Access to this full-text is provided by PLOS.
Content available from PLOS One
This content is subject to copyright.
RESEARCH ARTICLE
Cognitive reflection, 2D:4D and social value
orientation
Kobe MilletID*, Aylin Aydinli
Department of Marketing, Vrije Universiteit Amsterdam, Amsterdam, Netherlands
*kobe.millet@vu.nl
Abstract
The current study seeks confirmation for the hypothesis that 2D:4D (positively) predicts pro-
sociality when people are more likely to rely on intuition than deliberation. We assess intui-
tion and deliberation using the Cognitive Reflection Test (CRT) and measure prosociality via
the validated Social Value Orientation (SVO) slider measure. Although our results do not
provide collective evidence for our main proposition, we observe in the data that for low
(right) 2D:4D men, the more intuitive they are, the less prosocial they become, whereas for
high (right) 2D:4D men the thinking style does not affect their prosociality. Importantly, we
find that two alternative measures of cognitive reflection, CRT and CRT-2, differently relate
to prosocial decision making such that only CRT-2 (but not the classic CRT) positively pre-
dicts prosociality. Given that previous research on the role of cognitive reflection and 2D:4D
in prosocial decision making provided inconsistent results, the present study findings are
highly valuable to get a better understanding in this domain of study. Furthermore, some of
our findings invite further confirmatory tests, thereby opening up multiple avenues for further
research.
Introduction
Second to fourth digit ratio is the ratio of the index (2D) to ring (4D) finger, shortly referred to
as 2D:4D. It is a putative marker of prenatal exposure to testosterone [1,2] with a lower ratio
pointing towards exposure of higher testosterone levels during pregnancy. Direct evidence for
2D:4D as a biomarker for organizational effects of prenatal testosterone has been provided in
non-human mammals like rats [3] and mice [4,5] and much more (indirect) evidence in
humans has been provided as well. One of the most robust effects is the sexual dimorphism in
2D:4D with men having lower digit ratios than women [6], in line with the observation that
testosterone levels in amniotic fluid are higher for male than female fetuses [7,8]. At least, as
illustrated by the increasing number of publications in the last decades 2D:4D is commonly
accepted as an indirect biomarker for fetal testosterone exposure (despite a need for further
validation [9]).
Over the years, evidence has been provided for the relation between 2D:4D and a multitude
of personality traits, decisions and behavior, both in the lab and the field. For instance, 2D:4D
has been associated with sex role identity [10], depression [11], aggression [12], dominance
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 1 / 14
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Citation: Millet K, Aydinli A (2019) Cognitive
reflection, 2D:4D and social value orientation. PLoS
ONE 14(2): e0212767. https://doi.org/10.1371/
journal.pone.0212767
Editor: Valerio Capraro, Middlesex University,
UNITED KINGDOM
Received: July 17, 2018
Accepted: February 8, 2019
Published: February 22, 2019
Copyright: ©2019 Millet, Aydinli. This is an open
access article distributed under the terms of the
Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original
author and source are credited.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information
files.
Funding: The author received no specific funding
for this work.
Competing interests: The author has declared that
no competing interests exist.
[13,14], risk taking [15], social preferences [16,17], achievement in sports [18,19] and cogni-
tive tests [20], product preferences [21], gift giving [22], etc. However, there is uncertainty
about which relationships can be consistently replicated [23,24]. To our interest, the relation
between 2D:4D and social preferences has been doubted [25]. Given the paucity of clear evi-
dence in the field, it is important to continue the investigation of the association between
2D:4D and specific personality traits, decisions and behavior to get more insight into reliable
relationships.
One perspective that aims to clarify seemingly inconsistent results in 2D:4D research high-
lights the importance of the specific study context [23]. For instance, one interpretation of the
inconsistencies in research findings on the association between 2D:4D and risk taking, aggres-
sion and dominance focuses on the potential role of the status relevance of the behavior at
hand [26]. It has been suggested that some contexts (e.g. sports competition) entail status
enhancing behaviors which influence the relationship between 2D:4D and specific aggressive,
risky or dominant decisions [26]. At least, whereas previous research has typically focused on
direct relationships, a contextual perspective suggests that it may be fruitful to give more atten-
tion to potential interaction effects focusing on specific contextual characteristics that may
influence the relationship between 2D:4D and other variables. Therefore, the present research
focuses on prosocial decision making and investigates a potential moderator that is theoreti-
cally meaningful in studying the relationship between 2D:4D and prosociality.
Previous research on the effect of 2D:4D on prosociality yields mixed results. For instance,
low 2D:4D has been related to increased cooperation levels in public good [17] and dictator
[27] type of games. On the other hand, there is also some evidence for low 2D:4D and
decreased prosociality [28–30], albeit this may depend on the particular context of the decision
[23,28] or the incentivization of decisions [26]. Finally, in other type of games and contexts,
some null results and curvilinear relationships have been reported as well [16,31,32]. Given
this multitude of studies with different paradigms, it is still unclear how 2D:4D is actually
related to prosociality, suggesting the importance of studying potential moderators. As such,
we focus on a variable that may be of importance to understand the relation between 2D:4D
and prosociality: cognitive reflection (i.e. intuitive vs deliberate processing).
