According to a very widespread belief, human beings are not able to keep in memory more than 7 items at the same time, within a range of plus 2 or less 2, that is from 5 to 9. Is it true? But, above all, is it important to know the so-called “short-term memory limit” for the art of memory?

Imagines agentes, short term and working memory

Imagines agentes give us the chance to handle a certain number of mental elements, all at once. This feature can greatly improve the performance of our short-term and working memory. It remains to be evaluated when it is appropriate to use these kinds of tricks. In the example that we have learned from the ancient book Ad Herennium, showed in another piece, we have seen that the Author uses only a small number of elements to characterize his imagines agentes. For convenience, I propose again here the passage to which I refer since this is really the centre, the core of the art of memory; but to reconstruct the question as a whole I suggest you read the previously mentioned article.

We shall imagine the man in question as lying ill in bed, if we know him personally. If we do not know him, we shall yet take someone to be our invalid, but not a man of the lowest class, so that he may come to mind at once. And we shall place the defendant at the bedside, holding in his right hand a cup, in his left, tablets, and on the fourth finger, a ram's testicles. In this way we can have in memory the man who was poisoned, the witnesses, and the inheritance1.

And this is the explanation that F.A. Yates gives of the imagines agentes:

We have to suppose that we are the counsel for the defence in a law suit. 'The prosecutor has said that the defendant killed a man by poison, has charged that the motive of the crime was to gain an inheritance, and declared that there are many witnesses and accessories to this act.' We are forming a memory system about the whole case and we shall wish to put in our first memory locus an image to remind us of the accusation against our client. This is the image. [...] The cup would remind of the poisoning, the tablets, of the will or the inheritance, and the testicles of the ram through verbal similarity with testes – of the witnesses. The sick man is to be like the man himself, or like someone else whom we know (though not one of the anonymous lower classes). In the following loci we would put other counts in the charge, or the details of the rest of the case, and if we have properly imprinted the places and images we shall easily be able to remember any point that we wish to recall2.

There are in this sketch only a few composite images (the bed, the man lying ill, the ram’s testicles and a cup). In fact, there is only one scene, but there are some images that interact with it, so we could say that imagines agentes are a small number of images that interact in a single scene. So, it is very important to establish the maximum number of images that is useful to use in a single mnemonic composition. This point is pivotal: certainly, many of you know the article by George A. Miller The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information. The question raised by Miller was widely discussed after the publication of his essay. The argument of the capacity of our short-term memory is still nowadays a critical subject in the debate about the cognitive psychology of memory and involves a lot of other aspects of memory, first of all that of the scientific model of working memory.

However, in the simplified version3 of Miller's thesis that everybody knows, it seems that we can process 7 + ̶ 2 items (words, notions, concepts, things etc.) together. In reality, the number of pieces of information that our mind can nimbly handle at the same time is usually lower than 7. The question is quite complex and a lot of theorists of short-term and working memory capacity argue against measuring capacity in terms of a fixed number of elements, they prefer to consider the necessary time and the nature of information rather than a precise number of contents4.

This is largely a model of memory issue. However, here we don’t do scientific psychology; we use the discoveries of scientific psychology for mnemonic purposes. So, we have to realize how many elements to put into an efficient and incisive imago agens. I repeat the concept: no question about how much information we can retain in a rapid and exact sequence; it is important – to the contrary – how many “things” we can easily elaborate and organize at the same time and in the same “mental situation”. So, the right question we have to answer is: How many “things” (notions, words, items of information, concepts etc.) can we handle for mnemonic uses? In my opinion, the answer is not “seven”. To use our memory efficiently, we do not have to push our cognitive abilities to the limit because, in this way, the energies are exhausted soon and the effort is not sustainable for a long time.

The rule of five elements

Since imagines agentes are associations of images that create a logical and meaningful sequence, we have to point out that there are some profound differences between this technique and others like that of creating a storyline. After all, also a little tale is also a collage of mental pictures organized in sequence, but it requires too elaborate and long a plot. To the contrary, an image in the imagines agentes must be linked to an episode, and to only one.

Creating a long sequence of episodes without other elements that hold the story together doesn’t help the memory. Imagines agentes are visualizations of only one element of memory, one at a time, namely an episode. Like an episode, imagines agentes have a good identifiable, recognizable meaning, one and only one central semantic core. We can do a similar reasoning for the use of acrostics and acronyms. In a lot of books we can find advice about the large use of acronyms for improving memory. However, rarely can we find another important, very important, piece of advice: we shouldn’t exaggerate by thinking up words which are too long. We have to simplify and streamline the memory, not weigh down the memories.

