In this context, communication plays a crucial role for patients' well-being, healthcare decisions, and medical outcomes. As validated measures are lacking, we tested the psychometric properties of an observational instrument to assess Communication Behavior in Dementia (CODEM) in the acute care hospital setting. As a novel feature, we were also able to incorporate linguistic and social-contextual measures.

MATERIAL AND METHODS

Data were drawn from a cross-sectional mixed methods study that focused on the occurrence of elderspeak during care interactions in two German acute care hospitals. A total of 43 acutely ill older patients with severe cognitive impairment (CI group, Mage ± SD = 83.6 ± 5.7 years) and 50 without cognitive impairment (CU group, Mage ± SD = 82.1 ± 6.3 years) were observed by trained research assistants during a standardized interview situation and rated afterwards by use of CODEM.

RESULTS

Factor analysis supported the expected two-factor solution for the CI group, i.e., a verbal content and a nonverbal relationship aspect. Findings of the current study indicated sound psychometric properties of the CODEM instrument including internal consistency, convergent, divergent, and criterion validity.

CONCLUSION

CODEM represents a reliable and valid tool to examine the communication behavior of older patients with CI in the acute care hospital setting. Thus, CODEM might serve as an important instrument for researcher and healthcare professionals to describe and improve communication patterns in this environment.

A computerized test for the assessment of mild cognitive impairment subtypes in sentence processing

This study examines thesentence processing ability of mild cognitive impairment (MCI) subtypes. In addition to standard MCI neuropsychological tests, an experimental approach was applied to assess language. 133 people (93 MCI/40 controls) participated in novel computerized sentence processing tasks. Results presented statistically significant differences between MCI/controls andMCI subtypes (ANOVA):(a) duration F(2,92) = 19.259,p < .001) in sentence construction; (b) correct answers (F(2, 89) = 8.560,p < .001) and duration (F2,89) = 15.525,p < .001)in text comprehension; (c) correct answers (F(2, 92) = 8.975,p < .001) andduration (F(2, 92) = 4.360,p = .016) in metaphoric sentences comprehension; (d) correct answers (F(2, 92) = 12.836,p < .001) andduration (F(2, 92) = 10.974,p < .001) in verb form generation. Subtle changes in MCIsubtypes could affect sentence processing and provide useful information for cognitive decline risk estimation and screening purposes.

Predicting mild cognitive impairment from spontaneous spoken utterances

INTRODUCTION

Trials in Alzheimer's disease are increasingly focusing on prevention in asymptomatic individuals. We hypothesized that indicators of mild cognitive impairment (MCI) may be present in the content of spoken language in older adults and be useful in distinguishing those with MCI from those who are cognitively intact. To test this hypothesis, we performed linguistic analyses of spoken words in participants with MCI and those with intact cognition participating in a clinical trial.

METHODS

Data came from a randomized controlled behavioral clinical trial to examine the effect of unstructured conversation on cognitive function among older adults with either normal cognition or MCI (ClinicalTrials.gov: NCT01571427). Unstructured conversations (but with standardized preselected topics across subjects) were recorded between interviewers and interviewees during the intervention sessions of the trial from 14 MCI and 27 cognitively intact participants. From the transcription of interviewees recordings, we grouped spoken words using Linguistic Inquiry and Word Count (LIWC), a structured table of words, which categorizes 2500 words into 68 different word subcategories such as positive and negative words, fillers, and physical states. The number of words in each LIWC word subcategory constructed a vector of 68 dimensions representing the linguistic features of each subject. We used support vector machine and random forest classifiers to distinguish MCI from cognitively intact participants.

RESULTS

MCI participants were distinguished from those with intact cognition using linguistic features obtained by LIWC with 84% classification accuracy which is well above chance 60%.

DISCUSSION

Linguistic analyses of spoken language may be a powerful tool in distinguishing MCI subjects from those with intact cognition. Further studies to assess whether spoken language derived measures could detect changes in cognitive functions in clinical trials are warrented.

Propositional Density in Spoken and Written Language of Czech-Speaking Patients With Mild Cognitive Impairment

PURPOSE

Propositional density (PD) is a measure of content richness in language production that declines in normal aging and more profoundly in dementia. The present study aimed to develop a PD scoring system for Czech and use it to compare PD in language productions of older people with amnestic mild cognitive impairment (aMCI) and control participants matched on age, gender, and education.

METHOD

Groups of patients with aMCI and cognitively healthy control participants (N = 20 each) provided short spoken and written language samples. Two samples were elicited for each modality, 1 describing recent events and 1 describing childhood memories. Series of neuropsychological tests were administered. The groups were compared using t-tests and the relations between measures using correlation coefficients.

RESULTS

PD was lower in spoken productions of patients with aMCI, compared with control participants, but only in language samples using remote memories. PD in these samples was related to verbal fluency and education but not to working memory. PD in written samples did not differ between participants with aMCI and control participants.

CONCLUSIONS

PD in spoken language reflects the cognitive decline in people with aMCI, but the effect is relatively mild. The results support the existing findings that PD is related to verbal fluency.