Cognitive reflection as a potential moderator between 2D:4D and
prosociality
Alonso and colleagues [33] observe that low 2D:4D is related to less generous and more selfish
behavior in a multiple-rounds dictator-type of economic game when people are already in a
disadvantaged position (i.e. worst paid agent) and when they score low on the cognitive reflec-
tion test (hereafter CRT). CRT, introduced by Frederick [34], consists of a set of numerical
problems that all have an intuitive, yet incorrect, answer that may be selected by those who do
not reflect carefully enough. Therefore, CRT scores reflect people’s ability to resist reliance on
intuition in favour of reliance on deliberation. The question arises whether low 2D:4D people
have a tendency to be less prosocial when they rely on intuitive (vs. deliberative) decision mak-
ing. At least, some findings in the literature appear to be in line with this proposition. Some
evidence shows that low male 2D:4D is only related to lower prosociality and increased selfish-
ness in the dictator position when aggression has been primed [27]. Interestingly, the mere
interaction of men with a gun [35] as well as exposure to an aggressive video [36] leads to an
increase of circulating testosterone levels and it has been shown that higher testosterone levels
decrease men’s performance on a cognitive reflection test [37]. Thus, the moderating effect of
the aggression prime may not necessarily be driven by the aggressive content of the prime, but
possibly by its effect on cognitive reflection. Further, a positive relationship has been shown
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 2 / 14
between young (resp. 11- and 6-to-9—year old) children’s 2D:4D and prosociality in different
resource allocation paradigms [28,29]. As it has been shown that young children have lower
impulse control than adults [38], this is again consistent with the idea that reliance on intuition
(vs. deliberation) may drive the relationship between low 2D:4D and selfish preferences. Over-
all these findings suggest that part of the inconsistencies in literature could be explained by the
differences in the extent to which people’s decisions are based on intuition vs. deliberation.
More specifically, we will test the hypothesis that 2D:4D positively predicts prosociality
when people rely on intuition, but not when they think more deliberately. That is, we predict
that the lower 2D:4D the less prosocial people become when they score low on cognitive reflec-
tion. In contrast, for people scoring high on cognitive reflection, we do not expect that 2D:4D
has any effect on prosociality. To do so, we will explore the interaction between 2D:4D and
cognitive reflection on prosociality, as measured by the social value orientation (SVO) slider
measure (see Method section below for a clarification of this specific measurement). Social
value orientation (SVO) is the prevalent conceptualization of social preferences in psychology
[39] and has originally been defined as a personal characteristic of how people interact in social
dilemmas [40]. With the original SVO measure, people are typically categorized as either pro-
social or proself. The slider measure we make use of is a relatively recently developed, more
finegrained, continuous measure of prosociality and opens opportunities to use as a dependent
variable [41].
Cognitive reflection and prosociality
While the main focus of our study is on the potential moderating role of cognitive reflection
on the relation between 2D:4D and prosociality, our findings may also provide insight into the
direct relationship between cognitive reflection and prosociality. On the one hand, previous
research suggests that prosociality is an intuitive response, whereas deliberation may lead to
more selfish decisions [42–45]. Some studies identified moderators for this effect showing pro-
sociality is an intuitive response in women (as opposed to men) [46] and for people with (as
opposed to without) experience in economic games [47]. Others again provide evidence that
deliberation leads to more selfish decisions in men (but not women) [48]. Still, it is unclear
if people really act in accordance with this so-called ‘social heuristic hypothesis’ [49]. On
the other hand, a reflective model of prosociality claims that we need to overcome our prepo-
tent, selfish impulses [45,50]. And again, there is consistent evidence with this perspective
showing, for example, that automatic, spontaneous reactions in one-shot anonymous prison-
er’s dilemma games tend to be egostic [51]. Overall, the discussion is not finished yet and
therefore it is interesting by itself to explore if there is a direct relationship (and if so, how this
relationship looks like) between cognitive reflection and our SVO measure.
In our empirical study, we adopt two measures of cognitive reflection, the 3-item classic
cognitive reflection test [34] as well as an alternate, recently developed, 4-item cognitive reflec-
tion test [52]. Whereas the original CRT has been widely adopted as a measurement of cogni-
tive reflection, it has recently been criticized as relying too much on numeracy skills [53] and
behaving differently than the CRT-2 [54]. By adopting two CRT measures, not only we are
able to test our main interaction hypothesis and the direct relationship between cognitive
reflection and prosociality separately for each of the measures, but we also gain more insight
into potential differences between CRT and CRT-2.
Finally, we adopted the extended version of the (non-incentivized) SVO slider measure
[42] as it is not only a validated measurement of prosociality, but also because the additional
items in the extended version are specifically set up to identify to what extent the aim to maxi-
mize collective outcomes (‘joint maximization’) vs. the aim to minimize differences between
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 3 / 14
oneself and the other (‘inequality aversion’) drive prosociality. By using this extended version,
the current study may provide the opportunity to explore potential relationships between
inequality aversion vs. joint gain maximization, 2D:4D and cognitive reflection. A recent study
[55] provides some evidence in a group context that low CRT scores are associated with indi-
viduals’ concerns for their relative shares (i.e. distribution of shares among group members)
and high CRT scores are associated with individuals’ concerns for social efficiency (i.e. total
level of group resources). Following these results we would expect that increased cognitive
reflection is positively related to the maximization of joint gains and negatively related to
inequality aversion.
Method
Two hundred eighteen participants between 18 and 31 years of age took part in the study (127
men, 91 women) of which 167 in return for course credit and 51 in return for a hedonic food
reward (chocolate or pringles). All subjects gave informed consent in accordance with the Dec-
laration of Helsinki and the study was approved by the School of Business and Economics
Research Ethics Review Board (Vrije Universiteit Amsterdam). The study took place in the
experimental lab of the School of Business and Economics where every participant was
assigned to a computer in a partially enclosed carrel in which they could not see one another
and could not talk. A maximum of 14 students participated at the same time.
