In general, labyrinth B proved very satisfactory as a means of testing the ability of the dancer to learn a simple path. The narrow passages effectively prevented dancing, and the introduction of the electric shock as a punishment for mistakes developed a motive for escape which was uniform, constant, and so strong that the animals clearly did their best to escape from the labyrinth quickly and without errors. This maze was so simple that it did not tend to discourage them as did the one which is next to be described. It must be admitted, however, that, though labyrinth B is perfectly satisfactory as a test of the dancer's ability to learn to follow a simple path, it is not an ideal means of measuring the rapidity of habit formation. This is due to the fact that the preliminary trial and the first training test play extremely different roles in the case of different individuals. A dancer which happens to follow the correct path from entrance to exit in the preliminary trial may continue to do so, with only an occasional error, during several of the early training tests, and it may therefore fail for a considerable time to discover that there are errors which should be avoided. The learning process is delayed by its accidental success. On the other hand, an individual which happens to make many mistakes to begin with immediately attempts to avoid the points in the maze at which it receives the electric shock. I was led to conclude, as a result of the labyrinth-B experiments, that the path was too easy, and that a more complex labyrinth would, in all probability, furnish a more satisfactory means of measuring the rapidity of habit formation.



On the basis of the supposition that a maze whose path was so complex that the animal would not be likely to follow it correctly in the early trials would be more to the purpose than either A or B, labyrinth C was devised. As is shown in the plan of this maze, Figure 27, five mistakes in choice of path were possible on the forward trip. These errors, as a rule, were more difficult for the dancers to avoid than those of labyrinths A and B. Those which are designated by the numerals 2, 3, and 4 were especially difficult. Error 4 was much more troublesome for left whirlers than for right whirlers because, after turning around abruptly at the entrance to the blind alley, the former type of dancer almost always followed the side wall of the maze so far that it missed the correct path. Undoubtedly the various errors are not of the same value for different individuals; but it would be extremely difficult, if not impossible, to devise a maze which should be equally difficult for several normal individuals.

In order that records of the path followed by a mouse in test after test might be kept with ease and accuracy by the experimenter, the plan of this labyrinth, and also that of labyrinth D, were cast in rubber. The outlines of labyrinths C and D which appear in Figures 27 and 28 respectively were made with the rubber stamps which were thus obtained. Figure 27 is the reproduction of a record sheet which presents the results of the first, the fifth, the tenth, and the eleventh tests of No. 2 in labyrinth C. The path followed by this individual in the first test was far too complex to be traced accurately on the record sheet. The record therefore represents merely the number of errors which was made in each region of the maze. For the fifth test, and again for the tenth and the eleventh, the path was recorded accurately. This simple device for making record blanks which can readily be filled in at the time of the experiment should recommend itself to all students of animal behavior.

In labyrinth C ten pairs of dancers were given continuous training tests at the rate of one test per minute until they were able to follow the direct path correctly. Because of the difficulty in learning this maze perfectly, it was not demanded of the mice that they should follow the path correctly several times in succession, but instead the training was terminated after the first successful trip.



TABLE 39

RESULTS OF LABYRINTH-C EXPERIMENTS, WITH TWENTY DANCERS

MALES FEMALES

NO. OF NO. OF FIRST NO. OF NO. OF FIRST MOUSE CORRECT TEST MOUSE CORRECT TEST

2 11 29 15 30 33 49 34 50 49 57 15 52 22 59 15 58 16 215 10 60 17 415 10 76 3 75 8 78 6 77 11 86 5 87 9 88 25 85 11

AV. 18.7 AV. 13.8



The results of the experiments with this labyrinth as they are presented in Table 39 indicate that its path is considerably more difficult for the dancer to learn than that of labyrinth B, that the females learn more quickly than the males, and finally, that individual differences are just as marked as they were in the case of the simpler forms of labyrinth. It therefore appears that increasing the complexity of a labyrinth does not, as I had supposed it might, diminish the variability of the results. Certain of the individual differences which appear in Table 39 are due, however, to the fact that in some cases training in labyrinth B had preceded training in labyrinth C, whereas in the other cases C was the first labyrinth in which the animals were tested. But even this does not serve to account for the wide divergence of the results given by No. 2 and No. 50, for the latter had been trained in B previous to his training in C, and the former had not been so trained. Yet, despite the advantage which previous labyrinth experience gave No. 50, he did not learn the path of C as well in fifty tests as No. 2 did in eleven. The facts concerning the value of training in one form of labyrinth for the learning of another, as they were revealed by these experiments, may more fittingly be discussed in a later chapter in connection with the facts of memory and re-learning.



Labyrinth C is a type of maze which might properly be described as irregular, since the several possible errors are extremely different in nature. In view of the results which this labyrinth yielded, it seemed important that the dancer be tested in a perfectly regular maze of the labyrinth-D type. The plan which I designed as a regular labyrinth has been reproduced, from a rubber stamp print, in Figure 28. As is true also of the mazes previously described, it provides four kinds of possible mistakes: namely, by turning to the left (errors 1, 5, 9, and 13), by turning to the right (errors 3, 7, and 11), by moving straight ahead (errors 2, 4, 6, 8, 10, and 12), and by turning back and retracing the path just followed. The formula for the correct path of D is simple in the extreme, in spite of the large number of mistakes which are possible, for it is merely "a turn to the right at the entrance, to the left at the first doorway, and thereafter alternately to the right and to the left until the exit is reached." This concise description would enable a man to find his way out of such a maze with ease. Labyrinth D had been constructed with an exit at 10 so that it might be used as a nine-error maze if the experimenter saw fit, or as a thirteen-error maze by the closing of the opening at 10. In the experiments which are now to be described only the latter form was used.

Can the dancer learn a regular labyrinth path more quickly than an irregular one? Again, I may give only a brief statement of results. Each of the twenty dancers, of Table 40, which were trained in labyrinth D had previously been given opportunity to learn the path of C, and most of them had been trained also in labyrinth B. All of them learned this regular path with surprising rapidity. The numerical results of the tests with labyrinths B, C, and D, as well as the behavior of the mice in these several mazes, prove conclusively that the nature of the errors is far more important than their number. Labyrinth D with its thirteen chances of error on the forward trip was not nearly as difficult for the dancer to learn to escape from as labyrinth C with its five errors. That the facility with which the twenty individuals whose records are given in Table 40 learned the path of D was not due to their previous labyrinth experience rather than to the regularity of the maze is proved by the results which I obtained by testing in D individuals which were new to labyrinth experiments. Even in this case, the number of tests necessary for a successful trip was seldom greater than ten. If further evidence of the ease with which a regular labyrinth path may be followed by the dancer were desired, it might be obtained by observation of the behavior of an individual in labyrinths C and D. In the former, even after it has learned the path perfectly, the mouse hesitates at the doorways from time to time as if uncertain whether to turn to one side or go forward; in the latter there is seldom any hesitation at the turning points. The irregular labyrinth is followed carefully, as by choice of the path from point to point; the regular labyrinth is followed in machine fashion,—once started, the animal dashes through it.



TABLE 40

RESULTS OF LABYRINTH-D EXPERIMENTS, WITH TWENTY DANCERS

MALES FEMALES

NO. OF NO. OF FIRST NO. OF LAST OF NO. OF NO. OF FIRST NO. OF LAST OF MOUSE CORRECT TWO CORRECT MOUSE CORRECT TWO CORRECT TEST TESTS TEST TESTS

2 3 7 29 10 11 58 7 10 49 7 8 30 9 10 57 3 6 60 10 14 215 6 10 402 10 11 415 7 8 76 4 7 75 4 13 78 4 5 77 11 12 86 3 9 87 4 9 88 4 8 85 3 4 90 7 8 83 4 7

Av. 6.1 8.9 Av. 5.9 8.8



From the results of these labyrinth experiments with dancers I am led to conclude that a standard maze for testing the modifiability of behavior of different kinds of animals should be constructed in conformity with the following suggestions. Errors by turning to the right, to the left, and by moving forward should occur with equal frequency, and in such order that no particular kind of error occurs repeatedly in succession. If we should designate these three types of mistake by the letters r, l, and s respectively, the error series of labyrinth C would read l-l-r-s-l. It therefore violates the rule of construction which I have just formulated. In the case of labyrinth D the series would read l-s-r-s-l-s-r-s-l-s-r-s- l. This also fails to conform with the requirement, for there are three errors of the first type, four of the second, and six of the third. Again, in a standard maze, the blind alleys should all be of the same length, and care should be taken to provide a sufficiently strong and uniform motive for escape. In the case of one animal the desire to escape from confinement may prove a satisfactory motive; in the case of another, the desire for food may conveniently supplement the dislike of confinement; and in still other cases it may appear that some form of punishment for errors is the only satisfactory basis of a motive for escape. Readers of this account of the behavior of the dancing mouse must not infer from my experimental results that the electric shock as a means of forcing discrimination will prove satisfactory in work with other animals or even with all other mammals. As a matter of fact it has already been proved by Doctor G. van T. Hamilton that the use of an electric shock may so intimidate a dog that experimentation is rendered difficult and of little value. And finally, in connection with this discussion of a standard Labyrinth, I wish to emphasize the importance of so recording the results of experiments that they may be interpreted in terms of an animal's tendency to turn to the right or to the left. My work with the dancer has clearly shown that the avoidance of a particular error may be extremely difficult for left whirlers and very easy for right whirlers.

