Slashdot Mirror
Brain's Cache Memory Found
Shipud writes "Electrical activity in a single section of the brain has been linked to very short-term
working memory, as is
reported at Nature. Very short-term working memory capacity is thought to be related to intelligence. In the same way that a larger cache speeds processing time, people with a greater capacity for holding images in their heads are expected to have better reasoning and problem-solving skills. The localization of this ability is a surprising finding, as until now it was believed that STWM was diffused throughout the cortex, rather than localized."
Is this going to lead to benchmarking people?
Employer: I'm sorry sir you don't have a big enough cache for our needs. We are going to have to let you go.
Employee: Man this blows i would be really upset but i forgot what you just said.
A Fatal OE Exception has occurred, Sig will now reboot.
Does the article mention anything about expansion modules? I'd read it myself, but I can't remember what we're talking about here...
What was I saying again?
W
-------------------
This is my SIG. There are many like it, but this one is mine.
good news for pr0n hounds.
too bad it's addicting
It's interesting how we use rudimentary digital computing analogies to explain the workings of our brain. Like in most theories, I suppose one can extend this analogy only to a certain extent. Which, in this case, shouldn't be suprising considering how comlex the brain is...
Perhaps this explains why my head gets extremely hot when I do my Calculus exams.
Hmm.. Makes you wonder if it's possible to overclock your own brain. Some kind of implant with electric stimuli.. Or maybe some kind of chemical. Only, I can't quite figure out how to make a decent cooling solution, and I absolutely refuse to walk around with a heatsink attached to my forehead! ..Or if you find out how to stimulate that part, maybe some good oldfashioned brain exercise to increase your cache and speed. On second thought.. Nah.. Not really geeky enough ;)
There has been plenty of studies showing that people tend to remember things incorrectly. Could this very short term memory be part of the final proof needed to invalidate witness statements in legal cases? Or perhaps they can use the line and dots test on witnesses and see how likely they are to remember something that happens in a glance. If they check high on the test, they might be more likely to be able to remember an incident correct.
Apple built a platform for their ideas, Google built one for everyone's.
Thanks.
... not remember 4 spots of light.
... but I've also seen many of those same types fail when tasked with the above sorts of questions.
This seems to be quite questionable as far as any sorts of broad conclusions are concerned.
When people talk about "intelligence" they usually mean something like "being able to grasp two deep concepts and put them together"
Granted, I have seen a correlation between people who are capable of remembering 10 digit codes and intelligence
Maybe this is a red herring.
As my father lik@(munch munch)...
I don't know about you, but when I program, I *do* think visually about it. It's really hard to describe exactly how, but to me, writing in a programming language "feels" more akin to drawing a picture than writing an essay.
I don't think all programmers approach the task using the same kind of intelligence.
I think it would be interesting to check different disciplines against each other, but programming is a bit too all-encompassing to be nailed down to just one kind of intelligence. It's partly language thinking, partly spatial thinking, partly mathematical thinking, a little bit of art, etc...
Don't label something "offtopic" unless you know the topic well enough to tell what's on topic.
Most people can hold three or four things in their minds at once when given a quick glimpse of an image such as a collection of coloured dots, ...
Did it not also depend on what kind of (was it) chunks you store (if this is at all what is stored in should it perhaps be ultra-) STM ?
Where it "started":
The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information
by George A. Miller
originally published in The Psychological Review, 1956, vol. 63, pp. 81-97
CC.
TaijiQuan (Huang, 5 loosenings)
It's not too surprising that the brain's short-term visual cache would be closer to the visual cortex. What I would like to know is how closely the visual cache is related to intelligence. Does it need actual visual input, instead of just imagined, and if so... <facetious>do you become marginally dumber when you close your eyes?</facetious>
From reading Synaptic Self, the "general" cache and CPU area would seem to be the prefrontal cortex. It can activate memories to work on (the closer the current emotional state it was recorded in, the better), and hold a few things to work on. Perhaps there are many more specializations yet to be uncovered, but I'm struck at the sheer relative size of brain required to actively think and plan a next move. Considering that even a worm brain can get its owner around, you'd think our capacity for juggling thoughts would be encyclopaedic.
