What physically constitutes memory?

Biosyn

What physically constitutes memory? I don't understand how millions of tiny connections between neurons allows us to memorize things or picture something in our mind.

Any related videos would be appreciated :)

Pythagorean

This is not a simple question to answer. Firstly, there's lots of different kinds of memory. But also, we're not sure of all the mechanisms involved.

One example:

A system of neurons receives an input through the olfactory system that smells food; elsewhere, the system receives inputs about the sound vibrations in the atmosphere. There exists a neuron that both inputs connect to (let's call it the post-synaptic neuron) but the synapse between the the audio inputs and this neuron is a "silent synapse" because it's synaptic channels have blockage (Magnesium ions) in them. If the postsynaptic neuron fires (because it received an input from the olfactory system) it will expel the magnesium, allowing the audio input to fire the neuron for a short time after that.

Now, as the two stimuli continue to pair up on that one post-synaptic neuron, via Hebbian learning, the neuron starts importing a new kind of channel to this synapse that doesn't have Mg blocking it and the synapse stops being silent. And now the audio signal will directly fire the post-synaptic neuron all by itself.

This might be an example of Pavlov's dog: conditioning the audio signal to fire the same neuron that the olfactory signal does.

This is just one example of a mechanism for memory though, there are lots of other mechanisms (many still unknown) and some are embedded in genetic expression regimes, not just electrophysiology.

Pythagorean

Here are some youtube videos showing another way neurons encode information: Place cells.

http://www.youtube.com/watch?v=PGHRD...eature=related

http://www.youtube.com/watch?v=lfNVv0A8QvI

Pythagorean

Here's a more recent proposal: that prions play a role in long-term memories:

http://www.sciencedaily.com/releases...0127162409.htm

The peer-reviewed journal is referenced at the bottom. Not sure how robust the claim is, but it gives an idea of how many different little mechanisms come together in all the different cases of what we call "memory".

jackmell

Know what a purkinje (pa-kin-gee) neuron looks like? How about a large elm tree without the leaves? About the same. Really, it's trillions of connections and many of them are fed-back: the output of some neurons go back to be used as input to neurons along the connection sequence. So we have billions of non-linear (ala Hodgkin-Huxley) neurons connected in likewise non-linear configurations.

You know what strange attractors are and in particular their associated basins of attraction? A non-linear dynamic system, often contains islands of stability called "attractors" which are surrounded by a region such that if the system finds itself in this region, the dynamics will "fall" into the attractor and just stay there until it's "bumped" out of it's basin and perhaps falls into another basin surrounding some other attractor.

Ever tried to remember something but you can't initially recall it? You think of things related, and sometimes the memory "comes" to you. What's going on in the brain during that event? Well some believe memory is effected by strange attractors created by the dynamics of neuron firing patterns: the act of trying to remember sets up firing patterns in the brain, and if their dynamics "falls" into some basin of attraction, it leads to the attractor representing the memory. For example, see this reference:

http://leadserv.u-bourgogne.fr/peopl..._as_motion.pdf

You may be interested in searching the internet for "strange attractors in the brain".

Dotini

A clock's internal mechanism systematically marks seconds into minutes and hours. But the mechanism does not create or represent time. There is a similar problem with memory and the brain.

For memory to exist in the brain, there must be a way to represent information. The general theory is that patterns of synaptic connections create memory. However, all parts of the brain's bio-machinery (neurons, synapses, dendrites, etc.) are continually changing. Though neuroscientists have been studying memory for many decades, there is no integrated theory. Some of them may doubt memory exists in the brain.

Heisenberg said, "...what we observe is not nature itself, but nature exposed to our method of questioning". Godel's incompleteness theorem showed that nothing is certain concerning the universe. The rational mind may never know the ultimate truth.

Respectfully submitted,
Steve

Biosyn

Thank you guys for the in depth information! It was extremely helpful.

I'll be doing a ton of reading/researching this weekend.

ViewsofMars

I think this might be helpful and it's what teachers use for kids:

Simple and sweet.

nitsuj

A general question can have a general answer.

I know little to nothing 'bout memory, i keep forgetting it...but allot of it's about patterns. Billions of "specifics" adds up to a very specific "thing". in this case a memory...well the memory may only be a few "specifics" in which case...it's nearly forgotten, like my memory about memory.



"storage" must be the physical make-up/structure (brain itself, well it's "wiring"), of course a memory is not "reproduced" (I appreciate the sub-conscious fills in gaps) . No wonder memory comes across as very "malleable", and can believe in non-sense so easily.

ViewsofMars

I don't believe in non-sense but accept that some people have a great imagination!

nitsuj

oh snap!

anisotropic