There is no department of perception which does not furnish instances of associative supplementing. We perceive at once that a drawing in perspective is intended to represent an arrangement of objects in tridimensional space. We accept the rough brush- marks of a theatrical background as an adequate representation of a landscape. — How little we actually hear of what is said to us is shown by the difficulty which we find in understanding a conversa- tion in a foreign language, with which we are familiar only in its written form. We must wait till we are able to supplement the sounds heard, to supply by association the slurred and abbreviated syllables which the ear does not 'catch.' — When we are 'feeling' our way across a room in the dark, and come into contact with a hard object, we say at once: "That is the table! " The tactual perception, incomplete as it is, calls up a visual idea and its ver- bal expression. — The scent of sandalwood is supplemented by the 196 TJie Association of Ideas visual idea of a glove box or stamp case of sandalwood inlaid with ivory; the smell of roasting meat by the visual picture of the ' laid ' dinner table; and so on.
Method. — One of the commonest instances of associative sup- plementing is the right reading of words wrongly spelled. Even a practised proof-reader may overlook mistakes in very famihar words (§ 42). On the other hand, the misprints in a book which is written in a language not so familiar to us as our own attract our attention at once. We read English by general impression, sup- plementing what we see as we glance quickly over the printed words; we read German or French more accurately, because more slowly and toilsomely.
These facts suggest a method by which the conditions and attri- butes of associative supplementing may be investigated. Let an assistant prepare a number of cards, upon each of which is written a one-syllable word, more or less misspelled. For stage, e.g., he might take the following: siage, slave, seaye, seaue, etc. Series of, say, ten cards are formed. The larger part of these have mis- spelled words: i-/^^*? (stage), work (work), qtace (place), etc. To avoid the expectation error, however, one or two rightly spelled words must be included in each series. You are shown the ten cards, one at a time, for.2 to.5 sec, and required to read what is written on them.
To get at the conditions of the association in these cases, you must subject your results to a careful analysis: noting whether the familiarity of the word has anything to do with its supplementing, whether its form is of importance, whether first or last letters, vowels or consonants, long or short letters, etc., are most easily supplied.
The atti'ibtUes — extent and intensity — of the association can be determined by the assistant, if enough experiments are made. Thus, by varying the amount of the misspelling, as in the instance of stage given above, he can discover how extensive the alteration of the word may be, and yet be overlooked by the eye, — how many letters may be wrong while the word is still read aright by association. He can further determine the intensity of the supple- menting, either by questioning you closely as to the vividness of the letters which you say you saw, or by increasing the time of exposure till you are just able to read the misspelled word cor- rectly, that is, till the peripheral impression just outweighs the cen- tral supplement. The latter is the more reliable method. The quality of the associated ideas is always that of the given impres- sion: black letters on a white ground. Their duration can hardly be made out.
Auditory Localisation. — Some of the most striking instances of associative supplementing are afforded by the localisation of sounds. Sensations of tone and noise possess no spatial attribute, and our auditory perceptions cannot be arranged in space, as visual and tactual perceptions can. When we localise sounds, we do so by indirect means, by the help of secondary criteria. The auditory perception must be supplemented by other ideas.
Our idea of the direction in which a sound comes to us is based partly upon tactual sensations, proceeding from the skin and muscle of the external and middle ear, and partly upon the dif- ferences in the intensity of the sound, as it is heard by the two ears. A sound which comes from the right will evidently be louder to the right ear than to the left; while the impact of air- waves upon the right pinna will be the stronger, and the adjust- ment of the right tympanic membrane the more noticeable. Our idea of the distance of the sound is an idea of the distance of the source of sound, i.e., a visual or tactual — not auditory — idea of distance.
