The doctrine of formal discipline in the light of experimental investigation

Front matter

W~~rwW

■/yW

Class.

Part I. Historical

I. Aims of Investigators 4

Psychological 4

1. Experiments on the effect of training of one kind

of sensitiveness on other kinds of sensitiveness 4

2. Experiments on the accm-acy of voluntary ef-

fort and the effect of special training on the general rapidity and accuracy of motor adjustments 5

3. Experiments on the effect of special training on

the general rapidity and accuracy of memorizing 5

4. Experiments on cross-education or the transfer-

ence of practice from one member to a symmetrical one 6

Pedagogical 7

II. Means, Methods, Results and Conclusions of Experimental Investigations 9

Psychological Experiments 10

1. Effect of training of one kind of sensitiveness

on other kinds 10

2. Experiments on the accuracy of volimtary effort

and the effect of special training on the general rapidity and accuracy of motor adjustments 15

3. Experiments on the effect of special training on

the general rapidity and accuracy of memorizing 25

4. Data derived from experimentation in regard

to the effect of the training of one organ upon the bilaterally symmetrical one, or closely related member 28

vii

Vlll CONTENTS

Pedagogical Experiments 31

1. Mathematics 31

2. Spelling 36

3. English Grammar 36

4. Mental Traits 37

A. Memory 37

B. Habits 42

C. Concentration of attention 42

D. Observation 43

E. Quickness, accm-acy, attention, etc 43

F. Ideas of Method 44

G. Ideals 44

III. Summary of the Historical Review 46

Pakt II. Original Investigation

I. Introduction 49

II. Subjects 50

III. Time Schedule of Tests 50

IV. Practice Series 51

V. Method of Conducting the Tests 52

VI. Details of the Tests 53

VII. Marking the Tests and Use of the Scores 71

VIII. Tables and summaries 74

CONCLTJSIONS Ill

Comparison of Results with Those of Pedagogical Experiments 113

Bibliography 115

Appendix 119

The Doctrine of Formal Discipline in the Light of Experimental Investigation

Introduction

It would seem to be entirely superfluous to explain what is meant by the doctrine of formal discipline, for since the epoch-making experiments of James (25, 54)1 a little more than a decade ago, this theory has occupied the center of the controversial stage of pedagogical problems. Since the Middle Ages, when the doctrine flourished in justification of the classical learning, or even from earlier times as claimed by Locke (31) in his statement "Formal disciphne has been invoked from Plato's time to the present as a defense of the courses in mathematics," the theory of formal discipline has been held as almost axiomatic, but with the abandonment of the old ''faculty" psychology, it has lost its main prop and its subsequent subjection to experimental investigation has left the theory on a somewhat unstable foundation. We cannot estabhsh the truth or the falsity of the doctrine on a 'priori grounds in this scientific age, but must subject it to many careful, scientifically conducted and controlled experiments covering a wide range of school subjects and performed with school children of various ages, grades, mental attainments, and environments; for its reliability cannot be based longer upon

1 Reference numbers in parentheses refer to the bibliography at the end of the book.

Z THE DOCTRINE OF FORMAL DISCIPLINE

psychological experiments, pure and simple, performed upon a few adults in college laboratories with complex apparatus.

The experimental investigations until within the last few years have been mainly psychological, conducted by professors of colleges with college or university students or instructors as subjects, necessarily few in number; and only recently have tests been made with school children. Of these, a small number have been tried in this country, and they have been for the most part confined to the elementary schools. Judd (28) describes formal discipline as the "great problem of High-School education." The subjects of the elementary school are studied mainly for their content value and their practical relationship to the future normal life of the child fitted for the demands of his environment. Grammar may be a possible exception, although this should be taught from the standpoint of improving the child's English, rather than as a mental discipline. In the high school, however, nearly every study claims as one of the best reasons for its introduction into, or continuance in, the course of study, some definite mental training which will carry over into other walks of life, however remotely related. Mathematics is said to train the reasoning powers, science the observational abilities, literature and history the imagination, etc. Thus we can agree with Judd (28) that the ''final answer of the question will relate to the work of the High School." We must not argue or speculate about the matter, but we must investigate these claims for the various studies and support them by definite tests or refute them by similar carefully conducted and controlled experiments. Many experiments will be required

INTEODUCTION 6

upon all the subjects and under varying conditions, before we can generalize scientifically as to the validity of the doctrine. But with the question of the value of the disciplinary view of the studies in the balance, it would be well to look to the content side until such time as the theory has been definitely proved or disproved. Burt's (10) advice is to ''emphasize and extend the elements shared by the curriculum and the conditions of school in common with the conditions and requirements of life."

