The question can be answered only by an appeal to ex- periment. For it is only under experimental conditions that we can keep the object of attention absolutely simple, and so far rule out disturbing influences. Many experi- ments have been made, and all have led to the same result: that attention is not persistent, but intermittent, — rising and falling, waxing and waning, at quite short intervals. If we attend as closely as we can to a simple sense-impression, its quality is not made permanently clearer and more distinct; it becomes alternately clear and blurred, distinct and indistinct. The attention fluctu- ates.
The fluctuations of attention are usually irregular. The time of a single ' pulse ' of the attention — from distinct- ness to distinctness, or from indistinctness to indistinctness of the impression — has been found to vary, in the case of weak stimuli, between the limits of 6 and 24 sec.
Method. — It is best to work with the weakest possible, i.e., with just noticeable stimuli. For if any blurring or indistinctness occurs in a sensation which, at its best, is only just noticeable, it is plain that the sensation will disappear altogether. Waxing and waning of the attention will then mean appearance and disappear- ance of the sensation; and it is far easier to say that we do or do not see or hear something than to say that what we see or hear has grown more or less clear or distinct.
Paint a very light grey circle upon a square of white cardboard, and place the card so far from the eye that the grey is only just distinguishable, only just noticeably different from the white. Look steadily at the circle. It will be visible for a few moments; then the white of the card will seem to wash over it; then it will appear again; then disappear, and so on. Let an assistant hold a stop-watch. Each time that the grey becomes clear, tap the table with a pencil: the assistant will note and record the intervals between tap and tap. When you have accustomed yourself to the experiment, you may tap the table not only when the grey ap- pears, but also when it disappears, and compare the length of time during which attention is sustained with the length of time during which it is relaxed. The former time will probably be the longer.
The experiment may be repeated with a faint noise as stimulus — say, the tick of a watch removed so far from the ear that its sound is only just audible. In this case you must signal to the assistant by some different means, as the tap on the table would interfere with your attention to the watch. You might, e.g., have a string attached to your forefinger and to his wrist, and pull upon it when the ticking appeared and disappeared.
Since you are working with minimal stimuh, you must be care- 142 Conation and Atteiition ful to have the sense-organ exactly adjusted to the impression. A chance movement of the eye or a sHght turn of the head would cause the grey circle or the watch-tick to disappear, quite in- dependently of the attention. The duration of the attention can be inferred only from the behaviour of the sensation under abso- lutely constant bodily conditions.
It is to be noted that attention to centrally aroused sensations — e.g., an imagined watch-tick — is subject to the same fluctua- tions as attention to external stimuH.
The explanation of the intermittence of attention is to be sought in the nature of the physiological processes (pro- cesses of cortical reinforcement and inhibition) which cor- respond to it. When a nerve-cell acts upon another nerve- cell, it does this not gradually or continuously, as if in- fluence ' flowed ' from it, but suddenly and at once, as if the influence were 'discharged.' Indeed, physiologists speak always of the * discharge ' or ' explosion ' of a nerve- cell, when they refer to its exercise of function. We must imagine, therefore, that when the frontal lobes re- inforce or inhibit an excitation, they act by jerks: there is a jet or spurt of energy from them, — then a brief pause, during which they recuperate, lay up more energy, — then another jet, and so on. In the pauses between jet and jet, the cortex is at the mercy of other impressions than that which is the special object of attention: there is a relaxation of the attention, and the ideas correspond- ing to these other impressions take the place of the idea attended to.
* But whence do these impressions come 1 We are working under experimental conditions, and have ruled disturbances out.' We have ruled out disturbances, so far as we can. But we cannot rule out the rustling of our clothes which comes with the rise and fall of the chest in breathing, the noise of heart-beat, the pressure from the twitching of a muscle, the tickling of a hair, the tingling of the skin, etc. Still less can we rule out disturbing central excitations. Memories and imaginings of all sorts start up, — the counterparts of some chance excitation in a cortical centre, — and the attention is diverted from the given impression before we know that there is anything present to divert it.