First, participants received 11 questions to answer, seven of which were the test items of
two cognitive reflection tests: the 3-item CRT [34] and 4-item CRT-2 [52]. The CRT items
were intermixed in random order and the remaining four decoy questions [52] were presented
at fixed positions inbetween (2
nd
; 4
th
; 7
th
; 9
th
). For each subject, two separate total general cog-
nitive reflection scores were computed by summing up the relevant items of the CRT and
CRT-2. A total general cognitive reflection score ranges from 0 (intuitive decision making) to
3 (deliberate decision making) for the 3-item CRT and from 0 (intuitive decision making) to 4
(deliberate decision making) for the 4-item CRT-2. We measured the CRT items first as it has
been showed that students perform better in the beginning than at the end [56]. Thereafter,
participants completed the extended (15-item) Social Value Orientation Slider Measure [39,
41]. This measure exists of 15 different resource allocation dilemmas in which participants
indicate how they would distribute monetary resources between themselves and another anon-
ymous person. The responses to the six primary items determine a participant’s SVO angle,
where a higher value indicates higher pro-sociality. The main analyses will adopt the continu-
ous SVO angle as the dependent measure. We discuss the remaining secondary items of the
SVO slider measure in the supplementary materials.
Finally, both left and right hands were scanned to measure finger lengths. Participants
placed their hand palms on the glass plate of a scanner and finger lengths were measured by
two independent raters (by means of Adobe Photoshop) from the ventral proximal crease to
the tip of the finger. When there was a band of creases at the base of the digit finger lengths
were measured from the bottom crease. Both raters were asked to indicate any unclarity for
each specific finger measurement (i.e. resp. left and right index and ring fingers). After they
measured all fingers, both raters were requested to measure the right or left index and ring fin-
gers again (a) when one of the raters gave a comment on a specific measurement and/or (b)
the difference in 2D:4D between both raters was larger than.04. Moreover, raters were asked to
provide comments on each of the measurements they were asked to measure again. If deemed
necessary, a rater could advise not to include a specific 2D:4D measurement (because the
creases were not clear, etc.). If one of the raters recommended not to include a specific data
point, the participant was (only) excluded from the specific analysis in which the ratio was
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 4 / 14
included. Further, data points were also excluded of those ratios in which one of the raters
indicated it was not possible to clearly measure one of both finger lengths. Consequently, 8
data points were excluded from the data analyses that included left 2D:4D and 12 data points
were excluded from the analyses that included right 2D:4D. Assessments of both raters were
highly correlated (resp. r
left 2D:4D
= .93 and r
right 2D:4D
= .87), speaking towards the accuracy of
the measurements. Therefore, we averaged both assessments for left 2D:4D as well as for right
2D:4D to attain single left and right 2D:4D measures. Only these averaged 2D:4Ds are used in
the statistical analyses. Consistent with the prior literature, we treat left and right 2D:4D mea-
sures separately in our subsequent statistical analyses and focus on right 2D:4D in our main
text as this has been considered a better indicator of prenatal androgenisation than left-hand
2D:4D [6]. For the sake of completeness, we include results pertaining to left 2D:4D in the sup-
plementary materials (see S1 Table).
Results
We first explored sex differences (see Table 1) in (a) left and right 2D:4D (b) both CRT and
CRT-2 and (c) SVO angle. In line with previous literature, men have a lower 2D:4D [6], score
higher on the CRT [34,52] and turn out to be less pro-social than women [57]. We further rep-
licate the findings indicating that there is no sex difference on the CRT-2 [52] and that the
right 2D:4D sex difference is larger than the left 2D:4D sex difference [6]. Next, we performed
simple correlation analyses in men and women separately between left 2D:4D, right 2D:4D,
CRT, CRT-2, and SVO angle (see Table 2a and 2b). Interestingly, CRT and CRT-2 tend to
differ in their relation with the general SVO angle measurement. A correlational analysis
(irrespective of gender) indicates a positive relationship between SVO angle and the CRT-2
(r= .133; p= .050), but not the CRT (r= -.031, p= .649).
To test our main hypothesis, we regressed SVO angle on right 2D:4D (mean-centered), cog-
nitive reflection score (mean-centered) and the corresponding interaction. Table 3 reports
OLS regression results for SVO angle, disaggregated for (a) CRT-2 and CRT and (b) men and
women.
For men, the regression shows a positive effect of CRT-2 (β= 2.42, p =.054), but no signifi-
cant effect of right 2D:4D (p = .789). Specifically, prosociality in men increases with greater
levels of cognitive reflection. Importantly, the interaction term between the CRT-2 and right
2D:4D approaches marginal levels of significance (β= -67.08, p = .106). Therefore, we further
explored the nature of this interaction via what is known as spotlight analysis or simple slopes
analysis [58,59]. Specifically, we estimated and tested the conditional effect of 2D:4D at differ-
ent levels of CRT-2 and the conditional effect of CRT-2 at different levels of 2D:4D We fol-
lowed the convention of testing conditional effects at plus or minus one standard deviation
Table 1. Sex differences on the different measurements.
Males Females Sex Difference
N M SD N M SD t p
Right 2D:4D 121 .9574 .0289 85 .9720 .0329 3.35 .001
Left 2D:4D 123 .9617 .0267 87 .9698 .0283 2.09 .038
CRT 127 1.91 1.04 91 1.31 1.08 4.13 .0001
CRT-2 127 2.49 1.04 91 2.29 .99 1.45 .149
SVO angle 127 22.47 14.17 91 27.58 11.15 2.88
a
.003
a
as variances are not equal, equal variances are not assumed in the t-test
https://doi.org/10.1371/journal.pone.0212767.t001
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 5 / 14
from the sample mean of an interacting variable [60]. Comparing the effect of 2D:4D on SVO
angle at one standard deviation below and above the mean level of CRT-2 did not reveal any
significant effects. However, the effect of CRT-2 on SVO angle varied at one standard devia-
tion below and above the mean level of right 2D:4D. The results show a significant positive
effect of CRT-2 for male participants with low right 2D:4D (β= 4.36, p = .017), but not for
those with high right 2D:4D (β= .48, p = .768; see Fig 1). In other words, for male participants
with low 2D:4D, the higher (lower) the cognitive reflection, the more (less) prosocial behavior
they exhibit. However, for male participants with high 2D:4D, cognitive reflection did not
Table 2. Correlations between right 2D:4D, left 2D:4D, CRT, CRT-2, and SVO angle in men and women.
