Studies from the Psychological Laboratory
of the University of Chicago:
1. Reaction-Time: A study of Attention and Habit
James R. Angell and Addison W. Moore
assisted by J. J. Jegi
It is not without grounds that experimentation upon reaction time has been called the Lieblingsgegenstand of experimental psychology. The facts appear so simple and the interpretation so illusive that ingenuity has seemed piqued anew each time the matter has been opened. The fact has had its interest recently augmented by Professor James Mark Baldwin's challenge of the results and explanations which have hitherto passed current stamped with the mark of the justly revered Leipzig school. It must often have occurred to many readers following the Leipzig explanation of the alleged fact that the time of the so-called 'motor form' of reaction is faster than the 'sensory' to ask, why a brain-reflex should be established in the former case and not in the latter, or why, if established, it should be so much less effective in reducing the time consumed by the reaction; why occasionally, despite the assertion that such persons did not possess the necessary Anlage and so could not be regarded, some persons proved unable to make any distinction whatever in the forms or even showed faster time in the sensory attitude. The fact that these and other questions were left open by the Wundtian explanations appeared to leave room for further investigations.
We set out with the general conception that from the evi-
(246)-dence already in hand it was to be anticipated that each individual mind would, from influences already surrounding its growth, show itself possessed of certain coordinations which were customarily employed in the everyday business of life, and that these coordinations would afford pathways peculiarly pervious to rapid nervous discharges, i.e., they would form paths of least resistance; whereas certain other coordinations would be either wholly lacking or much less practiced and much more difficult of employment, yielding when actually pressed into service much slower results. Working under this general conception we had reached results in our experimentation very similar to those of Professor Baldwin, and we were just ready to publish when his very notable article upon the subject appeared in this Review of May, '95, showing essentially the same results as we had reached. Not only had Professor Baldwin anticipated our results up to that time; he had also anticipated almost completely our mode of procedure. This full acknowledgment of his priority is due him on every score. Although our time results have continued to confirm those of Professor Baldwin, yet as the investigation proceeded a standpoint of interpretation emerged, differing in some essential respects as much from Professor Baldwin's as from that of the Wundtian school. On the other hand, the interpretation here given on the basis of the interrelation of habit and attention seems to us to combine and reconcile some of the principal contentions of both sides of the 'type' discussion.[1] The explanation we have attempted is 'dynamo-genetic' rather than static as most interpretations appear to us very largely to have been.
The experiments were begun in March, 1895, and have continued, with a two months' summer intermission, up to the present time. While some differentiations have been made, both in the stimulus and the mode of response, the results of these are submitted not so much to establish the characteristics of these differentiations themselves as to furnish cumulative evidence of
(247) the general nature of the reaction, and perhaps point the way for more detailed research in the future.
The reagents of this series were the persons whose names appear at the head of the report. J., as indicated by the smaller number of his reactions in some of the series, came into the work recently; A. had taken part in no reactions for several years; M. and J. were entirely unpracticed.
The time was taken with the Hipp chronoscope. The clock was tested at each hour's work, and frequently twice during the hour, by a falling screen whose time was taken from a 1,000 V. König fork. The variable error of the machine for the whole series was .0004 sec.
We may hope to escape the recriminations generally hurled at all users of the Hipp chronoscope, inasmuch as the significance of the figures rests but little upon their time values. As will appear, the essential question is whether certain groups of reaction-time approach or recede from each other. As a matter of fact, however, the accuracy of our instrument permits much more stringent conclusions. But this is all we are immediately concerned in.
The experiments include responses by the hand, foot, and lips, to auditory and visual stimulations in both 'sensory' and 'motor' forms. In the auditory series a large number of the hand and foot reactions were further differentiated into those made in the light and those taken in the dark. The auditory stimulation was given in most of the series through a telephone. The visual stimulation was the movement, from a stationary position, of a black screen with a white center.
The hand responses were given by pressing downward with the first finger of the right hand. The foot reactions were made by downward pressure of the toe of the right foot, the foot being supported under the instep to prevent fatigue and complicating strain. The lip responses were given with a special key, the reaction being made by parting the lips.
Most of the visual series were not begun until after the auditory series were completed, and in the latter most of the lip and foot reactions were taken after the hand series had been finished. As the modes of response were the same in both the
(248) auditory and the visual series the reactions in the latter had the benefit of the practice secured in the former. This observation is of importance in considerations where the effects of practice are to be taken into account. It also accounts, in some measure, for the time of the visual series being faster than is usually reported. Indeed, if the mean variation had not in many of these cases remained so small. suspicion must surely have arisen as to whether the conditions desired were really being attained, but with small variations (6 to I5%) and clear distinctions in the sensory and motor forms, the figures appear trustworthy. Since the hand responses to sound were taken first, far outnumber the others, and were distributed over a much longer period of time, they are of most value in showing the development under practice.
While the reactions were not all taken at the same hour of day, each period of work was divided between the sensory and motor forms. So that, throughout the course as a whole, each sensory reaction is balanced by a motor under parallel conditions. The number of reactions taken under each mode of response, hand, foot and lips, was about equally distributed between the sensory and motor series. Beside the differentiations above mentioned, several minor ones, such as alterations -in the intensity and location of the stimuli, changes in position of body, response with the left instead of the right hand, etc., were made in the course of the work. These variations, to which reference will be made later, simply served to emphasize the importance of habit as a factor in attention.
Effects of Light and Darkness
Most of the reactions in the light were taken with the eyes resting upon the responding organ, save of course in the case of reactions to visual stimulus. In its external aspects this form of reaction corresponds to what Prof. Baldwin calls the 'visual motor,' as distinguished from the 'kinaesthetic motor' reaction. [2] In the latter attention is focussed upon the thought of the movement, the responding organ not being seen. He further says:
(249) "In subjects of the motor type the 'kinæsthetic motor' is shorter --the 'visual motor' time approximating the sensory reaction time." Taking Prof. Baldwin's definition of 'kinæsthetic motor' in terms of attention, as that form in which the attention is occupied with 'the thought of the movement,' we found the same results, viz.: that in reagents of the 'motor type' it was the fastest form. But the external conditions of this location of attention we found just the reverse of those stated by Prof. Baldwin. Both A. and J. found that the attention could be more completely centered in the reacting organ when the latter was visible. When blindfolded or in a dark room, there was a tendency toward the sensory form; and the time of the reaction was between that of the motor in the light, and the sensory. A.'s and J.'s sensory form was also retarded in the dark. M.'s sensory form was faster in the dark. But the motor was slower and tended very strongly to pass over into the sensory.
Effects of Practice
At the outset of the course, it was found in the hand reactions to sound that A. and J. returned the faster time for the motor reactions, and M. for the sensory. The first attempts to react in both ways showed in all the reagents -- especially in the cases of M. and J., who were entirely unpracticed, very little difference, save an occasional big jump or total failure to react. This taken with the testimony of introspection showed that most of the reactions still came in the habitual way, the other form of reaction not yet having emerged. Then, as the new form began vaguely to define itself, there arose a large time difference between the two series, and a large mean variation in the series of the new form. At this stage, in the attempts at the new coördination, in spite of the reagent's best effort, attention frequently jumped back into the habitual form. With further practice, however, the confusion began to disappear, the new form coming out clear, with its time and mean variation diminishing. Meanwhile the time of the old form also kept diminishing, but did not make such rapid progress in reduction as did the new one. This continued until, at the close of some of the
(250) series, the difference between the two forms was inside the mean variation for each series with perhaps a slight final balance in favor of the motor as the faster time. Thus both A. and M., who had the longer practice in the auditory-hand series, and who started in the one motor, the other sensory, came out at the same point, relative to the two series, i. c., both were a little faster in the motor form. J. who came into the work recently does not show this outcome so clearly, though in some of his series the approach of the two times is clearly marked. Meanwhile the decreasing time and continued approximation of the times of the two series were accompanied in each series by an ever increasing degree of reflexness. At this stage any extraordinary effort to concentrate on either form resulted in a confused and lengthened reaction.
To sum up the steps in the development, we have (I) different habitual forms of attention at the outset; (2) a period of confusion and wide time difference in evolving the new form; (3) a subsequent reduction of absolute time and mean variations in both forms; (4) an approximation of the time values of the two forms; with (5) a final possibly shorter time for the motor form. In short, to generalize these steps, the conclusion, to which the whole series points, is that continued practice in the two modes of coördination with a constant stimulus, under constant conditions, results in two highly reflexive forms, not of widely different, but of about equal times values. This result receives further confirmation from Professor Baldwin's report of his own case [3] in which he says his reactions "have only changed in that the distinction between the sensory and motor time is less marked than it used to be, and this I explain as probably due to habit and practice, as my theory again seems at least not to contradict."