I hope I have succeeded in making clear by the foregoing account of my experiments that the labyrinth method is more satisfactory in general than the problem method as a means of measuring the rapidity of habit formation in the dancer, and I hope that I have made equally clear the fact that it is very valuable as a means of discovering the roles of the various senses in the acquirement of a habit (Chapter XI). From my own experience in the use of the labyrinth with the dancer and with other animals, I am forced to conclude that its chief value lies in the fact that it enables the experimenter so to control the factors of a complex situation that he may readily determine the importance of a given kind of sense data for the formation or the execution of a particular habit. As a means of measuring the intelligence of an animal, of determining the facility with which it is capable of adjusting itself to new environmental conditions, and of measuring the permanency of modifications which are wrought in its behavior by experimental conditions, I value the labyrinth method much less highly now than I did previous to my study of the dancer. It is necessarily too complex for the convenient and reasonably certain interpretation of results. Precisely what is meant by this statement will be evident in the light of the results of the application of the discrimination method to the dancer, which are to be presented in the next chapter. The labyrinth method is an admirable means of getting certain kinds of qualitative results; it is almost ideal as a revealer of the role of the senses, and it may be used to advantage in certain instances for the quantitative study of habit formation and memory. Nevertheless, I think it may safely be said that the problem method and the discrimination method are likely to do more to advance our knowledge of animal behavior than the labyrinth method.



CHAPTER XIV

HABIT FORMATION: THE DISCRIMINATION METHOD

Discrimination is demanded of an animal in almost all forms of the problem and labyrinth methods, as well as in what I have chosen to call the discrimination method. In the latter, however, discrimination as the basis of a correct choice of an electric-box is so obviously important that it has seemed appropriate to distinguish this particular method of measuring the intelligence of the dancer from the others which have been used, by naming it the discrimination method.

It has been shown that neither the problem nor the labyrinth method proves wholly satisfactory as a means of measuring the rapidity of learning, or the duration of the effects of training, in the case of the dancer. The former type of test serves to reveal to the experimenter the general nature of the animal's capacity for profiting by experience; the latter serves equally well to indicate the parts which various receptors (some of which are sense organs) play in the formation and execution of habits. But neither of them is sufficiently simple, easy of control, uniform as to conditions which constitute bases for activity, and productive of interpretable quantitative results to render it satisfactory. The problem method is distinctly a qualitative method, and, in the case of the dancing mouse, my experiments have proved that the labyrinth method also yields results which are more valuable qualitatively than quantitatively. I had anticipated that various forms of the labyrinth method would enable me to measure the modifiability of behavior in the dancer with great accuracy, but, as will now be made apparent, the discrimination method proved to be a far more accurate method for this purpose.

Once more I should emphasize the fact that my statements concerning the value of methods apply especially to the dancing mouse. Certain of the tests which have proved to be almost ideal in my study of this peculiar little rodent would be useless in the study of many other mammals. An experimenter must work out his methods step by step in the light of the daily results of patient and intelligent observation of the motor capacity, habits, instincts, temperament, imitative tendency, intelligence, hardihood, and life-span of the animal which he is studying. The fact that punishment has proved to be more satisfactory than reward in experiments with the dancer does not justify the inference that it is more satisfactory in the case of the rat, cat, dog, or monkey. Methods which yielded me only qualitative results, if applied to other mammals might give accurate quantitative results; and, on the other hand, the discrimination method, which has proved invaluable for my quantitative work, might yield only qualitative results when applied to another kind of animal.

The form of the discrimination method whose results are to be presented in this chapter has already been described as white-black discrimination. In the discrimination box (Figures 14 and 15, p. 92) the two electric-boxes which were otherwise exactly alike in appearance were rendered discriminable for the mouse by the presence of white cardboards in one and black cardboards in the other. In order to escape from the narrow space before the entrances to the two electric-boxes, the dancer was required to enter the white box. If it entered the black box a weak electric shock was experienced. After two series of ten tests each, during which the animal was permitted to choose either the white or the black box without shock or hindrance, the training was begun. These two preliminary series serve to indicate the natural preference of the animal for white or black previous to the training. An individual which very strongly preferred the white might enter, from the first, the box thus distinguished, whereas another individual whose preference was for the black might persistently enter the black box in spite of the disagreeable shocks. First of all, therefore, the preliminary tests furnish a basis for the evaluation of the results of the subsequent training tests. On the day succeeding the last series of preliminary tests, and daily thereafter until the animal had acquired a perfect habit of choosing the white box, a series of training tests was given. These experiments were usually made in the morning between nine and twelve o'clock, in a room with south-east windows. The entrances to the electric-boxes faced the windows, consequently the mouse did not have to look toward the light when it was trying to discriminate white from black. All the conditions of the experiment, including the strength of the current for the shock, were kept as constant as possible.

Choice by position was effectively prevented, as a rule, by shifting the cardboards so that now the left now the right box was white. The order of these shifts for the white-black series whose results are quantitatively valuable appear in Table 12 (p. III). That the order of these changes in position may be criticised in the light of the results which the tests gave, I propose to show hereafter in connection with certain other facts. The significant point is that the defects which are indicated by the averages of thousands of tests could not have been predicted with certainty even by the most experienced investigator in this field.

In Table 41 are to be found the average number of errors in each series of ten white-black discrimination tests for five males and for five females which were trained by being given ten tests per day, and similarly for the same number of individuals of each sex, trained by being given twenty tests per day. Since the results for these two conditions of training are very similar, the averages for the twenty individuals are presented in the last column of the table. For the present we may neglect the interesting individual, sex, and age differences which these experiments revealed and examine the significant features of the general averages, and of the white-black discrimination curve (Figure 29).



TABLE 41

WHITE BLACK DISCRIMINATION TESTS. NUMBER OF ERRORS IN THE VARIOUS SERIES

MALES FEMALES

AVERAGES AVERAGES GENERAL AVERAGES AVERAGES GENERAL AVERAGES SERIES FOR 5, FOR 5, AVERAGES FOR 5, FOR 5, AVERAGES FOR ALL 10 TESTS 20 TESTS FOR 10 10 TESTS 20 TESTS FOR 10 (20) MALES PER DAY PER DAY PER DAY PER DAY AND FEMALES

A 5.8 6.0 5.9 5.8 5.8 5.8 5.85 B 5.6 6.2 5.9 5.8 5.6 5.7 5.8 1 5.0 5.0 5.0 5.6 4.6 5.1 5.05 2 2.6 4.6 3.6 4.4 5.0 4.7 4.15 3 3.0 3.4 3.2 3.4 3.4 3.4 3.3 4 2.6 3.8 3.2 2.4 2.2 2.3 2.75 5 2.4 2.0 2.2 2.6 1.8 2.2 2.2 6 1.6 1.6 1.6 1.0 2.2 1.6 1.6 7 1.0 1.4 1.2 2.0 0.4 1.2 1.2 8 0.2 0.6 .4 1.4 1.6 1.5 .95 9 0.2 1.0 .6 0.6 0.8 .7 .65 10 0 .8 .4 1.0 0.8 .9 .65 11 0 .8 .4 0.8 0 .4 .40 12 0 .6 .3 0.4 0 .2 .25 13 0 0 0 0 0 0 0 14 0 0 0 0 0 0 15 0 0 0 0 0 0



The preference series, A and B, reveal a constant tendency to choose the black box, whose strength as compared with the tendency to choose the white box is as 5.8 is to 4.2. In other words, the dancer on the average chooses the black box almost six times in ten. The first series of training tests reduced this preference for black to zero, and succeeding series brought about a rapid and fairly regular decrease in the number of errors, until, in the thirteenth series, the white was chosen every time. Since I arbitrarily define a perfect habit of discrimination as the ability to choose the right box in three successive series of ten tests each, the tests ended with the fifteenth series.

