backtop


Print 36 comment(s) - last by twjr.. on Feb 2 at 8:57 PM


New research reveals a single cell can store a memory, like RAM. Another study reveals that longer term memories are store on the brain's surface, while yet another reveals that newly born cells are used to timestamp memories by storing associated memories.  (Source: Elon's School of Communications)
Three key studies help reveal insight into how we remember things in the short and long term and when they happened

Memory is a fascinating field of research.  Some researchers want to discover how to improve memory, others want to help people forget, while others yet merely want to understand it.  However, all fields of memory research are still relatively unexplored due to a baseline lack of understanding of how memory works, something that is only know being slowly discovered and theorized.

Three new studies have yielded great advances in understanding how the human brain might work.  One deals with how we understand when memories happened.  Another focuses on short term memory, and another on long term memory.

Looking at short term memory, researchers at the University of Texas discovered that individual nerve cells in the frontal regions of the brain can store parts of memories for up to a minute.  The discovery was made by careful observations on mice.  The study also showed why abuse of substances such as cocaine can harm peoples' short term memories.  Neurons' ability to retain information is inhibited by dopamine, a brain chemical.  Dopamine surges to high levels when mice are given cocaine, and remain at higher levels afterwards, harming short term memory.

Senior author of the study, Don Cooper, assistant professor of psychiatry at the University of Texas Southwestern Medical Center says the study may help to improve medication targeting attention or decision making disorders.  He says that short term memory is a lot like computer RAM, stating, "(Memory is) more like RAM [random access memory] on a computer than memory stored on a disk.  The memory on the disk is more permanent and you can go back and access the same information repeatedly. RAM memory is rewritable temporary storage that allows multitasking."

The results of the work are published in the upcoming February edition of the journal Nature Neuroscience.

New research from the University of California, San Diego helps to confirm that different regions of the brain are primarily involved in short term and long term memories, something that was previously expected.  The test, done on human volunteers measured their brain activity when asked to recall memories going back 30 years, starting at the present.

The study showed that the hippocampus and its surrounding regions were the most active parts in storing and recalling memories less than a year old.  The frontal, temporal, and parietal cortices, located on the brain's surface were the most active during memories 13 to 30 years old, indicating this was the region for long term storage.  Intermediate storage saw a mix, with primarily the surface storage being used.  The study should help Alzheimer's research says Larry Squire, professor of psychiatry at the University of California.  He states, "It helps us understand that Alzheimer's disease begins with memory problems because the very same structures we're talking about here [the hippocampus and related structures] are the ones affected in the disease."

This study is published in the Journal of Neuroscience.

The last study looks at how time is associated with memories.  It indicates that the new generation of thousand of brains cells a day helps date memories by copying memories that occur around a similar time to associate with a current time.  By stepping through networks of associated memories, according to the researchers at the Salk Institute for Biological Studies in La Jolla, California, and the University of Queensland in Australia, people are able to understand chronologies of events.

The study, based on computer simulations, also reveals that memories or links between neurons storing memories are often formed when the same neuron is used for a short term memory at a particular time.  So if you went skiing and then went on a date to a coffee shop, you might remember the two as occurring on the same day.  These discoveries should also help with memory and neurological disorders like Parkinson's disease and Alzheimer's, which involve new brain cells being ceased to be born, which would not only hinder memory recollection, but also timestamping of new memories.

The third study is featured in the journal Neuron.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

RE: So I wonder..
By grath on 1/31/2009 1:04:33 PM , Rating: 4
Biology, like nature in general, is analog not digital. Its probably an impossible task to accurately quantify storage capacity as we think of it. Obviously there are a finite number of cells in any region of a brain, so there is a theoretical upper limit on the rate of possibile interactions that store and process information, but equating those interactions to a quantifiable amount of information is the problem.

Even if it were possible to determine capacity, a more important factor is compression. Take the example of learning foreign languages, say "learning French as a second language to fluency will fill X amount of brainspace, whereas learning German as a second language to fluency will fill Y amount." It is often observed that learning more languages is easier after learning a second, so one could arguably continue to say that "learning French as a second language and German as a third language would take Z amount of brainspace, where Z is less than the sum of X and Y."

The same could be said of nearly any experience that shares similarities with another experience. The way the brain compresses information, indexes it, and generates metadata as it were, is what they are trying to figure out.


RE: So I wonder..
By snowbro on 1/31/2009 1:47:24 PM , Rating: 3
Actually, nerves (which are the fundamental unit of the brain) work in a DIGITAL fashion. They either SEND or DON'T SEND a signal... Exactly like the binary code system of computers, 1's and 0's. This is the basic functioning of a simple nerve.

