Printer Friendly

Aggression, gender-typical childhood play, and a prenatal hormonal index.

Gender differences in a variety of behaviors--including aggression and style of play--have frequently been reported (reviewed by Hines, 2004) and several studies have related prenatal androgens to male-typical play and aggression in clinical populations (e.g., Berenbaum & Hines, 1992; Berenbaum & Resnick, 1997; Ehrhardt & Baker, 1974; Hines et al., 2003; Reinisch, 1981). A study of amniotic testosterone level and childhood play of 4-to 6-year-old children did not find a relationship between testosterone level and play, although they did find the expected overall sex differences in gender-typical play (Knickmeyer et al., 2005). Higher amniotic testosterone levels during gestation have been shown to be related to lower levels of empathy in young children (Chapman et al., 2006; Knickmeyer, Baron-Cohen, Raggatt, Taylor, & Hackett, 2006), lower quality of social relationships at age 4 (Knickmeyer, Baron-Cohen, Raggatt, & Taylor, 2005), lower amount of eye contact at 12 months (Lutchmaya, Baron-Cohen, & Raggatt, 2002a), and less developed vocabulary at 12 months (Lutchmaya, Baron-Cohen, & Raggatt, 2002b). These authors related their findings to the literature suggesting greater social cognition and language development in female than male groups.

Studies have reported that finger length ratios show consistent gender differences, with evidence suggesting that these finger length ratios are also related to androgenic effects, most likely during prenatal development (reviewed by Manning, 2002). The present study sought to evaluate whether or not finger length ratios are associated with measures of aggression and play.


Gender differences such that men and boys are more aggressive than women and girls, especially in terms of physical aggression, have frequently been reported (e.g., Eagly & Steffen, 1986; Feingold, 1994; Hyde, 1984; Reinisch & Sanders, 1986). In animals, there is strong evidence linking aggression to both organizational and activational effects of androgens (see Simon, 2002 for review). Organizational effects in humans were reported by Reinisch (1981) in terms of higher childhood physical aggression being associated with prenatal exposure to a synthetic progestin with androgenic effects. Berenbaum and Resnick (1997) evaluated several adult and child samples with congenital adrenal hyperplasia (CAH), as well as related controls. CAH is a syndrome marked by higher androgen levels, beginning prenatally. Berenbaum and Resnick reported that females with CAH (and not males with CAH) showed more aggression compared to same-sexed relatives on the Multidimensional Personality Questionnaire (Tellegen, 1982). Of the related control group, the male participants scored higher than the female participants. In the other adult and child samples reported in this study, the control groups did not show a pattern of higher male than female aggression on the Multidimensional Personality Questionnaire, but both the male control adults and children did show higher aggression than female control subjects on the Reinisch Aggression Inventory (Reinisch & Sanders, 1986). The CAH adolescent and adult female group also showed more aggression than the female control group on the Reinisch Aggression Inventory.


Many consistent gender differences in childhood play have been reported. Children typically prefer same-sex playmates (e.g., Hines & Kaufman, 1994; Maccoby, 1988), and have clear preferences for gender-typical toys such as dolls for girls and vehicles and guns for boys (e.g., Berenbaum & Hines, 1992; Snow, Jacklin, & Maccoby, 1983; Sutton-Smith, Rosenberg, & Morgan, 1963). Greater rough-and-tumble play, or play including wrestling and rough housing by boys compared to girls have also been widely documented (e.g., DiPietro, 1981; Hines & Kaufman, 1994; Maccoby, 1988). These gender differences in play appear as early as at 12 months (e.g., Snow et al.).

Several studies have shown that females with CAH show more male-typical play behavior than unaffected females do, suggesting organizational effects of higher than normal circulating prenatal androgens on play. For example, girls with CAH were observed to spend more time with male-typical toys such as vehicles than control girls did (Berenbaum & Hines, 1992), and questionnaires have also indicated masculinization of activity, playmate and toy preferences in CAH girls (Hines et al., 2003). Masculinization of play behavior in CAH girls has been reported in a variety of cultures including Japanese, Scandinavian, and North American samples (e.g., Ehrhardt & Baker, 1974; Iijima, Ariska, Minamoto, & Arai, 2001; Nordenstrom, Servin, Bohlin, Larsson, & Wedell, 2002; Slijper, 1984). Studies of play behavior in CAH boys show mixed results. In one study, using an interview, Slijper reported decreased male-typical behavior in CAH boys compared to unrelated males, but not male siblings and less roughandtumble play by CAH boys. However, similar toy and playmate preferences to control boys was reported in other studies (Berenbaum & Hines, 1992; Hines & Kaufman, 1994). As discussed, the study by Knickmeyer et al. (2005) showed the expected gender-typical differences in play behavior at age 4, but no correlation with amniotic testosterone level.

Finger Length Ratios

Finger length ratios have been shown to differ for male and female subjects from age two and older, such that a smaller ratio of the second to fourth finger is found in men compared to women (Burton, Henninger, & Hafetz, 2005; Ecker, 1875; George, 1930; Phelps, 1952; see Manning, 2002 for a review). Additional gender differences are found in the ratio of the second to fifth finger, and of the third to fourth finger, all in the same direction as the second to fourth finger, such that the ratio is larger for women than for men (McFadden & Shubel, 2002). Consistent sex differences in digit length ratios have also been found in mice, baboons, gorillas, and chimpanzees (Brown, Finn, & Breedlove, 2001; McFadden & Bracht, 2000a, b).

These gender differences in finger length ratios are thought to reflect effects of androgens. Homosexual men and women show finger ratio patterns in between that of heterosexual men and women (McFadden & Shubel, 2002). Further, female subjects with CAH show a finger length pattern in the direction of the male pattern (Brown, Hines, Fane, & Breedlove, 2002). These gender differences in finger length ratios are thought to reflect prenatal organizational effects (Manning, 2002), because hand patterns are set prenatally (see Manning, 2002; Garn, Burdi, Babler, & Stinson, 1975) and the ratio difference has been reported at such a young age. McIntyre, Ellison, Lieberman, Demerath, and Towne (2005) have also reported longitudinal data using radiographs indicating that sex differences arise before puberty and finger lengths in infancy and childhood are highly correlated with adult lengths.


The goal of the present study was to evaluate the relationships between finger length ratios and behaviors showing gender differences and direct relationships to androgenic effects. Current reports of aggression and reports of gender-typical play in childhood were obtained, and finger lengths were measured. Given the evidence that finger length ratios reflect androgenic effects, it was hypothesized that gender-typical finger length ratios would be associated with gender-typical patterns of play, and that more male-typical finger length ratios would be associated with higher aggression.



A sample of 134 university students was evaluated, after informed consent was obtained, in partial fulfillment of a research requirement. This sample included 93 female and 41 male participants (reflecting the ratio in classes), and has been described in Burton, Henninger, and Hafetz (2005) and in Burton, Hafetz, and Henninger (2007). The mean age of the male participants (19.3 years, SD of 1.1) did not differ from the mean age of the female participants (20.0 years, SD of 4.4).


Finger Length Ratios Photocopies were obtained from the students in the sample of both hands with fingers together and rings removed. The length in millimeters from the proximal crease at the base of the digit to the tip of the finger was measured with a ruler. Inter-rater reliabilities for two raters were calculated for a subset of 60 participants, and correlations between raters for each of the fingers ranged from .94 to .99. Given this very high reliability, subsequent calculations just used one of the raters. The ratio was calculated of the length of the second to fourth finger of the right (R24) and left (L24) hands, of the third to fourth finger of the right (R34) and left (L34) hands, and of the second to fifth finger of the right (R25) and left (L25) hands.

Gender-Typical Play Behavior Four questions were used to evaluate childhood play behavior, for which the subjects rated themselves on a 5-point Likert scale, as follows:

1. Compared to individuals of the same sex, as a child, did you engage in rough-and-tumble play (rough-housing)

much less (1) less (2) the same amount (3) more (4) much more (5)

2. Compared to individuals of the same sex, as a child, please rate the amount of injuries you had (e.g., stitches, etc.)

much less (1) less (2) the same amount (3) more (4) much more (5)

3. Compared to other children, did you play with children of the opposite sex

much less (1) less (2) the same amount (3) more (4) much more (5)

4. Compared to other children, how much did you play with toys typical of the opposite sex (i.e., girls playing with trucks, boys playing with dolls, etc.)?

much less (1) less (2) the same amount (3) more (4) much more (5)

In order to obtain an overall gender-typical play score, for the female participants, these numbers were simply added up. Scores could range from 4 to 20, with lower scores indicating greater or even exaggerated sex-typical behavior (and higher scores indicating less sex-typical behavior). For the overall play score for the male participants, the scores for items 1 and 2 were reversed (5 changed to 1, 4 changed to 2, 3 remained 3), and then added to items 3 and 4. Thus, the male overall score could also range from 4 to 20, and a lower male score would also reflect greater or exaggerated sex-typical behavior. It should be noted that the four questions asked subjects to evaluate their own behavior relative to same-sex peers. A score of 3 for any item, or 12 for the total, would indicate that the subject felt that his or her play behavior was the same as individuals of the same sex.

Physical Aggression Questionnaire The physical aggression questionnaire of Buss and Perry (1992) was used with slight modifications, including changing the Likert scale categories (never = 1, sometimes = 2, frequently = 3, very frequently = 4, always = 5), and changing the direction of rating to be consistently the same. The nine physical-aggression items queried being unable to control the urge to strike someone, hitting with provocation, hitting back, getting into fights more than average, resorting to violence to protect rights, coming to blows if pushed too far, having good reasons for hitting, threatening others, and breaking things in anger. An aggression total score was calculated by adding the item values. Scores could range from 9 to 45, with high scores indicating high aggression.



The mean male and female scores for the Physical Aggression and Gender-Typical Play Behavior Questionnaires, and finger length ratios were evaluated with t tests (shown in Table 1). For the Physical Aggression Questionnaire, the male subjects reported more total aggression than the female subjects. For gender-typical play behavior, there was no reason to expect any sex differences, as subjects were evaluating their behavior relative to same sex-peers (it should be noted that although questions 1 and 2 were reversed to calculate the overall play score for the male subjects, the tables and subsequent analyses reflect their unreversed scores). However, the male subjects reported playing with toys typical of girls less than the female subjects reported playing with toys typical of boys (question 4) and the overall play score was higher for the male subjects than the female subjects. It should be noted that the overall play score reflects gender stereotypy in play, such that lower scores reflect greater gender stereotypy for both sexes.

All of the gender differences in finger length ratios were in the direction reported in the literature, such that the ratios of the second to fourth and second to fifth digit were greater for female than male subjects. The ratios of the third to fifth fingers were not significant, consistent with the studies showing that this difference is found less consistently than the other gender differences in ratios (McFadden & Shubel, 2002), and these ratios were eliminated from further analyses.


Pearson correlations were performed between finger length ratios and the Physical Aggression and Gender-Typical Play Behavior scores, and interpreted with two-tailed probability values (Table 2).

Finger length ratios and aggression Lower finger ratios, the male-typical pattern, were associated with greater physical aggression for the combined male and female sample for R25 (r = -.25; p = .025) and L25 (r = -.22; .017). For the male participants alone, this same pattern was seen for R24 and R25 (see Table 2). These relationships were not significant for the female participants.

Finger length ratios and play In terms of reports of childhood play and finger length ratios, for the female subjects, all of the correlations were in the expected direction. The higher, female-typical, finger ratios were associated with less rough-and-tumble play, less play with toys typical of boys, less playing with boys, and a lower and thus more gender-typical overall play score. For the male subjects, the lower, male-typical finger ratio was associated with less play with toys typical of girls. The correlations between finger length ratios and play for the combined male and female subjects were not done because the play evaluations were comparisons to same-sex peers; the opposite direction of the male and female correlations would cancel effects if combined and would be meaningless.

Aggression and play The overall aggression and play scores were not correlated for the male or female participants. However, more reported rough-and-tumble play as a child was associated with higher current aggression overall score for the male participants (r = .43; p = .005), with a similar pattern at trend level for the female participants (r = .19; p = .069). As stated above, the correlations between aggression and play for the combined male and female subjects were not done because the play evaluations were comparisons to same-sex peers; the opposite direction of the male and female correlations would cancel effects if combined and would be meaningless.


