source: sciencedaily ai: your “um” and pauses could reveal early dementia risk
level: research
researchers from baycrest, the university of toronto, and york university found that subtle speech features such as pauses, filler words like 'um' and 'uh', and word-finding difficulties are closely linked to executive function. executive function includes memory, planning, attention, and flexible thinking. the study used ai to analyze recordings of people describing images, detecting hundreds of speech features. these features predicted cognitive test performance even after adjusting for age, sex, and education.
executive function naturally declines with age and is often affected early in dementia. traditional cognitive tests can be time-consuming and prone to practice effects. speech analysis offers a simpler, repeatable way to monitor cognitive health in daily life. it captures processing speed and real-world cognitive function without strict time limits. the researchers suggest speech-based tools could identify people whose cognitive decline is progressing faster than expected, potentially flagging higher dementia risk.
the study provides strong evidence linking natural speech patterns to key cognitive abilities. it builds on earlier work showing faster speech in older adults is tied to better thinking skills over time. the team believes speech analysis could eventually be used in clinics or at home for early detection. they note that more long-term studies are needed to track speech changes over time and distinguish normal aging from early disease. combining speech analysis with other health measures could improve accuracy and accessibility of early cognitive decline detection.
why it matters: speech-based ai tools could enable frequent, low-cost cognitive monitoring at home, helping catch dementia earlier when interventions may be more effective.
source: sciencedaily ai: your “um” and pauses could reveal early dementia risk