I'd be curious as to what connections this area has to the prefrontal cortex - I've heard of the spots tests before - I don't recall it being related to general intelligence.
Addressing the question of how cache gets spat out to hard drive, as it were, to keep thoughts in slightly longer-term storage, it looks like thoughts have to be put through the hippocampus and entorhinal cortex, where they will slowly get rewired (indexed?) over the course of about two weeks - about the length of memories you can lose under strong electroshock therapy.
So many small functional pieces of the brain; I'm struck by how independent the sections of the brain are, by and large. Large-scale coordination has to go through a secondary 'chemical drip' system, from neuromodulators released by non-connecting nerves throughout the brain. It's that level of coordination required to put your brain to sleep or wake it up, amongst other things.
I'm looking forward to more decoding of the brain's structures - narrowing down specific activities to a small area of the brain like they did is fantastic.
Binary geeks can count to 1,023 on their fingers
When people talk about "intelligence" they usually mean something like "being able to grasp two deep concepts and put them together" ... not remember 4 spots of light.
... but I've also seen many of those same types fail when tasked with the above sorts of questions.
Indeed. Intelligent people would be those who are excellent at conceptual blending. List of resources on this page.
Granted, I have seen a correlation between people who are capable of remembering 10 digit codes and intelligence
I'm currently reading Kandel & Squire's Memory.
Having a too-good memory is what you don't want. They relate the case of a hyperretentive memorist from Russia, who had almost supernatural retention skills, but was hopeless at appreciating metaphors, or pattern matching or generalizations. Which are the building blocks of analytical intelligence.
And what about the famous "magic number", 7 +/- 2? These people seem to be offering 4 +1/-2.
It was found that the famous "5-9 digits" resulted from a bogus test. Rather than testing short-term memory, it was testing the "auditory loop" -- people weren't remembering the digits, they were mentally replaying the sound of someone speaking the digits.
When people are given the digits via non-auditory means, 3-5 digits seems to be the norm.
Tarsnap: Online backups for the truly paranoid
Well, there's the obvious. Use it. A lot. The human machine is built around building up what gets "stressed." That goes for the brain too. For short term working memory exercise make references. Read a book, history or something like that, where you're bit over your head. Keep Google going while you do it and every time you hit something you don't understand do a search, follow the search to whatever else interesting it might lead to, bounce back and forth from the book to search materials.
/. on the side.
Now do it with two books, maybe even on different but related subjects, while you keep an eye on
This is pure "cache" work. Don't try to memorize any of it. That's a different "brain muscle." Isolate what you're exercising. You're just trying to keep the different threads of thought all going without losing them.
Now, remember what I said about getting stressed? Don't. Really, the biggest killer of working short term memory is any sort of tension. Tension is an attention grabber, and you only have a limited amount of attention at any one time. Learn to relax. Let it flow of its own accord. If you pick it it will never heal.
It's one of those zen things, where you hit the target by not being aware that the target is even there. The arrow releases itself.
Oh, and here's the nasty part. Just like stressing muscles to build strength, it's a use it or lose it deal. Yes, you can improve your short term working memory, but when you stop using it, the improvment will fade.
I really hate that part.
KFG
I don't know about you, but when I program, I *do* think visually about it. It's really hard to describe exactly how, but to me, writing in a programming language "feels" more akin to drawing a picture than writing an essay.
When I program, I hardly think visually at all. Then I've usually mapped a clear sequence 1. 2. 3. 4. that'll get me from A to B. Even if it doesn't work right, it's mostly just adding, subtracting or reorganizing the steps, in a purely linear fashion.
When I design, I primarily think in 3D. Or at least, more than 2D, I don't think in the form of trees and object hierarchies, but more like freeform 3D FPS. This objects connects to this and that and that and that, and I "see" how they interact around it.
