Scientists talked about the connection of speech features with brain aging

Speech patterns may reflect age-related brain changes, memory status, and other indicators of biological aging. This conclusion was reached by an international group of scientists who analyzed the data of almost 3,000 people from five Latin American countries. This was announced by Trinity College Dublin with reference to a study published in the journal Science Advances.

How scientists determined age by speech

The researchers have developed a machine learning model called a "speech clock." It evaluates a person's age based on a combination of acoustic and linguistic characteristics.: it takes into account how his speech sounds, what words he chooses and how he builds statements.

The study included 2,928 Spanish-speaking participants from Argentina, Chile, Colombia, Mexico and Peru. Among them were healthy adults, people with mild cognitive impairment, Alzheimer's disease and various forms of frontotemporal dementia.

The algorithm analyzed hundreds of features, including speech tempo, pauses, voice pitch, emotional content, vocabulary, accuracy of meaning transmission, volume and organization of utterances. Based on these data, the model calculated the estimated age of the person.

The researchers then compared the estimate with the participant's actual age. If the algorithm rated a person as older, it meant that their speech profile looked "older" than expected. It was this gap that the researchers compared with health indicators. By itself, it does not mean that the human body is literally older by the same number of years.

What did it show about the state of the brain

An older speech profile was associated with age-related brain characteristics determined by structural and functional neuroimaging data. The scientists also found a link with epigenetic age, an assessment of biological aging based on age—related changes in DNA methylation. Three independent models, known as epigenetic clocks, were used for this purpose.

Scientists have reported on the role of different brain regions in the perception of illness

The greater the gap between speech and actual age, the worse the overall indicators of cognitive functions, several types of memory, and the ability to cope with everyday tasks turned out to be. The communication also concerned executive functions necessary for planning and monitoring actions.

At the same time, the results were not limited to speech and language tasks. Speech age was also associated with the performance of tests that assessed other cognitive abilities.

According to him, speech can reflect both the passage of calendar time and changes in cognitive functions, the state of the brain and body, as well as the accumulated influence of the social environment.

What differences were found in dementia?

The smallest gap between speech age and actual age was observed in healthy participants. This indicator was higher in people with Alzheimer's disease and forms of frontotemporal dementia.

The comprehensive assessment, which combined acoustic and linguistic characteristics, distinguished clinical groups better than individual speech features. According to the abstract of the scientific work, among the groups of patients under consideration, the gap increased from Alzheimer's disease to forms of frontotemporal dementia, especially those with predominant language impairments.

In participants with Alzheimer's disease, the speech index was also associated with the concentration of p-tau217 in blood plasma, one of the biomarkers of pathological changes characteristic of this disease. However, the differences found so far do not allow the use of a "speech clock" to make a diagnosis. The study shows a link between speech characteristics and health status, but it does not prove that the algorithm is able to reliably detect a disease in a particular person.

How social conditions are reflected in speech

Scientists have also studied the relationship between speech age and living conditions. Among healthy participants and people with Alzheimer's disease, a more "senior" speech profile was associated with an unfavorable social exposure. This is what researchers call the totality of social influences that a person experiences throughout his life.

This assessment included education, financial situation, access to adequate food and medical care, as well as early life circumstances. The results indicate that speech patterns may be related not only to diseases and age, but also to the accumulated influence of living conditions.

At the same time, the study does not establish which factor causes changes in the speech profile. It reveals statistical relationships that need to be studied in more detail.

Keep it in mind: headphones and smart bracelets will detect brain disorders

How will the new wearable technology for assessing cognitive health help users?

Will it be possible to monitor aging by voice recording

The main practical advantage of the method is accessibility. Many ways to assess biological aging require brain examinations, blood tests, molecular studies, or special clinical tests. Speech can be recorded remotely and repeatedly, without invasive procedures and at low cost.

According to the authors, in the future, "speech clocks" could complement more expensive methods of assessing aging, including in combination with other biomarkers. This approach is particularly interesting for regions where access to complex surveys is limited.

However, additional studies are needed before clinical use. The work was mostly cross-sectional: scientists compared the indicators of participants, rather than tracking the development of changes over the years. Therefore, it is not yet known whether the "senior" speech profile allows predicting subsequent memory impairment or the development of dementia. In addition, the model needs to be tested in other languages, in different cultural settings, and when analyzing more natural everyday speech.