The experiments performed and described by the author have been undertaken with the object of contributing a mite to the countless ones necessary to determine the validity or falsity of the disciplinary conception^ especially as applied to high-school subjects.

PART I. HISTORICAL.

I. Aims of Investigators.

Reviewing the experimental researches on this topic, we find that while the aims of the investigators have been manifold, they may be grouped under several main divisions. First, we may classify them as primarily psychological or primarily pedagogical.

Psychological.

1. Experirnents on the effect of training of one kind of sensitiveness on other kinds of sensitiveness. These include those of Bennett (6) to test the effect on discrimination of length by the eye as a result of practice in discriminating length by arm^mov ements; those of Wallin (58, 60) to attempt to control the reversions in a number of reversible perspective outlines; those of Seashore and Jenner (42) on the training of the voice by the aid of the eye in singing; those of Coover and Angell (12) to test the general practice effect of special exercise, including experiments on the transfer of practice effects in sound to light discriminations, and the transference of practice in sorting cards to ''typewriter reactions"; those of Thorndike and Woodworth (53) on the effects of practice in estimating length of lines on estimating length of lines, areas, and weights; on the effects of practice in picking out and marking words of some one special sort on the observation of words containing certain combinations of letters and picking out and marking certain letters; and those of Urbantshitsch (56) on the effects of

AIMS OF INVESTIGATORS 5

practice with sound stimuli on tactual, gustatory, olfactory, and visual stimuli.

2. Experiments on the accuracy of voluntary effort and the effect of special training on the general rapidity and accuracy of motor adjustments. These experiments include those of Davis (14) to test the accuracy of voluntary effort in lunging at a target, those of Jastrow (27) to test quickness of response to touch and visual stimuli; those of Gilbert and Fracker (reviewed by Thorndike, 54) to test quickness in moving the finger at a given signal; those of Swift (52) to test the acquisition of skill in the complex muscular act of keeping three balls in the air, and in typewriting; those of Judd (2) to test the influence of training in the judgment of direction of lines, and the transfer of practice in the Mijiler-Lyer illusion; those of Foster (18) to test the effect of practice upon visualizing and upon the reproduction of visual impressions; those of Whipple (62) to test the effect of practice upon the range of visual attention and visual apprehension; those of Bergstrom (7) on sorting cards, wherein was tested the interference of habits formed with ability to perform opposite acts; those of Bair (21, 54) to test effects of special training on general power to meet new situations; those of Munsterberg (21, 35) to test whether a habit associated with a given sensory stimulus can continue automatically, while some effect of a previous and different habit associated with the same stimulus remains; those of Ruger (41) to test the transfer of specific motor habits. 3. Experiments on the effect of special training on the general rapidity and accuracy of memorizing. Here are included those of James (21, 25, 54) to test improvement in memory after special training in memorizing b THE DOCTRINE OF FORMAL DISCIPLINE

poetry; those of Ebert and Meumann (2, 21) to test the effect of practice in memorizing meaningless syllables on the power of immediate recall and retention of numbers, letters, monosyllabic nouns, words of Italian prose and poetry, visual signs, etc.; those of Fracker (19, 21) to test the effect of practice in memorizing four tones on memorizing poetry, four shades of gray, nine geometrical figures, nine numbers, extent of arm movements, etc.; those of Jansen (26) to test similar points to those of Prof. Meumann; those of Bennett (6) to test the effect of practice in memorizing poetry on memorizing rows of figures and names of places; those of Thorndike and Woodworth (21, 53) to test the influence of special training in memorizing on the general ability to memorize.

