About the SA Blog Network



Opinion, arguments & analyses from the editors of Scientific American
Observations HomeAboutContact

To sleep, perchance to dream–and learn

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

dream nap sleep learn memory improveDreams might be helping your brain do more than express Freudian fixations or practice escapes from prehistoric predators. They are there, in part, to help you learn, according to a new study from Harvard University.

The idea that dreams can help with creative problem-solving has been discussed for decades, along with observations that sleep boosts memory. But the new research, which will be published in May in Current Biology, provides evidence that dreams really do help people retain new information—but only if it is worked into the reverie.

"After nearly 100 years of debate about the function of dreams, this study tells us that dreams are the brain’s way of processing, integrating and really understanding new information," Robert Stickgold, director of the Center for Sleep and Cognition at Beth Israel Deaconess Medical Center at Harvard Medical School and senior author of the study, said in a prepared statement.

To peek into dreams’ roles in learning, researchers tested 99 college students with a three-dimensional computer maze task. The subjects were instructed to do their best to navigate through the maze as quickly as possible. After about an hour, the subjects spent the next 90 minutes either taking a nap or sitting quietly (watching videos or just relaxing). The sleep subjects, who were not allowed to go into REM sleep (which has previously been linked to learning and problem solving), reported if they had dreamed and what they had dreamed about, and the awake group periodically told researchers what they were thinking about. Five hours after the first test, all the subjects were tested again.

Nappers who reported dreaming about an aspect of the task later had a 10-fold improvement on their maze-navigating abilities compared to their compatriots who did not dream about it or who remained awake.

"Dreams are a clear indication that the sleeping brain is working on memories at multiple levels, including ways that will directly improve performance," Stickgold said.

Those who dreamed about the maze, however, did not report precise representations of the task, but rather "remote associations and memories thematically related to the task," the researchers described in their paper.

"These dreamers described various scenarios—seeing people at checkpoints in a maze, being lost in a bat cave or even just hearing the background music from the computer game," Erin Wamsley, a postdoctoral researcher at Beth Israel and Harvard Medical School, said in a prepared statement. Similar associative—but not identical—patterns have been observed in brain-activation patterns of rodents sleeping after completing an activity.

One of the most interesting findings, the researchers noted, was that motivation to do well in the task appeared to have little to do with how much subjects improved. In fact, subjects in the awake group who were most motivated to solve the puzzle quickly and thought about it between tests fared no better on the second test than those who didn’t care as much or didn’t report dreaming about it. But those in the nap group who dreamed about the maze were among those who did most poorly on the initial task and consequently improved the most.

Wamsley suggested this might be because "if something is difficult for you, it’s more meaningful to you, and the sleeping brain therefore focuses on that subject—it ‘knows’ you need to work on it to get better, and this seems to be where dreaming can be of most benefit."

So dream on—your memory and problem-solving abilities might just be the better for it.

Image courtesy of iStockphoto/AtnoYdur

Rights & Permissions

Comments 14 Comments

Add Comment
  1. 1. jtdwyer 8:50 pm 04/22/2010

    Information management in the brain may be closely related to the principles of information management systems established in the early 1970s.

    Database management systems often must optimize the initial storage of high volume transaction data. Background processes can then be scheduled to update permanent databases with the new data, constructing complex interrelationships between the new data and other databases. In this way critical information can be most quickly stored in short term memory, then migrated to complex data structures in long term memory during idle periods.

    To minimize storage resource usage, memories may be decomposed into lists of links to common, redundant information previously stored. As a result, the memory must be reconstructed on recall. If the common redundant memory segments are subject to change, those modifications could be the source of memory alterations over time.

    Establishing the logical links between new and prior memories during this migration and integration process may enable better understanding of the subject, aiding problem solution.

    Perhaps those researching human learning and memory management processes could benefit by investigating the methods devised by humans to manage computer system based information.

    Link to this
  2. 2. way2ec 12:23 am 04/23/2010

    My college roommates and friends were always "pissed" that I wouldn’t study as much as they did but in the end did better on tests and grades. The part that galled them the most was that I slept much more than they did. (I think they would have forgiven me if I was partying, but sleeping?!) I just sent this article to my son who is so time stressed at college that it is hard for him to give himself permission to sleep. There is a balance that must be maintained, we don’t sleep one third of our lives as some sort of luxury. Sort of like the stupid idea that we only use 10% of our brains.

    Link to this
  3. 3. pratul47 8:29 am 04/23/2010

    Does it have or how much does it have to do with the time of the dream and/or sleep? Some reports say that late morning sleep helps in consolidation of memory.

    Link to this
  4. 4. ramesam 9:49 pm 04/23/2010

    I admit I have not read the full paper. But the summary here talking about the role of dreams and sleep in learning is nothing new that Harvard should be proud as if it is for the first time they discovered it!

    Hope Kathy brings out the really really new thing of their discovery.

    Link to this
  5. 5. robert schmidt 11:48 am 04/24/2010

    @jtdwyer, human memory and computer memory are fundamentally different things. Computers are Von Neumann machines where memory and processors are separate whereas in the brain, memory and processors are connected. Computers store information as digital bits whereas the brian forms memories by strengthening or weakening connections between neurons. In computer memory the bits that make up one image are distinct from those that make up another image. In human memory different memories use the same neurons and synapses. Still, the branch of science called computational neuroscience is quite aware of how computers work.

    Link to this
  6. 6. robert schmidt 12:01 pm 04/24/2010

    @ramesam, what is new here is the proof and details about the relationship between dreaming and learning. Saying that there is nothing new here is like telling Isaac Newton after he formalized classical mechanics that he discovered nothing new because we already understood that things moved and objects fell. Having not read the article should have been your first clue that you didn’t understand it. It is silly to criticize something you don’t understand.

    Link to this
  7. 7. jack.123 8:37 pm 04/24/2010

    I saw no comments on the 3D aspect of human memory and dreams.This is something no computer can match,and it may be quite some time before they do.Because many other animals don’t see or dream in 3D.I wonder if it has a special use where humans are concerned?

    Link to this
  8. 8. freedomsway 9:22 pm 04/24/2010

    This reminds me of Henri Poincare who used to nap during the day and to which he would often attribute his breakthroughs.
    I find that my thinking is radically more assimilated after a short nap, particularly since I started practicing mindfulness.

    Link to this
  9. 9. freedomsway 9:24 pm 04/24/2010

    This reminds me of Henri Poincare who apparently used to nap during the day and to which he would often attribute his breakthroughs.
    I find that my thinking is radically more assimilated after a short nap (or for that matter a good night’s sleep), particularly when it is integrated with some daily mindfulness practice. Zeph

    Link to this
  10. 10. jtdwyer 7:42 am 04/25/2010

    robert schmidt – You’re attempting to lecture me (again) on the physical storage mediums and encoding methods used for recording of information by the brain and computers. I understood that 50 years ago, thank you; it is not relevant to the subject of my comment, which pertained to the logical management of information. Apparently you cannot comprehend the distinction – forget about it.

    Link to this
  11. 11. jtdwyer 10:08 am 04/25/2010

    robert schmidt – I would be interested in your complete dissertation on how "brian forms memories by strengthening or weakening connections between neurons." It sounds most enlightening…

    Link to this
  12. 12. AAkutagawa 3:57 pm 04/25/2010

    I found it fascinating that the dreamer reported "remote associations and memories thematically related to the task" as opposed to clear, lucid representations. I think it’s marked evidence that there’s some higher-level processing at work.

    Link to this
  13. 13. robert schmidt 8:51 pm 04/25/2010

    @jtdwyer, Brian does it the same way we all do it…or at least most of us anyway.

    Link to this
  14. 14. eeeee 4:46 pm 05/3/2013

    This is interesting, but I do not agree with it 100 percent. I actually learned in my psychology class that the way we explain our dream is not the way they really happen. In fact it is just a bunch of signals in our brain going off and then we combine these into our “dream” once we wake up. We are really just making up our dreams in the end.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article