off the work and fatigue of the sense organ and of the end plate in the muscle from the work of the connection system and to distinguish sensorial fatigue, intellectual fatigue and muscular fatigue. For the action of a sense organ or of an end plate is only partly, and the action of a muscle fiber is not at all, like that of the connecting neurone." He passes into the realm of classification of fatigue with a vengeance in the following note: 1 1 It will doubtless be better in the long run to subdivide the work of the human animal still further and to replace the terms sensory, mental and muscular work by the work of the accessory apparatus of the sense organs, the work of the peripheral end of the first sensory neurone, the work of conduction along a neu- rone, the work of conduction across a synapse from neurone to neurone, the work of changing the intimacy of synapses, the work of conduction from the ending of a neurone to a muscle, the work of the muscle fibers." (Italics in text omitted.)
(3) The final factor which has brought the psychology of THE ORGANISM AT WORK 351 fatigue to such a helpless position is the attempt to keep safe the tenets of psycho-physical parallelism. The discussion of this con- cept belongs to the realm of metaphysics. It can be left safely to the philosophers for discussion.
In approaching the problem of the exercise of a function both from the standpoint of research and for collecting *data on what has been done it seems safe to assume that no confusion can result and that nothing will be lost to science if psychology will discard the concept of fatigue altogether. All the investigator in this field needs to do is to name the function he is measuring and specify the conditions under which the act is taking place. The function studied may be bricklaying, typewriting, multiplying subvocally or any other function, and we may be as painstaking as we please about the care with which we specify the conditions under which such functions are exercised. For example, we may specify that the individual is multiplying one four-place number by another without being allowed to use pencil or paper or to put the numbers down in any way or to'speak aloud or move from his chair or eat or sleep until ten hours of such work has been accom- plished. We may still further specify by saying that he performs this work with eyes closed and with ears and nose plugged up and that after he has worked for five hours such and such a drug is administered. In bricklaying, for example, we may be as care- ful as we please about describing the kind of bricks the mason lays, the kind of mortar he uses, the kind of structure he is build- ing, how far* he has to stoop for his brick or mortar, what his family situation is, the wage he receives and similar conditions. When we have described the function being investigated, the con- ditions surrounding the worker and the method of measuring the function, we would then describe what the standard of efficiency is and how it was obtained and then state the effect the various control factors had upon the quantity and quality of the output.1 1 It is quite another problem but again worth while to specify the con- dition of the worker from time to time. For example, at the end of the fourth hour he may have become dizzy, nauseated or complained of a head- ache. The position we take here is not at all incompatible with studying the individual's organic condition from moment to moment or even with recording his verbal complaints..
352 PSYCHOLOGY The situation is thus not different in studying human functions from what it would be if we were studying activity in animals.
The point we would make in doing behavior work is to leave out of discussion all of those factors which are not touched or approached by the problem in hand; for example, what possible good does it do when discussing bricklaying or subvocal arithmetic to guess at what goes on in the synapse, in the efferent or afferent leg of the reflex arc or in the muscle itself? These are all worth- while problems, but they belong in the realm of physiology and this section of physiology has not yet been written. In preceding chapters we have given brief summaries of the few facts on the effect of continuous exercise of the nerve fiber, nerve cell and muscle concerning which there is some agreement. The total known facts are pitifully meagre. For students of behavior to devote hundreds and thousands (literally) of pages to the "physiological aspect" of fatigue problems tends in no way to the advancement of the study. The subject of the effect of continued use of the nervous system should be left to the neuro- physiologists or, better, to the joint work of the psychologists who will do the behavior work and the neuro-physiologists who will examine the nervous and muscular structures.
THE CURVE OF WORK.
THE CURVE OF WORK.
'General Considerations. — Assuming that a function is any well-established habit like that of typewriting, billiard playing or subvocal arithmetic, the question next arising in measuring the quantity or output of work done and its quality is whether there are well-marked fluctuations or rhythms or bursts of speed during the exercise of the function or whether the only marked change is the slow decrease in the output and possibly increase in errors due to the effect of continuous exercise. There is rather a persistent popular view that there are such fluctuations. In certain laboratories these views have apparently been confirmed. It is stated that there is (1) an initial spurt, (2) an end spurt and (3) a warming up or adaptation period.