Correlations
Males Females
1 2 3 4 1 2 3 4
1. Right 2D:4D
2. Left 2D:4D .702
c
.730
c
(n = 118) (n = 83)
3. CRT .038 -.046 .034 -.035
(n = 121) (n = 123) (n = 85) (n = 87)
4. CRT-2 -.078 -.112 .531
c
-.100 .011 .466
(n = 121) (n = 123) (n = 127) (n = 85) (n = 87) (n = 91)
5. SVO angle -.010 .036 -.012 .157
a
-.004 -.026 .080 .153
(n = 121) (n = 123) (n = 127) (n = 127) (n = 85) (n = 87) (n = 91) (n = 91)
Notes:
a
p<.10;
b
p<.05;
c
p<.001
https://doi.org/10.1371/journal.pone.0212767.t002
Table 3. SVO angle as a function of 2D:4D and cognitive reflection.
CRT-2 (4 items)
(male) (female)
Est SE p-value Est SE p-value
2D:4D -11.925 44.660 0.790 3.152 38.392 0.935
CRT-2 2.422 1.245 0.054 1.483 1.263 0.244
2D:4D�CRT-2 -67.078 41.144 0.106 -0.789 35.952 0.983
Constant 22.599 1.265 0.000 27.282 1.249 0.000
N 121 85
R2 4.79% 1.75%
CRT (3 items)
(male) (female)
Est SE p-value Est SE p-value
2D:4D -8.053 45.062 0.859 -3.877 38.019 0.919
CRT -1.218 1.283 0.344 0.668 1.150 0.563
2D:4D�CRT -34.310 44.982 0.447 39.574 31.303 0.210
Constant 22.792 1.286 0.000 27.238 1.240 0.000
N 121 85
R2 1.12% 2.25%
https://doi.org/10.1371/journal.pone.0212767.t003
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 6 / 14
affect their behavior. Interestingly, a similar regression analysis including the original CRT
measure does not reveal any significant effects (all ps >.344; see Fig 2). For women, both
regression analyses reveal non-significant results, either involving CRT-2 (all ps >.244) or
CRT (all ps >.209).
As a follow up on these results, we ran further regression analyses in order to explore the
possibility that the effect of cognitive reflection on prosocial behavior changes with the CRT
measure being used. Thus, we regressed SVO angle on CRT (mean-centered), gender (contrast
coded as -1 = male and 1 = female) and the corresponding interaction. Table 4 reports the
results for SVO angle, disaggregated for CRT-2 and CRT.
In both cases, the regression shows a positive effect of gender (β= 2.75, p =.002 and
β= 2.68, p =.005, respectively), suggesting that women are more prosocial than men. More
Fig 1. SVO as a function of 2D:4D and CRT-2 (Right-hand male sample).
https://doi.org/10.1371/journal.pone.0212767.g001
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 7 / 14
interestingly, we observe that CRT-2 significantly predicts prosociality (β= 1.97, p =.023)
whereas CRT does not have any significant effect on prosocial behavior (β= .25, p =.768).
The interaction between cognitive reflection and gender is not significant in both cases. These
results corroborate our previous findings in the sense that CRT-2 and CRT differently relate to
prosociality. The observed difference between the two measures of cognitive reflection is in
line with previous research showing that CRT-2 is a better predictor of behavior than CRT is
[54].
Discussion
The current study is of importance for the growing literature on the biological foundations of
prosocial behavior. Given the mixed findings currently reported in literature on how 2D:4D
Fig 2. SVO as a function of 2D:4D and CRT (Right-hand male sample).
https://doi.org/10.1371/journal.pone.0212767.g002
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 8 / 14
relates to prosocial behavior, more systematic investigation is needed to understand the
relationship between 2D:4D and prosociality. Moreover, it is at least as important to focus on
theoretically plausible moderators and to set up specific studies to understand if specific associ-
ations with 2D:4D emerge under particular circumstances. Therefore, the present study tested
the possible moderating effect of cognitive reflection (a marker for intuitive vs. deliberative
decision making) on the relationship between 2D:4D and prosociality. Further, the study pro-
cedure also allowed us to explore the potential direct relationship between cognitive reflection
(using two separate measures, CRT and CRT-2) and prosociality (using the validated SVO
slider measure). Speaking towards the reliability of our findings, we replicate previous findings
showing (a) a sex difference in social value orientation [56], (b) a sex difference in performance
on the classic CRT, which is attenuated for the CRT-2 [52] (c) a stronger sex difference in right
2D:4D than in left 2D:4D [6].
Our confirmatory analysis did not provide collective evidence for a moderating effect of
cognitive reflection on the relationship between 2D:4D and SVO. This raises the question
whether previously observed relationships between 2D:4D and social preferences or cognitive
reflection are just type I errors or if the relationships that are reported in literature depend on
other (omitted) variables. At least, null results are often not written up and therefore much
(absence of) evidence is unobserved [61], despite the potential detrimental impact on scientific
progress in the particular domain of interest. However, as suggested before, it is important to
realize that different contextual factors may be of importance for a relationship with 2D:4D to
occur. Recently, the findings on the relations reported between 2D:4D and risk taking, domi-
nance and aggression were re-interpreted from the perspective that the particular behavior at
hand in a specific context needs to be status relevant for a relationship to arise [26]. Therefore,
the status relevance of the context at hand may be one interesting avenue for future research to
explore relations between 2D:4D and prosociality.
