How does the brain manage abstract memory

Memory-dependent forms of memory

It is a myth that weekend trainers market at a profit that humans only use a very small part of their neural resources. This is nonsense: there are nowhere in the brain that lie fallow. If that were the case, tissue could be removed from there without having to fear any loss of functionality. But this is not the case.
Wolf Singer

When looking at memory, a distinction must be made between the sensory memory or register (with storage duration in the subsecond range), which can be understood as part of the perception apparatus and also as Ultra-short-term memory is referred to, the Short term memory and the Long-term memory. A subject matter cannot be suddenly stored in our memory in a single step. Rather, it comes first via a sensory register in a Short-term storage, in which it is held ready for a while before it is decided whether it should fade away without a trace or is worth preserving in a permanent form. Information or learning content that is to be permanently stored in our brain requires a fundamental restructuring process on the neurons that takes at least 24 hours. If new information arrives too quickly, its content will compete and delete one another. This can be represented graphically in a very simplified way:



Anyone interested in a tutorial on memory should visit the site "Paths to Psychology" by Gerd Mietzel and colleagues, where interactive exercises, experiments, additional drawings and examples are offered. This graphic - incidentally, a sample of didactically excellently prepared psychological theory! - comes from this tutorial.

Duration of memory according to Eccles

It must be emphasized at the beginning that an exclusively cognitive perspective on memory does not go far enough, because the evaluation of new information and thus the storage is particularly effective if it is used for practical or cognitive problem solving. The resulting consequences such as success or failure also act as motivating factors for further learning. The basis for this is an internal brain reward system that works with the neurotransmitter dopamine and, if successful, ensures a pleasant mood and thus an increase in motivation. Dopamine at the same time ensures gene activation in the neural systems that maintain the relevant short-term memory content and thus the synapse remodeling for long-term memory. The second effect is important for that memorybecause people experience many things and very little remains in long-term memory. The conversion from short-term memory to long-term memory is largely controlled by dopamine, because success is fun. Dopamine thus has a direct effect on memory formation and helps to automatically and sustainably save a successful strategy. At the same time, people retain emotionally tinged experiences better than neutral ones.

Before you read any further ...

what tasks the different memory stores have, make some of the short ones Memory span tests, the Bernhard Jacobs from Media center of the Philosophical Faculties of Saarland University among other interesting test procedures. On the one hand, you can use simple and more difficult memory material to carry out a targeted check of the content-specific memory spans; on the other hand, you can adaptively determine your own memory span, i.e. the program begins with a certain memory span and adapts the further specifications according to your performance. Just click on the brain in the graphic on the left ...

The orderly structure of the memory memory and thus the subsequent access to the knowledge stored there depends on the conceptualization of information, whereby the formation of categories plays a fundamental role, i.e. the possibility of classifying information into classes of meaning that become increasingly differentiated with age and experience. Completely new categories are likely to be difficult to form in old age unless new things are developed through practical problem-solving strategies that result in real successes and failures.

Sensory register - Ultra-short-term memory

It stores all sensory data very briefly (for approx. 0.5 - 2 seconds) in uncoded form and transfers them to a filter, the

  • selected according to certain characteristics
  • performs a first pattern recognition in the sense of "pre-attentive processes" and
  • bundles information in the sense of "chunking".

Pre-processed in this way, the information is stored in a short-term memory.

Short term memory

The method of pronouncing numbers in German is already 4000 years old and goes back to the Germanic ancestors, because at that time numbers were marked with I for one and X, for ten comparable to Roman numbers. The difference between Indo-European and Latin lay in the arrangement, because 24 was written with IIIIXX and not as in Latin XXIV. Then IIII and XX were read, so consequently twenty-four. In England, the pronunciation of numbers changed in the middle of the 15th century, because this should be economically advantageous and make learning easier for children, so that “fiveandfourty” became “fourtyfive”, but the change only took effect from the number twenty. The mathematicians Jakob Köbel and Adam Riese advocated a change in pronouncing all numbers from 13 onwards differently than usual, and Adam Riese suggested in his book for written arithmetic to pronounce 6789 as "six thousand seven hundred eighteen nine-one". However, this proposal never caught on. Miller (1956) found in his research that the Working memory is able to hold 7 (+/- 2) elements at the same time. The amount of information of these elements (chunks) depends on the Preprocessing from. You can try to remember individual numbers 1, 2, 4, 6, 9, 6, 7, 3 (...) or the "chunks" 12, 46, 96, 73 (...).

