This article uses number memorization as an example, so people here might find it interesting:
The problem with short-term memory—and the problem that Steve was coming face-to-face with—is that the brain has strict limits on how many items it can hold in short-term memory at once. For some it is six items, for others it may be seven or eight, but the limit is generally about seven items—enough to hold on to a local phone number but not a Social Security number. Long-term memory doesn’t have the same limitations—in fact, no one has ever found the upper limits of long-term memory—but it takes much longer to deploy. Given enough time to work on it, you can memorize dozens or even hundreds of phone numbers, but the test I was giving Steve was designed to present digits so fast that he was forced to use only his short-term memory. I was reading the digits at a rate of one per second—too fast for him to transfer the digits into his long-term memory—so it was no surprise that he was running into a wall at numbers that were about eight or nine digits long.
Still, I hoped he might be able to do a little better. The idea for the study had come from an obscure paper I had discovered while searching through old scientific studies, a paper published in a 1929 issue of the American Journal of Psychology by Pauline Martin and Samuel Fernberger, two psychologists at the University of Pennsylvania. Martin and Fernberger reported that two undergraduate subjects had been able, with four months of practice, to increase the number of digits they could remember when given the digits at a rate of about one per second. One of the students had improved from an average of nine digits to 13, while the other had gone from 11 to 15.
Skimmed over it due to work today, but I feel that it is the combination of two other psychological phenomena. The learning curve and problem solving.
The memory story is the problem solving part. Maybe the article adresses it to be different, I didnt read it fully, so correct me if I am wrong. I read a similar story before, what was noticed there was that the subject didn’t train his memory, he trained his brain to make shortcuts, chunks. Take the 10 digit number, 5718866610, I bet you if they showed him this number on the first day as even without advanced chuncking, this number is easily memorable with the following chuncks. 5-7-1-88-666-10, those are only six things to memorize.
What my studies into memory showed was something all of us already know: we don’t remember information, we remember meaning. I can show you a 27-digit number and once I tell you the meaning, you will never forget it.
365366125231302928457246060
these are the chuncks with their meaning:
365366 - a year can have 365 or 366 days
1252 - a year is 12 months / 52 weeks
31302928 - a month is 31, 30, 29 or 28 days.
45 - a month has between 4 and 5 weeks.
7 - a week has seven days
24 - a day has 24 hours
6060 - an hour has 60 minutes and a minute has 60 seconds.
If you can train your mind to find those connections, memorization becomes very easy. Even the 82-digit number has chuncks that I can immediately spot. 44344, 0222132, 010203, even 4345510355530 which my brain spotted as a single digit. The number had more digits but these were the chucks I found right away AND remebered without mnemonics. Probably harder when they are presented to you one by one, but then again I didn’t have that specific training.
That is the mnemonics bit, the rest of the article seems very focused on the learning curve story, and I am rolling my eyes at the fact that it is “again one article with the same information.” That blocked my focus a bit, the story has been told many times. Some tell it with the 4-minute mile, some tell it with the glass ceiling in weight lifting, this one with the marathon times.
They are interesting articles, but it is a subject that I have seen around too often unfortunately. This is by far the longest one I have seen, but to me it didn’t add much. Standard learning curve story. You start off exploring the thing you want to learn, then you start learning the thing you want to learn, then you hit a plateau because your body/mind does the task automatically and thus you are no longer forced to improve. “Going with the flow” proves to no longer improve you. Then if you focus on specific aspects rather than specific goals, you still improve. If I memorize cards with something that ticks every second, and with every tick, I have to put a card down, it shows me my weaknesses.
A similar thing I did while learning at school. I made flashcards and had 5 boxes. I’d grab a card and answer it, if I had answered it correctly, I was allowed to move it into box 1. Wrong, and it went back to the the pile, I had to answer it until I got it right. When all cards are in box 1, I do the same, grab one and put it into box 2 if I answered correctly. Back into box 1 if I was wrong. All the way until all cards were in box 5. That helps to focus on the weaknesses while also keeping the strengths in practice.
And it goes without saying that in the experiment, the people were not using mnemonic techniques (at least not established techniques.) If they had they would’ve improved so much more. Isn’t the record for spoken numbers over 400?
The box method referred to is the Leitner system, and is essentially a method of rote memory (the flashcards) and distributive practice (the box system)
.
I’m interested in any info that relates to short-term memory, especially now that I’m less young than I was many years ago. So, my comments refer only to the first part of the very long article that JC linked to: the part referring to Steve Faloon.
I googled for some terms that appear in the article: “chase carnegie mellon faloon”.
That link gave the name of the second test volunteer as “Dario Donatelli”, and the following string of quotes from the article:
Most people can remember between five and nine digits; unusual folks might recall 15 to 18. Imagine the surprise of Carnegie Mellon University researchers in the late 1970s when they came across a student who could master 82 digits.
Of course, he had a system.
A track and cross-country runner, Steve Faloon was accustomed to timed runs and, when forced to remember a string of numbers, tended to break them down into segments recognizable to him as typical times for a quarter mile, a mile, two miles, 10 kilometers, etc.
…
The lead researcher, William Chase, set out to prove that the student’s surprising skill in the digit span test could be passed on to another. So in 1979, he recruited another runner, Dario Donatelli, to see if he could duplicate Faloon’s feat.
“The whole point,” Donatelli said, “was that he used a coding system and he was able to teach me the system.”
…
…by the end, Donatelli was able to recall an astounding 113 digits.
So Faloon, IMHO, was NOT using short-term memory. He had a mnemonic system, which he taught to Donatelli.
