Most Disappointing Headline Ever — No, chocolate will not improve your memory

Fruit chocolate. Ian Musgrave

This morning I woke to the most disappointing headline ever. “A Bite to Remember? Chocolate Is Shown to Aid Memory” was prominent in my twitter feed, highlighting a recent paper in Nature Neuroscience. Excited, I quickly read the article only to have my hopes dashed, the headline was flat out wrong.

Other news sites were a bit better in their presentation, but basically at lot of spin was applied to an interesting but preliminary finding (and chocolate was nowhere to be seen).

The paper in question has a rather more prosaic title “Enhancing dentate gyrus function with dietary flavanols improves cognition in older adults”. The researchers looked at a memory test and the flow of blood through a brain region important for memory when people consumed anti-oxidant flavanol supplements.

Despite the headlines, no chocolate or cocoa drinks were consumed in the study, just capsules containing either 900 mg flavanols extracted from cocoa and 138 mg of the specific flavanol epicatechin (the high flavonol group) or 10 mg of cocoa flavanols and less than 2 mg of epicatechin (the low flavanol group).

The people consumed these capsules for 3 months, then underwent magnetic resonance imaging to determine blood flow in the Dentate Gyrus (the brain region important for memory I mentioned above), and then tried to remember various complex images flashed at them on a computer screen.

The blood flow in the Dentate Gyrus of the high flavanol group increased, while the low flavanol group showed no change. Blood flow in this area correlates with better performance on memory tasks.

Looking directly at memory, one type of memory test, remembering complex objects, was improved. People who took the high flavanol capsules were around 25% faster at recognising these objects than those who took the low flavanols (about two seconds rather than just over two and a half seconds). A different memory task (delayed retention) showed no difference.

The study had several good aspects; it was a blinded, randomised study (in some other studies people could easily tell they were getting the flavanol containing material, leading to bias). The researchers put a lot of effort into establishing the reproducibility of their study, and correlating age with the blood flow measurements and tests of memory. The memory test they used was for a specific type of memory (novel stimulus recognition) that declines with age.

On the other hand, the study numbers were small (only 37 people all up, with 19 in the high flavanol group and 18 in the low flavanol group).

The history of memory and Alzheimer’s research is littered with interventions that were promising in small studies, but evaporated like snow in summer when larger, more representative studies were performed. Also, the subjects weren’t that old (average ages 58 years for the high flavanol group, 59 for the low flavanol group).

As well, the statistical significance of the memory results was marginal. A false positive result cannot be ruled out, especially with the low subject numbers.

What does this all mean for the average person? Well, the results are consistent with other studies of flavanols in animals (high levels increase memory performance in snails and rats), and some human studies.

However comparing the doses and different memory tests is not straight forward. The current study had specific levels of epicatechin in their flavanols, other studies just looked at total flavanols and could have different compositions, also the doses varied quite a bit. And not all studies show an effect (see this one on rats and this one on humans).

Don’t rush out to buy flavanol capsules just yetIan Musgrave

While encouraging, this study doesn’t mean you should rush out and buy bottles of flavanol capsules, it is still way too early for that yet. Wait for larger, longer term studies.

But what about chocolate and cocoa, the hooks in the misleading headlines?

The flavanol concentrations in chocolate and cocoa are well below that in the high flavanol supplementation group. You will not get any memory boost from the low levels of flavanols present. But there is good evidence chocolate and cocoa produce small but significant beneficial effects on the heart and circulation (seehere and here), so there is no need to shun them.

So enjoy your chocolate and cocoa, your memory may not be better, but you don’t need that as an excuse.


Ian Musgrave is a molecular pharmacologist who works at the University of Adelaide, Australia. he has a broad interest in the understanding of how surface receptors can modulate neuronal function and survival, as well as interests in natural product pharmacology and drug design. His interests include evolutionary biology, science communication (committee member of SA Science Communicators) and am an avid amateur astronomer. His astronomy blog is astroblog.

Dr. Musgrave did his postdoctoral work in the Institute of Pharmacology at the Free University of Berlin in 1991-1994, returning to Australia in 1994. he has collaborated with Professor John Carver, Dr. Scot Smid and Professor Colin Barrow on using natural products to attack the neurodegeneration process in Alzheimer’s Disease. Dr. Musgrave was recently elected as convenor of the Toxicology Special Interest Group of the Australasian Society for Pharmacology and Toxicology and is also collaborating with Professor Roger Byard on the forensic implications of herbal medicines. He has previously held NH&MRC funding and is currently hold ARC funding.