Take the 2-minute tour ×
Biology Stack Exchange is a question and answer site for biology researchers, academics, and students. It's 100% free, no registration required.

Often we are able to memorize very precise bits of information (exact addresses, birthdays of inlaws, number of dirty socks under the bed), but over time our recollection tends to become fuzzy. We no longer recall the day and month a certain president was killed, only the year, or maybe the decade (despite being straight A students in junior high). Like the smoothing out of a delta function under the heat equation, instead of a pin-sharp value we're left with a range of values where the correct one is likely situated. However, most of the time we are perfectly satisfied with such approximate information and can work with them just fine (except in some extreme cases of function impairment, such as Alzheimer's or the aftermath of binge drinking).

Discrete data must be stored in low entropy structures in the brain that over time spontaneously shift to a higher entropy state. Meanwhile, our biology tries incessantly to slow down the decay by expending energy (with the frequently utilized neural pathways being better maintained than the abandoned ones). Even processes analogous to error correction take place, e.g. when we arrive at a certain lost piece of information by lateral thinking and piecing together available data.

My question is whether the fading of memory exactitude can be explained in terms of the thermalization of the digital-like structures where discrete information is stored, and if anyone can point me to studies in this direction.

share|improve this question

3 Answers 3

Interesting point, but I think there is a simpler and possibly more widespread reason for the decay of memories. Namely, as new memories need to be stored in the same neural networks, the weights, timings and other properties of the network links are constantly changed. In other words, there is competition for the limited amount of memory. This re-use of the same networks for multiple functions and memories is the same reason we have redundancy in our brain - it's also largely why they are as fuzzy in remembering as they are. And since memories are recalled associatively, the more connected two networks that encode different memories, the more likely it is that activations of one will activate the other. But that also blurs the boundaries between specific bits of information stored. It's a trade-off between fidelity of the memory stored and recalled and generality of the recall and associations possible.

share|improve this answer
Good point. There are various theories on memory loss but I wanted to address the problem of organic decay specifically, which seems prevalent in some situations. In particular, older people suffer from memory decay more than young people, even though they're arguably faced with less new information and the corresponding need to free up space is also smaller, so your argument doesn't provide a full explanation. –  suissidle Feb 4 at 14:32
No, you are right, it doesn't. But we could extend the point I'm making to older people as well: they have a significantly smaller number of neurons so that while there may be less new information to deal with, there is also more competition from the memories previously stored. Also, on a different note, it could be that there is memory decay due to a greater inability to sustain the current memories - for example, because of lower production of certain crucial cell-adhesion and receptor proteins. –  Martin Dinov Feb 4 at 14:37
Great, I'm not debating all these phenomena you mention, in fact they're all organic ways of decay, but let's look at how they are essentially able to turn digital values into fuzzy feelings of what the value should approximately be. I'm convinced that only a small, high-level, conscious part of the human mind deals works digitally as an emergent feature of a larger, fuzzy mechanism (without breaking it all the way down to quantum mechanics). –  suissidle Feb 4 at 14:52

Discrete data must be stored in low entropy structures in the brain that over time spontaneously shift to a higher entropy state

How on Earth could you possibly know that? You are overextending physics to the workings of the brain and there is nothing that says this is necessarily so. Memory decay is not fully understood but it is known that the more features(i.e sound ,taste) are used when forming the memory and the more it is associated within an existing knowledge framework, the higher the chance that the memory will persist.

share|improve this answer
What exactly are you disputing exactly? Information is pretty much synonymous with low entropy (i.e. ordered) structure. E.g. a magnetic tape whose domains are collinear is seen as holding information, whereas a random distribution of the domains corresponds to no information. It takes a lot of data (i.e. high entropy) to describe a randomly ordered structure. Roughly, total degrees of freedom = useful information + entropy. Finally, the second law of thermodynamics states that entropy spontaneously increases. –  suissidle May 17 at 18:17

My understanding is that there are two major reasons for not being able to recall where your socks are or when your anniversary is:

  1. The information was not committed to short term memory in the first place when you initially encountered it, so there was never anything to remember.
  2. The information is intact, but you are not able to effectively retrieve it, because other memories compete and act as noise when you try to remember a certain specific one.

Neither of these scenarios involves actual loss of long term memory, which is more commonly a result of concussions, or other sorts of serious brain damage.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.