Interpersonal behavior in women: Association with prenatal exposure to sex hormones

Abstract

The ratio between the second and fourth digit of the hand (2D:4D ratio) has been used as a marker of prenatal exposure to sex hormones. A previous study found that men with smaller 2D:4D ratios, relative to men with larger ratios, were more agreeable and less quarrelsome with women than with men. We aimed to replicate this study, using a similar procedure for sampling interpersonal behaviors but including participants who were first-year students (n = 81) rather than working adults in a long-term romantic relationship. Event-contingent recording of everyday social interactions yielded data on agreeableness, quarrelsomeness, dominance, and submissiveness in different social contexts (e.g., with male vs. female interaction partners). While we did not replicate the findings previously obtained in men, this might be due to the relatively low number of male participants. Female participants with larger 2D:4D ratios were more agreeable with women than with men; this was not found among women with smaller ratios. Future studies on the link between the 2D:4D ratio and interpersonal behavior should consider with whom interactions take place and examine additional factors such as participants' relationship status and hormonal contraceptive use.

Full text

Contents lists available at ScienceDirect
Personality and Individual Differences
journal homepage: www.elsevier.com/locate/paid
Short Communication
Interpersonal behavior in women: Association with prenatal exposure to sex
hormones
Michelle Spix
a,b,⁎
, Marije aan het Rot
a
a
Department of Psychology, Faculty of Behavioural and Social Sciences, University of Groningen, Netherlands
b
Department of Clinical Psychological Science, Faculty of Psychology and Neuroscience, Maastricht University, Netherlands
ARTICLE INFO
Keywords:
2D:4D ratio
Agreeableness
Estrogen
Event-contingent recording
Interpersonal behavior
Sex hormones
Social interaction
Testosterone
ABSTRACT
The ratio between the second and fourth digit of the hand (2D:4D ratio) has been used as a marker of prenatal
exposure to sex hormones. A previous study found that men with smaller 2D:4D ratios, relative to men with
larger ratios, were more agreeable and less quarrelsome with women than with men. We aimed to replicate this
study, using a similar procedure for sampling interpersonal behaviors but including participants who were first-
year students (n= 81) rather than working adults in a long-term romantic relationship. Event-contingent re-
cording of everyday social interactions yielded data on agreeableness, quarrelsomeness, dominance, and sub-
missiveness in different social contexts (e.g., with male vs. female interaction partners). While we did not re-
plicate the findings previously obtained in men, this might be due to the relatively low number of male
participants. Female participants with larger 2D:4D ratios were more agreeable with women than with men; this
was not found among women with smaller ratios. Future studies on the link between the 2D:4D ratio and
interpersonal behavior should consider with whom interactions take place and examine additional factors such
as participants' relationship status and hormonal contraceptive use.
1. Introduction
Prenatal exposure to sex hormones shapes brain development
(McEwen & Milner, 2017). An easy-to-measure marker of prenatal
testosterone relative to estrogen exposure is the 2D:4D ratio, re-
presenting the length of the second digit relative to the length of the
fourth digit. Men generally have smaller ratios than women, indicating
higher prenatal testosterone exposure (Manning, 2011).
Besides, within-gender variation in the 2D:4D ratio has been linked
to within-gender variation in interpersonal functioning. Moskowitz
et al. (2015) found that while men were generally more agreeable and
less quarrelsome towards women than towards other men, this was
particularly true for men with smaller 2D:4D ratios. Associations be-
tween 2D:4D ratios and interpersonal behaviors were not significant in
women. It was suggested that men exposed to more testosterone pre-
natally are more masculine and hence more motivated to connect with
women. In the present study, we aimed to replicate these findings.
While other studies have only found only weak associations be-
tween the 2D:4D ratio and psychological functioning (Fink et al., 2004;
Hönekopp & Watson, 2011), many used standard questionnaires, which
only allow for examining between-person differences. In contrast,
Moskowitz et al. (2015) used event-contingent recording of social
interactions. As this involves intensive repeated measures across mul-
tiple social contexts, sources of within-person variation in interpersonal
behavior (e.g., the gender of interaction partners) can be considered as
well.
Unlike Moskowitz et al. (2015), whose participants were working
adults in long-term heterosexual relationships, we studied first-year
students who varied in their relationship status. Nonetheless, we ex-
pected 2D:4D ratios to be associated with agreeableness and quarrel-
someness in men but not in women.
2. Method
2.1. Participants
First-year students participated in exchange for course credit
(n= 110). Twenty-nine participants were excluded (4 did not identify
as heterosexual, 4 provided no hand scans, and 21 did not complete the
event-contingent recording), leaving 81 participants for data analyses.
Descriptive data are in Table 1.
https://doi.org/10.1016/j.paid.2020.110379
Received 14 June 2020; Received in revised form 31 August 2020; Accepted 1 September 2020
⁎
Corresponding author at: Maastricht University, Universiteitssingel 40, 6229 ER Maastricht, Netherlands.
E-mail addresses: m.spix@maastrichtuniversity.nl (M. Spix), m.aan.het.rot@rug.nl (M. aan het Rot).
Personality and Individual Differences 168 (2021) 110379
Available online 25 September 2020
0191-8869/ © 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/BY/4.0/).
T

2.2. Measures
2.2.1. 2D:4D ratios
Digit lengths of both hands were derived from hand scans, gener-
ated using an Aficio MP3351 digital copier (Ricoh, Japan), by four
coders using AutoMetric 2.2 (Face Research Lab, University of
Glasgow). Each second and fourth digit was measured from the middle
of the ventral base crease to the tip. Intra-class coefficients (Shrout &
Fleiss, 1979) were ≥0.89 for the right and left hand separately and for
male and female participants separately. Thus, the right and left 2D:4D
ratios from the four raters were averaged. The correlation between right
and left ratios was strongly positive, r(81) = 0.80, p< 0.0001, and
right vs. left ratios did not differ significantly, t(80) = 1.91, p= 0.06.
2.2.2. Interpersonal behavior
For event-contingent recording (ECR; Moskowitz & Sadikaj, 2013)
of social interactions, we installed a link to an app (TEMPEST; Batalas
et al., 2018) on participants' smartphones. Using the app, either online
or offline, participants recorded their interactions for 14 days. First,
they provided information on contextual variables, including the
gender and role of their primary interaction partner. Subsequently, they
checked which behaviors, of a list of 12 behaviors, they engaged in. For
each interaction, to prevent fixed response sets, participants randomly
saw one of four different but comparable behavior lists. These lists were
derived from the Social Behavior Inventory covering agreeableness,
quarrelsomeness, dominance, and submissiveness (Moskowitz, 1994).
Corresponding to prior studies (e.g. Moskowitz, 1994), for every
interaction we first calculated the mean frequency of all checked be-
haviors and then subtracted this mean frequency from the frequency of
a specific type of behavior. The resulting ipsatized scores represent the
rate at which each type of behavior was checked, adjusted for general
behavior checking.
2.3. Procedure
First-year psychology students at the University of Groningen enroll
into an online research participation platform at the start of the first
semester and provide personal information including their sexual
identity, relationship status, and contraceptive use. Participants in the
present study were recruited via this platform, after the Departmental
Ethics Committee approved the study.
In a first meeting with a research assistant, participants received
detailed study information, provided consent, and completed several
questionnaires. Subsequently, TEMPEST was installed, the ECR proce-
dure was explained, and a sample social interaction (i.e., a spoken
conversation of at least 5 min) was recorded together with the assistant.
In a second meeting, within a week after the ECR period, participants
completed additional questionnaires and provided hand scans.
2.4. Analyses
Data were analyzed in SAS 9.4 (SAS, Cary, NC) using multilevel
modeling with an unstructured covariance structure. Social interactions
were nested within individuals. Degrees of freedom were calculated
using the Satterthwaite method.
Analyses were identical to Moskowitz et al. (2015): social interac-
tions involving parents or siblings were omitted; each participant's
2D:4D ratio was standardized using the mean ratio for participant's
gender and treated as a continuous between-person factor; and analyses
were conducted separately for each dimension of behavior, for men and
women, and for the right and left hand. The gender of the primary
partner during each social interaction was the within-person factor. In
the first analysis step, main effects of Ratio and Partner gender were
entered. In the second step, the interaction term was added. Significant
interaction terms were examined by estimating simple slopes and cal-
culating point estimates.
3. Results
The mean number of social interactions available for analyses was
46, with no significant differences between men and women, see
Table 1. Analysis outcomes are in Supplementary Table S1.
3.1. Analyses in the men
After omitting family interactions, the mean number of interactions
was 36. The main effect for the 2D:4D ratio and the interaction term
were never significant.
3.2. Analyses in the women
After omitting family interactions, the mean number of interactions
was 42. Again, we found no significant main effects for the 2D:4D ratio
and no significant interaction terms for dominance and quarrelsome-
ness. The interaction effect was significant for submissiveness in the
analysis involving the right-hand ratio but simple slopes for male vs.
female partners were not significant.
However, the interaction term was significant for agreeableness in
the analysis involving the right-hand ratio, see Fig. 1. Women with
larger ratios acted more agreeably towards other women than towards
men, b= 6.24, t(1336) = 3.44, p< 0.001, d= 0.19. This was not
found among women with smaller ratios, b= −0.56, t
(1325) = −0.29, p= 0.77, d= 0.02. Similarly, while the analysis
involving the left-hand ratio did not yield a significant interaction term
(0.05 < p< 0.10), post-hoc simple contrasts revealed that women
with larger left-hand ratios were more agreeable towards other women
than towards men, b= 5.08, t(1361) = 2.94, p= 0.003, d= 0.16, and
this was not found among women with smaller ratios, b= 0.58, t
(1349) = 0.30, p= 0.76, d= 0.02.
4. Discussion
Between-person differences in 2D:4D ratios have previously been
associated with differences in interpersonal behavior among working
adults: Moskowitz et al. (2015) found that men in their late twenties
with smaller 2D:4D ratios, considered a marker of higher prenatal
Table 1
Relevant descriptive and social interaction data.
Men (n= 28) Women (n= 53) tor X
2
test
Age in years 21 (1) 20 (1) t(73.2) = 3.69, p= 0.002
In a romantic relationship when entering university 39% 44% X
2
(1) = 0.18, p= 0.67
Hormonal contraceptive use – 70% –
Right-hand preference 75% 89% X
2
(1) = 2.54, p= 0.11
Right-hand 2D:4D ratio 0.97 (0.03) 0.98 (0.04) t(79) = −0.51, p= 0.61
Left-hand 2D:4D ratio 0.97 (0.03) 0.98 (0.03) t(79) = −1.40, p= 0.17
Number of social interactions available for analyses 41 (20) 49 (17) t(79) = −1.91, p= 0.06
Note. Data represent M (SD) unless indicated otherwise.
M. Spix and M. aan het Rot Personality and Individual Differences 168 (2021) 110379
2

exposure to testosterone, were more agreeable and less quarrelsome
towards women than men with larger ratios. It was suggested that
higher prenatal testosterone exposure increases men's engagement with
behaviors that, to women, indicate a willingness to invest in opposite-
gender relationships.
We aimed to replicate these findings in a younger sample, i.e. stu-
dents in their early twenties. However, we found no significant asso-
ciations between 2D:4D ratios and interpersonal behavior among male
participants: men acted more agreeably and were less quarrelsome to-
wards women than towards men, regardless of their 2D:4D ratios. Thus,
we did not replicate Moskowitz et al. (2015).
Nonetheless, 2D:4D ratios were positively associated with agree-
ableness among female participants. Women with larger ratios were
more agreeable towards other women than to men, while women with
smaller ratios did not show this contextual variation in agreeableness.
Following Moskowitz et al. (2015), this might mean that relatively low
prenatal exposure to testosterone increases the engagement of young
women with behaviors that indicate a willingness to invest in same-
gender relationships.
The diverging results of both studies might be due to several
aforementioned sample differences. Participants studied by Moskowitz
et al. (2015) had a mean age of 29 years and were employed and in a
long-term relationship. Participants in our study were students, on
average nearly 10 years younger, and mostly not in a relationship.
Being in a relationship has been associated with a reduced motivation
to be agreeable (Schultheiss et al., 2003), particularly in women (Suh
et al., 2004). Possibly, our female participants varied more in agree-
ableness than the women studied by Moskowitz et al. (2015).
Between-study differences in the use of hormonal contraceptives
(HCs) might also explain the results. Most of our female participants
used HCs. In the (Canadian) study by Moskowitz et al. (2015), this
percentage may have been lower (United Nations Department of
Economic and Social Affairs Population Division, 2019). HCs often
contain estrogen and may alter opposite-gender interactions (Little
et al., 2013). The 2D:4D ratio has previously been shown to predict HC-
associated side effects (Oinonen, 2009), suggesting that the impact of
HC use might differ between women with higher vs. lower 2D:4D ratios.
As HC use might have influenced the observed association between
female agreeableness and 2D:4D ratios, our study may be limited in that
we did not exclude HC users. However, neither did Moskowitz et al.
(2015). Another limitation of our study was the relatively small sample
size, resulting in relatively low statistical power (which prevented us
from examining HC use as a potential moderator). Relatedly, the low
proportion of male participants reduced the likelihood of finding
significant results in men. This is why the results for men should be
interpreted with caution. A final limitation was that the result on
agreeableness in women was larger for the right-hand 2D:4D ratio than
for the left-hand ratio. However, it has been argued that right-hand
ratios are better marker of prenatal sex hormone levels (Hönekopp &
Watson, 2010).
In conclusion, we did not replicate Moskowitz et al. (2015). None-
theless, we found that young women with larger 2D:4D ratios were
more agreeable towards other women than towards men, whereas this
was not seen among women with smaller ratios. While this result may
seem to contradict Moskowitz et al. (2015), it might also complement
their work. We highlighted several factors that might help explain both
studies' findings, including being in a (long-term) romantic relationship
and using HCs. These factors could be examined in future studies.
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.paid.2020.110379.
CRediT authorship contribution statement
Michelle Spix: Conceptualization, Validation, Writing - original
draft. Marije aan het Rot: Conceptualization, Methodology, Software,
Formal analysis, Writing - review & editing, Visualization, Supervision.
Declaration of competing interest
None.
Acknowledgements
The authors thank Renske Bosman, Minita Franzen, Sophie
Lammertink, Jana Uppendahl, and Lenka Wichmann for data collec-
tion.
References
Batalas, N., aan het Rot, M., Khan, V. J., & Markopoulos, P. (2018). Using TEMPEST: End-
user programming of web-based ecological momentary assessment protocols.
Proceedings of the ACM on Human-Computer Interaction, 2(EICS), 1–24. https://doi.
org/10.1145/3179428.
Fink, B., Manning, J. T., & Neave, N. (2004). Second to fourth digit ratio and the “big five”
personality factors. Personality and Individual Differences, 37(3), 495–503. https://doi.
org/10.1016/j.paid.2003.09.018.
Hönekopp, J., & Watson, S. (2010). Meta-analysis of digit ratio 2D:4D shows greater sex
difference in the right hand. American Journal of Human Biology, 22(5), 619–630.
https://doi.org/10.1002/ajhb.21054.
Hönekopp, J., & Watson, S. (2011). Meta-analysis of the relationship between digit-ratio
2D:4D and aggression. Personality and Individual Differences, 51(4), 381–386. https://
a. Right-hand ratio b. Left-hand ratio
Fig. 1. Interaction effect of 2D:4D ratio and partner gender for female agreeable behavior.
Note. *p< 0.05.
M. Spix and M. aan het Rot Personality and Individual Differences 168 (2021) 110379
3

doi.org/10.1016/j.paid.2010.05.003.
Little, A. C., Burriss, R. P., Petrie, M., Jones, B. C., & Roberts, S. C. (2013). Oral contra-
ceptive use in women changes preferences for male facial masculinity and is asso-
ciated with partner facial masculinity. Psychoneuroendocrinology, 38(9), 1777–1785.
https://doi.org/10.1016/j.psyneuen.2013.02.014.
Manning, J. T. (2011). Resolving the role of prenatal sex steroids in the development of
digit ratio. Proceedings of the National Academy of Sciences, 108(39), Article 16143.
https://doi.org/10.1073/pnas.1113312108.
McEwen, B. S., & Milner, T. A. (2017). Understanding the broad influence of sex hor-
mones and sex differences in the brain. Journal of Neuroscience Research, 95(1–2),
24–39. https://doi.org/10.1002/jnr.23809.
Moskowitz, D. S. (1994). Cross-situational generality and the interpersonal circumplex.
Journal of Personality and Social Psychology, 66(5), 921–933. https://doi.org/10.
1037//0022-3514.66.5.921.
Moskowitz, D. S., & Sadikaj, G. (2013). Event-contingent recording. In M. R. Mehl, & T. S.
Conner (Eds.). Handbook of research methods for studying daily life (pp. 160–175). New
York, NY: Guilford.
Moskowitz, D. S., Sutton, R., Zuroff, D. C., & Young, S. N. (2015). Fetal exposure to
androgens, as indicated by digit ratios (2D:4D), increases men’s agreeableness with
women. Personality and Individual Differences, 75, 97–101.
Oinonen, K. A. (2009). Putting a finger on potential predictors of oral contraceptive side
effects: 2D:4D and middle-phalangeal hair. Psychoneuroendocrinology, 34(5),
713–726. https://doi.org/10.1016/j.psyneuen.2008.11.009.
Schultheiss, O. C., Dargel, A., & Rohde, W. (2003). Implicit motives and gonadal steroid
hormones: Effects of menstrual cycle phase, oral contraceptive use, and relationship
status. Hormones and Behavior, 43(2), 293–301. https://doi.org/10.1016/S0018-
506X(03)00003-5.
Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater relia-
bility. Psychological Bulletin, 86(2), 420–428. https://doi.org/10.1037/0033-2909.86.
2.420.
Suh, E. J., Moskowitz, D. S., Fournier, M. A., & Zuroff, D. C. (2004). Gender and re-
lationships: Influences on agentic and communal behaviors. Personal Relationships,
11(1), 41–60. https://doi.org/10.1111/j.1475-6811.2004.00070.x.
United Nations Department of Economic and Social Affairs Population Division (2019).
Contraceptive Use by Method 2019: Data Booklet (Vol. ST/ESA/SER.A/435).
M. Spix and M. aan het Rot Personality and Individual Differences 168 (2021) 110379
4