Method. — Seat yourself in a chair, and let an assistant chalk upon the floor a circle of i m. radius, whose centre is the centre of the imaginary line joining your two ears. The circumference of the circle can easily be marked off into 72 parts, i.e., into units of 5°. Close your eyes. The assistant tests the accuracy with which you can localise sounds coming to you from different directions by holding a stop-watch, on a level with your ear, at various points of the circumference of the circle. Having taken up his position, he touches you upon the hand with a rod, and starts the watch. You hear its ticking, and point with another rod in the direction from which you think the sound comes.
198 The Association of Ideas If you close one ear with cotton-wool, you will find that your mistakes will be much larger than they are when both ears are open. You have no longer the different intensities of the ticking, as heard by the two ears, to guide you in localising it.
It is always easier to say whether a sound comes from the right or the left, than to say whether it comes from in front or behind. In the first case you have different tactual sensations and different intensities of sound in the two ears to assist your judgment; in the latter you can judge only from the absolute in- tensity of the sound. A sound in front is generally louder than a sound behind, because it is caught by the pinnae, and reflected into the ear-passages. Hence if you tie your ears back, by a piece of tape, and place your two hands in front of the ear-passages, the palm facing backwards, you will find your ordinary judgments of ^ before ' and ' behind ' reversed. The two hands act as two pinnae; but being turned in the opposite direction, they catch sounds coming from behind, and reflect them into the ear-passage, while they cut off sounds coming from in front. What before was loud, and therefore in front, is now weak, and therefore behind, and vice versa.
Our idea of the distance of a sound is accurate only when the source of sound is familiar, when we know by experience how far off the body must be to give rise to the sound which we hear. If the perception is unfamiliar, we may make ludicrous mistakes.
(2) The other form of simultaneous association — a form of extreme importance in the adult consciousness — is word association. The verbal idea contains both exten- sive and qualitative elements: in its most perfect form it consists of an auditory complex, a mixture of clang and noise (word heard), a strain complex due to the adjust- ment of larynx and mouth necessary for the emission of a particular sound (vv^ord spoken), a visual complex, a written or printed form (word seen), and the strain complex due to the adjustment of hand and fingers necessary for the production of this form (word written). The part played by the verbal idea in consciousness, under one or more of these four aspects, is always large, although its actual range differs with different mental constitutions (§ 35).
The verbal idea serves two purposes in simultaneous association. It may arise before associative supplement- ing is at an end. In this case, it aids materially in the supplementing, — sometimes, indeed, rendering all further supplements unnecessary. Or it may arise just as the supplementing is concluded, and clinch the association, putting the seal of finality upon it. In the latter case, it is oftentimes difficult to say whether the process is a simul- taneous or a successive association.
The verbal idea of a given consciousness does not contain all four elements (word heard, spoken, seen and written) in equal proportions. It resembles the note rather than the chord; one constituent predominates in the complex. Sometimes the sound heard is all that comes to mind: more often the word as heard and spoken. If the visual form is aroused, it nearly always brings the auditory idea with it. The writing-complex hardly ever occurs without the visual form, and therefore hardly ever without the audi- tory idea also. In every case, some one of the four components is predominant.
Method. — To test the power of the verbal idea as an aid in associative supplementing, a method may be followed similar to that described in the previous paragraph. Rough drawings, mere hints of the objects which they are intended to represent, are prepared. A word is called out to the subject, and then one of the drawings shown him, — a drawing of something closely related to the object denoted by the word called out. He supplements the drawing, by help of the verbal association, and so ' sees ' a great deal more than is actually upon the cardboard.
Or the original method may be followed still more exactly. Let us suppose that the extent of associative supplementing has 200 The Association of Ideas been determined, by means of a series of misspelled words. Similar series are again presented to the subject; but before each word is shown, a word related to that which the complex of letters is intended to represent is called out. Associative sup- plementing will go much farther than it did before. Without the verbal association, siaye may have been the most mutilated impression which could be read as stage. But if consciousness has been given the right direction by the calling out of the word ' theatre,' a form like aioye may be supplemented to stage.
