Preference for Darker Faces in Photographs at Different Phases of the Menstrual Cycle: Preliminary Assessment of Evidence for a Hormonal Relationship

PETER FROST

Département d'anthropologie

Université Laval

Summary. —Six pairs of photographs showing human faces of both sexes were presented to 98 women who had to choose the more pleasing one of each pair. Faces within each pair were identical except for a slight difference in complexion. For women not taking oral contraceptives, skin color preference differed significantly between two groups of subjects classified according to the current phase of their menstrual cycle: darker male faces were judged more positively by subjects in the phase when the estrogen/progesterone ratio was expected to be high than by those in the phase when this ratio was expected to be low. Female faces evoked no such cyclic response. Users of oral contraceptives showed no cyclic response to either male or female faces. These results suggest a mental mechanism whose inputs are (a) hormonal state, (b) visual identification of the sex of the face being observed, and (c) visual recognition of complexion, and whose output enters into evaluation of male and female faces. Replication with direct measures of hormonal state is recommended.

Frost, P. 1994b. "Preference for darker faces in photographs at different phases of the menstrual cycle: Preliminary assessment of evidence for a hormonal relationship", Perceptual and Motor Skills 79: 507-514.

Men have browner and ruddier complexions than do women, a result of differing melanin and hemoglobin levels in the skin's outer layers (Edwards & Duntley, 1939; Harrison, 1973; van den Berghe & Frost, 1986). This sex difference is universal and arises at puberty, as do analogous differences in skin texture, body hair, fat distribution, and other cutaneous characteristics (Frost, 1988; Kalla, 1973; Mesa, 1983).

Variation in skin color has been thought of primarily in terms of race ever since the postmedieval European expansion and the creation of multiracial colonial societies (Jordan, 1968). Previously, however, skin color was perceived in a more homogeneous milieu. Our expression "the fair sex" harks back to a time when people thought of skin color in terms of sexual identity. This premodern perception seems to have been universal: a cross-cultural survey by van den Berghe and Frost (1986) of the Human Relations Area Files shows that traditional societies viewed light complexions as more feminine and darker complexions as more masculine. Similarly, premodern art from a wide range of culture areas — Greco-Roman, Egyptian, East Asian, Aztec — invariably depicts women with paler hues and men with browner, redder ones (Capart, 1905, pp. 26-27; Pallottino, 1985, pp. 34, 45, 73, 76-77, 87, 93, 95, 105, 107, 115; Phan & Flandrin, 1984, p. 50; Soustelle, 1970, p. 130; Tegner, 1992; Thompson, 1989, p. 131; Wagatsuma, 1967, pp. 412, 414).

This sex difference apparently had meaning for people in premodern societies. It may, therefore, have structured how humans evaluate skin color, at least for minor variations in complexion (Guthrie, 1970, pp. 279-280). One outcome of this theoretical model would be that men and women should evaluate skin color differently. In our ancestral environment, an individual with a dark, male coloration would have meant one thing to a man and another to a woman. The man might have seen a potential rival; the woman, a current or future mate.

Some data point to such a behavioral sex difference. Feinman and Gill (1978) asked 549 male and 482 female white college students to indicate the complexions they liked and disliked in the opposite sex; see Table 1. Analysis showed significantly more preference by men for lighter female skin color and by women for darker male skin color. The contrast was more striking when subjects stated their dislikes.While 30% of the men disliked black complexion, only 10% of the women did so. Similarly, 82% of the women disliked the two lightest shades, while the corresponding proportion for men was 56%.

TABLE 1

Likes and Dislikes by Men and Women¹ regarding Complexion of the Opposite Sex²

The results of that study are consistent with anecdotal evidence, e.g., the "tall, dark, handsome man" of Hollywood movies. In Japan, according to Wagatsuma (1967, pp. 417-419), the male esteems a fair complexion as "a symbol of women, distinguishing them from men." Conversely, the Japanese female displays a certain ambivalence to male skin color and often distinguishes "a beautiful man" from " an attractive man". The former has white skin and, although admired almost aesthetically, is seen as effeminate. The latter has brown skin and is considered virile, dauntless, energetic, and self-assertive.

This pattern of likes and dislikes may begin as early as childhood. Asher and Allen (1969) asked 186 black and 155 white children from 3 to 8 years in age to choose between two puppets — one white in facial color and the other medium brown. Significantly more girls than boys chose the darker one; see Table 2.

TABLE 2

Dolls Chosen by Sex, by Race, and by Type of Question¹

Questions: (a) Which puppet is the nice puppet? (b) Which puppet would you like to play with? (c) Which puppet looks bad? (d) Which puppet is the nice color?

