The economy and training of memory

Preface

During the last twenty-five years the experimental study of memory has made considerable progress, and valuable results have been obtained. Many of these have an important bearing upon practice and upon the economy of effort. The aim of the following pages is to make these results more accessible to students and teachers, and to win for exact psychology some part of the interest and attention which

it will command in the future.

H. J. W.

February, 1909.

iii

Contents

CHAPTER PAGES

I. INTRODUCTION - - - 1—10

1. CAN THE NATIVE FACULTY OF MEMORY

Chapter I Introduction

1. Can the Native Faculty of Memory be Improved ?

There is almost a consensus of opinion at the present time that the native faculty of memory cannot be improved. Each man is born into the world with a certain capacity for remembering, and all he can do is to use his talent to the best effect. Now, while this opinion does correspond in some fashion to the present state of our knowledge, it can hardly be held to be an adequate expression of it. The fact is, we do not know what absolute memory means. We have knowledge of a number of circumstances which have a good or bad effect on the work of memory — which, in other words, increase or decrease the amount learned, or the amount remembered after a certain time, or the speed of learning. But we certainly do not yet know all these circumstances, and while we believe that, apart from helping or hindering circumstances, the native power of memory varies from man to man, we have not yet been able to give numerical expression to these variations. It is even possible that no such ultimate variations exist. Besides, if they do exist, there is no good reason why we should not hope one day to discover a means of improving memory, just as we have discovered the means of combating disease, and of increasing the health and strength of individuals. This would, however, simply mean that we had discovered factors hitherto unknown, whose temporary operation produced a permanent improvement in memory power. There is, indeed, little reason for the assertion that the native faculty of memory cannot be improved, except the fact that we do not often observe a spontaneous improvement of memory. Yet, in view of the fact that a spontaneous and permanent deterioration of memory is no uncommon occurrence, we might well believe that the native faculty of every normal man is the same — that is, the best possible — apart from the help or hindrance given by factors of temporary duration. For these reasons it is better at present not to suggest the possibility of an improvement in memory, but only to teach how the memory may be used most economically, how it can be got to do the most with the least effort.

A certain general vigour or weakness may be characteristic of the memory of any one person, but that this vigour is not so great for some things as for others is a matter of popular knowledge. It changes with the needs, habits, interests of the individual, and with a hundred other circumstances. Besides, to talk of a native faculty of memory suggests that we have accounted for all that might hinder or help memory, and have sifted out the essence of memory itself and measured its strength. We are, of course, still far from this goal of memory study. We shall have a better view

1—2 of it after we have travelled farther over the road of careful and accurate investigation, and it is probable that we shall then not talk of a faculty of memory, which might be improved as a whole, but of many processes of memory, each helping to build up the whole, and each brought about with the least expenditure of energy.

2. Whether we ever completely Forget Anything.

It is quite another question whether we ever completely forget anything. Most people imagine we do not. The preoccupations of our daily life and our vivid memory of many details of the past hide from us what we have forgotten, though every now and then the veil is lifted, and we recover with joy or dismay the memory of some forgotten word or deed. On the other hand, those who have lost hope and take refuge in the past, as well as those who are fretful of the past and brood over it repentantly and complainingly, soon come to the conviction that nothing is really beyond recall. To persistent brooding even the dimmest past will yield. There are many other motives for a belief in a complete memory record of the past. They need not be recounted here. This question, like the previous one, is insoluble because we have no absolute or complete record with which to compare our subjective memory record. We may, indeed, believe we never forget anything. This does not, however, mean that we can recall everything at will, but only that we cannot be sure that what we can usually specify merely as the event of yesterday or a year ago, or of such and such an occasion, will not return in detail to the mind as the direct object of memory.

3. The Definition of Memory : Distinction from Trace of Influence.

If a series of foreign words is learnt to-day with a certain amount of labour, and relearnt with a less amount of labour after many years of forgetfulness, it may well be assumed that the influence of memory shows itself in tie lessened labour of the second learning, although it must not be forgotten that intervening experiences may be the cause of the saving in labour. So it might be said that every thought and experience we have, has some effect on our minds, just as everything we eat has some effect upon our bodies, and that therefore nothing is forgotten. But this is not the proper meaning of the word " memory." To remember an event is to recall it clearly and consciously to mind with the help of some clue.