In the (1) initial spurt it is assumed that the -worker keyed up for his task assumes an initial rate of speed which cannot be THE ORGANISM AT WORK 353 maintained. Thorndike tested this matter by recording the work done by several subjects in sub vocal multiplication and in written addition. These subjects were educated adults working continu- ously for approximately two hours. In general, there was no initial spurt. It was found that the curve of work varied in each individual from day to day and that if any one were interested in proving an initial rise he could pick out occasionally a curve which would show it; but no spurt was characterstic of any of the workers when the several curves gathered from several days' works were examined. In Kraepelin 's laboratory the initial spurt was rather a constant occurrence and was supposed to endure for approximately five minutes. A very recent study of Chapman and Nolan tends to confirm the older view of Kraepelin rather than that of Thorndike. Subjects were tested in continuous addition for 16 minutes on 7 successive occasions. They found definite evidence of an initial spurt in beginning the work, with a high percentage of errors. The speed was so great that no system of penalties for errors could cause this spurt to disappear. The spurt was of short duration, the subject always taking on a higher rate than he could possibly maintain. "He very rapidly settles down to a normal rate at which he can work for long periods. It is the rapidity of this decrease in efficiency which has led to the fact that initial spurt has been overlooked and even denied." Thorndike again in his most recent work finds an absence of initial spurt.
In regard to (2) "end spurt, " the evidence shows in many cases that there is a slight increase in the output in the last few minutes of work where the subject is working against time and has been informed beforehand that the work is to cease after a specified number of minutes. Book in his study of typewriting where ten-minute periods of work were used shows that the last three minutes are slightly better than any previous three minutes, the output standing in the relation of 100, 101, 102. Aral's last ten minutes of work (twelve hours continuous work) on sub- vocal arithmetic (page 355) was done at a slightly faster rate than any rate attained in the previous half hour. The end spurt though real under such conditions is not an important factor 23 354 PSYCHOLOGY in the curve of work. There is only very slight evidence to show that spurts occur after interruptions or disturbances or that they occur after momentary periods of decreased efficiency.
The curve illustrating the output of the exercise of any given function is said by some to show a gradual increase in efficiency from the moment of starting (or slightly later, at the end of the decline following after the "initial spurt") up to the end of the first twenty to thirty minutes; to stay at this high level for a definite time and then to show a decline. This is called the (3) "warming-up" period. It is sometimes asserted that there is a slower and more permanent improvement running parallel with the warming-up phase but enduring for a longer time. This is A FIG. 65. — Curve showing the gradual increase in time required to multiply (subvocal) one four-place number by another. The work was done continuously from 11 A.M. to 11 P.M. The times of solving the first example on each of the four days are averaged, giving the first point in the curve; the times of solving the second examples on each of the four days are similarly averaged, giving the second point on the curve. This procedure was repeated to the sixty-seventh example, which was the number completed on each day. The curve was constructed by the author from Dr. Arai's tables. It is not corrected for errors. The vertical line gives time in minutes; the horizontal line, the number of examples. Had the time of completing each example not increased — that is, if the function had not decreased in efficiency — the curve would have remained parallel to the base line; as it is, the curve, allowing for slight fluctuations, remains constant for the first ten examples, then declines and remains constant on this higher level to the 34th example, then declines sharply and continuously (with fluctuations) to the 48th example, then improves markedly to the 61st example, then again declines.
called the "adaptation" process. Experimental confirmation for these changes is not at hand. An examination of the curve illustrating Dr. Arai's work (Fig. 65) shows the absence of such asserted improvement. Where there is apparent evidence for it the changes noted seem to be due to a period of improvement by practice in the function (actual learning). Where the function tested is a thoroughly grounded one the phenomenon is lacking. There is no very satisfactory experimentation upon the curve of work illustrating strictly manual functions. In baseball, track and crew work a process of limbering up or preliminary practice THE ORGANISM AT WORK 355 is universally indulged in and is apparently necessary. Prob- ably no manager of a baseball team would risk sending a pitcher in without giving him an opportunity to work up a sweat and to break in his arm. Possibly the main beneficial result comes from increased glandular action: the muscular area in operation in such functions is very great and the consumption of food is large, hence the by-products of activity are enormous. The need of adrenin for breaking up and releasing the stored glycogen and inducing better circulation seems essential. Where the muscular area in operation is small, as in laryngeal work, the warming up period is not essential.
A great deal of speculation has been indulged in in regard to the lack of "fatigue," so-called, in a laryngeal function as con- trasted with its very rapid appearance in manual work. Arai continued to work in subvocal arithmetic for a period of twelve hours with a decrease in efficiency possibly of not more than 25 per cent. No crew working as rapidly as possible could continue its rowing beyond a few miles with only a similar loss in efficiency. In view of the difference in the size of the muscular areas involved in the two types of work, the findings are in line with prediction.