Lack of Frank Agrammatism in the Nonfluent Agrammatic Variant of Primary Progressive Aphasia

Background/Aims

Frank agrammatism, defined as the omission and/or substitution of grammatical morphemes with associated grammatical errors, is variably reported in patients with nonfluent variant primary progressive aphasia (nfPPA). This study addressed whether frank agrammatism is typical in agrammatic nfPPA patients when this feature is not required for diagnosis.

Method

We assessed grammatical production in 9 patients who satisfied current diagnostic criteria. Although the focus was agrammatism, motor speech skills were also evaluated to determine whether dysfluency arose primarily from apraxia of speech (AOS), instead of, or in addition to, agrammatism. Volumetric MRI analyses provided impartial imaging-supported diagnosis.

Results

The majority of cases exhibited neither frank agrammatism nor AOS.

Conclusion

There are nfPPA patients with imaging-supported diagnosis and preserved motor speech skills who do not exhibit frank agrammatism, and this may persist beyond the earliest stages of the illness. Because absence of frank agrammatism is a subsidiary diagnostic feature in the logopenic variant of PPA, this result has implications for differentiation of the nonfluent and logopenic variants, and indicates that PPA patients with nonfluent speech in the absence of frank agrammatism or AOS do not necessarily have the logopenic variant.

Vocabulary Size in Speech May Be an Early Indicator of Cognitive Impairment

Little is known about the relationship between mild cognitive impairment (MCI) and changes to language abilities. Here, we used the revised Hasegawa Dementia Scale (HDS-R) to identify suspected MCI in elderly individuals. We then analyzed written and spoken narratives to compare the language abilities between study participants with and without MCI in order to explore the relationship between cognitive and language abilities, and to identify a possible indicator for the early detection of MCI and dementia. We recruited 22 people aged 74 to 86 years (mean: 78.32 years; standard deviation: 3.36). The participants were requested to write and talk about one of the happiest events in their lives. Based on HDS-R scores, we divided the participants into 2 groups: the MCI Group comprised 8 participants with a score of 26 or lower, while the Healthy Group comprised 14 participants with a score of 27 or higher. The transcriptions of both written and spoken samples for each participant were used in the measurement of NLP-based language ability scores. Our analysis showed no significant differences in writing abilities between the 2 groups in any of the language ability scores. However, analysis of the spoken narrative showed that the MCI Group had a significantly larger vocabulary size. In addition, analysis of a metric that signified the gap in content between the spoken and written narratives also revealed a larger vocabulary size in the MCI Group. Individuals with early-stage MCI may be engaging in behavior to conceal their deteriorating cognition, thereby leading to a temporary increase in their active spoken vocabulary. These results indicate the possible detection of early stages of reduced cognition before dementia onset through the analysis of spoken narratives.

Profiling spontaneous speech decline in Alzheimer's disease: a longitudinal study

OBJECTIVE

This study aims to document the nature and progression of spontaneous speech impairment suffered by patients with Alzheimer's disease (AD) over a 12-month period, using both cross-sectional and prospective longitudinal design.

METHODS

Thirty one mild-moderate AD patients and 30 controls matched for age and socio-cultural background completed a simple and complex oral description task at baseline. The AD patients then underwent follow-up assessments at 6 and 12 months.

RESULTS

Cross-sectional comparisons indicated that mild-moderate AD patients produced more word-finding delays (WFDs) and empty and indefinite phrases, while producing fewer pictorial themes, repairing fewer errors, responding to fewer WFDs, produce shorter and less complex phrases and produce speech with less intonational contour than controls. However, the two groups could not be distinguished on the basis of phonological paraphasias. Longitudinal follow-up, however, suggested that phonological processing deteriorates over time, where the prevalence of phonological errors increased over 12 months. Discussion Consistent with findings from neuropsychological, neuropathological and neuroimaging studies, the language deterioration shown by the AD patients shows a pattern of impairment dominated by semantic errors, which is later joined by a disruption in the phonological aspects of speech.

Spoken Language Derived Measures for Detecting Mild Cognitive Impairment

Spoken responses produced by subjects during neuropsychological exams can provide diagnostic markers beyond exam performance. In particular, characteristics of the spoken language itself can discriminate between subject groups. We present results on the utility of such markers in discriminating between healthy elderly subjects and subjects with mild cognitive impairment (MCI). Given the audio and transcript of a spoken narrative recall task, a range of markers are automatically derived. These markers include speech features such as pause frequency and duration, and many linguistic complexity measures. We examine measures calculated from manually annotated time alignments (of the transcript with the audio) and syntactic parse trees, as well as the same measures calculated from automatic (forced) time alignments and automatic parses. We show statistically significant differences between clinical subject groups for a number of measures. These differences are largely preserved with automation. We then present classification results, and demonstrate a statistically significant improvement in the area under the ROC curve (AUC) when using automatic spoken language derived features in addition to the neuropsychological test scores. Our results indicate that using multiple, complementary measures can aid in automatic detection of MCI.