4. Experiments on cross-education or the transference of practice from one member to a symmetrical one. These comprise those of Davis (14, 54) to test transference of motor ability from the practised right hand to the unpractised left in lunging at a target; those of Scripture (44) to test transference of ability from the right hand practised on a mercury dynamometer to the unpractised left hand; those of Scripture, Smith, and Brown (45, 54) to show transfer from the right hand, practised on inserting a needle into a hole without touching the sides, to the unpractised left hand; those of Raif (37, 44), who performed similar experiments in piano playing; those of Wallin (58, 60) to attempt to control the reversions in a number of reversible perspective outlines, in which he found that practice with one eye afforded practice for the other unused eye; those of Volkmann (21, 54), reviewed by Henderson and Thorndike, showing the effect of practice in training the sensitiveness of the skin of the left arm,

AIMS OF INVESTIGATORS 7

upon the right arm; those of Swift (52) to test the acquisition of skill in the complex muscular act of tossing three balls, in which he found that practice transferred from the right to the left hand; those of Starch (50), who found in the experiment of tracing the outline of a six-pointed star as seen in a mirror that improvement was transferred from the left to the right hand; those of Davis (14, 54) to show the transfer of tapping ability with the toe of one foot to the unpractised toe and to show the increase in girth of the right arm and the left arm through lifting a weight with the right arm; and those of Woodworth (54) showing a transfer of practice in hitting a dot with the right hand to the unpractised left.

Pedagogical.

In the pedagogical field, the aims of the experimental investigations have been more limited. The experiments with school children have been confined to few school subjects and few mental abilities. Miss Aiken's (1) pioneer experiments, described in 1896, and her astounding results argue well for the validity of the doctrine of formal discipline. Recently, Dallenbach (13) has made Aiken's experiments, together with the laboratory studies of Whipple (62) and of Foster (18) on visualization, the basis of an exhaustive scientific test on elementary school-children. His results tend to support those of Miss Aiken and lead to the conclusion that experiments with adults are not reliable tests of what occurs with school-children.

Pedagogical experiments relating to this topic may be grouped under those concerned with college students and those concerned with school children. The school subjects utilized for investigatory purposes have been arithmetic, spelHng, and grammar and the mental abiUties, memory, habits, and ideals. Arithmetical experiments with college students include those of Starch (49) to test the transfer of training in arithmetical operations; those of Lewis (30) and Collins (11) to ascertain the relation between mathematics and general reasoning; and those of Rietz and Shade (38) to discover the correlation of efficiency in mathematics and in other studies, while with elementary or high-school pupils Winch (67, 68) has experimented to discover if improvement in numerical accuracy transfers, Lewis (30) has found interesting data in regard to the relation between mathematical and practical reasoning, and Stone (51) has tested the relation between distinctive procedures in arithmetic work and the resulting abilities. Wallin (59) has tested elementary pupils to ascertain if spelling efficiency acquired in column drill transfers to dictated compositions, and Briggs (9) has investigated the disciplinary effects of formal English grammar.

With reference to mental abilities, memory has received most attention, although reason has been given some attention, as in the experiments of Lewis and Collins, cited above. College students have been used by Sleight for experimentation in regard to the transfer of memory, while elementary pupils have been made subjects of investigation by Winch to test immediate memory, both visual and auditory, and to test the transfer of improvement in memory; and by Sleight (46, 47) to ascertain if memory training is general or specific.

The effect of special habits on the general conduct or habits of school children has been tested by Squire (3) and by Ruediger (40). Squire's aim was to test the transfer of neatness and accuracy in arithmetic

AIMS OF INVESTIGATORS 9

papers to other school subjects, while that of Ruediger was to test the influence of ideals of neatness in improving the written work of the 7th-grade school children. Judd (2) in his target experiment has demonstrated the value of a recognized method.

There is urgent need for investigation in regard to other school subjects and other mental powers if the problem is to be solved satisfactorily, and it is to be hoped that the experiments may be undertaken by the classroom teacher, drilled in the methods of experimental pedagogy, for only then will there be sufficient trained investigators and optimal conditions for work that may produce unassailable conclusions.

11. Means, Methods, Results and Conclusions OF Experimental Investigations.

The means employed by the various investigators have been almost as numerous as their aims. Under this heading are included apparatus and materials used and the subjects of the investigation. In many cases, especially in the psychological experiments, the apparatus has been quite complicated, while the subjects have been generally few in number and usually those possessing psychologically trained minds. The same is true of several of the pedagogical experiments, while, as has been previously stated, few school children, especially in this country, have been used as reagents.

