Now certain questions arise here which are directly pertinent to our present topic: Is a person motor, visual, or auditory, in his speech, and in his reactions generally, because he has strengthened a particular kind of memories by the prevailing concentration of his attention upon them? Or does he give motor or sensory attention and reaction, because of the predominant strength of a certain class of his memories? Probably both of these positions are true; and each of them is of great importance in the education of speech, and other motor functions, as well as for the theory which I am developing. The case seems to be the exhibition, on a large scale, of what we find to be true of the relation of attention to sensations generally. Increased intensity of sensation tends to draw the attention; and the attention increases the intensity of sensations. It is one of those processes of 'reasoning in a circle' which charac- terize the growth of body and mind together. Another instance is this, for which we have already seen some probable reasons: pleasure arises from healthy function, while healthy function is directly assisted by pleasure.
The relation which we have now discovered, however, be- tween a person's 'type,' and the movements and habits of his attention, is capable of a clear psycho-physical explanation.
We know that increasing intensity of sensation liberates energy increasingly toward the motor centres. A strong 1 To quote my own case again — I find it impossible to think of a French sentence without keeping my attention on the visual picture of the printed signs; but I can follow a German sentence by memories of speech movements with no trace of visual attention.
462 The Origin of Attention, sensation tends to excite more movement than a weak one. It is probable, therefore, that a given degree of in- tensity of each particular sense-quality involves a motor ingredient, as an element in its conscious value — be it in part due to a setting-back process from the motor centres themselves, or in whole to the stirring up of revival pro- cesses in the kinassthetic centres. The distinction be- tween sensory and motor consciousness is largely logical; all consciousness is both. Every sensation reverberates outwards in the muscles, and this muscular resonance reacts back upon the sensory factor. But it is clear that the largest amount of the motor ' ingredient ' attaches to the most intense sensation.
Now we also know that the exercise of attention involves a large amount of motor process; its constant and necessary accompaniments are motor. Consequently the rising tide of motor incitation due to the rising intensity of sensation, is an increasing stimulus to the attention, by a radiation of processes in the centres of movement. So we have a valid reason for the general fact that an increase of intensity of sensation tends to draw and hold the attention.
On the other hand, the ordinary opinion is true, that the idea of a movement is already the beginning of that movement. In the light of this principle it is easy to see that, when I turn my attention to a sensation, I in so far start into more vigorous existence the motor ingredients and associations of that sensation. This in turn tends to bring out more intensely the sensory ingredients, and so the second aspect of our ' reasoning in a circle ' is made clear; i.e., that attention heightens the intensity of sensations.^ ^ After the original publication of the article containing this position, Professor Hoffding, in a private letter, called my attention to the following I The Development of Attention. 463 This process of radiation, or mutual overflow, among the different motor centres — if they be different — is not hypothetical. All theories demand it. It is simply a ques- tion, in any special case, as to how far the circle of influ- ence of one motor process may extend to neighbouring fibres and cells. And if my theory be true, that attention is just the most habitual of all forms of motor reaction — because extending far back in the race history of organic accommodation — then the direct arousing of the attention by changes in mental content is fully explained in the way I suppose.
To put the matter in a nutshell — just in as far as the motor ingredient of a mental content of any kind is large, that is, in as far as the sensory ingredient is intense, just to this degree will the direction of the attention be secured, and to this degree also will both the ingredients be intensified by this act of attention. The two facts, therefore, that intensity draws attention, and attention increases intensity, may be stated in terms of a single principle which I venture to call, in view of the doctrine of association already explained, the ' law of sensori-motor association,' i.e., every mental state is a complex of sensory and motor elements, and any hiflnence wJiich strengtliens the one, tejids to strengthen the other also.
quotation from his Outlines of Psychology (p. 316), which clearly takes the same general ground aS to the cause of heightened intensities when attention is aroused: " It is possible that impulses return from the centres with which the voluntary concentration of consciousness is linked, to the centres of sensuous perception (as in other cases to motor centres), in which way their effect may be strengthened. This would be the physiological form of the psycholoqical fact that an idea becomes clearer if ive give ourselves up to picturing itP (Italics mine.) See also his reference to Wundt {Physiol. Psychologies I, pp. 233 f.).
