THE ARCHITECTURE OF MEMORY

By Gregory Mitchell

In Mind Development we recognize the process of memorization has several sequential stages: Sensory Memory; Short-Term Memory; Medium-Term Memory; and Long-Term Memory - with further sub-categories of memory described. This understanding opens many doorways both to improved memorization techniques and to enhanced awareness of our mental processes, and hence the possibility of meta-cognitive control.

CONTENTS

Sensory Memory
All incoming information is held briefly (1/2 to 2 seconds) in Sensory Memory as a primitive and unanalyzed copy of the actual sensory information. This storage takes place in the Primary Sensory Cortex, which is why this type of memory is so vivid and lifelike, whilst it lasts.

Auditory (hearing) information is held in 'echoic' memory. To understand echoic memory, imagine someone rattling off a string of numbers, then stopping suddenly and asking you what were the last 5 numbers he or she said. To answer the question, you would have to 'replay' what you had heard. Fortunately, the last few numbers would probably still be in your echoic memory, and most likely you could answer the question correctly. If the person had stopped and waited a few seconds before asking the question, however, probably you would not remember any of the numbers - you could not 'replay the tape.' Echoic memories fade after a few seconds.

Visual information is held in 'iconic' memory. To understand iconic memory, look around you and then shut your eyes. For a short time, you retain a vivid image of everything you just saw. However, as you try to analyze the details, the image fades very rapidly, within 1/2 of a second.

If most of the information in sensory memory fades away and does not get processed further, what determines which information is selected for further processing? Considerable evidence from research suggests that it is the information that we pay attention to that moves on for further processing.

Short-Term Memory
The sensory information's next port of call is the Forebrain. It then becomes Short-Term Memory. The duration of Short-Term Memory depends upon the type of material being stored. Without active rehearsal, numbers are stored for 10 to 15 seconds, but colors are stored for up to 3 minutes.

When you are talking or answering questions on an exam, the information must be brought into Working Memory for you to manipulate, and your words and answers come out of Working Memory. As you try to understand the words on this page, your ability to understand these concepts depends on your Working Memory and also to some extent - when contextual information needs to be retrieved - from your Long-Term Memory.

Short-Term Memory is limited in terms of the both its capacity (amount of information it can hold) and its duration (length of time it can hold information). It is fluid and ever-changing - the focus of our consciousness.

Imagine that you are asked to remember a telephone number that is new to you. You could probably keep it in your memory for more than 30 seconds, but only by saying it over and over again 'in your head.' This is called 'rote rehearsal' or 'maintenance rehearsal.' This can help you to keep information in Short-Term Memory for more than 30 seconds but if anything happens to interrupt your rote rehearsal, the information will be lost, unless you have already succeeded through an effort of memorization to move the information into Medium or Long-Term Memory. You can probably think of real life examples where this has happened: you were trying to keep a telephone number in your head, but someone interrupted your thoughts and you lost the number forever. Or a number where you made an effort to remember it; in our course, we have many methods to help you do this easily.

People with an apparent Short-Term Memory deficit may in fact have a well functioning memory except that they are failing to filter useful information from bad. To use a computer analogy, their "spam filter" is not as good, therefore they fill up their working memory with background information, most of which is noise. For example, a person telling a joke may provide lots of contextual description; however those with good filters will be able to recognize and separate the actual storyline and save it in Short-Term Memory. They quickly discard the non-relevant bits of information and are left with the skeleton of the story story that fits within the available working memory; thus the relationship of the important parts of the story is clear - the person can "get the joke" and be able to repeat it to himself and then later his friend if required. A person who does not filter the story for its essentials may not remember the beginning of the joke when it's only half way told, and then certainly won't be able to get the point of the joke at its conclusion, nor be able to repeat it.

Latent inhibition is a preconscious filtering mechanism that allows people to ignore stimuli previously experienced as irrelevant. To repair inadequate filtering and to learn to inhibit the passing to working memory of unnecessary information (i.e. increasing latent inhibition) concentration exercises are effective, such as those included in the Mind Development Concentration Course.

Working Memory
Working Memory is the active system that temporarily stores and manipulates information that is needed in the execution of complex cognitive tasks, such as learning, reasoning and comprehension. Working Memory is all that is currently in view. There are two types of components: storage and central executive functions. The four storage systems within the model are the Articulatory Loop and Semantic Buffer (which are responsible for the temporary storage of verbal and semantic information respectively) and the Visuo-Spatial Sketch Pad and Episodic Buffer (which are responsible for the temporary storage of visual and sensory information). Our Short-Term Memory at any time contains the information with which we reason. Working Memory is that part of Short-Term Memory that contains the information we use for thinking: visually, verbally and numerically.

The main functions of Working Memory can be itemized as follows:

  1. The Articulatory Loop
  2. The Visuo-Spatial Sketch Pad
  3. The Central Executive
  4. Episodic & Semantic Buffers
There are many other subsidiary functions but these are the ones that should be readily accessible and whose function can be improved.

Not everybody has developed this range of functions to the limit of his ability. Circumstances may have resulted in your developing one faculty at the expense of the others. Development of all these functions, some of which reside in the Left and some in the Right hemispheres of the brain, means that we have to become specialized in these areas. Lack of specialized centers in the brain is called Bicamerality (two cameras), because the functions of the left and right hemispheres are the same and duplicate each other. This results in indecision and other problems of dual personality. Thus there is inefficiency, not only in executive action but in the duplication of information left and right. The primary objective of the Mind Development exercises at this level is to get rid of this Bicamerality, so that your will can be united and your brain becomes a more effective instrument.

1. The Articulatory Loop
The function of the Articulatory Loop is to rehearse speech sub-vocally, that is silently. This is necessary in order to maintain a memory trace as an electrical vibration in the Short-Term Memory. A process of sub-vocal rehearsal is used to refresh a fading memory trace before it decays into inaccessibility. The evidence for the existence of this loop is as follows:

A. Errors in speech often have an acoustic or phonemic similarity, for example F instead of S, B instead of G. It is harder to remember passages with similar speech sounds - the letter sequence of DBCTPG is harder to remember than KWYLRQ, test this for yourself.

B. The effect of articulatory suppression, that is, for example, forcing someone to repeat a common word like 'the' while they are trying to remember a series of digits. This has the effect of reducing the number of digits that can be recalled, i.e. the 'digit span'. These two factors mean that the length and accuracy of the digit span depends upon the ability to rehearse the material sub-vocally.

C. It is easier to remember short words than long ones. Short syllable words like 'bishop' or 'rivet' are easier to remember than words with long vowel sounds like 'harpoon' or 'Friday'. Slowly spoken words are not well remembered. It has been found that digit span levels for Welsh-speaking children are less than those for English-speaking children. Welsh schoolchildren are one year behind their English counterparts in mathematics when they are instructed in Welsh (Baddeley and Graham Hitch 1974). This is because the time required to read digits in Welsh is longer. The difference disappears when English numbers are used, and proves that English and Welsh speakers are intellectually equal. In contrast, Cantonese school children are better at mathematics than English schoolchildren, because they have a digit span of ten; the words for numbers in Cantonese are shorter than the number words in English, this gives Cantonese schoolchildren a numerical IQ advantage of about 20 points.

Thus one can say that the efficiency of the Articulatory Loop depends critically on the time taken to say the message, i.e. fast talkers remember better. In short, you can remember as much as you can say in 3 seconds, which is typically the length of the loop. (Our findings are that a child of 8 years speaks at 1.5 words per second and a child of 12 years speaks at 2.5 words per second. The child of 8 has a digit span of approx. 4 and the child of 12 has a digit span of about 7, so this indicates that 3 seconds is about the right figure.)

If I gave you a series of letters to remember, your ability to remember the series correctly would probably depend on the number of letters in the series, which you could 'fit into' the 3 seconds of the Articulatory Loop. Most people can remember a series of letters correctly if there are only 3 or 5 letters in the series. About half the people asked to remember a seven-letter series have difficulty. Relatively few people can remember a series consisting of 9 or 11 letters correctly. This finding, that the limit on Short-Term Memory is typically around 7 items, is one of the most consistent findings in all of psychology. George Miller, a very famous cognitive psychologist, coined the phrase “the magical number seven, plus or minus two,” to describe the capacity of Working Memory.

Working Memory is clearly essential if we are trying to understand a spoken sentence, or remember a string of digits. Working Memory is an active process where the goal is not necessarily to move the information into permanent storage, but to keep it available until it is put to use - think of a phone number you'll repeat to yourself until you can dial it on the phone.

We essentially 'think' out of our working memory, but the capacity of Working Memory is only seven (plus or minus two) items. If the limit on working memory was literally 7 letters, I could not write my own name, much less form a complete sentence. It is true that Short-Term memory has a capacity of 7 somethings, but the limit is not necessarily 7 letters. We can increase the working capacity of Short-Term Memory by combining bits of information into meaningful units, or chunks.

So the memory span of young adults is typically around seven elements, called chunks, regardless whether the elements are digits, letters, words, or other units. However that span does depend on the category of chunks used (e.g., span is around seven for digits, around six for letters, and around 5 for words), and even on features of the chunks within a category. For instance, span is lower for long than for short words. In general, memory span for verbal contents (digits, letters, words, etc.) strongly depends on the time it takes to speak the contents aloud, and on the lexical status of the contents (i.e., whether the contents are words known to the person or not).

A new piece of information may be connected by association with things that we already know or will readily be reminded of, or which are typically linked by logic or character. This connection then forms a single chunk in our symbol space. Many memorization techniques take advantage of this to effectively expand Working Memory.

Focusing on meaning therefore helps with chunking. Understanding the meaning of information to be learned involves understanding how new information relates to other new information or to information you already know.

Focusing on meaning not only helps with chunking, it also helps you to get information into and out of Long-Term Memory in the most efficient way. Repetition in itself is not a very efficient strategy for moving information from Short-Term to Long-Term Memory, but if we focus on the meaning of information to be learned and try to relate it to information that is already in our Long-Term Memory, this seems to help the new information to be effectively filed away. Since Long-Term Memory appears to 'file' information according to meaning, this will also help you to be able to find that information when you need quick access to it in the future.

The Articulatory Loop also appears to be a useful checking mechanism which is good at preserving the order in which information is to be given out. This is particularly noticeable when one is asked to suppress sub-vocalization by saying a word repeatedly under your breath. If two words of a given passage were then reversed, one would be less likely to notice whilst suppressing sub-vocalization. Thus one would use sub-vocalization when reading difficult prose, e.g. a legal document where meticulously accurate comprehension is necessary. But one might not sub-vocalize very much when reading a novel, since the concepts are easy to understand. What one does experience while reading romantic fiction, however, is a mental image of the sound of the characters speaking, as well as a visualization of their surroundings. These 'voices' are not the function of the Articulatory Loop, but of another system which produces auditory and visual imagery...

2. The Visuo-Spatial Sketch Pad
Visuo-Spatial perception refers to our ability to process and interpret visual information about where objects are in space. This is an important aspect of cognitive functioning because it is responsible for a wide range of activities of daily living. For instance, it underlies our ability to move around in an environment and orient ourselves appropriately. Visuo-Spatial perception is also involved in our ability to accurately reach for objects in our visual field and our ability to shift our gaze to different points in space.

The Visuo-Spatial Sketch Pad is used in the temporary storage and manipulation of spatial and visual information, such as remembering shapes and colors, or the location or speed of objects in space. It is also involved in tasks which involve planning of spatial movements, like planning one's way through a complex building.