The great importance of these distracting impressions is vouched for by introspection. Attend to the grey circle or watch-tick as before; but interrupt the experiment, as soon as the impression has disappeared, to ask yourself introspectively the reason for its disappearance. It may have vanished because your eye ' sHpped,' or your collar creaked. In that case try again. You will find that the impression disappears, however favourable the conditions for seeing or hearing; and that it disappears when the attention has been distracted by the ^ cropping up ' of some irrelevant idea.
Introspection cannot, however, show why it is that the irrele- vant ideas are allowed to crop up. The reason for that lies in the nature of the physiological processes which constitute the condition of attention.
It is natural to suppose that, if we could secure really constant conditions of experimentation, internal and external, we should find the fluctuation of attention to be regular. Under such conditions, discharge and reloading of frontal lobe cells should succeed each other with perfect regularity. The conditions are, unfortunately, almost impossible of realisation, — impossible, un- less a happy chance assist our efforts to regulate circumstances. It has been found, however, in experiments made with all con- ceivable caution, that the fluctuations may be regular; in these experiments, the time occupied by a single pulse of the attention was 3.5 sec.^ 1 It may seem strange that while the average pulse of the attention lasts from 6 to 24 sec, pulses so short as 3.5 sec. should be obtained under the most 144 Conation arid Attention This result agrees with a fact of general psychological experi- ence: that the signal for any experiment is best given to the observer some 1.5 to 2 sec. before the experiment is made. The interval allows the attention to come to its full strain, but ends before relaxation has begun. It is surely significant that it is just half of the fluctuation-time mentioned above.
§ 42. The Range of Attention. — Our problem is to determine how many impressions can be attended to together, without diminution of the clearness, intensity, etc., to which each one of them would attain if the atten- tion were directed to it singly. It is a common saying that ' nobody can attend to two things at once '; but experiment shows that the truth of the statement de- pends upon what the ' things ' are, whether complex or simple impressions; and, if they are complex, upon the degree of their complexity.
We can approach the problem in two ways, — by a si- multaneous and a successive method. We may present a number of stimuli to a sense-organ at the same time, gradually adding to them, until it becomes impossible to attend to all at once. This procedure is the best for visual and cutaneous stimuli (lines, letters, circles, bands of colour, etc., laid upon the same background; or simul- taneous pressures at different parts of the body). Oi: we may give the stimuli in succession, gradually increasing their number till the point is reached at which the first disappears from consciousness as the last is given. This favourable circumstances. It is probable that the pulse is of different duration in the case of different individuals. It may be, however, that the results which give 6 to 24 sec. are not wholly trustworthy; that the stimulus was not mini- mal throughout the experiments, and that accordingly one or more blurrings or fadings of the impression were overlooked by the observers. New investi- gations must be made, before the question can be finally settled.
§ 42. The Range of Attention 145 method answers best with auditory stimuli {e.g., beats of a metronome). In both cases the object of enquiry is the same: we wish to determine the limit of complexity at which the attention becomes unable to cope with the stimuli offered to a given sense-organ.
It has been found by the former method that four or five simple visual stimuli can be presented together, with- out distraction of the attention from any one, i.e., without diminution of its clearness, intensity, etc.; while the sec- ond method leads to the conclusion that a series of eight auditory impressions can just be grasped by the atten- tion.
Method: ( i ) Simultaneous stimuli. — Prepare a series of white cards, upon which are printed letters, lines, etc., in gradually in- creasing numbers. One of the cards must be set up, in each experiment, at a convenient distance from the eye of the ob- server. In front of it is an apparatus resembling the instantaneous shutter of a photographic camera. When this apparatus is set in action, the card becomes visible for a fraction of a second. The time of exposure must be very short, since otherwise the eye may sweep rapidly over the impressions, leaving some to be remem- bered while the others are directly attended to: in this case they would not be apprehended by ' an ' attention, but by a series of attentions. The card, too, must be so small as to be easily * taken in ' by the eye at a glance, without eye-movement. Otherwise we may be measuring not the extent of the ' field ' of attention, but that of the field of vision. Thus, if the field of vision were filled out with broad bands of colour, we might reach its hmits before we had reached the limits of the grasp of the attention; three bands, of red, green and blue, might fill the visual field, while the attention, as we have said, is able to grasp four or five simultaneously presented simple visual impressions, — colours, letters, lines, etc.