Instances of the power of the verbal idea to clinch or cement associative supplementing will be readily furnished from the reader's own experience. A striking illustration is that of the recognition (§ 70) of a friend, who has not been seen for some time, and whom one meets unexpectedly. The visual picture is supplemented by a number of ideas (ideas of past meetings, their circumstances, etc.); but the recognition does not become abso- lute and final until the phrase: " Why, it's Brown! " has come to one's lips or mind. It can be shown experimentally that those objects are best remembered and most easily recognised which can be denoted by specific names (§ 73).
Illusions. — Just as we have illusory ideas, ideas which repre- sent an object or process of the physical world in a way which the measuring eye cannot accept, so do we have illusory associa- tions of ideas. A given impression is supplemented, or calls up a verbal idea, under certain conditions; when these are reduced to standard conditions, the association proves to have put an erroneous interpretation upon the impression.
It is frequently impossible to disentangle, with any degree of certainty, the two possible factors in a particular illusion. It may be due to the structure or mode of working of a sense-organ: then we have an illusory perception or idea in the strict meaning of the phrase. But the illusion may also be due to associations. Take, for instance, the two semicircles of Fig. 8. The closed figure may suggest a strung bow, and the open an unstrung bow; and the illusion of their difference may result from this associative supplement, and not directly from eye movement. Or take the apparently simple illusion of the greater height of a square figure. This overestimation of the vertical lines may be due to the diffi- culty of eye movement in the upward direction. But it may also be ascribed to associative supplementing. The square is not broad-based enough to suggest a block of stone lying upon the ground; that which is to give us the idea of rest must be longer than it is high, — must resemble the prostrate figure of a man. The square seems to be striving upwards, to be raising itself, and to be ' holding itself together,' squeezing itself in, in the effort.
Stroboscopic illusions, again, might be occasioned by the per- sistence of after-images, without any associative supplementing of the photographs. But they are greatly assisted if we have a clear idea of the sort of movement which we are going to see, before we look through the sHts in the cylinder.
But while many illusions can be regarded either as illusory ideas or as illusory associations of ideas, there are some which undoubtedly have their source in association alone. Thus, the sun and moon look smaller to us when they are directly above our heads, at the zenith, than when they are in front of us, at the horizon. It is difficult to see any reason for this illusion in the structure or function of the eyes. On the other hand, ( i ) the out- line of the discs is more distinct at the zenith than at the horizon, because there is less air between them and us, and what there is is clearer, less misty and smoky. Hence they seem to be nearer. And (2) there are many objects — trees, houses, hills — between ourselves and the horizon; none between us and the zenith. Again, then, the bodies seem to be nearer. But if a nearer and a remoter object occupy the same space in the field of vision, the former must be smaller than the latter.
The same holds of certain illusions which involve qualitative ideas. The ventriloquist ' throws ' his voice into some inanimate object at a distance from him. To produce the illusion at which he aims, he keeps his lips as far as possible unmoved during articulation, raises or lowers his voice beyond its natural speaking pitch, and looks steadfastly at the object to which he wishes the sounds to be referred. The listener knows that he is being de- ceived; but the illusion may be so complete that it cannot be 202 TJie Association of Ideas wholly destroyed except by the visual perception that the muscles of the performer's throat are twitching, although his lips are still.
Illusions of melody are similarly produced. When we are wait- ing for the passage of a circus procession, we * hear ' the music of the band in the distance many times over, before it actually comes within the range of audition: some chance sound is as- sociatively supplemented, and so takes the form of a familiar melody.
§ 54. Successive Association. — We found only a single difference between the idea or perception and the simul- taneous association of ideas: the difference that the ele- mentary processes contained in the idea had never before been connected with others, while the elementary pro- cesses contained in the simultaneous association had already played a part in some idea. The same difference holds between the idea and the successive association of ideas. But there is a further distinction, which enables us to mark off the successive association both from the idea and from the simultaneous association, — the distinc- tion which is expressed by the word ' successive.' We cannot indicate any stages in the formation of the idea; when certain conditions, physical and mental, are realised, the idea emerges, takes shape at once. Nor can we indi- cate any stages in the formation of the simultaneous asso- ciation; we no sooner hear the noise, than the visual picture of the traction engine comes up; there is no * before ' and ' after ' in the experience. In the successive association, on the other hand, there is a clearly marked division of the whole process into stages; an idea arises, and tJien another idea is connected with it.