Subvocal Arithmetical Functions Continuously Used. — One of the most thorough investigations yet made upon the meas- urement of a function continuously exercised for a long period of time is that of Dr. Arai in subvocal arithmetic. Having by frequent practice reached the point where practice effects had almost disappeared, she worked on four days from 11 A.M. to 11 P.M. at multiplying four-place numbers by four-place numbers without extraneous aid. During the multiplication, the two four-place numbers (for example, X|^|| ) were not in sight but were referred to when necessary. The multiplication took place with eyes closed. The only rest taken was the few seconds re- quired to write down the answer and to learn the next problem. The work involved in making a single such calculation is for most people stupendous or even impossible. Approximately 17 sets, 4 examples in each set, were done during the 12 hours' continu- ous work. The whole table cannot be reproduced, but we can 356 PSYCHOLOGY compare the average time of solving the first set of 4 examples each day with the average of the last set of 4 as follows: Average time in min- utes of solving first First day Second day Third day Fourth day Average time in min- utes of solving last It probably would be fairer to compare the average time of solv- ing the first two sets of four examples (8 examples) with the average time of solving the last two sets of four.
An examination of the tables shows that there is a loss in effi- ciency as measured by the increase in time of solving the last 8 examples. The average increase in time for the last eight ex- amples over the first 8 examples on the 4 days was 119 per cent. While at first sight this loss in efficiency may seem large, a little consideration will convince us that the efficiency of the function even after 12 hours' continuous work was still very high. Dr. Arai thus required a little over twice as long to perform one such example at the end of the 12 hours' work as at the beginning. Even at the end she was performing a feat which could not be equaled probably by two persons out of a thousand engaged in arithmetical work. The second most important point coming from this research was the fact that the normal period of sleep completely restored the function as shown by the fact that the average time of solving the first four examples1 on each of the four days is about equal. As a matter of fact, there is a decrease in time which looks like a practice effect. The author has taken the figures of Miss Arai and plotted the curve given above, Fig. 65 (without correction for error). She worked four days and did 67 problems on each day. The points on the curve were obtained as follows: The times of solving all first problems (that is, the times of solving the first problems on each of the four days) were THE ORGANISM AT WORK 357 averaged; this gives the first point on the curve. Then the times of solving all second problems were similarly averaged to obtain the second point on the curve. This procedure was repeated for all 67 points. The curve shows in the main an increase of time. The first 10 problems show only individual fluctuations; then occurs a slight decline. From that point to the 34th, the function is maintained at fairly uniform efficiency but with upward and downward fluctuations. From the 34th example to the 48th there is a marked decline in efficiency. From the 48th to the 61st a marked improvement occurs, then a fall in efficiency extending to the 65th, and finally a slight rise.
• It would take us beyond the limits of this discussion to present all the data which has been collected upon this type of work. Such functions as the following have been tested: Writing sen- tences from dictation, learning nonsense syllables and digits, translating, marking words containing d and t and selecting titles for a bibliography. The interesting point about all of the results is that such functions can be exercised continuously for several hours without severe loss in efficiency.
Continuous Exercise of Manual Functions. — No careful measurement of the loss in efficiency in a manual function due to continuous exercise, comparable with Dr. Arai's work, has been made. That the loss is more rapid in manual work every one admits. Furthermore, short periods of rest seem to be advan- tageous in manual functions, whereas they are certainly not necessary and possibly disadvantageous in subvocal functions. Many of the experiments have been made upon the ergograph where a single group of muscles is singled out and made to work continuously in a way in which no daily task ever employs them. This work we summarized on page 167. It is a most unsatisfac- tory way to study the curve of work in manual functions.
In the factories loss of efficiency in manual functions as measured in terms of decreased output is being studied with some care. We quote the following, which appeared in The Iron Age: "At these works there was recently constructed a long incline up which heavy loads were to be wheeled in barrows, and pre- 358 PSYCHOLOGY • iniums were offered to the men who did or exceeded a certain amount of this labor. They attempted it vigorously, but none succeeded in earning any of the extra money; instead, they all.fell considerably below the fixed task.
' ' Prompt investigation by an expert disclosed that the trouble lay in the fact that the men were working without sufficiently frequent periods of rest. Thereupon, a foreman was stationed by a clock, and every twelve minutes he blew a whistle. At the sound every barrowman stopped where he was, sat down on his barrow, and rested for three minutes. The first hour after that was done showed a remarkable change for the better in accom- plishment; the second day the men all made the premium allow- ance by doing more than what had been too much; and. on the third day the minimum compensation had risen, on the average, 40 per cent., with no complaints of overdriving from any of the force."