I'm quite aware I got a fairly big "cache" to map out such problems in, I kinda doubt that works for everyone. I'm nothing like those people that manage to use long term memory to do insane math calculations, but well above average.
I remember I got it "wrong" on a math estimation test (i.e. not supposed to do any math on paper, no calculator) because I was too accurate. They suspected I was cheating, until I told them to give me a few bonus questions orally.
It's nice for doing wild tricks like:
Q:"What is the cube root of 53,582,633?"
A:
1. last digit = 3, from 7^3 = 343 (1-to-1 mapping) -> ends in 7
2. 3^3 = 27 begins with 3
3. a) 33 - 7^3 = 33 - 43 = 90 mod 100,
b) 3 * 7^2 * x = 9 mod 10
3 * 9 * x = 9 mod 10
7 * x = 9 mod 10
x = 7 -> middle is 7
A: "The answer is 377"
If you have the squares (1,2,4,16,25,36,49,64,81) and cubes (1,8,27,64,125,216,343,512,729) memorized you can do this in real-time, or at least I can. Trust me, it'll completely freak your friends out.
Kjella
Live today, because you never know what tomorrow brings
Well, one assumes a "hypertentive memorist" is dealing primarily with long-term memory. I interpret this article as dealing with the type of short term memory used in solving an equation, or writing a small code section. It's certainly possibly your HM was deficient in that area.
I think there's a valid point to be made about how much information someone can deal with in those contexts. The one caveat I'd make is whether the person is dealing with text or imagery - AFAIK there's quite a range there.
At any rate, I think it's clear that many intelligent people also have above average long-term memories.
Galileo: "The Earth revolves around the Sun!"
Score: -1 100% Flamebait
I'm studying for a degree in genetics at the moment and I was interested to hear your question.
.Whilst stem cells are what is known as 'pluri/toti-potent' - that is, they can give rise, under different conditions, to many (or in some cases all) types of bodily cell, (e.g. liver, spleen, pancreatic, brain neuron etc etc.), they need the correct stimulus, in the form of the correct chemical environment, to make them differentiate into that sort of cell.
:P
Simply injecting stem cells there wouldn't really do a lot
Increasing the size of that area would probably make some sort of difference to the STWM, so we'd need to approach it in a way that caused us to end up with not only more cells there, but more cells that actually perform the correct function there, and that tie in with the existing lot of cells. No use having a ball of cells of the right type there that just grow into a new mass. In fact, that's what we call a tumour. Never good in the middle of your head
Needless to say, that's not as easy as it sounds. You'd need to get some stem cells, and discover what is the exact stimulus that makes them, in the developing embryo, mature into 'STWM cells'. Since I think we can likely assume that your 'brain cache' doesn't grow in size throughout life (or you'd get progressively more logical and have an improved short term memory as you got older), we can also probably guess that this area is fully developed at birth and therefore the only place the correct environment for this differentiation would be likely to occur is in the developing foetus.
Which means that you'd have to take some developing foetuses apart to try and localise the correct chemical environment. And then you get into legal/ethical fluff. Currently, there's no way you'd get permission to take foetuses to bits to improve some adult's short term memory. Maybe in times to come, we'll be able to co-localise these factors and chemical environment electronically, or with some sort of prenatal scan. Until then, I'd think that stem cell therapy is unlikely to work correctly.
Sgt. Friday: "Are you sure this is the woman you saw in the post office?"
Burns: "Absolutely! Who could forget such a monstrous visage? She has the sloping brow and cranial bumpage of the career criminal."
Smithers: "Uh, Sir? Phrenology was dismissed as quackery 160 years ago."
Burns: (measuring Smither's head) "Of course you'd say that... you have the brainpan of a stagecoach tilter!"
"No fair, you changed the outcome by measuring it!" - Professor Hubert J. Farnsworth
Both the Nature article and the posting here on /. are exceedingly misleading (I don't blame the poster... he/she just reported what the Nature article said)
All that the two articles *may* have found is the location of a part of VISUAL working memory. This would be the area that tracks objects through space and binds features that are processed seperately by the visual system (say color and form) into the same object. This is NOT the seat of all intelligence.