If letters are employed as stimuli, they must form a meaning- L 146 Conation and Attention less series, such as RKZT. It has been found that a familiar word of four letters can be apprehended by the attention as if it were a single letter; it is attended to, not as a series of letters, but as one total impression (Ch. VII). For the same reason, the attention can deal better with figures than with disconnected letters. Any combination of figures ' makes sense,' represents a definite number: 4321 means something, just as much as 1234.
(2) Successive Stimuli. — The running weight upon the tongue of a metronome is so placed that the interval between beat and beat is about a quarter of a second. The experimenter marks off groups of beats by sounding a bell simultaneously with the first beat of each group. Two series are given in each experiment: beat ) beat ) and the observer is required to say whether the two groups are equal or unequal. He must not count, of course; counting would mean that attention was given to each beat separately, and, there- fore, that the series was apprehended by successive attentions, and not by * an ' attention. Accurate judgment is impossible in the case of series which consist of more than eight impressions.
But just as in the previous experiments a word of four letters was equivalent, for the attention, to a single letter, so here a group of impressions may be equivalent, for the attention, to a single impression. We have already referred to the influence of rhythm in judgments passed upon the relations of auditory stimuli (§ 29). Now when we listen to our series of metronome beats, it is impos- sible to avoid throwing them into a more or less complex rhythm. If the 8 impressions which constitute an experimental series are single beats, they are apprehended not as 8 but as 4 {beat beat, beat beat. If eat beat, t?eat beat; not beat, beat, beat, beat, beat, beat, beat, beat); so that, for the attention, they are 4 impres- sions. The limits of the grasp of attention lie between 8 impres- sions of 2 beats each (16 beats in all) and 5 impressions of 8 beats each (40 in all). The 8 beats in the latter case are broken up into 4 pairs, accented as trochees {cf. § 47).
The range of affection we found to be coextensive with con- sciousness. It is evident that the same cannot be said of the § 42. TJie Range of Attention 147 range of attention. The reason for this apparent anomaly is to be looked for in the fact (§ 2)Z) ^^^^ ^^ affection attaching to a single idea or group of ideas radiates from this to the other ideas of which consciousness is composed, even if they are in them- selves indifferent. Attention does not radiate from the objects attended to. — Cf. what is said of the composition of the feeling, PART II CHAPTER VII Perception and Idea § 43. Sensation, Perception and Idea. — We have hitherto used the terms 'perception' and 'idea' indifferently, to sig- nify a complex of sensations; and we have implied that such a complex process becomes, under certain conditions, a single item of mental experience, forms a coherent whole, — so that we can speak of its intensity, duration, etc., quite apart from the intensity or duration of the element- ary processes which enter into it. We must now ask how these complexes are formed; which of the four attributes of sensation are of the greatest importance for their production; and under what circumstances they acquire their unity or singleness for mental experience.
There is no fundamental psychological difference between the perception and the idea. It is customary to speak of ' percep- tion ' when the majority of the simple processes in the complex are the result of stimulation of a sense-organ, i.e., are peripherally aroused, and of ' idea ' when the greater number are the result of an excitation within the brain cortex, i.e., are centrally aroused. If I have a table before me, and my eyes open, I am said to ' per- ceive ' the table; if I close my eyes, and think of what I saw, to have an ' idea ' of the table. But we have seen that the sensations aroused centrally do not differ as psychological processes from those aroused peripherally (§ 7). Hence although we might be tempted for convenience' sake to follow^ the common usage, — to employ ' perception ' to denote what is now before us, and '■ idea ' to denote what is remembered or imagined, — we should be obhged constantly to remind ourselves that, in principle, the two processes are one and the same. And the danger of forget- ting this would far outweigh the convenience of separating the terms.