To achieve this functionality, the association areas of the visual cortex in humans are separated into two major component pathways, which are believed to mediate the different aspects of visual cognition...

  1. The inferotemporal region of the brain is believed to mediate our ability to process visual information about the shape, color and texture of objects. This may be called the "visual cache" and can be used, for example, for constructing and manipulating visual images, and for the representation of mental maps.

  2. The parieto-occipital region is believed to process Visuo-Spatial and Visuo-Motor types of information that deal with location, i.e. spatial and movement (kinaesthetic) information. It may be called the "inner scribe." It also rehearses information in the visual cache and transfers information to the central executive.
Try multiplying 8 x 13 in your head. You will find that to perform this calculation mentally you will need to 'carry' digits which have to be held over in the memory. To do this you may imagine a blackboard on which you chalk the figures; this then we can define as the Visuo-Spatial Sketch Pad. It is a specialized visualization function that involves both left and right hemispheres working in combination.

Three main findings provide evidence for the distinction between visual and spatial parts of the Visuo-Spatial Sketch Pad:

  1. There is less interference between visual and spatial tasks than between two visual tasks or two spatial tasks.

  2. Brain damage can influence one of the components without influencing the other.

  3. Results from brain-imaging show that working memory tasks with visual objects activate mostly areas in the left hemisphere, whereas tasks with spatial information activate more areas in the right hemisphere.
In addition to the visualization faculty there appears to be the ability to hold spatial recording in memory. Blind people are able to find their way about, bats can navigate in the dark by the reflections of the high-frequency sounds they emit, one can learn to touch type without seeing the keyboard.

Words which can be visualized are more easily memorized than those which cannot. It is therefore possible to use visual imagery as a mnemonic system for the purpose of remembering random sequences of numbers and words.

There is evidence that images are stored directly in the brain, which means that information can be accessed rapidly if it is linked to an image, or ideogram, and if perceived material is reduced to these ideograms. This probably does not apply to long-term memory; for reasons of data space the images are converted into an abstract code as they would be in a computer's memory. In order to display and operate upon various images stored in Long-Term Memory we need to have a recall system functioning through codes or keywords. Once the image is recalled to Short-Term Memory it can then run as a program would in a computer, creating the image in the mind's eye and operate upon it.

The ability to visualize is reduced if the subject is asked to point to something while seeing a picture, but heavy verbal tasks, such as grammatical analysis of sentences, do not interfere with the image. If you have ever tried to form a mental picture while driving a car you will realize how dangerous this can be. Nevertheless it is possible to hold conversations with other passengers or listen to the radio without danger. Visual imagery can also be affected by heavy demands upon one's mathematical and logical faculties. Being called upon to make a difficult decision seems to affect the area of executive control necessary to maintain the image.

Spatial memory is an essential part of our ability to function as an individual within our environment. For instance, if I were to walk into a darkened room that was very familiar to me, I would have a good idea where all the individual items in the room were so that I didn't walk into them. In order for me to be able to do this, I need to be able to remember what is in the room and where they all are in relation to my entry point, the door. In short, I need an internal map of the room. The map consists of three elements; what items are present, where they are in relation to each other and where I am, in relation to both those items and the room in general.

The generation and operation of such maps require the integration of multiple neural systems. In order to recognize and recall what is present in the room, two neural systems come into operation. A recognition system based in the Perirhinal Cortex tells me that the items present are familiar to me. A second system that deals with the arrangements of individual items based in the Hippocampus tells me how these items are arranged in relation to each other.

3. The Central Executive
When a person needs to simultaneously perform mental tasks such holding and manipulating data and images, and reasoning with that information, Working Memory must allocate available cognitive resources. Working Memory is like the control tower of a major airport, responsible for scheduling and coordinating all incoming and outgoing flights. To achieve this, there are the two 'slave systems' described above, the Articulatory Loop and the Visuo-Spatial Sketch Pad responsible for short-term maintenance of information, and a Central Executive responsible for the supervision of integrating the information and for coordinating the slave systems.

The Central Executive is, among other things, responsible for directing attention to relevant information, suppressing irrelevant information and inappropriate actions, and for coordinating cognitive processes when more than one task must be done at the same time.

Working Memory operations are both conscious and unconscious. Working Memory theories and research have focussed mainly on reportable, conscious functioning. However, we must acknowledge that a myriad of unconscious specialized operations carry out detailed Working Memory functions (Baars & Franklin 2003). Unconscious Working Memory is not subject to the capacity limitations of Conscious Working Memory, i.e. about three to six chunks. Anderson (1983) found that Unconscious Working memory can sometimes contain over 20 units at one time. To reconcile such a large capacity of Working Memory with the much smaller capacity of Short-Term Memory, Anderson argued as follows: The activation of elements decays very rapidly. For this reason the number of units that can be actively maintained long enough to be included in immediate recall is much less than all of the information activated at the start of recall.

Working Memory functions are able to operate below the level of consciousness because they have become automated. Unconscious automated processing is crucial to successful Working Memory performance, because it is believed that automated processing, such as Intuition, does not draw on the measurable capacity of Working Memory. Automated processes operating below the level of awareness tend to be readily accessible, being called into consciousness whenever effortful processing is required. Operations that were once conscious, but became unconscious as their function became automated are the most accessible. When Working Memory operations are brought into awareness, executive processes within and outside of Working Memory are usually called into play simultaneously.

4. Episodic & Semantic Buffers
Baddeley (2002) has postulated a fourth component in the Working Memory, in addition to the Articulatory Loop, the Sketch Pad and the Central Executive. There is a temporary store for information, with a limited capacity, from which material can be recalled with 90 to 100% accuracy, that Baddeley calls the Episodic Buffer. The Episodic Buffer holds representations that integrate episodic information (concrete, experiential data) into a unitary episodic representation. This information would include phonological, visual, and spatial information and data retrieved from Long-Term Memory.

The Episodic Buffer, then, is dedicated to linking information across domains to form integrated units of visual, spatial and verbal information with time sequencing (or chronological ordering), such as the memory of a story or a movie scene. The Episodic Buffer is also assumed to have links to Long-Term Memory and semantical meaning. The main motivation for introducing this component was the observation that some (in particular, highly intelligent) patients with amnesia, who presumably have no ability to encode new information in Long-Term Memory, nevertheless have good short term recall of stories, recalling much more information than could be held in the Articulatory Loop. A small number of densely amnesic patients could perform at a normal level on immediate recall of a prose passage, containing some 20 or more idea units, and hence considerably beyond verbal or spatial span. Such a passage would be approximately 100 words.

Recent research has found that some densely amnesic individuals could remember considerable information (up to 25 idea units) for up to an hour in the absence of any interference (in a quiet, dark room), even on trials in which they slept during the retention interval. This suggests that they must have used some sort of storage mechanism outside of the focus of attention that does not depend on Long-Term Memory, at least not as we normally describe it. This would be the function of an Episodic Buffer.

As yet, the storage location of the Episodic Buffer is not certain, but it is surmised that it may be the Frontal Lobes, as this would explain a lot of phenomena. In my opinion, the Episodic Buffer acts as a bridge between Short-Term Working Memory and the locations of Medium-Term Memory and Intermediate-Term Working Memory (readily accessible expert information). Medium-Term Memory is, for the most part, episodic and temporary, though some of its contents may be stored in Long-Term memory if they need to be. Intermediate-Term Working Memory is a temporary store of expert knowledge retrieved from Long-Term Memory for the task at hand. As Intermediate-Term Working Memory, in the case of experts, is nearly as fast and accurate as Short-Term Working Memory, and has much more in common with Medium-Term Memory than Long-Term Memory, I prefer to use the expression Intermediate-Term Working Memory rather than Medium-Term Working Memory or Long-Term Working Memory. In my opinion, Intermediate-Term Working Memory is a very much expanded Episodic Buffer that operates in a specific domain.

An Intermediate-Term Working Memory lasting for minutes to hours has recently been demonstrated, and even longer time intervals are expected to be documented in the future (see Dr. Merlin Donald's book, A Mind So Rare - 2001). Donald maintains that too much of the focus in consciousness research has been in exploring phenomena associated with Short-Term Memory. He maintains that Intermediate-Term Working Memory is the key to understanding how meta-awareness is maintained over the long term. As species become more complex, brains develop Short-Term memory, and later, Intermediate (or Medium-Term) control. The key to consciousness in humans is the huge growth in the tertiary or association areas of the cortex.

Semantic Buffer
Further research is supporting the existence of a Semantic Buffer, filling a similar role to the Episodic Buffer but for semantic material. The Semantic Buffer is a Working Memory center that stores meaning(s), rather then the surface features of words and phrases. In the Kintsch and van Dijk (1978) model and its successors, a limited-capacity, Short-Term Memory Buffer plays an important role in comprehension. By maintaining information in Working Memory, the Short-Term Memory Buffer facilitates the construction of a coherent textbase. A small number of propositions are selected at the end of each processing cycle and carried over in a buffer to be reprocessed with the input propositions from the next processing cycle. In this way, the theme or topic of an entire discourse may be held in mind, and one may paraphrase or translate to an alternative language. In short, there is a storage buffer for the lexical context of discourse. Current experimental work has suggested that such global structures of meaning play an important part in the cognitive processing of discourse, e.g. in comprehension and recall.

The Central Executive is very active in Working Memory and is responsible for the selection, initiation and termination of processing routines (e.g. encoding, storing, and retrieving). Encoding refers to the processes of placing items into memory. Retrieval refers to the processes through which we recover items from memory.

The Structure of Working Memory
Right
brain
  Structure Left
brain

The Central Executive is a limited capacity system used for a variety of purposes, including tasks that involve planning or decision making; trouble shooting in situations in which the automatic processes appear to be running into difficulty; novel situations; dangerous or technically difficult situations; and situations where strong habitual responses or temptations are involved.

The Frontal Lobes are probably the location of the Central Executive of Working Memory. Extensive damage to the frontal lobes such as lobotomy would result in severe impairment in Central Executive functioning. The classic frontal syndrome is characterized by “disturbed attention, increased distractibility, a difficultly in grasping the whole of a complicated state of affairs... well able to work along old routines but cannot learn to master new types of task, in new situations.”

Intelligence
An important function of the Working Memory is the ability to store and manipulate numbers. We can recognize similarities, like an 'And' function, we can add, subtract, multiply and divide. Our arithmetical digit span seems to depend upon speed - the faster one can operate with the information stored in Working Memory, the greater the span and accuracy.

Until recently, a person's IQ - a measure of all kinds of mental problem-solving abilities, including spatial skills, memory and verbal reasoning - was thought to be a fixed commodity largely determined by genetics. But recent modern research demonstrates that the very basic brain function called Working Memory in fact underlies our general intelligence, in particular our Fluid Intelligence, opening up the intriguing possibility that if you improve your Working Memory, you would boost your IQ too. This is what Mind Development has found to be the case, over many years experience.

Reasoning ability is central to intelligence. Fluid Intelligence relates to our ability to solve novel problems. Although many different cognitive processes may be executed in the solution of a task, individual differences in Working Memory efficiency, skills and resources play a crucial role in determining the speed and correctness of the results obtained.

Measures of Working Memory capacity are strongly related to performance in other complex cognitive tasks such as reading comprehension and problem solving skills, and with measures of the intelligence quotient. Some researchers have argued that working memory capacity reflects the efficiency of executive functions, most notably the ability to maintain a few task-relevant representations in the face of distracting irrelevant information. The tasks seem to reflect individual differences in ability to focus and maintain attention, particularly when other events are serving to capture attention. These effects seem to be a function of frontal brain areas.

Working Memory is the brain's short-term information storage system. It's a workbench for solving mental problems. For example if you calculate 71 - 7 + 6, your Working Memory will store the intermediate steps necessary to work out the answer. And the amount of information that the Working Memory can hold is strongly related to general intelligence. Of course, writing things down is an important way of extending Working Memory, and this enables organization of information and calculations to be accomplished that would not be possible for most people with information held within the mental space alone. However using notes to extend Working Memory does not exercise and improve the mental ability to hold and juggle a large number of items in the mind, with instant random access to each, as is needed for maximum fluid intelligence.

Recent studies suggest that Working Memory can be improved by training and that a period of such training increases a range of cognitive abilities and increases IQ test score, and indeed that working memory underlies general intelligence. It has been shown that, after training, measured brain activity related to working memory increased in the prefrontal cortex, an area that many researchers have associated with working memory functions. In 2009, it was reported in Science that working memory training led to measurable density changes for cortical dopamine neuroreceptors in test persons. Perhaps of greater importance, another study has found a period of Working Memory training increases a range of cognitive abilities and increases IQ test scores. Working Memory training - as accomplished, for example, by the memorization techniques included in our Memory Course - is therefore key to unlocking brain power.

5. Global Workspace Theory
Recent studies have led to the proposal that Working Memory operates not as a gateway between sensory input and Long-Term Memory but as a Global Workspace. Access to acquired knowledge and prior learning in Medium and Long-Term Memory occurs and is integrated in the Episodic Buffer with current sensory input before that complex of information becomes available to Working Memory.

The Episodic Buffer is assumed to be attentionally controlled by the Central Executive and to be accessible to conscious awareness. Baddeley regards the Episodic Buffer as a crucial feature of the capacity of Working Memory to act as a Global Workspace that is accessed by conscious awareness. According to this model, when Working Memory requires information from long-term storage, it may be downloaded into the Episodic Buffer, rather than simply activated within Long-Term Memory.

Global Workspace Theory was proposed by Bernard Baars in 1983. The function resembles the concept of Working Memory, especially the Episodic Buffer as described by Baddeley. The Global Workspace corresponds to subjective experience in the context of Working Memory, the inner domain in which we carry on the narrative of our lives, including our inner speech and visual imagery. Ideas and symbols remain in the Global Workspace for as long as the individual is working with them.

The easiest way to think about the Global Workspace is in terms of a the metaphor, "theater of consciousness." The entire stage of the theatre corresponds to Working Memory, in which we talk to ourselves, visualize places and people, and plan actions. The stage is illuminated by a spotlight of selective attention, revealing the contents of consciousness - actors moving in and out, making speeches or interacting with each other. The bright spot is further surrounded by a periphery of significant but vaguely conscious events. The audience remains in the dark, watching the play. Behind the scenes, also in the dark, are the director (the Central Executive processes), stage hands, script writers, scene designers and the like (corresponding to the Visuo-Spatial Sketch Pad, Phonological Loop and Episodic and Semantic Buffers) that shape the visible activities in the spotlight, but are themselves invisible.

The illuminated stage of of the Global Workspace corresponds to what we are conscious of, and this information is broadcast to a multitude of unconscious cognitive brain processes, which may be called receiving processes. Other unconscious processes, operating in parallel with limited communication between them, can form coalitions which act as input processes to the Global Workspace. Since globally broadcast messages can evoke actions in receiving processes throughout the brain, the Global Workspace may be used to exercise executive control to perform the voluntary actions of conscious experience.

The continuity of the "stream of consciousness" may in fact be illusory, just as the continuity of a movie is illusory. Nevertheless, the seriality of mutually incompatible conscious events is well supported by objective research over some two centuries of experimental work. A simple illustration would be to try to be conscious of two interpretations of an ambiguous figure or word at the same time. When timing is precisely controlled, as in the case of the audio and video tracks of the same movie, seriality appears to be compulsory for potentially conscious events presented within the same 100 msec. interval. The 100 ms time domain corresponds closely with the known brain physiology of consciousness, including brain rhythms in the alpha-theta-gamma domain, and event-related potentials in the 200-300 ms domain.

Consciousness
The Global Workspace allows for cooperative problem-solving by large collections of specialized programs; therefore the brain can be perceived as a "society of mind." Our knowledge of the cortex, in particular, is consistent with the hypothesis that much of the brain consists of highly specialized regions. Detailed cortical processing of visual information, for example, is largely unconscious, but the outcome of visual processes is a conscious experience of objects and scenes. As predicted by Global Workspace Theory, there is evidence that visual contents evoke highly distributed activity in non-visual regions of the brain. In Global Workspace Theory, this is called "broadcasting" of global messages to multiple target functions.

The idea that consciousness has an integrative function has a long history. The brain may be viewed as a massive parallel distributed system of highly specialized processors. In such a system coordination and control may take place by way of a central information exchange, allowing some specialized processors - such as sensory systems in the brain - to distribute information to the system as a whole. A sizable body of evidence suggests that consciousness is the primary agent of such a global access function and there is evidence to suggest that all the separate strands of consciousness in the Global Workspace come together in the thalamus. It is posited that this global control is effected via cortical 'gating' of a strategic thalamic nucleus. Gamma wave synchrony controls access to the content of consciousness. Synchronization of remote areas implicates the thalamus, and the thalamus represents a hub from which any site in the cerebral cortex can communicate with any other such site or sites.

thalamus

The thalamus is located deep within the cerebrum. It is an egg-shaped structure lying at the very top of the brain stem, above the hypothalamus. Almost all of the messages that are received by the cerebral cortex from the environment or from the body's internal receptors come through the thalamus and much current thought about perceptual processing is based on sensory pathways that relay in the thalamus. Recent research also suggests that the thalamus regulates the electrical rhythms that parts of the brain use to communicate with each other. It has been speculated that tips of the tongue experiences (when only part of a memory is recalled) may occur when the rhythms don't synchronize with the regions properly - which would put these memory failures at the door of the thalamus.

The thalamus is much like a brain within the brain... it is approximately brain shaped and has interconnecting tissue between its two lobes just like the corpus callosum. It shares the most central, most protected part of the brain with the hypothalamus and the third ventricle. The thalamus is tucked under the corpus callosum and it is cushioned by the third ventricle of the brain, where it sits on either side. The thalamus connects the optic, acoustic and bodily sensors with the appropriate areas of the cortex and serves as a central relay station for all the main sensory systems except for the olfactory system. Additionally it connects to the cerebellum and the hypothalamus, next to it. See The Triune Brain for more information about the "brains within a brain" of the human being.

It is postulated that in the thalamus, consciousness (the Ego) experiences the results of brain processes (not the processes themselves) and is able to control attention, thought and motor processes. Without input from the thalamus, consciousness is not maintained, but the final seat of consciousness may lie in the limbic system - not the frontal lobes, as people who have had a lobotomy still have some sort of Ego, and people with massive damage to the cortex do not lose consciousness. Currently, there is much debate as to which part of the brain is the center of consciousness. It is suggested the hypothalamus, particularly the posterior hypothalamus, which plays a critical role in the maintenance of consciousness or wakefulness, is the center of consciousness.

The thalamus has massive connections with the frontal lobes, the limbic system and the activating reticular formation of the brainstem. Recent theories have posited that the hippocampus and thalamus serve distinct, yet related, roles in episodic memory. Whereas the hippocampus has been implicated in Long-Term Memory encoding and storage, the thalamus, as a whole, has been implicated in the selection of items for subsequent encoding and the use of retrieval strategies. Scans reveal that introverts have more activity in the frontal lobes of the brain and anterior, or front, thalamus. Damage to the dorsomedial thalamus can impair memory formation, but the thalamus is not, however, a memory store; the thalamus is the Screen of Consciousness.

A higher-level consciousness (awareness of being aware), probably unique to humans, is possible if the brain is also capable of abstracting the relationship between the self and the non-self, and this can only happen through social interaction. This leads naturally to the development of linguistic faculties. Garald Edelman (in The Remembered Present, 1989) identifies the regions that are assigned to define self within a species (the amygdala, the hippocampus, the limbic system, the hypothalamus) and those that operate to define the non-self (the cortex, the thalamus and the cerebellum).

Note that, according to Edelman, concept-formation preceded language. Language was enabled by anatomical changes. What changed with the advent of language is that concepts became independent of time, i.e. permanent. And semantics preceded syntax: acquiring phonological capacities provided the means for linking the preexisting conceptual operations with the emerging lexical operations.

Piero Scaruffi, in his book The Nature of Consciousness, summarized our human makeup as follows: "Consciousness is the awareness of existing. Self is the awareness of lasting in space and time (of being an "I"). Sensations are bodily feelings such as pain, red, warmth. Emotions are non-bodily feelings such as anger, happiness, fear. Cognition encompasses the processes of reasoning, memory, learning, speaking, etc. Perception is the physical process of perceiving the world. Thought is the act of being conscious over an extended period of time."

Consciousness is actually the only thing we can be sure of. We are sure that "we" exist, and "we" doesn't mean our bodies, it means our consciousness. Everything else could be an illusion, but consciousness is what allows us to even think that everything else could be an illusion. It is the one thing that we cannot reject.

In Edelman's picture, human consciousness includes a liberation from the present moment. Animals tend to live in the present, simply reacting to stimuli. Only conscious animals can think about the past and about the future. Indeed humans have tended to take this too far and rarely remain mindful in the present. We need our past experience for the learning it offers; we don't need it to cling on to regrets or blame, that changes nothing. We need our future to make plans; we don't need it to worry about what might happen, that changes nothing. Like an artist creating a painting, we put our life together in the present moment, in the context of the wisdom that obtains from being also conscious of the past and future.

Mnemonics
The Ancient Greeks first outlined the practice of memory and their simple rules are still used today in modern memory systems; however they used a flat model of memory, such as associating the items to be remembered with locations on a route. In contrast, advanced systems, such as the Memory Cube (which is taught in our forthcoming Memory Course) is architecturally hierarchical, with nested Thought Maps and Chains. This is truly a "Thought Engine*," with fast access to up to 15,000 items - think Mega Memory! Such a thought engine can can extend the Global Workspace, because it is possible to overcome Short-Term Working Memory Limitations by setting up a Virtual Short-Term Memory with almost unlimited capacity. By using Fenaigle chains, Virtual Short-Term Memory can be increased to several hundred digits. The World record using this method is 400 digits in something like five minutes. With the exception of special applications like the Pixel method, which requires several hundred digits, a Virtual Short-Term Memory of 20 to 30 items is sufficient for most applications. In effect one has greatly expanded the stage upon the theater of consciousness. This has implications for consciousness. See the sections on Intermediate-Term Working Memory and Global Workspace Theory.

*Note: A Thought Engine enables users to input, store, search, navigate and output concepts (and their semantic relationships) within the Mnemonic System; as well as link to information stored elsewhere within memory or databases.

Mnemonics have a general application; all spheres of knowledge may be temporarily added to the Working Memory, creating an artificial Intermediate-Term Working Memory - for which we use the computing term "High Memory" because of its fast random access - that is available for any subject, not just for the domain-specific expert's use. I have chosen to reserve the term "High Memory" for the Artificial Intermediate-Term Memory created through the use of Mnemonics, whereas I reserve the term Intermediate-Term Working Memory for Natural Memory. For example, a person who is an expert on football would have developed an Intermediate-Term Working Memory for football facts, but his memory for cricket would probably be not much more than average. One solution is for him to develop a superordinate Intermediate-Term Working Memory for sport in which case he would have an expert memory for both football and cricket. Most students will eventually develop an Intermediate-Term Working Memory for several domains, but a Mnemonic System is the superordinate system par excellence, as this can cover most topics.

Intermediate-Term Working Memory
In unskilled people the Working Memory is a function of Short-Term Memory and as such it is domain general. Long-Term Memory may also impact on Working Memory in a domain-specific manner. It is generally observed that memory performance increases after an individual practices on memory tasks involving specific types of materials and that an individual's familiarity with a given type of material is related to the amount of material recalled. With more familiarity and experience with a particular type of stimulus material, subjects acquire over time a set of complex patterns in Long-Term Memory that allows them to represent subsequently presented information in terms of already acquired patterns (chunks) of elements rather than individual stimulus elements.

In the case of experts such as mathematicians, chess masters, physicists and people of that ilk, part of the Long-Term Memory has been recruited to create a virtual center that we call Intermediate-Term Working Memory (sometimes referred to as Long-Term Working Memory). By and large, this type of working memory is domain specific, in short it has a limited range of transfer. In contrast to the Short-Term Working Memory we cannot view the content all at once; however, if the material is repeatedly recalled, it can be stored indefinitely in Long-Term Memory. We have an attention window that permits us to perceive 6-10 items or chunks at the same time. The Intermediate-Term Working Memory is like a spreadsheet and the window of attention is a parasitic function of the Short-Term Working Memory.

Intermediate-Term Working Memory is a faculty enjoyed by experts in many fields. Recent research on memory performance shows that with practice and the acquisition of memory skills, subjects can improve their recall performance on a specific memory task with a particular type of stimulus material by 100% to 1,000%. In short, the capacity of Intermediate-Term Working Memory is up to ten times that of Short-Term Working Memory. Conscious retrieval from Long-Term Memory typically takes from 1 to 2 seconds, but an expert's direct retrieval takes only 400 milliseconds, because the expert is able to bypass Long-Term Memory limitations.

Intermediate-Term Working Memory vastly increases the expert's symbol field. It comes into being because the Forebrain parasites on the Parietal Lobes, thus the capacity of Short-Term Memory is enhanced. This is so, because Intermediate-Term Working Memory uses a relevant, fast access area of Long-Term Memory. The capacity of Intermediate-Term Working Memory may be vast, but it is domain specific. There is little transfer to related skills, such as typing and playing the piano; or from playing chess to playing dominoes.

An expert in a given domain of activity, such as medicine, chess, music or golf, is “one, who has acquired special skill in or knowledge about a particular subject through professional training and practical experience” (Webster's, 1976, p. 800). Experts will therefore, by definition, have a greater body of knowledge about their domain of expertise than other individuals.

More remarkable is the expert's accurate memory for new experiences in his or her domain. An elite athlete can, after a sports event, discuss the play-by-play action. Expert chess players can readily recall details of chess positions from their matches in recent tournaments. Early in the twentieth century it was believed that experts were innately talented with a superior ability to store information in memory. Numerous anecdotes were collected as evidence of an unusual ability to store presented information rapidly. For example, Mozart was supposed to be able to reproduce a presented piece of music after hearing it a single time. However, more recent research has rejected the hypothesis of a generally superior memory in experts and has demonstrated that their superior memory is limited to their domains of expertise and can be viewed as the result of acquired skills and knowledge relevant to each specific domain. Although, experienced chess players are better at remembering the positions of the balls on a pool table than novices, so there is a small measure of transfer to related tasks.

The key seems to be the use of mnemonic devices and other methods of imposing some sort of order or meaning on the information involved, such as chunking or grouping into patterns and hierarchies. To illustrate, a chess master can usually recall the positions of all the pieces on a chessboard after a quick glance. But if the chessmen are arranged randomly and meaninglessly, his memory is reduced to near-normal. The gist is that long practice and the application of mnemonic devices can vastly improve anyone's memory and, in consequence, memory prodigies are not really so anomalous.

The Peg System of mnemonics, for example, has much in common with the Intermediate-Term Working Memory system. Material encoded on Pegs is invulnerable to distraction, it may be stored and retrieved, after several hours, with random access and 100% accuracy. A Peg System opens up the possibility of reflecting on a hundred or more ideas, rather than the three to five that you can reflect on in Short-Term Working Memory. When Pegs are combined with Short Chains, to make a network of data encoded into mnemonics, their potential for advanced thinking is almost limitless.

Note: Short-Term Working Memory can import ideas (data), three to five at a time, from the Peg System and, in turn, the Short-Term Working Memory can export ideas (data), one by one, to the Peg System.

Intermediate-Term Working Memory is developed as a virtual center by mnemonics experts. Although Intermediate-Term Working Memory is domain specific - a memory expert only has a Intermediate-Term Working Memory for mnemonics - the application of mnemonics is more or less domain-general, so a mnemonics expert gets the best of both worlds.

Intuition is not based on Intermediate-Term Working Memory. With Intuition, you are tapping some sub-cognitive area of the brain, possibly the cerebellum, in which case, you do not know that you know until the answer to your query becomes visible: then you know.

In the case of Intermediate-Term Working Memory, you know already that you know. You know that you have certain items in store, and that you could retrieve them if you wished, and you also know that certain items are not in store, so you can't, in which case, you know you don't know. This is Metacognition. The only content you can actually see is the part that has been imported into Short-Term Memory. At a maximum, only about nine items of data or perhaps two or three patterns can be imported into Short-Term Memory from Long-Term Memory at any one time. Not only do you know what is being stored in Intermediate-Term Working Memory, at any given time, you know, also, where to find any particular item, so you can import any item or collection of items into Short-Term Memory, within Short-Term Memory capacity limitations. Once these items have been imported into Short-Term Memory - but not before - they become visible.

To draw an analogy, Intermediate-Term Working Memory is like a spread-sheet, only part of which is visible at any one time. It is like looking at the spread-sheet through the center of a toilet roll. If you wish to see another area, you have to shift the toilet roll, then you cannot see the area that you were looking at before, but you know it is there and you can return to it any time you want. This is the introspective experience.

In my opinion, little or nothing can be transferred to Long-Term Memory in the waking state, but material from Long-Term Memory can be retrieved in the waking state, and recalled rapidly when the subject has an expert memory in a specific domain of knowledge; so the storage site of Intermediate-Term Working Memory cannot be a department of Long-Term Memory, if the content is to be recalled after a short period: it must be stored in Medium-Term Memory.

Intermediate-Term Working Memory can only store information for a day or two, so it has much more in common with Medium-Term Memory than Long-Term Memory proper; like Medium-Term Memory it functions through the the Frontal Lobes, the Hippocampus and the Parietal Lobes. The Medium-Term Memory acts as a mixing pot: material from Long-Term Memory is contrasted and compared with information from Short-Term Memory.

Contrary to some popular views, Intermediate-Term Working Memory is not infinite in its capacity. Although it has a greater storage capacity than Short-Term Working Memory, it is very rigid and the speed of access is only about half of that of Short-Term Memory. There are also certain things that Intermediate-Term Working Memory cannot do, that Short-Term Working Memory can. Content can only be manipulated once it has been transferred from Intermediate-Term Working Memory to Short. From my introspective experience, the Intermediate-Term Working Memory stores patterns, rather than words and images. Words and images are constructed after the fact.

What one starts with is a sense of knowing. If what it has to show you cannot be put into some meaningful pattern, it will not be available for recall. Because the capacity of Intermediate-Term Working Memory is tied to the recall of accepted and meaningful patterns, it does not much amplify the capacity for creative thinking. On the other hand, Short-Term Working Memory is not subject to these limitations, so anything that can extend the capacity of Short-Term Memory must amplify the capacity both for reasoning and for creative thought.

How long does it take to gain an Intermediate-Term Working Memory? This depends on the domain in question. In the case of a very narrow domain, such as understanding and remembering the weather forecast, this may only take a few weeks. Most intelligent people can already do this, so they already have an Intermediate-Term Working Memory for weather forecasts. But in the case of broad domains, such as becoming an architect, a bishop or a research chemist - or indeed, the meteorologist who prepares weather forecasts - it could take five to ten years. It all depends on the breadth of the domain and the frequency with which Long-Term Memory is accessed in that domain, integrated and used constructively, alongside Short-Term Memory in the Working Memory.

I-tags
Within the mind, visual experiences can be distinguished from auditory experiences and from experiences of other sensory modalities. In turn, perceptual experiences can be distinguished from memories, emotions, thoughts, etc. This universal ability implies that certain features of data coming into the field of consciousness can be consciously noted and marked or 'tagged' as such.

Tags are thus seen to form the basis both of phenomenal consciousness and the self. Recently the concept of tagging has been recognized as a fundamental aspect of the creation of memories. Brian Lancaster (in 'Mind, Brain and Human Potential') explains the continuity of the self by means of 'I-tags.' These are the basic data from which our sense of identity is constructed. He proposed that when an event is consciously experienced, its representation in memory includes a reference to the 'I' which actually experienced the event. There is a personal connection and identification with the experienced perception. Subsequent voluntary recall consists of making a connection to the appropriate I-tag. When an I-tag exists there is no effort required to 'find the memory.'

At any given moment, a number of such I-tags are activated, as sensory systems interact with memory. We may intentionally evoke an I-tag or similarity of circumstances may remind us of stored I-tags. Thus I may be using a mouse which triggers one I-tag, sitting at my desk, which includes memories of its purchase, another I-tag, listening to favorite music, another I-tag, and so on. Each I-tag embodies my past identity state when the given entity was experienced previously. The 'I' is continually constructed from an endless flux of I-tags.

Each individual I-tag is the basis of meaning since it embodies the individual's involvement with some specific object or event in the past. It is this 'involvement' - of making an impression with the current sense of self - which makes the difference between experience passing by as a stream, and it's incorporation into our memory storage.

Unless a memory is I-tagged when it is transferred from Short-Term to Medium-Term Memory it cannot usually be recalled and it is filed directly into Trash files. Recall will then only occur in the presence of an appropriate environmental trigger (souvenir), or through retrieval by hypnosis. In short, we have to have an I-tag to retrieve something. Mnemonics provide such a tag.

6. External Working Memory
External Working Memory is an important further dimension, relating to practical representation of concepts and information with representative objects or graphical images. Examples of this method include demo kits, mind maps, the Roman Rooms loci system, the use of blackboards in a physicist's lab, and so on. These external representations are utilized as mnemonics, and this occurs naturally, to some extent, with those who are using their minds. However, conscious and extensive use of such methods can increase Natural Working Memory dramatically and raise IQ (which is highly, positively correlated with Working Memory) by 20 to 30 points, as well as enhancing lateral thinking and hence creativity. External Working Memory may be regarded as an extension of Natural Working Memory that preserves the capacity of random access in a way that Long-Term Working Memory does not. This is especially the case in the dimensions of certainty, flexibility and the essential speed.

Until about three or four hundred years ago, actors would memorize the theater they were going to act in and theaters were designed with this goal in mind. This gave an actor presence and provided him with a system of Loci. In the Middle Ages, this Memory Theater method was used as an External Working Memory System for the purposes of creativity, bringing the Renaissance more rapidly into being. A large Working Memory with random access is a primary requirement in the area of creativity.

The addition of External Working Memory to Internal Working Memory, under the direction of a Central Executive, results in a framework of Distributed Working Memory...