There are many different aspects to working memory: people have hypothesized that there is a phonological working memory, one involved in the spelling process, one involved in computing things like syntactic relations, etc. And yes, there is probably such a thing as a general-purpose working memory. All they may have found is the location of the visual-spatial component of working memory. This is a far cry from finding anything that limits one's intelligence, unless you define intelligence as "visual-spatial ability".
In fact, it is quite wrong to even suggest that the visual-spatial working memory is somehow related to intelligence. There are many instances of people with working memory deficits who are able to function quite normally in other domains.
For the sake of brevity I won't go into the finer about the studies themselves (one of the studies used the ERP recording technique, which is *awful* at localization) because the main point is that in and of themselves the studies are fine. It's this conclusion that they've somehow found "the RAM" or the thing that would limit intelligence that's exceedingly problematic.
Often I find myself going to type in the URL of a website, manage to get distracted by four things on the way to focusing on the location dialogue, and by the time I'm read to type, I've completely forgotten where I was going to go.
However, if at that point I just 'let my fingers go', they can usually type out the first 5 letters of whatever it was I was going to go to, even if they weren't in typing position.
This is extremely handy. Any idea what it's called?
I find the methodology of their research much more interesting than their results. I've done quite a bit of work in this area, including my dissertation, and from the very high level description of the tasks involved their results need to be interpreted in a much more limited sense than they are being presented.
The task that you are given for a specified stimulus is going to very much influence your performance on later tasks. If you are presented a slide and asked to count the number of dots, then later asked whether or not the number of dots on a particular slide was even / odd, then you are likely to do fairly well. But what if you are presented a slide and asked if there was a blue dot on the slide or not, how is your performance going to be on the even / odd task later on? What kind of curve are you going to get for each task when you vary the number of dots and can you really then imply a limit to the theory of memory?
Obviously, you need more details than is presented in the shorter article. The last paragraph below is particularly interesting, since such generalizations don't seem to follow very well from the methods described.
I also would wish people would stop making analogies between the mind and the computer. It is a useful analogy for teaching undergrads and for articles in pop psych magazines, but is very restricting in terms of actual research directions.
Included below is additional text related to the story:
"Visual short-term memory is a key component of many perceptual and cognitive functions and is supported by a broad neural network, but it has a very limited storage capacity," Marois said. "Though we have the impression we are taking in a great deal of information from a visual scene, we are actually very poor at describing its contents in detail once it is gone from our sight."
Previous findings have determined that an extensive network of brain regions supports visual short-term memory. In their study, Todd and Marois showed that the severely limited storage capacity of visual short-term memory is primarily associated with just one of these regions, the posterior parietal cortex.
Todd and Marois used functional magnetic resonance imaging (fMRI), a technique that reveals the brain regions active in a given mental task by registering changes in blood flow and oxygenation in these regions, to identify where the capacity limit of visual short-term memory occurs.
The brains of research participants were scanned with fMRI while they were shown scenes containing one to eight colored objects. After a delay of just over a second, the subjects were queried about the scene they had just viewed.
While the subjects were good at remembering all of the objects in scenes containing four or fewer objects, they frequently made mistakes describing displays containing a larger number of objects, indicating that the storage capacity of visual short-term memory is about four.
The article, and the researchers in the article are making an assumption about intelligence: they're assuming raw information processing power IS intelligence. I would argue that a more substantial defining factor is recall of previously processed information and the clarity of that recall. In school, the Cram -->Take Test --> Brain Dump method works but doesn't foster leaning in the way that creates "intelligence" by my definition. If everyone were to re-take their final exams from their senior year of high school/college TODAY I would argue that those doing the best overall were the most intelligent, particularly if their school-age years were long ago.
suposedly people with higher scores on these are somehow smarter,
Not smarter, just better able to navigate the rote kinds of query and response that measure success in academic environments.
blog