4. The Experimental Investigation of Memory.

The knowledge of our time suggests only the attempt to improve the working of memory. We must strengthen in it those influences which favour remembrance, and remove those which quicken forget fulness. This, of course, has been the endeavour of men of all times. But conviction and system-making have often outrun knowledge, and knowledge itself has been based upon persuasion, effort, and the cursory observation of a few cases rather than upon facts and generalisations, valid in prospect for all, because, in fact, true of so many.

Only within the last few decades have exact observations been made of the conditions which influence the working of memory. Psychologists have begun to investigate memory experimentally. There is nothing new or startling in this. A great many sciences now extend the basis of their knowledge by means of experiment, and all of these knew a time when generalisations were made hastily on a small number of casual and imperfect observations. In the young days of a science such hasty generalisations are unavoidable, because the motives urging to generalisation are at all times strong, and the desire and attempt to form theories is at all times laudable. Such an attempt may lead to renewed care in the establishment of facts already noted, to the discovery of new facts, and so to a better theory. It is not its method of generalisation which characterises modern science, but the demand for an ever-increasing exactness and number of data on which to base generalisations. Modern inductive method demands for its conclusions a basis of observed fact such as is provided either by a few observations under a limited number of known conditions, or by a large number of observations under known conditions and an accompanying complex of unknown conditions which are all of limited influence, more or less irregular occurrence, and irregular tendency. It is under these latter conditions that psychology experiments. It belongs not to the group of exact sciences like physics and chemistry, which can express laws with mathematical exactness, systematise these and deduce unknown ones mathematically. So far as psychology is quantitative, it is a science of probabilities, based on experimental statistics. As such it has been able to achieve some unity of opinion about the operations of memory and its economy, and so to get rid of the distressing divergence of views which was so general in the previous literature of the subject.

5. Systems of Memory Training.

Where knowledge of the natural basis of any important practice is imperfect, it is natural to attempt to complete it by rules and principles. So we find that well-nigh all the old theories about memory took the form of systems of memory training, each almost infallible in its own eyes, though differing in many points from any other memory system. There have been systems of all possible kinds, some offering greater, some less results. It is needless to describe them. They are well known. Any pamphlet on the subject will give a good idea of them all. And yet such systems are not all valueless, for, though based on rough trial and general impression, they have been handed down through many centuries, and bear the mark of the experiences of many kinds of men. Thus, both their foundation and their validity may be greater than it seems. But, of course, they were bound to miss many of the influences which affect memory, such influences especially as are too subtle and indirect to force themselves upon our notice. The impressiveness of these systems is derived partly from their seeming simplicity, and partly from the results they describe and lightly promise to everyone, although these are often more remarkable for their oddity than for their usefulness. In principle such systems of memory training are all out of date, although in some cases their contents have not yet been tested or outreached in psychological experiments. Their best effect is to stimulate the will to remember, although they often detract from this by rousing the desire to produce a big show with little substance — in short, how to seem to have a good memory. They have often served a good turn in that they have persuaded many who have followed them that they had, after all, not so bad a memory as they imagined, and so have saved them from themselves. Most of us are slaves to our bad opinions of ourselves.

6. Conclusion.

In order to use the memory to most advantage we must first know and understand how it works, which influences increase its power, and which act against it. Our practical efforts must thereafter consist in encouraging good influences and eliminating bad ones. We must therefore observe the memory at work, and measure its activity under varying influences. This is the experimental investigation of memory.

