Is rote learning useful or not? The answer is yes. But rote repetition is also the worst way to learn something. It is repetitive, tedious and requires only a minimal semantic encoding. So students often repeat paraphrasing the texts that they need to memorize. This is a good practice, especially if integrated with the use of periphrases and metaphor for what you have to repeat. If this form of active repetition is then integrated with examples and other forms of explanation, we can obtain a good tool to aid learning.[1]

Usually, the student pretends to be a teacher and explains the lesson to another real or imaginary student. So, for the student to repeat becomes a form of reworking that involves more than simply memorizing other mental processes such as reflecting on what he says, analytical and critical abilities etc. This methodology should be adopted but also integrated with other strategies, it has some positive aspects, but has some gaps from a structural point of view.

As you may have understood the issues, concepts, formulas, theorems and so on, if they are not fixed in a well-defined pattern of knowledge in the mind, they will always remain vacuous or require a further mnemonic effort to become permanent memories. Reprocessing and binding new notions does not guarantee remembering them if, at the same time, we don’t tie them to a stable and stratified “plane” (this term is used here metaphorically with reference to its geometrical meaning) of knowledge in our minds.

So, we must find a way of fixing the knowledge in mind and after, but only after, repeating them in an active manner that is revising, re-elaborating them. In the active repetition, it is essential to identify some key points (be they notions, images, concepts, etc.) and rotate the other elements of knowledge around them. When we actively repeat (paraphrasing and explaining what we are learning) is, therefore, important to find the cornerstones on which to fix the other knowledge.

The identification of these points should be part of the same active re-enactment. That is, when we repeat, interpreting and paraphrasing what we are learning, we must also consider the elements that can serve as points of support for all the others. The identification of these points should be part of this active re-enactment.

The principle is this: it is the criterion for “allocation” of memory in mind that determines the quality of the subsequent re-enactment. The memory to be effective should not be too complex, namely overly detailed. It should instead be 'dotted' with some simple mnemonic elements, and, if possible, interrelated within them. It is the overall pattern that they make up to make later, in the retrieval, a complete and realistic picture of what you memorized. So it is the “allocative repetition” (so I call it metaphorically), that places the notion repeated within a specific stable pattern in memory, to give the best guarantees.

Scientific psychology has concluded that the amount of time dedicated to learning is directly proportional to the amount of learning material: the more you study, the more you learn. Such a conclusion is common ground, or so it seems. But contrary to appearances, this does not mean that we should learn in a rote way; in fact, even if we study and engage all our cognitive resources the principle remains equally valid.

Baddeley and co-workers carried out a survey on behalf of the British postal service to understand how long it takes to get their employees to learn how to type. They discovered that devoting 4 hours a day to learning to type allows them to reach the rate of 80 beats per minute in 4 weeks with 80 hours of exercise (4 hours per working day, from Monday to Friday). The same result was achieved by those who practiced only 1 hour per day in 55 hours of exercises, but if we consider that they exerted only 1 hour a day for 5 days per week, they took 11 weeks to get the same results as the other group (i.e. the one that was exercised for 4 hours per day).

In a nutshell, we can say that distributed practice (the studying that breaks practice up into a number of shorter sessions) requires fewer hours devoted to study, but massed practice (which comprises fewer, longer, learning sessions) allows you to save time in absolute terms. Much therefore depends on the specific condition of each one, if we have a long period of time to plan our program of study distributed practice is better, but if we are bound by strict deadlines massed practice may become a necessity.

In the first case (that is you have enough time), one of the concepts that you have to keep in mind is the frequency of the practice. It is demonstrated that, in case of needing to learn a high quantity of notions, better results are obtained by splitting material and study times. This system is called microdistribution of practice. «If you are presented with a single item on two occasions, do you remember it better if it is presented and tested in rapid succession, or is recall better if the two presentation and test are spaced further apart?

Fortunately, the answer is clear; spaced presentation enhances memory. [...] The solution is to use a flexible strategy in which a new item is tested initially after a short delay, ensuring that it is still recallable. Then, as the item becomes better learned, the practice – test interval is gradually extended, the aim being to test each item at the longest interval at which it can reliably be recalled» (Baddeley, M.W. Eysenek, M. Anderson, Memory, New York, Psychology press, 2015, p. 112).

The principle is quite simple: the knowledge acquired should be tested at increasing time intervals, so that at the beginning, it stabilizes and then remains in the long-term memory. This is possible because testing means returning a notion to the awareness level. We must do this more often in the beginning, that is when the memory is weaker. Gradually, as the notion becomes more stable in the memory, test intervals can be extended.

Schematizing: Learning for the first time a notion (listening to or reading it) → repeat it after 1 hour → repeat it after 4 hours → repeat it the day after → repeat it after 2 days → repeat it after 4 days and so on until the information is definitively acquired.

Testing is thus a form of repetition, an active repetition, but with a cue. Testing at increasing time intervals is a good strategy because it reflects the natural way in which stable knowledge is formed in the long-term memory. But it’s a passive form of memorizing and must be integrated with a more “proactive” practice. It is easier to learn something if we bring it into consciousness without external aid. This ensures a better grip of the new information in our minds. This form of memory reconstruction carried out by the subject without cues is called “generation effect”: «if you succeed in remembering an item for yourself, this strengthens the memory more than if you have the item provided for you; this is known as the generation effect [...] that is the items that you yourself have generated successfully are remembered best» (Baddeley, M.W. Eysenek, M. Anderson, Memory, New York, Psychology press, 2015, pp. 111-112).

To improve memory we have to combine the principles of the distribution of practice and that of the generation effect. The aim is on the one hand to ensure that the subject can recall the information autonomously (without the cues) and on the other side to try out learning at increasing time intervals. Each time that the subject responds positively to the test the interval should be lengthened. If the subject receives confirmation of the accuracy of what he recalled or, on the contrary, an explanation of the error that he has committed, this affects the subsequent memory operations in a positive manner.

Evidently this activates a whole series of benchmark mechanisms, re-evaluation and redefinition of memory and those mental operations go on one side to strengthen it (the memory) and on the other side to perfect, remodelling erroneous memories into correct ones. In this way the recovery phase has been transformed into a new phase of encoding. The only rote repetition doesn’t much improve memory, in fact – if the feedback does not intervene to correct and redefine knowledge – the erroneous memory remains in mind for an indefinite period.

The generation of memory is, strictly speaking, a real phase of recording because it always involves a rework. We have to rummage in the depths of our minds searching for clues that cause us to reconstruct the memories. Feedback accentuates and perfects this encoding operation that, in the case of error, is also a procedure of redefinition of information and, in case of confirmation, a strengthening of the knowledge.

Also for this reason it is important to spend a short amount of time between the test and the feedback, so that the subject has time to be able to restructure the memory and then compare it and adapt it to the received feedback. Among the various types of tests should be preferred those in which we have to answer without having any support points (such as an open-ended question) than the tests with multiple-choice answers, which does not generate a response. A response that is already given to us, must only be acknowledged amongst others that are proposed to us.

In conclusion: the simple repetition, also when it leads to memory, does not facilitate the regeneration of the concepts by the subject, it does not allow the reworking of the content and, above all, it does not allow correcting the errors of memory that are made at the beginning and that tend to remain. The feedback we get from testing instead is useful for correcting errors and not only to aid memory retrieval. But the strategy to combine the practice of testing with that of the generation effect (that is produced by an autonomous effort by the subject) could be further improved if, before, in the coding phase and then in the revision, we identify some key concepts around which to spin all the others.