Sei sulla pagina 1di 5

YHBEH-03421; No.

of pages: 5; 4C:
Hormones and Behavior xxx (2012) xxxxxx

Contents lists available at SciVerse ScienceDirect

Hormones and Behavior


journal homepage: www.elsevier.com/locate/yhbeh

Men's preference for the ovulating female is triggered by subtle face


shape differences
Cora Bobst a, Janek S. Lobmaier a, b,
a
Institute of Psychology, University of Bern, Muesmattstrasse 45, 3012 Bern, Switzerland
b
Center for Cognition, Learning, and Memory, University of Bern, Muesmattstrasse 45, 3012 Bern, Switzerland

a r t i c l e i n f o a b s t r a c t

Article history: Recent studies have revealed that there may be perceptible cues to ovulation in humans. This study aims at
Received 13 April 2012 extending these ndings by using female faces that were shape transformed towards a late follicular (fertile)
Revised 17 July 2012 and a luteal (non-fertile) prototype. Fertile prototypes were created by averaging 25 photographs taken of
Accepted 20 July 2012
females during ovulation (as determined by ovulation tests); non-fertile prototypes were created by averag-
Available online xxxx
ing 25 photographs of the same women during the luteal phase. Twenty different (new) female faces were
Keywords:
then shape transformed towards the luteal prototype and towards the follicular prototype in 50% and 100%
Female cycle steps. The two 50% transforms and the two 100% transforms were paired, resulting in stimulus pairs of two
Fertility cue different difculties. Thirty-six male participants were asked to choose the more attractive (Task 1), the
Ovulation more caring (Task 2), and the more irtatious face (Task 3). In a nal task the participants were asked to
Face perception choose the woman with which the participant would have better chances to get a date (Task 4). For all
Shape transformation tasks we found a signicant preference for the follicular face. In trials with a 100% transformation towards
Face morphing the shape of the prototype, the preference for the follicular stimulus was signicantly stronger than in trials
with a 50% transformation. We conclude that subtle shape differences in faces are sufcient to trigger men's
preference for a woman in her fertile cycle phase.
2012 Elsevier Inc. All rights reserved.

Introduction showed that t-shirts worn by females during different phases of their
cycle were judged to smell more pleasant and sexier when collected dur-
Female gonadal steroids have been shown to modulate socially rel- ing the late follicular phase compared to the luteal phase. A similar study
evant aspects of face processing, namely perception and interpretation by Havlicek et al. (2006) used cotton pads to gather odours and found
of attractiveness (see e.g., Jones et al., 2008 for a review). Several studies that men rated the odour from the follicular phase to be more pleasant
have shown that females' preference for male faces changes across the and more attractive but less intense compared to the odours gathered
menstrual cycle. Penton-Voak et al. (1999) showed that females pre- in other phases. Havlicek et al. (2006) as well as Singh and Bronstad
ferred less feminized male faces during the follicular phase. Moreover, (2001) used self-report data to estimate females' cycle and therefore
a negative correlation between progesterone level and preference for no direct link between females' hormonal variation and men's odour
masculine faces was found for male and female faces (Jones, et al., judgements can be made. Furthermore, both studies only inspected the
2005). Similarly, Roney and Simmons (2008) found that females who odour of females not taking hormonal contraception. This common re-
were tested near ovulation and who therefore had a higher estradiol striction to females who have a natural cycle seems to be reasonable as
concentration reported stronger preferences for faces of males with a one study that included females with and without hormonal contracep-
higher testosterone level than females who were tested during their lu- tion found this odour effect only for females not taking hormonal contra-
teal phase of the cycle. ception (Kuukasjarvi et al., 2004). An analogous study was conducted on
Not only do females' preferences for masculinity change across the females' voices. Pipitone and Gallup (2008) asked females to count from
menstrual cycle, but the attractiveness of females changes across the one to ten in different cycle phases. Voices recorded during the fertile
menstrual cycle as well. Recent studies suggest that there may be some cycle phase were rated as more attractive both by males and females.
perceptible cues to ovulation in humans. These cues are very subtle Again, this was only true for voices of women not taking hormonal con-
and can affect different senses. Singh and Bronstad (2001) for instance traception. In a comparable study (Fischer et al., 2011), English speaking
males were asked to judge samples of females' free speech in German
and samples of their vowel /a/. Whereas a preference for the free speech
Corresponding author at: Institute of Psychology, University of Bern, Muesmattstrasse
45, CH-3012 Bern, Switzerland. Fax: +41 31 631 36 06.
collected in the fertile phase was found, no such preference was found
E-mail addresses: cora.bobst@psy.unibe.ch (C. Bobst), janek.lobmaier@psy.unibe.ch for the vowel /a/ samples, indicating that a single vowel does not provide
(J.S. Lobmaier). enough information about the female cycle to trigger males' preference.

0018-506X/$ see front matter 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.yhbeh.2012.07.008

Please cite this article as: Bobst, C., Lobmaier, J.S., Men's preference for the ovulating female is triggered by subtle face shape differences,
Horm. Behav. (2012), http://dx.doi.org/10.1016/j.yhbeh.2012.07.008
2 C. Bobst, J.S. Lobmaier / Hormones and Behavior xxx (2012) xxxxxx

Similarly, there is evidence that females' body shape changes across the We investigated whether the choices differ depending on the ques-
cycle. Scutt and Manning (1996), for example, compared individuals' tion asked. In line with previous research on attractiveness judgements
length of each ear and the length of left and right digits, revealing that we expected a preference for the fertile female when asked to choose
the length of these body parts was more similar and therefore more sym- the more attractive face (see Roberts et al., 2004; Samson et al., 2011).
metric on the day of ovulation than during other cycle phases. The day of We might also expect faces of fertile women to be rated as more caring,
ovulation was determined by ultrasonography. Additionally, a study by since a study by Schultheiss et al. (2003a) has shown that naturally cy-
Kirchengast and Gartner (2002) showed that females' waist-to-hip cling women had elevated implicit afliation motivation around ovula-
ratio (WHR) was lower during ovulation. Note that these cyclic changes tion. For the tasks asking to choose the female that was perceived as
were measured objectively in both studies, but whether they are also vis- more irtatious or more upfront for a date, the data currently available
ible to the naked eye was not assessed. are more ambiguous: Whereas Schultheiss et al. (2003b) found that im-
Little research has been conducted on cyclic changes in female faces. plicit afliation motivation was rather unexpectedly negatively cor-
A study by Law Smith et al. (2006) showed positive correlations be- related with intercourse, several studies have found that females near
tween females' oestrogen levels assessed via urine samples checking ovulation were more receptive to male courtship (e.g. Gueguen,
for oestrone-3-glucuronide and attractiveness, femininity and health 2009a,b). Insofar a preference for the ovulating women might also be
ratings of their faces. This effect was found for female and male raters. expected in these tasks. To exclude a potential learning effect, partici-
However, this result is based on interindividual differences in oestrogen pants were tested twice.
levels rather than on intraindividual differences, thus no conclusions
about the differences between cycle phases within the same female Method
can be drawn.
Roberts et al. (2004) showed that portraits of women with a natu- Participants
ral cycle were rated as being more attractive when taken during the
fertile phase than when taken in the luteal phase. Overall, this was Thirty-six male participants aged between 19 and 37 years (M=24.2
true for male and female raters. However, this study was conducted SD=3.8) took part in this study; 35 indicated to be heterosexual, 1
with two independent study groups and in one of these groups this reported to be bisexual. Participants received course credits or a snack
preference was only found for female raters when additional infor- in return for their attendance. All reported normal or corrected-
mation, such as hairstyle, was shown. For male raters of the same to-normal vision and all provided written informed consent to take
group this result occurred only if additional information was exclud- part in this study. They were naive regarding the purpose of the experi-
ed. The shown pictures were actual photographs of women, hence it ment and were fully debriefed at the end of the study. The research was
remained unclear how the faces differed from each other. To this approved by the ethics committee of the Faculty of Human Sciences of
point conclusions can be drawn about the perceived attractiveness the University of Bern and participants were treated according to The
of female faces from different cycle phases but not on what basis Code of Ethics of the World Medical Association (Declaration of Helsinki).
these differences occur. Samson et al. (2011) hypothesized that the
effect found by Roberts et al. (2004) could result from changes in
skin colour, since Symons (1995) had shown that females' skin is Materials
lighter near ovulation. Similar to Roberts et al. (2004) Samson et al.
(2011) took pictures of females with a natural cycle in their fertile Stimuli
and non-fertile cycle phases. Male participants then rated the attrac- Stimuli were prepared in two steps. First a late follicular and a lu-
tiveness of these pictures. In addition, the colour and lightness of spe- teal prototype were created. In a second step new female faces were
cic areas such as forehead and cheeks of these pictures were transformed towards both prototypes.
measured by means of a spectrophotometer. In accordance with
Roberts et al. (2004), Samson and colleagues found that the pictures Prototypes. Prototypes (see Fig. 1) were composites created from pic-
taken in the fertile phase were rated as more attractive than the pic- tures of women taking part in an unrelated experiment investigating
tures taken in the luteal phase. However they found no difference of the inuence of the cycle phase on a perceptual task. Twenty-ve of
skin colour or lightness in the pictures of the two phases. Bleske- these women had consented to having their photographs taken and
Rechek et al. (2011) failed to replicate the results found by Roberts were photographed once during their late follicular cycle phase (ovula-
et al. (2004) and Samson et al. (2011). Similar to the previous studies, tion) and once during their luteal cycle phase. These females were not
Bleske-Rechek et al. (2011) showed photographs of women's faces, taking any kind of hormonal contraception and the time of ovulation
taken during luteal and late follicular cycle phase, to male and female was determined by urine samples using WH Ovultell ovulation test
raters. Again, cycle phase was estimated based on self-report. In addi- strips. They were photographed once within 24 h after the peak of the
tion, the mentioned study failed to nd the WHR changes that were luteinizing hormone (LH) as indicated by the ovulation test strip and
found by Kirchengast and Gartner (2002). It remains unclear why once 7 days later. Additionally, these females provided saliva samples
Bleske-Rechek et al. (2011) were unsuccessful since their approach during both photo sessions (i.e., at both time points in the menstrual
did not differ vastly from others. Haselton and Gildersleeve (2011) cycle) using a commercially available sampling device (Salivette;
therefore concluded in their recent overview on detection of ovulation Sarstedt, Rommelsdorf, Germany). Specically, participants placed a
cues in humans that more research on such cues in faces is needed. synthetic cotton swab in their mouths and chewed it for 45 s to stimu-
The current study aims at further investigating whether there effec- late salivation. The saliva samples were then stored at 20 C. After
tively is a face preference as was reported by Roberts et al. (2004) and thawing, saliva samples were analysed by an independent laboratory
Samson et al. (2011). If there is such a preference, which are the under- (Dresden Lab Service GmbH, Dresden, Germany) using commercially
lying cues for such a preference? The current study used standardized available radioimmunoassay kits adopted for the analysis of salivary
stimuli differing only by shape between the luteal and late follicular samples (IBL International, Hamburg, Germany). The sensitivity of the
phase in order to exclude further dimensions as reasons for the prefer- progesterone assay is 2.8 pg/ml, and the sensitivity of the estradiol
ence for faces of fertile females. To expand the focus, the preference assay is 0.3 pg/ml. Inter and intra assay variances are below 18.9% for
measurement was not only based on attractiveness judgements, but both assays. These assessments revealed that progesterone levels
also included judgements such as choosing the more caring or the were signicantly higher during the luteal phase (M = 46.58 pg/ml,
more irtatious face, and which woman the participant would have bet- SE= 4.73) than at the time of ovulation (M = 30.81 pg/ml, SE= 4.31;
ter chances to get a date with. t = 2.726, p = .012). However, levels of estradiol did not differ between

Please cite this article as: Bobst, C., Lobmaier, J.S., Men's preference for the ovulating female is triggered by subtle face shape differences,
Horm. Behav. (2012), http://dx.doi.org/10.1016/j.yhbeh.2012.07.008
C. Bobst, J.S. Lobmaier / Hormones and Behavior xxx (2012) xxxxxx 3