464 The Origin of Attention, The reflex attention which follows upon increased in- tensity of sensory excitation may be considered, therefore, in conformity with what has already been said, the return wave of revived motor associates; and the increased in- tensity which follows the direction of the attention, is due to the direct influence of this return wave, by the reverse association.^ This principle also goes far to explain the relation to each other of the two so-called laws which are usually stated independently in connection with reaction times: (i) greater intensity of stimulus diminishes the reaction time, and (2) motor reactions are generally shorter than sensory. Both are ready deductions from the 'law of sensori-motor association.' As for the first law, that more intense stimulation gives a shorter reaction than less in- tense, the reason of it is now evident. It is because the more intense stimulus arouses more and stronger motor associates; or, put physiologically, because it has greater dynamogenic effect, and so facilitates motor discharge, both directly into the reacting muscles, and indirectly by its readier influence in getting the attention.
Now as for the second fact, which holds for the majority of people, its explanation also follows. Experiments show that the reaction time is shorter when the signal is fore- known, and the attention is consequently not drawn to it, 1 Wallaschek {^Zeit. fiir FsycJiologie, VII., heft I, March, 1894, p. 67) criticises my view on the ground that only in persons of the motor type — of speech, for example — would there be the necessary 'motor associates.' But this is the reverse mistake to that made by Fere, noticed above, who says that the law of dynamogenesis makes it necessary that all should be 'motors' in type (cf. footnote on p. 436, above). Both fail to distinguish between the general dynamogenic influence of a stimulus, which, by the law of 'sensori- motor association,' implicates the attention, and, on the other hand, the kines- thetic motor images of memory, which represent particular movements.
The Development of Attention. 465 but is left free to seek some further facilitating cue. This cue is found, of course, in persons accustomed to depend upon their motor memories for various voluntary actions, in the thoughts of the movements actually to be made in reacting. And so the ^ motor reaction' is directly prepared for. In these cases, a particular kind of motor associa- tion is emphasized by the direct act of attention. The motor associates are pictured, dwelt upon, emphasized beforehand, the motor centres are put into a state of high potential, the stimulus is left to discriminate itself without attention — and thus the reaction time is shortened. It is evident that in the sensory reaction, part, at least, of the dynamogenic influence of the stimulus goes with the atten- tion, for the discrimination of the signal, etc.; while, in the motor reaction, it all goes into the reaction, which is already prepared for by motor attention. ^ It is an evident corollary, also, that only in persons of the motor type would the motor reaction be shorter than the sensory; for it supposes a ready habit of using motor memories mainly in voluntary movement. Persons trained, however, to use auditory and visual memories as the in- strument of attention, find their reaction time lengthened ^ 1 It is only what we would expect that, when the stimulus (signal) is not intense enough to carry its own discrimination, either the reaction takes place upon a false stimulus, or the attention shifts from the movement, and the time is lengthened.
2 Cases in which the sensory time was shorter than the motor have, in fact, been reported by Cattell {Phil. Stud., VIII., 1892, p. 403), Flournoy {Arch, des Sci. Psy., 1892, No. 10, quoted in Rev. Philos., April, 1893, p. 444), and Baldwin {Medical Pecord, April 15, 1893, p. 455). The explanation given in the text was proposed independently by Flournoy and myself in the papers cited. See my proposed report of results, with discussion of those of Cattell and Flournoy, in The Psychological Review, II., 1895 C Studies from the Princeton Laboratory').
466 The Origin of Attention.
when they come to pay close attention to the movements which they are about to make.
Applying this thought to the rise of speech and its method, we find abundant reason for the variety of types found among adults. Visual, auditory, and motor memo- ries of words date back to early childhood, and do not arise synchronously. Visual pictures of figure arise and get comparatively fixed in childhood some months before the child begins to speak or write, as is shown by its rec- ognition of simple figures, animals, and later, letters. Its auditory images, however, date back still farther; this is seen in the very early recognition of words heard. Special motor memories, on the contrary, are the latest of all. The ability to trace outlines which have been already recognized,^ arises only after considerable progress has been made in speaking, and the progress in speaking is, in turn, relatively much later in its rise than visual and audi- tory recognition. So the probable order in which these different elements of the speech faculty would come under the jurisdiction of the Maw of sensori-motor association' is about this: auditory, visual, speech-motor, hand-motor (writing) memories. And a similar genetic analysis might be made out for other complex activities, if the facts were carefully observed.
