To sleep, perchance to dream, perchance to remember

By Ed Yong | April 22, 2010 12:00 pm

MazeIt seems obvious that thinking about something will help you to remember it better, but it might be more surprising to know that this process works even more efficiently when we’re asleep. Erin Wamsley from Harvard Medical School has shown that people who are trained to navigate a virtual maze learn the best route through it more quickly if they dream about their experiences.

The last decade of research has clearly shown that sleep is one of the best aide memoires that we have. During this nightly time-out, our brain can rehearse information that it has picked up during the day and consolidate them into lasting memories. Wamsley’s new study supports that idea but it also shows that dreaming while you nap can strengthen our memories even further.

Maze_taskShe asked 99 volunteers to learn the layout of a complex virtual maze so that they could reach a specific landmark after being dropped at a random starting point. Five hours later, they were tested again. Those who had stayed awake in the intervening time beat their previous times by 26 seconds, but those who had had a 90-minute nap improved by a whopping 188 seconds.

But those who dreamt about the task fared even better. Wamsley either asked her recruits directly about whether they dreamt about the labyrinth, or asked them to give an open-ended report of everything that was going through their mind while they were asleep. Either way, those who had thought about the maze during their short nap improved far more than those who didn’t. They also beat those who mentally replayed their training again while awake. These striking results suggest that there’s something special about the mental rehearsals that happen during dreaming sleep.

However, the dreams weren’t straightforward replays of previous experiences. When the volunteers described their dreams, they didn’t mention specific objects, locations or routes through the maze. Instead, some talked about isolated parts of their experience, like the music or the prospect of a re-test. Others discussed tangential memories, like other mazes or being stuck in a bat cave (heh). Interestingly, scientists have found the same thing in rodents. A sleeping rat will show similar brain activity to its prior bout of wakefulness, but the two patterns won’t quite match up.

It’s a very exciting set of results. We know that forcefully repeating pieces of information can make them stick in our minds. But this study shows that we can do even better through a passive unintentional process where the material to be learned is only tangentially referenced!

Wamsley doesn’t think that dreams themselves improve our memory – they’re a side effect of processes that do this. While our bodies lie still, our brain is busily working away processing the day’s memories. The brain doesn’t simply replay those memories, as the volunteers’ descriptions show. Instead, Wamsley thinks that it works the old into the new, slowly integrating parts of our recent experiences into our more established memory networks. Dreams, then, are like the tip of a mental iceberg – the visible sign of a tremendous body of work going on behind-the-scenes.

Certainly, people who dreamt about the maze at all included those who found the training most difficult. It seems that their initial difficulties meant that their brains were more likely to continue processing the information they learned after they nodded off. Based on this idea, it’s tempting to suggest that the best time to study intensely is just before you go to sleep, or that a quick nap after an afternoon swotting session is a good idea.

Reference: Current Biology

More on sleep:


Comments (8)

  1. karen

    A timely article for the students out there, and another reason to get a good nights sleep.

  2. Robert E Harris

    I used to tell my university students that the way to prepare for exams was to study and then get a good night of sleep. They generally thought I was nuts. I suppose the all night cramming still goes on, but it is stupid.

  3. badnicolez

    Great! I’m going to go to sleep tonight thinking about where I put my keys two weeks ago.

  4. Matt C.

    I’ve always felt this was an important part of the learning process. Memories are reinforced when they’re accessed. In many ways, our “concious” executive processes can get in the way of this activity by keeping our thoughts “in check”, moving through our head one at a time.

    Sleep, if I’m not citing outdated info, is characterized by waves of impulses moving across the brain, kind of tickling the whole thing, which seems to serve the twofold purpose of selectively reinforcing memories and “reseting” the brain.

    Of course, since these memories are just being fired off by the waves moving through the brain, there’s a lot less concious interference (great for focus, but not for absorbing new knowledge).

    I’ve seen the benefits of this myself, as a programmer. I’ll be stuck on a problem and fruitlessly ponder it during the day. That night, I’ll have a dream related to the programming or problem solving, and the subsequent day the solution will suddenly become obvious. All the bits of information related to the problem are suddenly all clear at once.

    That’s not to say the solutions in my dreams are help. They’re generally of the “I need to knit this code onto my monitor, but the cat is unraveling it!” variety… but of course, it’s not really your cat, it’s your third grade teacher. Stupid dreams.

  5. zackoz

    Did you get home OK?
    Anyway, I hope Australia was kind to you.

  6. Hi Ed,
    This is a great post, and it is great to see this research supporting and reinforcing what we say as educators. Before a big test, the best thing to do is get a good FULL night’s sleep! Do your cramming two-nights before, not the night before.

    Although teachers are wise at knowing what works for learning, it is a great time for research to actually prove and support the teachers intuitions and ‘gut’ feelings….mind-brain and education science field is here!

  7. timew

    People jumping to making a “connection” between dreams and memory have always been around, and you can’t justify it with such vague tests.
    I’m sure someone could find the same connection between memory and spicy food, with a test very similar to the one above. It doesn’t mean anything.

    On the other hand, cases like this: indicate that dreaming has as much to do with memory, or even less, as the digestive process.

  8. Great post!
    I am a university student with a roommate that sums up a whole semester in 6 hours. He will be taking his next paper in 30 minutes and he only started his revision marathon last night after watching a movie. That apparently doesn’t help but I guess he doesn’t care much about the health of his brain , but I wonder how much can he remember later~~ o.O


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Not Exactly Rocket Science

Dive into the awe-inspiring, beautiful and quirky world of science news with award-winning writer Ed Yong. No previous experience required.

See More

Collapse bottom bar