Despite a lack of collective evidence with regard to our main hypothesis that cognitive
reflection moderates the relationship between 2D:4D and prosociality, the interaction between
right 2D:4D and CRT-2 on prosociality in men approaching significance yields further
insights. Specifically, the effect of CRT-2 on prosociality in high and low 2D:4D men differs
such that for low 2D:4D men, the more intuitive they are, the less prosocial they become,
Table 4. Social value orientation as a function of cognitive reflection and gender.
CRT-2 (4 items)
Est SE p-value
CRT-2 1.969 0.862 0.023
Gender 2.746 0.891 0.002
CRT-2�Gender -0.211 0.881 0.811
Constant 24.582 0.878 0.000
N 218
R2 5.99%
CRT (3 items)
Est SE p-value
CRT 0.249 0.842 0.768
Gender 2.677 0.933 0.005
CRT�Gender 0.498 0.848 0.558
Constant 24.747 0.917 0.000
N 218
R2 3.85%
https://doi.org/10.1371/journal.pone.0212767.t004
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 9 / 14
whereas for high 2D:4D men the amount of cognitive reflection does not affect their prosocial-
ity. This observation is consistent with our main hypothesis that low 2D:4D people may
become less prosocial or more selfish when they make intuitive decisions. As this specific find-
ing awaits confirmation in a follow-up research, caution is needed in the interpretation of this
tentative result. That said, it is remarkable to see that this effect only seems to emerge (a) with
the CRT-2 measure (b) in right 2D:4D measurements and (c) in men. If reliable, the question
arises why we only found (at least part of) our expected pattern of results under these specific
conditions. First, it is important to be aware that the CRT-2 measure is less reliant on numer-
acy than the original CRT [52]. The original CRT to some extent has been criticized to reflect
numerical skills than deliberation [53]. Therefore, given that numerical skills by themselves
may influence how people respond in a ‘numeric’ resource allocation paradigm such as SVO,
this may have weakened potential relationship between the classic CRT and SVO. Further-
more, recent research provides evidence that CRT-2 behaves more in line with other measures
of deliberative thinking and has superior predictive value over CRT [54]. Second, given that
the gender difference is typically larger in the right than left 2D:4D, right 2D:4D has been con-
sidered a better indicator of prenatal androgenisation [6]. If anything, our results are at least
consistent with this consideration.
In line with our pattern of results, different studies focusing on dominance-related out-
comes observe stronger effects in men then women (e.g. relations between aggression and
2D:4D have been reported to be stronger in men than women [12,62–64]). Considering that
our prosociality measure is about the division of limited resources between oneself and
another, one may consider the selfish behavior as an expression of dispositional dominance. In
that case, the present data could suggest that the relation between dispositional dominance
and 2D:4D in men may not necessarily arise because of the activation of the dominance system
[14], but rather because of a reliance on intuition in challenge situations. However, as this sug-
gestion is highly speculative, it is worthwile to first provide stronger evidence for this modera-
tion between intuition and male 2D:4D on prosociality in a confirmatory study before
exploring this avenue further.
Finally, it is important to be aware that our prosociality measurement is not incentivized,
which may explain the collective lack of evidence for our main hypothesis. Whereas incentivi-
zation seems to be crucial to observe relationships between 2D:4D and financial risk taking
[15], the same may hold for prosocial decision making. However, the literature on 2D:4D and
prosocial decision making does not hint at a similar moderating role of incentivization on
potential relationships between prosociality and 2D:4D. Moreover, in line with both our tenta-
tive results in men as well as the reasoning developed above, 2D:4D may not predict risk taking
with real monetary incentives because of its potential status relevance (cf. [26]), but rather
because of its increased reliance on intuition when taking specific riskful decisions with real
(compared to fictitious) financial consequences. Again, this hypothesis awaits empirical testing
and is a potential avenue for further research.
When we focus on the direct relationship between cognitive reflection and prosociality, we
observe that CRT-2 (but not the CRT) positively predicts prosociality. Given that CRT-2 is
likely to be a more reliable measurement of deliberative thinking (see above) our findings
seem to be in support of a reflective, but not a heuristic, model of prosociality.
Summarized, our study highlights how important it remains to re-interpret many 2D:4D
findings from a contextual perspective and to test and report both hypotheses and results irre-
spective of significance levels. While it is important to be aware that we did not provide support-
ive evidence in our confirmatory test of our main hypothesis, this did not prohibit us to explore
the data further and provide a post hoc interpretation of the patterns of results. We find this a
fruitful approach as long as this is clearly communicated in the paper. As such, we followed this
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 10 / 14
approach and clearly distinguished between confirmatory and post hoc, exploratory research
findings. In the end, this allows us to get more reliable insights into which factors may be more
or less likely to be of importance in studying the relationship between 2D:4D and prosociality
(or any other behavior) as only then the true role of 2D:4D in (economic) decision making can
be discovered. Furthermore, our study results point to a positive association between delibera-
tion and prosociality. In this regard, our findings suggest that it remains important to further
examine the effect of cognitive deliberation on prosocial decision making in different settings
and with different operationalizations of the constructs at hand. Finally, we would like to allo-
cate some attention to the fact that the literature on 2D:4D has suffered from interpretations of
findings that are limited to one hand, but are not replicated in the other hand. If anything, the
more pronounced sex difference in right 2D:4D provides support to focus on only right 2D:4D
(and to make that choice before the start of the study) or if focusing on both measurements, the
observed effects should be replicated in both hands (and findings always reported if this choice
had been made before the start of the study). Still, more direct replication studies are needed to
test the robustness of these results regardless of which hand they pertain to. We speculate that
right 2D:4D effects will be more easily replicated than left 2D:4D effects.
