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 July 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. 29] CHAPTER III.


THREE groups of problems connected with the skin and muscle senses were investigated; touch and pressure, cutaneous space, and temperature. The subjects under each heading were as follows:

A. Touch and pressure.
    1. Threshold of impact on the volar side of the forearm.
    2. Threshold for pain through pressure on the right and left temples.
    3. Discriminative sensibility for pressure on the palm of the hand.
    4. Discriminative sensibility for lifted weights.

B. Cutaneous space.
    1. Discrimination of two points crosswise and lengthwise on the volar side of the forearm.
    2. Discrimination of areas on the volar side of the forearm.

C. Temperature.
    1. Discriminative sensibility at the physiological zero. Standard, 30° C.
    2. Discriminative sensibility for cold. Standard, 5° C.
    3. Discriminative sensibility for heat. Standard, 45° C.


1. Threshold of impact. -- The piece of apparatus used in the determination of the threshold of impact was one designed by Professor James R. Angell for use in the laboratory of the University of Chicago (1). In brief, the instrument is a delicate balance, and the stimulus is given by a cork surface at the end of a rod suspended from one arm of the balance. So long [p. 30] as the amount of fall of the balance arm and the weight producing the fall are constant, the force of the impact must be constant. The instrument is noiseless, and the stimulations can be given at any rate desired.

The area used in the present investigation was the middle of the volar side of the right forearm. The middle point between the wrist and the elbow was marked with an ink spot. The arm was then adjusted on padding under the balance arm; a flat uniform area, if possible with no hairs, was selected close to the ink spot; and the balance arm was brought over this area. The cork surface was raised 3 mm. above the area, and a trial weight of 20 mg. placed in the weight pan. The stimulations were given in series of ten, with intervals of from twenty to sixty seconds between stimulations. Any rhythmic regularity in the series was carefully avoided. The subject was directed to count aloud every time he felt the stimulations -- one for the first, two for the second, etc. The point at which he could count seven or eight in ten correctly was taken as the threshold. Not more than three series in succession were given without a rest and change of position.

This test was made in combination with the one for the absolute threshold for vision (see chap. vi, sec. A). The subject was obliged to have his eyes completely protected from the light for three-quarters of an hour before the threshold for light could be tested. This time was used for the determination of the touch threshold, and the discriminative sensibility for area (see sec. B, 2, below). During the entire test, therefore, the subject sat with his head in the [p. 31] dark box. Since the ventilation of the box was not good, the conditions were not so favorable for the concentration of attention as in the other tests, but this distraction was the same for all subjects.

Since the size of the contact surface (2 mm. square) and the height of the fall (3 mm.) were kept constant throughout the entire series of tests, and the weight in the weight pan was the only factor varied, the results can be recorded for comparative purposes in terms of weight only. The determination of the threshold was rendered very difficult in some cases by a tendency to imagine stimulations as soon as the threshold was approached. Almost all of the subjects put a few imaginary sensations into the series. In a few the tendency was very marked. One subject -- a man -- counted a whole series of fifteen non-existent stimulations. The tendency was partially counteracted by telling the subject of his error. The point where seven or eight of the real stimulations were correctly counted, regardless of the extra ones inserted, was finally taken as the threshold.

The curve for the threshold of impact (Fig. 13) shows a somewhat lower threshold in the women. In the region of very low thresholds the men and women are present in equal numbers, but the women are in excess in the middle lower ranges, and the men in the upper. The real difference is probably somewhat greater than [p. 32] the curve represents, because the region used -- the forearm -- is more plentifully supplied with hairs in men than in women. It is perhaps worth while to mention, in passing, a curious illusion experienced by several of the subjects in connection with this test. Although all the stimulations of any one series were given on exactly the same spot, several subjects volunteered the observation that they could feel the changes of position of the stimulations very distinctly. One subject said that the successive stimulations described a long oval from wrist to elbow, and that they differed in distinctness in different regions. There was no opportunity to examine further into this phenomenon.

2. Threshold for pain on the right and left temples. -- The thresholds for pain on the temples were taken with a spring algometer registering 4,000 g. (57). The subject was required to lay his head on a thin padding on the table, with one temple up. A piece of chamois skin was placed under the metal disc of the algometer to prevent temperature sensations.

The subject was told that the pressure would be increased gradually and that he was expected to indicate the point at which he first began to feel pain. It was carefully explained to him that he was not to wait until the experience was decidedly painful, but was to indicate the point at which he could just begin to detect a feeling of pain in addition to a mere pressure sensation. The ease with which the discrimination was made differed very widely in different cases. Some subjects had a sudden sharp transition from mere pressure to pressure plus pain, which made it a simple matter to indicate the advent of pain. Others had a [p. 33] very gradually increasing feeling of discomfort, almost from the beginning; to mark any point at which pain could be said to begin was extremely difficult -- in fact, almost arbitrary. In such cases, there was no criterion except the final judgment of the subject himself. The test was repeated four times on each temple. The results which appear in the curves are averages of the four readings. Sometimes the right temple was taken first, and sometimes the left. A period of four or five minutes' rest was allowed between the two.

The curves for the two temples (Figs. 14 and 15) are alike in general outline. In both cases the interpretation of the curve, as a whole, is lower pain thresholds for the women than for the men; but it is also true that both curves Show more men than women with very low thresholds. In general, more women than men are found in the middle ranges, and more men than women at both extremes, but the preponderance of men is most marked in the region of high thresholds.[p. 34]

The interpretation of the results offers the same apparently insurmountable difficulties as that of all similar experiments. Does it mean that women really feel pain more quickly than men, or that they are more apt to call a slightly disagreeable sensation painful than are men? There seems to be no possible criterion for a real decision of this question. The problem is not peculiar to the comparison of pain sensations of men and women. In attempting to make a quantitative comparison of the pain sensations of any two individuals the standard is absolutely subjective, and must be accepted as such. It must in consequence be admitted that quantitative measurements of pain are less capable of control and consequently less reliable than most quantitative measurements of sensation processes. But this fact does not justify the assumption of a difference in pain standard on the part of different classes of individuals. As far as we are capable of interpreting the results obtained, they indicate somewhat greater sensitiveness to pain on the part of women than on the part of men.

3. Discriminative sensibility for pressure on the palm of the hand. -- The apparatus used for testing the discrimination of pressure was a series of little wooden bottles weighted with shot. The series consisted of twenty-one bottles beginning with 80 g. and ending with 100 g. The uneducated hand, in most cases the left, was used for the test. The hand was supported on padding as comfortably as possible and was perfectly relaxed. It had to be placed in such a position that it offered a flat space large enough to allow of setting the bottles upright. A small cork disc was placed on the hand first, to give a smooth surface of [p. 35] contact and insure placing the bottles in approximately the same spot each time.

The two weights to be compared were placed successively on the same spot, and the subject, whose eyes were closed, was asked to say which of the two was heavier. The 100-g. weight was used as the standard. The subject was not told anything about a standard. He was merely asked to make the comparison between the two weights given him, and did not know that one of them each time was the 100-g. weight. The standard was put on sometimes first, and sometimes second, to avoid any constant errors of order. The series of tests began with the large differences, usually 80 and 100. If these were judged correctly every time, a smaller difference was tried. The number of tests with each pair of weights was increased as the limit of discrimination was approached. The point finally fixed upon as the discriminative sensibility was the point at which three-fourths of twelve or sixteen judgments were correct.

The results (Fig. 16) show no marked difference between the men and the women. The women are a little more numerous in both the upper and the lower ranges, but the average is about the same for both sexes.

4. Discriminative sensibility for lifted weights. -- The apparatus used for testing the discriminative sensi-[p. 36]bility for lifted weights was the series of wooden bottles used for the previous test. The subject was required to lift the bottles with the thumb and forefinger of the uneducated hand -- usually the left. He sat with his eyes closed, and his hand held in such a position that the bottle could be placed between the thumb and forefinger. After lifting the first one, he set it on the table, and it was at once replaced by the second. For the rest, the experiment was carried out in the same way as the preceding one.

The results represented in Fig. 17 show a much finer discrimination on the part of the men. There is a much larger proportion of men than of women who can discriminate a difference of 4 to 10 g., while the reverse is true in the range of differences greater than 10 g. Since the discrimination of lifted weights involves principally the joint sensations, this test is closely allied with the motor tests. The finer discrimination of the men for weights is in accord with their better developed motor ability in general.


1. Discrimination of two points crosswise and lengthwise on the volar side of the forearm. -- Jastrow's æsthesiometer was the instrument employed for testing the [p. 37] discrimination of two points on the skin. The instrument allowed, the distance between the two points to be varied from 1 mm. to 100 mm. The region used was the middle of the volar side of the right forearm. The arm was supported on a padding in a comfortable position. The subject was told that he would be touched in the region described, sometimes with one point and sometimes with two, and that all that was required of him was to tell whether he felt one point or two. The measurement of the discrimination crosswise of the arm was taken first, and the lengthwise test followed on another day.

In making the test, stimulations with one point were frequently inserted in the series as a control. Often the judgments seemed to be pure guesses when the difference was really below the discriminative sensibility of the subject. In these cases the distance was increased until a reliable judgment could be made. In a few instances a genuine illusion seemed to be involved, which caused the judgments to remain difficult and variable through a large range of differences. All that could be done was to fix an approximate point after a long series of experiments. Frequently the guessing process would be stopped, or at least much reduced by telling the subject that he was calling one point two. The series of tests was begun with a difference a little below the average discriminative sensibility and was increased or decreased as the case demanded, until the least difference was found at which three-fourths of the judgments of two points out of twelve or sixteen were correct.

The women proved to have a somewhat finer discrimination in the crosswise direction (Fig. 18) and a decidedly finer discrimination in the lengthwise direc-[p. 38]tion (Fig. 19). In the former case the two curves occupy the same range. The difference is shown by the preponderance of women with a small discriminative sensibility and the preponderance of men with a large discriminative sensibility. In the latter case (the lengthwise discrimination the difference is so great that the two curves occupy a different range; the women's curve from 20 to 65 mm., and the men's curve from 35 to 75 mm.

There are two factors which doubtless combine to decrease the apparent difference in the discriminative sensibility of the two sexes in the tests made across the arm. The first is that the curve of the arm makes it very difficult to put the instrument down crosswise in such a way that the two points strike simultaneously and exert the same pressure. Inequalities of time and pressure are therefore much more likely to assist the judgment in the crosswise test than in the lengthwise. The second factor is that the structure of the arm is much more differentiated crosswise than it is [p. 39] lengthwise. If one point rests on one tendon and the other on another it is easy to distinguish two points by an indirect judgment, because the difference between the two tendons is already known. The judgment ceases to be a pure skin discrimination and becomes a complex judgment based on other sorts of experience. In the lengthwise test, on the other hand, the two points fall upon a homogeneous substructure, a single muscle or tendon, and the discrimination is much more nearly a pure skin judgment.

2. Discrimination of areas on the volar side of the forearm. -- The apparatus used for determining the discriminative sensibility of the skin for area was a series of five cork blocks about 3 mm. thick, varying in size from 10 mm. square to 20 mm. square. A preliminary test was made with the blocks all weighted to the same amount -- 20 g. It was found that in this case the smallest block felt so much heavier than the largest that the difference in pressure interfered seriously with the judgment of size. Either the smaller block was called larger, because the factor of pressure was not clearly separated from that of size, or the subject reported himself unable to make any reliable size judgment because of the disturbing difference in weight. With the hope of remedying this evil the blocks were then weighted proportionately to their area, so that equal amounts of pressure should be exerted on equal skin areas in all stimulations. This attempt was only partly successful. The smallest block now felt lighter than the largest. The series of tests was nevertheless carried out with the latter blocks, because the difference of pressure was much smaller with them than with the former; but the [p. 40] results cannot be regarded as entirely reliable. Until the relation of pressure and area in judgments of area on the skin has been made the object of special investigation, and series of areas weighted to produce equal sensations of pressure have been determined, no thoroughly reliable results in this field can be obtained. This series of tests, as well as all previous investigation, suffers from this defect.

The area of skin employed for the discrimination of size was the same as that used for the touch and space thresholds -- the middle of the volar side of the right forearm. This experiment, like that for the touch threshold, was made during the time required for the fatigue of the retina for the threshold of light (see p. 76). The method was that used for all the experiments in discriminative sensibility; two areas were applied successively, and the subject was asked to report each time which of the two felt larger. The series began with the largest difference and worked down to the smallest difference, regarding which three-fourths of the judgments out of twelve or sixteen were correct.

The curves presenting the results of the test (Fig. 20) show a somewhat better discrimination for area on the part of the men. Their curve culminates at 15, and that of the women at 12.5. The outer limits of the two curves are the same.[p. 41]


The only aspect of temperature sensation experimented upon was that of the discriminative sensibility. It was tested with three different standards: one near the physiological zero, 30° C.; one approaching the pain threshold for cold, 5° C.; and one in the region of the pain threshold for heat, 45° C.

The method of giving the stimulus was the same in all three cases, viz., immersing the first two fingers of the right hand to the second joint in water. To facilitate preserving a constant temperature, a large mass of water was used. The apparatus consisted of two large zinc basins, eighteen inches long, ten inches wide, and six inches deep. They were filled to within an inch of the top. The basins were covered with asbestos jackets to prevent changes of temperature. Asbestos lids with openings for the thermometer and for the immersion of the fingers covered the basins. The thermometers, reading tenths of a degree, were hung very close to the place where the fingers were immersed, to insure the recording of the temperature of the water actually used in the stimulation. Each basin was set on a tripod, and supplied with a Bunsen burner for changing the temperature. For the cold stimulation, ice was used. The temperature changed very slowly, and by leaving a low flame, experimentally determined, under the basin, it was possible to keep the temperature constant through considerable periods of time. The changes of temperature required some time and patience. They could be produced rapidly enough, but it was difficult to bring them to a standstill at exactly the point required for the test. To economize time, the intervals required for chang-[p. 42]ing the temperature were employed by the subjects in writing their answers to the questions on general information (see chap. vii, sec. D), and in sorting the worsteds for the test on color-blindness (see chap. vi, sec. D). It required from half an hour to an hour to determine the discriminative sensibility with each standard.

The method of making the discrimination was the same as that used in the other tests. The subject was told to put his fingers first into one basin and then into the other, and tell which felt the warmer. The fingers were dried after each discrimination, and time was allowed for the effects of the extreme stimuli to disappear. It was not possible to make so many judgments for each stimulus difference as in the case of the other skin discriminations, partly because the effects of the extreme stimuli are so lasting that only a few tests can be made without long rest intervals, and partly because it was impossible to hold the temperature absolutely constant for many tests at a time. Consequently, three correct judgments out of four, or at most four out of six, were regarded as decisive. If further tests threw doubt on the accuracy of any determination, the same stimulus difference was tried a second time.[p. 43]

Since the order in which the stimuli are given is a very important factor in temperature discriminations, great care was taken to see that an equal number of judgments was made in each order. The summation of stimuli which tends to make the second stimulus feel more intense than the first is more marked in temperature than in any other sense. The difference required to make the more intense stimulus feel more intense when it was given first was frequently very large in the cold and hot ranges, whereas when it came second, a very slight difference was sufficient. In fact, when the two were of the same temperature, or the second a little less intense, the second was judged more intense. In the form of temperature test in which the subject is allowed to put his fingers back and forth from one basin to the other, much smaller absolute differences can be discriminated than those reported in this test, but the judgment made is not a simple sense discrimination comparable with those made in the other senses. For example, if a discrimination with two very cold temperatures is being made, and the subject is allowed to have each stimulation but once for each judgment, he will say that the second one is colder each time, but that the difference in temperature is much greater in one order than in the other; and that he therefore believes that the basin which when second is a colder [p. 44] second, is really colder. The same process in less conscious form is what takes place when the subject is allowed to change back and forth from one basin to the other. Each stimulus, as he gets it, feels colder than the previous one, but the difference is much more intense when he goes from the really less intense to the more intense than when he goes in the opposite direction. He reaches a correct judgment as to which is colder, but the judgment is not a simple temperature discrimination; it is an indirect judgment. The absolute values obtained for temperature discrimination are therefore largely dependent on the method. The results differ with the method far more in the extreme temperatures than in those near the physiological zero. The present results represent simple temperature discriminations, and show correspondingly large values for the discriminative sensibility in extreme temperatures.

The curves for the temperature tests (Figs. 21-23) show very slight variation in the sensibility of the two sexes. At the physiological zero no distinction can be made. In the two extreme temperatures the women have a slight advantage. They are grouped somewhat more toward the region of fine discriminations, but the difference is scarcely large enough to be regarded as significant.[p. 45]


There are no results directly comparable with the test on the threshold for impact. The so-called measure of the fineness of touch in the Italian investigations (Lombroso, Ottolenghi, Di Mattei) is an æsthesiometer test. The experiments on general sensibility, however, are often considered to be a measure of the delicacy of tactile sensations. Ottolenghi (66) calls general sensibility a sort of contact sensation. Its measure is the least amount of a faradic current which can be perceived. Lombroso (51 chap. iii) reports that women have a less keen general sensibility than men, while Dehn (20) experimenting with a small number of individuals, and Ottolenghi (66) from returns on eight hundred women and six hundred men, agree that women have a keener general sensibility than men. Di Mattei (21) corroborates this result for children of from four to twelve years. Griffin (31) has shown that sensitiveness to electrical stimulation and sensitiveness to pressure stimulation do not necessarily vary together. It is therefore impossible to argue directly from keener general sensibility to more acute touch.

There is a much greater mass of material for the comparison of men and women with reference to sensitiveness to pain. Two methods of inducing pain have been employed; one by electrical stimulation and the other by pressure. Lombroso (50, 51), Ottolenghi (66), Di Mattei (21), and Dehn (20) used the former method. The three Italians, the first two working with adults, and the last with children of from four to twelve years, all find the female less sensitive to pain [p. 46] through electrical stimulation than the male, while Dehn finds women more sensitive than men. The former result is based on a far greater mass of results than the latter.

The experiments made by the second method, pain through pressure, all agree with the present series of tests in showing a lower threshold for women than for men. Wissler (82) and MacDonald (54) experimented on adults, Carman (16) on children, and Swift (78) on both adults and children. The relation holds for all ages.

There is but one comparative test on passive pressure, that made by Dehn (20). He used an error method instead of a gradation method, but his results agree with ours in showing no difference between the sexes in this respect. The ability to discriminate lifted weights was found by Gilbert (quoted by MacDonald, 55, p. 1107) to be greater in boys than in girls between the ages of thirteen and seventeen, a result which corroborates ours. Wolfe (83) in experimenting on the effects of size on judgments of lifted weights, finds women much more subject to illusion than men. Gilbert (30) finds the same difference between boys and girls. Both are inclined to explain the fact by the greater suggestibility of the female. If it is true, however, that the actual ability to discriminate lifted weights is less in women than in men, this may explain in part the fact that they are more subject to the size-weight illusion than men. Other forms of test on the perception of weight do not agree with the discriminative tests in showing less accuracy on the part of women. Wissler (82) [p. 47] reports a test in which the subject was required to lift against a spring to 1 kg. as a standard, and then attempt to lift the same amount several times from memory. He found no difference in the ability of men and women to do this. Jastrow (38) required his subjects to estimate a pound and an ounce in shot with no guide or standard. He found women more accurate than men. The results suggest the generalization that men excel women in the direct discrimination of lifted weights, but are equaled or excelled by women in tests where the memory of a given weight is involved; but no stress can be laid on such a statement until more data are available.

There are several comparative tests at hand on two-point discriminations. Galton (27), Dehn (20), Lombroso (51, chap. iii, 50) and the Columbia University tests (82) dealt with adults. Galton measured about 1,200 men and women on the nape of the neck, using a method like that of the present test. His results are in accord with ours in showing a finer discrimination on the part of the women. Dehn's test and those on Columbia students failed to show any difference of sex in this respect. The method used is probably a sufficient explanation for the fact in both cases. The æsthesiometer points were kept a fixed distance apart, and the right and wrong answers on a small number of stimulations were recorded. The results thus yielded are too meager to give any reliable measurement. Lombroso, experimenting on 100 normal men and 100 normal women, finds women less sensitive than men. His subjects were of varying ages and social conditions, but he states that the general relation holds also for men and women of the educated class. He [p. 48] does not describe his method. The fact that in several other respects his results are contradictory to those of other observers, makes one hesitate to lay much stress on this discrepancy. There are two æsthesiometer tests on children, that by MacDonald (55, p. 1005) made on the palmar surface of the wrist, and that by Di Mattei (21) made on the index finger. MacDonald's method, and presumably Di Mattei's, though he is less explicit, were analogous to ours. Both sets of results agree with ours in showing the female to have a finer discrimination of two points than the male.

There are no previous data known to the author on the comparative ability of the sexes to discriminate area on the skin. One would expect to find that the class with the finer two-point discrimination was also the one with the greater ability to discriminate area on the skin, but this does not hold in the present case. Since the discrimination of area is a complicated judgment involving several factors, it is by no means sure that it need be correlated with a two-point discrimination. If the two results are contradictory, doubt should be thrown on the test in the discrimination of area rather than on the æsthesiometer test, since the conditions of the former were much less satisfactory than those of the latter.

There are two tests on temperature discrimination in which a comparison of the sexes has been made; one by Dehn (20) on adults, and one by MacDonald (55, p. 1005) on school children. The method in both cases differed from that employed in the present tests. It consisted in stimulation of the skin by metal surfaces of known temperature. Dehn used successive stimu-[p. 49]lations and MacDonald simultaneous stimulations. Dehn's temperatures were near the physiological zero. MacDonald gives no standard, but one of the stimuli was certainly above the physiological zero, since the test is called a discrimination of heat. Dehn finds women more sensitive than men. MacDonald finds boys, on the whole, slightly more sensitive than girls. The present tests show no difference of sex. It seems safe to conclude that sexual differences in ability to discriminate temperatures are very insignificant, if they exist at all.


The general outcome of the experimentation on the sensations mediated by the skin is to show that women have somewhat keener senses than men. This statement does not hold for all forms of sensation. The greater sensibility of women is marked in the two-point discrimination, in general sensibility, and in sensitiveness to pain through pressure; and is slight in delicacy of touch. In passive-pressure discrimination and in temperature there is no difference. In pain through electrical stimulation, the discrimination of lifted weights, and possibly, in the discrimination of area on the skin, men are more sensitive.