Automated free speech analysis reveals distinct markers of Alzheimer’s and frontotemporal dementia
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Dementia can disrupt how people experience and describe events as well as their own role in them. Alzheimer’s disease (AD) compromises the processing of entities manifested by nouns, while behavioral variant frontotemporal dementia (bvFTD) entails a depersonalized perspective, signaled by an increase of third-person references. Yet, no study has examined whether these patterns can be captured in spontaneous discourse via natural language processing tools (NLP). We asked persons with AD (n = 21), bvFTD (n = 21), and healthy controls (n = 21) to narrate a typical day of their lives and calculated the proportion of nouns, verbs, and first- or third-person markers via part-of-speech and morphological tagging. Inferential statistics and machine learning were used for group-level and subject-level discrimination. The above linguistic features were correlated with patients’ cognitive outcomes, captured through the Montreal Cognitive Assessment (MoCA). We found that, compared with HCs, AD (but not bvFTD) patients produced significantly fewer nouns, while bvFTD (but not AD) patients used significantly more third-person markers. Machine learning analyses showed that these features identified individuals with AD and bvFTD (AUC = 0.76). No linguistic feature was significantly correlated with MoCA scores in either patient group. Taken together, we suggest that differential markers of AD and bvFTD can be automatically detected in spontaneous routine descriptions. By targeting specific features linked to each disorder’s cognitive profile, our approach favors interpretability for enhanced syndrome characterization, diagnosis, and monitoring.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it