Conclusions
In general, the current study does not provide evidence for the main hypothesis that the rela-
tionship between 2D:4D and social preferences may be influenced by people’s cognitive style.
At the same time we urge for a confirmatory test of our observed pattern of results in men. At
least, previous findings on sex differences in (a) right vs left 2D:4D, (b) CRT vs CRT-2 and (c)
SVO are replicated and therefore speak towards the reliability of the results. Given the multi-
tude of inconsistent findings and the omission of potentially important variables in the study
on the connection between 2D:4D and social preferences, it is important to report studies that
focus on both direct and moderating effects, irrespective of the significance of results. There-
fore, the current study is of importance to the field. Finally, the relationship between CRT-2
and prosociality deserves further attention in future research.
Supporting information
S1 File. Supplementary materials.
(DOCX)
S2 File. The supporting dataset file.
(SAV)
S1 Table. Social value orientation as a function of 2D:4D and cognitive reflection (Left-
hand sample).
(DOCX)
Acknowledgments
We thank the Behavioral Lab of the School of Business and Economics (Vrije Universiteit
Amsterdam) for data collection support.
Author Contributions
Conceptualization: Kobe Millet, Aylin Aydinli.
Data curation: Kobe Millet, Aylin Aydinli.
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 11 / 14
Formal analysis: Kobe Millet, Aylin Aydinli.
Funding acquisition: Kobe Millet.
Investigation: Kobe Millet.
Methodology: Kobe Millet, Aylin Aydinli.
Project administration: Kobe Millet.
Resources: Kobe Millet.
Writing – original draft: Kobe Millet, Aylin Aydinli.
Writing – review & editing: Kobe Millet, Aylin Aydinli.
References
1. Manning JT, Scutt D, Wilson J, Lewis-Jones DI. The ratio of 2nd to 4th digit length: a predictor of sperm
numbers and concentrations of testosterone, luteinizing hormone and oestrogen. Hum Reprod. 1998;
13(11): 3000–3004. PMID: 9853845
2. Manning JT. Digit ratio: A pointer to fertility, behavior, and health. New Brunswick: Rutgers University
Press; 2002.
3. Zheng Z, Cohn MJ. Developmental basis of sexually dimorphic digit ratios. Proc Natl Acad Sci. 2011;
108(39): 16289–16294. https://doi.org/10.1073/pnas.1108312108 PMID: 21896736
4. Talarovičova
´A, Krs
ˇkova
´L, Blaz
ˇekova
´L. Testosterone enhancement during pregnancy influences the
2D:4D ratio and open field motor activity of rat siblings in adulthood. Horm Behav. 2009; 55(1): 235–
239. https://doi.org/10.1016/j.yhbeh.2008.10.010 PMID: 19022257
5. Auger J, Le Denmat D, Berges R, Doridot L, Salmon B, Canivenc-Lavier MC, Eustache F. Environmen-
tal levels of oestrogenic and anti-androgenic compounds feminize digit ratios in male rats and their
unexposed male progeny. Proc Biol Sci. 2013; 280(1768): 20131532. https://doi.org/10.1098/rspb.
2013.1532
6. Ho
¨nekopp J, Watson S. Meta-analysis of digit ratio 2D: 4D shows greater sex difference in the right
hand. Am J Hum Biol. 2010; 22(5): 619–630. https://doi.org/10.1002/ajhb.21054
7. Auyeung B, Baron-Cohen S, Ashwin E, Knickmeyer R, Taylor K, Hackett G, Hines M. Fetal testosterone
predicts sexually differentiated childhood behavior in girls and in boys. Psychol Sci. 2009; 20(2): 144–
148. https://doi.org/10.1111/j.1467-9280.2009.02279.x PMID: 19175758
8. Gitau R, Adams D, Fisk NM, Glover V. Fetal plasma testosterone correlates positively with cortisol.
Arch Dis Child Fetal Neontal. 2005; 90(2): F166–F169.
9. Voracek M. No effects of androgen receptor gene CAG and GGC repeat polymorphisms on digit ratio
(2D: 4D): a comprehensive meta-analysis and critical evaluation of research. Evol Hum Beh. 2014;
35(5): 430–437.
10. Csatho
´A, Osva
´th A, Bicsa
´k E, Kara
´di K, Manning JT, Ka
´llai J. Sex role identity related to the ratio of
second to fourth digit length in women. Biol Psychol. 2003; 62(2): 147–156. PMID: 12581689
11. Bailey AA, Hurd PL. Depression in men is associated with more feminine finger length ratios. Pers Indi-
vid Diff. 2005; 39(4): 829–836.
12. Bailey AA, Hurd PL. Finger length ratio (2D: 4D) correlates with physical aggression in men but not in
women. Biol Psychol. 2005; 68(3): 215–222. https://doi.org/10.1016/j.biopsycho.2004.05.001 PMID:
15620791
13. Manning JT, Fink B. Digit ratio (2D: 4D), dominance, reproductive success, asymmetry, and sociosexu-
ality in the BBC Internet Study. Am J Hum Biol. 2008; 20(4): 451–461. https://doi.org/10.1002/ajhb.
20767 PMID: 18433004
14. Ryckmans J, Millet K, Warlop L. The influence of facial characteristics on the relation between male
2D:4D and dominance. PloS ONE. 2015; 10(11): e0143307. https://doi.org/10.1371/journal.pone.
0143307 PMID: 26600255
15. Branas-Garza P, Galizzi M, Nieboer J. Experimental and self-reported measures of risk taking and digit
ratio (2D: 4D): evidence from a large, systematic study. Int Econ Rev. 2018; 3: 1131–1157.