On the whole, the outcome seems to agree with some of the results on both sides of the 'type' discussion. It indicates with Professor Baldwin's results that " the ground of origin of types is to be found in education, which must necessarily apply to single functions;" that, as so defined, in the sensory ' type' the sensory form of the reaction may be shorter than the motor even after the latter has clearly emerged in consciousness. That
(251) continued practice does not tend to widen the time difference at first manifest between the two forms, but on the contrary. On the other hand, it finds that when both forms have reached a high degree of reflexness, the motor form is probably somewhat faster, though not to the extent reported in the Leipzig results.
The following table shows the effects of practice in the time differences between the first and last thirds of each sensory and motor series. In all, except the hand series of A. and M., the
Table showing the Results of Practice
| Sen. Organ. | Mot. Organ. | Focus of Atten. | Reagents | ||||||||
| A. | M. | J. | |||||||||
| No. | Time in s. | No. | Time in s. | No. | Time in s. | ||||||
| First Third | Last Third | First Third | Last Third | First Third | Last Third | ||||||
| Ear. | Hand | Sen. Mot. |
560 540 |
195 149 |
133 127 |
420 380 |
163 178 |
132 134 |
160 165 |
185 169 |
173 159 |
| Foot | Sen. Mot. |
145 155 |
182 159 |
168 150 |
220 230 |
138 145 |
133 134 |
125 115 |
218 204 |
208 196 |
|
| Lips | Sen. Mot. |
130 120 |
132 125 |
122 116 |
125 125 |
117 112 |
108 106 |
160 140 |
169 157 |
155 146 |
|
| Eye. | Hand | Sen. Mot. |
100 100 |
206 193 |
173 150 |
100 100 |
153 176 |
125 130 |
100 100 |
180 193 |
173 163 |
| Foot | Sen. Mot. |
100 100 |
218 170 |
170 151 |
110 115 |
160 153 |
153 148 |
125 125 |
229 199 |
183 175 |
|
| Lips | Sen. Mot. |
130 120 |
141 133 |
135 127 |
125 125 |
144 136 |
138 133 |
100 1000 |
193 166 |
179 165 |
|
number of reactions is very inadequate to show anything like the full effects of practice. But even in the shorter series the drift is clearly indicated. In addition to the general results already pointed out, it further appears from the table, that for all three reagents, the ear-lip coördination is the fastest; also that there is less difference between the sensory and motor reactions at the outset of this series than at the beginning of the others. For A. and J., the reactions at the beginning of the eye-foot series, are the longest on the record, and show also the widest
(252) difference between the sensory and motor forms. A. shows at the beginning of all the series the faster time in the motor form; M.'s reactions are quite divided between the two forms showing the faster time in the sensory form in both hand series, and in the auditory-foot series, but is faster in the motor form for both lip series, and for the eye-foot series. J. is motor throughout except in the eye-hand series. A series of auditory and visual memory tests with nonsense syllables showed A. and J. quicker in the visual, and M. slightly faster in the auditory form.
Interpretation [4]
Taking the simple reaction as the type of voluntary action in general, and voluntary action as action under the direction of attention, it seemed that the key to any explanation adequate to all the facts, the individual peculiarities and the effects of practice, must be found in the functions of attention and habit in their relations to each other.
Not to go into too great detail, the process of attention in its essential outlines in, say, the auditory-hand reaction, appears something as follows: As the reagent receives his instructions for the reaction, he formulates in imagination what he is going to do. This formulation, the getting in mind what he is to do, is his attention to the act. Whatever may be the detail of imagery involved in this formulation, it involves primarily the coördination of two groups of incoming sensations, one from the ear, the other from the hand, started by the operator's descriptions. From this, two distinctions may be drawn: (I) As related to the act of attention, these two sensation groups are its stimuli; and each group is as much stimulus as the other — the sensations from the hand as much as those from the ear. The reaction ' e-s meaning the whole act to be performed is not the mere response of the hand to the ear, but the act of attention in coordinating the incoming stimuli from both the hand and the ear. Concerning the 'sensory-motor' distinction it follows that, since the stimulus, i. e., the material for the act, lies in
(253) these incoming currents from both hand and ear, as related to the whole act, both 'forms' may be regarded as equally 'sensory' or equally 'motor.' (2) In relation to each other inside the act of attention, most discussions of the subject appear to make the ear process merely a stimulus to which the hand adjustment is merely a response. But the question arises, What holds the ear to its work? Why does the reagent maintain his listening attitude ? It may be replied that it is ' because he is told to.' But he is not told to listen any more than he is told to move his hand. If the telling suffices in one case it should in the other. Moreover, he is not merely to listen, or even to listen just for the click, but to listen for the click as a pressure signal. It is this character of the click as a signal for pressure that keeps up the interest in it and the attention to it. (We are assuming here, of course, a case of sensory attention.) The hand therefore is stimulus as well as response to the ear, and the latter is response as well as stimulus to the hand. Each is both stimulus and response to the other. The distinction of stimulus and response is therefore not one of content, the stimulus being identified with the ear, the response with the hand, but one of function, and both offices belong equally to each organ. The reason the movement of the hand is so often treated as the mere response to the ear as its mere stimulus appears to be that the whole act, or 'reaction,' is identified with the movement of the hand. But the entire act is the act of attention in coordinating the two groups of stimuli coming from both hand and ear. To be sure, in the act of coördination there is, as we have seen, the interaction of the two elements as stimulus and response each to the other. But it must be kept in mind that this latter is a distinction falling inside the act, not between the hand movement considered as the act, and the sound considered as its external stimulus or 'cause.' In a word, the reagent reacts as much with his ear as he does with his hand.