In what follows, therefore, as in what has preceded, we shall use the words indiscriminately.
We classified sensations in the first place by reference to the sense-organs from v^hich they proceed, and second- arily by reference to the stimuli w^hich arouse them. We might now classify ideas in the same way, beginning with the great groups originated in a sense department (visual, auditory, olfactory, etc.), and subdividing these by the help of differences of stimulation within a department (ideas of colour, of brightness, of tone, of noise, etc.). But such a classification would be misleading. The sense- organs are, as a matter of fact, not separate instruments: they are instruments in the service of a single organism, and they are connected with one another, by way of the brain. So long as we are enquiring into the nature and number of the elementary conscious processes (§ 4), we may regard each group of sensations as separate and independent, and each member of a group as an individual process, possessed of its own attributes. But when we come to consider sensations as elements in ideas, we find, naturally enough, but little show of independence and individuality. The particular sensation, regarded apart from other sensations, is the product of scientific analysis, an abstraction from actual mental experience: the simplest item oj that experience is the idea. It was necessary for us, as psychologists, to see how the sense-organs would work if they were working separately. That donCj how- 150 Perception and Idea ever, we must go on to enquire how they really do work together for the benefit of the organism.
We never have, then, a perfectly simple mental experi- ence: consciousness is never composed of a single sensa- tion. Two points may be noticed, (i) On the one hand, several sensations, from different sense departments, may be combined into one idea. The contributions made by a particular sense department will, it is true, be predomi- nant in the idea; but the character of the whole process will nevertheless depend upon all the contributions sent in from the different departments concerned. My idea of lemonade is predominantly an idea of taste. But taste alone could not give me an idea of lemonade; there must be added to the taste qualities, sweet and acid, a pressure, a scent, a colour, a movement of gas-bubbles, etc. My idea of an arm-chair is predominantly visual, a picture of the chair; but it contains also the idea of softness, of the sitting position, etc., — elements of movement and press- ure. (2) On the other hand, not every sensation is called upon to assist, in equal measure, in the formation of every kind of idea. There is a division of labour. Thus visual sensations, which have the attribute of extent, are pre- eminently concerned in the formation of extensive (spatial) ideas; auditory sensations, which have no spatial attribute, contribute nothing directly to our ideas of space. We ' see ' how far off a thing is, in what direction it lies, how large it is, what form it has, etc. Auditory sensations, however, possess a well-marked duration; they begin abruptly and cease abruptly with the beginning and cessa- tion of stimulus; there is but little auditory after-image. This fact, in connection with the impossibility of their spatial arrangement, gives them an especial fitness to § 43- Sensatioiiy Perception and Idea 151 arouse temporal ideas, ideas of frequency, succession, rhythm, etc. In all such cases we have the elevation of one attribute of sensation at the expense of others; in the cases quoted, quality, the core or * self ' of the sensation, becomes subordinate to extent or duration in the idea. In others, quality may be the predominant attribute.
We shall classify ideas, for the purposes of the present chapter, as extensive, tejnporal and qualitative. And we may confine ourselves to the consideration of those ideas which are built up from sensations of pressure (cutane- ous, articular and muscular), of tone and of brightness. Pressure gives us all three classes of ideas in their earliest, most rudimentary form: the eye and the ear furnish the same ideas at their highest level of development.
