Does Light Exercise Help Memory? Ten Easy Minutes Woke Up the Hippocampus

After just ten minutes of very light exercise, people got better at telling similar things apart in memory, and the hippocampus, the brain’s memory centre, became more strongly linked with the regions around it. That is the result of a study by a team from the University of Tsukuba and the University of California, Irvine, published in PNAS (Proceedings of the National Academy of Sciences) in 2018.

In a Brain Mondo short video, I introduced this study in about 40 seconds. In this post I write about the procedure and numbers that did not fit in the video, and about how far we can trust the idea that moving a little makes things easier to remember.

Video: Exercise and the brain: Ten minutes of easy exercise (Japanese)

The video is in Japanese.

What the experiment was

The study was run by Kazuya Suwabe, Hideaki Soya and colleagues at the University of Tsukuba, together with Michael Yassa’s group at UC Irvine. The participants were 36 healthy young adults, split across two experiments. Experiment 1 had 20 people (mean age 20.6, 8 women), and experiment 2 had a different 16 people (mean age 21.1, 12 women).

In both experiments, each person went through an exercise day and a no-exercise day. The two days were at least 48 hours apart, and the order was randomised. For 24 hours beforehand, participants avoided exercise, alcohol and caffeine.

On the exercise day, they pedalled a recumbent cycle ergometer, a bike with a backrest, for ten minutes. The load was set to 30% of each person’s peak oxygen uptake, which the American College of Sports Medicine calls “very light”. The authors compare it to walking at a slow pace, or to yoga or tai chi. On the other day, participants sat on the same machine for ten minutes without pedalling.

The ten minutes was not arbitrary. The team’s earlier work had shown that ten minutes of exercise improves cognitive performance, so they used the same length.

What was measured: telling similar memories apart

The study looked at the ability to keep similar things apart as separate memories. In neuroscience this is called “pattern separation”, and it is thought to depend on a part of the hippocampus called the dentate gyrus.

In experiment 1, about five minutes after exercise or rest, participants looked at 196 photos of everyday objects, two seconds each, and said whether each was an indoor or outdoor object. Then they rested for 45 minutes watching a silent film. After that they saw 256 photos and answered “old”, “similar” or “new” for each. Of the 256, 64 were exact repeats, 128 were similar but slightly different, and 64 were new.

The key is the middle category. Say you saw a mug earlier, and now you see a mug of the same shape with a slightly different colour or detail. Can you say “similar” rather than “old”? The better you are at this, the more finely your memories are thought to be stored. The similar photos were analysed in three levels: high, medium and low similarity.

Result: the finer the difference, the bigger the gain

On the exercise day, performance on the high-similarity and medium-similarity photos was better than on the no-exercise day. On the low-similarity photos there was no difference.

That makes sense. Photos that are not very similar are easy to tell apart anyway, so there is no room for exercise to help. The difference showed up on the hard items, the ones that look alike and are easy to confuse.

In experiment 2, the other 16 people went into an MRI scanner within about five minutes of exercising and did a version of the same task. Again they did better on the exercise day, so the result of experiment 1 was repeated in a separate group.

What was happening in the brain

Experiment 2 used high-resolution functional MRI that can tell apart small subfields inside the hippocampus. The team compared brain activity when people correctly called a similar photo “similar”, on the exercise day versus the rest day.

On the exercise day, activity was higher in the DG/CA3 subfield of the hippocampus, in CA1 and the subiculum, and in the entorhinal and parahippocampal cortices next to the hippocampus. Other regions in the same scan, the perirhinal cortex, temporal pole and amygdala, did not change. So exercise did not just raise activity across the whole brain. Specific places changed.

On top of that, the activity of DG/CA3 moved more closely together with the left angular gyrus, left fusiform gyrus and left parahippocampal gyrus. This is called functional connectivity. The angular gyrus is involved in recollection, and the fusiform gyrus in recognising the shapes of objects.

This is the part the video described as “the more the connections strengthened, the bigger the memory gain”. The correlations between the increase in connectivity and the improvement in performance were 0.64 for the angular gyrus, 0.57 for the fusiform gyrus and 0.62 for the parahippocampal gyrus. With 16 people, there is no need to read too much into each number, but the pattern is clear: people whose connections strengthened more also improved more.