Numerous gender differences were found, consistent with other reports in the literature. Men reported more current aggressive behavior than the women, and showed smaller ratios than women of the second to fourth finger and second to fifth finger, for both the right and left hands--also consistent with the literature.

Finger length ratios were related to current aggression and childhood play. For the male participants, lower finger length ratios, the male-typical pattern, were associated with a greater number of reports of current aggression and report of less childhood play with toys typical of girls. For the female participants, the higher, female-typical, finger ratio was associated with less childhood rough-and-tumble play, less play with toys typical of boys, less playing with boys, and more gender-typical overall play.

The relationships between current aggression and reports of childhood play were also quite interesting. For the male participants, more frequently reported rough-and-tumble play as a child was associated with higher current aggression overall. For the female participants, a similar pattern of a higher incidence of current aggression and more reported rough-and-tumble childhood play was seen at a trend level. The relationships between reports of current aggressive behavior and childhood play behavior suggest a certain stability over time. Individuals, both male and female, who played more roughly as a child are now using rougher and more physical means to deal with others. Rough-and-tumble play can be thought of as a mild, childhood form of physical aggression, and it certainly seems that individuals who engaged in this more than others as a child are still using physical aggression more as an adult. Of relevance is a study by McIntyre et al. (2007) which found that finger length ratios were associated with aggression in video game play, such that a more male-typical pattern predicted an unprovoked attack in a war game.

This stability over time adds support to the idea that these patterns of aggressive behavior are developed quite early, perhaps reflecting, to some extent, organizational effects of hormones. The correlations between current aggression, childhood gender-typical play, and finger ratios also support the idea that aggression and childhood play styles develop early, since the finger ratios are thought to be determined prenatally and to be influenced by androgens.

Further research might utilize direct observation of behavior or reports of informants such as parents or teachers in younger samples to replicate these findings. Finger length ratios may be very useful, nonintrusive indices of organizational androgenic effects, and further studies might evaluate the relationship of these ratios to other androgen indices as well as to other variables known to be influenced by androgens.


Berenbaum, S., & Hines, M. (1992). Early androgens are related to childhood sex-typed toy preferences. Psychological Science, 3, 203-206.

Berenbaum, S., & Resnick, S. (1997). Early androgen effects on aggression in children and adults with congenital adrenal hyperplasia. Psychoneuroendocrinology, 22, 505-515.

Brown, W., Finn, C., & Breedlove, S. (2001). A sex difference in digit length ratio in mice. Hormones and Behavior, 39, 325.

Brown, W., Hines, M., Fane, B., & Breedlove, M. (2002). Masculinized finger length patterns in human males and females with congenital adrenal hyperplasia. Hormones and Behavior, 42, 380-386.

Burton, L., Hafetz, J., & Henninger, D. (2007). Gender differences in relational and physical aggression. Social Behavior and Personality: An international journal, 35(1), 41-50.

Burton, L., Henninger, D., & Hafetz, J. (2005). Gender differences in relationships of mental rotation, verbal fluency, and SAT scores to finger length ratios as hormonal indices. Developmental Neuropsychology, 28(1), 493-505.

Buss, D., & Perry, M. (1992). The Aggression Questionnaire. Journal of Personality and Social Psychology, 63(3), 452-459.

Chapman, E., Baron-Cohen, S., Auyeung, B., Knickmeyer, R., Taylor, K., & Hackett, G. (2006). Fetal testosterone and empathy: Evidence from the Empathy Quotient (EQ) and the "Reading the Mind in the Eyes" Test. Social Neuroscience, 1(2), 135-148.

DiPietro, J. (1981). Rough and tumble play: A function of gender. Developmental Psychology, 17, 50-58.

Eagly, A., & Steffen, V. (1986). Gender and aggressive behavior: A meta-analytic review of the social psychological literature. Psychological Bulletin, 100, 309-330.

Ecker, A. (1875). Some remarks about varying character in the hands of humans. Archives of Anthropology, 8, 68-74.

Ehrhardt, A., & Baker, S. (1974). Fetal androgens, human central nervous system differentiation, and behavior sex differences. In R. Friedman, R. Richart, & R. Van Der Wiele (Eds.), Sex differences in behavior (pp 33-51). New York: Wiley.