Distributed Working Memory

Medium-Term Memory
There are three major types of memory storage filing systems: Short-Term Memory (also called Working Memory or Immediate Memory); Medium-Term Memory (also called Intermediate Memory); and Long-Term Memory (also called Permanent Memory). Successful memory retrieval relies on a combination of all three memory storage systems at any given time.

Inability to recall information that is stored in the memory is called amnesia. When the Short-Term Memory is affected the person will have difficulty recalling the events that occurred in the preceding few seconds. Medium-Term Memory is affected when a person cannot recall events that happened from within a few seconds to a few days prior to the cause of the amnesia. With Long-Term Memory loss a person will be unable to recall events that occurred further back in time.

Medium-Term Memory covers the time span between a few seconds in the past and extending backward for 24 to 48 hours. This information is available in a "medium term" memory store that involves the Long-Term Memory system, but does not necessarily require consolidation into Permanent Memory, if it is not sufficiently salient or fails to be repeatedly recalled. It is much easier to recall what we had for dinner yesterday and the day before, but the memory of what we ate for dinner 3 weeks ago is gone.

Often when lay people speak of Short-Term Memory, they are referring to Medium-Term Memory (Intermediate Memory). Medium-Term Memory occurs once the information has been processed. It can be viewed as the part of memory which holds and mixes information from the different parts of the memory architecture. This will determine how we feel and what we will do about a given situation. It defines our ability to express actions.

Medium-Term Memory is used to transfer information from Short-Term Memory to Long-Term Memory. Material that has been I-tagged according to its type of information (semantic or episodic - data and concepts or sounds, sights, tastes, smells, etc.) is transferred to the hippocampus where the consolidation of the material in preparation for sending to Long-Term Memory takes place. The brain is generally thought to do all of this during sleep, specifically slow-wave sleep, when the brain is not busy with processing real-time inputs. The hippocampus is the storage site of Medium-Term Memory, which is stored for 24 to 48 hours. Medium-Term Memory is a separate system, intermediating between Short-Term Memory and Long-Term Memory. Short-Term storage is electrical, Medium-Term storage is chemical and Long-Term storage is protein based.

The main thrust of modern memory research, then, suggests that there are three types of biological memory, namely electrical, structural (protein based) and "calcium-sensitized" (chemical). The electrical type supports the span of immediate consciousness, the structural type supports permanent memory, and the calcium-sensitized type provides a means of the former "tagging" the latter with things which have just been accessed and might be needed again within the next hour or so - in other words the function of Medium-Term Memory. It is also the physiological mechanism which underlies the phenomenon of memory consolidation. It is the calcium-sensitized memory variant which allows direct access to items within Long-Term Memory, provided only that they are in the necessary state of heightened excitation.

Medium-Term Memory does not contain any knowledge at all, but rather maintains pointers (sometimes called "tags") to recently activated points within Long Term Memory. Biologically this "touch-and-glow" ability derives from synaptic sensitization processes such as "calcium switching" and "second messenger" neurotransmission, and psychologically it is the key to interfacing the electrical and the structural aspects of memory, and thus maintaining the continuity and coherence of thought. Retrieving memory objects via Medium-Term Memory is easier and quicker than obtaining them from Long Term Memory. After some hours the mental objects in this Intermediate Memory are typically no longer available, but must be retrieved from Long-Term Memory if we want to access them again. Long-Term Memory sensitization as Medium-Term Memory has an important role to play in understanding complex or prolonged items of communication.

Medium-Term Memory is a way station and sorting house where information is stored for a few hours up to 48 hours, in case it is useful. Most of this data is eventually classed as not useful and discarded; the remainder of this Hippocampal content is consolidated and passed on to Long-Term Memory. If we only had a Short-Term Memory and a Long-Term Memory, most of what we have learned would be forgotten in the first two minutes, not over a period of 48 hours or so, where it can be used and evaluated, and the most useful stored for later re-use.

Without Medium-Term Memory, we would not be able to find the car in the evening that we parked in the morning, before we started work. And we would not be able to enjoy a novel, because we would have forgotten the beginning of the page long before we got to the end; we would no longer have a global grasp of the story. Medium Term Memory enables us to quickly identify our surroundings without having to constantly visually scan our environment, since information about previous visual fixations is retained long enough to build up rich the phenomenological experience of a detailed scene. Many phenomena are explained by reference to Medium-Term Memory. One researcher suggests that dreams are forgotten within a minute because there is no Medium-Term store, only a Short-Term store, and that unless information is stored first in Medium-Term Memory and reviewed several times, it will not re-file as Long-Term Memory.

Note that Medium-Term Memory capacity can be increased by intensive practice over a long period, for example learning the intricate maps of London if you wish to be a taxi driver, or spending 10 hours every day memorizing the Koran. The hippocampus has the capacity to grow new cells, if it is used extensively, so the size of the hippocampus is increased in size, especially the posterior hippocampus.

The major difference between Short-Term Memory and Medium-Term Memory is that Short-Term Memory content is lost if we switch our attention to another task, but Medium-Term Memory content is not. It is still there when we return to it - material stored at this level is stored for hours. How long it is retained and whether it is then stored for long-term retrieval, depends on how actively we review and refer to the information.

No matter how diligently one studies, no matter how well material seems to have been learned a few minutes after completing the task, it is not long before time begins to erode our memory. An hour later, it may be possible to recall little more than half or even a quarter of what was committed to memory (and far less if no effort was made to pay attention to the material, understand its meanings and inter-relationships, and compare it to existing knowledge). A day later, almost everything seems to have evaporated from the mind, except perhaps for the most stand-out or impressive facts. People forget what they had tried so hard to remember at a rate corresponding to the following graph:

Passive Storage
The curve that represents the rate of forgetting sweeps swiftly downward, showing that only 30 minutes after information has been acquired, a significant amount of the new knowledge is already being lost. For the first 30 minutes, a lot of what we have learned may be retrieved from Medium-Term Memory, but after about 30 minutes, the first stage of consolidation begins so the rate of forgetting becomes rapid. Memory continues to suffer a rapid decline and after 4 hours about 50% of the content has been forgotten, and 90% has been forgotten after 24 hours - or even more if the material had not been properly paid attention to in the first place. Which means we retain at most about 10 percent of the information we hoped to commit to memory - unless a more active process of remembering is adopted.

Active Storage
Material that is rehearsed, because we are actively working on it (e.g. using the information at work) will be reliably stored for several hours and some residue will remain for several days. In the case of Active Storage, approximately 20% to 30% of the material being worked on is transferred to Long-Term Memory. It is impossible to give an exact figure; this depends on how impressionable is the data - its relative importance and emotional coloring.

Even more effective is a regular schedule of review, incorporating key words and key images, to retain 50% or more of the information.

Mnemonic Storage
When Mnemonics using powerful associations and images are made, Medium-Term Memory is 100% reliable for about 4 to 6 hours and about 90% reliable after 24 hours, if there has been no rehearsal of the encoded material. If material is to be available for a longer period and to be 100% accurate, rehearsal must take place. Unless the material in Medium-Term Memory is rehearsed or reviewed several times it will eventually be forgotten. In contrast, if Mnemonics have been used in conjunction with rehearsal - i.e. the information has been recalled, reviewed, actively used, thought about and integrated - most, if not all, of this encoded material is transferred to Long-Term Memory, in which case recall is nearly 100% even after several years.

Mnemonics enhance Medium-Term Memory, so unless the material that has been recoded in mnemonic form is reviewed three or four times it will, for the most part, be forgotten. Mnemonics enhance our capacity to duplicate; once we have duplicated a passage or whatever, we can play with it mentally. This mental play, a combination of comparing and contrasting will lead to understanding and once the material is understood it will become part of Permanent Memory.

Our potential for memorization is much greater than our usual practical ability. When the time of exposure to the source material and the time of recall is after only a short delay our recall of the material is almost as great as our full potential, but the gap starts to widen rapidly after as little as half an hour. With suitable prompts, most of the material could be recovered from potential Medium-Term Memory even after several days, although not much longer than that, because true forgetfulness has started to occur - much of the material in Medium-Term Memory is erased after a few days. This is why our recognition span is so good. When we recognize something the thing itself is the prompt. Mnemonics are a special kind of prompt. When we use the formal mnemonic systems to encode material, this material can be recovered even after several days, because the mnemonics act as a series of recognition prompts so the original memory is retrieved. Key Words behave in a similar manner; Key Words act as prompts or triggers, so the original wording of the passage can be retrieved from potential memory.

Why use Mnemonics if, by and large, storage is at the level of the Medium-Term Memory and forgetfulness will occur in three or four days? When used correctly Mnemonics can reduce learning time to about a third. To give you an example, there are forty and some American Presidents since the time of George Washington. Without Mnemonics you would have to rehearse this list of Presidents about 100 times to make a permanent memory. Learning the same list in conjunction with mnemonics would probably take less than half an hour. A memory expert could do something similar in about ten minutes. Okay, you may say, but I will forget this list in about three or four days, because this material is not actually in Long-Term Memory. This is true unless you use rehearsal and connection of the information with your wider knowledge base. If the material is already in Medium-Term Memory, however, you will only have to review it about four to six times to make the memory permanent, so there is a considerable saving in the time taken to learn the list and make it part of Permanent Memory.

Unless you are using mnemonics there is a bottleneck between Short-Term Memory and Medium-Term Memory: many repetitions are required to get the information into Medium-Term Memory. Mnemonics allow a person to overcome this.

Once material is in Medium-Term Memory you will have an illusion that you have learned it; this is not true, it will be forgotten in a few days at most, very little will have entered Long-Term Memory. To transfer material from Medium-Term Memory to Long-Term Memory involves over-learning: it must be repeated several times to make it important enough to transfer to Long-Term Memory and integrated with the existing knowledge base. The important dimension in Medium-Term Memory is how many times have you recalled the material in a meaningful context, not just the number of times you have been exposed to it. If something is recalled several times it becomes sufficiently important to be stored in Long-Term Memory.

Many students cram just before an exam. When they do this, they only have the material stored at the level of Medium-Term Memory. They believe they have learned the material, but they forget what they have crammed soon after the examination.

Mnemonics should not replace understanding unless you are doing party tricks, otherwise there will be little storage in Long-Term Memory. Long-Term Memory operates best with meaningful material.

Prospective Memory
Prospective Memory consists of recalling an action or an intention, triggered by either a stimulus or event or a time. Common examples are setting an alarm, making a shopping list or a to-do list. More subtly, you know you need to call your wife before leaving work, and the stimulus of the clock nearing the designated hour reminds you of the task. Meeting a friend (the cue) might remind you to pass on a message (the intention). Prospective Memory is one area where Mnemonics are very useful, and have a wide range of application. An image represents the cue (a predicted stimulus), and when that arises it immediately reminds you of an associated image or chain or images, representing the intended action.

Semantic & Episodic Memory
Material in Medium-Term Memory is sorted into three categories: Semantic Memory (meaningful information and concepts), Episodic Memory (episodes of life experience) and Trash. The Trash memory is forgotten as far as the conscious mind is concerned, although some of this content may continue to be stored in Implicit long-term storage and have subconscious effects. It may possibly be retrieved by a hypnotist and is the basis for Recognition memory (these mechanisms will be described in more detail later).