However, this rule only applies to them Western world, because in Chinese, for example, in which the numerals are shorter, remember Chinese nine digits on average. There are several reasons for this: Mandarin numbers have several advantages, including the brevity of the numerals. The Chinese 7 (qi) has one syllable, the English Seven and German Seven have two. If an American is asked to briefly look at a sequence of seven numbers and then write it on a piece of paper, they make a mistake 50 percent of the time, while this rarely happens to the Chinese. Also at Learning the numbers Chinese children have advantages: If three-year-olds from China and the USA are supposed to count in their respective language, then they get the same distance, usually up to eight or nine. At the age of four, the situation is completely different: Children from the USA can get to 15 with difficulty, while Chinese of the same age can reach 40 or 50. This difference is also in the strictly logical rules for numerals. While Americans say eleven and twelve, so for eleven and twelve as in German they also use their own words that the children have to learn like vocabulary. The Chinese, on the other hand, put the eleven and twelve together from the number words for ten and one or two. Eleven is called shiyi (ten-one), twelve shier (ten-two). Children's brains stumble over illogical structures

With the numbers 13 to 19 it becomes illogical in English and German: One says thirteen, fourteen and so on or thirteen, fourteen. First the ones are mentioned and then the tens - when writing down the numbers it is exactly the other way around. After all: From 21 onwards, the English numerals change to a logical structure (twenty one instead of one twenty), in German the one remains in front. Children's brains keep stumbling over this illogical structure. Incidentally, arithmetic is also easier for Chinese Kingers because of the simple syntax, because short numerals occupy the brain less and are easier to remember when you the little multiplication table learns. Miller's work, and later studies, have shown emphatically how closely language and mathematics are intertwined, and that many problems in arithmetic, ultimately, too Language problems are.

Incidentally, experiments have shown that already baby To be able to distinguish sets, i.e. to recognize when something exists only once or something is present multiple times, so that the sense of numbers is obviously innate. By the way, you can too Primates Recognize quantities up to three or four.

Waiter have a reputation for having good memories. In experiments with particularly "remarkable" waiters in Buenos Aires, George A. Miller's chunking hypothesis was confirmed, because waiters use a strategy in which they link the person, their place at the table and the order in their heads, and that exceptionally fast, which is both a matter of experience and practice.

Try it yourself with telephone numbers, whereby each sequence can only be read once and then played back. In practice, you can try it out on mobile phone numbers or international numbers, because these are usually composed of a country and area code or a provider code and a usually quite long number. The area codes usually also form well-memorized chunks due to the advertising or simply the frequency of their occurrence.

The Short-term memory capacity is therefore about seven objects. The value does not vary from one individual to another by more than about plus or minus two. This means that we can still keep an arbitrary sequence of digits of length 5 in short-term memory, for example 2 7 6 4 9, of a sequence of length 15, for example 2 7 6 5 8 3 7 5 8 4 3 6 6 7 5, but only fragments.

Kyriaki Sidiropoulou of Rosalind Franklin University (Chicago) found neurons from the prefrontal cortex of mice that continued to fire for up to a minute after receiving a short, rapid series of signals. These neurons can transmit signals for up to a minute when they receive a short series of signals, for which the metabotropic glutamate receptor 5 (mGluR5) in the membrane of the cells is responsible. If the receptor is activated by incoming signals, it sets in motion a cascade that leads to a longer-lasting postsynaptic depolarization. It is precisely at this point in the cortex that, in humans, new sensory impressions, for example, are linked with memory contents from different brain regions and temporarily stored until they can continue to flow. So the researchers probably have one Short-term memory mechanism discovered and identified with mGluR5 the receptor that sets this firing in motion.

The Storage period in the short-term memory is only very short if it is only stored once, a few seconds. If we want to keep the content longer, we have to repeat it in our minds. This is easiest with linguistic information, a little more difficult with other types of information. If we don't take special precautions, short term memory becomes like a snake managed. The first object that exceeds the capacity of the short-term memory will therefore displace the object from the short-term memory that has been there the longest. This can be countered by selectively repeating the content, treating it as if it were newly stored each time, so that it is possible to choose which object should be displaced first.

Due to the first precoding, information is represented acoustically, visually or semantically. However, deletion can also be due to Interference or by yourself Passage of time happen. The contents of the short-term memory are retained if one of the following two processes takes place:

  • Simple "sustaining repetition", e.g." saying "to yourself over and over again, feeds the information back into the short-term memory or
  • "Elaborate", e.g. through reorganization, categorization, linking to existing information,

transfers the contents of the short-term memory to the Long-term storage.

Every year, up to 5.4 billion gigabytes of new information are produced around the world. Statistically, every person produces around 800 MB of data annually - this value is significantly higher for the average American and European. The United States annually produces over 40 percent of the new information data that is saved on magnetic and optical storage media, with the vast majority of the printed information being produced by individuals (office supplies, e-mails, etc.). Printed products in the form of books, magazines, and newspapers are in the minority, a third of which is made in the United States. One suspects 170,000 GB of information on the Internet. Radio and TV produce around 3.5 million GB of new information in a year. By telephone (mobile and landline) around 17.3 exabytes (17.3 billion GB) of data are forwarded each year, most of which, however, are not stored.
source: (03-11-01)

The stored objects do not necessarily have to be as elementary as the digits in the previous example, because the information is stored in so-called Bundles (chunks) stored, the nature of which depends on what kind of content the long-term memory makes available. In any case, a bundle is a semantic unit, a concept. Words, for example, can be used for this. A sequence of five randomly chosen well-known words is as easy or difficult to remember as a sequence of five digits or five letters, although the five words together comprise considerably more than seven letters. With a long word list of more than nine words, the attempt becomes just as impossible as with a long list of letters.

The physicist Helmar Frank defined it Elementary quantity of human information processing: The memory span, measured in seconds, and the information processing speed, measured in bits per second, determine the storage capacity or the channel capacity of the short-term memory or working memory (cf. Lehrl et al. 1991). Long-term memory also depends on this working memory, i.e. the speed with which we call up and store information. The memory span is the number of elements we can think with at the same time. It depends on intellectual ability and amounts to three for a toddler, five for an IQ 100, seven for an IQ 115 and nine for gifted children. The "Erlanger School of Information Psychology" (WD Oswald, S. Lehrl, H. Frank) found that 5 percent of the test subjects when sorting cards, repeating monosyllabic words, capturing images or texts - also on the computer or in the newspaper - , is able to absorb, store or recall from memory twice the amount of information per unit of time than the rest of the population. About 30% of the population achieve medium results. It also turned out that the scientifically and technically gifted or highly skilled are identical to the "fast information processors" of the Erlangen psychologists (cf. Weiss 2002).

The model of short-term memory, which is only useful heuristically, has been replaced in recent years by the model of the Working memory replaced, which comprises three systems: A spatial-visual "notepad" for short-term storage of visual impressions, an "articulatory loop" for storing verbal information that is kept available for communication processes through internal repetition and a central "executive" that manages both systems and links the information from these systems with long-term memory.

The visual short-term memory for example, helps people to remember objects for a short period of time, even if these objects are no longer visible. Contrary to previously assumed, the visual short-term memory is based not only on one type of information about an object, i.e. only about its color or only about the name of the object, but several types of information can be maintained in the short-term memory at the same time. Apparently short-term memory is more complex than previously thought. Liu et al. (2929) recorded the brain activity in epilepsy patients with the help of electrodes, while the test subjects were presented with images of objects such as a banana and were asked to memorize them for a short time. It turned out that neural networks Process images in similar steps as humans, because if a human or a neural network sees a banana, simple properties such as the yellow color and the smooth texture are processed in the first step. As you look at it, the processed information becomes more and more complex, because this is how people and the network finally recognize that it is a special crescent shape, until the banana can be named in the end. The different processing steps of the neural network were compared with the brain data and determined which activity patterns belong to the processing of simple visual properties such as the yellow color of the banana, and which to more complex properties such as its name. The objects were not only depicted in one form in short-term memory, as originally assumed, but in several forms at the same time, namely when looking at simple features of the banana are initially kept present in the memory, and only then are more complex properties such as the name added. During the memorization phase, however, simple and complex information is maintained at the same time.

The short-term memory is already developed in the womb and 30-week-old unborn babies can remember events for minutes. The Brain of the unborn is known to develop slowly and is not yet fully developed at birth, but the structures from which memory is formed are created very early. Jan Nijhuis (Maastricht University) examined this with ultrasound Short term memory of fetuses (Unborn between 30 and 38 weeks) of healthy pregnant women using the habituation effect, which measures the number of impulses the fetus needs to get used to a new stimulus. The response of the fetuses to the stimuli was followed with ultrasound until they no longer responded. If the youngest fetuses were confronted with the same stimulus again ten minutes after the initial habituation, they apparently recognized it and reacted calmly to it. When the researchers did the same experiment with 34-week-old fetuses, they could still remember the stimulation four weeks after the first test (Strauch 2009).

The classic:
Miller, George A. (1956). The Magical Number 7, Plus or Minus Two: Some Limits on Our Capacity for Processing Information. Psychological Review, 63, pp. 81-97.

Other sources:
Dambeck, Holger (2010). Twenty-one beats twenty-one.
WWW: (10-02-23)


The following famous psychological experiment provides an impressive example of the efficiency of such abstractions: In 1965, DeGroot had his test subjects briefly examine chess positions on chessboards and then reproduce them from their short-term memory. As expected, very good chess players were often able to reproduce all pieces correctly, but always quite a few. Chess beginners, on the other hand, could only ever correctly remember very few pieces. But if you put the pieces on the board at random instead of specifying a real position from a chess game, the performance of the good players sank to the level of the beginners; the beginners were equally good in both cases. This shows that problem-specific abstractions and patterns in long-term memory can multiply the short-term memory capacity that is effectively available for a specific purpose. In the case of the game of chess, these abstractions are frequently occurring combinations of positions of several pieces that form a conceptual bundle.

[Image source:,property=inline.gif]


By the way: If human perception improves during learning, it is not so much because more information reaches the brain than in other people, but rather the brain learns to begin with the information more and more. In particular, plays with more and more Expertise the interpretation of the stimuli plays a more important role, with the learning process taking place primarily at the decision-making level. Experts therefore do not see or hear more or better than others, but their brains simply evaluate the hearing or visual information better and more thoroughly.

A little experiment on short-term memory: The "letter row".

You need at least two people for this, but it works best with a whole school class. Someone writes hidden rows of letters that make no sense, whereby the rows start with two letters (e.g. WL) and then always have one more letter (e.g. a row with 5 letters: KPZCA, or one with 10: BKLMNRSÜTQ).
These rows are read out slowly one after the other to the teammate or the class and after about five seconds each row is written down from memory. You start with the second row and work your way up to the 10 or 2 series. Most of them manage 7 or 8 letters without any problems, but then the first mistakes creep in.

Perceptual capacity of moving objects

To investigate how many moving objects people can follow at the same time, Franconeri & Alvarez showed the test subjects 16 moving points on a screen. At the same time, they told the test subjects which of these points they should keep an eye on when looking at the screen from half a meter away. If the points moved at a speed of one centimeter per second, the participants succeeded in following the movement of eight points exactly for one minute, which corresponds roughly to the ability to observe a football player in a group of other players at ten meters run far away at 700 meters per hour. However, the test participants failed to track more than eight objects, even when the speed was slower. Obviously there is an upper limit to how many objects a person can keep an eye on at the same time. At a speed of 15 centimeters per second, the participants could only follow one point; if the speed was even higher, they could not succeed at any of the points.

George Alvarez (Cambridge) and Steven Franconeri (Evanston): Journal of Vision, Vol. 7, No. 13, Article 14

What is a "chunk"?

Let us assume that we read individual rows of letters of varying length to our test subjects, e.g.

Then we will find out again: at some point you will become insecure and no longer be able to reproduce all letters correctly. That will be about

be the case, here there are 8 letters that have to be kept.

We conclude: humans can store a sequence of 7 letters in their short-term memory without errors. If you just order the 8 letters differently:

it is not a problem at all to memorize them. A single "higher order" piece of information (the word) is formed from disjointed pieces of information (letters). This process is called "chunking", and the term "chunk" refers to a unit of information that combines several elements into a single meaning. You can remember 7 simple words as well as 7 letters, so

The 7 chunks (here words) contain 35 letters that we could never remember as disjointed pieces of information. If you form a sentence from this:

you can also easily remember these. At some point you reach a limit when chunking, e.g. you can no longer remember seven sentences, seven stories and seven books. Chunking is one of the most effective strategies, which our brain developed in order to be able to memorize large amounts of information in a compressed form. Chunking is one intellectual achievement, so no mechanical process. Compressing pure data into something "meaningful" requires knowledge of meanings and the ability to link information.

In the Application of this capacity limitation of our short-term memory one should be careful with the “magic number” 7, because the capacity can be influenced by a combination of environmental influences such as noise stress, lower motivation or simply strive towards zero in a bad mood (cf. Wirth 2002). The subdivision into different memory systems, such as visual and phonological memory, must be taken into account, since the corresponding structuring of the information makes it easier or more difficult for people to form connections in memory.

Using: Wirth, Thomas: The memory span ... and the magic number 7.
WWW: (01-11-17)


A recognition of such template enables the storage of a significantly higher total number of facts in the short-term memory and thus the possibility of doing many tasks much faster and more safely, because no external memories such as notes or files have to be used and no errors occur that would otherwise make the job endanger the task.

Another possibility for improvement arises from observing that apparently any sensory system about a own short-term memory disposes. So far, only the speech / hearing system and the visual system have been thoroughly researched and evidence of two separate short-term memories for verbal and visual information has been provided, but there are probably other short-term memories for sounds, smells / tastes and tactile sensations.

Linguistic information is provided in the so-called phonological short-term memory stored regardless of whether they were delivered visually or acoustically. In contrast, non-linguistic visual information in the short-term visual memory saved.

This means that a simultaneous use of the two short-term memories and an associated increase in storage capacity can be achieved if combined linguistic and graphic modes of presentation are selected for the presentation of the information, provided that the graphic parts are not interpreted linguistically but graphically. It should be noted, however, that visually stored information is less precise than verbally stored information. Therefore, precautions should be taken as far as possible against errors in the visual storage, e.g. by adding all the visual information linguistically.

See: Learning strategies - learning types

Chewing gum improves memory - or it doesn't

Increasing one's own cognitive performance in the long term is a wish that probably all of us humans have. However, the systematic reviews already carried out to increase this ability have not been very successful. In a study, the medical psychologist S. Lehrl claims that chewing gum has massively increased the learning performance of his students. He also emphasizes in a Swiss journal that schoolchildren should actually be allowed to chew gum in a lesson if success is more important than aesthetics. Lehrl also tested the effect on wakefulness and learning performance in gum-chewing / non-chewing students. In the longer studies, the level of alertness of the non-chewers slipped below the initial level of alertness, with the chewers approaching a state of full alertness. With four studies he also tests whether crouching and non-crouching differ in terms of the increase in knowledge. The subjects with chewing gum fared significantly better than those without. Lehrl concludes from this that with this very simple measure, the learning performance in frontal teaching can be increased. In addition, Allen, Galvis and Katz published a study in 2004, where the positive influence of chewing gum could not be sustained. It should be added, however, that this study is only an abstract of a few lines long. Therefore, the quality cannot be assessed. The studies by Lehrl and Allen et al. are to be viewed very critically, however, as they cite some points of criticism (e.g. very small case numbers, missing sample description, etc.). It should also be noted that no solid, methodologically careful studies with large samples have yet been published (cf. Rost, Wirthwein, Frey & Becker, 2010, pp. 39-42). British scientists also claim to have found out that chewing gum promotes thinking. The studies were carried out at the University of Northumbria: 75 volunteers were subjected to various tests. A third of the subjects were given a piece of chewing gum, a third were allowed to chew with an empty mouth, and a third had to refrain from chewing movements. To everyone's surprise, the gum chewers fared clearly the best, with short and long-term memories even improving dramatically. The ingredients of the chewing gum have no effect, the researchers suspect that the act of chewing stimulates the brain. Heartbeat and pulse increase when you chew, which brings more oxygen to the brain. One theory says that the chewing process makes the brain believe that something is being eaten, which increases insulin production, which is also good for the brain.

Rost et al. (2010) refute in two experiments the thesis that chewing gum is smart because a short-term increase in intelligence is and remains an illusion. The first experiment examined the effects of chewing gum on cognitive performance. The sample consisted of 544 pupils in the fifth and sixth grade (Realschule and Gymnasium classes), with an average age of 11 years. In addition, the intention was to measure different facets of intelligence. The students were checked in various tests with different processing times, the children were randomly divided into two groups (chewers / non-chewers) and the survey was carried out in two separate rooms. However, the result did not produce any significant chewing gum effect. In the second experiment, the short-term effect of chewing gum was tested on N = 486 pupils in the fifth and sixth grades (two comprehensive schools and one elementary school with a special level). The mean age was 11.4 years. The survey of the short-term attention and concentration performance was measured by various tests (attention test, stress test, etc.). The second experiment was also carried out in two groups or by trained investigators and was carried out in a highly standardized manner. The statistical evaluation was exactly the same as the evaluation of the first experiment. In the second experiment there is a very interesting result. In some points (e.g. concentration performance) there were slight statistical advantages in favor of non-buyers. However, these are only very small and therefore there is no reason to assume that chewing gum has a positive effect on the memory performance of students in fifth and sixth grade. Based on these two experiments, the findings of Allen et al. and other researchers are not supported (see Rost et al., 2010, pp. 42-46).

Promoting cognitive performance is a common goal. More and more people want to increase this performance with little effort and therefore there are always research reports with the content of how one can achieve a significant increase in cognitive performance with simple measures. Educational psychology not only has the task of developing and improving theories, but also of evaluating assertions. It is thus assessed whether the findings do not consist of frivolous popularized effects but whether they also correspond to empirical standards. Even the famous Mozart effect (listening to pieces of music by Mozart) does not make you smart and this is exactly where the chewing gum effect can be classified. On the basis of the above-mentioned two experiments, which were carried out under sufficiently strict criteria, no advantages in favor of the chewers can be determined. On the contrary, in some tests there were small superiorities in favor of non-buyers. Thus, there is no assumption that chewing gum improves the memory performance of children of the age group tested. The results may look different in older pupils (students, adults) and lead to a positive chewing gum effect. Furthermore, the effect of the glucose supply through the sugar in chewing gum should also be taken into account in further studies. This would be easy to do because there are sugar-free chewing gums. These studies also show how easily criticizable studies are published in public as a kind of “panacea”, although they have not been carried out sufficiently (small samples, etc.). In order to increase or promote cognitive performance, it takes years of promoting the learning environment from birth and, in addition, it is very important to talk to your own children as much as possible.Furthermore, intensive preschool and school support that lasts for years is important, because neither listening to music or chewing gum can improve performance (cf. Rost et al., 2010, pp. 46-47).

source: (02-03-15)

Rost, D., Wirthwein, L., Frey, K. & Becker, E. (2010). Does chewing gum improve cognitive performance? Two experiments of a special kind. Journal for Pedagogical Psychology, 24, 39-49.

Long-term memory

While the contents of the Short term memory as Activations of neurons stored (i.e. as brain activity) are the contents of the Long-term memory in the form of Connections between neurons stored (i.e. as a brain structure) - the exact reality is a lot more complicated and therefore not yet fully known. The popular theory is that when you learn, memories are in stable protein chains get saved (Consolidation). However, according to the latest studies, the long-term memory is probably more unstable than previously assumed. Karim Nader, Glenn E. Schafe, and Joseph E. LeDoux (University of New York, 2000) suggest that every time memories are recalled from long-term memory, they have to be rewritten in the brain. The consolidation is not a one-time process afterwards, but the memories come into one every time they are called up chemically unstable phase. Then they have to be consolidated again, i.e., written into long-term memory. To do this, new protein chains have to be formed. This renewed consolidation takes place in that part of the nervous system which forms the actual memory. In this study on rats, the area of ​​the brain that stores the experience of fear was examined.

As we know today they are Almond kernels for example for the origin of the Fear memory essential, because the more they are activated, the more vivid the memory of an unpleasant experience. They respond quickly to possible dangers, because the almond kernels react in an experiment to a warning signal they have learned before this signal has been fully processed in the cerebral cortex and penetrates consciousness. In a sense, the stimulus takes a shortcut from the thalamus directly to the amygdala, i.e. people get scared before they even understand what is happening. While the hippocampus in particular stores conscious, explicit memories, for example of facts or personal experiences, in declarative memory, the amygdala primarily stores emotional memories that can later appear unconsciously in people, whereby they even outlive the information in declarative memory or themselves decouple from it. This can be seen from the fact that after years of entering a school people develop fear of school again or feel uncomfortable just because of the smell.

According to the current theory, the brain produces new proteins with each new long-term memory in order to stabilize the changes in the neural network associated with the information. In this way, the memories are firmly anchored in the brain and last a lifetime. According to Routtenberg & Rekart (Neuroscience 2005, 1, p. 12), however, long-term memory is constantly updated and is therefore much more flexible than previously assumed. Long-term memories are not saved through permanent changes in the nerve network, but with the help of temporary changes in the nerve contact points. The principle is therefore similar to that of short-term memory, except that various feedback effects between several networks keep the information in long-term memory permanently alive, which means that it can also be changed afterwards. According to a new, dynamic model, the shape, distribution or activity of existing proteins at the synapses, i.e. the contact points between the nerve cells, change during learning. These changes are not permanent, but reversible. In order to permanently store what has just been learned in the brain, the "hidden practice" of what has been learned is also necessary, i.e. a positive feedback system between different networks that continuously updates the previously stored information and constantly re-coordinates it. Since a reminder is stored in this way via different networks, it is also possible to quickly find memories that have not been accessed for a long time. Networks that have been destroyed in this way, of which information is still available in the other networks, could also be rebuilt.
However, there is also the phenomenon in human development that skills once acquired are forgotten or forgotten again. Olivier Pascalis et al. (University of Sheffield, Science 2002/5) found that people tune their brains to faces they see frequently in the first few years of life and compare new faces to the template of those old faces. Six-month-old toddlers can easily distinguish between individual human and individual monkey faces. Small children who are only three months older can distinguish similar-looking people, but monkeys are no longer able to do this. It is believed that there are remarkable changes in the fine-tuning of perception in the brain during the first few years of life, so that recognition of the features important to distinguish faces is acquired first and then lost when they are no longer useful for facial recognition walk.

This mechanism also establishes the two most important properties of long-term memory. It has a unlimited storage period and one almost unlimited capacity. A third property is important, but it is not that easy to explain. This is the fact that for the transfer of information into long-term memory apparently only a very low bandwidth is available. So relatively few things can be memorized over the long term within a given time. Here, too, the reality is again quite complicated because the bandwidth depends on the storage duration. We can memorize a lot in a short period of time for a few hours or days, but only a little that we keep for our entire life. This is because the conversion of information into brain structures is a process that takes several weeks, during which the information is temporarily stored in other, still volatile, ways.

The functionality of long-term memory consists of two parts. First, we can Recognize things and second, to these things are a multitude of Relationships stored, especially semantic such as numerous variants of "part of" or of "is a" ', as well as temporal and spatial such as the sequence of events in a story or digits in a telephone number.

Changes in the brain structure occur during storage - simply move the mouse over the image!

The storage always takes place in as possible abstract shapebecause a lot of details can be dispensed with and less space is required. Conversely, this mechanism means that it makes a big difference to the ease with which something can be retained, which abstractions are available to represent what is to be remembered - and are then also used, because this only happens partially automatically; to another part it is a conscious effort that as Elaboration referred to as. The more thorough the elaboration, the better the memory.

Not only is learning made easier, but also the retention of complex concepts, if appropriate Previous education is available. Therefore, one should always attach great importance to conveying basic concepts and clarify the connection between a fact and these basic abstractions.

Experiments have also shown that facts about which a Explanation or justification was included, were retrieved correctly better than others that were learned without justification. This is probably due to the fact that the reasoning is one further retrieval path opened to the desired fact: the fact could be remembered directly or the reason was found and from there the fact.

The same principle applies to almost everyone mnemonic systems underlying. Additional paths are created through which the desired information can be called up. Taxonomies, i.e. hierarchical arrangements of terms, are particularly helpful in this regard. An important finding from psychology on the subject of long-term memory concerns the Importance of the context of the content or also the spatial environment for successfully recalling memories. Godden and Baddeley tested the effects of changing environments on memory in 1975 by having divers learn a list of words on the beach and other divers in the water. Both groups were able to retrieve the words roughly equally well at the same place (beach or water). However, if the in-the-water learners call up on the beach or vice versa, the call performance sinks by 40 percent.

This effect comes from the fact that subconsciously elements of the environment are used to keep it Associations to the information to be memorized, which makes it easier to call up the information only if these elements are available as suggestions at the time of the call. The effect occurs not only in relation to local environments, but also for situational contexts on. The latter also include aspects of the internal situation, e.g. the emotional state or experiences recently made. As a consequence, the learning of knowledge and skills should, if possible, take place in application-oriented situations and rooms in order to optimize the later retrieval. (In this respect, learning in the lecture hall, which is typical for the university, is fundamentally more difficult, unless with regard to an exam that is written in the same lecture hall.)

Baddeley and Hitch (1974) extended the short-term memory model with modality-specific components, namely

  • the articulatory loop (phonological loop) to maintain linguistic information,
  • the pictorial-spatial notepad (visuo-spacial scetchpad) to maintain visual imaginations
  • a central executive (central executive) to control the other two subsystems (modality-unspecific)

and proved these mechanisms through experimental disturbances of individual components.

For example, after reading a computer manual, the information that you have to press Ctrl-Q to exit a certain program may remain in the long-term memory, but it may no longer be possible to reconstruct whether it said "To exit, enter Ctrl-Q . " or "If you want to exit the program, press the 'Ctrl' key, hold it down and then press the Q key at the same time". Although the two sentences are very different, they are mapped to the same representation in long-term memory because the abstraction of pressing a control key was already available and the details of the formulation are not important anyway.


See also Forgetting

Sourcelen & literature

Thomas Kramar: "Everything that counts takes three seconds" - How the brain structures the present. Die Presse, 3.2.2001, p. VIII.

Liu, Jing, Zhang, Hui, Yu, Tao, Ni, Duanyu, Ren, Liankun, Yang, Qinhao, Lu, Baoqing, Wang, Di, Heinen, Rebekka, Axmacher, Nikolai & Xue, Gui (2020). Stable maintenance of multiple representational formats in human visual short-term memory. Proceedings of the National Academy of Sciences, doi: 10.1073 / pnas.2006752117.

Pohl, Wolf (2001). Antonio R. Damasio: "I feel, therefore I am. The decoding of consciousness". A review. Enlightenment and Criticism 1/2001 (p. 168 ff.).

WWW: (01-07-10)

Strauch, Stefanie (2009). Memories from the womb. (09-07-14)

Weiss, Volkmar (2002). A critical contribution to the political evaluation of the PISA study.
WWW: (03-06-19)

Wirth, Thomas (2002): The magic number 7 and the memory span.
WWW: (04-06-19).

content :::: message :::: news :::: imprint :::: data protection :::: author :::: copyright :::: quote ::::