Chapter II Tee Experimental Investigation Of Memory

7. The Material.

How can we experiment upon memory ? It is not difficult. We must first obtain material to be learned, and since, in order to derive general statements which will hold good for everyone, a large number of experiments has to be made with different people, our material must be as uniform as possible. Experiments have been made with tones, with colours, with lines of various lengths, and other such simple things ; but experience has shown that the most elementary experience is not necessarily the most suitable material for showing the operations of memory. The use of very simple material is complicated by a variety of conditions which need not be enumerated here. 11

Of all material proposed, the one which most resembles the kind of thing we have to make an effort to remember in daily life is the nonsense syllable. It can always be of the same length, and can be constructed according to fixed rules, based on the experimental results obtained by using syllables constructed in various ways. By the use of such syllables, of which great numbers can be made, the very disturbing influence of the meaning of words with their manifold associations can be eliminated from the experiments. It is a familiar fact that words whose meanings are known, are remembered more or less easily for all manner of reasons. Nonsense syllables, having no meaning, provide little attraction to the memory, because they are almost devoid of associations. Thus, then, a large number of series of eight,* ten, twelve, or more syllables can be constructed, of such a kind that the same syllable either never occurs twice or

* Two examples of such a series of nonsense syllables :

recurs only after a long interval. So the conditions under which this uniform material is learned can be varied from time to time, and their effect on the rate of learning and the duration of memories observed. It is, of course, not necessary to confine oneself to nonsense syllables. A great deal of work on memory has been done with prose and poetry, and, indeed, the results obtained with nonsense syllables should always be tested and compared with those derived from the use of intelligible material.

8. Method of Taking into the Mind.

Further, a uniform method of taking into the mind what is to be learned, must be obtained. This is usually got by arranging the nonsense syllables or other matter in perpendicular columns on a drum which is made to revolve at different speeds, and before which a small screen is placed. In this screen there is a small slit, through which the learner looks, and in which he sees one syllable after the other appear gradually or suddenly, and disappear similarly, to be replaced by the next, and so on. The learner attends and attempts according to the instructions given him, which may prescribe silent or loud reading or the like, to learn the sequence of syllables as quickly as possible.

9. Memory Measures. (a) The Number of Repetitions.

As a measure of the rate of learning, the number of repetitions of the series of syllables, necessary before the series can be repeated by heart once or twice, may be used. Under different conditions this number will be different, and will serve as a basis for generalisation, if a sufficient number of experiments are made under the same conditions. So, too, the number of repetitions necessary before the series can be relearnt by heart after a certain interval — say twenty-four hours or a week — will serve as an index of the amount that had been forgotten in the interval. Further knowledge may thus be gathered. For it may not be taken for granted that the ability to repeat a series of words by heart once, always represents the same amount of mental work. It may be that, tested by the number of repetitions necessary to relearning the series next day, the effort may be seen to be of greater effect under one set of conditions than under the other. By such means a knowledge of the conditions of the most economical methods of learning may be gained.

10. (6) The Number of Specified Recalls.

Instead, however, of testing the memory by the number of repetitions which precede a free repetition by heart, the strength of memory may be determined after any given number of repetitions by presenting to the learner one of the words or syllables learned and requiring him to recall the one which followed or preceded it in the series in which it occurred, or the like. The number of syllables correctly recalled in this way is then a measure of his memory of what he had learned. Most valuable knowledge is obtained in this way. For it is obvious that while one may not be able to repeat a series of words quite correctly, one may have retained much of it. Or again, one might be quite familiar with a piece of verse and recognise every sentence and phrase of it, and yet be quite unable to repeat it by heart when given the first line as a clue. If, however, a test were made as to the ability to complete phrases from the piece, to put the right word after a given word or phrase, it might probably be found that a good deal of the verse had been retained. What does this mean ? It means that the ability to recall is not the only test of memory. We remember to some extent much that we are not able to recall in its entirety. There is, however, obviously no direct method of measuring the strength of such a memory.

11. (c) Recall Times.

The ability to recall is, however, the roughest test of memory, even though some degree of refinement is obtained by counting the specified recalls. But it is well known that two people with equally correct memories may show very different degrees of command of these. One who recalls well, answers promptly. Slow and hesitating answers in which careful deliberation is not involved, are the sign of a much less efficient memory. The townsman, running from street to street every day of the year, talking, discussing, and recalling perpetually, will direct you on your way in a quarter of the time the countryman needs, although the latter has much less to remember in the matter of roads and directions. In school, " Who answers first ?" will give the prompt memory its just opportunity. The time, therefore, which elapses between the giving of the question or clue and the first word of the answer, can be used as a good measure of memory. For a complex object of memory, of course, these recall times in conjunction with the correctness of recall afford a very elaborate and thorough measure of memory. If there is a fair number of short recall times and only a few long ones in a high percentage of correct reproductions, we may call the memory for the complex better than if in the same percentage of correct reproductions there be a smaller number of short recall times and a larger number of long ones. It is evident, then, that one may have a fair memory of a complex of which one is able to recall nothing voluntarily, and a poor memory of a complex of which a fair number of tags remain in the mind. This is one of the reasons why light theatrical pieces appeal to the mass of people more than others do.* Seeing that this method gives not only statistics of

