Open Access

JSON Export

{"created":"2022-01-31T15:21:10.623428+00:00","id":"lit28747","links":{},"metadata":{"alternative":"Studies from the Yale Psychological Laboratory","contributors":[{"name":"Johnson, W. Smythe","role":"author"}],"detailsRefDisplay":"Studies from the Yale Psychological Laboratory 6: 51-103","fulltext":[{"file":"p0051.txt","language":"en","ocr_en":"RESEARCHES IN PRACTICE AND HABIT,\nBY\nW. Smythe Johnson.\nI. Triangular movement.\nThe subject was required to tap continuously at the corners of an equilateral triangle whose sides measured 2o0m. This triangle was formed by a special triangular contact key, Fig. i, with knobs 2o0m from each other. This key was originally constructed at Dr. Scripture\u2019s suggestion for use in testing school-children by Gilbert, by whom, however, it was used merely for tapping and not as a habit-key.1\nThe key was placed in circuit with a 4 amp\u00e8re battery and the primary coil of a spark coil, the condenser being connected around the break. From the poles of the secondary coil, one wire led to the base of the recording drum, the other to the base of a ioo v.d. electric fork bearing a flexible point on one of its prongs. Pressure on one of the key-knobs closed the primary circuit for an instant. When the circuit was broken a spark passed through the smoked paper on the surface of the drum making a dot on the time line drawn by the fork. Each spark thus indicated a tap on one of the three key-knobs. The time between the sparks could be read to the thousandth of a second.2\nThe subjects included : K. (Kochi) and M. (Matsumoto), who were Japanese students of psychology ; P. (Powell), a student of English ;\n1\tGilbert, Researches on the mental and physical development of school-children, Stud. Yale Psych. Lab., 1894 II 40 (especially p. 40 and Fig. 5).\n2\tThe arrangement of the recording apparatus was identical with that of Exercise IX in Scripture, Elementary course in psychological measurements, Stud. Yale Psych. Lab., 1896 IV 113 ; it is shown in Scripture, New Psychology, frontispiece, London 1897.","page":51},{"file":"p0052.txt","language":"en","ocr_en":"52\nW. Smythe Johnson,\nL. (Lloyd), a student of philosophy; H. (Hawkins), a student of divinity; S. (Dr. Scripture); and J. (Johnson).\nBefore beginning the experiments the subjects were requested to devote their attention and energy to the exercise throughout each experiment and always to make as fast a record as possible. They were also instructed to practice no similar movements at any time outside that of experimentation. They were not allowed to inspect their records nor were they given the least intimation whether they were gaining or losing ; exceptions to this rule occurred in the cases of J. and S.\nThe person experimented upon was placed in a quiet room.1 The triangular key lay on a table. The subject stood while performing the experiment, as it was found that this position allowed the freest movement of the forearm. He fixed his gaze on the key. Before beginning an experiment a preliminary trial was allowed in order to acquaint the fingers with the relative positions of the keys. The subject was also asked to endeavor to regulate the tension of his muscles commensurate to the strain of tapping fifty successive times. Unless thus cautioned, he was almost certain to break down before the end. If he did not break down completely, he would at least miss some of the knobs in making the succession of movements. The caution was given only at the beginning of the whole set of experiments as it was merely intended to prevent the adoption of too high a standard in the beginning. It was not again mentioned lest the mental standard chosen in the first experiment should be changed. It was found that the subjects followed this request with respect to the left hand more closely than with respect to the right hand.\nWith all the subjects except S. and J. the experiment was first performed with the right hand, then with the left, and again with the right. In the case of J. only the right hand was used and in that of S. only the left hand.\nFatigue showed itself (i) when the subject completely broke down, (2) when he struck the knob of the key so inaccurately that he knocked it out of place, (3) when he missed one of the knobs of the key. The last case was the usual one and when it occurred the experiment was considered to have ended. By counting the records to this point we have a result practically free of fatigue.\nDaily average.\nThe average tapping time for each day is shown in Table I. Curves\n'Described by Scripture, New Psychology, 136, London 1897.","page":52},{"file":"p0053.txt","language":"en","ocr_en":"Researches in practice and habit.\n53\ncorresponding to the table are presented in Figs. 2, 3, 4, 5, 6. A comparison of these curves shows that, while each takes a direction determined by individual characteristics, yet they all follow closely the saine law of gain.\nTable I.\nAverage interval betiveen taps on successive days.\nHand\tSerial number of experiment.\nSubject.\tused.\tI\t2\t3\t4\tS\t6\t7\t8\t9\tIO\t11\nK.\tt R\t3\u00b0S\t283\t280\t277\t268\t242\t241\t213\t212\t209\t203\n\tX L\t3\u00b0\u00b0\t292\t3\u2018fi\t278\t276\t258\t243\t239\t234\t230\t226\nM.\t\\R\t294\t314\t302\t287\t271\t281\t242\t239\t235\t229\t206\n\tXL\t3H\t321\t3\u00b02\t28S\t262\t2S1\t274\t273\t253\t250\t187\nS.\tL\t2I9\t23s\t217\t201\t197\t188\t185\t. 182\t172\tI62\t\nJ-\tR\t224\t219\t210\t206\t230\t203\t195\t191\t171\t\t\nP.\t$ R\t244\t196\t175\tIS7\t152\t143\t\t\t\t\t\n\t< L\t226\t203\t173\t198\t186\t174\t\t\t\t\t\nL.\t\t245\t210\t194\t174\t169\t157\t\t\t\t\t\n\tXL\t26I\t244\t223\t199\t192\t1S6\t\t\t\t\t\nH.\t5?\t212\t169\t159\t\t\t\t\t\t\t\t\n\tX L\t24O\t174\t172\t\t\t\t\t\t\t\t\nThe unit of measurement is x<r = 0.001s.\nThe number of measurements in each case was from 40 to 50.\nThe probable errors of the determinations vary from \u00b15 \u25a0 to \u00b1 it.\nAn omission of one or more days did not in every instance materially effect the amount of gain resulting from practice. An interval of two days between the first and second experiments of P. and L. did not seem to effect their records ; an interruption occurring between the fifth and sixth days with the subjects K. and M., and between the fourth and fifth days with J. and S., showed its effect on J. and M. very clearly, but in only a small degree on K., and seemingly not at all on S.\nThe losses on the fourth, fifth and sixth days with P. (left hand) may be attributed to the fact that he unthinkingly reversed the original direction of the movement in using the left hand. In the three previous exercises he had made his hand move clockwise. All the subjects except J. were right-handed and the right hand was always moved counterclockwise, the left moving clockwise. The hands were thus moved symmetrically but in opposite directions. The subject J., however, moved his left hand counter-clockwise and his right hand clockwise. This would seem to indicate that the centers governing the rotary movements of the muscles in right and left handed persons are diametrically opposed","page":53},{"file":"p0054.txt","language":"en","ocr_en":"54\nW. Smythe Johnson,\nFig. 2.\nFig. 3.\nX, upper line, serial number of experiment. X, lower line, days of previous practice.\nV, tap time in thousandths of a second.","page":54},{"file":"p0055.txt","language":"en","ocr_en":"Researches in practice and habit.\nSS\nFig. 5.\nX, upper line, serial number of experiment.\nX,\tlower line, days of previous practice.\nY,\ttap time in thousandths of a second.\nto each other. A few experiments were made on another left handed person ; the same direction of the movement of either hand was chosen as with J. However, I caused him to reverse the direction of the movement of the right hand so as to move counter-clockwise. As a result the movements of the right hand were much slower and fatigue set in earlier. Moreover, there was less regularity in the movements of the right hand.\nDaily probable error.\nThe change in the probable error for successive days, given in Table II., may be considered as an expression.of the development of automatic\ncontrol over the movements of the hand.\nTable II.\nProbable error on successive days.\nSubject.\tHand\t\t\t\tSerial number of experiment.\t\t\t\t\t\t\t\n\tused.\ti\t2\t3\t4\t5\t6\t7\t8\t9\t10\tII\nK.\tj R\t50\t38\t23\t36\t32\t43\t\u2022 37\t1.5\t15\t17\t20\n\ti L\t80\t51\t18\t38\tto\t29\t20\t16\t27\t40\t]8\nM.\tJ A>\t3\u00b0\t37\t26\t47\t29\t36\t20\t20\t20\t14\t13\n\t\u25a0u\t35\t38\t19\t20\t24\t30\t22\t26\t27\t17\t42\ns.\tL\t39\t31\t37\t27\t31\t28\t37\t39\t27\t27\t\nJ-\tR\t39\t36\t48\t26\t35\t42\t29\t33\t3\u00b0\t\t\nP.\t{\u00ce\t30 26\t18 23\t18 19\t20 18\t15 15\t14 15\t\t\t\t\t\nL.\tJ R\t35\t3\u00b0\t29\t21\t17\t20\t\t\t\t\t\n\t\\L\t31\t28\t24\t23\t19\t13\t\t\t\t\t\nH.\t1R\t36\t21\t27\t\t\t\t\t\t\t\t\n\tXL\t27\t27\t32\t\t\t\t\t\t\t\t\nThe unit of measurement is I\u00b0 = o.ooi3.\nThis table gives the probable error of the tapping times from which Table I. was computed.","page":55},{"file":"p0056.txt","language":"en","ocr_en":"56\nW. Smythe Johnson,\nThe probable error P was calculated according to the formula\np=\nK + *,* +\n+\nn\u2014 i\nwhere vv vv . . ., vn are the errors for the \u00ab\u25a0 measurements. Table II. shows a decrease from day to day which closely corresponds to the average daily decrease in the intervals between taps.\nAn average of the decrease in the probable error of all th\u00e9 subjects for\nFig. 7.\nX, upper line, serial number of experiment. X, lower line, days of previous practice.\nV, tap time in thousandths of a second.\neach hand was made; a graphical representation of it is given in Fig. 7. Though the error for the left hand was larger than for the right, the curve for either hand takes about the same direction. Hence the centers governing the left-hand movements, though less developed, are susceptible to the same law of gain in automatic control.\nThe irregularities in the daily decrease of the error may be accounted for in part by the variations in the nervous condition of the subject from day to day. Moreover, the preliminary trials given just before beginning each experiment, which were always the same in number, were not sufficient in every case to arouse the nervous centers so as to get the best results from practice, for I observed that in a few instances the tapping time was very slow at first, but rapidly increased in speed during the first part of the experiment, which caused a larger probable error. Thus","page":56},{"file":"p0057.txt","language":"en","ocr_en":"Researches in practice and habit.\n57\nthe 43\u00b0' probable error of K. on the sixth day was principally due to the long intervals between the first four taps.\nThat the preliminary trial was not sufficient in every instance for him to acquire momentum, so to speak, is seen in the record for the sixth day, which ran as follows: 315, 289, 291, 295, 200, 245, 244, 240, 269, 242, 248,\t251,\t243,\t237,\t246,\t238,\t254,\t240,\t229, 254,\t218,\t219,\t260,\n222> 23I>\t239>\t235>\t2I6,\t223,\t242,\t229,\t217,\t212er. This shows a\tdif-\nference of 103er between the first and last taps, though the largest decrease occurred in the first four taps. As contrasted with this record, we give his record for the eighth day, which shows considerable gain in rapidity and regularity in\ttwo\tdays\u2019\tpractice:\t231,\t229, 225,\t220,\t219,\t223,\n229, 226,\t223,\t221,\t218,\t222,\t208,\t210,\t221,\t226, 209,\t204,\t229,\t221,\n2\u00b03, 207, 203, 200, 198, 217, 213, 209, 205, 202, 200, 195, 208, 203er. This record shows a difference of only 29er between the first and last taps. A graphical representation of the records for these two days may be seen in Fig. 8. According to the statement of the subject on the\nDiA Y\nFig. 8.\nX,\tupper line, serial number of experiments. X, lower line, days of previous practice.\nY,\ttap time in thousandths of a second.\nsixth day, he was conscious at the start that he was not doing his best, but somehow he could not get control of the muscles within the time allowed him in the preliminary trial. Hence the temporary effect on the first part of the record.\nIn those records which show a small variability the subjects invariably reported that they were feeling quite well and brisk. Moreover, I observed that as the movements came to be controlled by the automatic centers and the movements became more rapid,, the greater was the influence of any change in the nervous condition of the subject. An illus-","page":57},{"file":"p0058.txt","language":"en","ocr_en":"58\nW. Smyihe Johnson,\ntration of this may be seen in the results for J., sixth day, where he reached the highest probable error of his whole set of experiments, having inadvertently received an electric shock just before beginning the experiment.\nThe rapid decrease in the intervals between taps with P. (right hand, Fig. 4) emphasizes the influence which previous training of any one set of muscles has upon the formation of automatic control where the form of movement is changed only as to direction. It is even more manifest in the decrease of the probable error. Although the movements involved in this exercise were different from those required in piano playing, yet the years of previous training had developed the centers controlling the movements of the hand to respond quickly and with considerable regularity ; so when the movements became circular (as was necessary in tapping the three keys) it was only necessary to establish the one additional element in the automatic centers, the direction of the movement of the hand.\nAverages of successive taps.\nMy observations during different experiments each day convinced me that the subject gained in rapidity of tapping constantly from the beginning of the experiment until the close or until the setting in of fatigue. I tested and corroborated my conviction in the following way : Reviewing the protocols, I derived an average from all the first taps, then from all the second taps, and so throughout the whole series. That my conviction was correct is clearly shown by the fact that in every case the tapping time was decreased, ranging from to ioo<r.\nConsider ng the large number of experiments made and the number of subjects included in these tests, the constant increase in rapidity of muscular action during each experiment exceeded pur expectations. Indeed, it may be stated as a law of practice wherein rapidity is the objective point, the movement does not follow any rhythmical law of increase and decrease, but constantly increases in speed until the setting in of fatigue.\nA few experiments were made wherein the tapping time was long continued, which showed that when fatigue first sets in, the subject loses for a short time, then there is a renewal of effort and the speed is constantly accelerated for a certain period, but not so long as in the first. As each successive period of fatigue came on, the successive period of increase in speed was shorter ; this was continued until that state was reached where the alternations were so rapid as to effect almost every other tap. This accords with the assertions of other investigators who have directed their investigations especially to the study of fatigue.1\n1 Moore, Studies of fatigue, Stud. Yale Psych. Lab., 1895 III 68; Binet et Vaschide, Exp\u00e9riences de vit sse, Ann\u00e9e psychol., 1897 IV 267.","page":58},{"file":"p0059.txt","language":"en","ocr_en":"Researches in practice and habit.\n59\nThe lack of the development of the centers governing the movements of the left hand not only caused the tapping time to be slower than for the right hand, but the amount of gain during each experiment was also less. For instance, the average amount of gain for the right hand of K. was 9017 in 20 taps, while with the left he gained only 400'. Stated in words, the gain with the left hand of all of the subjects was much less in proportion to the rapidity of its movement.