Поиск по сайту




Пишите нам: info@ethology.ru

Follow etholog on Twitter

Система Orphus

Новости
Библиотека
Видео
Разное
Кросс-культурный метод
Старые форумы
Рекомендуем
Не в тему

список статей


Sex hormones and finger length: What does 2D:4D indicate?

David A. Putza, Steven J.C. Gaulinb, Robert J. Sporterc, Donald H. McBurneyc

1. Introduction

1.1. The present research

2. Methods

2.1. Study 1

2.1.1. Voice pitch

2.1.2. Social and physical dominance

2.1.3. Sociosexuality and sexual interest

2.1.4. Mating success

2.1.4.1. Number of potential conceptions: males

2.1.4.2. Number of sex partners

2.1.5. Digit lengths and fluctuating asymmetry

2.2. Study 2

2.2.1. Mental rotation ability

2.2.2. Mating success and sexual orientation

2.2.3. Number of potential conceptions: females

2.3. Statistical analysis

3. Results

3.1. Sex differences

3.2. Correlations with 2D:4D among females

3.3. Correlations with 2D:4D among males

3.4. Factor analysis

4. Discussion

References

Copyright

1. Introduction

The ratio of the lengths of the second and fourth manual digits (2D:4D) reliably differs by sex, with males having a lower 2D:4D (Manning, Scutt, Wilson, & Lewis-Jones, 1998). This sex difference exists in 2-year-old children (Manning et al., 1998) and may be established prenatally by the 13th or 14th week postconception Garn et al., 1975, Phelps, 1952. The same sex difference has also been reported for at least one other mammal (Brown, Finn, & Breedlove, 2002) and one bird (Burley & Foster, 2004). Because of its early emergence, sexual dimorphism in 2D:4D is thought to be influenced by, and hence to reflect, prenatal sex hormone regimes. In particular, high prenatal androgens, low prenatal estrogens, or both, may produce a low (masculine) 2D:4D (Manning et al., 1998). These assumptions are partially supported by studies of males and females with congenital adrenal hyperplasia (CAH), a condition characterized by the overproduction of prenatal androgens. Brown, Hines, Fane, and Breedlove (2002) and Okten, Kalyoncu, and Yaris (2002) offer evidence that CAH is associated with lower (more masculine) 2D:4D in both sexes, although the effect was not consistently observed in both hands. However, Buck, Williams, Hughes, and Acerini (2003) examined 2D:4D using radiographs of the left hand of control boys and girls, and girls with CAH (approximately 70 subjects per group). While they found the usual sex difference, they found no difference in 2D:4D between control girls and those with CAH.

For any trait affected by androgens, development depends not only on hormone levels but also on the activity of the androgen receptor (AR) molecule. In humans and other primates there is considerable allelic diversity in the AR gene, particularly as regards the number of CAG repeats in the N-terminal domain. Sensitivity to testosterone appears to be inversely proportional to the number of CAG repeats Chamberlain et al., 1994, Kazemi-Esfarjani et al., 1995. Building on this datum, Manning, Bundred, Newton, and Flanagan (2003) demonstrated that in a sample of 50 men, right 2D:4D was positively correlated with the number of CAG repeats: Relatively androgen-insensitive men had higher (less masculine) right 2D:4D.

Jamison, Meier, and Campbell (1993) suggested that prenatal sex hormone levels correlate with adult levels, and Manning et al. (1998) were able to demonstrate that right 2D:4D correlates negatively with adult testosterone levels in men. Lumping both sexes of adults, positive correlations were also observed between 2D:4D and estrogen levels (right hand), prolactin levels (right hand), and luteinizing hormone levels (both hands). Unfortunately, this analytical approach could yield spurious results due to sex differences in both hormone levels and 2D:4D. Manning et al. presented no data relating testosterone levels to 2D:4D in adult women. More recently, Neave, Laing, Fink, and Manning (2003) reported that neither right nor left 2D:4D was significantly related to current salivary testosterone levels in undergraduate men.

These reports have stimulated considerable research on the relationship between 2D:4D and various traits putatively linked to sex hormones. The results of these studies have been mixed (see Table 1). While some authors report significant correlations between 2D:4D and such diverse traits as fertility, sexual attitudes and orientation, status, and cognitive abilities, attempts at replication by other authors have often failed. Some failures to replicate may be due to variation in methodology. Indeed, most positive results (with the exception of sexual orientation) have derived from a single laboratory. However, the positive results are not as compelling as they might be for two reasons. First, digit ratio has been operationalized at least four different ways: left 2D:4D, right 2D:4D, mean 2D:4D, and the difference between right and left 2D:4D (called Dr−l). Other measures such as 4D/height have been used as well (Martin, Manning & Downick 1999). This use of multiple predictor variables within a single study greatly increases the risk of Type I error (i.e., concluding that a relationship exists where none does). Second, the examination of a large number of possible phenotypic correlates of fetal hormone regimes within a study likewise expands the risk of Type I error. For example, Austin, Manning, McInroy, and Mathews (2002) assessed the relationship between 2D:4D and 18 abilities and personality dimensions, for each sex and for the sexes combined. Of the 56 (3×18) resulting correlations, 2 were significant (see their Table 2), and these two are not fully independent because one is a subscale of the other. These facts suggest a cautious approach to claims of a relationship between 2D:4D and particular aspects of the phenotype. Some, but not all, of these studies have used corrections (e.g., Bonferroni) for multiple comparisons.

Table 1.