The discrimination curve, Figure 29, is a graphic representation of the general averages of Table 41.—It is an error curve, therefore. Starting at 5.85 for the first preliminary series, it descends to 5.8 for the second series, and thence abruptly to 5.05 for the first training series. This series of ten tests therefore served to reduce the black preference very considerably. The curve continues to descend constantly until the tenth series, for which the number of errors was the same as for the preceding series, .65. This irregularity in the curve, indicative, as it would appear, of a sudden cessation in the learning process, demands an explanation. My first thought was that an error in computation on my part might account for the shape of the curve. The error did not exist, but in my search for it I discovered what I now believe to be the cause of the interruption in the fall of the error curve. In all of the training series up to the tenth the white cardboard had been on the right and the left alternately or on one side two or three times in succession, whereas in the tenth series, as may be seen by referring to Table 12 (p.111), it was on the left for the first four tests, then on the right four times, and, finally, on the left for the ninth test and on the right for the tenth. This series was therefore a decidedly more severe test of the animal's ability to discriminate white from black and to choose the white box without error than were any that had preceded it. If my interpretation of the results is correct, it was so much more severe than the ninth series that the process of habit formation was obscured. It would not be fair to say that the mouse temporarily ceased to profit by its experience; instead it profited even more than usually, in all probability, but the unavoidably abrupt increase in the difficultness of the tests was just sufficient to hide the improvement.

As I have suggested, the plan of experimentation may be criticised adversely in the light of this irregularity in the error curve. Had the conditions been perfectly satisfactory the curve would not have taken this form. I admit this, but at the same time I am glad that I chose that series of shifts in the position of the cardboards which, as it happens, served to exhibit an important aspect of quantitative measures of the modifiability of behavior that otherwise would not have been revealed. Our mistakes in method often teach us more than our successes. I have taken pains, therefore, to describe the unsatisfactory as well as the satisfactory steps in my study of the dancer.



The form of the white-black discrimination curve of Figure 29 is more surprising than disappointing to me, for I had anticipated many more irregularities than appear. What I had expected, as the result of training five or even ten pairs of mice, was the kind of curve which is presented, for contrast with the one already discussed, in Figure 30. This also is an error curve, but, unlike the previous one, it is based upon results which were got from individuals of different ages which were trained according to the following different methods. Ten of these individuals were given two or five tests daily, ten were given ten tests daily, and ten were given twenty tests daily. The form of the curve serves to call attention to the importance of uniform conditions of training, in case the results are to be used as accurate measures of the rapidity of learning.

Examination of the detailed results of the white-black discrimination tests as they appear in the tables of Chapter VII will reveal the fact that some individuals succeeded in choosing correctly in a series of ten tests after not more than five series, whereas others required at least twice as many tests as the basis of a perfect series. In very few instances, however, was a perfect habit of discrimination established by fewer than one hundred tests. As the averages just presented in Table 41 indicate, fifteen series, or one hundred and fifty tests, were required for the completion of the experiment. One might search a long time, possibly, for another mammal whose curve of error in a simple discrimination test would fall as gradually as that of the dancer. It is fair to say that this animal learns very slowly as compared with most mammals which have been carefully studied. It is to be remembered, however, that quantitative results such as are here presented for the dancer are available for few if any other animals except the white rat. Neither in the form of the curve of learning nor in the behavior of the animal as it makes its choice of an electric-box is there evidence of anything which might be described as a sudden understanding of the situation. The dancer apparently learns by rote. It exhibits neither intelligent insight into an experimental situation nor ability to profit by the experience of its companions. That the selection of the white box occurs in various ways in different individuals, and even in the same individual at different periods in the training process, is the only indication of anything suggestive of implicit reasoning. Naturally enough comparison of the two boxes is the first method of selection. It takes the dancers a surprisingly long time to reach the point of making this comparison as soon as they are confronted by the entrances to the two electric-boxes. The habit of running from entrance to entrance repeatedly before either is entered, once having been acquired, is retained often throughout the training experiments. But in other cases, an individual finally comes to the point of choosing by what appears to be the immediate recognition of the right or the wrong box. In the former case the mouse enters the white box immediately; in the latter, it rushes from the black box into the white one without hesitation. So much evidence the discrimination tests furnish of forms of behavior which in our fellow-men we should interpret as rational.



Comparison of the error curves for the labyrinth tests (Figures 26 and 31) with those for the discrimination tests (Figures 29 and 30) reveals several interesting points of difference. The former fall very abruptly at first, then with decreasing rapidity, to the base line; the latter, on the contrary, fall gradually throughout their course. Evidently the labyrinth habit is more readily acquired by the dancer than is the visual discrimination habit. Certain motor tendencies can be established quickly, it would seem, whereas others, and especially those which depend for their guidance upon visual stimuli, are acquired with extreme slowness. From this it might be inferred that the labyrinth method is naturally far better suited to the nature of the dancer than is any form of the discrimination method. I believe that this inference is correct, but at the same time I am of the opinion that the discrimination method is of even greater value than the labyrinth method as a means of discovering the capacity of the animal for modification of behavior.

Inasmuch as my first purpose in the repetition of white-black discrimination tests with a number of individuals was to obtain quantitative results which should accurately indicate individual, age, and sex differences in the rapidity of learning, it is important to consider the reliability of the averages with which we have been dealing. Possibly two groups of five male dancers each, chosen at random, would yield very different results in discrimination tests. This would almost certainly be true if the animals were selected from different lots, or were kept before and during the tests under different environmental conditions. But from my experiments it has become apparent that the average of the results given by five individuals of the same sex, age, and condition of health, when kept in the same environment and subjected to the same experimental tests, is sufficiently constant from group to group to warrant its use as an index of modifiability for the race. This expression, index of modifiability, is a convenient mode of designating the average number of tests necessary for the establishment of a perfect habit of white-black discrimination. Hereafter I shall use it instead of a more lengthy descriptive phrase.

As an indication of the degree of accuracy of measurements of the rapidity of learning which are obtained by the use of 5 individuals I may offer the following figures. For one of two directly comparable groups of 5 male dancers which were chosen from 16 individuals which had been trained, the number of tests which resulted in a perfect habit of white-black discrimination was 92; for the other group it was 96. These indices for strictly comparable groups of 5 individuals each differ from one another by less than 5 per cent. Similarly, in the case of two groups of females, the indices of modifiability were 94 and 104. These figures designate the number of tests up to the point at which errors ceased for at least three successive series (30 tests).

The determination of the probable error of the index of modifiability further aids us in judging of the reliability of the measure of the rapidity of learning which is obtained by averaging the results for 5 individuals. For a group of 5 males (Table 43, p. 243) the index was 72 +- 3.5; and for a group of 5 females of the same age as the males and strictly comparable with respect to conditions of white-black training, it was 104 +- 2.9. A probable error of +- 3.5 indicates the reliability of the first of these indices of modifiability; one of +- 2.9, that of the second.

I do not doubt that 10 individuals would furnish a more reliable average than 5, but I do doubt whether the purposes of my experiments would have justified the great increase in work which the use of averages based upon so large a group would have necessitated.

Further discussion of the index of modifiability may be postponed until the several indices which serve as measures of the efficiency of different methods of training have been presented in the next chapter.

From the data which constitute the materials of the present chapter it is apparent that the results of the discrimination method are amenable to much more accurate quantitative treatment than are those of the problem method or the labyrinth method. But I have done little more as yet than describe the method by which it is possible to measure certain dimensions of the intelligence of the dancer, and to state some general results of its application. In the remaining chapters it will be our task to discover the value of this method and of the results which it has yielded.



CHAPTER XV

THE EFFICIENCY OF TRAINING METHODS

The nature of the modifications which are wrought in the behavior of an organism varies with the method of training. This fact is recognized by human educators, as well as by students of animal behavior (makers of the science of comparative pedagogy), but unfortunately accurate measurements of the efficiency of our educational methods are rare.

Whatever the subject of investigation, there are two preeminently important aspects of the educative process which may be taken as indications of the value of the method of training by which it was initiated and stimulated. I refer to the rapidity of the learning process and its degree of permanency, or, in terms of habit formation, to the rapidity with which a habit is acquired, and to its duration. Of these two easily measurable aspects of the modifications in which training results, I have chosen the first as a means to the special study of the efficiency of the training to which the dancing mouse has been subjected in my experiments.

The reader who has followed my account of the behavior of the dancer up to this point will recall that in practically all of the discrimination experiments the number of tests in a series was ten. Some readers doubtless have wondered why ten rather than five or twenty tests was selected as the number in each continuous series. I shall now attempt to answer the question. It was simply because the efficiency of that number of tests, given daily, when taken in connection with the amount of time which the conduct of the experiments required, rendered it the most satisfactory number. But this statement demands elaboration and explanation.

Very early in my study of the dancer, I learned that a single experience in a given experiment day after day had so little effect upon the animal that a perfect habit could not be established short of several weeks or months. Similarly, experiments in which two tests per day were given proved that even a simple discrimination habit cannot be acquired by the animal under this condition of training with sufficient rapidity to enable the experimenter to study the formation of the habit advantageously. Next, ten tests in succession each day were given. The results proved satisfactory, consequently I proceeded to carry out my investigation on the basis of a ten-test series. After this method had been thoroughly tried, I decided to investigate the efficiency of other methods for the purpose of instituting comparisons of efficiency and discovering the number of tests per day whose efficiency, as measured by the rapidity of the formation of a white-black discrimination habit, is highest.

For this purpose I carefully selected five pairs of dancers of the same age, descent, and previous experience, and gave them white-black tests in series of two tests per day (after the twentieth day the number was increased to five) until they had acquired a perfect habit of discriminating. Similarly other dancers were trained by means of series of ten tests, twenty tests, or one hundred tests per day. Since it was my aim to make the results of these various tests strictly comparable, I spared no pains in selecting the individuals, and in maintaining constancy of experimental conditions. The order of the changes in the position of the cardboards which was adhered to in these efficiency tests was that given in Table 12.

At the beginning of the two-test training I thought it possible that the animals might acquire a perfect habit with only a few more days' training than is required by the ten-test method. This did not prove to be the case, for at the end of the twentieth day (after forty tests in all) the average number of mistakes, as Table 42 shows, was 3.2 for the males and 3.0 for the females. Up to this time there had been clear evidence of the formation of a habit of discriminating white from black, but, on the other hand, the method had proved very unsatisfactory because the first test each day usually appeared to be of very different value from the second. On account of the imminent danger of the interruption of the experiment by the rapid spread of an epidemic among my mice, I decided to increase the number of tests in each series to five in order to complete the experiment if possible before the disease could destroy the animals. On the twenty- first day and thereafter, five-test series were given instead of two-test. Unfortunately I was able to complete the experiment up to the point of thirty successive correct tests with only six of the ten individuals whose numbers appear at the top of Table 42. That the results of this table are reliable, despite the fact that some of the individuals had to be taken out of the experiment on account of bad condition, is indicated by the fact that all the mice continued to do their best to discriminate so long as they were used. Possibly the habit would have been acquired a little more quickly by some of the individuals had they been stronger and more active.

It should be explained at this point that the results in all the efficiency-of-training tables of this chapter are arranged, as in the previous white-black discrimination tables, in tens, that is, each figure in the tables indicates the number of errors in a series of ten tests. In all cases A and B mark preliminary series of tests which were given at the rate of ten tests per series. The numbers in the first column of these tables designate groups of ten tests each, and not necessarily daily series. In Table 42, for example, 1 includes the results of the first five days of training, 2, of the next five days, and so on. The table shows that No. 80 made seven wrong choices in the first five series of two tests each. This method of grouping results serves to make the data for the different methods directly comparable, and at the same time it saves space at the sacrifice of very little valuable information concerning the nature of the daily results. It is to be noted, with emphasis, that the two-five tests per day training established a perfect habit after four weeks of training. This method is therefore costly of the experimenter's time.



TABLE 42

EFFICIENCY OF TRAINING. WHITE-BLACK TESTS AT THE RATE OF 2 OR 5 PER DAY

MALES FEMALES SETS 80 82 84 86 88 AV. 73 79 83 85 89 AV. OF 10

A 5 5 4 8 5 5.4 5 6 7 7 6 6.2 B 5 3 6 5 6 5.0 7 5 7 6 7 6.4

1 7 7 6 6 6 6.4 7 6 9 4 6 6.4 2 2 1 0 6 6 3.0 6 5 6 5 5 5.4 3 4 5 5 1 2 3.2 6 5 2 4 1 3.6 4 3 4 7 2 0 3.2 4 3 1 4 3 3.0 5 2 3 3 2 4 2.8 - 3 4 3 1 2.7 6 2 2 - 2 2 2.0 - 0 2 2 0 1.0 7 - 1 - 0 1 0.7 - 1 0 2 1 1.0 8 - - - 1 1 1.0 - 1 1 0 0 0.5 9 - - - 0 1 0.5 0 1 1 0 0 0.5 10 - - - 0 0 0 - 0 0 0 0 0 11 - - - 0 0 0 - 0 0 0 12 - - - 0 0 - 0 0 0



TABLE 43

EFFICIENCY OF TRAINING. WHITE-BLACK TESTS AT THE RATE OF 10 PER DAY

MALES FEMALES SETS 210 220 230 410 420 AV. 215 225 235 415 425 AV. OF 10

A 6 5 6 6 6 5.8 8 4 4 8 5 5.8 B 6 8 8 5 1 5.6 8 7 6 6 2 5.8

1 6 7 6 2 4 5.0 7 6 5 6 4 5.6 2 4 3 1 2 3 2.6 5 6 4 2 5 4.4 3 3 1 4 3 4 3.0 3 3 4 2 5 4.4 4 5 0 3 3 2 3.2 2 1 3 3 3 2.4 5 3 0 4 1 4 2.4 1 3 3 3 3 2.6 6 2 1 4 0 1 1.6 2 1 1 1 0 1.0 7 1 0 3 1 0 1.0 1 1 2 3 3 2.0 8 0 0 1 0 0 0.2 0 0 2 2 3 2.0 9 0 0 0 1 0 0.2 1 0 0 1 1 0.6 10 0 0 0 0 0 2 1 0 2 1.0 11 0 0 0 0 3 0 1 0 0.8 12 0 0 0 0 0 2 0 0.4 13 0 0 0 0 0 14 0 0 0 15 0 0



TABLE 44

EFFICIENCY OF TRAINING. WHITE-BLACK TESTS AT THE RATE OF 20 PER DAY

MALES FEMALES SETS 72 74 208 240 402 AV. 217 230 245 403 407 AV. OF 10

A 4 6 7 7 6 6.0 5 4 7 7 6 5.8 B 6 4 6 8 7 6.2 7 3 5 8 5 5.6

1 3 5 7 5 5 5.0 3 6 4 4 6 4.6 2 4 3 7 5 4 4.6 7 3 5 4 6 5.0 3 3 3 3 5 3 3.4 4 3 3 2 5 3.4 4 6 3 1 4 5 3.8 5 0 1 2 3 2.2 5 4 1 0 2 3 2.0 6 0 0 1 2 1.8 6 3 1 0 2 2 1.6 4 1 1 0 6 2.2 7 3 2 0 1 1 1.4 1 0 0 0 1 0.4 8 2 0 1 1 0.6 0 3 3 0 2 1.6 9 2 1 1 1 1.0 1 0 0 3 0.8 10 1 2 1 0 0.8 0 1 1 2 0.8 11 3 1 0 0 0.8 0 0 0 0 0 12 1 2 0 0 0.6 0 0 0 0 0 13 0 0 0 0 0 0 0 0 14 0 0 0 15 0 0 0

The results of the ten-test training as they appear in Table 43 need no special comment, for quite similar data have already been examined in other connections. In the case of this table it is to be remembered that each figure represents the number of errors for a single day as well as for a series of ten successive tests. The results of Table 44, on the other hand, appear as subdivided series, since each daily series was constituted by two series of ten tests, or in all twenty tests.

Finally, in Table 45 I have arranged the results of what may fairly be called the continuous training method. In connection with several of the labyrinth experiments of Chapter XIII continuous training proved very satisfactory. It therefore seemed worth while to ascertain whether the same method would not be more efficient than any other for the establishment of a white-black discrimination habit. That this method was not applied to ten individuals as were the two-five-test, the ten-test, and the twenty-test methods is due to the fact that it proved practically inadvisable to continue the tests long enough to complete the experiment. I have usually designated the method as one hundred or more tests daily. I applied this training method first to individuals Nos. 51 and 60. At the end of one hundred and twenty tests with each of these individuals I was forced to discontinue the experiment for the day because of the approach of darkness. In the table the end of a series for the day is indicated by a heavy line. The following day Nos. 51 and 60 succeeded in acquiring a perfect habit after a few more tests.