This means that auditory and visual memories get a good 'start* on the other varieties in the genetic process. They acquire considerable influence over the attention, which is largely reflex at that early period, and they become in turn relatively easy of revival, before the specific motor memo- ries are well begun. Here is sufficient reason for the existence of auditory and visual speech types. Habits 1 What I have called 'tracery imitation ' above, Chap. V., § I.
The Dcvclopincnt of Attention. 467 thus arise which, on the mental side, express the readiest sensori-motor associations. They amount to what some have called *pre-perceptions,' or better, perhaps, 'pre-ap- perceptions.' On the physical side these habits represent preferential dynamic tensions among those paths of dis- charge whose functions merge, in common, in that of the attention. The law signalized above tends, of course, as life advances, to consolidate these particular sensori-motor couples; and so one particular kind of attention becomes a permanent trait of the mental life, unless the other con- nections, which are subsequently brought into use, be of sufficient strength to supersede that originally most used. This latter, however, may happen in any of several in- stances: either from inherited tendency, or from the strength of other motor habits; or, in course of time, by dint of continued practice in one selected kind of attention.
It would seem, accordingly, that the ' auditory speech ' type should be found most frequently among unliterary people, and among those who have not had extended linguistic training, or large practice in writing and read- ing. The particular influences which are lacking in this type are present in the training which the attention gets in people of the * motor type.' ^ We have now reached, by the psychological and genetic analysis of speech, a result which, it is evident, confirms our general theory of attention. The law of * sensori- motor association ' is a generalization on the side of con- sciousness, from particular cases of dynamogcnesis, each of which shows, on the nervous side, the working of the law of 1 The important educational applications of this topic are reserved for my later volume.
468 The Origin of Attention.
* selection.' It is just by and for this, as we have seen, that attention has developed. It is a reaction of motor character upon sense qualities and mental contents generally, vary- ing in its degree of ease and effectiveness, according to the amount of habit and structural growth. On the other hand, the law of ' selection ' is, if my exposition be true, a generalization of the nervous process by which each of such habits gets started, as representing a new accommo- dation of the arganism to its stimulations.
Closer observation of states of attention also leads us to note some more facts and their explanations. We find on examining consciousness, that attention is not a fixed thing, a faculty, any more than are memory or imagina- tion. Yet in much of the literature of late years, in which the ' faculties ' have been scouted, I know of no author who has applied his own criticisms consistently to the at- tention. Attention is still treated as a constant quan- tity, a fixed thing, the same for all the exercises of it, and for all the contents to which it gives its reaction. Memory, on the contrary, is now known to be a func- tion of the content remembered; and not a faculty which takes up the content and remembers it. So we have no longer one memory, but many: visual, auditory, motor memories. Yet the very same thing is true of attention; we have not one attention^ but many. Attention is a function of content; and it is only as different contents attended to, overlap and repeat one another, that they have somewhat the same function of attention.
It is easy to see, however, why it is that attention has been left largely untouched in the recent reduction of mental functions to changes in content. It is for just the same reason that the notion of self has been left over by 1 The DevclopDiciit of Attention. 469 criticism likewise, as I showed in part above. ^ The reason is a genetic one. It is evident that here, as in many other cases, we have to note the tendency of many sensory stimulations to discharge themselves through common motor channels. The contrast between pleasure and pain tends, of course, to make a great line of division between the motor associates of some contents and those of others; such as that between reaching and withdrawing move- ments. As the senses develop, further divisions arise. But it nevertheless remains true that a balance of motor contraction, reverberation, effort, is common to all con- tents, and so becomes part of the fixed expression of all definite states of consciousness. This fixed grouping of muscular elements is, in its reaction upon the content which arouses it, the fixed element in attention (certain tensions of brow, jaws, skin of head, etc., — the yi element in the formula given above for attention 2); and this makes attention seem to be a faculty of constant value. So it is that certain organic and muscular feelings contribute a certain sameness of value to the sense of self.