\nAlthough P. was the most rapid, yet he gained less in amount than the other subjects during each experiment, for the right hand gained only So17 and the left possibly io<r. This demonstrates the fa> t that, as the muscles come more completely under control, the influence of practice becomes less during each experiment. With L., however, the results for the left hand are more favorable, namely, a gain of 40^ with the left hand, compared to a gain of so0, with the right. The results of H. are exceptional in that the gain was equal for both hands, namely, only 30^. With the subjects K., M. and P. the centers governing the left hand were slower in their activity and less influenced by training than with L. and H.\nIn like manner, the probable error for successive taps was determined. It showed a tendency to decrease as the experiment progressed. If the subject put forth the greatest effort in the beginning of the experiment, the error was correspondingly greater in the first part. But when the special effort was telaxed, the muscles reverted to their more accustomed speed of adjustment, and at the same time became more regular in their functioning. Therefore, a minimum gain in rapidity during any. practice period is the best condition for impressing that standard upon the nervous centers ; as a result the decrease of the probable error is correspondingly accelerated. This principle is well illustrated in the results obtained for the subjects J. and P. both of whom gained very little in rapidity of tapping during the progress of an experiment, but who, however, made considerable decreases in the probable error. As long as spasmodic accelerations are observable throughout a short series of practice, the automatic control over the muscles may be considered imperfect, the degree of imperfection being indicated by the amount of the probable error.\nRelative gain by practice.\nAlthough the gain from day to day varies in response to the mental and physical condition of the subject, yet we may suppose from the conformity of results of all the subjects'that the acquirement of muscular facility, or, physiologically, the transition from that state which demands","page":59},{"file":"p0060.txt","language":"en","ocr_en":"6o\nW. Smythe Johnson,\nconstant fixation of the attention to the state of automatic control follows as closely a mathematical law as do falling bodies. The difficulty in finding such an expression is partly due to the difficulty in getting the subject under the same physical and mental conditions at each experiment and partly to the lack of scientific results on allied subjects upon which the law of habit must also depend. For instance, some of the .personal factors of our problem causing variations in the results obtained from different individuals are : differences in muscular memory for different individuals ; different physical conditions due to differences in the constituents of the blood, etc. ; the rapidity of the heart-beat ; the temperature of the body ; the power of the fixation of the attention ; the interest in the experiment ; and the influence of emulation. The more nearly these various conditions approach the normal and the more accurate the measurements are, the nearer will the results for different individuals conform to the same law of gain in rapidity and regularity of muscular adjustments. This law of development may receive mathematical expression either by percentages or by algebraical formulae. Of these two methods, I have adopted the former. I have adopted a method 'somewhat similar to that used by Amberg1 for determining the percentage of gain by practice. It will be seen, however, that I do not adopt his method without modification.\nWe took the average tapping time at the first experiment as a measure of the subject\u2019s ability without any previous training. The difference between the averages of the first and second experiments was taken as representing the influence of practice during the first experiment. Likewise, the average tapping time of the third experiment was subtracted from the average of the first. This process was repeated throughout the series. Then taking the sum of the whole number of gains over the .average of the first experiment and dividing by the number of experiments, we obtained the average gain for the whole series over the results of the first experiment, which was expressed as a percentage of the average tapping time at the first experiment. But this did not complete the series, for at the succeeding experiment he began the practice anew just as at the first experiment. It was, therefore, necessary to repeat the same calculation, taking the average tapping time at the second experiment as the basis. In like manner, we continued the process for the whole number of experiments, thus obtaining the average percentage of gain, taking each succeeding experiment as the beginning of a series. Then dividing the sum of these percentages by the number of experiments\n1 Amberg, Ueber den Einfluss von Arbeitspausen auf die geistige Leistungsf\u00e4higkeit, Psychologische Arbeiten (Kraepelin), 1896 I 30.","page":60},{"file":"p0061.txt","language":"en","ocr_en":"Researches in practice and habit.\n61\nwe obtained the average percentage of gain for the whole series of experiments.\nThe formula for the computation of these percentages may be stated in the following way.\nLet the averages for the ist, 2d and 3d, \u2014, \u00bbth days be a, b, c, \u2014, l. Then take\n(i)\n(0\n(3)\n(1t \u2014 b) -f- (a \u2014 t') -f- \u00c7a \u2014 d) -f- T- (p \u2014 /) = A, (b-c) 4 (6-d) + \u2022\u2022\u2022+ (b-l) = B, \u00c7c \u2014 d ) -f -f- (<: \u2014 l) = C,\n0i - 0 = \u00e0,\nA -i- (pi \u2014 i ) a = M\u00b0/o, B -7- (n \u2014 2) b \u2014 N\u00b0/c, C-r- (n \u2014 3) c = 0%,\nk\t= j\nA+B+C+-+K\n(n \u2014i)a -f (n \u2014 2)b + (n \u2014 ^)c + \u25a0\u25a0\u25a0 + \u00a3\n= Ave. %.\nThe signification of such percentages is that they give us a true standard for the comparative influence of practice on different individuals. Although all practiced the same amount each day under similar conditions, yet we shall now see how differently the percentage of gain in speed of voluntary movements differed with each subject from day to day and how similar were the final results after the completion of the entire series.\nTable III.\nRelative gain in speed from day to day.\n\tHand\t\t\t\tSerial number of day.\t\t\t\t\t\t\t\nbject.\tused.\tI\t2\t3\t4\t5 \u25a0\t6\t7\t8\t9\t10\tAve.\nK.\t/R\t0.21\tO l6\t0.17\t0.18\t0.18\tO.II\t0.13\t0.02\t0.03\t0.03\t0.12\n\tu\t.14\t.12\t.21\t.12\t\u202213\t.09\t.04\t.04\t\u202203\t.02\t.09\nM.\t\t.12\t\u202219\t\u202217\t.16\t.12\t.18\t.06\t.07\t.07\t.10\t.12\n\t\\L\t.l6\t.18\t\u2022h\t.12\t\u25a005\t.12\t.12\t.16\t.14\t\u202225\t.14\nS.\tL\t.11\t.20\t\u202215\t.10\t.10\t.06\t.07\t.08\t.06\t\t.10\nJ-\tR\t.09\t.08\t\u25a005\t.04\t\u25a017\t.08\t.07\t.10\t\t\t.08\nP.\t;R\t.32\t.20\t.14\t.06\t.06\t\t\t\t\t\t.16\n\t\\L\t.17\t.IO\t\u2014.07\t.09\t.07\t\t\t\t\t\t.07\nL.\t\t.26\t.17\t\u25a013\t.07\t.07\t\t\t\t\t\t.14\n\tt L\t.20\t.l8\t.14\t\u25a005\t\u2022\u00b03\t\t\t\t\t\t.12\nH.\t{L\t.22 .28\t.06 .or\t\t\t\t\t\t\t\t\t\u202213 .14\nThe values in columns I, 2, io were calculated by formula (2), those in the column Ave. by (3). The figure for any particular day indicates the combined relative gain for all succeeding days over the record for that day.","page":61},{"file":"p0062.txt","language":"en","ocr_en":"6 2\nIV. Smythe Johnson,\nThe relative average daily increase in speed is given in Table III. For the right hand of K. the average percentage of gain on the second day was 21% over the speed made at the first experiment. But, as is shown in the table, the percentages decreased perceptibly until the close of the series of experiments, ending on the eleventh day with a gain of only 3 \u00b0Jo over the speed of the tenth day. The small percentage in the latter part of the series would seem to indicate that he had approximately reached his limit in rate of movement. So with all the other subjects, the percentage of gain in increase of speed constantly declined as the practice was continued from day to day. The large percentages in the first part of the series show that the greatest gains are to be made in the early part of practice.\nThe average percentage of gain given in the column of averages shows the comparative value of practice for each individual. Comparing K. and M., whose experiments extended over the same number of days, we see that practice had the same effect on the right hand of each of them, namely, 12%; but for the left, K.\u2019s average percentage of gain was less than for the right hand, while with M. it was even larger than for the right. So with S. and J., we see that practice was of more value for S. than J. by 2 %. Likewise P. gained 2 % more with his right hand than did L. ; but with the left hand L. gained 5 % more than did P.\nThe average daily decrease of error was also derived according to the formula given on page 61 ; the results are given in Table IV.\nTable IV.\nRelative average daily decrease of error.\n'jET\tHand\t\t\t\tSerial number of\t\t\tday.\t\t\t\t\n3 C/3\tused.\tI\t2\t3\t4\t5\t6\t7\t8\t9\t10\tAve.\nK.\t< R 0 45 \u20140.22\t\t\t0.28\t0.27\t0.23\t0.5 t\t0.55\t\u20140.16 \u2014\t-0.23 \u20140.18\t\t0.15\n\t( L\t.64\t.42\t\u2014 .68\t\u202229\t.11\t.10\t\u2014 -27\t\u2014 .9 \u2014\t\u20227\t.55\t. 10\nM.\t\\ L ) '\t\u2022!3 \u2022 25\t\u202233 \u202233\t\u25a005 \u2014 -37\t\u2022'54 \u2014 -33\t\u20223\u00b0 - -14\t\u202252 .11\t.16 \u2014 .27\t.21 \u2014 .11 \u2014\t\u202232 .9 \u2014\t.07 \u202259\t.26 \u2014 .11\nS.\tL\t.19 \u2014\t.02\t\u202217\t\u2014 -17\t\u2014 .02\t- -13\t.16\t\u25a031\t\t\t\u202205\nJ-\tR\t.11\t.04\t\u25a032\t\u2014 -3\u00b0\t.04\t\u202227\t\u2014 .09\t\u202299\t\t\t\u2022\u00b07\nP.\tS*\t\u202243\t.07\t\u2022c9\t.27\t.07\t\t\t\t\t\t\u25a019\n\t\\L\t\u202232\t.26\t\u202223\t\u202217\t.CO\t\t\t\t\t\t\u202219\nL.\t\\R\t\u202238\t\u202227\t\u202233\t.12\t\u2014 .18\t\t\t\t\t\t.18\n\t\\L\t\u202231\t.29\t\u202223\t\u202230\t\u202232\t\t\t\t\t\t.29\nH.\t\\R\t\u25a033 \u2014\t.29\t\t\t\t\t\t\t\t\t.02\n\t\\L -\t.09 \u2014\t.16\t\t\t\t\t\t\t\t\t.11\nThe explanation is the same as for Table III.\nComparing the decreases of the probable error for K., 15%, and M., 26%, we see that practice was more beneficial for the right hand of M. than","page":62},{"file":"p0063.txt","language":"en","ocr_en":"Researches in practice and habit.\n63\nfor the right hand of K. But for the left hand the reverse was true. Even if we cast out the last practice of M., on account of its poor effect, the \u2014 11 % changes to + 6 %. It should be noted that the percentages for the left hand of K. and M. are the reverse of those given in Table III. Likewise the same is true for S. and J., for while S. gained more in speed, J. gained more in regularity of movement. The small percentage of decrease of the probable error would seem to indicate that S. and J. kept the order to tap as quickly as possible in the foreground of the attention, for the percentage of gain in rapidity was larger for the whole series with each of them than was the reduction of the probable error. On the contrary with the subjects K. and M. who had approximately reached their utmost speed of voluntary movement, the percentage of the decrease of the probable error was larger than the percentage of gain in speed.\nWe may summarize the results given in Tables III. and IV. in the following way : P. made the greatest percentage of increase in speed with the right hand, and S. and L. the greatest with the left ; that of all the subjects J. possessed least ability for development of rapid movements with the right hand. When regularity, not rapidity is considered, the right hands of K. and M. and the left hands of K. and L. made the greatest gain in regularity of movement while S. made the least of all.\nRelative average daily gain in speed and decrease of error compared.\nThe average percentage of increase in speed and of decrease in the probable error from day to day for all the subjects were compared ; the iesults are expressed in the A and B curves of Fig. 9. These curves\nFig. 9.\nX, upper line, serial number of day.\nX,\tlower line, days of previous practice.\nY,\trelative increase in speed ( B ) and decrease in probable error ( A ).","page":63},{"file":"p0064.txt","language":"en","ocr_en":"6 4\nW. Smythe Johnson,\nshow very clearly that whereas the percentage of gain in rapidity of movement is larger in the first part of the series of the experiments, in the latter part the greatest influence is directed toward the reduction of the irregularity of the movements.\nConsidering the curves A and B as representatives of the influence of practice on seven subjects, indicating the gain made in six to eleven days, we are justified in making the following general statements concerning muscular action.\nFirst, the gain in rapidity and the gain in regularity of muscular action are greatest during the first periods of exercise ; one or both of these continues to diminish as practice continues at each experiment as well as at successive experiments.\nSecond, during the earlier periods of exercise practice has greater effect upon the rapidity of muscular action, but later its greater effect is in the reduction of the irregularity of voluntary movement.\nThird, the relations of rapidity and irregularity are largely affected by the relative complexity of the muscular movements, the number of muscles undergoing training, and the subject\u2019s power of concentration of attention.\nII. Drawing circles.\nThe object of this set of experiments in drawing circles was to show r (i) the gain in the reduction of imperfections in the drawings during each experiment and from day to day; (2) the influence that a copy placed before the subject had upon the size of his drawings.\nA package of ten sheets of paper, 12 by ioomm, was placed before the subject, sitting at a table. A sheet of paper with a circle of 6omm in diameter, described with a compass, was placed before him as a copy; hc.was asked to look carefully at it each time before commencing to draw his circle. The copysheet was the same size as those used by the subject. The ten slips of paper lay one upon the other at the start ; the experimenter stood by and removed them as the circles were drawn. Thus all visual comparison with previously drawn circles was prevented. It was impressed upon the subject that he should keep the true circle constantly in memory after he took his eyes off the copy to direct his hand. The circles were drawn with the free hand, no portion of the hand or arm being allowed to rest on the table during the process. No restrictions were imposed on the time consumed in the drawing of the circles. All distracting influences were removed. Before beginning, the subject was enjoined to do his best at each trial.\nThe subjects were J. F. and S. F., teachers in the primary grades of","page":64},{"file":"p0065.txt","language":"en","ocr_en":"Researches in practice and habit.\n65\nthe New Haven city schools, who had previously had considerable practice in drawing circles in the school-room ; M. C.., a girl of twelve ; B., C., H. and J., students in the university.\nA characteristic expression for the deviation of the drawn curve from the true circle may be found by comparing the longest diameter with the shortest one. The two diameters were measured in millimeters and theii difference was considered as the amount of error.