As widely different as are the aims and materials of the many investigators of the doctrine of formal discipline, just as widely divergent are the methods pursued.

The results and conclusions obtained will warrant a separation of the investigators into two groups — those in favor of the doctrine and those opposed.

10 THE DOCTRINE OF FOEMAL DISCIPLINE

Psychological Experiments.

1. Effect of training of one kind of sensitiveness on other kinds. The apparatus used by Bennett (6) in discrimination of length by the eye as a result of practice in discriminating length by the knowledge gained from arm-movements, consisted of 2 parallel, horizontal, steel rods, 2 feet long and 13^2 inches apart. On the lower rod were two spools fixed immovably 25 cm. apart, and on the upper rod were two spools which were freely movable. In the training the two subjects were required to move the two movable spools on the upper rod to equate their distance to that of the immovable spools on the lower rod, through the sensation derived by arm-movement. The final tests were visual estimates of length. The result of the experiment was negative; one subject showed improvement, but the other an equal degree of lack of improvement.

In the experiments of Wallin (58) to attempt to control the reversions in a number of reversible perspective outlines, the apparatus consisted of drawings of a parallelopiped, a book, a table, and a pyramid. Two subjects practiced in attempting to envisage uniformly the infrequent or non-predominant perspective, in all cases monocularly. Wallin found that perspectivity was subject to a high degree of practice control. In 9,246 trials, the per cent, of successful control in three groups amounted to 40, 62, and 82. There was an average gain of 42 per cent, between the first and last 20 days of the series. He also found that the practice effects were transferred to the unpractised eye. ''In a given series of experiments in which the figures were reversed in the direct and indirect visual fields, it appeared that the reversions occurred about 23^ times faster when the figures were directly regarded instead of being seen by the peripheral retina. But after practice with certain figures, it sometimes happened that the figures reversed most readily when a point outside the figures was fixated." This furnishes an instance of the transference to the peripheral retina of fixation motives attaching to the fovea. Here an acquired foveal disposition spread to the adjacent retinal elements, or the foveal tendency was transmuted into a ''generalized retinal habit."

Seashore and Jenner's (42) experiments on the training of the voice by the aid of the eye in singing, made use of the voice tonoscope. Each of the six observers sang before the speaking tube. The tests were divided into: (1) accuracy in striking a required pitch, (2) the least producible change in pitch (minimal change). The preliminary practice was given with six observers. Each period of practice consisted of 160 trials, which took about 45 minutes. Tests were given for 12 days; the first 5 days "without aid," the singer depended entirely on the ear; and the next 5 days with aid, that is, the record was read from the tonoscope in each trial ; the eleventh day the record was taken without aid again; and the record of the twelfth day was taken with the aid. Their conclusions pertinent to the topic in hand are: (1) "The aid enhances the ability to strike a tone which has been heard." There was 42 per cent, superiority of the aided series over the unaided. (2) "The aid enhances the ability to sing an interval." The aided series was superior to the unaided 50 per cent, in singing a major third, 50 per cent, in singing a fifth, and 60 per cent, in singing an octave. (3) The voluntary control of the pitch of the voice is improved 26 per cent, by the aid.

(4) ''There is probably some transfer of gain from the aided training to the following unaided singing."

(5) "There is no evidence of the transfer of gain in the accuracy of the memory image." (6) "The gain in the discriminative control of the pitch of the voice is fully transferred. (7) "Improvements in the ability to strike a tone, or an interval, and the ability to produce a minimal change are very much more pronounced and more rapid in the aided than in the unaided series." These seven points show that "by the use of the tonoscope we facilitate in a decided manner the ability to sing in true pitch and the improvement is almost immediate." (8) It seems probable "that a higher degree of accuracy of pitch in singing may be attained by aiding the ear in the training than would be possible to attain without such aid."