Here’s a link that demonstrates the difference between “trained memory” and “short-term memory”. The subject is a world champion: Alex Mullen.
The first sentence is:
"The reigning World Memory Champion, Alex Mullen, can memorize the order of a deck of cards in 17 seconds. But in some ways, he’s just as forgetful as the rest of us.
“I still forget plenty of basic things, like where I left my keys,” said Mullen"
So the training required to become World Memory Champion does not improve the short-term memory.
For me, that’s a disappointing conclusion. But it’s what I expected. The two types of memory are housed in different parts of the brain (see later).
Here are some links relating to long-term and short-term memory, separated by dashed lines:
On the other hand, short-term memories are stored in electrical “capacitance”. A common example of capacitance is the “starter” in a neon tube. The charge fired by the starter would be similar to the neurons fired by the transmitting synapse to initiate a short term memory. That memory then decays exponentially to a useless value in a very short period of time.
The link contains a lot words in Klingon and Greek. But if you mentally filter out those words, the article actually becomes quite understandable and interesting. Even if you only read the first section, “Abstract”, you can learn a lot.
If you search that previous link, you’ll find many references to the beneficial effects of calcium on something called “synaptic facilitation”. If you look at Figure 2c, you can see that synaptic facilitation increases the neuron firing rates to their original peak values at the start of the short-term memory cue.
The link contains the following paragraph:
George Miller published a paper in 1956 entitled “The magic number 7, plus or minus two: some limits on our capacity for processing information” [14]. A key issue was that the capacity of short term memory (for example our memory for a list of telephone numbers) is approximately 7±2 items. For visual short term memory, the capacity may be closer to 4 items
So Miller seems to be saying that the capacity of a short-term memory is 7 TELEPHONE NUMBERS, not 7 DIGITS.
Miller also says: “For visual short term memory, the capacity may be closer to 4 items”
I’m wondering if the Memory Palace guys would disagree with that statement
This article, which compares long and short-term memories, suggests that the capacitance decays to a useless value in 10-15 seconds. That short value explains Miller’s Law of 5-9 items per short-term memory. http://www.human-memory.net/types_short.html
I think capacitance depends on the distance between the “plates”, the area of the “plates”, and the dielectric (insulating material between the “plates”). For memory capacitance, one “plate” is a synapse that fires neurons to a second plate or synapse.
Since we are talking about “organic” capacitance rather than metals and wires, none of the dimensions of the capacitance can be changed by human action. That leaves only the dielectric, which can possibly be varied by the amounts of calcium, sodium, and potassium salts floating around in the brain.
So, the capacitance can also vary depending on the amount of calcium.
Maybe it’s the mention of “calcium, sodium, and potassium salts” that explains the “brain food” ads that are normally plastered all over websites related to memory (but not this site ).
Following link says that short-term memories are simply re-usable “scratch pads” with a life of only seconds
Just to clarify. Neurons do not fire signals to the post-synaptic neuron. Rather, bags of chemicals diffuse across the space, so the capacitor model only goes so far.
It is true that without rehearsal, short term memory decays quite quickly, but that is old news.
As you correctly conclude, anything above 9 items on average (items can be chunks of course) is not using short term memory but long term memory and to accomplish this, there will be a system lurking somewhere.
However, I think that many people when talking about poor short term memory, really mean issues with remembering things for a short time and not really actual short term memory. In this case, improvement can often be realized by training oneself to be originally aware ( mnemonics at the point of awareness can also help)
For example, the typical ‘where are my keys’ scenario, is most often a failure of encoding, not recollection. People can’t remember where they put their keys because they didn’t encode the information in the first place.
This is interesting because I also use chunks. The difference is that I don’t actually have any meaning behind the numbers I chunk. I just chunk numbers I like. The 27 digit example you gave, I would chunk it like this:
365-366-125-231-3029-28-45-72-460-60.
I am familiar with 365 like everyone else but I have no traditional meaning behind the other chunks besides that I just simply like the numbers this way.
My only explanation for this is that I have synesthesia because I also have a harder time remembering numbers I don’t like. I cannot chunk the 28,45 and 72 together, my mind forces me to chunk them seperately into 28,45 and 72, I really feel them that way.
Most synapses are chemical; these synapses communicate using chemical messengers. Other synapses are electrical; in these synapses, ions flow directly between cells
.
.
The following link says:
The signaling process is partly electrical and partly chemical.
Yes. But the neurons with direct electrical activity are quite rare in comparison with electrochemical communication, so I don’t think the capacitor model, which leaves out the majority of neurons, is that applicable to neural communication.
This is great.
I somehow recall this from a book i read in the past, not sure but i think it was jonathan hancok, not sure. I have so many books on memory improvements. It’s brilliant how easy we can memorise once we give it a meaning with something that we already know.
how do you do this without giving any meaning to it?
That was an awful lot of words to get to, the subject being tested did not increase his short term memory but instead developed a mental journey using familiar pegs and gamed the system. I suspect that in almost all cases including n-back that unusual results will be related to engaging the parts of your brain where you keep things longer and can retrieve them more readily. I would guess that MRI’s would probably show some significant differences between how folk think about these. The range of list size that most folk can repeat back under this kind of testing has shown to be fairly consistent and the range pretty small. 5-11 isn’t a stunning difference by my measure of things. Practicing an activity like practicing n-back provides minor improvements in the specific activity but without a system it has never shown to make a difference to working memory.
In the words of the Ronnie Coleman - “Everybody wanna be a bodybuilder but don’t nobody wanna lift no heavy ass weight”.