\nIt was deemed advisable to limit the number of circles drawn at one sitting to ten, so that the element of fatigue might be almost if not completely eliminated. Preliminary experimentation showed that the influence of fatigue was not perceptibly present with the right hand until the ninth or tenth repetition, and with the left hand until the fifth, to seventh repetitions. However, those curves in the series which most nearly approximated true circles resulted in attempts ranging from the fourth to the seventh. Consequently, under ordinary circumstances, the drawing of five circles would give sufficient practice for the subject to reach an approximate maximum of accuracy of adjustment. Hence it was deemed best to stop the experiment at a point where the subject was gaining in accuracy rather than to continue it until the errror began to increase.\nGain on successive days.\nThe average errors for each day are shown in Table V.\nTable V.\n\t\tAverage\terror on\tsuccessive days.\t\t\t\n\tHand\t\tSerial number of experiment.\t\t\t\t\nSubj ect.\tused.\ti\t2\t3\t4\t5'\t6\nB.\t[R\t12\t9\t12\t11\t8\t9\n\t\\L\t10\t9\t7\t5\t5\t8\nM.C.\t\\\u00ef\t10 13\t9 10\t7 7\t7 6\t6 S\t.4 4\nC.\t(R\t6\t6\tS\t4\t4\t3\n\tU\t10\t9\t9\t9\t8\t7\nJ.F.\tJR\t6\t6\t5\t5\t3\t2\n\tU\t11\t10\t9\t8\t8\t4\nS.F.\tJR\t6\t6\t6\tS\t4\t4\n\t\\L\t7\t6\tS\t5\t4\t4\nH.\tJR\t9\t9\t8\t7\t7\t6\n\tu\t7\t8\t10\t8\t9\t10\nJ-\tJR\t11\t6\t5\t5\t4\t3\n\tu\t9\t13\t8\t12\t7\t6\nThe unit of measurement is Imm.\nNumber of experiments each day, 5 with either hand.\nThe probable error of a determination varies from \u00b1 imm to \u00b1 o. imm.\n5","page":65},{"file":"p0066.txt","language":"en","ocr_en":"66\nW. Smythe Johnson,\nAs is shown in these figures, practice with the left hand gave more irregular results than that with , the right. How\u00e7ver, B.\u2019s record was an exception in that the error was not so large as that for the right hand. Moreover, the error decreased more rapidly in the practice with his left hand. The small and irregular gain with his right hand may have been due to. a tendency to accomplish the task with great rapidity and to previous practice in writing, for his penmanship was sharp and pointed. Hence the left hand had the advantage over the right since the right hand not only had to form a habit for a certain movement but to reform one.\nWith H. and J. the lack of improvement with the left hand was due to a decrease of effort. I observed that they made great effort on the first day ; but after that they declared it useless for them to practice with the left hand because improvement was hopeless. These two records plainly demonstrate the influence of confidence in one\u2019s ability upon one\u2019s development. Purposeful attention and persistent effort on the part of the subject are the two most essential elements in practice for the establishment of any definite mode of muscular action.\nThough the amount of gain withC. was the same for each hand, yet the magnitude of the error at the beginning of the series with the left hand gave more scope for improvement, and had the subject been equally skillful with his left hand the curve would have descended much more abruptly. On the contrary, M.C. gained more with the left hand than she did with the right. The very regular decrease of the error with this subject from day to day shows the influence of persistent effort ; I observed that she was very careful throughout the whole set of experiments.\nThough there was a gain of 7mm for the left hand of J.F. as contrasted with a gain of 4mm for the right, yet most of the decrease in the error for the left hand occurred at the last experiment, which was undoubtedly due to the extra effort put forth. Though the subject thought all along that she was doing her best while practicing with the left hand, yet after the close of the series she said that she was conscious of having made considerably greater effort at the last experiment. None of the subjects save J.F. knew that the sixth day\u2019s trial would close the set of experiments.\nGain in successwe circles.\nThe average of the errors for all first circles, that for all second circles, etc., were taken. The results, Table VI., show that With the most of the subjects the error became less as the practice was continued.","page":66},{"file":"p0067.txt","language":"en","ocr_en":"Researches in practice and habit.\n67\nTable VI.\nSubject.\tHand used.\t\tAverage error for successive circles. Serial number of circle. I\t2\t3\t\t\t4\t5\nB.\tf R\tII\t10\t9\t9\t10\n\tl R\t8\t5\t9\t7\t9\nM.C.\t\\R\t10\t9\t8\t7\t7\n\tl L .\t7\t6\t6\t7\t8\nC.\t\\\t\\R\t6\t5\t5\t5\t4\n\tl L\t10\t9\t9\t8\t7\nJ.F.\t\\\tr r lR\t6 ' 9\t5 9\t5 8\t4 7\t3 7\nS.F.\t-j\tr R\t6\t6\t5\t5\t3\n\tL R\t7\t6\t6\t4\t4\nH.\t! R\t8\t7\t7\t6\t6\n\tl R\t9\t8\t7\t9\t10\nJ- j\tf R\t7\t6\t6\t5\t4\n\tL R\t6\t9\t10\t8\t10\nThe unit of measurement is i\u201c\u201c.\nEach figure is the average of 6 experiments.\nThe probable error of a determination varies from \u00b1 0.9\u201c\"\u201d to \u00b1 o. Imm.\nIt has been proposed by Dr. Scripture to call the curve of change for a single continuous experimental session the \u201ccurve of practice\u201d and the curve of change for successive sessions the \u201ccurve of habit.\u201d In the present case the gain on successive days would be represented by a curve of habit and the gain in successive circles by a curve of practice.\nIn the case of M. C. the practice and habit curves are alike in that they each begin with an error of iomm and end with 7mm. The maximum decrease of error was not reached until the fourth circle was drawn with the right hand, while it was reached in the second with the left. The control of the movements of the left arm was evidently not sufficiently developed to prevent fatigue during the time required to make five circles.\nThe similarity between the practice and habit curves seems to indicate that the development during each practice period follows closely the same law as does the daily progress, for in both alike the gain is most rapid in the first part of the exercise. However, the amount of gain for the different subjects varies considerably; this I attribute to the differences in the muscular memories of different individuals. With some, the error of the first circle on successive days was only slightly larger than the aver-, age for all the circles made on the preceding day, while with others it was in some instances equal to that at the beginning of the preceding experiment but rapidly decreased. The law of practice and the law of","page":67},{"file":"p0068.txt","language":"en","ocr_en":"68\tW. Smythe Johnson,\nhabit are presumably of the same general form with different constants for different circumstances.\nComparison of the drawn curve with the original circle.\nIn order to indicate the size of the curve drawn as a circle, four diameters were measured and their average taken. The diameters measured were the shortest, the longest and the two 450 from these. These measurements furnished an approximate estimate for determining how much change, if any, there had been in reproducing the copy circle of 6omm in diameter.\nTable VII.\nA. Size of successive circles.\nHand\tSerial number of circle.\nSubject.\tused.\tI\t2\t\t3\t4\t5\nB.\t\t62\t64\t\t66\t63\t62\n\t\\L\t59\t62\t\t60\t62\t60\nM. C.\tJR\t64\t6l\t\t58\t57\t54\n\t\\L\t61\t58\t\t59\t59\t62\nC.\tJ R\t60\t6l\t\t62\t64\t68\n\t\\L\t58\t62\t\t61\t65\t69\nH.\tJR\t61\t60\t\t58\t57\t55\n\t\\L\t62\t63\t\t64\t63\t62\nJ.\tJ R\t61\t62\t\t63\t65\t68\n\t\\L\t58\t59\t\t61\t60\t60\nS. F.\tfR\t61\t62\t\t64\t65\t67\n\t\u25a0u\t61\t62\t\t62\t63\t64\nJ. F.\tJR\t63\t58\t\t60\t60\t60\n\tu\t63\t63\t\t65\t60\t63\n\t\tB. Daily average size of circle.\t\t\t\t\t\n\tHand\t\tSerial number of experiment.\t\t\t\t\nSubject.\tused.\tI\t2\t3\t4\t5\t6\nB.\tJR\t65\t59\t62\t63\t67\t65\n\tU\t57\t61\t61\t62\t63\t60\nM. C.\tJR\t66\t62\t57\t55\t55\t54\n\t\\L\t64\t61\t61\t60\t57\t56\nC.\tJR\t61\t62\t63\t63\t65\t65\n\tU\t61\t62\t62\t64\t64\t65\nH.\tf R\t60\t58\t58\t57\t57\t56\n\tU\t61\t63\t63\t62\t63\t63\nJ-\tJR\t68\t65\t64\t63\t63\t62\n\t\\L\t59\t63\t57\t63\t61\t55\nS. F.\t{i\tO O\t65 64\t65 61\t63 59\t62 59\t60 58\nJ. F.\t( R\t62\t61\t60\t60\t60\t60\n\t\\R\t66\t64\t63\t62\t60\t62\nThe unit of measurement is imm.\nIn section A, each figure is the average of six experiments.\nIn section B, each figure is the average of five circles.\nThe probable error of a determination varies from \u00b12.lmm to \u00b10.3\u2122\u201c.","page":68},{"file":"p0069.txt","language":"en","ocr_en":"\u25a0Researches in practice and habit.\n69\nThere were two points to which the subject was required to direct his attention in making each figure : the roundness of the figure and its correspondence in size to the copy. We shall now answer the question whether the subject directed his attention more especially to the former or the latter point.\nThe results given in the above table show three types of practice : ( 1 ) that in which the subject decreased the size of the circle both during the progress of each experiment and from day to day, (2) that in which the size of the circle was increased during the experiment but decreased from day to day, (3) that in which there was but little variation either during the experiment or from day to day. The first two classes characterize those who regarded more carefully the smoothness of contour of their own drawn curves than they did the correspondence in size to that of the copy. The third class were those who directed their attention more especially to the size of the curve, and who observed the copy closely each time before beginning to draw their own curves.\nThe measurements for B. showed that during the first three experiments he began by making the circle smaller than the copy, but increased its size during the experiment until it was larger than the copy, whether the right or the left hand was used. During this period he made the circles by moving each hand counter-clockwise. During the practice of the last three experiments, for some unknown reason, he reversed the direction of the movement for each hand and as a result the size of the successive circles was constantly decreased. The subject was not conscious of having made any change in the direction of the movement of either hand, and was unable to say which way he had moved his hand in the last experiment until he had made the motion as if making a circle, so unconsciously was the movement performed. Questioning the subject at the close of the series brought out the fact that this reversal may have been due to practice on the inverted oval as seen in the capital W. Excepting the change in the size of successive circles with the change in the direction of the movement of the hand, all the changes occurring during any one practice may be easily accounted for. If, for instance, after making the first circle, he judged that one to have been smaller or larger than the copy, then in the following he proceeded to correct it.\nThe figures given in 'fable VII. show that he made greater effort to approach the copy when practicing with the left hand than he did with the right. Moreover, I observed that he regarded the copy more carefully each time when practicing with the left hand. The care directed toward accuracy of contour and size of circles drawn with the left hand caused the time consumed in making the circles with the left hand to be about one-fourth longer than with the right hand.","page":69},{"file":"p0070.txt","language":"en","ocr_en":"7\u00b0\tIV. Smy the Johnson,\nWith M. C., the right hand was directed counter-clockwise, as was the case with all the other subjects save B., already mentioned ; the left was moved in the opposite direction. Starting with a diameter of 7 imra for the first circle with the right hand, she reduced the diameter to 6imm during the first experiment. Likewise the decrease in the size of successive circles continued throughout the series of experiments, becoming less, however, at each successive experiment. The average of all the experiments, Table VII., shows a decrease of iomm in the size of successive circles ; and in the daily averages there was a total decrease of i2mm. These figures show that the subject decreased\u201c the size of the circle with the right hand both during the process of the experiment and from day to day. Likewise the average of successive circles for the left hand shows that the size of the circle was decreased in the second circle, but increased thereafter. The decrease of the size of successive circles for the left hand was not so large as for the right, which indicates that the subject directed her attention more especially to the size of the circle when practicing with the left hand, and to the smoothness of contour of her own curves when practicing with the right.\nThe subject C. seemed to give closer attention to the copy when practicing with the left hand but more especially at the beginning of eachexperi-ment. As was the case with most of the others, the principal consideration at the beginning of each experiment seemed to be to make the circle the same size as that of the copy ; but when the copy did not correspond to the natural adjustments of the muscles and when an attempt to follow the size of the copy caused an increase of the error, the subject\u2019s aesthetic feeling for smoothness of contour in his own curves soon gained control of the movements, and consequently the size of the copy was wholly disregarded.\nSince S. F. made such small gains in the decrease of the angularities of her curves, we are justified in concluding that she directed her attention more especially to making her curves the same in size as the copy. Though there was a constant tendency to increase the size of successive circles throughout the experiment, yet the table shows that she decreased the average size of the circles in six days\u2019 practice with the right hand from 7omm to 6omm, and with the left from 7omm to 58\u2122\". I observed that during the first experiment she paid more attention to the smoothness of her curves than to their size, but with practice, however, she became more regardful of their size. In order to make a test on the point whether the size of the copy had any influence On the error in the drawings, I made a few experiments on this subject in which no copy was used. As a result the circles were somewhat larger than those made at","page":70},{"file":"p0071.txt","language":"en","ocr_en":"Researches in practice and habit.\n71\nthe first experiment and the error was less. These results for S. F. furnish an example of the development of the inhibitory powers in overcoming the muscular adjustments for larger movements of the arm.\nThe. most important feature of practice with J. F. was the approach in the size of the circles to that of the copy. Questioning the subject at the close of the series brought out the fact that she had fixed her attention more especially on the size of the circle. Her statement is confirmed by the large number of alternations in successive circles, so numerous in fact, that there were scarcely two successive circles of the same size. There was, moreover, no tendency to increase or decrease the size of successive circles as with the other subjects. The correcting process was continued throughout the series but the corrections decreased in amount as practice continued day by day. For instance, there was a variation of 8 m in the size of the circles drawn with the right hand on the first day while there was only 3mm variation on the last day. Such a gradual reduction of the variations in the size of successive circles from day to day demonstrates the influence of practice in giving control over muscular adjustments. The exaggerated adjustments are not in my opinion so much due to the lack of judgment as to the inertia of the muscles ; when they do yield to the will, such momentum is acquired that the movement is as likely as not to be exaggerated. The great difference in the development of control over muscular adjustments between J. F. and the other subjects was due to the close attention that she gave to the copy before attempting her circle. She not only observed the copy carefully at the beginning of the exercise but also before attempting each circle.\nDaily decrease of error expressed in percentages.\nThe percentages given in Table VIII. were computed according to the method given on page 61 ; they show the comparative effects of practice on fhe different subjects. We notice that equal amounts of practice had influences on the different subjects ranging in value from \u2014 12 % to + 37 %\u2022 In the case of B. the average gain was 8 % for the right hand and 13 % for the left. This signifies that although losses occurred on some days yet upon the whole, the good effects of practice overbalanced the evil effects of practice.\nThe percentages for M.C. show a decrease from day to day with each hand. It is particularly noticeable in the case of the left hand where the gain of the last over the first day amounted to 50 %, which, however, decreased very rapidly until the sixth day when there was no gain over the preceding day. These percentages show that practice had 3 % greater effect in the development of control over the movements of the","page":71},{"file":"p0072.txt","language":"en","ocr_en":"72\nW. Smythe Johnson,\nleft hand than for the right. On the contrary, with C. the gain was 12 % larger for the right hand than for the left. Moreover, the percentages from day to day showed no diminution in the influence of practice.\nTable VIII.\nRelative average daily decrease of error due to practice.\n\tHand\t\tSerial number of day.\t\t\t\t\nSubject.\tused.\ti\t2\t3\t4\tS\tAve.\nB.\t/f\t0. r8\t\u2014O.It\t0.22\t0.23\t\u20140.12\t0.08\n\t\\ L\t.60\t\u202231\t.14\t\u2014 -3\u00b0\t\u2014 -32\t\u202213\nM. C.\t\\Rr\t\u202232\t\u202231\t.14\t.21\t\u202217\t\u202223\n\t\t\u2022SO\t.42\t.24\t\u25a0t7\t.00\t\u202227\nC.\tH\t.27 .16\t\u202233 .08\t\u202227 .11\t.12 \u202217\t\u202225 .12\t\u202225 \u202213\nJ- F.\tIf\t\u202230\t\u202237\t\u202233\t\u202250\t\u202233\t\u25a037\n\t\t.29\t\u202227\t.26\t\u202225\t\u20225\u00b0\t\u202232\nS F.\tIf\t\u202217\t.21\t.28\t.20\t.00\t\u202217\n\t\\L\t\u202231\t\u202225\t\u202213\t.20\t.00\t.18\nH.\tIf\t.18\t.22\t\u202217\t.07\t\u202214\t\u2022IS\n\t\\L\t\u2014 -23\t\u2014 -iS\t.10\t\u2014 .19\t\u2014 .11\t\u2014 .12\nJ-\tIf\t.58\t.29\t.20\t\u202230\t\u202225\t\u202232\n\t\\L\t\u2014 .02\t\u202236\t\u2014 .04\t.46\t.14\t.18\nThe percentages given in this table were computed from the results given in Table V. according to the formulas explained on page 61. The values in columns 1, 2, ..., 5, are calculated by formula (2), those in the last column by (3).\nThe very considerable effect of practice on J. F. showed itself both in the increase of the percentage from day to day and in her making a larger average percentage of gain with each hand than any of the other subjects. As already observed, S. F. varied from the copy more with the left hand than with the right, the reduction of the error being 1 (1Jn greater for the left hand than for the right. Considering the fact that she was decidedly right-handed, this is important as it shows how a change in the fixation of attention alters the influence of practice. On the other hand, with H. we find that practice had an opposite effect, for with the right hand there was a gain of 15 %, while with the left there was a loss of 12 \u00b0J0.\nError for successive circles expressed in percentages.\nIn the same manner as in the preceding section, the percentage of decrease in the error for successive circles was determined, which showed in every case a gain with the right hand during each practice period, ranging in values from 1 % to 27 %. With the left hand, the percentages varied from \u2014 20 % to 21 %, though in four out of seven cases,","page":72},{"file":"p0073.txt","language":"en","ocr_en":"73\nResearches in practice and habit.\nthe percentages showed a loss for the left hand. This demonstrates very clearly that with untrained muscles a constant standard can usually be sustained only for a very short interval of time. However, the ability for continuing the exercise with increased regularity for a longer time increases with practice.\nTable IX.\nRelative average decrease of error for successive circles. Hand\tSerial number of circle.\nSubject.\tused.\ti\t2\t3\t4\tAve.\nB.\t\t14\t7\t\u2014 6\t\u201411\ti\n\tU\t6\t\u201417\t11\t\u201429\t\u201420\nM. C.\tIf\t22\t17\t6\t0\t11\n\tl L\t4\t\u201417\t\u201415\t\u201414\t\u201410\nC.\tIf\t21\t7\t10\t20\t19\n\tl L\t17\t11\t17\t12\t14\n:j- f.\tlR\t29\t20\t3\u00b0\t25\t26\n\t\\L\t14\t18\t12\tO\t11\nS. F.\tJ f\t21\t27\t20\t40\t27\n\tXL\t28\t22\t33\tO\t21\nH.\tIf\t19\t9\t14\tO\t10\n\t\t6\t\u2014 8\t\u20143\u00b0\t\u2014 II\t\u201412\nJ-\tJ*\t25\t17\t25\t20\t22\n\tXL\t\u201454\t\u2014 4\t10\t\u201425\t\u201418\nThe percentages given in this table were computed from the results given in Table VI. \u2022 according to the formulas on page 61.\nObservations and deductions.\nI observed that when a special effort was made it was usually accompanied by unnecessary movements of the body. For instance, C. in the first part of the series of experiments would contract the jaw muscles ; J. would bite his lips ; some would twist the mouth ; others would knit the eyebrows. As practice continued, however, and the action became more habitual, these distortions for the most part disappeared. Undoubtedly whenever there is a tension of some muscles while others are being vigorously exercised, they become influenced in proportion to this tension.\nIn some instances I have observed that when the subject noticed an irregularity in his figure, a desire to improve upon this seemed to excite the nervous centers so much that the following effort would not be so good as the preceding. Although the gain in proficiency is not entirely a physiological process, yet any chain of actions must be repeated a number of times before it becomes established in the automatic centers. However, the amount of practice required for any chain of actions to be","page":73},{"file":"p0074.txt","language":"en","ocr_en":"74\nW. Smythe Johnson,\ncarried on with the minimum amount, of attention and effort is a psychological factor, depending upon the intellectual, vigor of the individual.\nThese experiments bring out some striking differences in the development of the movements of the right and left hands. For instance, the practice curve for the right hand followed closely the direction of the habit curve ; while with the left, instead of there being a gain during each practice, there was, in some instances, a decrease in accuracy of the drawings as the practice continued. This difference in the practice curves for each hand, I attribute to the special effort of attention called forth at the beginning of the experiment with practice of the left hand. When practice is carried on until the movements become irregular, the practice becomes injurious, for the irregular movements become incorporated into the chain of reactions as certainly as do those which are purposefully directed. Therefore, practice may tend to establish irregular adjustments as well as regular ones. Speaking figuratively, the capital on hand at the beginning of each succeeding practice period is the sum of the preceding practices. Consequently, the larger the probable error of the average of all the preceding practices, the more irregular will be the movements of the muscles at the succeeding practice. Hence better results might have been obtained in those cases where the error increased after the third circle if the practice periods had been shorter in the first part of the series and more prolonged in the latter part.\nOwing to the fact that some of the subjects increased the size of successive circles and others decreased them, we are justified in concluding that there is a certain adjustment of the muscles in writing and drawing most suitable for each individual which should be taken into consideration when training the muscles for accurate adjustment. If the size of the copy corresponds to the natural adjustment, the subject needs only to direct his attention to the smoothness of his figures, otherwise, he has to contend with the distracting element of the size. Therefore the amount of gain in accuracy of adjustment will be influenced thereby. The general conclusion is that, in the earlier stages of muscular development, the size of the copy should be adjusted to the natural movements of the muscles. If this is not done it may prove such a distracting element that the subject will discard it altogether, for, as we have seen, the attention is always directed first toward smoothness of contour, or freedom from angularities. Psychologically the order of development is in such movements as writing and drawing, ( i ) reduction of irregularities, ( 2 ) correctness in size; and in movements where agility is involved, (x) rapidity, (2) regularity.\nThe correspondence in the decrease or increase of the size of the","page":74},{"file":"p0075.txt","language":"en","ocr_en":"Researches in practice and habit.\n75\ncircle and the average daily error, as shown in these experiments, indicates (i) that some distinct relation exists between the error and the size of the circle ; (2) that the subject\u2019s attention, especially in the case of the right hand, was usually directed to the decrease of the error rather than to making the circles of a size corresponding to that of the copy ; (3) that the muscles of the right hand, trained to make certain movements, found it difficult in some instances to establish an entirely new set of reflexes.\nFinally, these results support the principle that a short\" exercise often repeated is the best method of practice for rapid development of accurate adjustments of the muscles. There is no doubt that many of the long exercises in writing and drawing and other subjects in the school-room often engender habitual inattention in the pupils. They are often compelled to write continuously for several minutes ; the wisdom of this is doubtful when we consider that in the case of well-developed persons five trials at one time in the experiments with circles gave the best general results attainable at one sitting. Hence long practice at writing, drawing, piano-playing, etc., seems to be time and energy wasted, for not only are inattentive habits cultivated, but every wrong adjustment of the muscles gains a place in the chain of subconscious memories and therefore delays the development of the control over the muscles for accurate adjustments. The practice at each sitting should last only so long as the movements are purposefully directed.\nIII. Development of control over untrained muscles.\nThe object of this set of experiments was to ascertain the influence of practice on entirely untrained muscles and less adaptable joints. \u25a0 The experiments, made on the left large toe of Mr. Davis, covered a period of ten days. Four phases of the toe\u2019s movement were recorded, the time of the downward motion, the downward rest, the upward motion, and finally the upward rest. As each movement of the digit required the exercise of several muscles, the object of measuring each of these four phases of the toe\u2019s movement was to show: (1) the differences in time for the phases of the movement ; ( 2 ) the influence of practice on the phases ; (3) the influence of practice on the shortening of the time of the entire movement of the toe.\nApparatus.\nThe apparatus was virtually the same as that previously described in these Studies.1 The single 100 v. d. fork was, however, replaced by the\n1 Scripture, Elementary coutse in psychological measurements, Stud. Yale Psych. Lab., 1896 IV 113.","page":75},{"file":"p0076.txt","language":"en","ocr_en":"76\nW. Smythe Johnson,\nPfeil and Depeez markers from which wires led into the quiet room1 and were there connected with a double-contact key. The markers wrote directly on the smoked surface of the drum ; the points were placed parallel to each other and in a line tangent to the surface of the drum. When the drum was revolved, the distance between the marks caused by movements of the armatures could be measured with great accuracy.\nThe key knob moved through a distance of smm. At the beginning of the downward movement of the toe the back contact was broken ; at the end of the downward movement the front contact was made ; at the beginning of the upward movement, the front contact was broken ; at the end of the upward movement the back contact was made. Thus the limits were marked for the four phases, downward movement, lower rest, upward movement, upward rest. The latent times were compensated.\nDaily averages of the tapping time.\nThe average tapping times for successive days are given in Table X. The series of taps during each experiment was divided up into three parts. The first section in Table X. includes the averages of the first 30 taps ; the second, the 31st to the 50th tap ; and the third, the 51st to the close of the experiment.\nThe results show an almost constant increase in speed from day to day. Beginning with an average of 43 6er on the first day, the average tapping time became 212\u00b0 after ten days\u2019 practice. Only on one day was there a loss and then the subject was indisposed.\nFatigue was always present after about the 50th tap, appearing sooner on some days than on others, as indicated by the probable errors in section C. of Table X.\nPractice generally has its greatest effect between the 30th and 51st taps. The first 30 taps formed the training period of the muscles in which the tapping was constantly accelerated. When the exercise was continued until the muscles were partially overcome by fatigue, the tapping became slower and much more irregular. The portion of the experiment lying between the 30th and 51st taps was chosen arbitrarily as a uniform measure for comparison. The most regular part of the day\u2019s practice corresponded closely to that marked off in these boundaries. The tendency was, however, for this zone of regularity to move each day slightly farther away from the beginning.\nOf the different phases of the toe\u2019s movement, we notice that the movement upward was longer than the downward one ; likewise the upward rest was longer than the downward rest. The tension of the spring\n1 See above, p. 52.","page":76},{"file":"p0077.txt","language":"en","ocr_en":"Researches in practice and habit.\n77\nof the key, although made too slight to be considered, may have also contributed to make the downward rest shorter. That the touch stimulus was the main factor appears likely because by practice the upward rest tended to decrease more rapidly than the downward rest.\nTable X.\nDaily average of tapping times.\nDate,\t\tMovement downward.\t\t\t\t\t\tDownward\t\trest.\t\tMovement upward.\t\t\t\t\t\n1898.\t\tA I\t\u2022 e.\tB\tp.e. C\tp.e.\tA\tp.e.\tB p.e.\tC\tp.e\tA\tp.e.\tB\tp.e.\tc\tp,e.\nXII\t-14\t45\t14\t39\t6 47\t12\t56\t13\t97 40\t62\t20\t88\t38\t84\t40\t59\t18\nit\t15\t31\t3\t23\t7 45\t14\t63\t16\t57 14\t116\t41\t40\t12\t27\t9\t61\t18\nit\t16\t31\t5\t4t\t11 42\t12\t87\t41\t69 28\t77\t33\t40\t9\t38\t9\t41\t8\n* \u2018\t17\t30\t5\t32\t5 41\t12\t56\t10\t56 15\t71\t17\t31\t9\t31\t5\t44\t12\ni (\t18\t29\t6\t35\t10 32\t7\t86\t33\t80 25\t107\t40\t30\t6\t33\t4\t43\t19\ni i\t19\t34\t9\t32\t5 34\t8\t57\t13\t66 21\t63\t19\t37\t10\t36\t4\t46\t14\nit\t20\t46\t13\t58\t21 45\t11\t66\t19\t70 13\t77\t16\t41\t13\t38\t14\t48\t10\ni <\t21\t29\t3\t3\u00b0\t3 37\t9\t73\t25\t69\t9\t77\t25\t27\t3\t28\t3\t35\t6\nii\t22\t36\t10\t33\t8 37\t9\t74\tJ9\t75 10\t83\t17\t36\t9\t26\t4\t3\u00b0\t8\n11\t23\t29\t3\t31\t6 26\t4\t64\t26\t74 18\t75\t13\t24\t3\t25\t4\t25\t4\n\t\t\t\t\t\t\t\t\t\t\t\tEntire\tmovement.\t\t\t\t\n\t\t\t\t\tUpward rest\t\t\t\t\t\tTap-\t\tp.e.\t\tp.e.\t\t\n\t\tA\t\tp.e.\tB\tp.e\t\tC\tp.e.\t\t\ttime\tinv\t\tas%\t\t\n\t\t267\t\t94\t272\t98\t\t231\t70\t\t\t436\t103\t\t24\t\t\n\t\t153\t\t56\t169\t64\t\t175\t67\t\t\t322\t73\t\t23\t\t\n\t\tio\u00bb\t\t60\t98\t40\t\t147\t69\t\t\t296\t74\t\t25\t\t\n\t\t171\t\t99\t150\t62\t\t135\t52\t\t\t284\t68\t\t24\t\t\n\t\t112\t\t44\t102\t45\t\t134\t54\t\t\t281\t61\t\t22\t\t\n\t\t109\t\t49\t138\t58\t\t153\t65\t\t\t273\t62\t\t23\t\t\n\t\t157\t\t62\t97\t31\t\tIC9\t45\t\t\t275\t68\t\t25\t\t\n\t\t105\t\t43\t93\t37\t\t120\t53\t\t\t241\t54\t\t22\t\t\n\t\t81\t\t22\t76\t25\t\t83\t21\t\t\t221\t36\t\tl6\t\t\n\t\t99\t\t29\t73\t22\t\t71\tl8\t\t\t212\t35\t\t17\t\t\nUnit of measurement, i<r = 0.001s.\nThe probable error of a determination varies from \u00b1 23V to \u00b1 0.50'.\nA,\tthe daily average of the first 30 taps.\nB,\t\u201c\t\u201c\t\u201c\t\u201c \u201c following 20 taps.\nC,\t\u201c\t\u201c\t\u201c\t\u201c \u201c remainder.\np.e., probable error.\nThe most noticeable effect of practice consisted in the change of the probable error of the upward rest, which decreased after ten days\u2019 practice from 94ff to 29^ for the first 30 taps; for next 20 taps from 98^ to 22ff ; and after first 50 taps, from 70er to 18*. Likewise, the movement upward shows a greater gain in regularity than the movement downward ; it decreased in section A., from 38er to 3*; in section B., from-400' to 40'; and in section C., from I80, to 40'. The conclusion to be drawn","page":77},{"file":"p0078.txt","language":"en","ocr_en":"7\u00bb\nW. Smythe Johnson,\nfrom the disparity in the amount of decrease of the probable error for the different phases of the toe\u2019s movement is that, since in the movement upward and in the upward rest there was a greater amount of voluntary\n\u2022 I > t\n770\t77s TU 7Z7 in 177 no ns TJo 777\nFig. io.\nX,\tserial number of tap.\nY,\ttap time in thousandths of a second.","page":78},{"file":"p0079.txt","language":"en","ocr_en":"Researches in practice and habit.\n79\neffort, they were more.irregular in the beginning; and as the movement came to be controlled by the automatic centers, the influence of practice was more manifest in these two phases of the toe\u2019s movement. Consequently we conclude that in beating time that phase of the movement which causes the greatest irregularity consists in the change of the motion upward to the motion downward.\nThe daily average tapping time, Table X., shows a decrease of the probable error from 103\u00b0\u2019 to 35*. Moreover, when the error is expressed-as a percentage of the tapping time, it shows that the increase in regularity was larger than the gain in speed, for the relative gain decreased from 24% to 17%.\nThe average daily increase in speed, according to the formula given on page 61, was determined, with the following results: 169, 62, 40, 33, 37) 39> 5\u00b0> 25> 9<r- This expressed as percentages gives: 39, 19, 14, I2> I3) 14) 18, 10, 5%. These figures show that the first day\u2019s practice resulted in the largest gain, which, however, rapidly decreased as the practice was continued from day to day. They moreover show that in simple movements, as in tapping, the effect of practice is greater in proportion to the undeveloped state of the muscles.\nPhysiological effects of practice.\nThe differences in the fluctuations in the curves of Fig. 10 show the influence of practice. The A curve is the practice curve of the last experiment ; B, that of the second experiment.\nThese curves show how the periods of slight paralysis varied during the two experiments. In curve B the variations follow each other in rapid succession. In curve A they are not so frequent and the recovery is much more rapid. The fluctuations, though larger throughout the series of curve B, increase in amplitude and frequency toward the close of the experiment. In curve A the fluctuations are not so pronounced in the second half even as in the first half. Hence the most prominent physiological effect of practice is to delay the arrival of entire paralysis of the muscles and the reduction of fluctuations in the practice curve. This depends, however, to a large extent on the nervous condition of the subject, for on some of the intervening days absolute paralysis of the muscles occurred before the 125th tap.\nAt the beginning of the series of experiments intense pain was felt in the calf of the leg after a short period of practice. As exercise continued, the pain became sympathetically induced also in the calf of the other leg.\nThe coming on of fatigue was. characterized by the following stages : CO a feeling of strain throughout the tendon above the heel ; (2) slight","page":79},{"file":"p0080.txt","language":"en","ocr_en":"8o\nW. Smythe Johnson,\npain in the calf of the left, or exercised, leg; (3) pain was felt in the muscles controlling the upward flexion of the toe ; (4) paralysis gradually spread over the whole of that leg while the pain continued to increase in the above mentioned muscles ; ( 5 ) the pain was finally felt in the calf of the right leg. On the fourth day the pain did not recur in the calf of the leg but was felt in the tendon immediately over the second joint of the toe ; but before the close of the experiments it had ceased altogether.\nLikewise, in a series of dumbbell exercises wherein I acted as subject, on first to fourth days inclusive the pain was so intense that I was able to exercise only a short time, but after the fourth day I was able to raise one dumbbell over 1000 times without experiencing any actual pain, the only noticeable effect being considerable fatigue.\nThroughout all the experiments, I noticed that on some days there was a tendency to decrease the speed from the very start, while on others there was at first a period of acceleration for a limited time before the decrease began. I consequently performed some experiments upon myself in order to determine whether this was connected with the rate of the heart-beat. The experiments were made in tapping with the toe and in the exercises with the dumbbells. Although the tests were not entirely conclusive yet in a general way I found that there was at least some relation between them. When the movements were as rapid as possible there was a decrease in the heart\u2019s action. I also observed that when the nervous system was in a relaxed state there was a rapid increase in the pulse beat in the early part of the exercise, the pulse becoming full and firm. Accompanying this there was an increase in the rate of voluntary activity.\nThe fluctuations in the practice curves may be due to the following causes : ( 1 ) fluctuations of the attention caused simply by the lack of effort on the part of the subject ; (2) local paralysis of the centers governing the muscles brought on by exercise; (3) mental paralytic strokes causing fluctuations, which are generally ascribed to mere fluctuations of attention; (4) general physiological fatigue of the whole body; (5) general mental fatigue.\nThe above conditions are ever changing, for the physiological effects of1 practice and exercise depend upon the supply of energizing material stored up in the muscular tissue, the supply of oxygen to the blood, temperature, etc. The mental condition is, moreover, more or less dependent on the physiological condition.\nGeneral mental fatigue was distinguished from general muscular fatigue by the feeling that resulted after a period of violent exercise. The muscles of the subject sometimes became fatigued almost to the de-","page":80},{"file":"p0081.txt","language":"en","ocr_en":"Researches in practice and habit.\t81\ngree of paralysis, but no special mental effects were felt. On one day after the subject had been engaged with some difficult computations, the change of exercise seemed at first to be a relief, but in a few seconds it became quite difficult for him to control his attention. Though the rate of tapping on that day was as rapid as on any preceding day, yet the mental fatigue was much greater.\nA distinction can be made between mental and muscular fatigue by the fact that the coming on of mental fatigue is spasmodic and irregular. An aberration of the attention marks the initial stage of mental fatigue. This soon passes over into a stage which materially effects the regularity of the muscular action.1 No such irregularities characterize muscular fatigue which is governed by physical laws controlling the breaking down of muscular tissue and the dissemination of waste products.2 Mental fatigue, however, is subject to the effects resulting from fixation of the attention and thus partakes of all its fluctuations and irregularities.\nEffects of practice on muscular action.\n1.\tPractice causes increased circulation, furnishing thereby a large supply of food materials to the muscles, and more rapid dissemination of waste products. Thus the muscles are able to store up energy and give it out more readily on demand. But the storing of energy is not all. For example, gymnastic exercise has a higher purpose in view than to bring the body merely to that state of perfection found in the case of the common laborer. The development of muscular tissue must be supplemented by an education of the nervous centers in order that they may respond precisely to acts of will. Indeed, the increase of muscular tissue may be looked upon as merely an accidental accompaniment to that mental process which begins by constant fixation of the attention and ends, even where the achievement is most complicated, in automatic and subconscious control of the muscular movements.\n2.\tThe time before the coming on of mental and muscular fatigue depends upon the amount of muscular energy and upon the concentration of the attention, both of which are greatly influenced by practice.\n3.\tWhen for any set of actions the development of the centers has reached an automatic condition, the maximum rate of either mental or muscular rapidity is not reached in the first few seconds of the exercise because a chain of actions cannot be remembered by any act of will but requires the exercise of the muscles themselves to re\u00ebstablish the chain of subconscious reflexes.\n1\tMoore, Studies of fatigue, Stud. Yale Psych. Lab., 1895 III 89.\n2\tLombard, in the American Text-book of Physiology, nr, Philadelphia 1897.\n6","page":81},{"file":"p0082.txt","language":"en","ocr_en":"82\nW. Smythe Johnson,\n4.\tThe duration of maximum rapidity is dependent upon the power of the mind for sustained attention. As soon as the attention is diverted the movement comes into the control of the automatic centers. The speed is consequently decreased, for only by a special effort is speed, either muscular or mental, increased above the limit acquired by habit. The utmost speed can be maintained only for a few seconds at a time at first, but the period may be lengthened by practice. Hence the fluctuations in the practice curve are generally due to slight mental fatigue. Although the subject recovers very rapidly at first, yet as the exercise continues the fluctuations recur more frequently and the periods of recovery are lengthened.\n5.\tAnother very important element connected with the duration of one\u2019s ability for continuing the exercise, is his knowledge of the time that the exercise is to last. If the exercise is to last only a short time, greater effort will be put forth in that period than when the exercise is to continue an hour or more. In long periods of exercise the subject will unconsciously measure out the energy in proportion to the duration of the practice period. For instance, in the experiments made by Oehrn1 in memorizing syllables, in making successive additions, and in counting letters in groups of threes wherein the exercise was continued from one to two hours, the maximum was-reached only in the first instance after 24 minutes, in the second after 28, and in the third after 59. Contrasted with our experiments, his results show that the maximum point depends upon the effort put forth in the beginning of the experiment. The results that I obtained point to the fact that if Oehrn had shortened the practice period, the maximum point would not only have been reached in about one tenth of the time, but the progress no doubt would have been greater.\nNor does the maximum rate of voluntary movement depend upon an .innate sense of rhythm as Schaefer2 implies but is, as Camerer3 states, one of constant acceleration until the setting in of mental fatigue. Instead of a rhythmical fluctuation in the voluntary effort, maintained by some investigators, I found no remarkable regularity and am led'to consider it as dependent upon several psycho-physiological processes too complicated to have a regular period of oscillation.\n6.\tPractice does not always mean an absolute gain in efficiency ; it may\n1\tOehrn, Experimentelle Studien zur Individualpsychologie, Psychologische Arbeiten (Kraepelin), 1896 I 92.\n2\tSchaefer, Canney and Tunstall, On the rhythm of muscular response to volitional impulses in man, Jour. Physiol., 1886 VII 96.\n3\tCamerer, Versuche \u00fcber den zeitlichen Verlauf d. Willensbewegung, 41-45, T\u00fcbingen 1866.","page":82},{"file":"p0083.txt","language":"en","ocr_en":"Researches in practice and habit.\n83\neven produce negative results. It is generally taken to signify improvement ; but observation and the results of experiments discussed in preceding pages, show that, when either muscular rapidity or regulated movement is required, practice may be even detrimental to development. Every action leaves its trace on the nervous matter ; every effort put forth tends to establish itself so that acts immediately succeeding it follow as a matter of inertia. If the effort put forth is small and the action slow, habit then establishes itself for that mode of action. The law of the growth into habitual automatic control takes into account every activity. If a higher speed of activity or a regulated movement or \u00e0 certain readiness in mental grasp is desired, then every thoughtless action tends to establish itself and delays attainment to the desired standard of efficiency.\nIt requires the same effort to overcome the condition occasioned by bad effects of practice as it does to establish a new mode of functioning. Hence energy is wasted when the practice is not thoughtfully directed. Therefore, we conclude that not only is that part of practice efficient for growth in regulated movement, in speed of muscular adjustment, or in quickness of mental grasp, which is accompanied by conscious effort, but the unconscious adjustments also have their effects and should be directed properly.\n7. The feeling of satisfaction or of having attained one\u2019s limit is an-other'not less important element in the development of rapidity in either mental or muscular activity. Every advancement either in mental quickness1 2 or muscular activity requires a certain effort, depending on the stage of development already attained. The greater the speed and the smaller the probable error, the less the gain becomes for the same expenditure of energy. As this developed state is approached, a person feels that his efforts are not sufficiently rewarded, and finally there comes a period when he feels that he has reached the limit of his development. In fact, this constitutes the difference between the novice and the expert. The \u201c plateaus \u201d mentioned by Bryan 2 in the habit curve would seem rather to indicate resting periods in the effort. If the subject can be induced to sustain the same effort day by day, there would not be any \u201cplateaus \u201d in the habit curve.\nIf this law be expressed by the general equation y = fix'), where y indicates the amount of gain by practice, we must regard x as containing constant elements of (1) time, (2) complexity of\n1 Amberg, Ueber den Einfluss von Arbeitspausen auf die geistige Leistungsf\u00e4higkeit, Psychologische Arbeiten (Kraepelin), 1896 I 30.\n2Bryan, Studies in the physiology and psychology of telegraphic language, Psych. Review, 1897 IV 27.","page":83},{"file":"p0084.txt","language":"en","ocr_en":"84\tW7. Smythe Johnson,\nthe movements, (3) the number of muscles undergoing training, and (4) the growth of automatic control. The last may be resolved into various personal factors such as mental grasp, endurance for sustained effort, and the vividness of the impression. According to this principle, then, the growth of intellectual habits should be more rapid with those who possess the strongest intellectual powers, since, by their power of holding the attention, they succeed in getting the same impression repeated oftener in the chain of reactions. If with intellectual power there is combined strong individuality, or perhaps more properly speaking a strong will, such persons possess- superior ability after breaking down one habit and reforming another.\nIV. Estimation of time.\nThe apparatus consisted of a kymograph, to which was attached the Wundt time-sense apparatus with the Meumann star-contacts by means of which adjustments could be made so as to give any interval of time desired. The arrangement of the apparatus was the same as that described in these Studies.1 The sound of the 100 v. d. fork was conveyed to the subject in the quiet room by means of a telephone receiver. The sound to be estimated came first. After an instantaneous interruption the sound began again. When the subject thought that it had lasted as long after the interruption as before it, he pressed a key which recorded a spark on the drum of the kymograph. The contacts were so\nFig. ii.\narranged that an experiment occupied the first part of every period of 18a, the remainder of the period serving for rest. This was intended to avoid as far as possible any influence of fatigue. A series of such records is given\n'Scripture, Elemtntary course in psychological measurements, Stud. Yale Psych. Lab., 1896 IV 127.","page":84},{"file":"p0085.txt","language":"en","ocr_en":"Researches in practice and habit.\n85\nin Fig. 11. A represents the beginning of the tone; B, the point where it was interrupted ; and C, the point where the interval of time from B to C was equal to that from A to B. The dots represent the first ten estimates of 1002 by the subjects A.F., R.E. and H.J. on the first day ; the group at the top of the figure being the estimates for A.F., the second those for R.E. and that at the bottom those for H.J. This may be said to represent in a general way the ability of different individuals to judge intervals of time without practice.\nA number of preliminary experiments demonstrated the fact already established by a number of investigators that the time-judgment varied considerably for the same interval with different individuals and with the same individual for different intervals. Attempts have been made to establish a definite interval as that which can be estimated with the smallest amount of error, but the point has varied with each investigator.\nThe three intervals, 82s, roo2 and 1642, were selected because they covered the field of most accurate estimates according to the results obtained by previous investigators.\nThose who served as subjects for the experiments were W.J. (Jump), R.E. (Evans), E.F. (Furguson) and B.B. (Brown), students in the Theological Seminary ; A.F. (Fisher) and C.S. (Smith), steward and mechanic of the Psychological Laboratory.\nDaily average estimates.\nThe average estimates for successive days are given in Table XL, and are graphically represented in Figs. 12, 13 and 14.\nEvery precaution was taken to prevent the subjects from counting or moving any part of the body by which they might measure off the time through muscular energy, the object being to ascertain whether a person has an actual time-sense regardless of any form of muscular activity or mental calculation.\nAn inspection of the results for A. F. shows that there was a. constant decrease in the time estimate from day to day. Beginning with 88s on the first day, he gradually reduced the estimate during 16 days\u2019 practice to 552, which is but little more than one-half of the fime to be estimated. The same facts were brought out in the case of R. E., whose average estimate of 1002 was on the first day 1592, and 1022 on the eighth day. This shows a decrease of about one-half in the average estimate. H. J. increased his average for the first day, 99s, until the third day when he reached his maximum point, 1162, after which the estimate decreased until next to the last day. The very large estimate, 1372, on the last day, no doubt, was due to nervousness. The record for this day should not be","page":85},{"file":"p0086.txt","language":"en","ocr_en":"86\nW. Smythe Johnson,\nTable XI.\nAverage estimates on successive days.\n\t\tA. F.\t\t\t\tR. E.\t\t\tH.\tJ\t\t\t\nDate\tEsti-\t\t\tDate Esti-\t\t\t\tEsti-\t\t\t\t\t\n1898.\tmate on\tp. e.\tp. e. as\t1898. mate on\t\tp. e.\tp. e. as\tmate on\tP-\te\tp. e, as\t\t\n\t1005\t\t%\t\tIoo5\t\t%\tioo2\t\t\t%\t\t\nXI-26\t88\t1.4\ti-7\tXII-\t\u25a015 159\t1-5\t\u00b0-9\t99\tI.\t4\t1.4\t\t\n27\t91\ti-7\ti-9\t\t16 130\t1.6\t1.2\t103\ti.\t6\ti-5\t\t\n28\t73\t1.1\ti-5\t\t17 138\t1.8\ti-3\t116\tI.\ti\t1.0\t\t\n30\t88\t1.4\ti-7\t\t18 118\ti-3\t1.1\t114\tI.\ti\t1.0\t\t\nXII-i\t74\t1.6\t2.2\t\t19 114\t1.1\t1.0\t114\ti.\t5\ti-3\t\t\n2\t73\ti-5\t2.0\t\t20 II7\t1.1\t1.0\t109\ti.\t2\t1.1\t\t\n3\t58\ti-3\t2.2\t\t21 107\t0.9\t0.8\t103\ti.\t0\t1.0\t\t\n4\t54\ti-7\t3-i\t\t22 102\t0.8\t0 8\t137\t2.\t3\ti-7\t\t\n5\t53\t1.2\t2-3\t\t\t\t\t\t\t\t\t\t\n6\t58\t0.8\t1.4\t\t\t\t\t\t\t\t\t\t\n7\t61\t1.0\t1.6\t\t\t\t\t\t\t\t\t\t\n8\t53\ti-3\t2.5\t\t\t\t\t\t\t\t\t\t\n9\t55\t0.8\ti-5\t\t\t\t\t\t\t\t\t\t\n10\t58\t1.1\ti-9\t\t\t\t\t\t\t\t\t\t\n11\t55\t1.0\t1.8\t\t\t\t\t\t\t\t\t-\t\n12\t55\t1.2\t2.2\t\t\t\t\t\t\t\t\t\t\n\t\tB. B.\t\t\tE. F.\t\tA. F.\t\t\t\t<\ts.\t\nDate\tEsti-\t\t\tEsti-\t\t\tEsti-\t\t\t\tEsti-\t\t\n1898.\tmate\t\tp. e.\tmate\t\tp. e.\tmate\t\tp. e.\t\tmate\t\tp. e.\n\ton\tp. e.\tas\ton\tp. e.\tas\ton\tp. e.\tas\t\ton\tp. e.\tas\n\t825\t\t%\t825\t\t%\t1645\t\t%\t\t1645\t\t%\nI-26\t\u202297\t1.1\t1.1\t73\t1-5\t2.0\t91\t6-5\t7.1\t\t147\t3-3\t2.2\n27\t99\t1.1\t1.1\t70\t1.0\t1.4\t99\t2-3\t2.3\t\t157\t2.0\t1-3\n28\t102\t1.8\t1.8\t69\t0.9\t1.4\tI\u00b05\t2-5\t2-4\t\t164\t2.6\t1.6\n29\t113\tr-7\ti-5\t82\ti-5\t1.8\t119\t2.2\t1.8\t\t162\t2.1\ti-3\n. 31\t112\t1.6\t1.4\t80\t1.1\t1.4\t118\t2-5\t2.1\t\t167\t2.2\ti-3\nII- I\tH3\ti-5\ti-3\t82\t0.9\t1.1\t157\t2.6\t1.6\t\t166\t3-9\t2-3\n2\t115\t0.8\t0.7\t80\ti-5\ti-9\t154\t2.4\t1.6\t\t180\t2-5\ti-3\n3\t107\t\u00cf-5\t1.4\t92\t1.8\t2.0\t155\t3-4\t2.2\t\t207\t2.8\ti-4\n4\t101\ti-3 '\t\u2018 i-3\t99\t1.2\t1.2\t169\t2.8\t1.6\t\t175\t2.7\ti-5\n5\t116\t1-3\t1.1\t99\t1.4\t1.4\t150\t3-1\t2.T\t\t205\t2.7\ti-3\n6\t118\ti-5\ti-3\t87\t1.8\t2. i\t159\t3-5\t2.2\t\t190\t2.4\ti-3\nUnit of measurement, i5 \u2014 0.01s.\nNumber of estimates on each day, 30 to 50.\nThe probable error of a determination varies from \u00b1 ioff to \u00b1 2\" . p. e., probable error of each estimate from the average.","page":86},{"file":"p0087.txt","language":"en","ocr_en":"Researches in practice and habit.\n87\nconsidered-in making up our deductions to be drawn from the results, for the subject was in a very nervous state when the experiment was made. I included the record in the table only to contrast the difference between the nervous and tranquil frame of mind in respect to the estimate of time.\nX, upper line, serial number of day.\nX,\tlower line, days of previous practice.\nY,\testimate of an interval of iooS.\nIn the cases of B. B. and E. F., Table XL, where 82*was the interval given, we notice that with each of these subjects, the time estimate was continually lengthened as practice continued. B. B. began with an average estimate of 97s on the first day, which was larger than the interval to be estimated and continued to increase the interval throughout the series of experiments. E. F. followed the same course, but he began with an average estimate of 73s on the first day, which increased daily, until his estimate exceeded the correct amount by 15s. Hence there was with","page":87},{"file":"p0088.txt","language":"en","ocr_en":"W. Smythe Johnson,\nX, upper line, serial number of day.\nX,\tlower line, days of previous practice.\nY,\testimate of an interval of 822.\nFig. 14.\nX, upper line, serial number of day.\nX,\tlower line, days of previous practice.\nY,\testimate of an interval of 1642.","page":88},{"file":"p0089.txt","language":"en","ocr_en":"Researches in practice and habit.\n89\nE. F. a total increase of 24s in his estimate, and with B. B. 212. These amounts divided by the number of days ( 11 ), would give an increase of over 2s per day. Likewise A. F. decreased his average daily estimate by 33s in 16 days, a daily decrease of 22 per day. R. E. decreased his estimate 572 in eight days, a decrease of 72 per day. The correcting process present in the case of H. J., no doubt, caused the irregularities in his habit curve, for the subject was aware of the tendency to underestimate long intervals and to overestimate short intervals. Consequently he continually corrected himself. But even in his case, we notice that after the third day\u2019s practice there was a tendency to shorten the interval.\nWhen 1642 was estimated, the same facts were found to exist. However, it should be noted that where A. F. decreased his estimate when 1002 was the given interval, he increasedit from 912 to 1592 (the highest point being 169s, 9th day) in estimating the 1642 interval; C. S., likewise, increased his estimate of 1642, approximately 42 per day.\nWhen the 1002 interval was estimated, all may be said to have decreased their estimates of its duration by practice,varying in amount with each individual. Moreover, each had a different conception of its duration in the very beginning of the experiments, ranging from 882toi592. When 82s and 1642 were the intervals estimated all the subjects increased their estimates of it by practice. The estimates of 82s on the firs: day were 972and 73s. Of 164s, the estimates were 9i2and 1472.\nIn Table XI. are given the probable errors for each day\u2019s estimate. In the third column the probable error is expressed as a percentage of the estimate. Though there was a decrease in the absolute probable error with each of the subjects, yet when it is expressed as a percentage there was only one case, C. S., in which the relative probable error decreased more rapidly than the average estimate decreased or increased as the case may be from day to day. This fact demonstrates that the changes in the daily average estimates were not due to fatigue. If the effects of fatigue had been more prominent in the early part of the series than in the latter part then there would have been a decrease in the percentage of error. But there was little cause for fatigue, since the subject was required to make only one judgment every 18 seconds.\nThe facts brought out in these experiments are important as throwing some light on the various \u201cindifference points\u201d obtained by different investigators. The estimate is a personal factor which varies with each person and with the same person at different periods of practice. To make a test of this point, after the the close of the series I changed the interval of 1002 to 75s and gave A. F. a few tests. The results were: 50, 49\u00bb 38\u00bb 49\u00bb 49\u00bb 59\u00bb 36\u00bb 43\u00bb 47\u00bb 39\u00bb 53\u00bb 35\u00bb 39\u00bb 33\u00bb 49\u00bb 51\u00bb 38\u00bb 43\u00bb 33\u00bb","page":89},{"file":"p0090.txt","language":"en","ocr_en":"9\u00b0\nW. Smythe Johnson,\n36 (ave. 43s; p. e., 42). These figures, however, may have been influenced by practice on the ioo5 ; the large probable error would seem to indicate it.\nThe above figures, as well as the changes in the time estimate due to practice, Table XI., would seem to indicate that the interval which we judge with maximum accuracy is a changeable one, depending upon the amount of previous practice. Nor does it seem that our \u201ctime-sense,\u201d when unaided by some form of muscular action, is sharpened by practice ; on the other hand it may become less accurate. Without doubt, if some additional basis were allowed, as in counting or moving of the hand, practice would undoubtedly produce good results, in which instance it becomes a measure of muscular strain when carried into overt acts, or of strain of the attention when only the impulse is felt. The differences then in the time estimate of different individuals depend partly upon their nature, whether impetuous and nervous or temperate and deliberate ; partly upon the direction of the attention.\nWhile the nervous temperament may account for the difference in time estimate of different individuals, it does not explain the variations in estimate day by day. For instance, in the case of A. F. (the only subject practiced on two different periods) the case is reversed in the two estimates. He not only underestimated the 100s at the very beginning of the series of experiments but continued to reduce the estimate throughout the whole series. On the other hand, in estimating 164s his underestimate was larger in the beginning of the series than at its close. Consequently it may be inferred that the change in the estimates of the two periods was not due so much to the nervous temperament of the person, as to the conception of the length of the interval which he entertained at the beginning of the series of experiments.\nSummary.\nx. The estimate of a given interval varies for different individuals both with and without practice.\n2.\tThe estimate varies for different intervals with the same person both with and without practice.\n3.\tPractice on the same interval may cause the variation from the given interval to increase with one person and decrease with another. Or it may cause the variation from the given interval to increase with one interval and decrease with another interval.\n4.\tTime estimate is a personal factor depending upon (a) the nature of the person whether of an impulsive or quiet temperament ; (b) upon the point of the fixation of the attention whether to the sensory or to the","page":90},{"file":"p0091.txt","language":"en","ocr_en":"Researches in practice and habit.\n91\nmotor side ; (c) upon the acuteness of the sense of the person to strains accompanying the fixation of the attention and muscular tension.\n5.\tThe change that practice produces in the estimation of time is probably due to fixing the attention in the beginning of the series of experiments : (a) to the movement to be performed, in which case the time estimate is shortened in accordance with the growth of automatic control ; (b) upon the sensory side in which case the time estimate is made longer; (c) on the idea which the subject entertains of the interval on beginning the experiment\u2014if he considers it very short, he will at first underestimate it and will increase the time estimate by practice ; on the other hand, if he considers it very long he may shorten it by practice.\n6.\tOur sense of time may not be sharpened by practice ; on the contrary, it may become less accurate.\n7.\tThere is no \u201c indifference point \u201d from which the subject does not vary with long continued practice.\n8.\tThe overestimation of small intervals may also be due to the fact that the subject does not take into consideration his reaction time. The underestimation of longer intervals may also be due to the fact that the impulse to react keeps ripening and soon reaches such a degree of intensity that the subject cannot withhold the reaction.\nV. Regulated rhythmical action.\nApparatus.\nBy regulated rhythm we understand such actions as beating with a signal. The sound which guided the subjects in beating time was produced by a telegraph sounder, arranged in series with a make contact on a revolving drum. At a point on one of the upright standards supporting the drum a spring was attached. To the drum a small projecting arm was attached. When this arm moved round to the spring attached to the standard supporting the drum, it pressed a lever down, thus closing the circuit through the telegraph sounder, and thereby producing a click in the sounder. Consequently, the time elapsing between successive clicks of -the sounder depended upon the speed at which the drum was revolved ; the regularity in the frequency of the sounds depended upon the regularity in the speed of the drum.