Feingold, A. (1994). Gender differences in personality: A meta-analysis. Psychological Bulletin, 116, 429-456.

Garn, S., Burdi, A., Babler, W., & Stinson, S. (1975). Early prenatal attainment of adult metacarpalphalangeal rankings and proportions. American Journal of Physical Anthropology, 43, 327-332.

George, R. (1930). Human finger types. Anatomy Records, 46, 199-204.

Hines, M. (2004). Brain gender. New York: Oxford University Press.

Hines, M., Fane, B., Pasterski, V., Matthews, G., Conway, G., & Brook, C. (2003). Spatial abilities following prenatal androgen abnormality: Targeting and mental rotation performance in individuals with congenital adrenal hyperplasia. Psychoneuroendocrinology, 8, 1010-1026.

Hines, M., & Kaufman, F. (1994). Androgen and the development of human sex-typical behavior: Rough and tumble play and sex of preferred playmates in children with Congenital Adrenal Hyperplasia (CAH). Child Development, 65, 1042-1053.

Hyde, J. (1984). How large are gender differences in aggression? A developmental meta-analysis. Developmental Psychology, 20, 722-736.

Iijima, M., Ariska, O., Minamoto, F., & Arai, Y. (2001). Sex differences in children's free drawings: A study on girls with Congenital Adrenal Hyperplasia. Hormones and Behavior, 40, 99-104.

Knickmeyer, R., Baron-Cohen, S., Raggatt, P., & Taylor, K. (2005). Fetal testosterone, social relationships, and restricted interests in children. Journal of Child Psychology and Psychiatry, 46, 198-210.

Knickmeyer, R., Baron-Cohen, S., Raggatt, P., Taylor, K., & Hackett, G. (2006). Fetal testosterone and empathy. Hormones and Behavior, 49, 282-292.

Knickmeyer, R., Wheelwright, S., Taylor, K., Raggatt, P., Hackett, G., & Baron-Cohen, S. (2005). Gender-typed play and amniotic testosterone. Developmental Psychology, 41(3), 517-528.

Lutchmaya, S., Baron-Cohen, S., & Raggatt, P. (2002a). Fetal testosterone and eye-contact in 12 month old infants. Infant Behavior and Development, 25, 327-335.

Lutchmaya, S., Baron-Cohen, S., & Raggatt, P. (2002b). Fetal testosterone and vocabulary size in 18 and 24 month old infants. Infant Behavior and Development, 24, 418-424.

Maccoby, E. (1988). Gender as a social category. Developmental Psychology, 24, 755-765.

Manning, J. (2002). Digit ratio: A pointer to fertility, behavior, and health. New Brunswick, NJ: Rutgers University Press.

McFadden, D., & Bracht, M. (2002a). The relative lengths and weights of metacarpals and metatarsals in baboons. Hormones and Behavior, 42(2), 347-355.

McFadden, D., & Bracht, M. (2002b). Sex differences in length ratios from the extremities of humans, gorillas, and chimpanzees. Hormones and Behavior, 41, 479.

McFadden, D., & Shubel, E. (2002). Relative lengths of fingers and toes in human males and females. Hormones and Behavior, 42, 492-500.

McIntyre, M., Barrett, E., McDermott, R., Johnson, D., Cowden, J., & Rosen, P. (2007). Finger length ratio and sex differences in aggression during a simulated war game. Personality and Individual Differences, 42(4), 755-764.

McIntyre, M., Ellison, P., Lieberman, D., Demerath, E., & Towne, B. (2005). The development of sex differences in digital formula from infancy in the Fels Longitudinal Study. Proceedings: Royal Society of London, Biological Sciences, 272(1571), 1473-1479.

Nordenstrom, A., Servin, A., Bohlin, G., Larsson, A., & Wedell, A. (2002). Sex-typed play behavior correlates with the degree of prenatal androgen exposure as assessed by CYP21 genotypes in girls with Congenital Adrenal Hyperplasia. Journal of Clinical Endocrinology and Metabolism, 87, 5119-5124.

Phelps, V. (1952). Relative index finger length as a sex-influenced trait in man. American Journal of Human Genetics, 4, 72-89.

