PARIS - Bioactive ingredients found in cocoa sharply reversed age-related memory decline in a group of volunteers, scientists reported on Sunday.
The compounds, called flavanols, were taken in a specially prepared cocoa drink, according to an experiment published by the journal Nature Neuroscience. Over three months, 37 healthy volunteers aged 50-69 had a daily drink containing either a high dose of flavanols - 900mg - or a low dose, 10mg.
The scientists carried out brain imaging, measuring blood volume in a key part of the hippocampus called the dentate gyrus, a region of memory formation whose performance typically declines as one ages.
They also carried out memory tests before and after the volunteers started with the drink.
The tests entailed a 20-minute pattern-recognition exercise designed to assess a type of memory controlled by the dentate gyrus.
The high-flavanol group notched up major memory improvements and an increase in blood flow to the dentate gyrus.
"If a participant had the memory of a typical 60-year-old at the beginning of the study, after three months, that person on average had the memory of a typical 30- or 40-year-old," said Scott Small, a professor of neurology at Columbia University Medical Centre in New York.
More work in a bigger group is needed to verify these early findings, he cautioned.
Flavanols have excited great interest. They dangle the possibility of tackling age-related memory loss in the world's fast-growing population of elderly without using drugs.
The compounds exist in grapes, blueberries and other fruit, as well as in some vegetables and teas, but the type of flavanol and the amount vary widely.
Previous studies in mice showed that the class of flavanols found in cocoa boosts the performance of the dentate gyrus.
"The dentate gyrus in humans and mice are very similar," Prof Small said in an e-mail exchange with AFP.
"I suppose that our study does show, for the first time, that flavanols improve the function of humans' dentate gyrus, particularly in ageing humans."
The findings apply to normal memory loss - things such as forgetting the names of new acquaintances or where one has left one's glasses - which usually becomes noticeable when people reach their 50s or 60s. They do not apply to memory loss caused by diseases such as Alzheimer's.
The cocoa drink was prepared by a large United States food corporation, which partly supported the research. The study was published in a peer-reviewed journal.
The firm used a proprietary process to extract flavanols from cocoa beans. During conventional processing, most of the flavanols are lost from the raw plant.
Prof Small said it was still too early to make dietary recommendations for flavanols, but "certainly I would not suggest that people consume more chocolate".
"That would be a mistake," he said. "The amount of flavanols that are found in chocolate is minuscule compared to the very high amount of extracted flavanols that our subjects consumed.
The same is true for most other food or teas," said Prof Small. "Hopefully, in the future, a food source or a specific diet will be identified that contains very high amounts of the specific flavanols we studied."