(Boston)—Clinicians and researchers use standardized tests to assess memory and thinking ability. In one commonly used memory test, a short story is read to patients and then they try to recall it from memory. Their response is scored by counting the number of story details the patient remembers. However, this scoring ignores a wealth of information about 'how' the person responded; qualities such as the use of complex or simple sentences, repetition of story details, or admitting they forgot pieces of the story with comments such as "I forgot her name."
A new study from researchers at Boston University Chobanian & Avedisian School of Medicine has shown that language patterns during a memory test can signal current cognitive impairment and also predict future cognitive performance. They believe this early warning system gives doctors a better idea of a patient's future cognitive health, allowing for earlier interventions to help prevent further decline.
"Our method allows us to go beyond right or wrong scoring and capture qualities of the spoken test responses which shows us how a person thinks and remembers information and whether they notice their own mistakes," explains lead author Seho Park, a postdoctoral associate at the school. "These behavioral differences have great potential to help us detect cognitive impairment earlier and pinpoint different types of cognitive impairment."
The researchers analyzed digital voice recordings of participants' spoken responses to a story recall test in the Long Life Family Study . Using natural language processing methods, the team identified subtle differences, such as a lower recall of certain story details and the addition of comments not related to the story, that were more common in the responses from people with cognitive impairment than those with normal cognition. Next, they grouped these language markers into a comprehensive speech profile. They found that this speech pattern successfully predicted lower cognitive performance seven years later.
The team hopes that these and other computational methods will lead to more widespread use of sensitive, clinically-valid tests of cognition beyond specialized clinical and research settings such as in primary care settings and smartphone apps for self-assessments of cognitive function. "Gold standard tests of cognitive function require significant mental effort to complete which helps draw out subtle signs of cognitive dysfunction. With this work, we are developing ways to not only automate scoring of these tests, but to also get rich behavioral information that may be an earlier marker of cognitive dysfunction, well before someone starts scoring poorly on the test," says corresponding author Stacy Andersen, PhD, associate professor of medicine.
These findings appear in the Journal of the International Neuropsychological Society.
Funding for this research was provided by the National Institute on Aging (grants U01AG023746, U01AG023712, U01AG023749, U01AG023755, U01AG023744, U19 AG063893, and K01 AG057798).