* The simplicity of the lighter piece and the repetition of favourite songs give many details a sure footing in the mind of the listener. The more artistic work, on the other hand, demands sustained attention, is more complex, and often leaves no place for the repetition of parts. It is lovely in its details, of course, but it also aims at producing large unitary impressions, which may often not become quite clear till several hearings and some thought have been spent on the work.

what is remembered, but also time statistics, it is evident that we have here a much wider basis on which to raise generalisations. This method is therefore much more fruitful in detailed results than the others.

12. {d) Other Measures.

There are several other methods by which memory may be investigated. The ability to recognise what has been presented to the mind of the learner once or oftener may be tested. We do not recognise immediately everything we have seen once. There are people who even forget after a short time whether they have read a novel or not. We all know how many a book read after an interval for a second time appears almost new. Unfortunately, however, no very reliable test for recognition is known. A learner can either recall a word or he cannot. He will hardly omit to say the word if he can. But he may say he recognises a word, either without being sure that he does so, or without really recognising it at all. Not

2—2 everyone is equally fitted to be the subject of psychological experiment, and, therefore, where a subject may deceive himself, as we all do unwittingly at times, it is better to hold to methods that can be controlled objectively. Such are the methods of recall already described, and about these there can be no illusion. Another such method of somewhat less value consists in encouraging the learner to repeat what he has learned, even though he cannot do so completely or with confidence. The cues find clues given to him to help him along are then counted, and these numbers form the statistics. It is, however, not very clear just how much help such a clue does give, and the method is, consequently, not of any great value or reliability. Then, again, instead of counting the number of words which can be recalled when each preceding word is given, one may simply count the number of words that can be recalled voluntarily out of the whole complex. Obviously this is a measure of a certain kind of memory, but it neglects all that cannot be recalled spontaneously, which, as shown above, may be a good deal. It also omits to record what might have been recalled, had a clue been given. Still other methods are possible, but enough has been said to describe the general manner of investigating and measuring memory.

13. Association: Its Definition.

What is the central element in memory which is better in one individual and worse in another, and which may be strengthened or weakened by a variety of factors ? It is association. Association is one of the fundamental conceptions of psychology. This conception has been formed to explain the relation of our momentary, fleeting states of consciousness to one another, as forming simultaneous and successive states of a conscious individual who thinks, feels, and acts more or less as a unity. Now, there has been a great deal of trouble in the history of philosophy and psychology over the question whether ideas are independent, self-contained unities, merely linked together into bunches and chains by association, without affecting or changing one another thereby, or whether they influence one another in some way — as, for instance, a red background will influence a grey patch lying on it. The attempt has been made to work out the question consistently from both sides, of course for long without an exact investigation into the facts of the case by means of observation and experiment, and so far, one must say, all attempts have failed, and have often landed in absurdities those who have made them. Material for the solution of the problem is still being gathered, so that no final statement can be made. Fortunately it is unnecessary to venture upon any here, for the conception of association is necessary to a comprehension and analysis of mental states, whether these do or do not affect one another by coming together. There can be no doubt that ideas, whether these be mental images of colours, forms, sounds, tastes, smells, and other sensations, feelings, or thoughts, do get arranged differently at different times, and that a certain amount of mental work, of effort, is necessary to hold them fast in any one given arrangement. Once brought into any one arrangement, whether by outward impression or by voluntary effort, they tend to remain in it, so that under certain circumstances the recall of one or more of the ideas will tend to bring them all, together or in succession, to mind. Some conception is necessary in order to think about and discuss coherently such arrangements of ideas, and this conception is association. Association is what first makes the reproduction of one idea by another possible.