There are two principal forms of successive association: the train of ideas and association after disjunction. The former corresponds to the associative supplementing, the latter to the verbal association of § 53.
(i) In the train of ideas we have a continuous series of processes, idea following idea along the line of least mental resistance, without restriction of any kind. Ideas come and go, as they come and go in the ' inattentive ' consciousness of the child or the animal (§ 40); there is no concentration, no converging of tendencies; consciousness is conditioned by the accidents of the moment. This form of successive association appears whenever we fall into a reverie, or grow drowsy, or give ourselves up to the influence of our surroundings, — setting out on a country walk, e.g., with all thoughts of the routine of daily occupa- tion banished from our minds.
Method. — The course which the train of ideas follows in dif- ferent consciousnesses may be tested by experiment. Series of words are prepared, care being taken that the words forming a particular series differ as much as possible in meaning and char- acter; thus, two verbs should not be placed side by side, some substantives should be abstract and some concrete, etc. The words may be printed on cards, which are shown to the subject in succession, or may be merely pronounced by the experimenter; in the former case the time of exposure must be kept constant and must be short, — say, about 2 sec. After each word has been presented, a pause of some 10 sec. is made, during which the subject writes down the ideas which the word has suggested to him, i.e., which have been associated to it in his consciousness in the 10 sec.
The experiments may be made individually, upon a single sub- ject, or collectively, upon a number of individuals. In either case, the results must be carefully analysed, at the conclusion of the series, by experimenter and experimentee. The subject must go over his list of written associations, noting (i) the sense department from which each idea was drawn, (2) the period of 204 The Association of Ideas life to which it belongs, and (3) the idea which suggested it. The experimenter must then, in his turn, work over the list, noting (i) the relative quickness and readiness of association in different individuals, or in the same person under different circum- stances, and (2) the various kinds of association involved, — the association of one general or particular idea to another (co-ordina- tion), of a particular idea to a general (subordination), and of a general idea to a particular (superordination).
Suppose, e.g., that the first word of a printed series was the word horse. One list of associations, within the 10 sec, might read as follows: horse, Prince, heels, stable, strata, cow, dog. The subject would analyse this list, at the conclusion of the series of experiments, somewhat in this way: — (i) horse: auditory idea; present time; suggested by the written horse; a simultaneous association (verbal associa- tion): (2) Prince: mainly visual, idea of a particular horse; childhood; suggested by horse (successive): (3) heels: mainly visual, idea of a particular incident connected with Prince; childhood; suggested by Prince (successive): ^~ (4) stable: visual and olfactory; childhood; suggested by Prince (seemed to arise simultaneously with heels'): (5) straw: visual; childhood; suggested by i'/^/^/*? (successive): [here the train of ideas switched off from the Prince associations, and the original idea (visual and auditory) of horse came to mind:] (6) cow: general idea, auditory (verbal) and visual; no time reference; suggested by hoi^se (successive): (7) ^^S' general idea, auditory (verbal) and visual; no time reference; seemed to be suggested by cow, though pos- sibly due to horse.
From a large number of series, worked over in this way, we can discover how much the different sense-organs contribute to the furnishing of a particular mind with ideas, how observant and retentive the mind is, and how far it is accustomed to pursue a single topic without allowing itself to be interrupted by irrelevant ideas.
The experimenter now takes the same list, and notes that Prince is a particular idea associated to a general idea, ho7-se (subordinate); that heels, in the same way, is subordinate to Pj'ince; and that straw is subordinate to stable. The three ideas of horse, cow, and dog are co-ordinate. The relation of stable to F?'ince is doubtful: the two may be co-ordinate, or stable may be superordinate, — the home of a series of particular horses. By calculating the proportions of the three types of association in a large number of experiments, the experimenter can ascertain the way in which the subject ordinarily thinks, i.e., his intellectual constitution (§ 35). He also notes that in this case there were seven ideas aroused in the 10 sec. {^Prince, heels, stable, straw, horse, cow, dog).
(2) Association after disjunction consists, as its name implies, in the coming together again of ideas which were originally together, but have somehow become separated. The best illustration of this form of successive association is the connection of auditory ideas in the sentence. The whole 'thought,' i.e., complex of ideas, which the sentence expresses must form part of our consciousness, however vaguely, before we begin to speak; otherwise we could not carry the sentence to its conclusion without hesitation and mistake.
The disjunction is due to the attention; the rejoining is a successive association. Suppose that I say to myself: ** That chord contains the notes c, e, g\ " The chord is given as a total impression; it is a complex of simultane- ously sounding tones. But the attention fixes for some reason (§ 38) upon one of the constituent tone complexes, the note c. This is rendered prominent and distinct, while the remaining constituents are blurred and weak- 2o6 The Association of Ideas ened. The impression is thus spUt up, its components dissociated. The attention soon relaxes from its first object, and the other two notes receive, in turn, their share of notice. The whole complex is thus reviewed, part by part, and put together again in the sentence: '' It contains the notes ^, e, g'' The successive association in this and similar instances has the character of completeness or finality {cf. the verbal association of § 53). The * thought ' is complete when a certain number of words have been uttered; the chord is done with when the three constituent notes have been re-associated; the melody is com- plete when a certain number of chords have sounded. The final- ity is a necessary consequence of the fact that the association is based upon a foregoing ' dissociation; the whole is given before its parts are discerned \ the associative process comes to its nat- ural end when the dissociated parts have been put together again. The train of ideas, on the other hand, is absolutely lacking in finality; it never dies a natural death, but must be violently interrupted, if it is to come to a conclusion.
It must not be supposed, however, that the association after disjunction is a mere putting together of what has been pulled apart, of the original raw material. The attention, in singling out some factor in the original complex, renders it more promi- nent and therefore more liable to be associatively supplemented. The parts which are put together again, by way of successive association, are put together only after they have been modified, worked over, by way of simultaneous association. If I swallow a draught of lemonade, and, finding it very sweet, say: " How sweet this stuff is! " I am not simply putting together sweetness and the other constituent lemonade-processes as they were given in the original idea. The sweetness has attracted the attention; it has been disjoined from its surroundings, and supplemented by ideas of the right amount of sweetness, by sensations of nau- sea, etc. We have no longer a single idea, containing the ele- ment * very sweet '; we have two ideas, a successive association §§ 54> 55- Successive Association; Law of Association 207 of ideas, — the original idea having been followed by the worked- over idea of sweetness. In the same way, the statement of a scientific theory is not a simple re-collection of facts which have been presented together, but separately attended to: it is the re-collection of these facts after they have been associatively supplemented, i.e., referred to their conditions.
Psychologically regarded, all instances of judgment fall under this second heading of successive association. Take, e.g., the judgment: "The waste-paper basket is under the table." Here we have an original whole, a visual complex including local ideas of basket and table. The two constituents are disjoined by the attention, and reunited after the idea of position has been made explicit. Or suppose that we walk into a strange village and say: " That must be the hotel! " We have a visual complex, the idea of a certain house, from which the attention dissociates all the hotel-like elements. These are supplemented, and form the hotel idea, which succeeds the original house-hotel complex.
Method. — To test the formation of successive associations of this type, the following plan may be adopted. Show the subject, for a short time, a complex visual impression, — the picture of a street or ceremony or landscape, — an impression, i.e., which is too complicated to be grasped by one pulse of the attention. Let him then write a description of it, trying to reconstruct it as a whole, and putting down his ideas in the order in which they occur to him; that is, in the order in which the various parts of the picture attracted his attention. As he writes, more and more ideas will occur to him; so that the process of reconstruction will take the form of a successive association.