TABLE 45

EFFICIENCY OF TRAINING. WHITE-BLACK TESTS AT THE RATE OF 100 OR MORE PER DAY

SETS 51[1] 60 87 Av. OF 10

A 5 5 6 5.3 B 5 3 7 5.0

1 6 6 5 5.7 2 3 2 5 3.3 3 5 4 7 5.3 4 7 4 5 5.3 5 6 2 3 3.7 6 1 1 3 1.7 7 4 2 3 3.0 8 3 3 0 2.0 9 2 2 3 2.3 10 5 0 2 2.3 11 1 2 2 1.7 12 2 1 1 1.3

13 4 1 2 2.3 14 1 2 1 1.3 15 3 1 5 3.0 16 3 3 2 2.7 17 1 0 1 0.7 18 2 0 1 1.0 19 0 0 2 0.7 20 0 0 0 21 0 1 0.3 22 - 23 - 24 -

[Footnote 1: Age of No. 51, 22 weeks. Age of No. 60, 17 weeks. Age of No. 87, 8 weeks.]

The results of the continuous training method for these two mice were so strikingly different from those yielded by the other methods that I at once suspected the influence of some factor other than that of the number of tests per day. The ages of Nos. 51 and 60 at the time of their tests were twenty-two and seventeen weeks, respectively, whereas all the individuals used in connection with the other efficiency tests were four weeks of age. It seemed possible that the slow habit formation exhibited in the continuous training experiments might be due to the greater age of the mice. I therefore selected a healthy active female which was only eight weeks old, and tried to train her by the continuous training method. With this individual, No. 87, the results were even more discouraging than those previously obtained, for she was still imperfect in her discrimination at the end of two hundred and ten tests. At that point the experiment was interrupted, and it seemed scarcely worth while to continue it further at a later date. The evidence of the extremely low efficiency of the continuous method in comparison with the other methods which we have been considering is so conclusive that further comment seems superfluous.

We are now in a position to compare the results of the several methods of training which have been applied to the dancer, and to attempt to get satisfactory quantitative expressions of the efficiency of each method. I have arranged in Table 46 the general averages yielded by the four methods. Although these general results hide certain important facts which will be exhibited later, they clearly indicate that an increase in the number of tests per day does not necessarily result in an increase in the rapidity of habit formation. Should we attempt, on superficial examination, to interpret the figures of this table, we would doubtless say that in efficiency the two-five-test method stands first, the continuous-test method last, while the ten-test and twenty-test methods occupy intermediate positions.



TABLE 46

EFFICIENCY OF TRAINING

Number of Errors in White-Black Series for Different Methods of Training

SETS OF 10 2 OR 5 TESTS 10 TESTS 20 TESTS 100 OR MORE PER DAY PER DAY PER DAY TESTS PER DAY

A 5.8 5.8 5.9 5.3 B 5.7 5.7 5.9 5.0

1 6.4 5.3 4.8 5.7 2 4.2 3.5 4.8 3.3 3 3.4 3.2 3.4 5.3 4 3.1 2.5 3.0 5.3 5 2.7 2.5 1.9 3.7 6 1.5 1.3 1.9 1.7 7 0.9 1.5 0.9 3.0 8 0.7 0.8 1.1 2.0 9 0.5 0.4 0.9 2.3 10 0 0.5 0.8 2.3 11 0 0.4 0.4 1.7 12 0 0.2 0.3 1.3 13 0 0 2.3 14 0 0 1.3 15 0 0 3.0 16 2.7 17 0.7 18 1.0 19 0.7 20 0



We may now apply to the results of our efficiency-of-training tables the method of measuring efficiency which was mentioned at the end of the preceding chapter as the index of modifiability (that number of tests after which no errors occur for at least thirty tests). By taking the average number of tests for the several individuals in each of the Tables 42, 43, 44, and 45 we obtain the following expressions of efficiency:—



METHOD INDEX OF MODIFIABILITY (EFFICIENCY)

Two-five-test 81.7 +- 2.7 Ten-test 88.0 +- 4.1 Twenty-test 91.0 +- 5.3 Continuous-test 170.0 +- 4.8



Since the difference between the indices for the ten-test and the twenty- test methods lies within the limits of their probable errors (+-4.1 and +-5.3) it is evident that it is not significant. Except for this, I think these indices may be accepted as indications of real differences in the value of the several methods of training.

A somewhat different interpretation of our results is suggested by the grouping of individuals according to sex. In Table 47 appear the general averages for the males and the females which were tested by the several methods. The most striking fact exhibited by this table is that of the high efficiency of the twenty-test method for the females. Apparently they profited much more quickly by this method than by the ten-test method, whereas just the reverse is true of the males. I present the data of this table merely to show that general averages may hide important facts.



TABLE 47

EFFICIENCY OF TRAINING

CONDITION MALES FEMALES INDEX OF MODIFIABILITY INDEX OF MODIFIABILITY 2 or 5 tests per day 85.0 80.0 10 tests per day 72.0 104.0 20 tests per day 94.0 88.0 100 or more tests per day 160.0 180.0



From all considerations that have been mentioned thus far the reader would be justified in concluding that I made a mistake in selecting the ten-test method for my study of the modifiability of the behavior of the dancer. That this conclusion is not correct is due to the time factor in the experiments. If the dancer could acquire a perfect habit as a result of twelve days' training, no matter whether two, five, ten, or twenty tests were given daily, it would, of course, be economical of time for the experimenter to employ the two-test method. But if, on the contrary, the two-test method required twice as many days' training as the five-test method, it would be economical for him to use the five-test method despite the fact that he would have to give a larger number of tests than the two- test method would have demanded. In a word, the time which the work requires depends upon the number of series which have to be given, as well as upon the number of tests in each series. As it happens, the ten-test method demands less of the experimenter's time than do methods with fewer tests per day. The twenty-test method is even more economical of time, but it has a fatal defect. It is at times too tiresome for both mouse and man. These facts indicate that a balance should be struck between number of tests and number of series. The fewer the tests per day, within the limits of two and one hundred, the higher the efficiency of the method of training, as measured in terms of the total number of tests necessary for the establishment of a perfect habit, and the lower its efficiency as measured in terms of the number of series given. The greater the number of tests per day, on the other hand, the higher the efficiency of the method in terms of the number of series, and the lower its efficiency in terms of the total number of tests. By taking into account these facts, together with the fact of fatigue, we are led to the conclusion that ten tests per day is the most satisfactory number.

If my time and attention had not been fully occupied with other problems, I should have determined the efficiency of various methods of training in terms of the duration of habit, as well as in terms of the rapidity of its formation. As these two measures of efficiency might give contradictory results, it is obvious that a training method cannot be fairly evaluated without consideration of both the rapidity of habit formation and the permanency of the habit. A priori it seems not improbable that slowness of learning should be directly correlated with a high degree of permanency. By the further application of the method which I have used in this study of the efficiency of training we may hope to get a definite answer to this and many other questions concerning the nature of the educative process and the conditions which influence it.



CHAPTER XVI

THE DURATION OF HABITS: MEMORY AND RE-LEARNING

The effects of training gradually disappear. Habits wane with disuse. In the dancer, it is not possible to establish with certainty the existence of memory in the introspective psychological sense; but it is possible to measure the efficiency of the training to which the animal is subjected, and the degree of permanency of habits. The materials which constitute this chapter concern the persistence of unused habits, and the influence of previous training on the re-acquisition of a habit which has been lost or on the acquisition of a new habit. For convenience of description, I shall refer to certain of the facts which are to be discussed as facts of memory, with the clear understanding that consciousness is not necessarily implied. By memory, wherever it occurs in this book, I mean the ability of the dancer to retain the power of adaptive action which it has acquired through training.

I first discovered memory in the dancer, although there was previously no reason for doubting its existence, in connection with the ladder-climbing tests of Chapter XII. In this experiment two individuals which had perfectly learned to escape from the experiment box to the nest-box by way of the wire ladder, when tested after an interval of two weeks, during which they had remained in the nest-box without opportunity to exercise their newly acquired habit, demonstrated their memory of the method of escape by returning to the nest-box by way of the ladder as soon as they were given opportunity to do so. As it did not lend itself readily to quantitative study, no attempts were made to measure the duration of this particular habit. At best the climbing of a wire ladder is of very uncertain value as an indication of the influence of training.

Similarly, the persistence of habits has been forced upon my attention day after day in my various experiments with the mice. It is obvious, then, that the simple fact of memory is well established, and that we may turn at once to an examination of the facts revealed by special memory and re- learning experiments.