14. Quantitative Distinction between Associations.

Now, as has been already remarked, we observe frequently that one arrangement or sequence of ideas may be harder to impress on the mind than another, and that certain ideas can be recalled more quickly and with greater certainty than others. Here we find a basis for quantitative distinctions between associations. This is provided by assuming that one association can be stronger or weaker than another. If one idea is recalled by another, we say the association between both is stronger than it would be if the idea could no longer be recalled. If one idea recalls another more quickly in one case than in another, we also say the association is stronger in the first instance, and so on.

15. Problems Stated.

Now, what makes an association strong or weak ? This is just what the experimental investigation of memory must try to find out. What influences determine the strength of an association ? Under what conditions can the strongest association be produced with the least amount of effort in the shortest time, and what will secure the recall of an idea at the most appropriate time, and at any desired time ? We shall proceed to enumerate these factors directly, but before doing so one or two general questions must be discussed briefly.

Chapter III Some General Questions

16. Is All Memory Mechanical ? Is there such a thing as a purely intelligible memory, or does all memory rest on the mechanical basis indicated in the last chapter ? The general tendency at the present time is to hold that memory rests on a purely mechanical basis, that certain connections have to be made, and that there are certain conditions under which such connections are made strongest and most lasting. A good memory means, then, either the capacity for very strong associations, or the natural or acquired habit of producing the strongest and most lasting associations with the least labour. The question is of interest to the teacher. The answer just given does not mean, of course, that in learning a piece of verse, for instance, it is necessary only to get 25 up the words parrot-like without minding the sense. We shall see later on how wrong this is. It means that a however many thoughts are brought together, the necessity of acquiring the sequence of words by means of association of one word with the other will never be removed. On the other hand, these associations may be formed more rapidly by means of the ideas with which the words are already firmly connected in the mind — that is, by means of their sense. The thoughts of the verse, then, serve to bind groups of words together easily, and to .unite these groups into a whole. Any ideas introduced into the verse beyond what is suggested or made necessary by the verse itself, will not reduce the work to be done in forming the required association. They will, if anything, increase it by the amount of memory work to be done in linking the superfluous ideas to words which do not suggest them, and to one another, and then in getting rid of them. For if the verse is to be learnt properly and fluently, they must be got rid of. Memorising by the sense is easy, because the associations between the words and their sense have already been made, and it would mean more work to bind the mere words together by association than to associate their sense in its already developed form — complex thoughts. Yet the association between the word and its sense is in itself a mechanical one — that is, it is itself merely an association ; there is no intelligible intermediary binding the sense to the word. Any word could mean anything. This means that only the association with a meaning gives a word meaning. All memorising rests, therefore, ultimately on a mechanical basis, whether, trains of words or thoughts be memorised. Whether there is not also some form of intelligible memory, more especially in the sphere of thought, remains to be seen. Thoughts certainly do bear relations to one another other than those of mere contiguity or succession in the mind. Possibly these relations might under certain circumstances fulfil functions analogous to those of memory. If so, we shall find out about them in good time.

17. Has the Child a Better Memory than the Adult?

There is a very widespread popular opinion which says that the memory of the child is stronger than that of the adult. Evidence of this is the fact that the child learns a language in a few years — a feat which a large number of adults feel to be beyond their power— not to speak of acquiring the accent properly. See how quickly the child retains small verses and fairy stories, and how ineradicable the memories of childhood are, even in the last years of the longest life. Our memories must have been stronger when we were young, we think. We rarely learn a verse of poetry now. We forget the contents of a novel in a month or two, if not in a day or two, after we have read it ; and as for remembering the past, why, it is often hard enough to remember what we did yesterday. Against all this, however, one may put the fact that we remember far more of the scenes of the year just gone than the child does. Moreover, the child learns very little of his language in the first few years, and any adult of average intelligence who might be surrounded by as much careful attention, and taught so persistently and carefully as the child, would learn in the same time far more than the child does. The child, however, has in so far the advantage, as his mind is unoccupied. It is to his ever-increasing advantage to learn continually, and he has little to forget while doing so. On the other hand, the adult is settled in life ; he has adapted himself to his surroundings, and usually does not want to learn anything new.