Linking silent pause locations in speech to amyloid and tau deposition in cognitively unimpaired individuals.
Level 4 - case-series / case-control
Propensity score-matched cross-sectional case-control comparison of biomarker-positive and biomarker-negative individuals
PubMed 40934091 · doi:10.1177/13872877251371720
What was done
From the Wisconsin Registry for Alzheimer's Prevention, 26 cognitively unimpaired individuals with amyloid and tau PET positivity (A+/T+) were propensity-score matched 1:1 on age, sex, literacy, and race with 26 biomarker-negative (A-/T-) individuals (total n = 52; mean age 67.5 years, 75.6% female, 91% non-Hispanic white). Connected speech was recorded during the Cookie Theft picture description task closest to PET imaging. Silent pauses were annotated for rate and duration at utterance-initial, clause-initial, and within-clause locations (stratified by nouns, verbs, and adjectives/adverbs) and evaluated using Mann-Whitney U tests with Cliff's delta (δ).
What was found
A+/T+ individuals had statistically significantly longer adjective/adverb-initial pauses compared to A-/T- individuals (p = 0.016, Cliff's δ = 0.39). The preclinical Alzheimer's cognitive composite did not significantly differentiate the two groups (p = 0.347, Cliff's δ = 0.15).
Why it matters
Fine-grained analysis of silent pause locations during natural speech may serve as a non-invasive digital biomarker for preclinical Alzheimer's neuropathology, potentially showing sensitivity before standard cognitive composite batteries.
Limits
The sample size was small (26 pairs, 52 participants). The cohort lacked demographic diversity, consisting predominantly of female and non-Hispanic white participants. The study was cross-sectional and evaluated speech from a single picture description task rather than longitudinal or naturalistic conversation.