A preliminary investigation of the associations between personality, cognitive ability and digit ratio

https://doi.org/10.1016/S0191-8869(02)00002-8Get rights and content

Abstract

The ratio between second and fourth digit lengths (digit ratio) is known to be sexually dimorphic, with males having lower values. It is believed that digit ratio acts as a marker of pre-natal testosterone exposure and a number of studies of its correlates support this idea. In the present study, associations between digit ratio and a number of cognitive and personality tests which are known to show male–female score differences were investigated. Evidence for associations in the expected direction was found for sensation seeking, psychoticism and neuroticism, in all cases for females only. No association was found between digit ratio and cognitive test scores.

Introduction

The ratio between second and fourth digit lengths (digit ratio) is known to be sexually dimorphic, with mean values being lower for males than for females (Baker, 1888, George, 1930, Manning et al., 2000a, Phelps, 1952). Digit ratio is also known to be stable between childhood and adulthood and to be determined in utero by about the 14th week (Garn et al., 1975, Manning et al., 1998, Phelps, 1952). There is accumulating evidence that in both males and females digit ratio acts as a marker of the levels of testosterone and oestrogen to which the developing foetus was exposed. These findings include correlations in the expected direction between digit ratio and (1) adult testosterone and oestrogen levels (Manning et al., 1998), (2) reproductive success in males and females (Manning et al., 2000) (3) female waist:hip ratio (Manning, Trivers, Singh, & Thornhill, 1999) (4) a condition (congenital adrenal hyperplasia) associated with high prenatal androgen (Brown, Hines, Fane, & Breedlove, 2001) and (5) negative correlations between maternal digit ratio and androgen in the amniotic fluid of the foetus (Manning, in press). It has been suggested (Manning et al., 1998) that the underlying mechanism for such correlations is via the action of the Homeobox genes, which control the differentiation of digits, testes and ovaries. This common control of the differentiation of digits and gonads may allow aspects of gonadal function such as the production of testosterone and oestrogen to affect the development of the digits. However, it is possible that the sexual dimorphism in the digit ratio may arise prenatally as the result of factors other than that of in utero sex steroids. For example Arnold (1996) has emphasised the role that the Y chromosome itself may play in sexual differentiation and resultant sex differences.

The effects of prenatal testosterone levels feature in Geschwind and Galaburda's (1985) theory that high levels of foetal testosterone may compromise the development of the left cerebral hemisphere (leading to left-hand preference, language impairments and autism) and facilitate that of the right hemisphere (leading to enhanced musical, spatial and mathematical abilities). Evidence linking this theory to digit ratio has been found. Left-hand preference in a peg moving task in children has been shown to be related to low digit ratios (Manning, Trivers, Thornhill, & Singh, 2000b), autistic children have been found to have lower digit ratio values than the general population (Manning, Baron-Cohen, Wheelwright, & Sanders, 2001), whilst a study of male musicians in a British symphony orchestra showed that they had lower digit ratios than controls and that high-ranking players had lower values than their low-ranking colleagues (Sluming & Manning, 2000). These studies show effects in the direction that would be expected from the Geschwind–Galaburda theory if digit ratio is a marker for prenatal testosterone.

There are known to be a number of replicable male–female differences in personality and cognitive ability. The above review suggests that it is of interest to investigate the associations of digit ratio with personality and ability traits for which mean scores differ between the sexes, especially those for which a theoretical or empirical link to levels of sex hormones has been established. In personality, males have consistently been found to score higher than females on psychoticism (Eysenck & Eysenck, 1976) and on aggression scales (Harris, Rushton, Hampson, & Jackson, 1996). Males also score higher than females on sensation seeking and on the impulsive unsocialised sensation seeking (P-ImpUSS) dimension derived by Zuckerman (1991). Females have been found to score higher than males on neuroticism and on measures of depression (Eysenck and Eysenck, 1976, Hawkins et al., 1989, Sowa and Lustman, 1984) and to have a higher incidence of diagnosis of clinical depression (Brems, 1995). In the case of aggression, associations between aggression scores and testosterone levels have been found for both sexes (Harris, 1999, Harris et al., 1996), whilst positive associations between sensation seeking scores and testosterone levels have been reported for males (Gerra et al., 1999). Negative associations between testosterone levels and neuroticism have also been found in males (Dabbs, Hopper, & Jurkovic, 1990).

In cognitive ability, a number of male–female differences have been documented. Males obtain higher scores on some types of spatial task with male-female differences being most marked on mental rotation tasks (Voyer, Voyer, & Bryden, 1995). There is evidence that performance on spatial tasks varies with testosterone level, but the relationship appears to be curvilinear, with optimum performance being found when testosterone levels are in the low male range (studies in this area are reviewed by Kimura, 1996). Females are found to score higher than males on some verbal tasks. One example of a task with superior female performance is verbal fluency, although the effect size is small (Kimura, 1999).

From the above discussion it is possible to formulate hypotheses about the expected signs of the correlations between digit ratio and personality and cognitive tests. Assuming that personality and cognitive scores which show male–female differences are affected by pre-natal testosterone levels, the prediction is for negative correlations with digit ratio for tests for which male scores have been found to be higher (mental rotation, aggression, sensation seeking, psychoticism) and for positive correlations for traits where female scores have been found to be higher (verbal fluency, neuroticism, depression-proneness). In the case of mental rotation a previous study (Manning & Taylor, 2001) has found that performance was negatively correlated with digit ratio in a male sample.

In this paper two studies on relationships between digit ratio and personality and cognitive ability measures are described. Although some measures were common between the two studies, we do not present any combined data, as a preliminary ANOVA analysis indicated significant group differences in digit ratio. Since such a difference might be related to gene frequency differences in the two populations sampled, the use of separate analyses is appropriate.

Section snippets

Participants

The participants were 165 undergraduate students (79 male, 86 female) at the University of Edinburgh. The mean age of the males was 20.1 years (standard deviation 1.1 years) and the mean age of the females was 20.6 years (standard deviation 2.5 years).

Verbal fluency

Two short tests of verbal fluency were administered. Participants were first asked to write down as many words as they could think of beginning with the letter ‘F’ in 1 min (Miller, 1984). In the second test participants were given 2.5 min to

Participants

The participants were 49 male and 51 female students at the University of Liverpool. The mean age of the males was 19.1 years (standard deviation 0.99 years) and the mean age of the females was 21.8 years (standard deviation 7.7 years).

The Eysenck Personality Questionnaire-Revised

As in Study 1.

The Aggression Questionnaire

As in Study 1.

Procedure

Digit ratio was measured as for Study 1, and measurement was blind to questionnaire results. Both left and right hands were measured and all hands were measured twice to assess measurement repeatability. Participants were tested in

Discussion

The two studies presented here extend the study of the correlates of digit ratio to investigate its associations with personality traits where mean scores differ between the sexes. Some evidence for associations in the expected direction for sensation seeking, psychoticism and neuroticism was found. No significant associations were found for the cognitive tests, with the association between digit ratio and mental rotation performance found by Manning and Taylor (2001) not being replicated. In

References (35)

  • W.M Brown et al.

    Masculinized finger length ratios in humans with congenital adrenal hyperplasia (CAH)

    Hormones & Behavior

    (2001)
  • A.H Buss et al.

    The aggression questionnaire

    Journal of Personality and Social Psychology

    (1992)
  • H.J Eysenck et al.

    Psychoticism as a dimension of personality

    (1976)
  • S.M Garn et al.

    Early prenatal attainment of adult metacarpal-phalangeal rankings and proportions

    American Journal of Physical Anthropology

    (1975)
  • R George

    Human finger types

    Anatomical Record

    (1930)
  • G Gerra et al.

    Neurotransmitters, neuroendocrine correlates of sensation-seeking temperament in normal humans

    Neuropsychobiology

    (1999)
  • N Geschwind et al.

    Cerebral lateralization. Biological mechanisms, association and pathology: a hypothesis and a program for research

    Archives of Neurology

    (1985)
  • Cited by (0)

    View full text