The visual discrimination method, which proved invaluable as a means of measuring the rapidity of habit formation, proved equally serviceable in the measurement of the permanency or duration of habits. Memory tests for discrimination habits were made as follows. After a dancer had been trained in the discrimination box so that it could choose the correct electric-box, white, red, blue, or green as it might be, in three successive daily series of ten tests each, it was permitted to remain for a certain length of time without training and without opportunity to exercise its habit of visual discrimination and choice. At the expiration of the rest interval, as we may designate the period during which the habit was not in use, the mouse was placed in the discrimination box under precisely the same conditions in which it had been trained and was given a series of ten memory tests with the box to be chosen alternately on the right and on the left. In order that the entire series of ten tests, and sometimes two such series given on consecutive days, might be available as indications of the duration of a habit, the mouse was permitted to enter and pass through either of the electric-boxes without receiving a shock. Had the shock been given as punishment for a wrong choice, it is obvious that only the first test of the memory series would be of value as an indication of the existence of a previously acquired habit. Even under the conditions of no shock and no stop or hindrance the first test of each memory series is of preeminent importance, for the mouse tends to persist in choosing either the side or the visual condition (sometimes one, sometimes the other) which it chooses in the first test. If the wrong box is chosen to begin with, mistakes are likely to continue because of the lack of punishment; in this case the animal discriminates, but there is no evidence that it remembers the right box. Likewise, if the right electric- box is chosen in the first test, correct choices may continue simply because the animal has discovered that it can safely enter that particular box; again, the animal discriminates without depending necessarily upon its earlier experience. I have occasionally observed a series of ten correct choices, made on the basis of an accidental right start, followed by another series in which almost every choice was wrong, because the animal happened to start wrong.

As the results of my tests of memory are of such a nature that they cannot advantageously be averaged, I have arranged in Table 48 a number of typical measurements of the duration of visual discrimination habits. In this table I have indicated the number and age of the individual tested, the habit of discrimination which had been acquired, the length of the rest interval, the result of the first test (right or wrong), and the number of errors made in each series of ten memory tests.



TABLE 48

MEASUREMENTS OF THE DURATION OF A HABIT

Memory

ERRORS NO. AGE NAME OF TEST REST FIRST FIRST SECOND INTERVAL CHOICE SERIES SERIES

1000 25 weeks White-black 4 weeks Right 0 5 27 White-black 4 Right 5 7 210 15 White-black 8 Right 5 220 15 White-black 8 Right 4 230 15 White-black 8 Wrong 5 215 15 White-black 8 Right 5 225 15 White-black 8 Right 2 235 15 White-black 8 Right 7 410 15 White-black 8 Wrong 4 415 15 White-black 8 Wrong 6 420 15 White-black 8 Wrong 3 425 15 White-black 8 Right 3 2 28 Black-white 4 Wrong 9 7 17 Black-white 2 Wrong 1 7 21 Black-white 6 Right 1 7 27 Black-white 10 Right 1 6 998 18 Black-white 2 Wrong 3 998 22 Black-white 4 Right 0 998 28 Black-white 10 Right 5 5 13 10 Black-white 4 Right 3 14 10 Black-white 4 Right 3 15 10 Black-white 4 Right 2 16 10 Black-white 4 Right 4 1000 25 Light blue-orange 4 Right 4 2 28 Light blue-orange 2 Wrong 5 5 28 Light blue-orange 6 Wrong 4 6 3 25 Light blue-orange 4 Wrong 8 10 24 Light blue-orange 2 Right 8 10 26 Light blue-orange 2 Right 5 11 25 Light blue-orange 2 Right 6 11 27 Light blue-orange 2 Wrong 5 151 13 Green-red 2 Right 1 0 152 13 Green-red 2 Right 5 1



This quantitative study of the duration of simple habits of choice showed that in the majority of cases a perfectly acquired habit persists for at least two weeks. To be perfectly fair to the animal I must restrict this statement to visual conditions other than colors, for the dancer exhibited little ability either to acquire or to retain a habit of distinguishing spectral colors. Altogether, I made a large number of white-black and black-white memory tests after rest intervals of four, six, eight, or ten weeks. The results for the four-week interval show extreme individual differences in memory. Number 1000, for example, was able to choose correctly every time in a series of white-black tests after a rest interval of four weeks, whereas No. 5 was wrong as often as she was right after the same interval. I have placed the results for these two individuals at the head of the table because they suggest the variations which render averages undesirable. Number 1000 had a perfect habit at the end of four weeks of disuse; No. 5 had no habit whatever. I shall reserve further discussion of age, sex, and individual differences in the permanency of habits for the next chapter.

With Nos. 7 and 998 memory tests were made after three different rest intervals. At the end of two weeks the black-white habit was present in both individuals, although it was not perfect. After six and four weeks, respectively (see Table 48), it still persisted; in fact, it apparently had improved as the result of additional training after the earlier memory tests. At the expiration of ten weeks it had wholly disappeared. In her first series of memory tests after the ten-week interval No. 7 made only one error, but a chance choice of the black (right) in the first test and the subsequent choice of the box in which no shock had been received serve to account for results which at first appear to be indicative of memory. That this explanation is correct is proved by the fact that a second memory series, in which the first choice happened to be wrong, resulted in six mistakes. Evidently she had lost the habit.

In no instance have memory tests definitely indicated the presence of a habit after a rest interval of more than eight weeks. It is safe, therefore, to conclude from the results which have been obtained that a white-black or black-white discrimination habit may persist during an interval of from two to eight weeks of disuse, but that such a habit is seldom perfect after more than four weeks.

The measurements of memory which were made in connection with color discrimination experiments are markedly different from those which were obtained in the brightness tests. As might have been anticipated (?), in view of the extreme difficulty with which the dancer learns to discriminate colors, the habit of discriminating between qualitatively different visual conditions does not persist very long. I have never obtained evidence of a perfect habit after an interval of more than two weeks, and usually, as is apparent from Table 48, the tests indicated very imperfect memory at the end of that interval. It seems probable that even in these so-called color tests discrimination is partly by brightness difference, and that the imperfection of the habit and its short duration are due to the fact that the basis of discrimination is inadequate. This is the only explanation which I have to offer for the difference which has been demonstrated to exist between the duration of brightness discrimination habits and color discrimination habits.

The duration of a discrimination habit having been measured with a fair degree of accuracy, I undertook the task of ascertaining whether training whose results have wholly disappeared, so far as memory tests are in question, influences the re-acquisition of the same habit. Can a habit be re-acquired with greater facility than it was originally acquired? Is re- learning easier than learning? To obtain an answer to the question which may be asked in these different forms, ten individuals were experimented with in accordance with a method whose chief features are now to be stated. In each of these ten individuals a perfect white-black habit was established by the use of the standard series of tests the order of which is given in Table 12. At the expiration of a rest interval of eight weeks precisely the same series of tests were repeated as memory and re-training tests. In this repetition, the preliminary series, A and B, served as memory tests, and the subsequent training series, as re-training series.

) is the error curve of the original learning process; the broken line (———) is that of the re-learning process, after an interval of eight weeks.]

The striking results of this investigation of re-learning are exhibited in the curves of learning and re-learning of Figure 32. These curves make it appear that the mice re-acquired the white-black discrimination habit much more readily than they had originally acquired it. But in addition to furnishing the basis for some such statement as the foregoing, the curves suggest a serious criticism of the experiment.

In the original tests, the preliminary series indicated a strong preference for black. In series A it was chosen on the average 5.8 times in 10, and in series B, 5.7 times. This preference was rapidly overcome by the training series, and at the end of 130 tests discrimination was perfect. All this appears in the curve of learning (solid line of figure). On the other hand, these preliminary series when repeated as memory tests, after a rest-interval of eight weeks, gave markedly different results. Series A indicated preference for white (5.6 times in 10) instead of black, and series B indicated only a slight preference for black. In brief, series A and B show that the preference for black was considerably stronger at the beginning of the training than at the beginning of the re-training.

In the light of these facts it is fair to claim that the effects of the white-black training had not wholly disappeared as the result of eight weeks of rest, and that the experiment therefore fails to furnish satisfactory grounds for the statement that re-learning occurs more rapidly than learning. I accept this criticism as pertinent, although not necessarily valid, and at the same time I freely admit that the results have a significance which I had not anticipated. But they are not less interesting or valuable on that account. Granting, then, that at least some of the ten individuals which took part in the experiment had not completely lost the memory of their white-black training at the end of eight weeks, it is still possible that an examination of the individual results may justify some conclusion concerning the question which was proposed at the outset of the investigation. Such an examination is made possible by Tables 49 and 50, in which I have arranged separately the results for the males and the females.



TABLE 49

WHITE-BLACK TRAINING. TEN TESTS PER DAY

Males TRAINING RETRAINING 210 220 230 410 420 AV. 210 220 230 410 420 AV.