The two primitive sense qualities are, in all probability, those of pressure and pain (§ 21). Pain, from its very nature, has but a small part to play in the formation of ideas. Its appearance is an indication that some sense-organ is being damaged, and it is always unpleasant. Hence a consciousness composed of pain ideas could accompany only a pathological bodily state, — a state of localised injury and general nervous deterioration, a state in which catabolic processes had the upper hand in a particular organ and in the nervous system generally. If this bodily state and this consciousness were of frequent occurrence, the organ- ism's life would be short.^ Pressure, on the other hand, may be expected to form the foundation for all classes of ideas. It is a primitive sensation, 1 It may be objected that invalids whose life is a continual pain often live to a good old age. But it must be remembered that they are cared for in a way which is unknown to the lower animals; that their pain is mitigated by medical treatment; and that they are capable of looking forward to recovery (§ 40), while the animal by its very constitution cannot anticipate the future. *' While there's life, there's hope " holds only of mankind, because mankind alone can form a conscious plan of life; while, on the other hand, the fact that hope is possible robs pain of a part of its destructiveness.
152 Perception and Idea the first material out of which an idea can be shaped. It pos- sesses all four sensation attributes: quality, intensity, extent, and duration. And its quality is common to several great groups of sensory nerves, — nerves of skin, mucous membrane, muscle and joint.
We find, accordingly, that tactual ideas — ideas built up from pressure sensations, cutaneous and organic — are of all three kinds: extensive, temporal and qualitative. Since it is endowed with the spatial attribute of extent, pressure can naturally serve as the basis of the various spatial ideas: ideas of size, direction, form, position, etc. Since it is the quality aroused by movement of a limb, by friction of the articular surfaces against each other, it can serve as the basis of temporal ideas: ideas of rhythm, of rapidity of movement, etc. And though it is not qualitatively variable, though, i.e., it remains the same 'pressure' whether it proceed from muscle or joint or skin, it blends with other quali- ties from other sense departments to form qualitative ideas: with organic sensations to form ideas of hardness, resistance, etc. (§ 16), with taste sensations to form ideas of astringency, pun- gency, etc. (§15).
Vision and audition, the senses which are richest in sensation qualities, may also be expected to give rise to a great variety of ideas. These senses stand at the other extreme of the de- velopmental series from that occupied by pressure and pain; they are the highest products of mental evolution in the sphere of sense. Visual and auditory ideas are cast in the same mould as tactual, formed in the same way and used for the same general purposes. But they are more ' finished ' and at the same time more comprehensive. Whenever a simultaneous appeal is made to the two groups, the final decision rests with vision and audition: we estimate size by look, and not by ' feel '; we take our rhythm in dancing from the music rather than from the sensation com- plexes set up by bodily movement.
Having considered the ideas formed from the most simple and the most highly differentiated sense materials, we shall have no need to consider any others. No sensations, except those of eye and skin, have the spatial attribute of extent. No sensations, except those of hearing and pressure, possess a well-marked and clear-cut duration. No other sense, not even that of smell, is so rich in qualities as are vision and audition. Hence when we have discussed our three groups of ideas in these three departments, we shall have given an outline of the formation of ideas in general.^ 'But what of intensive ideas?' it may be asked. ' If the attri- butes of quality, duration and extent form the nucleus round which certain ideas gather, why cannot intensity serve as the nucleus of certain other ideas? ' The answer is to be found partly in the nature of the attributes themselves, partly in the adjustment of the organic functions to the needs of practical life.
Quality, we have said, is the absolute and individual attribute of sensation; the others are relative or comparative, common to all sensations alike (§ 26). Quality, then, will naturally stand alone; qualitative ideas are a matter of course. My idea of lemonade is an idea built up from qualities of sensation; it does not matter how long those qualities last, or how much lemonade there is, or into how wide a glass it is poured. My perception of a musical chord is qualitative, again; duration and intensity do not occur to me, as I listen to it, — or, if they occur, are entirely subordinate to the quality of the total impression. Lemonade and the chord ^-^-^ are, first of all, themselves (§ 8); they are not so much of something, but something, different from other things.
But the ' how large ' and ' how long ' of things are often im- portant. Hence extent and duration are made absolute, by reference to an arbitrarily selected unit, - — centimetre, second, — for the purposes of everyday life. We must know at what hour a train goes, how many go in the course of a day, at what rate they run, etc. We must know how many miles it is to the next town,