The problem is, for a specific function, MANY nerves are involved, and you have situations where there a thousand nerves entering and a thousand nerves leaving a synapse, and its the general pattern of excitatory or inhibitory neurotransmitter release that decides the ultimate outcome. Not to mention the myriad of pre-synaptic and post-synaptic modulating factors/processes that can take place.

So a better way to understand the human brain is that at its most basic level, it is a DIGITAL on/off system, with significant ANALOG modulation and complexity.

Some people have actually tried to quantify the actual storage capacity of the human brain.

I like this explanation:
"The human brain contains about 50 billion to 200 billion neurons (nobody knows how many for sure), each of which interfaces with 1,000 to 100,000 other neurons through 100 trillion (10 14) to 10 quadrillion (10 16) synaptic junctions. Each synapse possesses a variable firing threshold which is reduced as the neuron is repeatedly
activated. If we assume that the firing threshold at each synapse can assume 256 distinguishable levels, and if we suppose that there are 20,000 shared synapses per neuron (10,000 per neuron), then the total information storage capacity of the synapses in the cortex would be of
the order of 500 to 1,000 terabytes. (Of course, if the brain's storage of information takes place at a molecular level, then I would be afraid to hazard a guess regarding how many bytes can be stored in the brain. One estimate has placed it at about 3.6 X 10 19 bytes.)"

I hope this clarifies things...


RE: So I wonder..
By William Gaatjes on 1/31/2009 1:59:25 PM , Rating: 2
Are there also not variables as firing rate ( or frequency)and intensity.
With firing rate it is possible to build some kind of pulse widht modulation. Thiw PWM can be used to send many values all with 1 bit. an 8 bit pwm can send 256 different pulsewidths. Maybe our braincells uses pwm too.


RE: So I wonder..
By William Gaatjes on 1/31/2009 2:00:57 PM , Rating: 2
Sorry for the many errors. Accidently pressed post comment while still reading the preview.


RE: So I wonder..
By Cullinaire on 1/31/2009 2:53:28 PM , Rating: 3
You'd think one'd be able to run chkdsk on his own brain, to see how many blocks are available.

But how many people come preloaded with chkdsk? Now that's the issue.


RE: So I wonder..
By rudolphna on 1/31/2009 3:30:21 PM , Rating: 2
I read somewhere I cant remember where though, that each strand of our DNA contains roughly 750MB of information...


RE: So I wonder..
By mbarren on 1/31/2009 9:53:41 PM , Rating: 2
was it on 23andme website?!


RE: So I wonder..
By ekv on 2/1/2009 2:43:48 AM , Rating: 2
dude, if you can't remember you may be down to just 500MB

sry 8)


RE: So I wonder..
By clockworkjpm on 1/31/2009 5:49:51 PM , Rating: 2
Good summary and reference, snowbro. There are lots of ways to try to encapsulate the theoretical information storage capacity of the brain. But we should all keep in mind that, as elegant as it is, the brain is biological, and thus the actual capacity is a much less quantifiable number (at this time).

It's clear that memories (or at least associations) are stored at synapses between cells, and that many neurons, including the famous "Halle Berre" or "grandmother" neurons seem to retain relationships between their excitability and abstract stimuli over long periods of time (years, etc.). A lot of recent evidence has demonstrated that much of our cellular memory seems to reside in the brain as "traces" or serial patterns of activity. But then again, each of these serial patterns has numerous extrinsic connections which, given certain modulatory influences (drugs, age, disease, etc.) could either switch or fail to propagate.

And obviously action potentials are near-binary (don't let anyone tell you they're 100% binary, by the way, my dissertation is devoted in part to this topic) but there is still considerable debate in the field whether, as a whole, action potentials are truly analogous to "bits" in a computer.

What was the topic again?


RE: So I wonder..
By tyson766 on 1/31/2009 8:12:14 PM , Rating: 2
More discoveries are being made every year. Neurons act in a similiar way to an AND gate. If a neuron gets enough signals it will fire. We now know that neuron signals can also be negative, ie a signal received along a certain connection will reduce the chance of the neuron firing. To make things even more complex chemicals can be released in certain sections of the brain which can also enhance neuron output in that particular area.

How all this affects the complexity of the brain I have no idea, but it will be a while before we are anywhere near the computational power needed to reproduce it.


RE: So I wonder..
By spread on 1/31/2009 11:19:41 PM , Rating: 2
If you look at it that way;

ANALOG signals are DIGITAL also. They are either SENDING or NOT SENDING.

But we all know its not that simple.


RE: So I wonder..
By Wierdo on 2/1/2009 11:21:55 AM , Rating: 2
I thought the human brain cells behaved like memristors, so in a sense they're analog-ish in nature as they can adapt their output based on provided input.

http://en.wikipedia.org/wiki/Memristor


"If a man really wants to make a million dollars, the best way would be to start his own religion." -- Scientology founder L. Ron. Hubbard














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki