Classics in the History of Psychology

An internet resource developed by
Christopher D. Green
York University, Toronto, Ontario
ISSN 1492-3713

(Return to index)

The Mental Traits of Sex

Helen Bradford Thompson (1903)
Posted June 2000

Classics Editor's note: The numbers appearing in parentheses throughout the text refer to the items
in the bibliography at the end of the volume.

[p. 8] CHAPTER II.


THE subjects of the tests on motor ability were as follows:

A. Reaction times.
    1. Auditory.
    2. Visual.

B. Rapidity of finger movement and rate of fatigue.

C. Co-ordination.
    1. Formation of a co-ordination. Card-sorting.
    2. Accuracy of a formed co-ordination.
        (a) Striking a target.
        (b) Precision of movement in drawing lines.

D. Motor automatisms.


Two sets of simple reaction times, the first auditory and the second visual, were taken from each subject. The Hipp chronoscope was used for both. The auditory stimulus was a, click in a telephone receiver, made by breaking the circuit. A flash of pale purple light in a Geissler tube served as the visual stimulus. Between forty and fifty reactions of each kind were made by each subject. While reacting, the subject sat alone in the reaction room, placed in as comfortable a position as possible. His right forearm was supported on the table, and the forefinger of his right hand rested lightly on the button of a break key which was in the chronoscope circuit. The telephone for the auditory stimulus was adjusted to the ear of the [p. 9] subject. The Geissler tube for the visual stimulus was placed on the table in front of the subject. It was suspended before a black background, against which the flash of pale purple light was thrown out sharply. A warning signal consisting of two clicks of a telegraph instrument was given about two seconds before the stimulus. The time was varied slightly to counteract the widespread tendency to premature reactions. The signal for the release of the key after the reaction consisted of one click of the telegraph instrument. The reactions were taken in series of ten to twelve, with four- or five-minute periods of rest between. The subject was given no instructions about the direction in which his attention should be concentrated. He was merely told to make the reaction as nearly instantaneous as possible. After the experiment he was asked in what direction his attention had been concentrated while reacting.

The results are recorded in terms of the mean reaction time and mean variation (74). Series of the first forty-one unquestioned reactions of each kind served as the basis of the calculation. Both the mean reaction time and the mean variation appear in the

results in terms of sigma (.001 seconds). The number of subjects represented by each ordinate in the [p. 10] curve is the number whose reaction times fell between the point on the curve at which the ordinate is erected and the next previous one. Thus the number of subjects represented at 150s is the number whose reaction times fell between 140s and 150s.

The curves represented in Figs. 1 and 2 show that the men have, on the whole, shorter reaction times than the women. In both auditory and visual reactions there are several men with shorter times than any of the women, and several women with longer times than any of the men. Moreover, the men are decidedly more numerous than the women in the region of short times. The

average of the mean variation (Table I) is also smaller for the men than for the women. The difference is not apparent in the lowest ranges. It is shown by the smaller number of women in the middle ranges, and their greater number in the region of very large variations. The fact that the mean [p. 11] variations of the visual reactions are in both sexes less than those of the auditory, is doubtless partly due to the fact that the auditory reactions were taken first. The effect of practice is shown in the greater evenness of the visual reactions.

The shorter reaction time of the men is at least partly explained by special training in athletics. The man who made the most rapid reaction both to the auditory and to the visual stimuli was one of the best players on the football team. The other three men who made very quick auditory reactions were track athletes, one a bicycle rider and the other two runners. Two of these latter had visual reactions also which were shorter than the visual reactions of any woman.

The type of the reaction was recorded under one of the three general heads, sensory, motor, and central. The central rubric includes all cases in which the subject reported that his attention had been equally divided between stimulus and movement.

There is a decided preponderance of women with a sensory type of reaction. The adherents of the Leipsic school would doubtless say that the shorter reac-[p. 12]tion time of the men is to be explained by the greater proportion of motor reactors among them. Probable as this theory looks from the tables, it is not borne out by a detailed examination of results. The men with the shortest reaction times were in most cases of the sensory type, while several of the motor type were among those with longest times. The real explanation of the greater frequency of the motor type among men is rather to be sought in the fact that they lead more active lives on the whole than women, and are more interested in learning new movements of various sorts. For this reason their attention is more likely to be directed to the technique of movement than is that of women.


The apparatus used to ascertain the rapidity of finger movement, and the rate at which the finger becomes fatigued, was a counting machine worked by a rod bearing a disc on which the finger rested. A dial on the front of the machine registered the number of times the rod was pressed. The machine was fastened in a wooden support on a table, with the rod projecting upward. The wooden support was extended into a rest for the arm. The subject sat at the table with his forearm from elbow to wrist resting on the support, and the index finger of his right hand on the disc of the rod. When in this position, every downward movement of the finger pressed the rod down and was registered on the dial of the machine. The arm was bound in position at the wrist and at the elbow to confine the movement as much as possible to [p. 13] the finger muscles. In spite of this precaution the arm came into play somewhat, particularly after fatigue set in. But although it was impossible to limit the motion strictly to the finger muscles, still they were principally involved, and the conditions were the same for all subjects. In pressing down the rod, the finger was working against a considerable resistance -- about that of a stiff-action piano key.

The subject was told that the object of the test was to find out how rapidly he could make the movement. He was not told how long he was to continue it. His only instruction was to start the instant the signal was given, and keep up the movement until he was told to stop. The dial readings were taken every twenty seconds by the second hand of a watch. The subject was stopped at the end of two minutes, if he had not already given out. The movement had by this time become painful in every case. The test was made twice. The second time the subject of course knew that it would have to be continued until it became painful, but he was told not on that account to try to save his strength by going slowly at first, but to go as fast as possible at the start, and let the running-down process take its natural course. The results which appear in the curves are averages of the two tests.

[p. 14] The results given in Figs. 3 and 4 and Table III, show a striking advantage on the part of the men, both in the initial rate of the movement, and in the ability to sustain it. The men made on an average about ten more taps in twenty seconds than the women. Only two men gave out before the end of two minutes, while eight women did so. One of the two men had had his arm permanently weakened by a fracture. The men had an average of about twenty taps in twenty seconds faster than the women at the close of the test.

It is interesting to note in connection with this test that it has been shown by Professor Oscar Reif of Berlin (70) that the rate of movement of the separate fingers is not greater in piano players than in other people. The only way in which piano practice would give an advantage in this test is by increasing endurance through the general strengthening of the hand. In so far as this factor affects the results, it is in favor [p. 15] of the women, since there were more piano-players among them than among the men.

Professor Féré (24) makes the suggestion that probably the force and the rate of voluntary movements vary together. The present series of tests certainly tends to corroborate this theory. The amount of force required for the movement was even at the outset well within the limits of strength for both sexes, but the rate appears constantly as a function of the strength. The same relation between force and rate may account for the faster reaction times of men.


1. Formation of a co-ordination. -- The apparatus used for testing the ability to form a co-ordination was one of the boxes of the Jastrow card-sorting apparatus (39). Its four divisions were marked with discs of the four pure colors, red, blue, green and yellow. There were forty cards in the pack, ten of each color. Before each test, the pack was so arranged that no two cards of the same color followed one another. The directions given to the subject were to sort the pack as rapidly as possible, throwing each card into the division marked with its own color, making no stops for mistakes and no attempt to correct them. The signal to start was the word "go," after a count of three. The time was taken with the second hand of a watch. The test was made three times for each subject. To shut out the effects of practice and insure a fresh co-ordination each time, the colors on the divisions of the box were arranged differently for each trial. The results are given in terms of the average time of the three trials in seconds and the average number of mistakes.[p. 16]

The curves for the card-sorting test (Figs. 5 and 6) show that the women are decidedly more rapid than the men. The best record is that of a woman. The women's mean rate is about two seconds faster than that of the men, and there are several men with longer times than any woman. The women have also a somewhat higher degree of accuracy than the men.

To ascertain whether or no the handling of playing cards gave an advantage to the card players in this test, the subjects were questioned as to their habits of card-playing. The fact that those who made the best records, both men and women, were people who played cards very little or not at all, indicates that practice in card playing is not of great importance in this test. In so far as it is a factor, it would be in favor of the men, since there were more card-players among them than among the women.

In two cases of abnormal slowness among the men, a decided color-blindness is doubtless responsible: None of the subjects were so color-blind that they could not distinguish between the pure colors used on the cards, but in the two worst cases of color-blindness the discrimination was probably slower than the normal. The subject with the longest time reported a feeling [p. 17] of slowness in recognizing the colors, but none of the others were conscious of this difficulty. That the poorer color discrimination of the men (see Fig. 47) could account for their slowness in sorting the cards is impossible, since there proved to be no co-ordination between the rate of card-sorting and the fineness of color discrimination. Several subjects with excellent color discrimination were slower than the average, while several of those with slight partial color-blindness were much faster than average.

The two factors of time and accuracy showed no co-ordination. Some subjects with the shortest times had also the highest degree of accuracy, and some with the longest times were very inaccurate.

2. Accuracy of a formed co-ordination. -- The first of the tests on the accuracy of a formed co-ordination consisted in striking at the center of a target with a rapid free-arm movement. The target was a sheet of paper on which were inscribed nine concentric circles. The central circle had a diameter of 2cm., the next one 4, the next 6, etc., giving a total diameter of 18 cm. to the target. The four radii at right angles were marked at each intersection with a circle, with the number of millimeters from the center; the first one 10, the second 20, etc. The target was hung on the wall at such a height that its center was on a level [p. 18] with the hand when the arm was stretched out straight from the shoulder. The subject then took his stand at such a distance that when his arm was extended before him the point of a pencil held in the hand just touched the center of the target. He was required to attempt to strike the inner circle with the pencil, in rapid thrusts from the shoulder. The rate of movement was timed with a metronome. Before beginning the test the subject was allowed to practice a few strokes on a blank paper for the purpose of learning the rhythm. He was then required to hit the target fifty times.

The results were calculated by counting the number of strokes which fell within each successive 5 mm. section of the target, measured from the center along the radii. Table IV gives the results in full, in terms of the percentage of dots falling within each section of the target.

[p. 19] The general outcome of the test may most readily be seen by comparing the two curves (Figs. 7 and 8) plotted from Table IV. The first one is a graphic representation of the percentage of dots falling within the innermost section (i. e., 5mm.) and the second one of the percentage of dots falling within the 15-20 mm. section, which was the outermost one in which all subjects (except one) placed dots. The first curve shows a greater number of women than men with small percentages of dots in the center of the target, and a greater number of men than women with large percentages. The second curve shows the reverse to be true for the outer section of the target. The men with small percentages are somewhat more numerous than the women, and the women with large percentages than the men. The two curves agree in showing better co-ordination on the part of the men,[p. 20] showing the range of the dots farthest from the center, corroborates this conclusion. There are more men than women who put no dots outside the 20 mm. circle, and more women than men with dots falling beyond 20 or 25 mm., although the best record was that of a woman. The differences, though small in each case, are in accord in showing better co-ordination on the part of the men.

The second of the tests on the accuracy of a formed co-ordination was made with an apparatus modeled after that used by Bryan (11) in his tests on school children for determining the precision of movement. This apparatus, however, was made on a much larger scale than Bryan's, and was used for free-arm movements instead of finger movements. It consisted of two thin strips of copper 21 cm. long fastened to a glass surface in such a way that they were in contact at one end, and diverged very gradually toward the other, where they were about 5 mm. apart. A brass writing-point ending in a small knob was connected by a flexible wire with a battery whose circuit was closed whenever the writing-point touched either of the strips of metal. The

closing of the circuit was announced by the click of a telegraph instrument. The point where [p. 21] the strips of metal were such a distance apart that the knob of the writing-point when placed on the glass between them just made the contact with both strips of metal, was called the zero point. >From the zero point to the ends of the strips millimeter scales were marked on both pieces of metal. The total length of the scale was twenty cm. The glass on which the metal strips rested was sunk into a board and set with putty on a level with the surface of the board. The board thus afforded a support for the hand in making the movement.

For the first test the subject was seated at a table with the apparatus before him in such a position that the strips of metal converged toward him. He was told to start at the twenty cm. point of the scale and attempt to draw a line on the glass between the strips of metal without touching either one. In this position the movement was, of course, toward the body. The subject was allowed to hold the writing-point as he chose, and take his own rate of movement. The only regulation was that the movement [p. 22] must be continuous from start to finish, and must be a free-arm and not a finger, wrist, or elbow movement. As soon as a click of the telegraph instrument indicated a contact, the subject stopped and began again. The point on the scale where the click occurred was noted each time. The subject was allowed two or three trials to see how the apparatus worked, and then the readings of five successive trials with each hand, first the right and then the left, were taken. For the second test the apparatus was turned around, and the movement was made away from the body five times with each hand.

Each of the four sets of results obtained from each subject (i. e., right hand, toward and away; left hand, toward and away) was averaged, and its average variation reckoned.

In all four movements the men have a somewhat greater degree of precision than the women (Figs. 9-12). The right-hand movements are better than the left for both sexes, and the movements toward the body better than those away from it.[p. 23]

The average variations (Table VI) for the sexes approximate one another more closely than the averages. In the movements away from the body neither sex can be said to have greater uniformity. In the movements toward the body the variation of the men is somewhat less wide than that of the women.


The object of the test on motor automatisms was merely to discover whether or not a tendency toward automatic movements was present in the subject. The apparatus employed was that used by Miss Stein (76) in her experiments in this field. It consisted of an oblong board suspended from a hook in the ceiling by ropes attached to its four corners. When adjusted, the board hung in a horizontal position about two inches above the surface of a table, on which was [p. 24] placed a large sheet of rough manila paper. The subject sat at the table with his right arm, from wrist to elbow, resting on the board. He held in his hand, which hung over the edge of the board, a soft black lead pencil, whose point rested lightly on the paper. The board responded instantly to any movement of the arm. Each movement was registered on the paper by means of the pencil.

The instructions given the subject were to place himself in a perfectly comfortable position, such that the arm would have no tendency to move through strain, and then to let his arm do as it pleased, -- move if it wished or stay still if it wished, -- not to inhibit any impulses to movement which arose, nor make any voluntary movements. The subject's attention was distracted during the test by asking him the series of questions on personality given in chap. viii. These questions proved to be a very efficient method of distraction, since the subjects were universally interested in them.

The results were classified under four heads, with reference to the presence of automatisms, i. e., Absent, Doubtful, Present, and Marked. The cases where no movement was registered, or only such movement as was evidently due to slight changes of position, were marked "absent." Those where the amount of movement was greater, but still possibly due to changes of position, or to gradual accommodation to slight strain, were called " doubtful." Cases where the movements were unquestionably automatic arm movements, but slight in extent and number, were marked "present," while those having movements of considerable extent and variety were classified as "marked."[p. 25]

The results, given in Table VII, show a somewhat greater tendency on the part of women to display motor automatisms than on the part of men. The tendency is shown most clearly in the last column of the table.


There are several researches on reaction time to compare with the present experiments. Lewis (46), after experimenting on a large number of American men and women, using both visual and auditory reactions, found that men are quicker than women in both kinds of reaction, and have a smaller mean variation. The Columbia University tests (82) included five auditory reactions for each subject. In these the women were slower than the men. The remaining reaction time tests of which we have a record were made on children. Gilbert (30) has shown that boys are quicker than girls at all ages in auditory reaction, and that boys of over ten years have a smaller mean variation than girls. MacDonald, from his work on the school children of Worcester, Mass. (55, p. 1106), reports a longer reaction time for girls of all ages. Herzen (33, Appendix), from a much less extended series of observations than those on school children, concludes that young girls are quicker in their reac-[p. 26]tions than boys, but that after adolescence the relation is reversed. As far as adults are concerned, therefore, the experimental evidence agrees unanimously with the present series of tests in showing that men have a shorter reaction time and a smaller mean variation than women. The same relation as to rate probably holds for children.

The only comparable tests on the rapidity of finger movement and the rate of fatigue are those performed on school children by Bryan (11, p. 173),Gilbert (30), and Bagley (3); and those reported by MacDonald (55, p. 1105). They are all in accord with the series of tests here reported in finding greater rapidity of finger movement among males than among females. Gilbert also reports that boys are somewhat less easily fatigued than girls, a conclusion which is again in accord with the present results. The Columbia University tests (82) on fatigue show no difference between men and women in this respect. But in this case the experiment was performed with Cattell's ergometer and the subject was required to make fifty pressures on the instrument at the rate of one a second, conditions certainly not so well calculated to produce fatigue, as those of the tapping test. The failure to indicate any sexual difference in fatigue may be due to the fact that the amount of fatigue induced by the experiment was so slight. In the tests recently made on Chicago school children (18) the boys surpassed the girls in both strength and endurance at all ages.

Bagley (3) in his experiments on school children used the card-sorting test in several forms as a test of mental ability. He reports that he found girls some-[p. 27]what superior to boys in mental ability -- a result which is in accord with that of the present test on card-sorting. Another experiment which, although not directly comparable with the card-sorting test, is still of the same type, is the one called a test for the rate of perception in the Columbia University series (82). The subject was given a card containing five hundred printed letters, of which one hundred were A's, arranged haphazard, and was asked to mark out all the A's as rapidly as possible. Here, again, the women were more rapid than the men. The essence of the test in this case also is the formation of a new eye-hand co-ordination.

Both Bryan (11, pp. 192-6) and Bagley (3) find boys slightly superior to girls in precision of movement. Bryan's test was, like the present one, the drawing of a single straight line. Bagley's consisted in tracing a pattern. Bryan also found boys slightly superior to girls in a target test. All these tests on the accuracy of a formed co-ordination agree in showing the male child and adult slightly superior to the female.

The experiments on involuntary movements, and movements influenced by the sight of moving objects carried on by Tucker (81, p. 404) with Jastrow's automatograph revealed no difference of sex in children. Miss Stein's experiments (76) on college students, in which she used the same apparatus which was used in the present tests, produced results which are in accord with those given above, in so far as they are comparable with them. She finds a somewhat greater proportion of women than men who display spontaneous motor automatisms.[p. 28]


All the tests on motor ability of which we have record agree in showing that in most phases this ability is better developed in the male than in the female. Men have a shorter reaction time, with a smaller mean variation than women. They have a greater rapidity of movement than women, and become fatigued less rapidly. They have a somewhat greater accuracy of movement than women. Women excel men in the formation of a new motor co-ordination, such as that of card-sorting and of marking out A's, and are slightly more subject to motor automatisms than men.