Of course, an episode can and always should have multiple elements that interact into it. As a certain number of “agents” play a role in an episode of our life, so there are some specific traits, agents or things in a composition of imagines agentes that interact and that give meaning to them. Imagines agentes are motion pictures with a low number of things that act on them. Above all, these elements are contained in a temporally and spatially circumscribed fact. They are a sequence of images, but there is only one “meaning” to memorize. Usually a mnemonic scene has a small number of elements that interact, no more than five, frequently less, seldom more.

The principle of three elements

A mental image can have many traits that distinguish and define it. I am not arguing that we must represent excessively abstract or stylized images; the more a representation has sensitive details, the better it will remain in memory. I believe, however, that we don’t have to accomplish excessively complex figures or scenes. The advice we can find in a lot of manuals – that affirms that we have to focus on very strong and very composite pictures – is not completely correct. We risk, doing so, transforming rich and meaningful thoughts to burdensome and tiresome mental exercises. The mnemonic images have to be intense and vivid, but not excessively heavy. So, whatever the elements they are composed of, they must have only a small "number" of "memorable" aspects. I said five pivotal elements, but this is the maximum number, I find it more profitable to use only three elements in each image and, above all, in a single composition of imagines agentes. If we build a scene with some images that interact, it is better if we focus our attention and our imagination only on a few aspects, usually only on three particulars that have something that can be a good clue for memory. Three clues that we can associate with each other and with a specific, single, clear meaning. Three free associations for a single mnemonic image are a well-proportioned effort for our mind. Three connections inside a single composition of mnemonic images (imagines agentes) are the right way to strengthen the memories without forcing the memory excessively.

Three things, three aspects of an image that we can remember at the beginning of our mental training are the best choice. The number three is perfect because it allows us to tie two different things through a third element, which I call a “median”. This element allows us not only to relate two different things, but also to be able to pass from one to another without excessive cognitive changes. By “cognitive changes” I mean the shift of the attention or of the concentration of the mind from one mental object to another. We have to do it in this way without making excessive effort. It is possible by connecting two things that directly have no obvious, relevant aspect in common with a third thing that instead has aspects in common with both. The third element serves precisely to define a medium term, something that combines the thing A with the thing B. The number 3 is the number of the relationship.

Every connection, as complex as it may be, is based on the union of two different elements through a third that binds them; or it consists of a chain of connections, each of which is composed of two elements linked by a third part. We have always to bear in mind that every complex thought and all memories are above all a chain of mental images.

A primary imagines agentes

Since two different elements are connected together through a third different (but related) mental object, we can say that a basic arrangement of “imagines agentes” is a mental chain composed of three different elements. They are structured in this way: the subject (the one who carries out something), the action executed by the subject and, third, the object through which the action is accomplished5.

If I have to learn that “book” in Italian is “libro” I can imagine, for example, Nostradamus (the famous astrologer) giving to Monica Bellucci (a famous Italian actress) a book with the symbol of Libra on its cover. In this mental sketch we have three elements and only one meaning. The meaning is the concept/image of the book. The three elements are here so identified: 1) “Nostradamus” that indicates to me that 2) the scales (the second meaningful element) pictured on the cover of the book must be interpreted in their astrological sense. So the word we have to memorize is “Libra” since it has an assonance with the Italian word “libro”. 3) “Monica Bellucci” is used to remind us that we have to consider that the letter “a” is often used in Italian at the end of female words when “libro” is masculine. So, we have to substitute the letter “a” with “o” in order to pass from “Libra” to “libro”. Finally, Monica Bellucci is Italian, I have in this way a further clue to remember that the word “libro” is an Italian word that denotes the object “book”. These kinds of examples are more difficult to describe than to think, nevertheless they have an intrinsic complexity such that it is better not to make them too pretentious. It is mainly for this reason that I advise not to build compositions of images (or little tales) with over complex plots. Moreover, the example just shown revolves around the translation of a word from one language to another. This type of strategy is often used and shown in mnemonics books to teach how to learn the vocabulary of a foreign language. I have often done so myself, although I do not think this is the best way to learn a new language. However, I prefer to use this kind of trick since it is just a good strategy to make clear examples.

As I will say in the course of this article and in the next one, these images should not remain in mind forever, but they have to be only an indication in such a way that the concept could naturally pass from episodic to semantic memory. If we think of the past events that we remember best, we will notice that we do not remember exactly everything about that event, but only specific aspects. Generally, these aspects, in addition to being specific, are also few. Thus, images constructed in this way are intended to fix some semantic contents passing through episodic memory. As we do not remember exactly the day when we learned the multiplication tables, but we remember the multiplication tables, so we will learn that the word “libro” means book although having forgotten the example that I have shown here. Usually, it is sufficient to memorize examples like this and then, after an adequate use of the word (or the concept or whatever else), we can forget the example we have learned without forgetting what we wanted to learn. In other words: we take the meaning although we leave the episode thought which we learned it. So, the rule is simple: use images to fix meanings in memory and after wiping away the images (the episode) hold onto the acquired knowledge.