¹From Asher and Allen, 1969.

A study using dolls yielded similar results from South African Bantu children three to seven years of age (Gregor & McPherson, 1966). No such sex difference has emerged in doll studies limited to children six years of age and younger (Renninger & Williams, 1966; Williams & Roberson, 1967; Williams & Rousseau, 1971).

It may be that male and female evaluation of skin color diverges during late childhood under the influence of diverging hormonal input, i.e., the sex difference in production of androgens and estrogens from seven years of age onwards. Frost and Lamontagne (1988) asked a sample of 225 preschool children to choose the nicer of two dolls, one of which was slightly darker than the other in complexion. Although no relation was found between the subjects' sex and the chosen doll's complexion, the children choosing the darker-colored doll had significantly more body fat (cf. also Frost 1989). This choice may have been influenced by a higher estrogen/androgen ratio: fatty tissue contains an enzyme, aromatase, that converts an androgen to estrogen; when the gonads are relatively inactive, as is the case below 7 years of age, estrogen blood levels correlate with the body's fat mass (Judd, Davidson, Frumar, Shamonki, Lagasse & Ballon, 1980; Kirschner, Schneider, Ertel & Worton, 1982; Siiteri & MacDonald, 1973; Vermeulen & Verdonck, 1978, 1979). This source of estrogen is not negligible; in prepubertal boys, excessive aromatase activity can even induce breast development and other signs of feminization (Hemsell, Edman, Marks, Siiteri & MacDonald, 1977).

The present study was undertaken to see whether this apparent hormonal effect may be observed in adult subjects. Instead of hormonal variance due to fatty tissue, the present study relied on phases of the menstrual cycle which generally correspond to hormonal fluctuations: estrogen levels rise from Day 0 to Day 14, fall gradually with a secondary maximum around day 20 and then decline markedly; progesterone levels increase from Day 14 to Day 22 and remain high until Day 27 (Winter, 1978).

Progesterone acts as an antiestrogen by reducing the number of estrogen receptors at the cellular level ( Klaiber, Broverman, Vogel, Kennedy & Nadeau, 1982, p. 269; Linkie, 1982, p. 19; Sanders & Reinisch, 1985, p. 184). Skin color preference, if influenced by estrogen, should vary with the ratio of estrogen to progesterone (E/P). Preference for darker complexions should be stronger during the first two-thirds of the menstrual cycle when the E/P ratio is high and weaker during the last third when E/P is low. Because oral contraceptive users have consistently high E/P throughout the cycle (Wilson & Foster, 1985 , p. 455), their skin color preferences should resemble that of nonusers in the high E/P phase.

METHOD

Subjects

A total of 98 women were tested during the 1990-1991 fall and winter terms at Université Laval. They were recruited from first- and second-year undergraduate courses in the health sciences and from a university-based choir. Almost all were French-Canadians from 19 to 23 years of age. Their reported menstrual cycles fell within the normal range with few abnormally short or long cycles; see Table 3.

TABLE 3

Self-Reported Length of Menstrual Cycle of Subjects

Materials

Six pairs of photographs depicting human faces were prepared: three male faces and three female ones. All individuals depicted in the photographs were French-Canadian. Although each pair was developed from the same negative, one photograph was left in the developing solution for a standardized longer period of time than the other. As a result, one of the faces appeared somewhat darker in complexion than the other. The pairs of photographs were then mounted on cardboard, numbered from 1 to 6, and placed in a box in this order.

Procedures

The women were tested individually. The subject was asked to choose the more pleasing photograph of each pair (laquelle des deux vous plaît le mieux). She was then left alone to write down her choices as well as the number of days since the first day of her last menstruation, the average length of her menstrual cycle, and whether she was currently taking oral contraceptives or other hormone-based medication.

Analysis

After the test had been completed, information on the questionnaire was used to estimate the phase of the subject's menstrual cycle and her skin color preference. Phase of menstrual cycle was calculated by dividing number of days since last menstruation by average length of menstrual cycle. Skin color preference was rated for male and female faces separately on a scale of 0 to 3, where 0 = no darker faces chosen and 3 = all three darker faces chosen. The data were then divided into two groups: subjects not taking hormone-based medication and subjects taking such medication (comments written on the questionnaire or made verbally indicate that the latter group was on oral contraceptives). Within each group, mean skin color preferences were calculated for the first two-thirds of the menstrual cycle and the last third. Then t tests were used to assess significance of differences between means for subgroups representing different phases of the menstrual cycle.