A train of illusory ideas is termed a hallucination. Hallucinations do not occur in the normal mind. We have instances of them in dreaming and in the visual phantasies of alcoholic delirium. Illu- sory judgments are \.^xv(\^A fallacies, when formed in a normal con- sciousness; delusiotis, when they appear as a symptom of insanity.
§ 55. The Law of Association. — The fundamental law of the association of ideas may be stated in almost the same 2o8 The Association of Ideas words as those which we used in accounting for the sin- gleness or unity of the idea. What the organism finds together in the world in which it lives, we said, remains together in perception or idea. But one and the same kind of elementary mental process may be concerned in many different adjustments to physical surroundings, and will therefore have a tendency to connect with processes which form part of many different ideas. This fact is the key to association. )^// the connections set tip betzveen sen- sations by tJie formation of ideas tend to persist, even when the original conditions of connection are no longer fulfilled.
Let us apply this law to the four cases of association which we have described.
(i) Associative Stipplementiftg. — Here we have a complex of sensations, abc, some or all of which have been connected, in past experience, with other elementary processes, xyz. Hence, when- ever ab or abc appears, xyz tends to appear with it.
A rumbling noise, abc, is heard. Our idea of a heavy vehicle includes, as part-processes, the noise, abc, and a complex of visual sensations, xyz. Hence as the noise is heard, the visual complex is aroused also; the noise is supplemented by the other compo- nents of the idea of a heavy vehicle.
Or we have the idea of a clearly outlined hill, abc. Our idea of nearness, xyz, has been connected, in past experience, with the idea of clearness of outline, ab. Hence when we see the hill, abc, we have at once the idea of its nearness: ab\^c'] is supplemented by xyz.
(2) Verbal Association. — Verbal association takes place in precisely the same way as associative supplementing. The only reasons for separating the two processes are, first, that the verbal idea is the most important of all the supplementary ideas, — some- times, indeed, as in the instance of a verbal Mocal sign' (§ 44), rendering all further supplement unnecessary, — and secondly, that the verbal association is on the boundary line between the simultaneous and the successive association, {a) The word is important because it is, so to speak, the common denominator of all ideas alike; words are the medium by which we communi- cate ideas to one another, whatever the ideas may be. Hence the verbal idea is the richest of all ideas in habits of connection; it has the greatest tendency to associate, as well as the greatest range of association. At the same time, it is the word which, as the single expression of a complex of sensations, gives definite- ness or finahty to that complex, {b) That verbal association lies on the border-line between the simultaneous and the successive forms of association is shown by the two instances of the traction engine and the hotel. Had the noise been a little less definite in its suggestion, we might have thought for a moment, and come to the conclusion that it was due to a traction engine (successive association). Had the village hotel been a little more definite in its suggestion, a little more clearly a hotel, the word ' hotel ' might have arisen in our minds as soon as we saw the building (simultaneous association).
We may refer to our first illustration. The noise abc has aroused the visual complex, xyz. The words ' traction engine,' which we may represent by pqr, have been constantly connected with this visual complex. Hence given abc, and we have ahcxyz; given xyz, and we have xyzpqr. Given the noise, and we have visual picture and name of the vehicle.
(3) The Train of Ideas. — This is easily reduced to the same formula. The written word horse is supplemented by the audi- tory idea of horse; abc becomes abcxyz. But there are some elements, x, common to my ideas of horse and of Prince; on the one hand I have abcxyz, on the other, say, xdef. When the former is given, therefore, the latter comes up. But again, there are elements, /, common to my idea of Prince and to my idea of stable; on the one hand I have xdef, on the other,?>:x.y,fgh. When the former is given, therefore, the latter comes up. And so on.
(4) Association after Disjunction. — We have a complex, abed. This is divided up by the attention into ab and cd. The former is supplemented to abxy, the latter to cdjyq. We then have the 2IO The Association of Ideas