Coover's and Angell's (12) investigations to test the general practice effect of special exercise included experiments on the transfer of practice effects in sound to light discriminations and the transfer of practice in card-sorting to "typewriter reactions." In the former experiment, four reagents were trained by means of a sound pendulum in discrimination of intensities of sound for 17 days during a period of 57 days, and each reagent made 40 judgments in each day's training. Before and after training, the reagents were tested in the discrimination of shades of gray. Each test consisted of three series, of 35 judgments each, delivered on three separate days. Control reagents were given the tests. It was found that there was an improvement in three of the trained subjects of 7, 5, and 15 per cent, respectively, but a failure in one of them; the control reagents showed a

EXPEEIMENTAL INVESTIGATIONS 13

loss in discriminative ability in the final tests. Coover and Angell concluded that the '^ improvement seems to consist of divesting the essential process of the unessential factors," that "efficiency of sensible discrimination acquired by training with sound stimuli has been transferred to the efficiency of discriminating brightness stimuli and that the factors in the transfer are due in great part to habituation and to a more economic adaptation of attention, i. e., are general rather than specific in character." In their second experiment just referred to, in which reaction with discrimination (choice) was tested, four reagents were trained through four days in card-sorting, during which 4000 cards were distributed by each reagent. They were tested for five days on 'typewriter reactions" — about 3000 reactions in each case; and after training in card-sorting, an after test was given for three days which included about 1800 reactions. The results warranted the conclusion that ''training in card-sorting is the cause of the increased ease and facility experienced by the regular reagents in the second trial in 'typewriter reactions'"; but this improvement is not considered due to "identical elements." "The general condition that is common to both is the habit of stripping the essential process of unnecessary and complicating accessories." The cause of transference of facility is the formation of a habit of reacting directly to a stimulus, which results in an equitable distribution of attention and the consequent power of concentrating the attention.

Thorndike and Woodworth (53) in their tests of improvement in mental functions used magnitudes, such as lines of various lengths, areas of different sizes, and weights of several magnitudes; the practice work was on estimating magnitudes of the same general sort. They also tested the ''influence of training in observing words containing certain combinations of letters or some other characteristic, on the general ability to observe words." The subject practised picking out and marking words of some one special sort until a high degree of proficiency was attained. From four to six subjects were used in the various experiments, all of which tested the "influence of improvement in a function on other functions closely allied to it."

Their conclusions were as follows: "It is misleading to speak of sense discrimination, attention, memory, observation, accuracy, quickness, etc., as multitudinous separate individual functions are referred to by any one of these words." "There is no reason to suppose that any general change occurs corresponding to the words 'improvement of the attention,' or 'of the power of observation,' or 'of accuracy.'" "Improvement in any single mental function rarely brings about equal improvement in any other function, no matter how similar, for the working of every mental functiongroup is conditioned by the nature of the data in each particular case." "The general consideration of the cases of retention or of loss of practice effect seems to make it likely that spread of practice occurs only where identical elements are concerned in the influencing and influenced functions."

Urbantschitsch (56) studied pathological cases with diseases of the eye and ear and found that stimulation also affects parts seemingly not immediately concerned. An hour's operation on the right eye showed on the left a relative enhancement of the capacity to see. In many patients with chronic catarrh of the middle ear, the observer was surprised to find that an important pathological influence was transferred from the ear to vision. In a later investigation he demonstrated that the sensitivity for tactual, gustatory, olfactory, and visual stimulation can be increased by practice with auditory stimulation.

To summarize the conclusions of the investigators upon the topic of the effect on other kinds of sensitivity of training one kind of sensitivity, we find that Wallin, Seashore and Jenner, Coover and Angell, and Urbantschitsch subscribe to the doctrine of transfer, but Thorndike and Woodworth are opposed. Bennett's tests must be disregarded, as the two reagents obtained results diametrically opposed. The results of Wallin and Bennett could be set aside on the ground of having too few subjects, unless their results were subsequently verified by other observers. In all of the tests, the subjects were too few in number, but the results of Seashore and Jenner, of Coover and Angell, and of Urbantschitsch, seem to place the preponderance of scientific conclusion on the affirmative side.

2. Experiments on the accuracy of voluntary effort and the effect of special training on the general rapidity and accuracy of motor adjustments. Here, many diverse methods have been used, with resultant dissimilarity of results and inferences.