RESULTS AND DISCUSSION

When judging male faces, the 56 nonusers of oral contraceptives differed in response according to the current phase of their menstrual cycle. The darker male face was preferred by a significantly higher percentage of subjects in the high E/P phase (M=.9; t= 1.91, p<.05) than in the low E/P phase (M=.4); see Tables 4 and 5. The difference was only relative; in neither phase was the darker male face more strongly preferred than the lighter one. When judging female faces, the nonusers of oral contracpetives showed no differences in response by phase of menstrual cycle.

TABLE 4

Mean Scores for Darker Face Preference in Photographs:

Differences between Subjects by Phase of Menstrual Cycle¹

^¹Darker face preference ranges from 0 to 3, where 0 = no darker faces chosen and where 3 = all three darker faces chosen.

^²One-tailed.

 

TABLE 5

Mean Scores for Darker Face Preference in Photographs:

Evaluation of Male Faces by Nonusers of Oral Contraceptives¹

¹Darker face preference ranges from 0 to 3, where 0 = no darker faces chosen and where 3 = all three darker faces chosen.

For both male and female faces, the 42 users of oral contraceptives showed no difference in response by phase of menstrual cycle; see Table 4. Consequently, these users in the last third of their cycle displayed significantly more preference for the darker male complexion than did non-users (M = 1.3; t = 2.91, p <.05); see Table 6.

Self-reported cycle phase, actual cycle phase, and actual hormonal state cannot be assumed to be equivalent categories because of a number of factors: errors in subject's evaluation of cycle length, variation in cycle length, absence of ovulation, etc. It is worth noting, however, that such factors would have introduced noise into the data without producing the results observed in this study. In the future, these results should be replicated by using blood or saliva tests to measure the subjects' hormonal state directly. Failing this, a longitudinal study could be conducted to see whether individuals change in skin color preference over the menstrual cycle.

The present results indicate that women at different reported phases of their menstrual cycle differ in skin color preference, suggesting that a hormonal mechanism may be involved. It should be noted that aversion for the darker male was equal in intensity to that for the darker female during the high E/P phase of the cycle. In the low E/P phase, aversion for the darker male was much higher. This suggests that estrogen may inhibit a mechanism that promotes a general distancing from a recognizably male image. Behavioral antiestrogens, such as progesterone, would allow this mechanism to function by releasing the estrogenic inhibition. Alternatively, antiestrogens may stimulate attraction to lighter skin in the opposite sex; in this second scenario, estrogen would block attraction by competing for the same receptor sites.

Finally, the results of this study may support the view that mental processing of skin color images has been structured to reflect conditions that once prevailed in our ancestral environment, i.e., when sex and age differences in complexion accounted for most observable variation in skin color. These results are also consistent with an emerging view that human evaluation of facial features is to some extent influenced by "hardwired" mental structures (Perrett, May & Yoshikawa, 1994).

REFERENCES

ASHER, S.R., & ALLEN, V.L. (1969) Racial preference and social comparison processes. Journal of Social Issues, 25, 157-166.

CAPART, J. (1905) Primitive art in Egypt. London: H. Grevel.

EDWARDS, E.A., & DUNTLEY, S.Q. (1939) The pigments and color of living human skin. American Journal of Anatomy, 65, 1-33.

FEINMAN, S., & GILL, G.W. (1978) Sex differences in physical attractiveness preferences. Journal of Social Psychology, 105, 43-52.

FROST, P. (1988) Human skin color: a possible relationship between its sexual dimorphism and its social perception. Perspectives in Biology and Medicine, 32, 38-58.

FROST, P. (1989) Human skin color: the sexual differentiation of its social perception. Mankind Quarterly, 30, 3-16.

FROST, P. (1990) Fair women, dark men: the forgotten roots of colour prejudice. History of European Ideas, 12, 669-679.

FROST, P. (1992) A fleur de peau: gender and complexion on Ile-aux-Coudres. Canadian Psychology, 33, 363.

FROST, P., & LAMONTAGNE, J. (1988) Une relation entre l'adiposité et les attitudes envers la couleur de la peau: témoin des influences hormonales? Annales de l'Association canadienne-française pour l'avancement des sciences, 56, 79.

GREGOR, A.J., & MCPHERSON, D.A. (1966) Racial preference and ego-identity among white and bantu children in the Republic of South Africa. Genetic Psychology Monographs, 73, 217-253.

GUTHRIE, R.D. (1970) Evolution of human threat display organs. Evolutionary Biology, 4, 257-301.

HARRISON, G.A. (1973) Differences in human pigmentation: measurement, geographic variation and causes. Journal of Investigative Dermatology, 60, 418-426.