This illustrates the fact that where the task is severe there is a growing tendency to allow a short period of rest after definite intervals of work. The whole question of decrease in the quantity and quality of output is bound up in daily life with a great many emotional factors such as whether the man's pay is sufficient 1x> afford him an outlet for play, marriage and the like; one's personal and family adjustment; political convictions; the sociological theories the worker may have; open- and closed- shop policies; and most important probably, the rate at which the individual works. The heart teaches us a lesson. The heart functions continuously from the beginning of the third month of foetal life to death but at such a rate and with such a rest period between functions that its efficiency is uniformly main- tained. While not very much of value on the problem of effi- ciency in manual functions has yet come from a psychological laboratory, the solution has to be sought there, since work in the shop and in commercial houses generally cannot be controlled in such a way as to yield very safe results. But the type of work which the laboratory undertakes will have to be considerably broadened in scope. Our own view is that many of the basal laws can be established by using animals. We can force the THE ORGANISM AT WORK 359 animal to form, very much, more complicated habits than would appear from our ordinary laboratory studies. If punishment is used instead of food, the stimulating value of the situation as a whole would probably not decline very much from hour to hour. Given such a set of functions in animals, it would be possible to vary the conditions, such as underfeeding, undersleeping and the administration of drugs, and note the effect of the varying con- ditions upon the efficiency of the function under observation.
Physiological Aspects of the Continuous Exercise of a Function. — In previous chapters we have discussed certain phases of the physiological findings in regard to the effect of continuous works as follows: Chemical changes occurring in a nerve fiber due to continuous stimulation (page 121); structural and chemical changes in nerve cells (page 121); chemical and form changes in muscle due to work (page 168); effect of adrenin upon the prod- ucts formed by continuous activity of the muscle (page 221). We briefly summarize these findings again: The question as to whether there are any structural and chemical changes in the nerve fiber due to continued functioning is still in doubt. The question of the changes in the nerve cell is even more problem- atical; in general, form, size and chemical changes are believed to occur, but the amount of exercise of the function necessary to bring about such changes is tremendously large. The products of muscular activity are slightly better known: there is the elim- ination of CO., in greater amounts when the muscle is active than when inactive; lactic acid is formed and possibly potassium phosphate. These products of activity get into the blood stream, are carried to muscles which have not been active and decrease the amount of work they are capable of doing. In regard to the influence of adrenin, it is stated that due to the increased supply of blood to the muscles such products are rapidly washed out or else neutralized in EFFECT OF DRUGS ON THE FUNCTIONING OF ESTABLISHED HABIT SYSTEMS.
General Considerations. — Several factors should be taken into account in the discussion of experimentation and measure- 360 PSYCHOLOGY • ment of any function: (1) only in rare cases has a function been practiced sufficiently in laboratory studies before controlled tests were introduced to gauge the effects of drugs, continuous exer- cise, loss of sleep and other factors. In other words, the subject improves in skill through practice — this may mask the effects of a drug or the loss of sleep or continuous exercise. Either the func- tion should be practiced to the point where no improvement can be observed before control tests are introduced or else a meas- urement of the improvement factor should be found. (2) The situation for the worker in such experiments is not normal. He is being tested under unusual conditions, and emotional factors with their reinforcing or inhibiting effects may enter in. This has vitiated many experiments upon alcohol, tobacco and caffeine especially. To eliminate this effect various devices may be used. The subject may be told that he is being given caffeine in a cap- sule, whereas only flour or meal may be given; or he may be told that he is being given alcohol and some other masked non-alco- holic substance is actually given. On the other hand, he may be told nothing, the experimenter depending upon his own ability to introduce the control substance or condition in such a way that the subject is not able to note whether the drug or control factor has been administered. Rivers first emphasized the im- portance of the emotional aspect of such experimental conditions, and efficiency experiments may be said to start with his work in 1906. (3) Too narrow functions have been studied — rarely has any of the work been done upon functions used in daily life. As an example, we may mention the fact that in the great mass of experimentation upon manual functions the ergograph has been used. (4) The functions have been followed through for entirely too short periods for wide generalizations to be made.
In the summary of the literature which follows these factors should be taken into consideration and the results cited should be looked upon as being merely provisional.
Alcohol. — On account of various partisans for and against the use of alcohol, it is difficult to obtain unequivocal data upon this subject. The following summary presents the unparti- san view of most scientific men who have taken up the study THE ORGANISM AT WORK 361 of alcohol from a psychological and physiological standpoint.
1. There is a wide range of individual variation in the ability of different people to consume alcohol without showing its effect upon their curve of work. The amount may vary from 20 c.c. to 40 c.c. of absolute alcohol (a very much heavier dose than is usually taken for purely social purposes).
2. The effect of alcohol upon all reflexes when any is noted is depressing. This has been confirmed by many investigators working upon the patellar reflex, reflex eye movements and others.