A 6 5 6 6 6 5.8 5 4 5 4 3 4.2 B 6 8 8 5 1 5.6 8 4 5 4 6 5.4

1 6 7 6 2 4 5.0 3 3 4 7 3 4.0 2 4 3 1 2 3 2.6 2 4 2 5 3 3.2 3 3 1 4 3 4 3.0 1 4 1 4 1 2.2 4 5 0 3 3 2 2.6 0 1 0 1 2 0.8 5 3 0 4 1 4 2.4 0 2 0 2 0 0.8 6 2 1 4 0 1 1.6 0 1 0 0 2 0.6 7 1 0 3 1 0 1.0 0 0 0 0 8 0 0 1 0 0 0.2 0 0 1 0.2 9 0 0 0 1 0 0.2 0 0 0 10 0 0 0 0 1 0.2 11 0 0 0 0 0 12 0 0 0 0 13 0 0 14 15



Only three of the ten individuals failed to re-acquire the habit of white- black discrimination more quickly than it had originally been acquired, and, in the case of these exceptions, No. 220 required exactly the same number of tests in each case, and No. 420 was placed at a slight disadvantage in the re-learning series by an interruption of the training between the seventh and the eighth series. Had his training been completed by the sixth series he too would have had the same number of tests in training and re-training. Moreover, and this is of preeminent importance for a fair interpretation of the results, in several instances even those individuals which exhibited as strong a preference for the black in the memory series as in the preliminary series re-learned more quickly than they had learned. Number 210, for example, although he gave no evidence of memory, and, in fact, chose the black more frequently in the memory series than he did in the preliminary series, re-acquired the discrimination habit in less than half the number of tests which had been necessary for the establishment of the habit originally.



TABLE 50

WHITE-BLACK TRAINING. TEN TESTS PER DAY

Females

TRAINING RE-TRAINING 215 225 235 415 425 Av. 215 225 235 415 425 Av. A 8 4 4 8 5 5.8 5 2 7 6 3 4.6 B 8 7 6 6 2 5.8 8 5 6 4 3 5.2 1 7 6 5 6 4 5.6 4 1 5 4 3 3.4 2 5 6 4 2 5 4.4 1 1 1 2 3 1.6 3 3 3 4 3 4 3.4 1 0 3 6 0 2.0 4 2 1 3 3 3 2.4 0 0 3 3 1 1.4 5 1 3 3 3 3 2.6 0 0 died 2 0 0.5 6 2 1 1 1 0 1.0 0 1 0 0.2 7 1 1 2 3 3 2.0 0 0 0 8 0 0 2 2 3 1.4 1 0.2 9 1 0 0 1 1 0.6 0 0 10 0 2 1 0 2 1.0 0 0 11 0 3 0 1 0 0.8 0 0 12 0 0 0 2 0 0.4 13 0 0 0 0 0 14 0 0 0 15 0 0



The facts which have been presented thus far become more significant when the indices of modifiability for the learning and the re-learning processes are compared.

INDICES OF MODIFIABILITY

LEARNING RE-LEARNING

Females . . . . . . . 104 42.5 Males . . . . . . . . 72 54

The behavior of the mice in the experiments, the detailed results of Tables 49 and 50, and the indices of modifiability together justify the following conclusions. Most of the ten dancers, at the end of a rest interval of eight weeks, had so far lost the habit of white-black discrimination that memory tests furnished no conclusive evidence of the influence of previous training; a few individuals seemed to possess traces of the habit after such an interval. In the case of each group of individuals re-training brought about the establishment of a perfect habit far more quickly than did the original training. This suggests the existence of two kinds or aspects of organic modification in connection with training; those which constitute the basis of a definite form of motor activity, and those which constitute the bases or dispositions for the acquirement of certain types of behavior. There are several indications that further study of the modifiability of behavior will furnish the facts which are necessary to render this suggestion meaningful.

Closely related to the facts which have been revealed by the re-training experiments are certain results of the labyrinth experiments. For the student of animal behavior, as for the human educator, it is of importance to learn whether one kind of training increases the efficiency of similar forms of training. Can a dancer learn a given labyrinth path the more readily because it has previously had experience in another form of labyrinth?

The answer to this question, which my experimental results furnish, is given in Table 51. In the upper half of the table have been arranged the results for six individuals which were trained first in labyrinth B, then in labyrinth C, and finally in labyrinth D. Below, in similar fashion, are given the results for six individuals which were trained in the same three labyrinths in the order C, B, D, instead of B, C, D. My purpose in giving the training in these two orders was to ascertain whether labyrinth C, which had proved to be rather difficult for most individuals, would be more easily learned if the training in it were preceded by training in labyrinth C.



TABLE 51

THE INFLUENCE OF ONE LABYRINTH HABIT UPON THE FORMATION OF ANOTHER LABYRINTH B LABYRINTH C LABYRINTH D NO. OF NO. OF NO. OF NO. OF NO. OF NO. OF NO. FIRST COR- LAST OF FIRST COR- LAST OF FIRST COR- LAST OF RECT TEST FIVE COR- RECT TEST FIVE COR- RECT TEST FIVE COR- RECT TESTS RECT TESTS RECT TESTS

76 8 14 3 19 4 7 78 5 20 6 14 4 5 86 13 22 5 12 3 9 75 4 15 8 19 4 13 77 7 11 11 29 11 12 87 12 22 9 20 4 9

AV. 8.2 17.3 7.0 18.8 5.0 9.2

LABYRINTH C LABYRINTH B LABYRINTH D

58 16 — 2 14 7 10 60 17 — 13 37 10 14 88 25 35 9 22 4 8 49 34 — 1 5 7 8 57 15 — 3 20 3 6 85 11 18 2 11 3 4

AV. 19.7 26.5 5.0 18.2 5.7 8.3



The results are sufficiently definite to warrant the conclusion that experience in B rendered the learning of C easier than it would have been had there been no previous labyrinth training. Those individuals whose first labyrinth training was in C made their first correct trip as the result of 19.7 trials, whereas those which had previously been trained in labyrinth B were able to make a correct trip as the result of only 7.0 trials. Similarly the table shows that training in C rendered the subsequent learning of B easier. To master B when it was the first labyrinth required 8.2 trials; to master it after C had been learned required only 5 trials. In addition to proving that the acquisition of one form of labyrinth habit may facilitate the acquisition of others, comparison of the averages of Table 51 furnishes evidence of the truth of the statement that no results of training can be properly interpreted in the absence of knowledge of the previous experience of the organism.



CHAPTER XVII

INDIVIDUAL, AGE, AND SEX DIFFERENCES IN BEHAVIOR

All dancers are alike in certain important respects, but to the trained observer of animal behavior their individual peculiarities are quite as evident, and even more interesting than their points of resemblance. Omitting consideration of the structural marks of individuality, we shall examine the individual, age, and sex differences in general behavior, rapidity of learning, memory, and discrimination, which have been revealed by my experiments. Observations which bear on the subject of differences are scattered through the preceding chapters, but in no case have they been given sufficient prominence to force them upon the attention of those who are not especially interested in individual peculiarities. It has seemed worth while, therefore, to assemble all the available material in this chapter for systematic examination and interpretation.

In the pages which follow, individual, age, and sex peculiarities are discussed in turn. Within each of these three groups of differences I have arranged in order what Royce has appropriately named the facts of discriminating sensitiveness, docility, and initiative. Individuals of the same age and sex no less than those which differ in sex or age exhibit important differences in ability to discriminate among sense impressions ("discriminative sensitiveness"), in ability to profit by experience ("docility"), and in ability to try new kinds of behavior ("initiative").

Individual differences in sensitiveness to visual, auditory, tactual, and olfactory stimuli have been revealed by many of my experiments. The brightness discrimination tests conclusively proved that a degree of difference in illumination which is easily detectable by one dancer may be beyond the discriminating sensitiveness of another. Both the tests with gray papers and those with the Weber's law apparatus furnished striking evidence of individual differences in the kind of visual sensitiveness which throughout this book has been called brightness vision. I suspect that certain of the differences which were observed should be referred to the experience of the individuals rather than to the capacity of the visual organs, for training improves visual discrimination to a much greater extent than would ordinarily be thought possible. To the truth of this statement the results of the Weber's law experiments with No. 51 bear witness. Likewise in color discrimination there are individual differences, examples of which may be discovered by the examination of the results given in Chapters IX and X.

No differences in auditory sensitiveness appeared in my adult dancers, for in none of them was there definite response to sounds, but among the young individuals differences were prominent. I may call attention to the data on this subject which Table 5, p. 89, contains. The mice in four out of twelve litters gave no indications of hearing any sounds that I was able to produce; the remaining individuals responded with varying degrees of sensitiveness. I made no attempt to measure this sensitiveness, but it obviously differed from mouse to mouse. I feel justified, therefore, in stating that the young dancers exhibit extreme individual difference in sensitiveness to sounds.