Episodic memory and imagines agentes

I speak about “The rule of the five elements” and “The principle of the three elements”. What's the difference? It consists essentially in the fact that “memorizing five aspects of an image” is only a rule because it is a question of evaluating the appropriate number for an effective storage. Instead, I speak about a principle of the three elements because it is more than a rule since it is based on a simple logic principle (but not only logic, also cognitive, didactic, etc.) according to which in order to connect two different mental objects we must use at least a different (but partly also similar) third object.

Keeping in mind three things at the same time is also the basis for creating episodes mentally. All the fantasy tales and novels – however complex they may be – are nothing but a chain of episodes, each of which is made up of two figurative elements linked by a third that acts as a “median” between the first and the second. In other words, the episodic memory is based on at least three cognitive elements linked to each other.

It is in this way that imagines agentes are based on episodic memory: from a certain point of view even the episodes of our life, as they are memorized, are imagines agentes. We do not remember all our lives, but only individual meaningful facts, that are meaningful for us. With the same mechanism the imagines agentes work and that is why they are so effective. Those who study mnemo-techniques know that it is advisable to insert imagines agentes into the theatres of memory. However, often we don’t pay due attention to the importance of this integration between imagines agentes and mnemonic theatre.

While in a mnemonic story the images (and hence the “things”) are linked to each other by giving them an artificial logic sequence (that is not based on any element in our thinking); inserting imagines agentes into mnemonic places dispenses us to load our memory to remember the correct sequence of images since it is guaranteed by the mnemonic theatre. We can state that it is appropriate to use only a few elements at a time when we have to memorize. We must organize these few elements in a single episode, they have to be visualized in our imagination through imagines agentes, better only in a little sequence of imagines agentes. Finally, those episodes must be laid in specific places inside a mnemonic theatre.

In a lot of books I find long lists of names and numbers that the reader has to memorize with the aim of developing his own memory. From my point of view I judge this procedure as contradictory; I’m not arguing that it doesn’t work; in some circumstances it runs well, sometimes very well. However, the right question we have to ask ourselves is: do we want to become a storage machine? Or rather, do we want to develop our method of studying and learning, a good method to think better? There is a lot of difference between the Renaissance books about the art of memory and the nineteenth-century treatises on the mnemo-techniques. The latter are permeated with a dull positivist mentality or, worst of all, they are effects of the flat, boring didactics widespread in that century. The ideological position affirmed in the late twentieth century against the study based on the mechanical learning of notions, against every rote method of study is a late reaction to that mentality. The question was badly placed from the start; in reality our mind makes no distinction between storing notions and memorizing images. It does both and often it does one through the other. Notions are just tools to develop our intelligence: we cannot think without them, but they are not all our thought.

1 Ad Herennium, III, xx.
2 F.A. Yates, The art of memory, Routledge & Kogan Paul, London - New York, 1966 p. 3.
3 I say “simplified” since “George Miller (1956) suggested that memory capacity is limited not by the number of items to be recalled, but by the number of the chuncks”. Baddeley, M.W. Eysenek, M. Anderson, Memory, New York, Psychology press, 2015, p. 43.
4 Cowan, Nelson (2001), The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral and Brain Sciences. 24 (1): 87–114; discussion 114–85; Tarnow, Eugen (2010), There is no capacity limited buffer in the Murdock (1962) free recall data. Cognitive Neurodynamics. 4 (4): 395–7; Murdock, Bennett B. (1962), The serial position effect of free recall. View Table of Contents and Online First Publication Journal of Experimental Psychology. 64 (5): 482–8; Bays, P. M.; Husain, M. (2008). Dynamic shifts of limited working memory resources in human vision. Science. 321: 851–854; Ma, W. J.; Husain, M.; Bays, P. M. (2014), Changing concepts of working memory. Nature Neuroscience. 17 (3): 347–356; Baddeley, A (1992), Working memory. Science. 255 (5044): 556–9; Baddeley, Alan (2000), The episodic buffer: a new component of working memory?. Trends in Cognitive Sciences. 4 (11): 417–23.
5 This schema is at the basis of the so called "p.a.o." system (P - person, A - action, O - object), a method of memorization that we will study in a next article. In reality, "p.a.o." system is a simplification of the imagines agentes strategy and, above all, of some Renaissance mnemonics built on the imagines agentes. For the time being, in this article, I would like only to put the attention of the reader on the chain of mental connections required to remember and, of course, to remember well.