Semantic mapping in Alzheimer's disease: Measuring semantic distance using Natural Language Processing
Bibliographic record
Abstract
BACKGROUND: Semantic knowledge deficits are a common characteristic in people with dementia, yet how the brain maps words and meaning remains unclear. Semantic distance is a measure used to quantify the distance between two concepts in semantic space2. Through the use of verbal fluency data, we examined the semantic distance of animal names to gain insight into semantic mapping patterns. Our objective was to determine whether semantic distance can serve as a linguistic marker to identify differences in semantic representations between individuals with Alzheimer's disease (AD) and healthy controls. METHOD: Using verbal fluency data from DementiaBank1, Natural Language Processing (NLP) techniques were applied to generate the semantic distance for individual responses for all participants. We then compared the average semantic distance across groups using two models. Model 1: the AD group (N = 76) and the healthy control group (N = 76) were matched on age. Model 2: the AD group (N = 51) and healthy control group (N = 51) were matched for age, sex, and education. Analysis was completed using Python. RESULT: Results in Model 1 showed that the AD group had a significantly shorter semantic distance than the healthy control group, with a medium effect size. Findings in Model 2, which involved a more strict participant matching process, showed similar results to Model 1, with a small effect size. CONCLUSION: Semantic distance, as measured using NLP techniques, has the potential to distinguish individuals with Alzheimer's disease from those without. A shorter semantic distance, which was observed in the AD group, may reflect an inefficient and disrupted semantic network. An impaired semantic system may restrict access to semantic representations that are more distant from the core concept. Enhancing understanding of semantic systems in AD may lead to a more nuanced linguistic profile that can be used to monitor disease progress. This study contributes to the growing evidence for the role of computational tools in analyzing language-based assessments to advance understanding of semantic deficits in dementia. References 1. DementiaBank. https://dementia.talkbank.org/ 2. Reilly, J., et al. (2024). https://doi.org/10.3758/s13423-024-02556-7.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.021 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".