Finding clues for Alzheimer’s disease in online language - Poster Presentation - Biomarkers

Q: What data source and study approach were used to investigate Alzheimer’s disease online language?

The study used Reddit posts identified through a Self-Disclosure approach and focused on self-reported Alzheimer’s disease cases.

Q: How were linguistic features scored in the analysis?

Texts were analyzed with an automatic scoring method based on DistilBERT that quantifies internal (episodic) details and external details (repetitions and metacognitive statements).

Q: What language differences did the study find between patients and controls?

Patients used significantly more internal words and showed increased use of past tense, emotional, function, and social words, while controls used more analytic language reflecting logical and formal thinking.

Sep 30, 2026

Upload by 4398046744996

Natural Language Processing (NLP) can detect linguistic markers linked to MCI and Alzheimer’s disease, yet prior studies often relied on cross-sectional designs and curated texts such as novels and speeches. This work examines whether cognitive decline emerges in more ecologically valid, continuous social media behavior. Using Reddit self-disclosures, researchers analyzed posts from 42 users (20 diagnosed, 22 controls) to test linguistic shifts over time.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages2

Reading Time5 min

AccessFree to download

Abstract

Background

Method

Results

Conclusion

DOI: 10. 1002/alz70856_105272 BIOMARKERS POSTER PRESENTATION BIOMARKERS (NON-NEUROIMAGING) Finding clues for Alzheimer’s disease in online language Shira Darchi1 Yuval Pinter2 Talya Sadeh2 1 Ben Gurion University of the Negev, Beer Sheva, Israel, Israel 2 Ben Gurion University of the Negev, Beer Sheva, 8410501, Israel Correspondence Shira Darchi, Ben Gurion University of the Negev, Beer Sheva, Israel, Israel. [Email:](Email: shirah770@gmail.com)[ shirah770@gmail.com](Email: shirah770@gmail.com) Abstract Background: Natural Language Processing (NLP) has shown promise in identifying linguistic markers of MCI and Alzheimer’s Disease (AD). Prior studies have achieved classification accuracy of 80%-85% but were often limited by cross-sectional designsand reliance on curated texts like novels and speeches. Social media data offer a more ecologically valid, continuous assessment of linguistic changes. This study explores whether cognitive decline can be detected through online activity on Reddit, contributing to developing amore complex model capable of detecting early changes indicative of cognitive deterioration from online activity. Method: We applied a Self-Disclosure approach, commonly used in research for identifying self-reports of various mental illnesses. We identified self-reported AD cases on Reddit and extracted their posts. Data from 42 Reddit users (20 diagnosed with AD, 22 matched controls) were collected and analyzed to examine both online engagement patterns and linguistic shifts over time. Result: We analyzed the texts using an automatic scoring method based on DistilBERT. This method measures the amount of two types of details in a text: internal details (episodic information)and external details(repetitionsand metacognitive statements). A two-way MANOVA showed a significant group effect on word usage (F(2, 4243) = 41. 90, p < .001). Follow-up analyses revealed that this effect was driven by internal words, with control using significantly more internal words than diagnosed individuals (F(1, 4244) = 52. 78, p < .001; mean difference = -13. 30). No significant differences were found for external words between the groups (F(1, 4244) = 0.09, p = . 76). An analysis of LIWC psychological categories found that patients and controls differed in their language. Most interestingly, compared to controls, patients displayed increased use of past tense, emotional, function, and social words. The controls used more analytic language reflecting logical, formal thinking, cognitive processes such as reasoning and analysis. Conclusion: Significant differences in online activity and language patterns between AD patients and controls suggest that analyzing social media activity can provide This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2025 The Alzheimer’s Association. Alzheimer’s & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer’s Association. Alzheimer’s Dement. 2025;21(Suppl. 2):e105272. [https://doi.org/10.1002/alz70856_105272](https://doi.org/10.1002/alz70856_105272) wi[leyonlinelibrary.com/journal/alz](leyonlinelibrary.com/journal/alz) 1of2 valuable insights into cognitive decline. By leveraging such linguistic and behavioral markers, future models may enable early detection of dementia in a scalable and non-invasive manner.

FAQs

What data source and study approach were used to investigate Alzheimer’s disease online language?

The study used Reddit posts identified through a Self-Disclosure approach and focused on self-reported Alzheimer’s disease cases.

How were linguistic features scored in the analysis?

Texts were analyzed with an automatic scoring method based on DistilBERT that quantifies internal (episodic) details and external details (repetitions and metacognitive statements).

What language differences did the study find between patients and controls?

Patients used significantly more internal words and showed increased use of past tense, emotional, function, and social words, while controls used more analytic language reflecting logical and formal thinking.

Sep 30, 2026

Upload by 4398046744996