The way people speak during everyday conversations may provide valuable insights into brain health, according to new research from Baycrest University, the University of Toronto, and York University. Scientists have discovered that subtle speech characteristics, such as pauses, fillers such as “um” and “um”, and difficulty with word retrieval, are closely related to executive function, the mental abilities involved in memory, planning, attention, and flexible thinking.
The findings provide some of the strongest evidence to date linking natural speech patterns and key cognitive abilities. This study also extends previous research showing that older adults who speak faster tend to maintain strong thinking skills over time (Wei et al., 2024).
“The message is clear: Speech timing is not just a matter of style, but a sensitive indicator of brain health,” said Jed Meltzer, Ph.D., a senior researcher at Baycrest’s Rotman Institute and lead author of the study, titled “Natural Speech Analysis Can Reveal Individual Differences in Executive Function Across the Adult Lifespan.”
AI analysis reveals hidden cognitive clues
In this study, participants were shown detailed images and asked to describe them in their own words. They also completed established tests designed to measure executive function.
The researchers then used artificial intelligence to closely examine the audio recordings. The AI system detected hundreds of subtle audio features, including the length and frequency of pauses, the use of filler words, and timing-related patterns in the audio. These markers consistently predicted participants’ performance on cognitive tests, even after researchers controlled for factors such as age, gender, and education.
Speech patterns and dementia risk
Executive function naturally weakens with age and is often affected in the early stages of dementia. However, standard cognitive tests can be difficult to repeat frequently because they are time-consuming and often improve simply by becoming familiar with the test.
Natural voices may offer an easier alternative. Because speaking is part of everyday life, it can be measured repeatedly, discreetly, and on a large scale. The researchers also noted that audio provides valuable insight into processing speed and overall cognitive function in real-world situations without the strict time limits that are common in many traditional cognitive assessments.
The research team believes that speech analysis could eventually become a practical way to identify people whose cognitive decline progresses faster than expected and who are at high risk of developing dementia.
“This study sets the stage for an exciting opportunity to develop tools to help track cognitive changes in the clinic and even at home. Dementia involves progressive brain degeneration, which can slow its progression, so early detection is critical for treatment and intervention,” said Dr. Meltzer.
Future research on brain health monitoring
The researchers say further long-term studies are needed to track changes in speech over time and distinguish between normal aging and early signs of disease. They also suggest that combining voice analysis with other health measures could make early detection of cognitive decline more accurate, practical, and widely available.
This research was supported by the Mitacs Accelerate program and the Natural Sciences and Engineering Research Council of Canada (NSERC).