With the reaction now interpreted as essentially constituted by the act of coördinating the ear-hand activities; with the distinction of 'stimulus' and ' response' interpreted as wholly functional, falling inside the act, the question still remains, —why, in the act of coördination, is attention occupied more imme
(254)-diately with one of these processes than with the other? This question, again, does not ask whether the attention shall be given to the sound or to the movement as such, but where in the total ear-hand process the focus of attention shall fall. This point, wherever it is, must be determined not by the solicitation of the point in itself considered, but by the demands of the whole act of coördination. Whether the attention be 'in the hand' or 'in the ear,' it is ' there' in order to bring off successfully the ear-hand adjustment.
But why is it ' in' one or the other? This leads to the consideration of another function or rather another phase of the function of attention, namely, its function as the adjustor, the mediator, of the tension between habitually established coördinations and new conditions under which they have to express themselves. An habitual process, such, e.g., as walking, comes into consciousness as, i..e., under attention, only when some new set of conditions, some obstacle, arises, adjustment to which lies outside the scope of the habit. Then only so much of the process comes into consciousness as needs readjustment to the new conditions. Habit is still left to do all it can, and in every voluntary act there is always something left for it to do. No matter with how minute a portion of a process attention may be occupied, it always will be found giving direction to a group, no matter how small, of already coordinated activities. Any attempt, therefore, to leave habit out of the account in voluntary action makes such action impossible. It would be affirming a process of adjustment with nothing to adjust. If attention, as such, then, is the process of mediating the tension between habit and new conditions, its focus must be where this tension is strongest, i. e., where habit is least able to cope with the situation. The position of this point will depend upon the extent to which the different parts of the whole ear-hand process can be left to habit. If the ear element of the process, that is, the sound, be unfamiliar and the movement of the hand be familiar, the point of tension will fall 'in the ear' and vice versa. With the sound and movement both familiar or both unfamiliar, the balance between them will be determined by education, inherited structure, etc. Let it be noted again, that in
(255) leaving one phase more than the other to habit, the former is not left out of the act, nor out of the process of attention. In attention to the sound, the movement of the hand is present in the character of the sound as a pressure signal; and in attention to the movement, the sound is present in the very fact of its being a movement in response to the sound.
Concerning the process of 'shifting' the attention and the accompanying time variations it follows from the very nature of attention that it is only from an external point of view, the point of view of an observer, that we can speak of 'shifting' the attention in the same act. For the reagent a 'shift' of attention is a change in his act; it means a different process of coördination. For him, therefore, the sensory and motor forms of attention are not 'two forms' of attention for the same act; they are essentially two different acts; and the time question for him is, which of the two acts is the shorter. Regarding the act, however, in an objective way, we should expect the time of the reaction to be shortest when the attention is upon that part of the process which is least habitual, or in which habit encounters the most new conditions. When one who reacts spontaneously to sound in the sensory way attempts to transfer his attention to the hand, two things are involved: (I) leaving the ear adjustment for habit to take care of; (2) a breaking up, in attending to the hand process, of an already efficiently established coördination. This means that for the performance of the 'act,' regarded objectively, unnecessary work is being done. The focus of the attention upon the more habitual phase of the process means its resolution into elements. Now the moment these elements are called out as unit groups, they bring with them their own train of associated groups, all of which have to be inhibited. This increased and, from the objective view of the act, unnecessary complication, means, of course, an increase in time and mean variation, and accounts for the exceedingly 'artificial' feeling that accompanies the effort. [5]
(256)
Indeed, whether the attempt succeeds at all, depends upon the extent to which the ear
adjustment can be left to habit. If the sound be very strange the reagent finds he cannot
attend to the hand until, as he says he 'gets used' to the sound, and this ' getting used'
to the sound is, of course, the ear adjustment becoming habitual.