\nThe drum was run by a small motor to which it was attached by a thread belt. The motor was run by a current produced by three Edison-Lalande batteries, regulated by introducing a resistance of small German-sil ver wire. In this way the speed of the motor could be adjusted with","page":91},{"file":"p0092.txt","language":"en","ocr_en":"92\nW. Smythe Johnson,\ngreat accuracy.1 This method of producing sounds at regular intervals required a motor and drum of great regularity. A careful investigation was, therefore, made on several kinds of drums and recording arrangements.\n1.\tEdison phonograph. The metal cylinder of the phonograph may frequently be used as a recording drum. Three tests at different speeds were made on the form known as the Home Phonograph. At 267.5\u201c\"\" per second, the probable error was x.85\u201c\u201d', or 0.7% ; at a speed of 174.2mm per second, it was 0.42\"1\u201c, or 0.2%; at a speed of 354.3\u201d\u201d\" per second, it was o.84mm, or 0.2%. The drum is thus a very regular one. The small size of the drum is inconvenient, but the works are so strong that they might be used in running a much larger drum.\n2.\tLudwig kymograph made by Baltzar. At the slowest speed, o.47mm per second, the probable error was 0.008 mm, or 2 \u00b0fo at a speed of 0.9\"\u201d\", the probable error was o.oimm, or 1.0% ; at 28.9\"\"\", the probable error was 0.14\u201d\"\", or 0.2\u00b0J0 ; at 254.2\u201c\u201c\u201c, probable error was 1.64\"\"\u201c, or 0.6%. I observed that at the slower speeds one is likely to keep the spring wound up too tightly, in which case the probable error is increased. Even one-half turn of the handle which winds the spring will produce considerable variation in its speed.\n3.\tEdison motor run by Edison-Lalande batteries. Four tests were made at different speeds apd on different dates. Three cells, arranged in series, were used to run the motor. A thread from the motor ran the regular recording drum. The first test was made when the cells were fresh. At the high speed of 1037.7\"\"\u201d per second (over two revolutions per second), the probable error was 0.3%.\nThe next test was made nine days later ; the cells had been in constant use in the meantime. The same arrangements were used, but a slower speed was tried. At 159.3\u201c\"' per second, the probable error was 0.12%. Ten days later another test was made at a speed of 265.5\"\u201d\" per second, the probable error was 0.08%. In another ten days a test was made at a speed of 150.5\u201d\"\", the probable error was 0.13%.\nThese tests show that an Edison motor run by the Edison-Lalande batteries properly arranged, is a very regular source of power. It should be stated that the cells must work for several minutes before they become constant. In no case was I able to get a favorable'record until the cells had been working at least five minutes. The above tests were made after the cells had been working ten minutes or more.\n1 The remainder of the apparatus used was identical with that described in Scripture, Elementary course in psychological measurements, Stud. Yale Psych. Lab., 1896 IV 121.","page":92},{"file":"p0093.txt","language":"en","ocr_en":"93\nResearches in practice and habit.\n4.\tLatent ti?ne and regularity of the Pfeil marker.\n( i ) Break contact. When the break contact was used and the magnetic cores were removed as far as possible from the armature to which the metallic point of the marker was attached, the latent time was found to be 1.1\", with a probable error of 0.09'\u2019\u2019. The current was not strong enough to force the armature to react promptly ; hence the large probable error. When the cores were placed in close contact with the armature, the remanent magnetism was so strong that it caused the latent time to be considerably lengthened, namely to 14.7ff; but the probable error was reduced to 0.03\u00b0\u2019.\n(2) Make contact. When the cores were distant, the latent time was i. 8*, with a probable error of 0.12\u00b0. With the cores close up to the armature, the latent time vvas reduced to 1.3er, with a probable error of 0-5\"\nIf one desires the smallest probable error possible, then the adjustment with the cores close up to the armature on the make contact, and with them distant on the break contact is the most favorable. Several tests were made on intermediate amplitudes of movement of the armature, with the result that the latent time varied approximately with the distance of the cores from the armature, and the probable error inversely with the distance.\n5.\tLatent time and regularity of the Deprez marker. The results of the test on the make contact showed the latent time to be 2.5\u00b0\", with a probable error of o. 6\\a. The latent time in the case of the break contact with the same adjustment as above was 3.8*, with a probable error of o.opr. Owing to the delicacy of the instrument, changes in the amplitude of the armature did not affect the latent time so much as they did in the case of the Pfeil marker. It did, however, vary somewhat with the strength of the current passed through it, the latent time increasing with the strength of the current.\n6.\tRegularity of the spark record. The probable error of spark records under the usual conditions at the Yale Laboratory is o.2Sram. With the same apparatus and the same kind of paper the above amount is constant whether the drum is still or in motion. Consequently the amount that it will vitiate the record depends upon the speed of the drum. With a fast drum it is negligible ; with a slow drum it must be taken into consideration. In these records the sharp metal point of the marker was so bent that it stood perpendicular to the surface of the drum. If it is placed at an acute angle to the surface of the drum, the error is largely increased. When thus placed the spark often-times leaps off at the side rather than at the end of the pointer. The thinnest glazed paper is used.","page":93},{"file":"p0094.txt","language":"en","ocr_en":"94\nIV. Smy/he Johnson,\n7. Noiseless key. For these experiments a noiseless key was necessary because it was found that, while beating time, the subject would quite as often be guided by the sound produced by the key which he used as by the sound coming from the telegraph sounder. After various attempts to get a noiseless key, that which was found best adapted to the purpose of our experiments consisted of a small band of pendulum-wire soldered to a piece of brass which was fastened in a handle. The band rested against a platinum point so that when it was moved the circuit was interrupted. An insulated wire from one pole of the battery was connected with the platinum point through the handle, while the band was connected with the other pole. A soft substance for the finger to strike against in beating time was fastened over the end of the band. By this method we were able to do away with the guidance which the sound from an ordinary telegraph key gave to the ear in directing the beats. Consequently it was necessary for the subject to direct his attention more to the muscular feeling than to the coincidence or the variance between two sounds as would have been the case had a sounding key been used.\nThe experiments were made on I. M. (Ishiro Miyake), A. F. (A. Fisher), and W. J. (W. Johnson). The time interval was one second.\nExperiments.\nA number of preliminary tests on different persons showed a uniform tendency to anticipate the signal made by the telegraph sounder with this form of key, whereas with a sounding key the reverse was quite as often\nTable XII.\nDaily averages.\n\ti.\tM\tA.\tF\tIV. J.\t\nDate 1898.\tConstant\tProbable\tConstant\tProbable\tConstant\tProbable\nNov.\terror.\terror.\terror.\terror.\terror.\terror.\n14\t\u2014 us\t39\t-158\t44\t\u2014 1.59\t36\n15\t\u2014 136\t35\t\u2014 69\t37\t\u2014148\t29\n18\t\u2014 40\t3\u00b0\t\u2014 80\t42\t\u2014 150\t22\n19\t+ i\t66\t+ 27\t26\t\u2014 15\t39\n20\t\u2014 4\t29\t+ 65\t48\t\u2014 24\t17\n21\t+ 12\t28\t\u2014 4\t39\t\t 26\t20\n22\t+ h\t3t\t+ 10\t31\t\u2014 37\t15\nUnit of measurement, i<r = o.oois.\nNumber of beats at each experiment, 40.\nThe sign \u2014 indicates that the subject beat before the signal was heard ; -f, that he beat after the signal was heard.\nThe probable error of a determination varies from \u00b1 4^ to \u00b1 9<r.","page":94},{"file":"p0095.txt","language":"en","ocr_en":"Researches in practice and habit.\n95\ntrue. A series of such results with a sounding key has been previously published in these studies.1\nLikewise the results with the noiseless key given in Table XII. show that all the subjects anticipated the signal, not only on the first day, but also during the first three days in the cases of I. M. and A. F., while W. J. anticipated the signal throughout the series of experiments.\nIn each case the anticipation of the signal decreased until, in the latter part of the series, two of the subjects waited until they heard the signal before beating. The figures given in the table each represent the averages of 40 beats. The constant error was derived by adding all the + and \u2014 errors in each experiment, taking their difference and dividing by the number of experiments.2\nConclusions.\nThe probable error for each of the subjects was in the beginning of the series of experiments much less than the constant error. This shows that the tendency of the subject at first was to be governed by the muscular rhythms or his own most natural rate of rhythmical movement. With practice, however, he learned to accommodate the muscular rhythm to the time interval of the signal. However, the diverting of the attention to the signal prevented the rapid decrease of the probable error as was observed in all the other experiments. The psychological order of development in all regulated rhythmical movements is (a) the change from the ordinary rate of muscular action to that of the given rate, and (b) then, the decrease of the probable error.\nVI. Free rhythmical action.\n\u2019 The apparatus consisted of a revolving drum on the smoked surface of which the metal point of a Deprez marker wrote. The spark method was used. Wires from a battery of 4 amp\u00e8res were connected with the mercury cups of a Kronecker interrupter. As the vibrating arm of the interrupter dipped down into the mercury, it closed the circuit through the spark coil. The arm of the interrupter was kept vibrating by a separate self-interrupting circuit through the magnets on either side of the arm. From the secondary coil of the spark coil, one wire led to the drum and the other was connected with the support of the Deprez marker so that at each make of the interrupter there was produced a spark on the line drawn by the point of the marker. The interrupter was adjusted to vibrate 10 times a second.\n1\tSCRIPTURE, Elementary course in psychological measurements, Stud. Yale Psych. Lab., 1896 IV 123.\n2\tScripture, New Psychology, 182, London 1897.","page":95},{"file":"p0096.txt","language":"en","ocr_en":"\u00c7\u00d4\nW. Smythe Johnson,\nIn order to prevent irregularities in the spark record and to keep the mercury from oxidizing, a stream of water was allowed to flow over the surface of the mercury. According to the original plan two flasks, connected by rubber tubing with the mercury cups, were filled with water and placed above at sufficient height to get the necessary amount of pressure. This, however, was replaced by a much more convenient reservoir, arranged by Dr. Scripture, which was fastened to a tripod and placed a little above the interrupter. The reservoir was connected by rubber tubing with a hydrant. Two rubber tubes were led from the base of the reservoir to the two mercury cups connected with the interrupter. Any amount of water could in this way be supplied. In the event of too large a supply there was a waste pipe connected with the reservoir which prevented it from overflowing.\nThe Deprez marker was connected with a break contact key in the quiet room. Each movement of the key made a break in the circuit, and this in turn produced a movement in the armature of the marker. As in the preceding section, the noiseless key was used because it was desired to ascertain the rhythm of the person when unaided by the ear. The key was held between bags of sand placed on the lap. No restrictions were placed on the amplitude of the movement of the finger or hand in beating time. The subject was directed to choose his own rate of movement and was requested to retain the same speed throughout the series of experiments. All explanations were made before the beginning of the experiments. The subjects were not informed of their rate of movement. In most instances the watch which the subject carried in his pocket was laid aside because it was found that the ticking of the watch caused distraction of the attention with some subjects.\nDaily averages.\nThe figures given in Table XIII. show that all the subjects constantly decreased the interval between beats as practice was continued. Though each practice lasted during the time required to make from 250 to 400 beats, only sections of each practice are given in the table. Section I. gives the averages of the first 40 beats; section II. the averages of the next 40 beats; and section III. that portion included between the 160th and 201st beats.\nThe shortening in the average time of the beat is apparent in every case. The shortening was probably due to the increasing ease with which the subject performed the experiment. His mental processes probably became more fluent as the result of practice. It was also noticed in several cases that the amplitude of the movements of the hand also de-","page":96},{"file":"p0097.txt","language":"en","ocr_en":"Researches in practice and habit.\n97\nTable XIII.\nAverage daily rhythmical movement.\nate 189 Name.\tcr n> \u00a3 p & cn '\tp. e.\tp. e. as %\tP.E.\tcr to n d. !\u201d O\tp. e.\tp. e. as %\tP.E.\tV' 0 0 pa \u00d4 \u201c tO da O 0\tp. e.\tp. e. as %\tP.E.\nCO\tO\t\t\t\t\t\t\t\tP 3\t\t\t\nXI \u201420\t73\t3-2\t4.4\t0.5\t64.\t34\t5-3\to-5\t53\ti-7\t3-2\to-3\n21\t53\t2.4\t4-5\t0.4\t52\t2.0\t3-9\to-3\t49\t2. i\t4-3\to-3\n22\t46\ti-5\t3-3\t0.2\t45\t1.2\t2.7\t0.2\t45\t1.0\t2.3\t0.2\nA. F.\t23\t43\t1.9\t4.4\to-3\t\u2022 42\ti-9\t4-5\to-3\t45\t2 0\t44\t0-3\n24\t40\ti-7\t4.2\to-3\t37\t1.2\t3-2\t0.2\t33\t1.0\t3-o\t0.2\n25\t36\t1.2\t3-3\t0.2\t35\ti-3\t3-7\t0.2\t34\t2.2\t6-5\t0.4\n26\t32\t1.1\t34\t0.2\t32\ti-3\t4.0\t0.2\t32\tj.i\t34\t0.2\n27\t25\t1.2\t4.8\t0.2\t25\t1.1\t44\t0.2\t25\t1.1\t44\t0.2\nX-25\t99\t3-i\t3-2\t0.6\t91\t3-o\t3-3\to-5\t89\t34\t3-8\t0.6\n27\t73\t3'2\t4.4\t\u00b0'5\t76\t2.9\t3-8\to-5\t71\t4.1\t5-8\t0.7\n28\t69\t2.6\t3-8\t0.4\t64\t2.7\t4.2\t0.4\t61\t2.2\t3-6\t0.4\nW. S.\t29\t62\t1.9\t3-i\t0-3\t60\ti-5\t2.5\to-3\t60\t1.7\t2.8\t0.3.\n30\t63\t2.7\t4-3\t0.4\t60\t1.8\t3-o\t0-3\t57\t1.9\t3-3\t0-3\nXI\u2014 I\t61\t2.2\t3-6\t0.4\t60\t1.2\t2.0\t0.2\t56\ti-7\t3-o\to-3\n2\t59\ti-5\t2-5\t0-3\t58\t1.0\ti-7\t0.2\t55\t1.6\t2 9\to-3\nX \u201421\t84\t3-1\t3-7\t0.5\t86\t2.9\t34\t0.5\t87\t3-3\t3-8\to-5\n22\t56\t1.4\t2.5\t0.2\t55\t1.1\t2.0\t0.2\t54\t1.6\t3\u2019\u00b0\to-3\n23\t63\t1-7\t2.7\to-3\t65\ti.7\t2.6\t03\t62\t2.0\t3-2\to-3\nC. S.\t25\t59\ti-7\t2.9\to-3\t60\t1.4\t2-3\t0.2\t58\ti-5\t2.6\to-3\n26\t55\t2.4\t44\t0.4\t54\t2.1\t3-9\t0.3\t54\ti-7\t3-2\to-3\n27\t54\t2.1\t3-9\to-3\t53\ti-7\t3-2\to-3\t51\t1.6\t3-i\to-3\n28\t52\ti-7\t3-3\to-3\t52\t1.2\t2-3\t0.2\t51\t1.4\t2.8\t0.2\nX-23\t185\t4.6\t2-5\t0.7\t175\t5-2\t3-o\t0.9\t163\t5-i\t3-i\to.S\n24\tno\t3-5\t3-2\t0.6\t108\t4.0\t3-7\t0.7\tior\t6.2\t6.1\t0.9\nE. F.\t25\t133\t5-5\t4.2\t0.9\t137\t3-5\t2.5\t0.6\t143\t3-7\t2.6\t0.6\n26\t122\t3-5\t2.9\t0.6\t117\t2.4\t2.1\t0.4\t123\t2.9\t2.4\t0.5\n27\tit4\t2.5\t2.2\t0.4 '\t112\t2.0\t1.8\to-3\t108\t2-5\t23\t0.4\nX-23\t78\t3-*\t4-0\t0.5\t75\t1.9\t2.5\to-3\t75\t3-5\t4-7\t0.6\n24\t66\t2-5\t3-9\t0.4\t64\ti-9\t3-o\t0-3\t63\t1.6\t2.5\to-3\nE. W.