Brain Memory Storage: Understanding the Atomic Level Process

[.ultimate]

At atomic level, can someone clarify how memory is stored in the brain..

 

[sas3]

Memory is not stored at an atomic level. Memories are the connections between neurons and when a neuron makes and then reinforces that connection with another neuron that is a memory.

 

[.ultimate]

What do you mean by making 'connection'?

'Where' is memory stored?

 

[sas3]

Neurons connect to other neurons through dendrites you can see some pictures at the link below.
Also a much better explanation then I can give you.
http://en.wikipedia.org/wiki/Neurons

As for where memories are stored: I was being very simplistic in my previous post it takes clusters of neurons working together to form a memory. Also memories may be stored or scattered through out the brain, that is why a smell might trigger an old memory because that smell is part of that old memory.

I don't think it is possible to pin down the exact location of a memory.

Hope this helps..

 

[Math Is Hard]

What do you mean by making 'connection'?

'Where' is memory stored?

sa3 gave you a good answer. He/She is referring to synaptic transmissions of neurotransmitters between neurons. As to exactly "where" specific memories are stored, it's still not completely understood. You can google "Penfield" and "memory" if you want to read about some of the research that has been done in this area.

 

[sas3]

Think of it like this:

Think about a “Rose”
It is a flower, so you need the memory of what a flower is and that is stored in one part.
It has a shape and that is stored somewhere else.
It is a certain color again somewhere else.
It has a stem, stored somewhere
It has a smell
It has a texture
It is a living thing
It has thorns
It can cause pain
It was given to a girl at one time.
It was used as an example in a forum to explain memory to me once.
And so on.

 

[sas3]

Of Corse now you will never be able to look at a “Rose” again without thinking about this thread.

Sorry about that.

 

[venger]

So would it be true that a person with good memory would have a better bond between neurons. If so, how can one improve his/her memory?

 

[denverdoc]

well memory is still a poorly understood phenomenon, esp in the area of Long term memory. There are a number of processes that can help to explain short term memory, such as facilitation between synapses. But it remains for the most part a mystery. Part of this has to do with Penfields experiments at McGill where using a very small electric stimulus was able to provoke vivid memories in the patient. Yet a large lesion to the brain does not knock out memories of say 1975 to 1980. (drugs might), but leads usually to fuzzier recall across the boards; think of a holographic image here. The film can stand up to significant damage, and unlike a photo negative, you don't lose all of a piece of the image, more like a piece of all of the image.

 

[LondonW]

I know I've read somewhere that exercising and doing new things can help build connections and can possibly slow down Alzheimer's disease. It does make sense that someone suffering from Alzheimer's disease slowly loose their memory over time and not at once because the connections are slowly falling apart. As far as the exact place or places where a memory is located, I don't think that has been discovered or understood yet.

 

[phyzguy]

I think memory may be better understood than many are giving it credit for on this thread. Artificial neural networks have been created that mimic memory storage in the brain. In these networks (and I believe, in the brain), memory is stored by the pattern of synaptic weights - that is, how easily one neuron can cause another neuron to fire. One can train these artificial neural networks through learning processes, and they learn and make mistakes in ways similar to human subjects. One can then look at the resulting synaptic weights to see what has been modified, and one sees that the memory is distributed across the network in such a way that it is not stored at any particular location, as others have said in this thread.

It's not certain that this is how memory is stored in the brain, but I think this is a very reasonable hypothesis.

Below are some references.

http://ulcar.uml.edu/~iag/CS/Intro-to-ANN.html
http://en.wikipedia.org/wiki/Neural_network
courses.cit.cornell.edu/bionb330/.../Associative%20Memories.pdf

 

[Pythagorean]

Memory is still poorly understood, and there are a lot of cellular mechanisms involved that go beyond neural networks and include the protein networks that underlie neuromodulation. There's several cascading, overlapping events with lots of apparent degeneracy at multiple scales.

 

[Ryan_m_b]

Biology, as always, is fuzzy. Neurons form networks but as Pythagorean says their are layers of biology going on underneath (and above) this. There is no clear cut off line for "this is memory, this is support" that we know of.

So would it be true that a person with good memory would have a better bond between neurons. If so, how can one improve his/her memory?

No. If a person had strongly bonded neurons I imagine they would express severe learning difficulties. The same would be true of someone who has weakly bonded neurons. Newborns have very plastic synapses (i.e. they are not bonded strongly) but over time the synapses become less plastic. It is thought this is important in learning, if you neurons were strongly bonded you would not be able to process input well at all.

Having a good memory IMO is probably less about biology and more about psychology (with exceptions). Everybody has roughly the same brain, like sas3 says memories are stored as a linked map. For some people how they link new knowledge to old is more useful than how others do it.