This explains why at the outset of the series M's reactions were all made in the sensory form. As the ear phase, however, grows more and more reflex the breaking up of the hand process becomes possible and the motor form of attention emerges. A precisely parallel process takes place in the development of the sensory form for a reagent spontaneously motor. Under practice the new form continues, of course, to grow more and more reflex, and its time and mean variation steadily decrease. In the case of M., for whom the new form was the motor, its time kept diminishing in the hand series until, at the close, it was a few sigma faster than the sensory. A similar development occurred in J.'s visual hand series. The sensory was faster at the beginning, the motor at the close. In all the series where the motor was faster at the beginning, it still remained so at the close, though not by nearly so large a margin as reported in the Leipzig tables. In saying that continued practice on both these forms rendered them 'more reflex,' we mean that when at the close the reactions were made in the fastest time it was with a much less amount of tension, and consequently less attention, than at first. Any extraordinary attempt, at this stage, to concentrate upon either form resulted in great irregularity and increase of time, for the reason, as already stated, that all there is for attention to do is to break up and re-establish processes already unified under habit.
But why even the small margin in favor of the motor time at the close of a course of practice on the two forms ? The reason usually given is that ' attention to the movement is the beginning of it.' But if the whole act is not the mere hand movement, but the coördination of ear and hand, it is difficult to see why the sensory form is not as much the beginning of the act as is the motor. If attention to the hand is the 'beginning of the hand movement' it is no less true that attention to the ear is
(257) the beginning of the sound. And this is not mere ' arm chair' psychologizing, for as a matter of fact, at the beginning of the series, nearly all M's premature responses occurred in the sensory form, and at the close of the series about as many occurred in one form as in the other. When the premature response came, it was always preceded by a feeling of great tension in the ear, and succeeded by a corresponding feeling of relief in the ear, and when in this case the true signal came it was frequently lost entirely by the reagent. Again, it is said, that in attention to the hand a much larger portion of the whole pathway of discharge is ' innervated ' then in attention to the ear. This seems to assume that only that part of the total pathway is 'innervated' which is represented in the focus of attention. But if innervation' is necessary to movement, and the focus of attention is necessary to ' innervation,' how, e. g., does walking go on when the focus of attention is elsewhere? The answer is, that we must interpret the focus of attention not as a point of innervation merely but as a point of conflicting innervations demanding adjustment. It is not the need of innervation as such, but of adjustment of innervations, that determines the focus. But, even supposing that attention means increased innervation, there appears no reason why, assuming the nervous structure homogeneous, and the amount of innervation the same, its application should be more effective at one point than at another. As to the distribution of the innervation over a larger area, if the amount is the same, it must be correspondingly weakened at each point, and so nothing be gained.
It appears, then, that for an explanation of the fact that at the end of the course of practice, when both forms had became highly reflex, the motor form was little the faster, we must appeal again to the relation between Habit and Attention, still regarding, in the objective way, the sensory and motor reactions as ' two forms' of the ' same' act, this fact of the shorter motor time means, (I) in terms of Attention, that the stimulus presented by the ear adjustment affords less material for the continued exercise of attention than that presented by the hand; (2) in terms of Habit it means that the ear process becomes more rapidly and more completely habitual than that of the hand. It takes but a
( 258) short time to 'get used' to even the strangest sound. After this the character of the sound is comparatively fixed. It cannot tee changed through further adjustment. This appears due largely to the more stable character of the inherited ear mechanism. On the other hand, the phase of the stimulus presented by the hand is not nearly so fixed and stable. Here there is much more opportunity for continued variation, hence more ground for the continued exercise of attention. Applying what has already been said, we should, then, expect that act to be faster in which the focus of attention is upon the less stable phase of the hand element of the coördination rather than when it is more ' artificially' occupied in breaking up the more completely established ear adjustment. In a word the time question is not a case of a ' sensory ' vs. a ' motor' reaction, but of a sensori-motor less habitual vs. a sensori-motor more habitual.
As stated at the outset, this interpretation in terms of Habit and Attention seems to us to combine elements from both sides of the 'type discussion.' On the Princeton side it would say: (I) that the ' type' of attention and its accompanying time are determined by the relation between the individual's stock of coördinations, inherited and acquired, already on hand, and the particular coördination required by the reaction; (2) that the ' sensory form' may still be the faster even after the 'motor form' has clearly emerged in consciousness. On the Leipzig side it would say that under practice, in both forms, upon the same coördination, the sensory phase passes more completely under the control of habit and thus leaves the faster time to the motor form.