But this is not all. The actual content of attention feeling is more thaji half different from sense to sense. We have — i.e., I have — a content so different when I attend to a sound from that when I attend to a light, that it is with the greatest difficulty that I find any strains or stresses in head, body, or limb quite the same in the two. And when we come to the difference between attention to any such sense content and attention to an ideal con- 1 See above, Chap. XI., § 3. The recent chapter of Bradley {Appear- ance and Reality, Chap. IX.) is a remarkable exception to this charge, however.
470 The Origin of Attention.
tent, — even though the latter be the memory of the very same sense-thing, — the whole feeling of attention is again extraordinarily changed. In all these cases the content felt as attention is motor; but it is yet as varied as all the other habitually varied motor responses which have been found useful in the race history of the organism. Its vari- able elements are the a values of our formula.
Very cursory observation of certain animals shows these facts in forms fixed by their varied habits of life. One has only to ride an intelligent horse regularly to be con- vinced not only that most of his mental processes are conducted through his ears, — an effect exaggerated, per- haps, by the ' blinders ' which are put over horses' eyes when in harness, — but also that his attention is auditory. He shows his hopes, fears, expectations, curiosities, etc., by ear movements. In the rabbit and other animals in whom the olfactory lobes are largely developed for pur- poses of utility, a distinct type of memory and attention is probably developed in connection with smell, an olfac- tory type. The constant movements of the tip of the snout in many such animals when exploring for food, etc., by smell, shows the development of delicate smell-motor reflexes analogous to our eye-motor reflexes and the horse's ear-motor. Attention in these cases is probably reactive largely, but for that reason its direct connection with one sense is all the more simple and striking.
Cases from pathology, also, show the actual dependence often of a particular motor function upon the single sense which trained the attention in the learning of this action. Bastian ^ quotes the case of an aphasia patient, who spelt aloud a word wrongly when he wrote it {candd for cat), but 1 Brain as Organ of Mind, pp. 60-62.
The Development of Attention. 471 at the same time, pronounced it correctly, as he heard it. This means that his spelHng movements, letter by letter, had been learned in association with the making of the letters and the sight of them, while the learning of the word pronunciation, as a whole, has been in connection with its sound.
But further still, in the same line. I do not think that we ever — even in successive attentions to the very same thing under the most uniform conditions — have exactly the same attention feeling twice. Why should not atten- tion, like everything else, be subject to the changing effects of habit and accommodation } Indeed, it is the very outcome and exponent of these principles, as I have just been arguing. And then, too, dynamogen- esis, the basis of all the excess energies which are crys- tallized into habits, still works on, and is working on in every attentive reaction which we make. For all these reasons, we see that no two acts of attention can be just the same.^ And the variable element is the a of our formula.
One additional point may be noted here merely; it gets enforcement in the following chapters. We would expect this change in motor reaction content, from act to act of attention, to have some equivalent in consciousness; some equivalent apart from change in the particular content itself which stimulates the attention — some generalized, vague, unanalyzable feelings. And so we have found. Recognition is one such feeling, and Belief is another. I have argued independently over them both — apart from the genetic 1 I think it would not be difficult to test this theory of attention by the dynamogenic method of experiment suggested by Miinsterberg, The Psycho- logical Review, 1894, 441 fF.
472 The Origin of Attention.
aspect of the case — and found them to be just this — felt attitudes toward particular contents. ^ § 4. Voluntary Acquisition and Control.
We are now in a position to see that voluntary move- ment has three distinct stages of development in each individual. We find the mind at first occupied with an objecty presentation, or stimulus, which starts a muscular reaction, either native, acquired, or at random. Then a little later we find the mind occupied with -Bl presentation or idea of the moveiuent thus made, which, with its associates, tends to stimulate the corresponding motor processes, and thus to brinsf about the same movement. And at last we o find the mind occupied with an object again, for the attain- ment of which the movement is a necessary but now a subconscious means.
The original ' end ' of volition, therefore, is simply the image or picture which starts the imitative reaction. Sug- gestion turns out to be an original motor stimulus in voli- tion, as truly as in the lower activities. The child attempts to speak, for example, with no attention to his organs of 1 Chap. X., § 3. On Belief, see my Handbook, II., Chap. VII.; the genetic theory of belief is reserved for the later volume of ' Interpretations.' The doctrine of Recognition, based on the law of ' sensori-motor association,' was published in the Philos. Review, July, 1893. Professor Hoffding, in a pri- vate communication, makes the criticism that, on my view, we would con- fuse two qualities which had been repeated the same number of times. This would mean that we have no differences of attention for the different sense qualities. But it is evident that that is not true, if I am right in saying that the actual motor content is different for each quality, and that we so have different attentioiis., just as we have different memories, etc. His criticism shows — what I said above — that even the best psychologists still look upon attention as a relatively fixed ' faculty,' rather than as a shifting function of content.
Vohmtary Acquisitioii and Coulrol. 473 speech. He then learns that it is by muscular effort, by persistent imitation, that he must proceed. Accordingly, the muscular movement now becomes his end. He strains to set his vocal organs properly. His efforts to control the organs, however, throw him, at first, into great confusion and failure. But after more muscular control is acquired, the third stage gradually follows, as the movements become habitual. The end is now again a picture or object, and the muscular consciousness falls into the background, as, for example, in our developed adult speech, when we think only of the ideas which we wish to express.
The theory of motor development now worked out throws much light also on the whole vexed question of muscular control — the regulation of movement in amount and direction, and its suppression, etc. It is easy to see that the material of volition, the ideas or copies attended to and imitated, are the means of holding the course of each movement in check by association. I can repeat a movement only because I am able to reinstate in memory, the feeling of it, the copy elements of it. But by asso- ciation, as we have seen, other elements, such as visual, or auditory, or touch, memories, may stand for the mus- cular memories. The whole management of a movement, therefore, depends upon the getting hold by the attention of the series of positions desired for the limb moved, and this can be done only by filling up the attention with the proper copy elements of sight, hearing, or other, which release the proper series of motor discharges, and these discharges only. And negative control, or inhibition, rep- resents, in general, the limitations which old organic ways of action impose upon new ways: the new must conform, if possible, to old organic * copy.* 474 ^^^^ Origin of Attention.
The current theory of * control ' lends itself directly to this view, hinging, as it does, upon the matching, term by term, of the movements being accomplished with a re- membered series, whether of sight, sound, or what not. The control of handwriting described above is a good instance.^ The current theory, however, lacks all account of the process by which the series to be matched is vividly held up for voluntary imitation.
This lack we are now able to supply. The view of atten- tion given in what precedes, teaches us that the motor reaction of attention is a function of the content attended to, on the one hand; but, on the other hand, it is a part of the motor process in which the whole content finds its dynamogenic expression. The office of attention, there- fore, is that of fixing the content steadily, on the sensory side, and at the same time of releasing the associated discharge movements, on the motor side. Attention has, in each case, as we have seen, grown uj) in exactly this way, both as an expression of motor reverberation from typical and constant accommodations, and also as itself the very beginning, by the law of ' excess,' of the useful discharges which, in their acquisition, got to be associated with the content in question. ) Attention is the go-between between the copy imi- tated, and the imitation which copies it. It is, therefore, the central and essential fact in all voluntary muscular control. I This theory is so important in the sequel, and so instruc- tive in its applications, that I take space for its summary statement here. Its development is not necessary to the clear statement of our general evolution theory. I shall Vohiutary Acquisition and Control. 475 accordingly return to it in detail in the later volume, which deals with applications and inferences.^ 1 I intimated this theory of control in my article in the Philosophical Re- vieiv, II., p. 406, from which I may quote: " The correlation of various images in the attention, through their respective * motor ingredients,' is necessary for voluntary activity; and where a particular class of images is lost, the damage it works in the mental life is not alone the narrowing of the content of consciousness, but it is in many cases the withdrawing of that sup- port, without which the voluntary function can not proceed at all. It is the coordination of the attention, therefore, — what I have elsewhere called ' volitional apperception,' — that every one of the incoming sensory elements must have part, at least, of its regulating effect upon the efferent discharge. This is shown so clearly, as a matter of fact, in the elaborate article by Pick on the loss of voluntary movement by certain anaesthetics when the eyes or ears are closed {^Die sogenawite ' conscience inusculaire^ Zeitsch. fUr Psych., IV., 1892, 161 ff.), that I need not do more than recognize the support which my article gets from his. A collection of cases which show the extreme depend- ence of attention and voluntary movement, in persons of the visual type, upon