HEMSELL, D.L., EDMAN, C.D., MARKS, J.F., SIITERI, P.K., & MACDONALD, P.C. (1977) Massive extraglandular aromatization of plasma androstenedione resulting in feminization of a prepubertal boy. Journal of Clinical Investigation, 60, 455-464.

JORDAN, W. (1968) White over black: American attitudes toward the negro 1550-1812. Williamsburg, N.C.: Univer. of North Carolina Press.

JUDD, H.L., DAVIDSON, B.J., FRUMAR, A.M., SHAMONKI, I.M., LAGASSE, L.D., & BALLON, S.C. (1980) Serum androgens and estrogens in postmenopausal women with and without endometrial cancer. American Journal of Obstetrics and Gynecology, 136, 859-869.

KALLA, A.K. (1973) Ageing and sex differences in human pigmentation. Zeitschrift für Morphologie und Anthropologie, 65, 29-33.

KIRSCHNER, M.A., SCHNEIDER, G., ERTEL, N.H., & WORTON, E. (1982) Obesity, androgens, estrogens, and cancer risk, Cancer Research (suppl.), 42, 3281-3285.

KLAIBER, E.L., BROVERMAN, D.M., VOGEL, W., KENNEDY, J.A., & NADEAU, C.J.L. (1982) Estrogens and central nervous system function: electroencephalography, cognition, and depression. In R.C. Friedman (Ed.), Behavior and the menstrual cycle. New York: Marcel Dekker. Pp. 267-290.

LINKIE, D.M. (1982) The physiology of the menstrual cycle. In R.C. Friedman (Ed.), Behavior and the menstrual cycle. New York: Marcel Dekker. Pp. 1-22.

MESA, M.S. (1983) Analyse de la variabilité de la pigmentation de la peau durant la croissance. Bulletin et mémoires de la société d'anthropologie de Paris, 13(10), 49-60.

PALLOTTINO, M. (1985) La peinture étrusque. Paris: Flammarion.

PERRETT, D.I., MAY, K.A., & YOSHIKAWA, S. (1994) Facial shape and judgements of female attractiveness. Nature, 368(6468), 239-242.

PHAN, M.C., & FLANDRIN, J.L. (1984) Les métamorphoses de la beauté féminine. L'Histoire, 68, 48-57.

RENNINGER, C.A., & WILLIAMS, J.E. (1966) Black-white color connotations and racial awareness in preschool children. Perceptual and Motor Skills, 22, 771-785.

SANDERS, S.A., & REINISCH, J.M. (1985) Behavioral effects on humans of progesterone-related compounds during development and in the adult. Current Topics in Neuroendocrinology, 5, 175-205.

SIITERI, P.K., & MacDONALD, P.C. (1973) Role of extraglandular estrogen in human endocrinology. In S.R. Geiger (Ed.), Handbook of physiology. Washington DC: American Physiology Society. Sect. 7, vol. 2, part 1, pp. 615-629.

SOUSTELLE, J. (1970) The daily life of the Aztecs. Stanford: Stanford Univer. Press.

TEGNER, E. (1992) Sex differences in skin pigmentation illustrated in art. American Journal of Dermatopathology, 14, 283-287.

THOMPSON, L.A. (1989) Romans and blacks. London: Univer. of Oklahoma Press.

VAN DEN BERGHE, P.L., & FROST, P. (1986) Skin color preference, sexual dimorphism, and sexual selection: a case of gene-culture co-evolution? Ethnic and Racial Studies, 9, 87-113.

VERMEULEN, A., & VERDONCK, L. (1978) Sex hormone concentrations in post-menopausal women. Clinical Endocrinology, 9, 59-66.

VERMEULEN, A., & VERDONCK, L. (1979) Factors affecting sex hormone levels in postmenopausal women. Journal of Steroid Biochemistry, 11, 899-904.

WAGATSUMA, H. (1967) The social perception of skin color in Japan. Daedalus, 96, 407-443.

WILLIAMS, J.E., & ROBERSON, J.K. (1967) A method for assessing racial attitudes in preschool children. Educational and Psychological Measurement, 27, 671-689.

WILLIAMS, J.E., & ROUSSEAU, C.A. (1971) Evaluation and identification responses of negro preschoolers to the colors black and white. Perceptual and Motor Skills, 33, 587-599.

WILSON, J.D., & FOSTER, D.W. (1985) Williams' textbook of endocrinology. Philadelphia: Saunders.

WINTER, J.S.D. (1978) Prepubertal and pubertal endocrinology. In F. Falkner & J.M. Tanner (Ed.), Human growth. Vol. 2. New York: Plenum. Pp. 183-213.