Exercise Can Erase Memories

Running causes rodents to forget their fears in part because of increased hippocampal neurogenesis, a study shows.

By Ruth Williams | May 8, 2014

New neurons (white) integrating into hippocampus.JASON SNYDERAdult mice that exercised on a running wheel after experiencing an event were more likely than their inactive mates to forget the experience, according to a paper from researchers at the University of Toronto, published in Science today (May 8). The results suggest that the production of new neurons—neurogenesis—prompted by the exercise wiped out the mice’s memories. They might also explain why human infants, whose brains exhibit abundant neurogenesis, do not have long-term memories.

“In general, hippocampal neurogenesis has been thought to be the basis for memory and they’re suggesting that it’s the basis for amnesia,” said Thomas Insel, director of the National Institute of Mental Health. “That’s a very controversial and provocative concept.”

Infantile amnesia is common to all humans. Children typically do not develop long-term memories until age three or four. But why is that? Sheena Josselyn and her husband Paul Frankland, who are both neuroscientists at the University of Toronto, pondered precisely that question after noticing that their two-year-old daughter could easily remember things that happened within a day or two, but not several months in the past.

More specifically, they wondered whether it might have something to do with neurogenesis in the hippocampus—a brain region involved in learning and memory. Hippocampal neurons are produced rapidly during infancy, but neuronal generation in the region slows to a trickle in adulthood. “This inverse relationship between the levels of neurogenesis and the ability to form a long-term memory got us thinking that maybe one is due to the other,” said Josselyn.

Running is known to boost neurogenesis in mice. So, to test whether neurogenesis might impair memory, Josselyn and Frankland first taught mice to fear a particular environment—the researchers placed the animals in a distinctive box and gave them electric shocks—and then provided them with access to a running wheel or let them remain sedentary. When the mice were returned to the box after a day or a week, both groups of animals tended to recognize the now-familiar environment and freeze—a fear response. But if the mice were returned to the box after two weeks or more, only the sedentary mice froze. The exercisers seemed to have forgotten their fears.

Running imposes physiological changes aside from neurogenesis, of course, but the team saw the same failure in memory recall when they specifically increased neurogenesis pharmacologically in the mice. They also found that inhibiting neurogenesis in exercising mice and in infant mice made the animals better at remembering.

The team also showed that rodents such as guinea pigs, which have reduced neurogenesis in infancy compared with mice, tended to remember a fearful experience for much longer than did infant mice. And boosting the guinea pigs’ neurogenesis caused them to forget their fears more readily.

As Insel pointed out, previous studies have indicated that neurogenesis in adults is beneficial to learning and memory—a finding that seems at odds with Josselyn and Frankland’s. However, said René Hen, a professor at the Kavli Institute for Brain Science at Columbia University in New York, “previous findings have mostly been dealing with the role of neurogenesis in encoding novel information”—that is, learning and remembering something new. “Now, in the Frankland study, they are looking not at the ability of encoding novel information, but at forgetting older information. So one way to reconcile the two is to think of it as a trade-off: if you get better at acquiring new stuff it maybe at the detriment of keeping old stuff.”

Should the results of this study on rodents cause people to worry that training for a marathon might make them forgetful? “People do always say that running clears your mind,” said Josselyn, “and in a sense I would say that’s true.” But clearing one’s mind is not necessarily detrimental, she added. “For instance, I don’t want to remember where I parked my car two weeks ago because that’s going to interfere with me remembering where I parked it today. . . . We think that neurogenesis and forgetting is an important part of healthy memory. We don’t want to remember absolutely everything.”

K. G. Akers et al., “Hippocampal neurogenesis regulates forgetting during adulthood and infancy,” Science, 344:598-602, 2014.

Add a Comment

Avatar of: You



Sign In with your LabX Media Group Passport to leave a comment

Not a member? Register Now!

LabX Media Group Passport Logo


Avatar of: Godagesil


Posts: 2

May 9, 2014

Might the loss of memories also be related to the stress induced in the animals by the exercise they are forced to perform? Stress is known to suppress memory, which is probably a defense mechanism in the brain. Studies of soldiers and police have shown unequivocably that memory is compromised due to selective impairment or suppression of human senses depending on the conditions of the situation under which the stress was induced. Ex: auditory sense increases under low light conditions and visual senses are suppressed.

May 9, 2014

Interesting inverses!!

But from a common sense spatial consideration;.within the unyielding space of the intra cranium, isn't 'more' for some neurons automatically meaning 'less' for some others. If the intracranial volume has capacity for only X mls of neuronal tissue, and there is y mls of long term memory; that leaves X-y mls for short term. Step up the volume for one, expect step down for the other given a constant X. We see this clinical scenario in intracranial growths or even post traumatic intracranial collections with bleeding. The more blood collecting within the locked space of the intra cranium, the more the chances of loss of neurons if appropriate evacuation or decompression is not timely attained.

This I consider a possible spatial theory to this observation.

There is a confounding factor to this spatial theory in human infants whose skull is not yet everywhere bony and Fontanelles exist providing a give in response to demand: but even then only for so much or for so long

But there also could be cybernetic factors involved with a k factor somewhere within the inverse relations

But it is also known that memory for bad experiences stick around longer than for good.

The hippocampus tends to hold on to unpleasant memories far longer than the pleasant ; the explanation offered for this is an evolutionary survival mechanism to ensure that man does not get himself trapped in the same unpleasant situation twice by forgetting too soon his experience of the first. So the cave man retained enough memory of all his mis-steps within his savage terrains to avoid a repeat. The hippocampus holds on less to pleasant evolutionarily non threatening experiences.

This might be a chance to test this out in this survey.

Avatar of: JToeppen


Posts: 37

May 10, 2014

The effects of cannabis on the hippocampus are interesting and tied to memory;   http://www.jci.org/articles/view/25509   http://en.wikipedia.org/wiki/Removal_of_cannabis_from_Schedule_I_of_the_Controlled_Substances_Act   Some people would like to see more research on this topic.
Avatar of: dsp


Posts: 1

May 12, 2014

Regarding the following:

"As Insel pointed out, previous studies have indicated that neurogenesis in adults is beneficial to learning and memory—a finding that seems at odds with Josselyn and Frankland’s. However, said René Hen, a professor at the Kavli Institute for Brain Science at Columbia University in New York, “previous findings have mostly been dealing with the role of neurogenesis in encoding novel information”—that is, learning and remembering something new. “Now, in the Frankland study, they are looking not at the ability of encoding novel information, but at forgetting older information."

Isn't it just as likely that the difference between the two experimental contexts is not the difference between "remembering something new" and "forgetting old information" but rather the difference between remembering a new fact/piece of information and becoming incapable of generalizing from the previously attained information?  In other words, we have no way of knowing if the mice had completely forgotten their experiences in the shock environment, all we know is that they no longer exhibited a fear response when re-exposed to the environment, so isn't it possible that what they lost was not information but rather the ability to form links between pieces of information?  If so, couldn't this resolve, to some extent, the discrepancies noted in the results of research on this topic?  

Avatar of: LabRat 1

LabRat 1

Posts: 1

December 26, 2014

Amnesia. Is not having an underlying learning structure firmly in place to retain memories actually the same as amnesia?

Out of my league, however: I think more explanation for the conclusion is warranted. I see a theory without enough evidence presented. Did they shock the babies, in order to have an equivalent test?

The inverse effect needs to be investigated: if 'hippocampal neural regeneration slows to a trickle.' Does anything on that end of the spectrum behave in keeping with the conclusions?

What role is played by the fact that a 1-3 year old often has marginal language skills to reinforce whatever memories they DO have in those early years and what would those memories essentially consist OF? (Again, was it an equivalent test?)

I know I have a few memories that were of events when I was 3, but they had to be rather pivotal ones & I had to revisit them once I had language skills to help frame them better. (That's my claim, without evidence). The language framing is what locked them down into adulthood.

(Aside: However, it is also possible to blithely create false memories by language articulation as well, as a child's memory is reinforced by the adults in their lives retelling the anecdotes & subsequently dreaming into being a memory of the event they failed to remember before its depiction).

The mouse data in the article seems to hinge on expecting mice to freeze in terror. Perhaps the exercise-enabled mice were instead empowered by exercise & thus did not freeze in panic.

Sedentary mice in the interim, it seems logically to me, had no countervailing experience in the interim & only had the shocking experience as a point of reference.






Avatar of: N K Mishra

N K Mishra

Posts: 60

December 28, 2014

What is that one wants to forget. and what is that one does not want?

In this contex exercise might exert different responses.

One thing appers straightforward, neurogenesis offers a scope to store new memory, but whether it is  a cost of the loss of old memmory is not clear.

Avatar of: asdfzxh


Posts: 1

February 5, 2015


Popular Now

  1. How to Separate the Science From the (Jerk) Scientist
  2. Could a Dose of Sunshine Make You Smarter?
  3. Prevalent Form of Childhood Leukemia May Be Preventable
  4. Conservation Biologist Ben Collen Dies of Bone Cancer