Summary of previously tested phenotypic correlations with 2D:4D

Domain/Trait Correlation of trait with 2D:4D Reference
Sign Comments
Prenatal hormones
CAH Girls, both hands Okten et al. (2002)
Boys, right hand only Okten et al. (2002)
Girls, right hand only Brown et al. (2002)
Boys, left hand only Brown et al. (2002)
0 Only left hand evaluated Buck et al. (2003)
Maternal WHR Sexes combined Manning et al. (1998)
Otoacoustic emissions + Left side, homosexual men only; but no correction for 24 tests McFadden and Shubel (2003)
Current hormones
Testosterone Men sampled from infertility clinic Manning et al. (1998)
0 Men, both hands Neave et al. (2003)
Estrogen + Sexes combined Manning et al. (1998)
Luteinizing hormone + Sexes combined, right hand only Manning et al. (1998)
Progesterone + Sexes combined, right hand only Manning et al. (1998)
WHR Women, right hand only Manning et al. (2000)
0 Men Manning et al. (2000)
Fertility
Sperm count Right hand only Manning et al. (1998)
Ejaculate quality 0 Four variables including sperm count Firman et al. (2003)
Reproductive success Men, four of seven samples not significant Manning et al. (2000)
+ Women, five of eight samples not significant Manning et al. (2000)
Offspring sex ratio Effect significant only when ethnicity is controlled Manning et al. (2002)
Sexual orientation and identity
Homosexuality Men, left hand only; bisexuals lower than both homosexuals and controls Robinson and Manning (2000)
Men, both hands Rahman and Wilson (2003)
Women, both hands Rahman and Wilson (2003)
Women, right hand only McFadden and Shubel (2002)
+ Men, left hand only McFadden and Shubel (2002)
Women, right hand only Williams et al. (2000)
0 Men Williams et al. (2000)
Twins discordant for homosexuality Hall and Love (2003)
Sex role identity “Butch” lower than “femme” lesbians Brown et al. (2002)
Bem Sex Role Inventory Csathó et al. (2003)
Health
Autism Parents and sibs also lower than controls Manning et al. (2001)
Breast cancer + Age-of-onset effect, right hand only Manning and Leinster (2001)
Myocardial infarction + Age-of-onset effect, right hand only Manning (2002)
Fetal growth + Placental weight Ronalds, Phillips, Godfrey, and Manning (2002)
Neonatal crown–heel length Ronalds et al. (2002)
Vigor
Soccer ability League players vs. controls Manning and Taylor (2001)
Athletic ability Right hand only Manning and Taylor (2001)
Running speed 800 and 1500 m, right hand only Manning (2002)
Cognitive abilities and personality
Spatial ability 0 Men, water-level task Manning (2002)
Men, MRT, unusual scoring method Manning (2002)
Women, latency to problem solution Csathó et al. (2003)
+ Women, spatial recall Csathó et al. (2003)
Men, MRT, unspecified scoring method Manning (2002)
0 Women, MRT, unspecified scoring method Manning (2002)
0 Men, MRT, large sample Coolican and Peters (2003)
0 Women, MRT, large sample Coolican and Peters (2003)
Verbal fluency + Men, Varley Test (semantic fluency) Manning (2002)
0 Women, Varley Test (semantic fluency) Manning (2002)
0 Men, FAS Test (phonological fluency) Manning (2002)
0 Women, FAS Test (phonological fluency) Manning (2002)
Musical ability Men, symphony members vs. controls Sluming and Manning (2000)
0 Women, symphony members vs. controls Sluming and Manning (2000)
Left handedness Right hand only Manning et al. (2000)
Psychoticism Sexes combined, average of both hands Austin, Manning, McInroy, and Mathews (2002)
Neuroticism + Sexes combined, average of both hands Austin et al. (2002)
Sensation seeking Women Austin et al. (2002)
0 Men Austin et al. (2002)
Depression 0 Men (Martin 1999)
0 Women Martin et al. (1999)
Appearance
Attractiveness 0 Men Neave et al. (2003)
Masculinity Men Neave et al. (2003)
Height 0 Men Neave et al. (2003)
Status
SES + Men, right hand only Manning (2002)
0 Women Manning (2002)
Dominance Men Neave et al. (2003)

+=significant positive correlation; −=significant negative correlation; 0=no significant relationship; MRT=mental rotations test; for dichotomous traits (e.g., CAH) correlation indicates direction of change in 2D:4D when trait is present.

As Table 1 shows, many reported relationships are restricted to one sex, sexual orientation, or hand. Moreover, there are puzzling inconsistencies. For example, spatial ability is sexually dimorphic and, on both experimental and clinical evidence, appears linked to the ontogenetic effects of sex hormones Gaulin, 1995, Kimura & Hampson, 1993. For example, hyperandrogenized females with CAH exhibit elevated spatial ability compared to matched controls Hampson et al., 1998, Perlman, 1973, Resnick et al., 1986, whereas hypoandrogenized females with Turner's syndrome exhibit a specific cognitive deficit in the area of spatial problem solving Buchsbaum & Henkin, 1980, Rovet & Netley, 1982. Against this coherent background of research on spatial ability, the 2D:4D data are less patterned. The study of spatial navigation by Csathó et al. (2003) used a testing arena designed to be an analogue of the Morris water-maze (Morris, 1981). The Morris water maze reliably yields a male advantage (measured as shorter latencies to problem solution), but Csathó et al. found that shorter (more masculine) 2D:4D was significantly associated with longer solution latencies. Manning (2002) found no relationship between 2D:4D and performance on the Water-Level task, another spatial task that typically yields a male advantage. The Mental-Rotation Test Shepard & Metzler, 1971, Vandenberg & Kuse, 1978 produces the largest sex difference of any spatial task, and Manning does report that male (but not female) performance is negatively correlated with 2D:4D. However, with large samples of men and women, Coolican and Peters (2003) found no such relationship in either sex, nor did Austin et al. (2002).

At least for males, the sexual orientation data are also inconsistent. Rahman and Wilson (2003) found that both sexes of homosexuals have masculinized 2D:4D compared to same-sex heterosexuals. McFadden and Shubel (2002) agree regarding females but found that male homosexuals had feminized 2D:4D, whereas Williams et al. (2000) found no significant relationship between 2D:4D and male sexual orientation. Finally, Robinson and Manning (2000) found that the 2D:4D measures of male homosexuals were significantly masculinized, but that those of bisexual men were even more so. These relationships are far from clear at present.

1.1. The present research

In the two studies reported here, we reassessed the relationship between 2D:4D and several variables previously claimed to be related to 2D:4D (see Table 1), including sexual orientation, spatial ability, status, physical prowess, and components of reproductive success. In addition we examined the relationship between 2D:4D and several other traits whose expression is thought to be related to sex hormones, including voice pitch, sociosexuality, mating success, and fluctuating asymmetry (FA).

Low voice pitch is a male secondary sex characteristic whose pubertal development depends on elevated testosterone (T) levels Beckford et al., 1985, Hollien, 1960, Hollien et al., 1994). Voice pitch continues to correlate negatively with salivary T levels in young adult men (Daabs & Mallinger, 1999) and decreases with androgen treatment (Need et al., 1993). Thus, voice pitch indicates postnatal (pubertal and current) androgen levels.

Sociosexuality is a dimension of personality that indexes interest in casual or uncommitted sex. Sociosexuality is thought to differentiate separately within sex, such that high androgen and high estrogen levels lead to high sociosexuality for males and females, respectively (Mikach & Bailey, 1999). Like sociosexuality, the behaviors that lead to mating success are believed to differentiate separately within sex, such that high androgen and high estrogen lead to high mating success in males and females, respectively (Mikach & Bailey, 1999).

Sluming and Manning (2000) found a significant relationship between 2D:4D and musical ability. Because musical ability may be an honest signal of male genetic quality (Miller, 2000), Manning (2002) suggested that, in addition to signaling fertility, 2D:4D may also be a reliable signal of genetic quality. Fluctuating asymmetry (FA; Van Valen, 1962) is probably the most widely employed proxy measure of underlying genetic quality Gangestad & Thornhill, 1999, Gangestad & Thornhill, 1997, Møller, 1997, Møller & Thornhill, 1997a, Møller & Thornhill, 1997b.

2. Methods

Subjects were 230 male and 120 female undergraduates between the ages of 18 and 30 recruited from the psychology subject pool at the University of Pittsburgh. Subjects participated in one of two studies approved by the governing human subjects committee. To maximize our ability to detect relationships between 2D:4D and other aspects of the phenotype, we did not use Bonferroni correction of P values.

2.1. Study 1

One hundred eleven self-reported heterosexual males (mean age 18.9, standard deviation=1.2) participated in the first study. After signing a consent form that outlined the experimental procedure, each subject was seated at a computer monitor and given a combination headphone/headset microphone and a short written passage. The experimenter's recorded voice instructed the subject to read the passage aloud, ostensibly as a test of the recording equipment. The subject was then told that he would be competing with a male in another room for a lunch date with a woman in a third room, a protocol similar to that of Simpson, Gangestad, Christensen, and Leck (1999). In actuality, the male competitor was audio recorded, and the female was video recorded. The subject's voice was recorded as he (1) read the passage aloud (“baseline” recording), (2) described himself to the female (“courtship” recording), and (3) described why other men might respect or admire him to his competitor (“competitive” recording). From their consent forms subjects were aware of the upcoming “dating game” when they made their baseline recording; thus this knowledge may have influenced acoustic parameters of their voices. After this dating game scenario, each subject was given a questionnaire designed to assess social and physical dominance, sociosexuality, sexual interest in the potential date, and previous mating success.

2.1.1. Voice pitch

Average fundamental frequency (F0), the acoustic parameter most closely related to pitch, was later measured for each participant's three recordings using the voice analysis software program, Praat.

2.1.2. Social and physical dominance

Dominance entails access to rewards, such as mates and resources, that is not typically challenged by others. In most animals, dominance is largely a matter of intimidation through threats of aggression. In humans, dominance may also be achieved through skillful leadership and persuasion (Henrich & Gil-White, 2001). Thus, dominance may result from the ability to utilize physical force (physical dominance) or from leadership ability and prestige (social dominance).

Social and physical dominance were assessed via self-ratings. Following Mazur and Booth (1998), subjects were instructed that “a [socially] dominant person tells other people what to do, is respected, influential, and often a leader; while submissive people are not influential or assertive and are usually directed by others.” Subjects were then asked to rate their own dominance/submissiveness on a 6-point scale. To measure physical dominance, subjects were asked about their level of agreement with the following statement: “If you got in a fistfight with an average undergraduate male, you would probably win.” Subjects were also asked to rate the social and physical dominance of their competitor. Because every subject was exposed to the same competitor, a subject's rating of his competitor could be used as a baseline from which to evaluate his self-rating. The difference between a subject's rating of his own dominance and his rating of his competitor's dominance is here referred to as relative (social or physical) dominance.

2.1.3. Sociosexuality and sexual interest

According to the sociosexuality orientation inventory (SOI) developed by Simpson and Gangestad (1991), sociosexuality is measured by the subject's level of agreement or disagreement with a series of statements (e.g., “It is better not to have sexual relations until you are married”) The statements are counterbalanced by polarity, such that for some items agreement indicates high sociosexuality, whereas for others disagreement does so. The inventory is scored by first inverting the choices of the reverse-polarity items, assigning values for “strongly disagree” through “strongly agree” and then summing these values over all items. Low scores on the sociosexuality inventory indicate low interest in uncommitted sex. Subjects were asked to respond to a 12-item sociosexuality questionnaire. Also, as a measure of general interest in courtship, subjects were asked about their level of interest in going on a date with the (video recorded) woman who interviewed them.

2.1.4. Mating success

Mating success is a function of copulatory frequency with fertile partners. In this study, mating success was measured in two ways: number of potential conceptions and number of sex partners.

2.1.4.1. Number of potential conceptions: males

Mating success was first evaluated by estimating the number of conceptions that would likely result from a particular copulatory pattern in the absence of contraception. The number of potential conceptions (NPC; Pérusse, 1993) was calculated according to the equation

(1)
where the exponents p1pn represent the number of coital acts with partners 1…n, respectively. The constant 0.97 represents the probability that a single unprotected copulatory act, at a random time during the female's ovulatory cycle, will fail to result in a conception. This constant is derived empirically from the fertility research of Bongaarts and Potter (1983) and Leridon (1977). Thus, 0.97p is the probability that p copulatory acts will fail to produce a conception, and therefore 1–0.97p is the chance that p copulatory acts will produce a conception. The NPC calculation simply sums this quantity across sexual partners. NPC was calculated for subjects during the past year because the NPC calculation assumes that each female mate can conceive only once—a reasonable assumption over a single year.

2.1.4.2. Number of sex partners

Although number of sexual partners is an important component of male mating success, it is not perfectly correlated with NPC. This occurs because males who have had sex with many women may nonetheless have low NPC scores if they copulated few times with each partner. The number of sexual partners each subject had in the past year was therefore used as a measure of desire for and ability to obtain multiple sex partners. We chose an interval of 1 year (a) to ensure some variance, (b) to maximize recall accuracy, and (c) for comparability with other research (e.g., Faurie, Pontier, & Raymond, 2004) in the field.

2.1.5. Digit lengths and fluctuating asymmetry

After each subject completed this questionnaire, measurements and hand photocopies were obtained for assessing 2D:4D and FA. Each subject's wrists were measured at their maximum diameter, alternating left and right until each wrist had been measured twice. The correlation between repeated measurements was r=.985 for the left wrist and r=.982 for the right. Next, each subject's hands were photocopied, alternating left and right until each hand had been photocopied twice. Second through fifth digit lengths were later measured from these photocopies using a digital caliper accurate to 0.01 mm. Each digit was measured from tip to basal crease, and measurements from the two photocopies were averaged to obtain a final length measurement for each digit. Ten percent of digit lengths were remeasured from photocopies to assess intrameasurer reliability. Correlations between repeated measurements ranged from r=.990 to .998.

Four measures of 2D:4D were calculated from measurements of digit lengths. These included right 2D:4D, left 2D:4D, average 2D:4D, and the difference between right and left 2D:4D (Dr−l).

The asymmetry of each trait (wrists and four digits) was calculated as the absolute difference between left and right sides, divided by the average of the left and right measurements for the trait. A composite measure of fluctuating asymmetry was obtained by averaging the asymmetries of all five traits. Thus, total FA was obtained via the formula

(2)
where L and R represent left and right measurements of trait i. Traits that subjects reported having injured were discarded, and FA measurements from the remaining traits were averaged.

2.2. Study 2

One hundred nineteen male and 120 female subjects participated in this study. Palmar surfaces of each subject's hands were photocopied, and second and fourth digit lengths were later measured from these photocopies to be used in calculating FA and 2D:4D (see above). Unlike Study 1, FA was calculated only for the second and fourth digits. The average FA of these two traits alone is here referred to as FA2,4 to distinguish it from FA of the five traits measured in Study 1, referred to simply as FA.

Over 10% of the sample (30 subjects) was remeasured for assessing intrameasurer reliability. Correlations between first and second measurements ranged from r=.990 to .996.

2.2.1. Mental rotation ability

The original paper-and-pencil version of the Vandenburg and Kuse (1978) mental rotation test was administered to subjects, who were given 7 min to complete the task. This task consists of 20 items, each with two correct and two incorrect choices, and is conventionally scored as correct minus incorrect to adjust for guessing. When scored in this way, the MRT yields a reliable sex difference of approximately .7 standard deviation Linn & Petersen, 1985, Rosenthal & Rubin, 1982.

2.2.2. Mating success and sexual orientation

Following the mental rotation test, subjects completed the 12-item sociosexuality questionnaire described in Study 1 and answered a short sexual history questionnaire, which yielded the data necessary for calculating NPC and number of sexual partners in the past year (see above) and for assessing sexual orientation. Subjects were asked about their sexual orientation using a 5-point scale ranging from 1 (exclusive sexual interest in members of the opposite sex) to 5 (exclusive sexual interest in members of the same sex). Subjects who answered 1 or 2 on this scale were considered heterosexual, while those who answered 4 or 5 were considered homosexual.

2.2.3. Number of potential conceptions: females

Because the NPCs of females and males are differentially affected by sex with multiple partners, NPC was calculated differently for the two sexes. If .97 is the probability that a single unprotected copulatory act, at a random time during a female's ovulatory cycle, will fail to result in a conception, then a female's probability of conception can be estimated by the equation

(3)
where k is her total number of copulatory acts since her last conception. In this study, values for k were derived by summing the numbers of copulations each female reported for each of her mates.

2.3. Statistical analysis

Data from Study 1 and Study 2 were lumped wherever methods were identical in order to increase our statistical power to detect any 2D:4D relationships. One-tailed statistical tests were used when directional (i.e., masculine/feminine) predictions could be made. This included all correlations and sex differences, with the exceptions of sex differences in FA2,4, partner number, and NPC, which were analyzed using two-tailed t tests (see Discussion).

3. Results

3.1. Sex differences

The various 2D:4D measures and the candidate correlates of 2D:4D (mental-rotation performance, partner number, NPC, sociosexuality, and FA2,4) were first analyzed for sex differences using t tests. (Table 2).

Table 2.

Sex differences

Females Males t value P value
Mean N Mean N
MRT 13.5 120 20.0 119 −5.15 <.0001
No. of partners 2.08 120 1.57 222 2.19 .015a
NPC .480 120 .369 219 2.47 .014a
Sociosexuality 2.75 120 3.77 229 −9.66 <.0001
Left 2D:4D .979 120 .955 218 6.21 <.0001
Right 2D:4D .974 120 .949 220 7.00 <.0001
Mean 2D:4D .977 120 .951 213 7.23 <.0001
Dr−l .005 120 .001 214 .610 .271
FA2,4 .022 120 .018 213 2.97 .004a
a

Two-tailed P value.

Significant sex differences were found for all measures of 2D:4D (6.21≤t≤7.23; P<.0001), except Dr−l (the difference between right and left 2D:4D), which did not differ significantly between the sexes (t=0.610, P=.271). This result agrees with that of Manning et al. (2000), who found no significant sex difference in Dr−l in a sample of 567 males and 661 females from seven populations. Thus, Dr−l is unlikely to be a reliable marker for prenatal sex hormone regimes and will not be examined here as an independent variable.

All dependent variables showed significant sex differences. Male scores on the mental rotation test (20.0±0.921) were significantly higher than those for females (13.5±0.901) (t=−5.15, P<.0001), whereas females had significantly lower sociosexuality scores (t=−9.66, P<.0001), indicating lower interest in casual sex than males, they reported having had significantly more sexual partners in the past year (t=2.19, two-tailed P=.030) and had significantly higher NPC scores (t=2.47, two-tailed P=.028). Females also had significantly higher FA2,4 than did males (t=2.97, two-tailed P=.004).

3.2. Correlations with 2D:4D among females

Sexual orientation was significantly related to left 2D:4D (r=−.161, P=.040) and marginally significantly related to mean 2D:4D (r=−.143, P=.061) (Table 3). 2D:4D (left, right, and mean) did not significantly predict female NPCs, number of sex partners in the past year, sociosexuality, or FA2,4 (F tests), although left 2D:4D was almost significantly related to FA2,4 (r=.146, P=.060). However, females with higher (more feminine) left and mean 2D:4D had significantly better MRT scores (r=.157 and 0.167, P=.043 and .034, respectively), and right 2D:4D was marginally significantly related (r=.147, P=.054) to MRT score in the same, unexpected direction (Table 3).

Table 3.

Correlations between digit ratio and other measures in women

Left 2D:4D Right 2D:4D Mean 2D:4D
N r P N r P N r P
MRT 120 .157 .043 120 .147 .054 120 .167 .034
No. of partners 120 −.044 .316 120 .075 .207 120 −.064 .244
NPC 120 −.003 .487 120 −.068 .230 120 −.037 .344
Sociosexuality 120 .134 .073 120 −.042 .326 120 .057 .269
Sexual orientation 120 −.161 .040 120 −.095 .152 120 −.143 .061
FA2,4 120 .146 .056 120 .021 .411 120 .093 .155

3.3. Correlations with 2D:4D among males

Male sexual orientation was significantly related to left 2D:4D (r=−.125, P=.034) (Table 4). Among the (self-identified heterosexual) males from Study 1, no 2D:4D measure was significantly related to courtship-, competitive-, or baseline-voice pitch; relative social or physical dominance; interest in the female interviewer; or FA. Among all male participants, MRT, NPC; number of sex partners in the past year, and FA2,4 were not significantly related to any measure of 2D:4D. However, left 2D:4D was significantly related to sociosexuality (r=.127, P=.031) and mean 2D:4D was marginally significantly related (r=.093, P=.088) to sociosexuality (Table 4). On the possibility that spatial ability is curvilinearly related to androgen dosage in men (Hines et al., 2003) we also tested for a quadratic relationship between 2D:4D and MRT performance. This relationship was nonsignificant for left, right, and mean 2D:4D (F=0.49, 0.98, and 0.93, respectively, all n.s.).

Table 4.

Correlations between digit ratio and other measures in men

Left 2D:4D Right 2D:4D Mean 2D:4D
N r P N r P N r P
MRT 119 .013 .446 119 −.091 .162 119 −.042 .324
No. of partners 211 .014 .420 212 −.062 .186 206 −.001 .497
NPC 207 −.039 .290 209 −.023 .368 202 −.008 .458
Sociosexuality 218 .127 .031 219 .013 .424 213 .093 .088
Sexual orientation 217 −.125 .034 219 .020 .385 212 −.061 .187
Interest in female 98 .035 .383 100 .050 .408 93 .044 .338
Relative social dominance 99 −.019 .271 101 −.097 .273 94 −.061 .259
Relative physical dominance 98 .139 .114 100 .052 .323 93 .110 .146
Baseline F0 99 .048 .493 101 .037 .300 94 .047 .325
Courtship F0 99 .032 .415 101 .026 .367 94 .032 .380
Competitive F0 99 .037 .357 101 .062 .265 94 .054 .302
FA 99 .032 .340 101 −.029 .312 94 .002 .493
FA2,4 213 .051 .482 213 −.044 .250 213 .004 .316

3.4. Factor analysis

Factor analysis offers another analytical approach to examining relationships among these various phenotypic measures. Those traits subject to similar control mechanisms should cluster together. To maximize the data available for analysis, three separate varimax-rotated factor analyses were performed: one for the Study 1 participants, and one for each sex of participant in Study 2. Each yielded four or five factors and explained approximately 75% of the variance.

In each of these analyses, the three digit-ratio variables loaded heavily onto a single factor, and no other variables loaded heavily onto that same factor. In general, the other obvious constructs each loaded onto a single factor. For example, in Study 1, the three F0 variables loaded onto one factor. Sociosexuality, number of sex partners in the past year, and NPC loaded together. MRT score formed a separate factor, except in females, where it clustered with FA2,4.

4. Discussion

The significant sex differences in 2D:4D found in this study support the role of sex hormones in the development of digit lengths. Sex differences in mental rotation performance and sociosexuality were also highly significant in the predicted directions. Interestingly, number of sex partners, number of potential conceptions, and FA2,4, showed significant sex differences, as well, with females exceeding males for all three variables.

NPC and number of mates should exhibit no sex difference in a population with an unbiased sex ratio. This is because every time a female conceives or has heterosexual intercourse, a male does as well. On the other hand, given females' apparent underreporting of their sexual activity and/or males' overreporting (Baker & Bellis, 1995), one might predict the opposite of our findings. Three nonmutually exclusive explanations for these results suggest themselves: First, copulations with females may be concentrated among a few males who largely escape detection in a sample of this size, while copulations with males may be more evenly distributed among females (but see Baker & Bellis, 1995, for the reverse prediction). Second, because females sexually mature earlier and undergo menopause, their sexual activity may be concentrated earlier in life than is the sexual activity of males. Finally, females in their late teens and early twenties are expected to be more attractive as mates than are males of the same age (e.g., Buss & Schmitt, 1989). Thus, females in this age group may engage disproportionately in sex with older (more desirable) males.

Another interesting and unexpected finding was a sex difference in FA2,4. Across many vertebrate species, males are more parasitized than are females, perhaps because of males' riskier behavior or higher circulating levels of the immunosuppressant hormone, testosterone (Zuk & McKean, 1996). All else being equal, higher parasitic infection rates in males should cause greater FA in males, but the reverse was found. Greater levels of FA in women than in men have also been found in dermatoglyphics Arrieta et al., 1995, Micles & Kobyliansky, 1991 and dentition (Mizoguchi, 1988), although Perzigian (1977) and Townsend (1981) found no sex difference in levels of dental FA. Though not the focus of the present research, the observed sex differences in sexual activity and FA are intriguing and await replication and further investigation.

With respect to the main focus of the present research, our results suggest that the usefulness of 2D:4D as a predictor of other sex-hormone-mediated traits is limited. The 18 female correlations, 6 (Traits)×3 (Measures) of 2D:4D, produced only three significant results, and the 39 total male correlations, 13 (Traits)×3 (Measures) of 2D:4D, produced only two significant results at the .05 level. Of the three significant female correlations, two were counter to the predicted direction. Lower (more masculine) 2D:4D measures were expected to correlate with higher MRT scores in females, but the reverse was found. Likewise, one of the two significant male correlations was counter to expectation; males with higher (more feminine) 2D:4D ratios reported higher sociosexuality scores (greater interest in casual sex). Thus, only 2 correlations out of 57 were significant in the predicted direction. These indicated a link between sexual orientation and left 2D:4D that was manifested in both sexes. Additionally, factor analysis revealed that 2D:4D measures did not cluster with other measured traits.

The question arises: How can 2D:4D be a valid predictor of sex hormone levels and yet be unrelated to so many traits that also depend on sex hormones for their development? The answer may involve developmental timing. Sex hormone levels fluctuate substantially during human growth and development, and various sexually dimorphic traits differentiate at different times. Thus, traits that differentiate under the same hormonal influences may nonetheless be uncorrelated in their expression if they differ in developmental timing. Indeed, our factor analyses indicate that most of these sexually dimorphic traits do not cluster together. Although Manning et al. (1998) found a significant correlation between 2D:4D and adult androgen levels, their finding was limited to males, and the only attempt at replication of which we are aware (Neave et al., 2003) failed. If this result turns out to be spurious, then 2D:4D may indicate little more than sex hormone levels at the time when 2D:4D differentiates sexually.

One of the most robust correlates of 2D:4D is sexual orientation. Several independent laboratories have noted such a relationship, both for male sexual orientation Rahman & Wilson, 2003, Robinson & Manning, 2000 and, more consistently, for female sexual orientation Brown et al., 2002, Hall & Love, 2003, McFadden & Shubel, 2002, Rahman & Wilson, 2003. The present study supports these findings.

On the other hand, spatial ability does not appear to correlate reliably with 2D:4D (Austin et al., 2002, Coolican & Peters, 2003; this study) and yet probably differentiates sexually under the influence of the same hormones affecting sexual orientation. This suggests that 2D:4D and sexual partner preference differentiate sexually at similar times, but that their developmental timing does not coincide with the sexual differentiation of spatial ability. Abnormal fluctuations in fetal sex hormone levels (due to maternal hormonal fluctuations or factors intrinsic to the developing fetus) may primarily affect traits that differentiate sexually around the time of the fluctuations. Because CAH girls are exposed to consistently higher androgen levels prenatally, they tend to be masculinized for many traits that sexually differentiate prenatally. Thus, CAH girls exhibit masculinized spatial abilities, sexual orientation, childhood play patterns (Berenbaum, 1999), and 2D:4D.

In conclusion, 2D:4D may be statistically unrelated to the expression of many sex-hormone-dependent traits because (1) the timing of sexual differentiation in these traits differs from the timing of 2D:4D sexual differentiation and (2) individual differences in sex hormone regimes during 2D:4D sexual differentiation are statistically unrelated to individual differences in sex hormone regimes during sexual differentiation in these other traits.

References

Arrieta et al., 1995 1.Arrieta MI, Martinez B, Nunez M, Gil A, Criado B, Telez M, et al. A–b ridge count in a Basque population: fluctuating asymmetry and comparison with other populations. Human Biology. 1995;67:121–133.

Austin et al., 2002 2.Austin EJ, Manning JT, McInroy K, Mathews E. A preliminary investigation of the associations between personality, cognitive ability and digit ratio. Personality and Individual Differences. 2002;33:1115–1124.

Baker & Bellis, 1995 3.Baker RR, Bellis MA. Human sperm competition: copulation, masturbation, and infidelity. London: Chapman and Hall; 1995;.

Beckford et al., 1985 4.Beckford NS, Schain D, Roor SR, Schanbacher B. Androgen stimulation and laryngeal development. Annals of Otology, Rhinology and Laryngology. 1985;94:634–640. MEDLINE

Bongaarts & Potter, 1986 5.Bongaarts J, Potter RG. Fertility, biology and behavior: an analysis of proximate determinants. New York: Academic; 1986;.

Brown et al., 2002 6.Brown WM, Finn CJ, Breedlove SM. Sexual dimorphism in digit-length ratios of laboratory mice. The Anatomical Record. 2002;267:231–234.

Brown et al., 2002 7.Brown WM, Finn CJ, Cooke BM, Breedlove SM. Differences in finger length ratios between self-identified “butch” and “femme” lesbians. Archives of Sexual Behavior. 2002;31:123–127. MEDLINE | CrossRef

Brown et al., 2002 8.Brown WM, Hines M, Fane BA, Breedlove SM. Masculinized finger length in human males and females with congenital adrenal hyperplasia. Hormones and Behavior. 2002;42:380–386. CrossRef

Buchsbaum & Henkin, 1980 9.Buchsbaum MS, Henkin RI. Perceptual abnormalities in patients with chromatin negative gonadal dysgenesis and hypogonadotropic hypogonadism. International Journal of Neuroscience. 1980;11:201–209. MEDLINE

Buck et al., 2003 10.Buck JJ, Williams RM, Hughes IA, Acerini CL. In-utero androgen exposure and 2nd to 4th digit length ratio—Comparisons between healthy control and females with classical congenital adrenal hyperplasia. Human Reproduction. 2003;18:976–979. MEDLINE | CrossRef

Burley & Foster, 2004 11.Burley NT, Foster VS. Digit ratio varies with sex, egg order and strength of mate preference in zebra finches. Proceedings of the Royal Society, Series B. 2004;271:239–244.

Buss & Schmitt, 1989 12.Buss DM, Schmitt DP. Sexual strategies theory: an evolutionary perspective on human mating. Psychological Review. 1989;100:204–232. CrossRef

Chamberlain et al., 1994 13.Chamberlain NL, Driver ED, Meisfeld RL. The length and location of CAG trinucleotide repeats in the androgen receptor N-terminal domain affect transactivation function. Nucleic Acids Research. 1994;15:3181–3186. MEDLINE

Coolican & Peters, 2003 14.Coolican J, Peters M. Sexual dimorphism in the 2D/4D ratio and its relation to mental rotation performance. Evolution and Human Behavior. 2003;24:179–183.

Csathó et al., 2003 15.Csathó Á, Osváth A, Karádi K, Bisák É, Manning J, Kállai J. Spatial navigation related to the second to fourth digit length in women. Learning and Individual Differences. 2003;13:239–249.

Daabs & Mallinger, 1999 16.Daabs JM, Mallinger A. High testosterone levels predict low voice pitch among men. Personality and Individual Differences. 1999;27:801–804.

Faurie et al., 2004 17.Faurie C, Pontier D, Raymond M. Student athletes claim to have more sexual partners than other students. Evolution and Human Behavior. 2004;25:1–8.

Firman et al., 2003 18.Firman RC, Simmons LW, Cummins JM, Matson PL. Are body fluctuating asymmetry and the ratio of the 2nd to 4th digit reliable predictors of semen quality?. Human Reproduction. 2003;18:808–812. MEDLINE | CrossRef

Gangestad & Thornhill, 1997 19.Gangestad SW, Thornhill R. Human sexual selection and developmental stability. In:  Simpson JA,  Simpson DT editor. Evolutionary personality and social psychology. Hillsdale, NJ: Erlbaum; 1997;p. 169–195.

Gangestad & Thornhill, 1999 20.Gangestad SW, Thornhill R. Individual differences in developmental precision and fluctuating asymmetry: a model and its implications. Journal of Evolutionary Biology. 1999;12:402–416. CrossRef

Garn et al., 1975 21.Garn SM, Burdi AR, Babler WJ, Stinson S. Early prenatal attainment of adult metacarpal–phalageal rankings and proportions. American Journal of Physical Anthropology. 1975;43:327–332. MEDLINE | CrossRef

Gaulin, 1995 22.Gaulin SJC. Does evolutionary theory predict sex differences in the brain?. In:  Gazzaniga M editors. The cognitive neurosciences. Cambridge, MA: MIT Press; 1995;p. 1211–1225.

Hall & Love, 2003 23.Hall LS, Love CT. Finger length ratios in female monozygotic twins discordant for sexual orientation. Archives of Sexual Behavior. 2003;32:23–28. MEDLINE | CrossRef

Hampson et al., 1998 24.Hampson E, Rovet JF, Altman D. Spatial reasoning in children with congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Developmental Neuropsychology. 1998;14:299–320.

Henrich & Gil-White, 2001 25.Henrich J, Gil-White FJ. The evolution of prestige: freely conferred deference as a mechanism for enhancing the benefits of cultural transmission. Evolution and Human Behavior. 2001;22:165–196.

Hines et al., 2003 26.Hines M, Fane BA, Pasterski VL, Mathews GA, Conway GS, Brook C. Spatial abilities following prenatal androgen abnormalities: targeting and mental rotation performance in individuals with congenital adrenal hyperplasia. Psychoneuroendocrinology. 2003;28:1010–1026. Abstract | Full Text | Full-Text PDF (182 KB) | MEDLINE | CrossRef

Hollien, 1960 27.Hollien H. Some laryngeal correlates of vocal pitch. Journal of Speech and Hearing Research. 1960;3:52–58. MEDLINE

Hollien et al., 1994 28.Hollien H, Green R, Massey K. Longitudinal research on adolescent voice change in males. Journal of the Acoustical Society of America. 1994;96:2646–2654. MEDLINE | CrossRef

Jamison et al., 1993 29.Jamison CS, Meier RJ, Campbell BC. Dermatoglyphic asymmetry and testosterone levels in normal males. American Journal of Physical Anthropology. 1993;90:185–198. MEDLINE | CrossRef

Kazemi-Esfarjani et al., 1995 30.Kazemi-Esfarjani P, Trifiro MA, Pinski L. Evidence for a recessive function of the long polyglutamine tract in the human androgen receptor: possible pathogenic relevance for the (CAG)n-expanded neuropathies. Human Molecular Genetics. 1995;4:523–527. MEDLINE

Kimura & Hampson, 1993 31.Kimura D, Hampson E. Neural and hormonal mechanisms mediating sex differences in cognition. In:  Vernon PA editors. Biological approaches to the study of human intelligence. Norwood, NJ: Ablex; 1993;p. 375–397.

Leridon, 1977 32.Leridon H. Human fertility. Chicago: University of Chicago Press; 1977;.

Linn & Petersen, 1985 33.Linn MC, Petersen AC. Emergence and characterization of sex differences in spatial ability: a meta-analysis. Child Development. 1985;56:1479–1498. MEDLINE | CrossRef

Manning, 2002 34.Manning JT. Digit ratio. New Brunswick, NJ: Rutgers University Press; 2002;.

Manning et al., 2000 35.Manning JT, Barley L, Walton J, Lewis-Jones DI, Trivers RL, Singh D, et al. The 2nd:4th digit ratio, sexual dimorphism, population differences, and reproductive success: evidence for sexually antagonistic genes?. Evolution and Human Behavior. 2000;21:163–183.

Manning et al., 2003 36.Manning JT, Bundred PE, Newton DJ, Flanagan BF. The second to fourth digit ratio and variation in the androgen receptor gene. Evolution and Human Behavior. 2003;24:399–405.

Manning et al., 2002 37.Manning JT, Martin S, Trivers RL, Soler M. 2nd to 4th digit ratio and offspring sex ratio. Journal of Theoretical Biology. 2002;217:93–95. MEDLINE | CrossRef

Manning et al., 1998 38.Manning JT, Scutt D, Wilson J, Lewis-Jones DI. The ratio of the 2nd and 4th digit length: a predictor of sperm numbers and concentrations of testosterone, luteinizing hormone and oestrogen. Human Reproduction. 1998;13:3000–3004. MEDLINE | CrossRef

Manning & Taylor, 2001 39.Manning JT, Taylor RP. Second to fourth digit ratio and male ability in sport: implications for sexual selection in humans. Evolution and Human Behavior. 2001;22:61–69.

Martin et al., 1999 40.Martin SM, Manning JT, Dowrick CF. Fluctuating asymmetry, relative digit length, and depression in men. Evolution and Human Behavior. 1999;20:203–214.

Mazur & Booth, 1998 41.Mazur A, Booth A. Testosterone and dominance in men. Behavioral and Brain Sciences. 1998;21:353–397. CrossRef

McFadden & Shubel, 2002 42.McFadden D, Shubel E. Relative lengths of fingers and toes in human males and females. Hormones and Behavior. 2002;42:492–500. CrossRef

McFadden & Shubel, 2003 43.McFadden D, Shubel E. The relationships between otoacoutic emissions and relative lengths of fingers and toes. Hormones and Behavior. 2003;43:421–429. CrossRef

Micles & Kobyliansky, 1991 44.Micles S, Kobyliansky E. Asymmetry and diversity of dermatoglyphics. Homo. 1991;42:21–42.

Mikach & Bailey, 1999 45.Mikach SM, Bailey JM. What distinguishes women with unusually high numbers of sex partners?. Evolution and Human Behavior. 1999;20:141–150.

Miller, 2000 46.Miller G. The mating mind. New York: Doubleday; 2000;.

Mizoguchi, 1988 47.Mizoguchi Y. Degree of bilateral asymmetry of nonmetric tooth crown characters quantified by the tetrachoric correlation method. Bulletin of the National Science Museum, Series D, Anthropology. 1988;14:29–49.

Møller, 1997 48.Møller AP. Developmental stability and fitness: a review. American Naturalist. 1997;149:916–932. CrossRef

Møller & Thornhill, 1997a 49.Møller AP, Thornhill R. A meta-analysis of the heritability of developmental stability. Journal of Evolutionary Biology. 1997;10:1–16. CrossRef

Møller & Thornhill, 1997b 50.Møller AP, Thornhill R. Developmental stability is heritable. Journal of Evolutionary Biology. 1997;10:69–76. CrossRef

Neave et al., 2003 51.Neave N, Laing S, Fink B, Manning JT. Second to fourth digit ratio, testosterone and perceived male dominance. Proceedings of the Royal Society, Series B. 2003;270:2167–2172.

Need et al., 1993 52.Need AG, Durbridge TC, Nordin BE. Anabolic steroids in postmenopausal osteoporosis. Wien. Med. Wochenshr. 1993;143:392–395.

Okten et al., 2002 53.Okten A, Kalyoncu M, Yaris N. The ratio of second- and fourth-digit lengths and congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Early Human Development. 2002;70:47–54. Abstract | Full Text | Full-Text PDF (95 KB) | MEDLINE | CrossRef

Perlman, 1973 54.Perlman SM. Cognitive abilities of children with hormone abnormalities: screening by psycho-educational test. Journal of Learning Disabilities. 1973;6:21–29.

Pérusse, 1993 55.Pérusse D. Cultural and reproductive success: testing the relationship at proximate and ultimate levels. Behavioral and Brain Sciences. 1993;16:267–322.

Perzigian, 1977 56.Perzigian AJ. Fluctuating dental asymmetry variation among skeletal populations. American Journal of Physical Anthropology. 1977;47:81–88. MEDLINE | CrossRef

Phelps, 1952 57.Phelps VR. Relative index finger length as a sex-influenced trait in man. American Journal of Human Genetics. 1952;4:72–89. MEDLINE

Rahman & Wilson, 2003 58.Rahman Q, Wilson GD. Sexual orientation and the 2nd to 4th finger length ratio: evidence for organizing effects of sex hormones or developmental instability?. Psychoneuroendocrinology. 2003;28:288–303. Abstract | Full Text | Full-Text PDF (87 KB) | MEDLINE | CrossRef

Resnick et al., 1986 59.Resnick SM, Berenbaum SA, Gottesman II, Bouchard TJ. Early hormonal influences on cognitive functioning in congenital adrenal hyperplasia. Developmental Psychology. 1986;22:191–198. CrossRef

Robinson & Manning, 2000 60.Robinson SJ, Manning JT. The ratio of 2nd to 4th digit length and male homosexuality. Evolution and Human Behavior. 2000;21:333–345.

Ronalds et al., 2002 61.Ronalds G, Phillips DI, Godfrey KM, Manning JT. The ratio of second to fourth digit length: a marker of impaired fetal growth. Early Human Development. 2002;68:21–26. Abstract | Full Text | Full-Text PDF (78 KB) | MEDLINE | CrossRef

Rosenthal & Rubin, 1982 62.Rosenthal R, Rubin DB. Further meta-analytic procedures for assessing cognitive gender differences. Journal of Educational Psychology. 1982;74:708–712. CrossRef

Rovet & Netley, 1982 63.Rovet J, Netley C. Processing deficits in Turner's syndrome. Developmental Psychology. 1982;18:77–94. CrossRef

Shepard & Metzler, 1971 64.Shepard RN, Metzler J. Mental rotation of three-dimensional objects. Science. 1971;171:701–703. MEDLINE

Simpson & Gangestad, 1991 65.Simpson JA, Gangestad SW. Individual differences in sociosexuality: evidence for convergent and discriminant validity. Journal of Personality and Social Psychology. 1991;59:971–980. CrossRef

Simpson et al., 1999 66.Simpson JA, Gangestad SW, Christensen PN, Leck K. Fluctuating asymmetry, sociosexuality, and intrasexual competitive tactics. Journal of Personality and Social Psychology. 1999;76:159–172. MEDLINE | CrossRef

Sluming & Manning, 2000 67.Sluming VA, Manning JT. Second to fourth digit ratio in elite musicians: evidence for musical ability as an honest signal of male fitness. Evolution and Human Behavior. 2000;21:1–9.

Townsend, 1981 68.Townsend GC. Fluctuating asymmetry in the deciduous dentition of Australian aboriginals. Journal of Dental Research. 1981;60:1849–1857. MEDLINE

Van Valen, 1962 69.Van Valen L. A study of fluctuating asymmetry. Evolution. 1962;16:125–142. CrossRef

Vandenberg & Kuse, 1978 70.Vandenberg SG, Kuse AR. Mental rotations: a group test of three-dimensional spatial visualization. Perceptual and Motor Skills. 1978;47:599–604. MEDLINE

Williams et al., 2000 71.Williams TJ, Pepitone ME, Christensen SE, Cooke BM, Huberman AD, Breedlove NJ, et al. Finger length ratios and sexual orientation. Nature. 2000;404:455–456. MEDLINE | CrossRef

Zuk & McKean, 1996 72.Zuk M, McKean KA. Sex differences in parasite infections: patterns and processes. International Journal for Parasitology. 1996;26:1009–1023. MEDLINE | CrossRef

a Department of Anthropology, University of Pittsburgh, Pittsburgh, PA 15260, USA

b Department of Anthropology, University of California, Santa Barbara, CA 93106, USA

c Department of Psychology, University of Pittsburgh, Pittsburgh, PA 15260, USA

Corresponding author. Tel.: +1-412-648-7500.

PII: S1090-5138(04)00021-2

doi:10.1016/j.evolhumbehav.2004.03.005



2007:11:13