Medium-Term Memory is sometimes called Transient Episodic Memory in the literature; this is in fact a good definition, because Medium-Term Memory has a high imagery content, whereas, in contrast, Permanent Memory or Long-Term Memory proper has a large Semantic (conceptual) dimension. For example if you were to go to a lecture, then recall it in the bus on the way home, in most cases this recall would be accompanied by imagery of the lecture, whereas if you were to try to recall the lecture a month later, for the most part your recall would be Semantic: most of the perceptual details will have been forgotten. Contrary to popular belief much of our autobiographical memory is Semantic; we tend to create imagery after the fact, little of the accompanying perceptual imagery is the original perceptions. One exception for many people, however, is the auditory imagery of music. Songs and music, in common with Episodic memory, are stored in the right hemisphere, whereas Semantic memories are stored in the left.

Until the age of about 14, Episodic Memory (memory for events) is predominant. From the age of 14, Semantic Memory (memory for information) starts to play a larger role and by the age of 18 at the latest, Semantic Memory plays the predominant role. During childhood the right hemisphere dominates in most case, but after the age of 18 the left hemisphere is the dominant hemisphere for 90% of the population.

Most adults may have a good recall of events of the last couple of days, because most of the content of Medium-Term Memory is Episodic, but they will tend to have a poor recall of events from last week or last year, because they tend to rely on longer-term Semantic Memory stored in the left hemisphere. They may know that they went to Brighton the week before and they may know they went by train, because this is the usual way to get to Brighton, but they may have little or no memory of the train journey itself, because they are using the more abstract Semantic Memory to reconstruct the episode of going to Brighton, instead of getting in the right-brain mode to relive the experience.

In terms of general intelligence, researchers often differentiate between fluid and crystallized intelligence. Fluid intelligence relates to our ability to solve novel problems and is intrinsic to the functioning of Working Memory. Crystallized intelligence, on the other hand, refers to our accumulated knowledge and experiences and how well we can access and use these, as well as practical intelligence, or the ability to solve to deal with everyday problems and situations. Semantic Memory stored in the left hemisphere is the basis of Crystallized intelligence. Fluid intelligence tends to be negatively affected by age, although this varies among individuals. Crystallized intelligence is generally well preserved and may even improve in some areas in old age, which supports the idea that wisdom comes with age and experience. In my opinion, increases in Crystallized intelligence across the life-span more than compensate for the decrements in Fluid intelligence that occur after the age of forty. The secret of making large increases in Crystallized intelligence is to continue to study until we are on our death bed!

The Locale and the Taxon Memory Systems
There are two fundamental systems for acquiring memories in the process of learning:

  1. The locale memory system is that which we use for memorizing location and scenarios. Locale memory is a form of episodic memory, capable of storing semantic type information, as there is a gradient between locale and semantic memory; however, it stores information in context. Consider arriving in a new town and wandering around its streets to familiarize yourself with it. You do not consciously learn anything, yet as you visualize the town and use your imagination with the scenarios in your Short-Term Memory, you start to memorize that city's layout very quickly - it becomes part of Medium-Term Memory and the highlights are passed to Long-Term Memory. This is the locale memory system in operation, utilizing right-brain storage, and it works seemingly without effort. The locale system evolved so that we could quickly update where we are in our surroundings so that when danger threatens we can instantly know which way to move to get away. Since it is vital to survival that this memory be constantly updated as we move around, this system is easily accessed and revised.

  2. The taxon memory system is the form of semantic memory used for rote learning a list of facts or words (as in a definition or piece of poetry for example), or a sequence of actions, such as those we use to play tennis or drive a car, or to solve a specific problem. In contrast to semantic memory, it can also store meaningless information, such as a list of nonsense syllables, in which case there is no need for understanding. To compare this process with the locale memory system, consider arriving in a town, as above, but this time instead of walking around and taking in your surroundings, you try to learn the names of the shops and offices, the street names and so on, using only written lists and directions. Using this left-brain analytical method, it takes a considerable effort of repetition to establish anything in memory, but still, the taxon memory system is of great value. Understanding is not required; it is repetition that is key. The taxon system evolved so that we could memorize courses of action needed frequently for our survival. In particular, those actions that are often repeated are the ones that are stored in Long-Term Memory, rather than the million and one things we might do that are of lesser importance.
Note: It is now known that much language and memory retention, in particular taxon memory, is stored in the left part of the brain, whereas locale memory is stored in the right. There are many memory systems, each dealing with a different kind of cognitive learning. Until puberty the corpus callosum connecting the two hemispheres of the brain is not fullt developed, so children have trouble gathering information in the right side of their brains and accessing the answer in the left side, where mathematical facts, the spelling of words, vocabulary lists and phone numbers are stored.

People who are intrinsically motivated, engaging in a task for its enjoyment value, have a better memory than people who are extrinsically motivated, doing tasks because they are told to do so. Intrinsic motivation is associated with enhanced performance, improved conceptual and creative thinking, and superior memory recall. Field-Independents tend to be intrinsically motivated. The acquisition of locale memory, with its use of visualization and imagination, and real-world involvement, tends to be associated with intrinsic motivation, or self-directed learning. The acquisition of learning by the taxon memory system tends to be extrinsically motivated, part of required education, and may be resisted because of the need for boring repetition. Therefore effective learning methods do well to incorporate use of right-brain visualization and imagination techniques, alongside the rote learning that is required, so learning is both more enjoyable and more effective. The use of mnemonics, i.e. making visual associations that assist in remembering, with their corresponding application of imagination and visualization, is an example of rote learning made both easier and more effective.

Autobiographical Memory
Semantic Memory, because it contains general information such as facts, names, and important historical dates, could be described as a person's knowledge of the world. Episodic Memory refers to a person's memory of events. Autobiographical Memory is a large and important subset of episodic memory containing those events that constitute the story of one's life. It exists as an integrated system that also contains elements of Semantic Memory to describe the autobiographical facts, names, dates and accompanying interpretation of events.

People recall few personal events from the first years of their lives. The loss of these first events is called infantile amnesia, caused by the immature brain development of infancy, and the preponderance of delta, theta and alpha waves in infancy in comparison to the mature adult's preponderance of beta frequencies during wakefulness. In contrast, people tend to recall many personal events from adolescence and early adulthood. This effect is called the reminiscence bump. Finally, people recall many personal events from the last few years. This is called the recency effect. For adolescents and young adults the reminiscence bump and the recency effect coincide. This is an ideal time to do Mind Development!

Autobiographical Memory is constructed as an evolving record of one's past history. A person's autobiographical memory is fairly reliable, although distortions may occur when memories are suppressed or elaborated upon. It provides an important component of one's identity and sense of self. The milestones in one's life are most prominent in Autobiographical Memory. If the first meeting with a loved one involves going to a movie, that event stands a good chance of becoming part of Autobiographical Memory. Other occasions on which movies are attended, however, will be remembered for a short time but probably will not become part of Autobiographical Memory. Instead, those events will contribute to generic memory. Generic memory contains memory for frequently occurring events such as brushing teeth or climbing the stairs. When asked about such events, it is unlikely that a specific instance of toothbrushing or climbing the stairs will be remembered.

Remembering an autobiographical event usually involves both retrieving the content of the event (remembering what) and placing it in time (remembering when). Of course, memory for both fades over time. Autobiographical Memory can be either reproductive or reconstructive. When it is reproductive, virtually all the details are retrieved from memory. When it is reconstructive, a few major points are retrieved from memory and the rest is constructed from generic memory. People are very good at reconstructing memory from generic events, and they are usually not aware that they are doing so. One of the consequences is that memory for old events is often distorted. Sometimes the error is minor and sometimes it is not. Memory researchers have shown that memory for the content of the event gradually changes from being almost entirely reproductive to being, after about a year, almost entirely reconstructive. By contrast, memory for when an event occurred is almost always entirely reproductive.

Experts learn new material in their field much faster than novices, and they retain that material much better as well. The reason for their outstanding performance in learning and memory is that they have a highly organized and detailed memory for their area of expertise. This allows them to relate new material to one or more pieces of information that they already know. Metaphorically speaking, they have many potential pegs on which they can hang new information. When they have to retrieve the new information, they can follow a well-beaten path to that information. Autobiographical Memory is also a highly organized and detailed memory. When it is possible to relate new information to life events, Autobiographical Memory functions in the same way as an expert system. The new information will be learned faster and remembered better than information that cannot be related to life events.

Functional neuroimaging studies of episodic memory have provided extensive evidence suggesting that regions of the prefrontal cortex (PFC) play a role in episodic memory retrieval. A review of PFC activations reported in imaging studies of Autobiographical Memory and generic Episodic Memory reveals patterns of similarity but also substantial differences. Episodic Memory studies often report activations in the right mid-dorsolateral PFC, but such activations are absent in Autobiographical Memory studies. Additionally, activations in the ventromedial PFC, primarily on the left, are almost invariably found in Autobiographical Memory studies, but rarely occur in studies of Episodic Memory.

It is suggested that these two regions mediate different modes of post-retrieval monitoring and verification. Autobiographical Memory relies on quick intuitive 'feeling of rightness' to monitor the veracity and cohesiveness of retrieved memories in relation to an activated self-schema. Episodic Memory for generic events requires more conscious elaborate monitoring to avoid omissions, commissions and repetitions.

Verbatim Memory & Gist Memory
Real, accurate, memory lasts for about 45 minutes. After that memories are filtered through the 'program' of beliefs and experiences that we have built up from earliest childhood. Often we 'remember' things that never happened, or seek explanations and contexts for fragments of 'memory' for which we have none. While material is in the Episodic Buffer, it can be recalled with nearly 100% accuracy, and it is retained for 30 to 45 minutes without rehearsal, so there is Pure Memory for a short period. However, once the material has moved on to Medium-Term Memory proper forgetting occurs, then the material is colored by beliefs held in Long-Term Memory. Of course, if the material in the Episodic Buffer is recalled repeatedly, it can be stored indefinitely. Mnemonics help to maintain a Pure Memory.

In the 1930s, the psychologist Bartlett performed recall experiments using material such as stories about Red Indians. He would read a short story of about 100 words to a student; then after various delays he would ask the student to recall the story. If there was only a short delay, say five minutes, the student could recall the story with nearly 100% accuracy, but if the delay was greater than about 30 to 45 minutes, confabulation would occur - there was a failure to recall unfamiliar details, and material would be added through rationalization that was not in the original story but in line with the student's cultural expectations, so the story read more like a typical English story.

The psychologist Valerie Reyna has concluded there are two distinct types of Episodic Memory: Verbatim, which allows us to recall what specifically happened at any given moment, and Gist, which enables us to put the event in context and give it meaning. When an event occurs, Verbatim Memory records an accurate representation. But even as it is doing so, Gist Memory begins processing the information and determining how it fits into our existing storehouse of knowledge. Information is paraphrased and summarized and put into a meaningful pattern. Accurate solutions to reasoning problems depend primarily on Gist Memory abilities (extracting the correct gist from problem information, focusing on that gist during reasoning, and accessing reasoning operations that process that gist).

In normal people, Verbatim memories generally die away within a day or two, as they are probably only stored in Medium-Term Memory leaving only the Gist Memory, which records the event as we interpreted it. Gist Memory also enables a Pure Memory to be revised. For example, a bird-watcher remembers seeing a lapwing last week though he only later looked up the appearance in a book and learned how to identify lapwings. Another case is the lady who remembers that her childhood home faced west even though as a girl she merely noticed that it faced the setting sun. The re-interpretation is an impure factual memory: a compound of pure factual memory and further inference or realization. Later knowledge or inference is mixed with the memory proper. We can only attend to so much while perceiving. Our ability to replay old experiences lets us add further meaning to them in the future.

When creating a Gist Memory, the mind simplifies or generalizes the Pure Memory, whereby only significant and basic characteristics are remembered clearly, giving a "fuzzy" memory trace of the full event. Pure Memory, i.e. the Verbatim Memory only lasts for about thirty minutes or so, then details start to be lost, and there is a process of reevaluation and revision that gives rise to Impure Memory or Gist Memory, and before long the neocortex spins its web of memories of associations of connotations of the experience. After a couple of days, many of the perceptual details of a Verbatim Memory are lost, and the mind relies on the Gist Memory.

There is a continuum in memory between completely context dependent episodes to truly general knowledge. This fuzzy boundary between Episodic and Semantic Memory is particularly illustrated in the cases of spatial memory and remote memory for personal events. Therefore the distinction between Episodic and Semantic Memory is more complex than classic memory models suggest. In real life, most of our episodic recall is Gist Memory, and most often the recollection is an Impure Memory containing some comforting descriptions that match our aspirations. Gist Memory, then, is an Episodic Memory for an event that is tinged by Semantic Memory, yet it is not Semantic Memory, in the sense of memory of a fact or knowing something, in an abstract sense without a location in time and space. The boundaries of the Episodic and Semantic Memory systems overlap, and the processes of these systems interact.

Long-Term Memory
The brain needs to organize complex information in Working Memory before it can be encoded into Long-Term Memory. In this process of organization, the meaningfulness or emotional content of an item may play a role in its retention into Long-Term Memory. If it is considered important, the information will be held temporarily in Medium-Term Memory, particularly if one has used memorization techniques such as chunking or mnemonics, and if repeatedly accessed there it will be transferred to Long-Term Memory.

The process of further consolidation in Medium-Term Memory starts after about 30 minutes. From the Hippocampus, the selected memories are passed on to the Parietal Lobes. Semantic Memory to the Left Parietal Lobe and Episodic to the Right Parietal Lobe. During our waking hours some information can be transferred from the Hippocampus to the Parietal Lobes (the first stage of consolidation), but there is little or no transfer from the Parietal Lobes to the Temporal Lobes of the Cortex during the waking state. Further transfer to the Temporal Lobes occurs during sleep. The Temporal Lobes are the true site of Long-Term Memory.

Note that the digit span of Long-Term Memory storage without repetition or rehearsal is only between three and four digits. If you were to give me a three digit or four digit number, I would remember this easily a few days later, but if you gave me five digits, I would only remember it 50% of the time, which indicates that five digits is beyond the Long-Term Memory Span, for most people. Typically, the true span for Long-Term Memory is one Chunk consisting of three or four digits. This explains why many senile people can remember a short string of digits, or a sentence, retrieved from Long-Term Memory, even though their Short-Term Memory Span has now effectively been reduced to zero. As a consequence of Long-Term Memory Span, many senile people can learn new material if it is presented slowly enough, perhaps one item of data every ten seconds, and most senile people can hold a conversation of some sort of the other.

When material is stored in Long-Term Memory, the rate of forgetting drastically slows down; it is as though the information is no longer fluid but has become crystallized. Perhaps 10% of the content is lost in a year and many memories endure for a lifetime. It is the basis of our knowledge base.

For the most part, transfer from the Hippocampus to the Cortex occurs during sleep, and particularly the REM periods. Zhang (2004) proposed that sleep has two different stages: NREM (non-rapid-eye-movement) sleep for processing the Declarative (conscious, recallable) memory, and REM sleep for processing the Implicit (subconscious, unrecallable) memories. He further suggested that there are two types of dreams. The type I dream, a thought-like dream, is the result of the memory replay when the Declarative memory is transferred from the Medium-Term Memory to the Long-Term Memory during NREM sleep. The type II dream, a more 'dream-like' dream, mainly occurs when the Implicit, subconscious memories are transferred from the Medium-Term Memory to the Long-Term Memory during REM sleep.

The capacity of the Medium-Term Memory is finite, it can only store a few days experience at most. Sleep deprivation for as little as four nights is enough to turn some people insane and eight days deprivation will turn most people insane. In addition, recent studies provide compelling evidence that the Amygdala is critically involved in modulating the consolidation of Long-Term memories, and particularly of emotional experiences.

Long-Term Memory stores our knowledge about the world. This in turn affects our perceptions of the world, and influences what information in the environment we attend to. Long-Term Memory provides the framework to which we attach new knowledge - it allows retrieval of information decades after it is stored, and it appears to be essentially unlimited in its capacity.

Schemas are mental models of the world. Psychologists believe that information in Long-Term Memory is stored in large, interrelated networks of these schemas, which form intricate an knowledge network. Related schemas are linked together, and information that activates one schema also activates ones that are closely linked. This allows relevant knowledge to be called up when information is presented.

Long-Term Memory influences what aspects of a situation we pay attention to, allowing us to focus on relevant information and disregard what is not important, which allows our senses to function efficiently.

Recent Long-Term Memory and Remote Long-Term Memory
Recently, experts have made a distinction between Recent Long-Term Memory and Remote Long-Term Memory. The latter are core memories that are so deeply etched into your being that they are a part of you, information such as what a cookie is, how to put on a shoe, the words to a childhood prayer, or lullaby that you pass on to your own children, and the name of your country, your mother, or your first dog. They give continuity to your life and help form your unique personality. Who we are is laid down in layers of memory and experience, both real and imagined.

The normal memory loss that accompanies aging usually affects Recent Long-Term Memory, but not Short-Term or Remote Long-Term Memory. This is called Ecmnesia: a loss of memory for recent events that does not extend to more remote ones, a common symptom of old age. Very recent memory is stored in neurotransmitters, but the older and better learned the memory, the more it resides in the protein-synthesised brain structure.

Patients with schizophrenia have significant reductions in measures of Short-Term Memory and Recent Long-Term Memory, but Remote Long-Term Memory remain substantially intact. In contrast, patients with Huntington's disease have impaired Remote Long-Term Memory. Recent Long-Term Memory is relatively unimpaired, events of up to a few years or so ago can be remembered, but many events of earlier life are lost, indicating that there are at least two layers of Long-Term Memory.

Recent Long-Term Memory and Medium-Term Memory should not be confused. Medium-Term Memory stores information for a day or two at most, whereas Recent Long-Term Memory is used to store information that occurred in the close past - it includes information stored for days, weeks or months, such as what you ate for breakfast yesterday, who you visited two weeks ago, or the events of last Christmas.

Research is pointing towards the idea that Long-Term Memory is structured like an onion with several distinct layers, and that knowledge is processed with respect to its surrounding context; each layer represents a different context and keeps related knowledge. Transfer of information into a permanent Long-Term Memory store may entail multiple-stage consolidation processes rather than a single-stage, unitary consolidation process.

Long-Term Memory includes both our memory of recent facts, which is often quite fragile, as well as our memory of older facts, which has become more consolidated. A gradual transition takes place from Episodic to Semantic Memory. In this process, Episodic Memory reduces its sensitivity to particular events, so that the information about them can be generalized.

Recent Long-Term Memory material is stored in the hippocampus for one to five years, over which time it is gradually transferred to cortical storage. To achieve this, the hippocampus operates at six to seven times the normal speed during sleep, as it educates the cortex. When information is in Recent Long-Term Memory, there is considerable forgetting over a period of several years, especially of information that is not accessed nor revised, then what remains is transferred to Remote Long Term Memory, which is independent of the hippocampus. The memories that remain are very stable and very little further forgetting occurs, and this material can often be retained for a lifetime.

Very Remote Long-Term Memory
Very Remote Long-Term Memory is a form of implicit, unconscious memory that was acquired early in life, or as a consequence of overlearning. Since this becomes the foundation for new memories and may be linked to many further more recent memories, such memory is less subject to change and/or loss. As the Hippocampus does not come on line until a child is two or three, storage is probably cerebellar and to a large extent preverbal, so the memory is implicit. There is some evidence, however, to suggest that this layer of memory can be retrieved by hypnosis or by using depth psychanalysis and the memories converted to hippocampal episodic memories through a process of reconsolidation.

The cerebellum plays a role in both Episodic and Semantic Memory. Research results suggest that besides its known role in verbal working memory, the cerebellum contributes to episodic long-term encoding. Also recent evidence suggests the cerebellum may subserve cognition, including the search or retrieval of lexical and semantic knowledge. The role of the cerebellum in these human functions has tended to be obscured by the traditional preoccupation with the motor functions of the cerebellum. In addition to being the site of early childhood memory, cerebellar storage may represent a third and final stage of Long-Term Memory consolidation for sufficiently overlearned material, such as a familiar song, in which Implicit Memory ("knowing how") and Explicit Memory ("knowing that") are connected with each other via the cerebellum, blurring the distinction between Implicit and Explicit Memory.

The cerebellum is more than a backup system for brain programs, the cerebellum creates shadow models of other parts of the brain. Once material has been sufficiently overlearned, the cerebrum delegates responsibility to the cerebellum, as the cerebellum is about ten times as fast as the cerebrum; then recall becomes effortless and automatic. For decades researchers have known that the cerebellum houses procedural memory - what is sometimes called "muscle memory." Essentially this is our "how to" learning. How to ride a bike, how to drive a car, how to jump rope, how to swim, and so forth... are stored as memories in the cerebellum. Scientists have also discovered that the cerebellum is the site of memories of many learned situations that have become automatic, but not necessarily associated with muscles. For instance, the cerebellum stores the alphabet after we learn it. A role in a play, multiplication tables, the skill of decoding words, and the stimulus-response effects, such as knowing opposites (I say "hot" and you automatically say "cold"), are probably also stored here. The cerebellum therefore plays an important role in cognition and psychiatric disorders; immediate and recent memory loss and poor mental arithmetic have been associated with cerebellar dysfunction.

Note: One's memory of experience as an infant before the acquisition of language is in images and kinesthetic 'body memory.' This is true also for much of traumatic experience. Verbal probes of nonverbal memory yield fragmentary results. We can help people draw out those memories by teaching them that "the hand remembers what the head forgets." Finally those unfinished traumatic experiences become comprehended in verbal narrative and become history.

Permanent Memory
Permanent Memory stores information for up to a lifetime. Once material is in Permanent Memory the rate of forgetting is negligible.

It is beneficial for memorization to connect the items to be remembered to other related information (e.g., elaborating on sentences to be remembered, mnemonic systems, or rhyming). In studying new materials it is best to:

  1. Preview the material
  2. Make up questions that you will look for answers to
  3. Read, trying to answer the questions
  4. Reflect while you read. Think of examples, relate it to what you know.
  5. Recite or write a synopsis of the information in each section after you've read it. Re-read what you can't recall.
  6. Review the major points and the answers to your questions at the end.
In other words, the few minutes that it takes for you to review and think about what you are trying to learn is the minimum length of time that is necessary to allow thought to become a lasting, more easily retrievable memory.

Reading, observing and listening are good methods for absorbing data and information into Working Memory. However, this information does not instantly become knowledge once we have absorbed it since it has not been encoded for long-term storage. It can easily become forgotten if it is not accessed for some time or if new information, using the same stimuli and associations, replaces the old. To help make it part of one's permanent knowledge base, one needs to take it through a 'learning cycle' which may include:

  • Comprehending and reflecting.
  • Forming concepts (models, frameworks, generalizations).
  • Fitting it in with previous experience and knowledge.
  • Testing in new situations.
  • Gaining experience in application.
Recent research has demonstrated that 7-10% of what we study is typically retained in Long-Term Memory, subject to it being studied with attention and interest, but this figure may be increased greatly if mnemonics and other aids to memory such as periodic revision are utilized.