Davis (14) used a target and a fencer's foil, and worked with six subjects. He had 10 thrusts made with the right hand and recorded on a paper target. After five minutes, the record of 10 thrusts with the left hand was made on another paper target. The initial tests were repeated after 10 days' practice of 10 thrusts daily with the right hand. It was found that the subjects could be educated in accuracy and co-ordination of voluntary movement. The left hand improved from 57.2 per cent, to 36.5 per cent, nearer the center of the target ; while the right hand improved from 50.5 per cent, to 24.3 per cent. Increase in accuracy is due to practice in lunging. His experiments did not test the transfer of this special training to other motor adjustments, but only to the bilaterally symmetrical organ.

Jastrow (27), in testing quickness of response to tactile and visual stimuli, experimented with two *'Sleight-of-Hand Experts," Kellar and Hermann, and with miscellaneous individuals for the control. The tests included measurements with the aesthesiometer, judgments of weights of different magnitudes, and tests of sensitiveness to texture, for which coils of different sized wire were utilized. The results showed that Hermann's and Kellar's tactual and muscular perception were below normal. With the Weber compass, Hermann could distinguish the points at 3.5 mm. distance, Kellar at 2.5 mm., and the average for miscellaneous individuals at 2 mm. Herman was unsuccessful, Kellar successful, in arranging in correct order five weights, each of which weighed y^ more than the previous weight. In a general series of tests, 92 per cent, of those tested arranged the series correctly. In the tests with the different sized wires, "both Mr. Hermann and Mr. Kellar succeeded in arranging both series correctly, but this was also done by nine out of ten persons who were tested in the same way." Quite a number of tests of the quickness of movement and of mental processes, and of visual perception, were made. Some of the ordinary forms of reaction experiments were also tried. Tests to find out the comprehensiveness of perception were likewise given. Selecting those tests in which the records of Hermann and of Kellar differ markedly from the normal, Jastrow finds: "In the quickness of response to a touch and a visual stimulus both the special subjects, and Mr. Kellar as well in response to an auditory stimulus, excelled to a considerable extent the average individual." But in the ''most complicated reaction they both fall considerably below the normal." There are decided indications of unusual quickness for both; but no very decided excellence appears in the scope and accuracy of visual perception. "In tests involving mainly tactual and muscular perception, the indication is rather that they are below than above the normal." Thus Jastrow concludes that "the positive results of the investigation are thus small, but as far as they go they are consistent with the forms of dexterity that are utilized in sleight-ofhand performances." "They also indicate that it may well be that special skill in one very specialized form of training may be only slightly influential upon other forms of capacity."

In Gilbert's and Fracker's (54) experiments, reviewed by Thorndike, two subjects were tested for their quickness in moving the finger, (1) when they heard a certain sound, (2) when they felt a certain electric shock, (3) when they felt a certain blow, and (4) when they saw a blue surface. They were also tested for their quickness in moving the finger at these same stimuli when either the given sound or one less loud, either the given shock or one less intense, either the given blow or one less hard, and either a blue or a red might appear. They were trained for a number of days in quickness in reacting to the sound, (a) when only it was given, (b) when either it or the weaker

18 THE DOCTEINE OP FORMAL DISCIPLINE

sound might be given. Then they were tested as before. It was found that there was an improvement in all cases, except one case with one observer. One observer was trained only in (a) quickness in reacting to sound. He improved markedly in the corresponding tests, but not so much as the others in the second set of tests. Thorndike explains this improvement on the ground of "identical elements." Elements common to all these tests are concentrated attention, alertness, effort, co-ordination of sensory and motor areas, and rapidity of discrimination. -' Swift's (52) experiments on the testing of the acquisition of skill in keeping two balls going with one hand, were made with solid rubber balls. There were six subjects — five university students and one professor. Ten series were given; each of which consisted in throwing until failure to catch one or both balls. Practice was with the right hand, but there was a preliminary test with the left. He discovered that progress was at first slow and then more rapid. There was great irregularity of advance. Progress was by ''jumps." A ''warming up" is necessary but "steady and calm intensity makes for progress." He also found transference of practice effects from the right hand to the left. Swift, himself, performed an experiment which combined physical and mental tests, in the acquisition of skill in typewriting. He wrote one hour each day from copy. Here, he also found that the rise of the practice curve is rapid, but that there is irregularity from day to day. In both of Swift's experiments, there is no test of the transfer to other abilities of special ability gained by practice. It was discovered that training of the right hand was effective upon the left, but he does not believe that this gives any support to the doctrine of ''formal education." "There is no evidence to show that training has general value."