Stimulating the Olfactory System as a Way to Improve Late Life Cognitive Function
The research noted here adds to evidence for the lack of use of the senses to contribute to age-related declines in cognitive function: a sort of "use it or lose it" proposition for the brain that becomes especially pronounced in later life. This effect is better studied in the context of age-related deafness, given the sizable amount of data on hearing aid use. Here, however, researchers focus on the sense of smell, and find that cognitive function can be improved by stimulation via scents.
When a fragrance wafted through the bedrooms of older adults for two hours every night for six months, memories skyrocketed. Participants in this study reaped a 226% increase in cognitive capacity compared to the control group. The project involved men and women aged 60 to 85 without memory impairment. All were given a diffuser and seven cartridges, each containing a single and different natural oil. People in the enriched group received full-strength cartridges. Control group participants were given the oils in tiny amounts. Participants put a different cartridge into their diffuser each evening prior to going to bed, and it activated for two hours as they slept.
People in the enriched group showed a 226% increase in cognitive performance compared to the control group, as measured by a word list test commonly used to evaluate memory. Imaging revealed better integrity in the brain pathway called the left uncinate fasciculus. This pathway, which connects the medial temporal lobe to the decision-making prefrontal cortex, becomes less robust with age. Participants also reported sleeping more soundly.
Scientists have long known that the loss of olfactory capacity, or ability to smell, can predict development of nearly 70 neurological and psychiatric diseases. These include Alzheimer's and other dementias, Parkinson's, schizophrenia, and alcoholism. Evidence is emerging about a link between smell loss due to COVID-19 and ensuing cognitive decrease. Researchers have previously found that exposing people with moderate dementia to up to 40 different odors twice a day over a period of time boosted their memories and language skills, eased depression and improved their olfactory capacities. The team decided to try turning this knowledge into an easy and non-invasive dementia-fighting tool.
"The study was supported by Procter & Gamble."
"People in the enriched group showed a 226% increase in cognitive performance..." what does that even mean?
'*Scientists have long known* that the loss of olfactory capacity, or ability to smell, can predict development of nearly 70 neurological and psychiatric diseases. These include Alzheimer's and other dementias, Parkinson's, schizophrenia and alcoholism. ' !!!
What a fun read!
That 226% number is the relative improvement, virtually meaningless in this context. Impossible for me to figure out if there was actually a substantial improvement in the treatment group as I am unfamiliar with the test used.
Not to be a bummer, but looking at the table in the paper https://www.frontiersin.org/articles/10.3389/fnins.2023.1200448/full , they ran a dozen closely related cognitive tests, found an statistically significant improvement on only one, and based the paper's argument (and paper coverage) on that one. You'd expect significant improvements across at least a few of those tests if the treatment had an effect on cognition...
There's a model in which there is a consistent, important treatment effect narrowly focused on whatever is measured specifically by that test, which would be a rather striking discovery indeed. But by far the likelier reading of this is that they tried it, ran a lot of tests, and chose to talk only about the one who showed a difference. Without more testing, I can't say I trust there's anything there.
Kudos to them for including the table for all tests. Many negative kudos for only using a graph with the one that worked - never mind the way it was written.
If anyone is wondering, the multiple-testing correction gives a p-value of 0.24 for that test:
R> p.adjust(c(0.89, 0.79, 0.76, 0.73, 0.22, 0.02, 0.76, 0.06, 0.07, 0.80, 0.53, 0.61))
[1] 1.00 1.00 1.00 1.00 1.00 0.24 1.00 0.66 0.70 1.00 1.00 1.00
Ugly. I wouldn't bet a bent penny on this replicating.