My observations of differences in sensitiveness to other forms of stimulation were made in connection with training tests, and although they are not quantitative, I venture to call attention to them. Indeed, I am led by the results of my study of various aspects of the dancer's behavior to conclude that the race exhibits individual differences in discriminating sensitiveness to a far greater extent than do most mammals, not excepting man. The importance of this fact (for I am confident that any one who carefully examines the detailed results of the various experiments which are described in this book will agree that it is an established fact) cannot be overlooked. It alters our interpretation of the results of training, memory, heredity, and discrimination experiments, and it leads us to suspect that the dancing race is exceedingly unstable. I do not venture to make comparison of my own observations of the dancer's sense equipment with those of Cyon, Rawitz, Zoth, and Kishi, for the differences are too great in many instances to be thought of as other than species or variety peculiarities. It has seemed fairer to compare only individuals of the same breed, or, as I have done and shall continue to do throughout this chapter, of two lines of descent.

With respect to docility individual differences are prominent. We need only turn to the various tables of results to discover that in modifiability of behavior, in memory, in re-learning, not to mention other aspects of docility, dancers of the same sex and age differed strikingly. Let me by way of illustration cite a few cases of difference in docility. Number 1000 learned to discriminate white from black more quickly and retained his habit longer than any other dancer with which I have experimented. I should characterize him as an exceptionally docile individual. Table 44 offers several examples. Numbers 403 and 407, though they were born in the same litter and were alike in appearance and in conditions of life, acquired the white-black habit with a difference in rapidity which is expressed by the indices of modifiability 50 and 100. In other words, it took No. 407 twice as long to acquire this habit as it took No. 403. Similarly the ladder-climbing tests revealed important individual differences in ability to profit by experience. In the tables of labyrinth tests (38, 39, 40) individual differences are too numerous to mention. It required forty-nine tests to establish in No. 50 a labyrinth-C habit which was approximately equal in degree of perfection to that which resulted from twenty-two tests in the case of No. 52. The figures in this and other instances do not exaggerate the facts, for repeatedly I have tested individuals of the same litter, the same sex, and, so far as I could judge, of the same stage of development, and obtained results which differ as markedly as do those just cited. If space limits permitted, I could present scores of similar differences in docility which the problem, labyrinth, and discrimination methods have revealed.

In examining the detailed individual results of the various tables for differences of this sort, it is important to bear in mind that sex, age, and descent should be taken into account, for with each of them, as will be shown clearly later in this chapter, sensitiveness, docility, and initiative vary. I have therefore based my statements concerning individual differences in docility upon the results of comparison of mice of the same litter, sex, and age. It is safe to say that human beings similarly selected for comparison do not exhibit greater differences in ability to profit by experience than did these dancing mice.

The facts concerning individual differences in initiative which I have discovered are not less definite than those of the preceding paragraphs. From the beginning of my study of the dancer I observed that what one individual would readily learn of his own initiative another never learned. For example, in the ladder-climbing experiment No. 1000 distinguished himself for his initiative, whereas Nos. 4 and 5 never acquired the habit of escaping from confinement by using the ladder. I noticed, in this test of the animal's ability to learn, that while one individual would be scurrying about trying all ways of escape, investigating its surroundings, looking, sniffing, and dancing by turns, another would devote all its time to whirling, circling, or washing itself. One in the course of its activity would happen upon the way of escape, the other by reason of the limited scope of its activity, not the lack of it, would fail hour after hour to discover even the simplest way of getting back to its nest, to food, and to its companions. Hundreds of times during the past three years I have noticed important individual differences in initiative in connection with the discrimination experiments. The swinging wire doors which one dancer learned to push open before he had been in the box five minutes, another might not become familiar with through his own initiative for hours or days. In fact, it was not seldom that I had to teach an individual to pass from one compartment to the other by gently pushing him against the door until it opened sufficiently to allow him to squeeze through. Occasionally a mouse learned to pull the doors open so that he could pass through the openings in either direction with facility. This was a form of individual initiative which I had not anticipated and did not especially desire, so I did not encourage its development, but, nevertheless, at least one fourth of the mice which I experimented with in the discrimination box learned the trick. The other three fourths, although they were used in the box day after day sometimes for weeks, never discovered that they might return to the nest-box by pulling the swing-door through which they had just passed as well as by entering one of the electric-boxes.

Another indication of individual initiative in action appeared in the tendency of certain mice to climb out of the experiment boxes or labyrinths. It would have been extremely easy for any of the mice to escape from the labyrinths by scaling the walls of the alleys, for they were only 10 cm. in height, and when a dancer stood on its hind legs it could easily reach the top with its nose. But, strange though it will seem to any one who has not worked with the dancer, not more than one in ten of the animals which I observed made any attempt to escape in this manner. They lacked initiative. That it was not due to a lack of the power to climb, I abundantly demonstrated by teaching a few individuals that a scramble in one corner meant easy escape from the maze of paths. I do not think any one of the mice was physically incapable of climbing, but I am confident that they differed markedly, not only in the willingness to try new modes of action, but in the readiness with which they could climb. I have already said that individuals differ noticeably in the scope of their activity. By this statement I mean that they try a varying number of kinds of activity. As in the case of men, so in mice, one individual will do a greater number of things in a few hours than another will in weeks or months. The dancers differ in versatility, in individual initiative, as do we, albeit not so markedly.

Important differences which may with certainty be described as age differences are not so obvious as are such marks of individuality as have been set forth in the preceding pages. I have noted few changes in discriminative sensitiveness, other than those with regard to auditory sensitiveness, which could be correlated with age. In certain instances adults appeared to be able to discriminate more accurately and more easily than young mice, but it is difficult to say whether this change belongs under sensitiveness or docility. I have not made an ontogenetic study of the senses, and I am therefore unable to describe in detail the course of their development and decline. Of one important fact I am certain, that discriminative sensitiveness increases up to a certain point with age and with training.

Differences in docility which are obviously to be correlated with age abound. In the prime of its life (from the second to the tenth month) the dancer is active, full of energy, quick to learn; in its senility (during the second year) it is inactive, but at times even more docile than during the period of greatest physical development. Frequently I have noticed in connection with labyrinth tests that individuals of the age of a year or more learn much more quickly than do individuals of the age of two or three months. But, on the other hand, I have contradictory observations, for now and then I obtained just the opposite result in experiments to test docility. Evidently this is a matter which demands systematic, quantitative investigation. Casual observation may suggest conclusions, but it will not justify them.

Early in my investigation of the behavior of the dancer I conceived the idea of determining the relation of modifiability of behavior (docility) to age. The question which was foremost in my mind and for which I first sought an answer may be stated thus: can the dancer acquire a given habit with the same facility at different ages? Since the visual discrimination experiment seemed to be well suited for the investigation of this problem I planned to train, in the white-black discrimination experiment, five pairs of dancers at the age of one month, and the same number for each of the ages four, seven, ten, thirteen, sixteen, and nineteen months.[1]

[Footnote 1: I have not been able thus far to determine the average length of the dancer's life. The greatest age to which any of my individuals has attained is nineteen months.]

To test the same individuals month after month would be the ideal way of obtaining an answer to our question, but I could devise no satisfactory way of doing this. The effects of training last so long, as the results of the previous chapter proved, and the uncertainty of their entire disappearance is so serious, that the same training process cannot be used at successive ages. The use of different methods of training is even more unsatisfactory because it is extremely difficult to make accurate quantitative comparison of their results. It was these considerations that forced me to attempt to discover the relation of docility to age by carrying out the same experiments with groups of individuals of different ages.

As my plan involved the execution of precisely the same set of tests with at least seventy individuals whose age, history, and past experience were accurately known, and of which some had to be kept for nineteen months before they could be trained, the amount of labor and the risk of mishap which it entailed were great. To make possible the completion of the investigation within two years, I accumulated healthy individuals for several months without training any of them. In March, 1907, I had succeeded in completing the tests for the age of one month, and I had on hand for the remaining tests almost a hundred individuals, whose ages ranged from a few days to eighteen months. Had everything gone well, the work would have been finished within six months. Suddenly, and without discoverable external cause, my mice began to die of an intestinal trouble, and despite all my efforts to check the disease by changing food supply and environment, all except a single pair died within a few weeks. Thus ended a number of experiments whose final results I had expected to be able to present in this volume. However, the work which I have done is still of value, for the single pair of survivors have made possible the continuance of my tests with other individuals of the same line of descent as those which perished, and I have to regret only the loss of time and labor.