blocks each. In all blocks the same 80 stimulus pairs were used but
shown in a fully randomized order. The order of the four blocks however
was not randomized over the two sessions. Tasks were in two-
alternative choice format presented with the experimental software
SuperLab 4.0 (www.cedrus.com). Specically, a face pair consisting of
one luteal and one late follicular transform of the same identity was
presented. Each trial started with the presentation of a xation cross
(1000 ms) which was replaced with a stimulus pair displayed until
the participant's response. In the rst block participants were asked to
choose the more attractive face, pressing the keys F or J with their left
(F) and right (J) index ngers, indicating that the left or right face was
perceived as more attractive. In addition, after every judgement partici-
pants were asked to rate the condence of their judgement by pressing a
number between 1 and 4, where 1 indicated that participants were
very uncertain and 4 that they were very certain. In the remaining
blocks participants were asked to decide which face looked more caring
(Block 2), more irtatious (Block 3) or with which one they would have
Fig. 1. Follicular and luteal prototype. Composite of 25 women during ovulation (a) and better chances to get a date with (Block 4). In contrast to Block 1, no con-
composite of the same 25 women during the luteal phase of their menstrual cycle (b). dence rating was requested in Blocks 24.

the two phases (Movu = 3.63, SEovu = .27; Mlut = 3.83, SElut = .33; p = Results
.284).
Using PsychoMorph computer graphics software (Tiddeman et al., The proportion of follicular faces chosen (choice) and reaction times
2001), the shape of each face was manually dened with 178 facial land- (RTs) were analysed. Bonferroni corrected one-sample t tests revealed
mark points, marking the shape and position of eyes and brows, nose, that participants chose the follicular stimulus signicantly more often
mouth, ears, cheekbones, chin, as well as the outer face shape. We then than would have been expected by chance alone (.50); this was true
averaged the 25 face shapes of the images taken during ovulation to cre- for all four tasks, both sessions and both pair-differences; all p's b .05
ate a late follicular (i.e. fertile) prototype. The same was done with the 25 (see Fig. 3). Furthermore a 4 (tasks) 2 (session) 2 (shape transfor-
images of the luteal phase resulting in a luteal prototype. Averaging a mation condition) repeated measures analysis of variance (ANOVA)
group of faces into one image reveals consistent characteristics of this was performed separately for choice and reaction times. The Huynh
group while characteristics that are not shared are averaged out.

Stimulus transformation. Twenty frontal photographs of female faces


showing a neutral expression were selected for further use (derived
from the Karolinska Data Base KDEF; Lundqvist et al., 1998). We then as-
similated the shapes of these faces to the luteal and the late follicular pro-
totype (see above). Using these prototypes as opposite end points on a
continuum (ranging from fertile to non-fertile), each of the 20 stimulus
faces was shape transformed in two steps (50% and 100%) towards the
luteal and follicular prototypes, respectively (see Sprengelmeyer et al.,
2009 and Lobmaier et al., 2010 for similar procedures). For this, we
again manually dened the shape of each individual face with facial
landmarks (see above). Each of the 20 stimulus faces was then
shape-transformed using the differences between each landmark point
and the corresponding landmark point of the late follicular and luteal
prototypes. To receive the two steps, 50% and 100% of the difference be-
tween the landmark points of the respective prototypes were added to
the individual landmarks of each photograph (see Supplementary mate-
rials for a broader description of the transformation process). The trans-
formation was a shape transformation only, which implies that
transformed stimuli exclusively differ in shape but not in any other di-
mension, such as colour, luminance or contrast. More details on the
transformation process can be found by Tiddeman et al. (2001).
We paired the two 50% transforms (50% shape transformation con-
dition) and the two 100% transforms (100% shape transformation con-
dition). Each stimulus pair was shown in both lateralizations: showing
the luteal stimulus once on the left and once on the right. We thus
used a total of 80 (2 2 20) face pairs. Note that even with pairs
where the faces were transformed towards the prototypes by 100%
the differences between the two pictures were still very subtle (for an
example see Fig. 2).

Design and procedure


Fig. 2. Sample stimuli. Stimulus pair where each face was transformed 50% towards
Every participant came to the laboratory twice with an interval of each prototype (top panel) and 100% (bottom panel), respectively, the luteal stimulus
14 days. The two sessions were identical, consisting of four different is shown on the left. (This image shows Karolinska face af01nes.)

Please cite this article as: Bobst, C., Lobmaier, J.S., Men's preference for the ovulating female is triggered by subtle face shape differences,
Horm. Behav. (2012), http://dx.doi.org/10.1016/j.yhbeh.2012.07.008
4 C. Bobst, J.S. Lobmaier / Hormones and Behavior xxx (2012) xxxxxx

Fig. 3. Means of choices. Means of the choice of the follicular stimulus for the different tasks and transformation conditions (50%, 100%), separately for each session. Error bars depict
standard errors.

Feldt epsilon correction for heterogeneity of covariances (Huynh and participants pairs of female faces that differed in such a way that one
Feldt, 1976) was used when sphericity could not be assumed. For face was shape-transformed towards a prototypical ovulating woman,
post-hoc pairwise comparisons we used the Bonferroni correction. and the other was shape transformed towards a non-ovulating
For choice, there was a signicant effect of task, F2.773, 97.063 = 4.250, woman. Participants were asked to make four different choices: In
p = .009. The preference for the fertile face was highest in the irtatious Task 1, the men were asked to choose the more attractive face, in Task
task (M = .64, SD= .11) followed by the caring task (M = .63, SD= .10), 2 they had to choose the face that looked more caring, in Task 3 the
the dating task (M = .62, SD= .12) and the attractiveness task (M = .59, men were asked to choose the face that looked more irtatious, and -
SD= .14). However, post-hoc pairwise comparisons (Bonferroni nally in Task 4 the participants had to choose the woman with which
corrected) did not reveal any signicant differences between the they thought they would have a higher chance to get a date. Participants
tasks, all p's > .05. The effect of transformation condition also reached consistently preferred the fertile transform in all tasks. Reaction times
statistical signicance, F1, 35 = 51.484, p b .001, attributable to the fact as well as choice data indicate that the manipulation of stimuli was suc-
that in trials with 100% shape transformation the preference for the fol- cessful: for pairs with 100% shape transformation participants not only
licular stimulus was signicantly stronger than in trials with a 50% responded faster but they also more often chose the ovulating female
transformation. The main effect of session was not signicant, F1, 35 = and were more condent in their judgements compared to pairs with
2.009, p = .165. None of the interactions between any of the factors a 50% transformation. This nding is not surprising, as it reects that
reached statistical signicance (all p's > .08.). the task became easier as the difference between the two stimuli in-
Reaction times were z-transformed for every participant and all creased. While participants became faster over time both from task
values with a z higher than 2 and lower than 2 were excluded from to task as well as from session to session neither their answer patterns
further analyses. All main effects in the 4 2 2 ANOVA were signi- nor their condence judgements changed from Session 1 to Session 2.
cant; task, F1.887, 66.059 =46.312, p b .001, session, F1, 35 = 27.520, Hence, participants only became faster but not better in terms of choos-
p b .001, and transformation condition, F1, 35 = 11.264, p= .002. Partici- ing the face transformed to the follicular phase, implying that this pref-
pants responded slowest in the attractiveness task (M= 4125 ms, erence is stable over time.
SD= 969 ms) followed by the caring (M=3509 ms, SD =868 ms) and These ndings are in line with previous ndings (Roberts et al.,
the irtatious task (M= 3382 ms, SD =912 ms) and responded fastest 2004; Samson et al., 2011), who also found that women's faces were
in the dating task (M =3195 ms, SD =793 ms). Participants responded perceived as more attractive during the fertile window of their men-
faster in the second session (M= 3322 ms, SD =918 ms) than in the strual cycle. Importantly though, the present results extend the ndings
rst session (M= 3825 ms, SD= 857 ms) and they also responded of Roberts et al. (2004) and Samson et al. (2011). In the present study
faster in trials with a 100% shape transformation (M= 3521 ms, SD= we used face pairs that differed only in shape, all other aspects of the
845) than in trials with a 50% shape transformation (M= 3521 ms, stimulus faces were kept constant (e.g., facial colour). Whereas
SD= 846). The task session interaction was signicant, F1.743, 60.994 = Samson et al. (2011) found no evidence that colour is a reliable cue to
9.332, p =.001; attributable to the fact that reaction times were dispro- ovulation on the basis of which observers make their choices, we
portionately slower in Task 1 (attractiveness choice) of Session 1 com- found that shape is indeed a reliable cue: a face that was assimilated
pared to Session 2, while the reaction times of the other tasks did not to the shape of a prototypical ovulating woman was perceived as
differ as much between Session 1 and Session 2. None of the other inter- more attractive than a face that was assimilated to the shape of a
actions were signicant (all p's >.05). non-ovulating woman (i.e. a woman in the luteal cycle phase).
A 2 2 (session transformation condition) repeated measures A second aspect that broadened the scope of previous research is that
ANOVA was run on the condence judgements collected in Task 1 (at- we investigated preferences in four different judgements: participants
tractiveness choice). Whereas the main effect of transformation condi- were asked which face was perceived as more attractive, caring or irta-
tion was highly signicant, F1, 35 = 19.240, p b .001, neither the effect tious and with which female they would have better chances to get a date.
of session, F 1, 35 = 0.641, P = .429, nor the session transformation con- Our participants showed a general preference for the ovulating female,
dition interaction, F1, 35 = 2.501, p = .123, was signicant. Condence irrespective of the dimension on which the faces had to be judged.
judgements were higher in trials with a 100% shape transformation These results are in agreement with our hypotheses. Since we found a ro-
(M = 2.63, SD= 0.31) than in trials with a 50% shape transformation bust preference for the face transformed towards the ovulating phase in
(M = 2.47, SD= 0.37). all four tasks, it might well be that participants could not differentiate be-
tween the questions and instead chose the face for which they had an
Discussion unspecic preference. In order to solve this matter, further studies need
to investigate concepts that are clearly associated with the luteal cycle
The present study investigated whether men can detect subtle phase. Furthermore it will be interesting to more closely investigate the
cues to ovulation in female faces. Specically, we presented male operationalisation of the concepts explored in the present study.

Please cite this article as: Bobst, C., Lobmaier, J.S., Men's preference for the ovulating female is triggered by subtle face shape differences,
Horm. Behav. (2012), http://dx.doi.org/10.1016/j.yhbeh.2012.07.008
C. Bobst, J.S. Lobmaier / Hormones and Behavior xxx (2012) xxxxxx 5

Specically, the concept of caring and irtatious as well as having References


better chances to get a date with should be reconsidered in order to
Bleske-Rechek, A., Harris, H.D., Denkinger, K., Webb, R.M., Erickson, L., Nelson, L.A.,
draw reliable conclusions. For example, it could have been expected 2011. Physical cues of ovulatory status: a failure to replicate enhanced facial attrac-
that a person being judged as irtatious would not necessarily also be tiveness and reduced waist-to-hip ratio at high fertility. Evol. Psychol. 9, 336353.
judged as caring. However, the difference between these two attributes Fischer, J., Semple, S., Fickenscher, G., Juergens, R., Kruse, E., Heistermann, M., Amir, O., 2011.
Do women's voices provide cues of the likelihood of ovulation? The importance of sam-
was not apparent in the choices made in the present study. pling regime. PLoS One 6, e24490, http://dx.doi.org/10.1371/journal.pone.0024490.
We note that our nding suggesting that shape differences are suf- Grammer, K., Renninger, L., Fischer, B., 2004. Disco clothing, female sexual motivation,
cient to provoke males' preference for the fertile female face does and relationship status: is she dressed to impress? J. Sex Res. 41, 6674.
Gueguen, N., 2009a. Menstrual cycle phases and female receptivity to a courtship solic-
not imply that men can reliably spot an ovulating female. Further- itation: an evaluation in a nightclub. Evol. Hum. Behav. 30, 351355, http://
more, our ndings do not reveal the causal reason for these shape dif- dx.doi.org/10.1016/j.evolhumbehav.2009.03.004.
ferences, nor do we know what these shape differences essentially Gueguen, N., 2009b. The receptivity of women to courtship solicitation across the men-
strual cycle: a eld experiment. Biol. Psychol. 80, 321324, http://dx.doi.org/
characterize. High estradiol levels have been associated with female
10.1016/j.biopsycho.2008.11.004.
facial characteristics such as softer jaw and suppressed eyebrow Haselton, M.G., Gildersleeve, K., 2011. Can men detect ovulation? Curr. Dir. Psychol. Sci.
growth and women high in estradiol have been shown to be more at- 20, 8792, http://dx.doi.org/10.1177/0963721411402668.
tractive (Law Smith et al., 2006). However, the preferences for the Havlicek, J., Dvorakova, R., Bartos, L., Flegr, J., 2006. Non-advertized does not mean
concealed: body odour changes across the human menstrual cycle. Ethology 112,
fertile face shape observed in the present study are unlikely to be 8190, http://dx.doi.org/10.1111/j.1439-0310.2006.01125.x.
caused by changes in estradiol levels, since the saliva concentration Huynh, H., Feldt, L.S., 1976. Estimation of the box correction for degrees of freedom from
of estradiol was comparable in the two phases in which the photo- sample data in randomized block and split-block design. J. Educ. Stat. 1, 6982.
Jones, B.C., Perrett, D.I., Little, A.C., Boothroyd, L., Cornwell, R.E., Feinberg, D.R.,
graphs were taken. Instead, the shape difference that drives the pref- Tiddeman, B.P., Whiten, S., Pitman, R.M., Hillier, S.G., Burt, D.M., Stirrat, M.R., Law
erence observed in the present study is more likely to be linked to Smith, M.J., Moore, F.R., 2005. Menstrual cycle, pregnancy and oral contraceptive
differences in progesterone levels, which were signicantly higher use alter attraction to apparent health in faces. Proc. Biol. Sci. 272, 347354,
http://dx.doi.org/10.1098/rspb.2004.2962.
during the luteal phase than during ovulation. Furthermore, the Jones, B.C., DeBruine, L.M., Perrett, D.I., Little, A.C., Feinberg, D.R., Smith, M.J.L., 2008. Ef-
shape differences also may have been caused by the LH surge during fects of menstrual cycle phase on face preferences. Arch. Sex. Behav. 37, 7884,
the late follicular phase. We note that we did not directly measure LH http://dx.doi.org/10.1007/s10508-007-9268-y.
Kirchengast, S., 1993a. Anthropometric-hormonal correlation patterns in fertile
levels; we merely determined the LH surge by means of an ovulation
and postmenopausal women from Austria. Ann. Hum. Biol. 20, 4765, http://
test. Finally, various other hormones that uctuate during the men- dx.doi.org/10.1080/03014469300002492.
strual cycle may mediate the shape difference. Kirchengast, S., 1993b. Body shape and sex-hormone levels in fertile and postmeno-
pausal women from eastern Austria. Homo 44, 145167.
Alternatively, these hormonal effects may not be direct but might
Kirchengast, S., Gartner, M., 2002. Changes in fat distribution (WHR) and body weight
instead be mediated by behavioural differences. Given that women's across the menstrual cycle. Coll. Antropol. 26, 4757.
behaviour especially towards men is inuenced by their menstru- Kuukasjarvi, S., Eriksson, C.J.P., Koskela, E., Mappes, T., Nissinen, K., Rantala, M.J., 2004. Attrac-
al cycle (e.g. Grammer et al., 2004; Gueguen, 2009a,b) it is possible tiveness of women's body odors over the menstrual cycle: the role of oral contraceptives
and receiver sex. Behav. Ecol. 15, 579584, http://dx.doi.org/10.1093/beheco/arh050.
that females in their fertile cycle phase might have been irting Law Smith, M.J., Perrett, D.I., Jones, B.C., Cornwell, R.E., Moore, F.R., Feinberg, D.R.,
more with the camera compared to when they were photographed Boothroyd, L.G., Durrani, S.J., Stirrat, M.R., Whiten, S., Pitman, R.M., Hillier, S.G.,
in their luteal phase. If so, this might have slightly inuenced face 2006. Facial appearance is a cue to oestrogen levels in women. Proc. Biol. Sci.
273, 135140, http://dx.doi.org/10.1098/rspb.2005.3296.
shape. It will have to be the aim of future studies to further investi- Lobmaier, J.S., Sprengelmeyer, R., Witten, B., Perrett, D.I., 2010. Female and male re-
gate the underlying mechanisms of these differences. Specically it sponses to cuteness, age and emotion in infant faces. Evol. Hum. Behav. 31,
will be interesting to further explore how hormonal uctuations are 1621, http://dx.doi.org/10.1016/j.evolhumbehav.2009.05.004.
Lundqvist, D., Flykt, A., Oehman, A., 1998. The Karolinska Directed Emotional Faces
linked to facial morphology. There is some evidence suggesting that KDEF, CD ROM from Department of Clinical Neurosciences, Psychology Section.
inter-individual hormone level differences have an effect on mor- Karolinska Institutet. ISBN 91-630-7164-9.
phology (e.g. Kirchengast, 1993a,b; Raschka, 1993), however to our Penton-Voak, I.S., Perrett, D.I., Castles, D.L., Kobayashi, T., Burt, D.M., Murray, L.K.,
Minamisawa, R., 1999. Menstrual cycle alters face preference. Nature 399,
best knowledge, no research has yet examined the interrelationship 741742, http://dx.doi.org/10.1038/21557.
between cyclic hormonal uctuations and changes in facial morphol- Pipitone, R.N., Gallup Jr., G.G., 2008. Women's voice attractiveness varies across
ogy during the menstrual cycle. the menstrual cycle. Evol. Hum. Behav. 29, 268274, http://dx.doi.org/10.1016/
j.evolhumbehav.2008.02.001.
To our best knowledge the present study provides rst evidence
Raschka, C., 1993. Sports anthropological knowledge of relationships between individual
that the shape of a female face changes subtly across her menstrual hormone levels and anthropometric measuring values (Sportanthropologische
cycle and that these changes are perceivable to the male eye. Howev- Erkenntnisse uber Bezuge zwischen individuellem Hormonspiegel und
er, our results do not imply that men can in fact identify the female's anthropometrischen Messwerten). Schweiz. Z. Sportmed. 41, 6774.
Roberts, S.C., Havlicek, J., Flegr, J., Hruskova, M., Little, A.C., Jones, B.C., Perrett, D.I., Petrie, M.,
fertile cycle phase, since the preference for the fertile female face can- 2004. Female facial attractiveness increases during the fertile phase of the menstrual
not be equated with explicitly recognizing fertility cues. cycle. Proc. Biol. Sci. 271, S270S272, http://dx.doi.org/10.1098/rsbl.2004.0174.
Roney, J.R., Simmons, Z.L., 2008. Women's estradiol predicts preference for facial cues of men's
testosterone. Horm. Behav. 53, 1419, http://dx.doi.org/10.1016/j.yhbeh.2007.09.008.
Samson, N., Fink, B., Matts, P., 2011. Does a woman's skin color indicate her fertility level? Pre-
Acknowledgments liminary ndings. Swiss J. Psychol. 70, 199202, http://dx.doi.org/10.1024/1421-0185/
a000057.
Schultheiss, O.C., Dargel, A., Rohde, W., 2003a. Implicit motives and gonadal steroid
The authors thank Jessika Golle for her helpful comments on a pre-
hormones: effects of menstrual cycle phase, oral contraceptive use, and relationship
vious version of this manuscript and Fabian Probst and Laura status. Horm. Behav. 43, 293301, http://dx.doi.org/10.1016/s0018-506x(03)00003-5.
Bachofner for their help with collecting the photographs for the pro- Schultheiss, O.C., Dargel, A., Rohde, W., 2003b. Implicit motives and sexual motivation and
totypes. We also thank Clemens Kirschbaum for his help with the behavior. J. Res. Pers. 37, 224230, http://dx.doi.org/10.1016/s0092-6566(02)00568-8.
Scutt, D., Manning, J.T., 1996. Symmetry and ovulation in women. Hum. Reprod. 11,
hormone analyses. This study was supported by a grant from the 24772480.
Swiss National Science Foundation awarded to JL (grant number Singh, D., Bronstad, P.M., 2001. Female body odour is a potential cue to ovulation. Proc.
PZ00P1_121622/1). Biol. Sci. 268, 797801.
Sprengelmeyer, R., Perrett, D.I., Fagan, E.C., Cornwell, R.E., Lobmaier, J.S.,
Sprengelmeyer, A., Young, A.W., 2009. The cutest little baby face: a hormonal
link to sensitivity to cuteness in infant faces. Psychol. Sci. 20, 149154, http://
Appendix A. Supplementary data dx.doi.org/10.1111/j.1467-9280.2009.02272.x.
Symons, D., 1995. The Evolution of Human Sexuality. Oxford University Press, Oxford.
Tiddeman, B., Burt, M., Perrett, D., 2001. Prototyping and transforming facial textures
Supplementary data to this article can be found online at http:// for perception research. IEEE Comput. Graph. Appl. 21, 4250, http://dx.doi.org/
dx.doi.org/10.1016/j.yhbeh.2012.07.008. 10.1109/38.946630.

Please cite this article as: Bobst, C., Lobmaier, J.S., Men's preference for the ovulating female is triggered by subtle face shape differences,
Horm. Behav. (2012), http://dx.doi.org/10.1016/j.yhbeh.2012.07.008

Potrebbero piacerti anche