The speed of recall of information from Long-Term Memory depends in part on how recently that information has been activated and it also depends on the amount of practice, i.e. the frequency and durations of reviews. A well known psychologist and researcher, Ebbinghaus, has reported that each additional recitation (after you have become familiar with the material) engraves the mental trace deeper and deeper, thus establishing a base for long-term retention. For many people over-learning is difficult to practice because, by the time they achieve bare mastery, there is little time left and they are eager to drop the subject and go on to something else. But reciting the material even just one more time significantly increases retention, so try to remember this and utilize the technique when you can.

Several different types of memory are included in Long-Term Memory. One way to divide up Long-Term Memory is into Explicit memory, Implicit memory and Automatic Memory...

Long-Term Memory

Explicit Memory
Explicit memories are memories that we can consciously remember or 'declare' - they are declarative and can readily be restored to Working Memory, i.e. they are consciously available. Most of what we commonly consider 'memory' is explicit memory. Answers you give on an exam are a product of explicit memory. Everything you 'know' you remember is explicit memory.

Episodic and semantic memories are explicit. Episodic memories (stored in right brain for the most part) are personal, autobiographic memories of experience, such as what your first day of school was like, or what you did on your last vacation - it is memory for specific events in time, and may include emotional and multi-sensory data. Medium-Term Memory may also be called, Temporary Episodic Memory.

Semantic memories (stored in left brain for the most part) are facts or meanings, such as names and dates or ideas and concepts. They are fast changing: quick to acquire but also quick to be lost, if unused for a period of time. Semantic Memory is, by and large a product of a literate culture, thus of recent evolutionary origin, whereas in contrast the other types of memory have an evolutionary history of millions of years.

Implicit Memory
Implicit memories are non-declarative - memories that we do not consciously remember, which nonetheless can be shown to influence our behavior. For example, the process of conditioning. You may have been exposed to many adverts of a particular brand and taken little notice, but when visiting a supermarket, this exposure may affect your purchasing decision. This is an example of memories that we are not aware of influencing our daily behavior as a result of conditioning.

The Implicit memory system is common to all vertebrates and accounts for associative learning; the Explicit memory system is unique to humans and requires language. The association between a subject and a predicate in language is structurally different from the associations that animals are capable of. Animals can learn associations between stimuli, but cannot infer subject-predicate associations, and that is the prerequisite to acquiring a language. Language allows humans to think in terms of “representations”, of “aboutness”. Animals, who are not endowed with language, cannot grasp this “aboutness”. The “aboutness” relationship is the fundamental grammatical requirement for language.

Implicit Memory is much larger than Explicit or Declarative Memory. These are further examples of Implicit Memory...

Procedural Memory represents motor or skill learning, which is memory without verbal mediation. This type of memory is encoded and probably stored by the cerebellum. It includes learning how to drive a car or tie your shoelace. Such memories are slow to acquire but more resistant to change or loss. Initial storage of Procedural Memory may be in the motor cortex and there is evidence that this storage may last for up to half an hour. The maximum rate of forgetting takes place during this period. Finally, the memory content is transferred from the motor cortex to the cerebellum. When the content has been over-learned it may be stored for a lifetime. One never forgets how to ride a bike.

Very Remote Memory simply refers to memories that were acquired early on. Since early acquired information is the foundation for new memories and may be linked to many further more recent memories, such memory is less subject to change and/or loss. We have little or no conscious memory of events before the age of two. During the sensory-motor phase of development, between shortly after birth and the age of about two, memory is almost entirely procedural, hence the site of memory storage at that age is the cerebellum. This is why most traumatic material is also stored in the cerebellum: most traumatic events occur before the age of two.

Trash Memories contain information of an episodic nature that was held and used in Working Memory but not noticed significantly enough to tag it for long-term storage. It is not lost, as may appear to be the case, but is passed on to the Implicit storage of Long-Term Memory. It is the basis for Recognition Memory. The Trash file content is not open to introspection except under special circumstances, such as external triggers (e.g. a diary entry, a photograph, hypnosis or deep psychoanalysis) but it has a vast if not unlimited capacity. In addition to the Declarative and Procedural memory systems, the Trashfile/Recognition Memory system represents a third route to Long-Term Memory.

Years ago Dr. Wilder Penfield found that while using an electric probe on the brain cortex to find epileptigenic foci, patients would bring forth vivid, buried memories when the cortical probe was stimulated. Now comes a report from Dr. Lozano, a neurosurgeon in Toronto, Canada, that while using experimental deep brain stimulation to try to decrease an obese man's appetite as a last resort treatment endeavor, a surprising 'accidental' finding emerged. While using deep brain stimulation to identify potential appetite suppressant points in the hypothalamus, the patient was suddenly bombarded with accurate, vivid recall of events of 30 years earlier that he had completely 'forgotten,' or at least so he thought. The more intense the stimulation, the more copious and vivid the memories. This effort is similar to Penfield's earlier findings, although this experiment's probing is at a much deeper level in the brain. Maybe we do all have a continuous tape of life events deeply buried!

Memory can be categorized in many different ways. One such way to look at your new memories is in terms of recognition and recall (or declarative) memory. These two types of memory simply represent the depth with which you remember the new material. Recognition memory is a superficial memory - if you have this type of memory for a concept, you will recognize the concept when you encounter it and may be able to generate some of the material on your own if you are prompted or given clues. Recall memory is a much deeper level of memory. If you have this type of memory for a concept, you should be able to generate the concept at any time, without any prompting or clues.

Recognition Memory takes effect because of the large implicit recorded memory storage that cannot be recalled voluntarily, but with the correct stimulus it becomes available. This process is called Priming. People can recognize the faces of people shown in pictures if they have seen them before, even if they were long 'forgotten.'

When there is an appropriate trigger, implicit recall may be strong enough to break into conscious awareness and take on the character of Explicit Memory.

Recognition Memory

Recognition memory can be a trap. This can occur in the area of study among other things. When you come to revise some materials, often you will recognize what you are reading and think that you already know it when actually you do not. The truth is you do not know it and would not be able to answer questions about it, because the information has not been sufficiently deeply processed for encoding to Long-Term Memory to have occurred.

Without recognition memory life would be extremely difficult, as much of what we do depends upon it. Recollection, as defined by memory specialists, is the ability to call up specific details about an encounter when one is reminded about it, while familiarity is simply knowing that someone or something has been encountered before. Both are elements of recognition memory and both, new research suggests, are functions of the brain's Perirhinal Cortex. You won't run out of space because that organ is estimated to have a capacity of at least 100 billion images!

Automatic Memory
Automatic Memory - identified just recently - is often referred to as conditioned response memory. Certain stimuli automatically trigger the information or memory. After you hear the first few words of a song from years past, you might remember all the words of that song. Using flashcards or songs in order to learn facts are ways of putting information into this system.

Automatic Memory is found in the cerebellum. Automatic Memory, as far as the cognitive dimension is concerned, is much more than Procedural Memory. The cerebellum is viewed as making an essential contribution to automatic behavioral control. It accomplishes this by copying and rehearsing or 'practicing' the contents of Working Memory; a third stage of Long-Term Memory consolidation. In Mind Development, this principle is applied to the concepts of expertise and giftedness. Familiar songs, frequently recalled information, the multiplication tables, the alphabet, frequently used words and decoding skills are stored in Automatic Memory; these are not Procedural Memory.

It also appears that the cerebellum is involved in word association and puzzle solving. Recall from the cerebellum has a shorter communication lag than recall from Working Memory, because the cerebellum is much faster than the cortex, and the operation of the cerebellum is not so much constrained by Working Memory limitations. Proof of this is that expert calculators, who have a digit span of between twelve and twenty, have a much faster response than untrained persons. This is demonstrated in experiments conducted by Mind Development, first commenced 30 years ago, in which communication lag was averaged over 100 responses to stimuli.

The cerebellum is a remarkable co-processor in its own right. It manages all motor activity that is "overlearned" - that is, so well learned that it no longer needs conscious attention. Walking, talking, speaking, and reciting familiar information are handled by the cerebellum, leaving the cerebrum free to manage other, more complex activities.

More than a backup system for brain programs, the cerebellum creates shadow models of other parts of the brain, opening possibilities of its managing the interweaving of explicit and implicit memory, parsing the domains and structures of Freud's topographic and structural systems, and directing limbic emotion toward meaningful actions.

The cerebellum learns to handle coordinated motor activities by mimicking the electrical patterns that occur in the cerebral cortex as you learn to serve a tennis ball, play a guitar chord, or sing a song. Once you've learned the procedure thoroughly, the cerebral cortex "delegates" the task to the cerebellum, which usually handles it afterward.

Problems can arise when you become anxious about your performance, as with a critical point in a tennis match or presenting detailed data from memory. Under anxiety, the cerebral cortex tries to take over the activity, not trusting the cerebellum to carry it out expertly. Bad tennis serves, bad golf shots, forgotten words to songs, missed comedy lines, and many other "flubs" occur at this instant of conflict between the cerebrum and the cerebellum.

Transactive Memory
As described in Adult Intellectual Development, individuals do not stand alone, they work with others in groups and share information. Individual thoughts may be shared with others and distributed by various means. Whereas intelligence is based on the knowledge and cognitive processes within the brain, extelligence is based on the information, skills and understanding that one can readily access from external sources, the pooled sum of human knowledge. There is distributed cognition within the group, between its members.

Extelligence requires a transactive memory system, through which groups collectively encode, store, and retrieve knowledge. According to Daniel M. Wegner, who coined the term, a "transactive" memory system consists of the knowledge stored in each individual's memory combined with metadata describing each participant's domains of expertise. This ranges from couples in close relationships and families, who need to coordinate information and tasks at home, through teams, to larger groups and organizations who develop their "group mind" utilizing a memory system that is more complex and potentially more effective than that of any of the individuals that comprise it. Complementing the brain power of individuals, computer storage and communication networks come to our aid, most notably corporate Intranets and the Internet. In this way, a transactive memory system can provide the group members with more and better knowledge than any individual could access on his own.

Creative Memory Course
In most civilized societies the development of language centers in the left hemisphere of the brain will produce dominance on that side, while spatial, visual and intuitive problem-solving skills, which are based on right-hemisphere relational processes, will be underdeveloped.

Though a highly developed memory and intuitive skills are not essential for life in modern society, they were important survival skills for primitive man who had no reference books to look up when he forgot something, no maps to guide him on long journeys, and was often in perilous situations where intuitive insight made the difference between life and death. To further evolve, we need to reclaim this heritage, which depends on the restoration and integration of our right-brain processes.

Without memory there is no knowledge, without knowledge there is no certainty and without certainty there is no will. We need a good memory to be able to orient ourselves in a rich network of all that we know and understand, to make sense of it and to move forward to attain goals that are based in reality and true to our selves.

You will learn advanced memory techniques in the Creative Memory Course that utilize the amazing powers of the right brain, which enable you to "file away" any new piece of information so that it is readily accessible for future immediate access.

As you continue to use the methods of cumulative perception taught in this course, this kind of random access memory begins to become second nature. Many memory experts call this the "soft breakthrough" because it happens almost imperceptibly at first, instead of hitting you like a mental bolt of lightning. Everything you find important is given its own unique mental file. Just like the executive whose desk has been buried in paper for years, who suddenly discovers his computer can do a much better job of storing and arranging information, a filed, organized mind suddenly begins to perform impressive recall tasks on demand.

Click here to begin the Creative Memory Course