It was not stressful

The paper has a few details the video did not mention.

The first is stress. The team measured cortisol, a stress hormone, in saliva, and it did not differ between the exercise day and the rest day. The exercise was confirmed to be stress-free for the body. On the other hand, salivary alpha-amylase, a marker that the body has switched into a slightly more active mode, went up after exercise.

The second is mood. After exercise, people felt more “aroused”, in the sense of awake and lively. The more someone’s arousal rose, the more their performance on the high-similarity photos improved. The authors suggest that light exercise switches on the brain’s arousal system, and that this helps the hippocampus work.

The idea that light exercise is the right kind comes from the team’s animal work. In studies with rats running on a treadmill, stress-free mild exercise increased neuronal activity in the hippocampus, helped new neurons form in the dentate gyrus and improved spatial memory, while intense exercise suppressed those effects. This study tested that finding in people.

A year earlier, in 2017, the same team tested ten minutes of moderate exercise, at 50% of peak oxygen uptake. In young adults, it also improved the ability to tell highly similar photos apart. The 2018 study checked whether the same thing happens when the intensity is lowered even further.

There is also a broader look at exercise and memory. In 2013, Roig and colleagues published a meta-analysis of studies on aerobic exercise and human memory. A single bout of exercise had a moderate to large effect on long-term memory (standardised mean difference 0.52). A long-term exercise habit, by contrast, showed no clear effect on long-term memory. For short-term memory, a single bout had a moderate effect (0.26) and a long-term habit a small one (0.15).

The authors see a single bout and a long-term habit as two different but complementary strategies. A single bout gives a time-limited boost to memory for what you learn right after it. A habit prepares the groundwork of the memory system itself. The 2018 study is about the single bout.

What this study does not show

The result is interesting, but there are things to keep in mind.

The participants were healthy young adults around university age. This study did not test older adults or people with illnesses. The authors write that it may matter most for older people, but that is left for future work.
The samples were small. Twenty people in experiment 1, sixteen in experiment 2. It is reassuring that both showed the same trend, but each is a small study.
The effect was measured within about an hour of exercise. We do not know if it lasts for days.
It could not separate an effect on learning from an effect after learning. The authors list this themselves as a limitation.
The best intensity is unknown. Effects were seen at both 30% and 50%, but which level works best has not been settled.
What was measured was memory for telling similar things apart. It was not how many English words you can memorise, or the amount of memory as such.

This is not a story of “walk for ten minutes and get smarter”. It is a study showing that, after very light exercise, the memory for fine differences worked better for a while.

Using it in daily life

Even so, this study is easy to bring into daily life. All it needs is ten minutes and somewhere you can walk slowly. Here are some ways to use it.

Walk slowly for ten minutes before something you need to remember. The exercise in the study was very light, about the pace of a slow walk. You do not need to sweat. A lap of the neighbourhood, or walking one station further, is probably close to the right intensity.
Use it before memorising things that are easy to mix up. What improved in this study was the ability to tell similar things apart. It can be used for things like words with similar spellings, medicines with similar names or parts with similar model numbers.
Do not overdo it. In the animal studies, intense exercise suppressed the effect. We do not know if the same holds for people, but there is at least no reason to think that only hard exercise counts.

Brain Mondo is a series of daily short videos about the brain, each based on a research paper. BrainDojo is an app for training calculation, memory, attention and more in a few minutes a day, and it includes memory training where you keep in mind something you saw a moment ago. Trying one right after you come back from a walk makes a small experiment of your own out of this study.

References

Suwabe K, Byun K, Hyodo K, et al. Rapid stimulation of human dentate gyrus function with acute mild exercise. Proc Natl Acad Sci U S A. 2018;115(41):10487-10492.
Suwabe K, Hyodo K, Byun K, et al. Acute moderate exercise improves mnemonic discrimination in young adults. Hippocampus. 2017;27(3):229-234.
Roig M, Nordbrandt S, Geertsen SS, Nielsen JB. The effects of cardiovascular exercise on human memory: a review with meta-analysis. Neurosci Biobehav Rev. 2013;37(8):1645-1666.