\t25\t70\t2-3\t3-3\t0.4\t68\t1-7-\t2-5\t\u00b0-3\t63\t2.0\t3-2\t0.3\n26\t64\t1.8\t2.8\t0-3\t64\ti-5\t2.4\to-3\t60\t2.0\t3-3\to-3\n27\t63\t1.6\t2.6\to-3\t62\t1.2\t1.9\t0.2\t59\t1.6\t2.7\t0.3\nXI-23\t44\t3-o\t6.8\to-5\t40\t1.8\t45\t0-3\t37\ti-5\t4.1\t0.2\n\u25a0\t24\t40\t2.0\t5-o\to-3\t37\t1.0\t2.7\t0.2\t33\t1.0\t3-0\t0.2\nJ- P-\t25\t40\ti-5\t3-8\t\u00b0-3\t38\t1.2\t3-2\t0.2\t34\t1.0\t2.9\t0.2\n26\t37\t1.4\t3-8\t0.2\t36\t1.0\t2.8\t0.2\t33\ti-3\t4.0\t0.2\n27\t37\t0.9\t2.4\t0.2\t35\t0.8\t2.3\t0.1\t33\t1.0\t3-\u00b0\t0.2\nX-31\t53\t3-o\t5-7\t0.5\t49\t3-i\t6-3\to-5\t46\ti-9\t4.1\t\u00b0-3\nXI\u2014 i\t56\t3-8\t6.8\t0.6\t54\t2-5\t4.6\t0.4\t50\t2.8\t5-6\t0.5\n2\t54\t2-3\t4-3\t0.4\t52\t2.0\t3-9\t\u00b0-3\t46\ti-9\t4.0\to-3\nW- J-\t3\t50\t2.4\t4.8\t0.4\t49\t1.8\t3-7\t\u00b0-3\t46\t1.9\t4.1\to-3\n4\t48\t1.6\t34\t0.3\t48\t1.2\t2-5\t0.2\t46\t1.5\t3-3\t0.3\n5\t47\t1.2\t2.6\t0.2\t47\t1.0\t2.1\t0.2\t45\t1.6\t3'6\to-3\n6\t47\t1.0\t2.1\t0.2\t46\t1.0\t2.1\t0.2\t45\t1.2\t2.7\t0.2\nThe unit of measurement is 12 = o.oiB.\nThe number of experiments at each practice ranged from 250 to 400. The probable error of a determination varied from \u00b19(7- to \u00b1 2<r. p. e., probable error of each beat from the average on that day.\nP. E., probable error of determination.\n. 7","page":97},{"file":"p0098.txt","language":"en","ocr_en":"9\u00a7\t'\tW. Smythe Johnson,\ncreased ; this was probably due to the increasing ease of the muscular adjustments.\nThe results also show for most subjects a decrease in the relative probable error, but for others a nearly constant probable error. This indicates for the former class an increase in regularity. Those who did not improve to any extent in regularity were those who most shortened the average beating-time. It may be suggested that the former subjects directed their attention mainly to regularity and the latter mainly toward ease of movement. According to the record all the subjects were reported to have kept the same speed according to the directions given. Since their attention was fixed more closely on the regularity of the movement any shortening of the interval if it did not materially effect the regularity of the movement would not be noticed.\nThe interval for each individual, then, varies with the development of automatic control. Unless the subject is directed by some external sound to a certain time interval, until a definite muscular movement has become habitual, the tendency will be to decrease the interval between successive beats until that stage is reached when the movement is guided by the natural rate of the response of the nervous system.\nIn every instance, at the beginning of the series of experiments, each of the subjects had in mind some musical composition which guided him in beating time. As soon as the subject felt that he could beat time more regularly by being guided by the muscular feeling then the other was given up. When the subject thought of some musical composition the movement was much slower and more irregular than it was when it became automatic.\nPreliminary tests with a sounding key showed that when the interval was guided by the ear there was a tendency to emphasize certain beats depending upon their frequency. This, of course, was due to the fact that one is accustomed to emphasize certain sounds in music and speech.\nThat the physiological rhythm is different from the emphasis rhythm is very clearly established by the results of this set of experiments. When the subject thought of a musical composition the movement corresponded to that which Bolton1 defined as the \u201c 2-groupings. \u201d However, in the latter part of the series, when the movement came to be controlled by the muscular feeling, the rate of movement corresponded to that which he called the \u201c grouping by fours.\u201d\nOur observations justify us in stating that rhythmical movement rests not only on the rate of breathing, heart-beat, etc., but also on other physiological and mental processes. It is a mental process in that the\n1 Bolton, Rhythm, Am. Jour. Psych., 1893 VI 215.","page":98},{"file":"p0099.txt","language":"en","ocr_en":"Researches in practice and habit.\n99\nduration between successive beats is mentally estimated ; it is a physiological process in that the time interval can be most accurately measured by muscular action.\nRelative decrease of error.\nIn order to see how the error was influenced by practice, the decrease of the errdt from day to day was computed according to the formula given on page 61. The results are given in Table XIV. The numerals\nTable XIV.\nDaily decrease of error shown in percentages.\nSubject.\tSection of beats.\tI,\t2\tSerial number of experiment. 3\t4\t5\t6\t\t\t\t7\tAve\nA.F.\tI\t50\t37\t5\t32\t31\t0\t\u2014 9\t21\n\tII\t.58\t33\t\u201412\t35\t\u2014 3\t8\t15\t19\nW.S.\tI\t24\t32\t20\t\u201412\t32\t32\t\t21\n\tII\t38\t44\t49\t11\t39\t17\t\t33\nE.F.\tI\t18\t\u201410\t45\t29\t\t\t\t20\n\tII\t43\t34\t37\t16\t\t\t\t32\nC.S.\tI\t34\t-38\t\u201434\t\u201433\t18\t15\t\t\u201414\n\tII\t46\t-47\t6\t\u201419\t31\t3\u00b0\t\t14\nE.W.\tI\t34\t24\t26\t11\t\t\t\t26\n\tII\t17\t23\t21\t20\t\t\t\t20\nJP-\tI\t51\t37\t23\t34\t\t\t\t36\n\tII\t44\t0\t17\t20\t\t\t\t20\nW.J.\tI\t32\t55\t33\t47\t31\t17\t\t36\n\t\t5\u00b0\t44\t37\t41\t17\t0\t\t31\n\tAvg.\t38\t19\t19\t17\t24\t15\t\t\nI and II represent the percentages for the first and second groups of 40 beats respectively. The percentages in this table are based on the results given in Table XIII.\nI and II, indicate respectively the first and second groups of 40 beats. Briefly, the percentages in that table show the average amounts of gain from the first, second, etc., days until the close of the series of experiments. The averages in the vertical column show that only with one subject, C.S, did practice produce negative results; and this occurred only in the first 40 beats. In the second group of 40 beats, however, the average daily gain in regularity amounted to 14%. Undoubtedly the large variations in the first group were occasioned by the irregularities in the nervous condition of the subject ; it was necessary to take him at times when he was very busily engaged at work.\nTwo general characteristics mark the results given in Table XIV.\nFirst. As rhythmical movement became more automatic, the percentage of decrease in the probable error proportionately decreased. This may be seen, not only by an inspection of the records for each of the","page":99},{"file":"p0100.txt","language":"en","ocr_en":"100\nIV. Smythe Johnson,\nsubjects, but is expressed, in a general way, by the average daily percentage of decrease as given at the bottom of the table. The same thing is seen here as in all our previous experiments and that is that the beneficial effect of practice in the reduction of the irregularities of movement is considerably greater in the first experiment than in any succeeding experiments. This decreases with considerable regularity as practice continues from day to day.\nSecond. The average daily gain in regularity of movement varies with each individual. The averages given in the vertical columns show variations ranging from \u2014 14% to + 36%. The greater uniformity, as well as the usually larger percentages of the second group of 40 beats show that it requires some time before the muscles can adjust themselves to the steadiest and evenest action. Consequently the period of the first 40 beats may be considered as the training period of the muscles.\nFinally, our results'show that the principal effects of practice consisted mainly in overcoming the previous effects of emphasis rhythm, the rhythm of speech, poetry and music. As this was overcome the beating-time became more rapid and at the same time more regular.\n[VII. General summary.\nThe object of this investigation was to ascertain the results of practice on voluntary movements by repeating the same movements an equal number of times each day until approximately the highest degree of perfection attainable was reached.\nI. Tria?igular movement of the arm.\nThe first experiment consisted in tapping continuously at the corners of an equilateral triangle whose sides measured 20\u201c. The tests each day lasted only a short time ; they were performed from 6 to 11 days by seven persons.\nThe results of the experiment showed that the greatest gains in rapidity of triangular movements of the hand as well as in the regularity of successive movements were made in the early part of the practice. The percentage of gain in speed rapidly decreased, being 20% for the second day, 10% for the fifth, and 5% for the ninth day. The probable error was used as a measure of irregularity. The percentage of decrease in irregularity of successive movements was not so large in the first part of practice as the perc\u00e8ntage of gain in speed ; but after the fourth day the percentage of decrease had grown until it exceeded the percentage of increase in rapidity, thus demonstrating that the psychological order of development in voluntary movement is (1) rapidity, and (2) regularity.","page":100},{"file":"p0101.txt","language":"en","ocr_en":"Researches in practice and habit.\tloi\nThe results also showed that during each practice period the subject constantly increased in speed and regularity of movement until the setting in of fatigue. However, where the exercise was continued, after a short interval there was a renewal of the effort and the same results were observed to occur, though the period was much shorter than in the former case. These periods of renewal of energy were observed to become shorter each time until they came to effect almost every alternate movement.\nII. Drawing circles.\nThis experiment consisted in making circles with the free arm movement. A true circle, drawn with a compass, 6omm in diameter was placed before the subject as a copy. Preliminary tests showed that ten circles at one sitting gave the best general results. The tests were made on seven subjects, extending over six days.\nThe results showed that with the right hand most of the subjects gained in smoothness of contour in their drawings both during the progress of each practice and from day to day ; with the left hand the results were more irregular.\nThough all gained in the smoothness of contour of their curves, yet all did not make them of a size corresponding to that of the copy. These results brought out three types of practice : ( i ) that in which the subject decreased the size of the circle both during the progress of each experiment and from day to day; (2) that in which the size of the circle was increased during the experiment but decreased from day to day : (3) that in which there was but little variation from the copy either during the progress of the experiment or from day to day. The first two classes were those who regarded more carefully the smoothness of contour of their own drawn curves than they did their correspondence in size to that of the copy. The third class were those who directed their attention more especially to the size of the curve, and who closely observed the copy each time before beginning to draw their own curves.\nThe results also showed an important principle bearing on pedagogy : that a short exercise often repeated is the best method of practice for rapid development of accurate adjustment of the muscles. Long practice at writing, drawing, etc., seems to be time and energy wasted. Not only are inattentive habits cultivated, but every wrong adjustment gains a place in the chain of subconscious memories, and therefore delays the development of the control over the muscles for accurate adjustments.\nIII. Development of control over untrained muscles and less adaptable joints.\nThis experiment consisted in tapping continuously with the large toe","page":101},{"file":"p0102.txt","language":"en","ocr_en":"102\nW\u25a0 Smythe Johnson,\nuntil it was completely fatigued. The make and break contacts of an electric key were connected with markers so that each movement of the key was recorded on the smoked surface of a revolving drum. - In this way each phase of the toe\u2019s movement could be measured; the phases were four, namely, the downward movement, the downward rest, the upward movement and the upward rest.\nThe average tap-time of the subject studied was on the first day 436er ; this very regularly decreased until at the close of the practice it was 212\u00ae'. Likewise, the probable error decreased from 103\u00ae to 35\u00ae. Moreover, the upward rest was longer in the first part of the practice than the other three phases combined ; but at the close of the series, it was the same as the downward rest, thus showing that the greatest gains in voluntary activity are those resulting from the practice of the weakest and less exercised muscles.\nIV. Estimation of time.\nAfter a number of preliminary tests, the intervals, 82s, 1002, and 1642 were chosen. The practice lasted from 8 to 16 days on seven subjects.\nThe results justify the following conclusions: (1) The estimate of a given interval varies for different individuals both with and without practice. (2) Practice on the same interval may cause the variations from the given interval to increase with one person and decrease with another. (3) Time estimate is a personal factor depending upon (a) the nature of the person, whether of an impulsive or quiet temperament, and (b) upon the point of the fixation of the attention, whether to the sensory or the motor side. (4) There is no \u201c indifference point \u201d from which the subject does not vary with long continued practice. The changes that practice produces in the estimation of time are probably d\u00fce to fixing the attention on the movement to be performed, in which case the estimate is shortened in accordance with the growth of automatic control, or to the sensory side in which case the time-estimate is made longer by practice.\nV. Regulated rhythmical action.\nIn arranging apparatus for this experiment the probable error was found for the Edison phonograph to range from 0.2% to 0.7%; for the Ludwig kymograph by Baltzar, from 0.2% to 2 % ; for a drum run by an Edison motor driven by carefully tended Edison-Lalande batteries, for o. 1 % to 3%. The Pfeil marker was found at a break of the circuit to have a latent time ranging from 1.1\u00ae \u00b1 0.09\u00ae with the magnet cores distant from the armature to 14.7\u00ae \u00b1 0.03\u00ae with the cores close to","page":102},{"file":"p0103.txt","language":"en","ocr_en":"Researches in practice and habit.\t103\nthe armature. At a make the latent time ranged from i.8\u00b0\u2019\u00b1o.i<r to i.3ff \u00b1 0.50'. With this marker the make is nearly as good as the break except for its slightly greater irregularity. The Deprez marker from Verdin showed a latent time at the break of 3-8,7\u00b1 0.07er and of 2-517 \u00b1 0.64er at the make. Changes in the adjusting spring did not make any great changes in the figures. The probable error of the spark records was found to be \u00b1 0.25\u2122\u201c independent of the speed of the drum.\nIn beating time in unison with a sounder click each subject had his own constant error ; this was generally negative, that is, the subjects generally beat time before the click occurred. With practice the constant error tended steadily to decrease, to become positive and to increase positively. The irregularity steadily decreased.\nVI. Free rhythmical action.\nThe seven subjects were required to beat time without any objective signal. The interval chosen at the start was unintentionally shortened with the progress of the experiment ; it was also shortened from day to day. The irregularity decreased in like manner.","page":103}],"identifier":"lit28747","issued":"1898","language":"en","pages":"51-103","startpages":"51","title":"Researches in practice and habit","type":"Journal Article","volume":"6"},"revision":0,"updated":"2022-01-31T15:21:10.623433+00:00"}