16. Branas-Garza P, Kova
´řı
´k J, Neyse L. Second-to-fourth digit ratio has a non-monotonic impact on altru-
ism. PLoS ONE. 2013; 8(4): e60419. https://doi.org/10.1371/journal.pone.0060419 PMID: 23593214
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 12 / 14
17. Millet K, Dewitte S. Second to fourth digit ratio and cooperative behavior. Biol Psychol. 2006; 71(1):
111–115. https://doi.org/10.1016/j.biopsycho.2005.06.001 PMID: 16024156
18. Tester N, Campbell A. Sporting achievement: what is the contribution of digit ratio? J Pers. 2007; 75(4):
663–678. https://doi.org/10.1111/j.1467-6494.2007.00452.x PMID: 17576354
19. Ho
¨nekopp J, Schuster M. A meta-analysis on 2D:4D and athletic prowess: substantial relationships but
neither hand out-predicts the other. Pers Individ Diff. 2010; 48(1): 4–10.
20. Bosch-Domènech A, Brañas-Garza P, Espı
´n AM. Can exposure to prenatal sex hormones (2D: 4D)
predict cognitive reflection? Psychoneuroendocrinology. 2014; 43(1): 1–10.
21. Aspara J, Van Den Bergh B. Naturally designed for masculinity vs. femininity? Prenatal testosterone
predicts male consumers’ choices of gender-imaged products. Int J Res Mark. 2014; 31(1): 17–121.
22. Nepomuceno MV, Saad G, Stenstrom E, Mendenhall Z, Iglesias F. Testosterone & gift-giving: Mating
confidence moderates the association between digit ratios (2D: 4D and rel2) and erotic gift-giving. Pers
Individ Diff. 2016; 91: 27–30.
23. Millet K. An interactionist perspective on the relation between 2D: 4D and behavior: An overview of
(moderated) relationships between 2D: 4D and economic decision making. Pers Individ Diff. 2011;
51(4): 397–401.
24. Voracek M, Loibl LM. Scientometric analysis and bibliography of digit ratio (2D: 4D) research, 1998–
2008. Psychol Rep. 2009; 104(3): 922–956. https://doi.org/10.2466/PR0.104.3.922-956
25. Branas-Garza P, Espin A, Garcia T, Kovarik J. Digit ratio (2D:4D) predicts pro-social behavior in eco-
nomic games only for unsatisfied individuals. MPRA paper. 2018. https://mpra.ub.uni-muenchen.de/
89142/8/MPRA_paper_89135.pdf
26. Millet K, Buehler F. A Context dependent interpretation of inconsistencies in 2D: 4D findings: The mod-
erating role of status relevance. Front Behav Neurosci. 2018; 11: 254. https://doi.org/10.3389/fnbeh.
2017.00254 PMID: 29403365
27. Millet K, Dewitte S. The presence of aggression cues inverts the relation between digit ratio (2D: 4D)
and prosocial behaviour in a dictator game. Br. J. Psychol. 2009; 100(1): 151–162.
28. Cecchi F, Duchoslav J. The effect of prenatal stress on cooperation: Evidence from violent conflict in
Uganda. Eur. Econ. Rev. 2018; 101: 35–56.
29. Horn L, Hungerla
¨nder NA, Windhager S, Bugnyar T, Massen JJ. Social status and prenatal testosterone
exposure assessed via second-to-fourth digit ratio affect 6–9-year-old children’s prosocial choices. Sci-
entific Reports. 2018; 8(1), 9198. https://doi.org/10.1038/s41598-018-27468-0
30. Kaltwasser L, Mikac U, Bus
ˇko V, Hildebrandt A. No robust association between static markers of testos-
terone and facets of socio-economic decision making. Front Behav Neurosci. 2017; 11: 250. https://
doi.org/10.3389/fnbeh.2017.00250 PMID: 29326567
31. Galizzi MM, Nieboer J. Digit ratio (2D: 4D) and altruism: evidence from a large, multi-ethnic sample.
Front Behav Neurosci. 2015; 9: 41. https://doi.org/10.3389/fnbeh.2015.00041 PMID: 25755639
32. Sanchez-Pages S, Turiegano E. Testosterone, facial symmetry and cooperation in the prisoners’
dilemma. Physiol Behav. 2010; 99(3): 355–361. https://doi.org/10.1016/j.physbeh.2009.11.013 PMID:
19954750
33. Alonso J, Di Paolo R, Ponti G, Sartarelli M. Facts and misconceptions about 2D: 4D, social and risk pref-
erences. Front Behav Neurosci. 2018; 12: 22. https://doi.org/10.3389/fnbeh.2018.00022 PMID:
29487510
34. Frederick S. Cognitive reflection and decisionmaking. J Econ Perspect. 2005; 19(4): 25–42.
35. Klinesmith J, Kasser T, McAndrew FT. Guns, testosterone, and aggression: an experimental test of a
mediational hypothesis. Psychol Sci. 2006; 17(7): 568–571. https://doi.org/10.1111/j.1467-9280.2006.
01745.x
36. Kilduff LP, Hopp RN, Cook CJ, Crewther BT, Manning JT. Digit ratio (2D: 4D), aggression, and testos-
terone in men exposed to an aggressive video stimulus. Evol Psychol. 2013; 11(5): 953–964. PMID:
24113579
37. Nave G, Nadler A, Zava D, Camerer C. Single-dose testosterone administration impairs cognitive
reflection in men. Psychol Sci. 2017; 28(10): 1398–1407. https://doi.org/10.1177/0956797617709592
PMID: 28771393
38. Steinberg L, Albert D, Cauffman E, Banich M, Graham S, Woolard J. Age differences in sensation seek-
ing and impulsivity as indexed by behavior and self-report: evidence for a dual systems model. Dev Psy-
chol. 2008; 44(6): 1764–1778. https://doi.org/10.1037/a0012955
39. Murphy RO, Ackermann KA. Social value orientation: Theoretical and measurement issues in the study
of social preferences. Pers Soc Psychol Rev. 2014; 18(1): 13–41. https://doi.org/10.1177/
1088868313501745 PMID: 24065346
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 13 / 14
40. Van Lange PAM. The pursuit of joint outcomes and equality in outcomes: an integrative model of social
value orientation. J. Pers. Soc. Psychol. 1999; 77(2): 337–349.
41. Murphy RO, Ackermann KA, Handgraaf M. Measuring social value orientation. Judgm Decis Mak.
2011; 6(8): 771–781.
42. Rand DG, Greene JD, Nowak MA. Spontaneous giving and calculated greed. Nature. 2012; 489: 427–
430. https://doi.org/10.1038/nature11467 PMID: 22996558
43. Rand DG. Cooperation, fast and slow: Meta-analytic evidence for a theory of social heuristics and self-
interested deliberation. Psychol Sci. 2016; 27: 1192–1207. https://doi.org/10.1177/0956797616654455
PMID: 27422875
44. Rand DG, Peysakhovich A, Kraft-Todd GT, Newman GE, Wurzbacher O, Nowak MA, Greene JD.
Social heuristics shape intuitive cooperation. Nat Commun. 2014; 5: 3677. https://doi.org/10.1038/
ncomms4677 PMID: 24751464
45. Zaki J, Mitchell JP. Intuitive Prosociality. Curr Dir Psychol Sci. 2013; 22: 466–470.
46. Rand DG, Brescoll VL, Everett JAC, Capraro V, Barcelo H. Social heuristics and social roles: Intuition
favors altruism for women but not for men. J Exp Psychol Gen. 2016; 145: 389–396. https://doi.org/10.
1037/xge0000154 PMID: 26913619
47. Capraro V, Cococcioni G. Social setting, intuition, and experience in laboratory experiments interact to
shape cooperative decision-making. Proc R Soc Lond B Biol Sci. 2015; 282: 20150237.
48. Espinosa MP, Kovarik J. Prosocial behavior and gender. Front Behav Neurosci. 2015; 9: 88. https://
doi.org/10.3389/fnbeh.2015.00088 PMID: 25926783
49. Bouwmeester S, Verkoeijen PPJL, Aczel B, Barbosa F, Begue L, Branas-Garza TGH, et al. Registered
Replication Report: Rand Greene & Nowak (2012). Perspect Psychol Sci. 2017; 12: 527–542. https://
doi.org/10.1177/1745691617693624
50. DeWall CN, Baumeister RF, Gailliot MT, Maner JK. Depletion makes the heart grow less helpful: Help-
ing as a function of self-regulatory energy and genetic relatedness. Pers Soc Psych Bull. 2008; 34:
1653–1662.
51. Capraro V, Cococcioni G. Rethinking spontaneous giving: Extreme time pressure and ego-depletion
favor self-regarding reactions. Sci Rep. 2016; 6: 27219. https://doi.org/10.1038/srep27219
52. Thomson KS, Oppenheimer DM. Investigating an alternate form of the cognitive reflection test. Judgm
Decis Mak. 2016; 11(1): 99–113.
53. Sinayev A, Peters E. Cognitive reflection vs. calculation in decision making. Front Psychol. 2015; 6:
532. https://doi.org/10.3389/fpsyg.2015.00532 PMID: 25999877
54. Yilmaz O, Saribay SA. The relationship between cognitive style and political orientation depends on the
measures used. Judgm Decis Mak. 2017; 12(2): 140–147.
55. Capraro V, Corgnet B, Espı
´n AM, Herna
´n-Gonza
´lez R (2017) Deliberation favours social efficiency by
making people disregards their relative shares: Evidence from USA and India. R Soc Open Sci. 2017;
4: 160605. https://doi.org/10.1098/rsos.160605
56. Branas-Garza P, Kujal P, Lenkeim B. Cognitive reflection test: Whom, how, when. ESI Working Paper.
2015. 15–25. http://digitalcommons.chapman.edu/esi_working_papers/174
57. Grosch K, Rau HA. Gender differences in honesty: The role of social value orientation. J Econ Psychol.
2017; 62: 258–267.
58. Spiller SA, Fitzsimons GJ, Lynch J, McClelland G. Spotlights, floodlights, and the magic number zero:
Simple effect tests in moderated regression. J Mark Res. 2012; 50(2): 277–288.
59. Hayes AF. Introduction to Mediation, Moderation, and Conditional Process Analysis. A Regression-
Based Approach( 2nd edition): Guilford Press; 2018.
60. Aiken LS, West SG. Multiple regression: Testing and interpreting interactions. Newbury Park, CA:
Sage; 1991.
61. Franco A, Malhotra N, Simonovits G. Publication bias in the social sciences: Unlocking the file drawer.
Science. 2014; 345(6203): 1502–1505. https://doi.org/10.1126/science.1255484
62. Hoskin AW, Meldrum RC. The association between fetal testosterone and violent behavior: Additional
evidence using the 2D:4D digit ratio. Pers Individ Diff. 2018; 134: 293–297.
63. Butovskaya M, Fedenok J, Burkova V, Manning J. Sex differences in 2D:4D and aggression in children
and adolescents from five regions of Russia. Am J Phys Antr. 2013; 152(1): 130–139.
64. Ho
¨nekopp J, Watson S. Meta-analysis of the relationship between digit-ratio 2D:4D and aggression.
Pers Individ Diff. 2011; 51(4): 381–386.
Cognitive reflection, 2D:4D and social value orientation
PLOS ONE | https://doi.org/10.1371/journal.pone.0212767 February 22, 2019 14 / 14
Available via license: CC BY
Content may be subject to copyright.