Reinisch, J. (1981). Prenatal exposure to synthetic progestins increases potential for aggression in humans. Science, 211, 1171-1173.

Reinisch, J., & Sanders, S. (1986). A test of sex differences in aggressive response to hypothetical conflict situations. Journal of Personality and Social Psychology, 50, 1045-1049.

Simon, N. (2002). Hormonal processes in the development and expression of aggressive behavior. In D. Pfaff, A. Arnold, A. Etgen, S. Fahrbach, & R. Rubin (Eds.), Hormones, brain and behavior (pp. 649-798). San Diego, CA: Academic Press.

Slijper, F. (1984). Androgens and gender role behavior in girls with congenital adrenal hyperplasia. In J. De Vries, J. De Bruin, H. Uylings, & M. Corner (Eds.), Progress in brain research (pp. 417422). Amsterdam: Elsevier.

Snow, M., Jacklin, C., & Maccoby, E. (1983). Sex of child differences in father-child interaction at one year of age. Child Development, 54, 227-232.

Sutton-Smith, B., Rosenberg, B., & Morgan, E. (1963). Development of sex differences in play choices during pre-adolescence. Child Development, 34, 119-126.

Tellegen, A. (1982). Brief manual for the Multidimensional Personality Questionnaire. Unpublished manuscript, University of Minneapolis, Minnesota, USA.

Leslie A. Burton, PhD, Department of Psychology, University of Connecticut, CT, USA and Weill Medical College of Cornell University, New York, NY, USA; Debra Henninger, Jessica Hafetz, and John Cofer, Fordham University, Bronx, NY, USA.

Appreciation is due to reviewers including: Sari van Anders, PhD, Department of Psychology, Simon Fraser University, RCB 5246, 8888 University Drive, Burnaby, BC V5A 1S6, Canada, Email: saria@

Please address correspondence and reprint requests to: Leslie Burton, PhD, Psychology Department, University of Connecticut, One University Place, Stamford, CT 06901, USA. Email: Leslie.

Leslie A. Burton University of Connecticut, CT, USA and Weill Medical College of Cornell University, NY, USA Debra Henninger, Jessica Hafetz, And John Cofer Fordham University, NY, USA
Table 1


                                Male     Female        t         p

Physical aggression Total      17.59     13.34        5.80     <.001
                               (4.65)    (3.50)

Finger Length Ratios
R24                             .957      .971       -1.86      .067
                               (.037)    (.045)
R25                            1.163     1.195       -2.45      .016
                               (.066)    (.073)
L24                             .949      .977       -3.83     <.001
                               (.036)    (.043)
L25                            1.175     1.211       -2.66      .010
                               (.070)    (.067)
Gender-typical Play Total      12.02     10.79        2.82      .006
                               (1.82)    (3.16)
P1 (rough-and-tumble play)      2.66      2.67
                                (.85)    (1.17)
P2 (injuries)                   2.37      2.16
                               (1.16)    (1.07)
P3 (opposite-gender friends)    3.00      3.24
                                (.87)    (1.03)
P4 (opposite-gender toys)       2.05      2.67       -3.19      .002
                               (1.05)    (1.01)

Table 2

                                       Male participants

                                  R24        R25        L24        L25
Aggression Total score           -.34       -.32
                                 (.031)     (.043)
Gender-Typical Play Total

P1 (rough-and-tumble play)

P2 (injuries)
P3 (opposite-gender friends)

P4 (opposite-gender toys)                               .38

                                      Female Participants

                                  R24        R25      L24        L25
Aggression Total score

Gender-Typical Play Total                    -.28
P1 (rough-and-tumble play)                   -.26
P2 (injuries)
P3 (opposite-gender friends)                 -.28
P4 (opposite-gender toys)         .33        -.25
                                 (.003)     (.027)
COPYRIGHT 2009 Scientific Journal Publishers, Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Burton, Leslie A.
Publication:Social Behavior and Personality: An International Journal
Article Type:Report
Geographic Code:1USA
Date:Feb 1, 2009
Previous Article:A prediction of self-esteem and life satisfaction by social problem solving.
Next Article:An empirical investigation of the moral judgment development of Taiwanese procurement executives.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters